【剑桥哲学指南】休谟 558页

thecambridgecompaniontoHUMESecondEditionEachCambridgeCompaniontoaphilosophicalfigureismadeupofspeciallycommissionedessaysbyaninternationalteamofscholars,providingstudentsandnonspecialistswithanintro-ductiontoamajorphilosopher.Theseriesaimstodispeltheintimidationthatreadersmayfeelwhenfacedwiththeworkofachallengingthinker.DavidHumeisnowconsideredoneofthemostimportantphilosophersoftheWesternworld.Althoughbestknownforhiscontributionstothetheoryofknowledge,metaphysics,andphi-losophyofreligion,Humealsoinfluenceddevelopmentsinthephilosophyofmind,psychology,ethics,politicalandeconomictheory,politicalandsocialhistory,andaesthetictheory.Thefif-teenessaysinthisvolumeaddressallaspectsofHume’sthought.Thepictureofhimthatemergesisthatofathinkerwho,thoughoftencriticaltothepointofskepticism,wasnonethelessabletobuildonthatskepticismaconstructive,viable,andprofoundlyimportantviewoftheworld.AlsoincludedinthisvolumeareHume’stwobriefautobiographiesandabibliographysuitedtothosebeginningtheirstudyofHume.ThissecondeditionofoneofourmostpopularCompanionsincludessixnewessaysandanewintroduction;theremainingessayshaveallbeenrevisedandupdated.DavidFateNortonisProfessorofMoralPhilosophyEmeritusatMcGillUniversityandAdjunctProfessorofPhilosophyattheUniversityofVictoria.HehaspublishedwidelyonHumeandeighteenth-centuryBritishphilosophyandrecentlycoeditedthefirstcriticaleditionofHume’sTreatiseofHumanNature.JacquelineTaylorisAssociateProfessorofPhilosophyattheUni-versityofSanFrancisco.ShehaspublishedonHume’sphiloso-phyinHumeStudies,Topoi,theJournalofEthics,TheBlackwellGuidetoHumesTreatise,andFeministInterpretationsofDavidHume.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nothervolumesintheseriesofcambridgecompanionstophilosophyABELARDEditedbyjeffreye.browerandkevinguilfoyADORNOEditedbythomashuhnANSELMEditedbybriandaviesandbrianleftowAQUINASEditedbynormankretzmannandeleonorestumpARABICPHILOSOPHYEditedbypeteradamsonandrichardc.taylorHANNAHARENDTEditedbydanavillaARISTOTLEEditedbyjonathanbarnesATHEISMEditedbymichaelmartinAUGUSTINEEditedbyeleonorestumpandnormankretzmannBACONEditedbymarkkupeltonenSIMONEDEBEAUVOIREditedbyclaudiacardBERKELEYEditedbykennethp.winklerBRENTANOEditedbydalejacquetteCARNAPEditedbyrichardcreathandmichaelfriedmanCRITICALTHEORYEditedbyfredrushDARWINEditedbyjonathanhodgeandgregoryradickDESCARTESEditedbyjohncottinghamDUNSSCOTUSEditedbythomaswilliamsEARLYGREEKPHILOSOPHYEditedbya.a.longEARLYMODERNPHILOSOPHYEditedbydonaldrutherfordFEMINISMINPHILOSOPHYEditedbymirandafrickerandjenniferhornsbyFOUCAULTSecondEditionEditedbygaryguttingFREUDEditedbyjeromeneuGADAMEREditedbyrobertj.dostalGALENEditedbyr.j.hankinsonGALILEOEditedbypetermachamerGERMANIDEALISMEditedbykarlameriksContinuedaftertheIndexCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheCambridgeCompaniontoHUMESecondEditionEditedbyDavidFateNortonMcGillUniversityandUniversityofVictoriaJacquelineTaylorUniversityofSanFranciscoCambridgeCollectionsOnline©CambridgeUniversityPress,2009\ncambridgeuniversitypressCambridge,NewYork,Melbourne,Madrid,CapeTown,Singapore,SaoPaulo,Delhi˜CambridgeUniversityPress32AvenueoftheAmericas,NewYork,NY10013-2473,USAwww.cambridge.orgInformationonthistitle:www.cambridge.org/9780521677349cCambridgeUniversityPress1993,2009Thispublicationisincopyright.Subjecttostatutoryexceptionandtotheprovisionsofrelevantcollectivelicensingagreements,noreproductionofanypartmaytakeplacewithoutthewrittenpermissionofCambridgeUniversityPress.Firstpublished1993Secondeditionpublished2009PrintedintheUnitedStatesofAmericaAcatalogrecordforthispublicationisavailablefromtheBritishLibrary.LibraryofCongressCataloginginPublicationDataTheCambridgecompaniontoHume/[editedby]DavidFateNorton,JacquelineTaylor.–2nded.p.cm.Includesbibliographicalreferencesandindex.ISBN978-0-521-85986-8(hardback)–ISBN978-0-521-67734-9(pbk.)1.Hume,David,1711–1776.I.Norton,DavidFate.II.Taylor,Jacqueline.B1498.C262008192–dc222007044296ISBN978-0-521-85986-8hardbackISBN978-0-521-67734-9paperbackCambridgeUniversityPresshasnoresponsibilityforthepersistenceoraccuracyofURLsforexternalorthird-partyInternetWebsitesreferredtointhispublicationanddoesnotguaranteethatanycontentonsuchWebsitesis,orwillremain,accurateorappropriate.Informationregardingprices,traveltimetables,andotherfactualinformationgiveninthisworkarecorrectatthetimeoffirstprinting,butCambridgeUniversityPressdoesnotguaranteetheaccuracyofsuchinformationthereafter.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\ncontentsListofContributorspageixPrefacetotheSecondEditionxiiiMethodofCitationxvListofAbbreviationsxvii1AnIntroductiontoHume’sThought1davidfatenorton2Hume’sNewScienceoftheMind40johnbiro3HumeandtheMechanicsofMind:Impressions,Ideas,andAssociation70davidowen4Hume’sTheoryofSpaceandTimeinItsSkepticalContext105donaldl.m.baxter5HumeonCausation147martinbell6HumeandtheProblemofPersonalIdentity177janel.mcintyre7Hume’sSkepticism209robertj.fogelin8Hume’sMoralPsychology238terencepenelhumviiCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nviiiContents9TheFoundationsofMoralityinHume’sTreatise270davidfatenorton10Hume’sLaterMoralPhilosophy311jacquelinetaylor11TheStructureofHume’sPoliticalTheory341knudhaakonssen12Hume’sPrinciplesofPoliticalEconomy381andrews.skinner13HumeontheArtsand“TheStandardofTaste”:TextsandContexts414peterjones14DavidHume:“TheHistorian”447davidwootton15HumeonReligion480j.c.a.gaskinAppendix:Hume’sAutobiographies515I.AKindofHistoryofMyLife(1734)515II.MyOwnLife(1776)522SelectedBibliography531Index539CambridgeCollectionsOnline©CambridgeUniversityPress,2009\ncontributorsdonaldl.m.baxterisProfessorofPhilosophy,UniversityofCon-necticut.HeistheauthorofHumesDifficulty:TimeandIdentityintheTreatise(NewYork:Routledge,2008),aswellasvariouspapersinearlymodernWesternphilosophyandinsystematicmetaphysics.martinbellisProfessorofHistoryofPhilosophy,ManchesterMetropolitanUniversity.HispublicationsonHume’sphilosophyappearasjournalarticlesandaschaptersineditedcollections,includingReadingHumeonHumanUnderstanding(2002),Impres-sionsofHume(2005),NewEssaysonDavidHume(2007),andACompaniontoHume(2008).HealsoeditedHume’sDialoguescon-cerningNaturalReligionforPenguinClassics(1990).johnbiroiscoeditorofSpinoza:NewPerspectives(1978);Mind,BrainandFunction(1982);Frege:SenseandReferenceaHundredYearsLater(1995);andSpinoza:MetaphysicalThemes(2002).Heisalsotheauthorofpapersonavarietyoftopicsinepistemology,thephilosophyofmind,andthephilosophyoflanguage.robertj.fogelinisShermanFairchildProfessorintheHumani-tiesandProfessorofPhilosophyEmeritus,DartmouthCollege.HeisauthorofHumesSkepticismintheTreatiseofHumanNature(1985),Wittgenstein(2nded.,1987),PyrrhonianReflectionsonKnowledgeandJustification(1994),BerkeleyandthePrinciplesofHumanKnowledge(2001),ADefenseofHumeonMiracles(2003),WalkingtheTightropeofReason(2003),andmanyarticlesonawiderangeofphilosophicaltopics.j.c.a.gaskinisaFellowofTrinityCollegeDublin,wherehealsoheldapersonalchairinphilosophy.HispublicationsincludeHumesixCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nxContributorsPhilosophyofReligion(2nded.,1988)andVarietiesofUnbelief(1989).HeistheauthorofnumerousarticlesonHumeandonthephilosophyofreligionandtheeditoroftheWorld’sClassicseditionofHume’sworksonreligionandHobbes’sLeviathan.Heisalsotheauthoroftwovolumesofghoststories,TheDarkCompanion(2001)andTheLongRetreatingDay(2006).knudhaakonssenisProfessorofIntellectualHistoryandDirectoroftheSussexCentreforIntellectualHistory,UniversityofSussex.HisbooksincludeTheScienceofaLegislator(1981);(ed.)ACul-tureofRights(1991);NaturalLawandMoralPhilosophy(1996);(ed.)EnlightenmentandReligion(1996);(ed.)TheCambridgeHis-toryofEighteenth-CenturyPhilosophy(2006);(ed.)TheCambridgeCompaniontoAdamSmith(2006);andeditionsofworksbyFrancisHutcheson,Hume,Smith,andThomasReid.peterjonesisProfessorEmeritusofPhilosophy,UniversityofEdin-burgh,wherehewasalsoDirectoroftheInstituteforAdvancedStud-iesintheHumanities.Hehaspublishedextensivelyonphilosophi-calandculturaltopicsincludingeighteenth-centuryaesthetics,theRussiannovel,Italianopera,andarchitecture.AmonghisbooksarePhilosophyandtheNovel(1975);HumesSentiments:TheirCicero-nianandFrenchContext(1982);(ed.)TheReceptionofDavidHumeinEurope(2005);andOveArup:MasterbuilderoftheTwentiethCentury(2006).janel.mcintyreisProfessorEmeritaofPhilosophy,ClevelandStateUniversity,andapastpresidentoftheHumeSociety.HernumerouspublicationsargueforthecentralityofHume’saccountofthepassionstohisoverallphilosophicalposition.Notablearticlesinclude“PersonalIdentityandthePassions”(1989)and“Character:AHumeanAccount”(HistoryofPhilosophyQuarterly1990).davidfatenortonisProfessorofPhilosophyEmeritus,McGillUniversity,andAdjunctProfessorofPhilosophy,UniversityofVic-toria.HeistheauthorofDavidHume:Common-SenseMoralist,ScepticalMetaphysician(rev.ed.1984)andmanyarticlesonHumeandeighteenth-centuryScottishphilosophy;coauthorofTheDavidHumeLibrary(1996);andcoeditorofDavidHume:Philosophi-calHistorian(1965),McGillHumeStudies(1978),andtheOxfordCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nContributorsxiPhilosophicalTextsandClarendoneditionsofHume’sATreatiseofHumanNature(2000,2007).davidowenreceivedhisB.Phil.andD.Phil.fromOxfordUniver-sity.HehastaughtinScotland,England,andCanadaandcurrentlyteachesphilosophyattheUniversityofArizona.HeistheauthorofHumesReason(1999)andeditorofHume:GeneralPhilosophy(2000),andhehaspublishedmanyarticlesinthehistoryofearlymodernphilosophy,especiallyonLockeandHume.terencepenelhumisProfessorEmeritusofReligiousStudies,Uni-versityofCalgary,wherehewasformerlyProfessorofPhilosophyandDirectoroftheCalgaryInstitutefortheHumanities.HisbooksincludeHume(1975),GodandSkepticism(1983),Butler(1985),DavidHume:AnIntroductiontoHisPhilosophicalSystem(1992),andThemesinHume(2000).andrews.skinnerisDanielJackProfessorofPoliticalEconomyEmeritus,UniversityofGlasgow.Inadditiontomanypapersonthehistoryofeconomicthought,heistheauthorofASystemofSocialScience:PapersRelatingtoAdamSmith(1979)andcoeditorofEssaysonAdamSmith(1975)andofSmith’sAnInquiryintotheNatureandCausesoftheWealthofNations(1976).jacquelinetaylorisAssociateProfessorofPhilosophy,Univer-sityofSanFrancisco.ShehasservedontheExecutiveCommitteeoftheHumeSocietyandcurrentlyservesasaneditorforHumeStud-ies.HerarticlesonHume’sphilosophyandHumeanethicshaveappearedinjournalsandinvariousanthologiesonHume,includ-ingFeministInterpretationsofDavidHume(2000),ThePracticeofVirtue(2005),TheBlackwellGuidetoHumesTreatise(2006),andTheCambridgeCompaniontoHumesTreatise(forthcoming).davidwoottonisAnniversaryProfessorofHistory,UniversityofYork.HeistheauthorofPaoloSarpi:BetweenRenaissanceandEnlightenment(2003)andBadMedicine:DoctorsDoingHarmsinceHippocrates(2006).HehaspublishedwidelyonRenaissanceandEnlightenmentpoliticaltheoryandiscurrentlywritingabiographyofGalileo.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nprefacetothesecondeditionThissecondeditionofTheCambridgeCompaniontoHumeincludesfiveentirelynewessays.Theseare,intheirorderofappear-ance,“HumeandtheMechanicsofMind:Impressions,Ideas,andAssociation,”byDavidOwen;“Hume’sTheoryofSpaceandTimeinItsSkepticalContext,”byDonaldL.M.Baxter;“HumeonCau-sation,”byMartinBell;“HumeandtheProblemofPersonalIden-tity,”byJaneL.McIntyre;and“Hume’sLaterMoralPhilosophy,”byJacquelineTaylor.Inaddition,“TheFoundationsofMoralityinHume’sTreatise,”byDavidFateNorton,bearsonlyafamilyresem-blancetotheessay“Hume,HumanNatureandtheFoundationsofMorality”foundinthefirstedition.OneessayincludedinthefirsteditionofthisCompanion,AlexanderRosenberg’scomprehensive“HumeandthePhilosophyofScience,”hasbeenomittedhereinfavorofthemoredetailedessaysbyOwen,Baxter,andBell.Inaddition,theessaysbyJohnBiro(“Hume’sNewScienceoftheMind”)andPeterJones(“HumeontheArtsand‘TheStandardofTaste’:TextsandContexts”)areinthiseditionsubstantiallyrevised,whilethepapersbyTerencePenelhum,KnudHaakonssen,AndrewSkinner,andJ.C.A.Gaskinhavebeenrevisedandupdated.More-over,mostessaysinthisvolumeconcludewithanewsetofsugges-tionsforfurtherreadingrelatedtothetopicoftheessay,whiletheSelectedBibliographylistsHume’sprincipalpublicationsandrec-ommendsamanageablesetofbooks,bothanthologiesandmono-graphs,thatofferdiverseandimportantinterpretationsofHume’sthought.xiiiCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nxivPrefacetotheSecondEditionTheeditorsgratefullyacknowledgeBeccaChase,MiriamMc-Cormick,NormanTaylor,andespeciallyMaryJ.Norton,fortheirassistanceinpreparingtheseessaysforpublication.AtCambridgeUniversityPresswehavebeengraciouslyandprofessionallyassistedbythelateTerryMoore,andalsobyBeatriceRehl,HelenWheeler,andDavidAnderson.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nmethodofcitationReferencestoHume’stextsarenormallygivenparenthetically,butonsomeoccasionsthesereferencesmaybeplacedinanote.Paren-theticalreferencesfollowtheformoftheseexamples:ATreatiseofHumanNature:TfollowedbyBook,Part,Section,andparagraphnumbers1oftheOxfordUniversityPresseditions(seeTinthelistofabbreviations)fromwhichallquotationsofthisworkaretaken,toproduce,forexample,theform(T3.1.1.26).AnEnquiryconcerningHumanUnderstanding:EHUfollowedbytheSectionandparagraphnumbersoftheOxfordUniversityPresseditions(seeEHUinthelistofabbreviations)fromwhichallquota-tionsofthisworkaretaken,toproduce,forexample,theform(EHU12.12).AnEnquiryconcerningthePrinciplesofMorals:EPMfollowedbytheSectionandparagraphnumbersintheOxfordUniversityPresseditions(seeEPMinthelistofabbreviations)fromwhichallquota-tionsofthisworkaretaken,toproduce,forexample,theform(EPM3.12).Essays:E-,followedfirstbyanabbreviationofthetitleofthepar-ticularessaycited(fortheseabbreviationsseeEinthelistofabbre-viations),thenbythenumbersoftheparagraphscited,andlastbytherelevantpagenumber(s)oftheeditionofHume’sEssaysMoral,1TheuseofparagraphnumbersinthiswayallowsthecitedmaterialstobefoundinnearlyallofthemanyeditionsofHume’sworks.xvCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nxviMethodofCitationPolitical,andLiterarydescribedatE,toproduce,forexample,theform(E-ST3,228).TheHistoryofEngland:HEfollowedfirstbychapterandparagraphnumbers,andthenbythevolumeandpagenumbersoftheeditiondescribedatHE,toproduce,forexample,theform(HE66.43,6:307–8).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nlistofabbreviationsAAnAbstractofaBooklatelyPublished;EntituledATrea-tiseofHumanNature,&c.WhereintheChiefArgu-mentofthatBookisfartherIllustratedandExplained(firstpublished1740).CitedfromtheClarendonEdi-tionofATreatiseofHumanNature(Oxford:ClarendonPress,2007),orfromthe9thandsubsequentimpres-sionsoftheOxfordPhilosophicalTexts(OPT)Edition(Oxford:OxfordUniversityPress,2005)ofthissamework;seeTlaterinthislist.DADialogue(firstpublished1751),citedfromtheOxfordPhilosophicalTextsEdition(Oxford:OxfordUniversityPress,1998)ortheClarendonEditionofAnEnquiryconcerningthePrinciplesofMorals(Oxford:Claren-donPress,1998),bothed.T.L.Beauchamp.Althoughthesetwoeditionsincorporatenotablydifferentedito-rialmaterials,theirtextsofDareidentical.DNRDialoguesconcerningNaturalReligion(firstpublished1779),ed.N.K.Smith(Oxford:ClarendonPress,1935;2nded.London:ThomasNelson&Sons,1947;3rded.,NewYork:LibraryofLiberalArts[1964?]).Thethreeeditionsareuniformlypaginated.DPDissertationonthePassions(firstpublished1757),citedfromADissertationonthePassionsandTheNatu-ralHistoryofReligion,ed.T.L.Beauchamp(Oxford:ClarendonPress,2007).EEssaysMoral,Political,andLiterary,ed.E.F.Miller(Indi-anapolis:LibertyClassics,rev.ed.,1987).AbbreviationsxviiCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nxviiiAbbreviationsoftheindividualessayscitedinthisvolume,withdateoffirstpublication,followthisentry.E-BGWhethertheBritishGovernmentinclinesmoretoAbsoluteMonarchy,ortoaRepublic(1741)E-BTOftheBalanceofTrade(1752)E-CLOfCivilLiberty(1741)E-CoOfCommerce(1752)E-CPOftheCoalitionofParties(1760)E-CRACharacterofSirRobertWalpole(1742;withdrawnafter1768)E-DMOftheDignityorMeannessofHumanNature(1741)E-DTOftheDelicacyofTasteandPassion(1741)E-EpTheEpicurean(1742)E-FPOftheFirstPrinciplesofGovernment(1741)E-IMOfImpudenceandModesty(1741;withdrawnafter1760)E-InOfInterest(1752)E-IPOftheIndependencyofParliament(1741)E-IPCIdeaofaPerfectCommonwealth(1752)E-ISOftheImmortalityoftheSoul(1777,afterhavingbeenwithdrawnin1757)E-JTOftheJealousyofTrade(1760)E-LPOftheLibertyofthePress(1741)E-MoOfMoney(1752)E-NCOfNationalCharacters(1748)E-OCOftheOriginalContract(1748)E-OGOftheOriginofGovernment(1777)E-PAOfthePopulousnessofAncientNations(1752)E-PCOfPublicCredit(1752)E-PGOfPartiesinGeneral(1741)E-PGBOfthePartiesofGreatBritain(1741)E-PRThatPoliticsmaybereducedtoaScience(1741)E-PSOftheProtestantSuccession(1752)E-RAOfRefinementintheArts(1752)E-RPOftheRiseandProgressoftheArtsandSciences(1742)E-ScTheSceptic(1742)E-SEOfSuperstitionandEnthusiasm(1741)E-SHOftheStudyofHistory(1741;withdrawnafter1760)CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAbbreviationsxixE-SROfSimplicityandRefinementinWriting(1742)E-STOftheStandardofTaste(1757)E-SuOfSuicide(1777,afterhavingbeenwithdrawnin1757)E-TaOfTaxes(1752)E-TrOfTragedy(1757)EHUAnEnquiryconcerningHumanUnderstanding(firstpublished,1748,asPhilosophicalEssaysconcerningHumanUnderstanding).AllquotationsandreferencesaretotheOxfordPhilosophicalTextsEdition(Oxford:OxfordUniversityPress,1999)ortotheClarendonEdi-tion(Oxford:ClarendonPress,2000),bothed.T.L.Beauchamp.Althoughthesetwoeditionsprovidesig-nificantlydifferenteditorialmaterials,theirtextsofEHUareidentical.EPMAnEnquiryconcerningthePrinciplesofMorals(firstpublished1751).AllquotationsandreferencesaretotheOxfordPhilosophicalTextsEdition(Oxford:OxfordUniversityPress,1998)ortotheClarendonEdition(Oxford:ClarendonPress,1998),bothed.T.L.Beauchamp.Althoughthesetwoeditionsprovidesignificantlydifferenteditorialmaterials,theirtextsofEPMareidentical.HETheHistoryofEngland(firstpublished1754–62),6vols.(Indianapolis:LibertyClassics,1983).HLTheLettersofDavidHume,ed.J.Y.T.Greig,2vols.(Oxford:ClarendonPress,1932).KHLAKindofHistoryofMyLife,citedfromtheappendixofthisvolume.LALetterfromaGentlemantoHisFriendinEdinburgh:containingObservationson...thePrinciples...saidtobemaintaindin...ATreatiseofHumanNature(firstpublished1745),citedfromATreatiseofHumanNature,ed.D.F.NortonandM.J.Norton(Oxford:ClarendonPress,2007).MOLMyOwnLife(firstpublished1777),citedfromtheappendixofthisvolume.NHLNewLettersofDavidHume,ed.R.KlibanskyandE.C.Mossner(Oxford:ClarendonPress,1954).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nxxAbbreviationsNHRTheNaturalHistoryofReligion(firstpublished1757),citedfromADissertationonthePassionsandTheNatu-ralHistoryofReligion,ed.T.L.Beauchamp(Oxford:ClarendonPress,2007).TATreatiseofHumanNature(firstpublished1739–40).AllquotationsofthisworkarefromtheClarendonEdition(Oxford:ClarendonPress,2007)orthe11thandsub-sequentimpressionsoftheOxfordPhilosophicalTexts(OPT)Edition(Oxford:OxfordUniversityPress,2006).Althoughthesetwoeditionsprovidesignificantlydif-ferenteditorialmaterials,theirtextsoftheTreatiseareidentical.IWDavidHume:ThePhilosophicalWorks,ed.T.H.GreenandT.H.Grose(London:Longman,1882–6;reprintedDarmstadt:ScientiaVerlagAalen,1964).1AcompletesetofthecorrectionsandrevisionsmadetotheOxfordPhilosophi-calTexts(OPT)EditionofHume’sTreatiseandAbstractmaybeseenathttp://digital.library.mcgill.ca/hume/Corrections.pdf.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\ndavidfatenorton1AnIntroductiontoHumesThoughtMuchofwhatDavidHumesaidaboutawiderangeofsubjectsremainsofgreatimportancetoday.Inthefirstvolumeofhisfirstwork,ATreatiseofHumanNature,aworkinwhichhearticulatedanewscienceofhumannature,Humefocusedonaninterrelatedsetofissuesintheoryofknowledge,metaphysics,andphilosophicalpsychology.Moreparticularly,heexplainedhowitisthatweformsuchimportantconceptionsasspaceandtime,causeandeffect,externalobjects,andpersonalidentity.Atthesametime,heofferedanequallyimportantaccountofhoworwhywebelieveintheobjectsoftheseconceptionsanaccountofwhywebelievethatcausesarenecessarilyconnectedtoeffects,thatthereareenduringexternalobjects,andthatthereareenduringselveseventhoughthehumanmindisunabletoprovideasatisfactoryproofthatthesephenom-enaexist.InthesecondvolumeoftheTreatiseHumeexpandedhisaccountofhumanpsychology,focusingontheoriginandroleofthepassionsandthenatureofhumanfreedom.Inthethirdandfinalvolumeofthisworkheexploredtheoriginsandnatureofmorality.Inlaterworkshereturnedtomanyofthesephilosophicalissues,buthealsomadesubstantialcontributionstoourunderstandingofpoliticaltheory,aesthetics,economics,andphilosophyofreligion.Inaddition,hewroteaninfluential,six-volumeHistoryofEngland,aworkpublishedinover175editionsintheeighteenthandnine-teenthcenturies,andstillinprint.i.lifeandwritingsHumewasborninEdinburgh,Scotlandscapital,on26April1711.TheyearsofhisyouthweredividedbetweenthatcityandNinewells,1CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n2davidfatenortonhisfamilyssmalllandholdingatChirnside,avillageneartheborderwithEngland.LittleisknownaboutHumeschildhood.HisfatherdiedwhenDavidwastwo;hismotherthereafterdevotedherselftoherthreechildren.ItislikelythatHumebeganstudiesattheCol-legeofEdinburghin1721(whenabouttwoyearsyoungerthanthetypicalenteringstudent)andcontinuedtherethroughthespringof1725,whenhewouldhaveturnedfourteen.1Afterleavinguniver-sityheapparentlymadeadesultoryeffortatlearninglaw,butsoon1WelackadetailedaccountofHumesearlyreadingandeducation,buttheout-linesofhisfouryearsattheCollegeofEdinburghareknown.HumewouldhavestudiedLatinduringhisfirstyear,andfollowedthiswithayearstudyingGreek.Hewouldhavefollowedinhisthirdyearacourseinlogicandmetaphysics,andinhisfourthandfinalyearacourseinnaturalphilosophyorganizedaroundthewritingsofRobertBoyle.Theplansoriginallydrawnupforthiscoursein1708includedprovisionforsomeinstructioninethics,butthereisnofirmevidencethatethicswereincludedinthe17245sessionHumewouldhaveattended.Inaddi-tion,inDecember1724Humejoinedaprivatelibrary(thePhysiologicalLibrary)thatgavehimaccesstoawiderangeofbooksonthesciencesthenstudied.HumewaslatertoreportthatinthethreeyearsendingaboutMarch1734hehadreadmostofthecelebratedBooksinLatin,French&English,andalsolearnedItalian(KHL6).Anylistofthosehavingasignificant(althoughnotnecessarilypositive)impactonhisearlythoughtwouldlikelyincludenotonlythosewritersoftenmen-tioned(JohnLocke,GeorgeBerkeley,IsaacNewton,andFrancisHutcheson,forexample),butalsoagreatmanyothers,includingsuchrelativelywell-knownfiguresasVirgil,Cicero,Plutarch,andSenecafromamongtheancients,andMichelMon-taigne,FrancisBacon,HugoGrotius,ReneDescartes,PierreGassendi,BlaisePas-´cal,SamuelPufendorf,RobertHooke,NicolasMalebranche,PierreBayle,AnthonyCollins,thethirdEarlofShaftesbury,SamuelClarke,BernardMandeville,JosephButler,BaronMontesquieu,andLordBolingbroke,aswellasmanyotherindividu-alsnowlesswellknown,fromtheearlymodernperiod.Consequently,despitehisobviouspreferenceforwhathecalledtheexperimentalmethodofreasoning,nosinglewriterorphilosophicaltraditioncanbereliedontoprovideacomprehen-sivekeytoHumesthought.TolearnmoreaboutHumesearlyeducation,seeM.A.Stewart,HumesIntel-lectualDevelopment,171152,inImpressionsofHume,ed.M.Frasca-SpadaandP.J.E.Kail(Oxford:ClarendonPress,2005),1158,esp.1125;andMichaelBarfoot,HumeandtheCultureofScienceintheEarlyEighteenthCentury,inOxfordStudiesintheHistoryofPhilosophy,ed.M.A.Stewart,3vols.(Oxford:ClarendonPress,19902000)[hereafterOxfordStudies],1:15190,esp.1516.ThelatteralsoprovidesmuchdetailaboutthePhysiologicalLibrary.ForamoregeneralaccountofthecurriculumattheUniversityofEdinburgh,andoftheviewsofthosewhotaughtit,seeR.L.Emerson,ScienceandPhilosophyintheScottishEnlighten-ment,alsoinOxfordStudies1:1136.Emersonarguesthatbyasearlyas1710theexperimentalismofBacon,Boyle,andNewtonwaswellknowninEdinburgh.ForextensivesuggestionsregardingthoseofhispredecessorsHumemayhaveread,seeEditorsAnnotations,inATreatiseofHumanNature,ed.D.F.NortonandM.J.Norton,2vols.(Oxford:ClarendonPress,2007)[hereafterNortonandNorton],2:685978.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought3enoughwasdevotinghisprincipaleffortstophilosophy,andespe-ciallytotheissuesthatbecamecentraltohisphilosophicalclassic,ATreatiseofHumanNature.In1734,discouragedbyhisinabilitytopresenthisviewsinsatisfactoryform,hetriedthemoreactivelifeofamerchantsassistant.Withinmonthsheabandonedthisexperi-mentandtraveledtoFrance,whereheremainedforthreeyearsandatlastfinishedadraftofhislongwork.InSeptember1737hesettledinLondonandcontinuedtorevisetheTreatise.InJanuary1739thefirsttwovolumesoftheworkwerepublishedandHumereturnedtoScotland,whereherevisedthemanuscriptofthethirdandfinalvolumeoftheTreatise.ThisvolumewaspublishedinlateOcto-ber1740.Twovolumesofhisessays(Essays,MoralandPolitical)appearedin17412.TohelpsupporthimselfduringthenextfifteenyearsHumetookpositionsfirstascompaniontoamentallyunbalancednobleman,thenasaide-de-campandlatersecretarytoaBritishgeneral,andfinallyasKeeperoftheAdvocatesLibraryinEdinburgh.AlthoughthemanyworksHumepublishedfrom1748to1762madehimfinan-ciallyindependent,heacceptedtwofurtherpublicserviceappoint-ments:from17635hewasatfirstSecretary,thenchargedaffaires´totheBritishEmbassyinParis,andin17678hewasUndersecre-taryofState(NorthernAffairs)intheBritishgovernment.HethenretiredtoEdinburghwhereheliveduntilhisdeathin1776.InhiswillHumeleftinstructionsforthepublicationofhisDialoguescon-cerningNaturalReligion.ii.experienceanditslimitsHumesmostoftencitedworksincludethethreevolumesoftheTreatiseofHumanNaturementionedabove;anAbstractofvol-umes1and2oftheTreatise(1740);acollectionofapproximately50essays,EssaysMoral,Political,andLiteraryandPoliticalDis-courses(mostofwhichwerefirstpublishedfrom1741to1752);AnEnquiryconcerningHumanUnderstanding(1748);2AnEnquiry2ThisworkwasfirstpublishedasPhilosophicalEssaysconcerningHumanUnder-standing;itwasretitledbyHumein1758.Fromthatdateandonthroughthenineteenthcentury,HumesessaysandAnEnquiryconcerningHumanUnder-standing,OfthePassions(retitledasADissertationonthePassions),AnEnquiryconcerningthePrinciplesofMorals,andTheNaturalHistoryofReligionwerepublishedtogetherasEssaysandTreatisesonSeveralSubjects.ForahistoryofCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n4davidfatenortonconcerningthePrinciplesofMorals(1751);OfthePassionsandTheNaturalHistoryofReligion(1757);hissix-volumeHistoryofEng-landfromRomantimesto1688(175462);abriefautobiography,MyOwnLife(1777);andDialoguesconcerningNaturalReligion(1778).3Theseworksspananexceptionallywiderangeoftopicsandthusareinsomewayssignificantlydifferentfromoneanother.Theyarenonethelessunifiedbyatleastonefundamentalcharacteristic:theirauthorscommitmenttotheexperimentalmethod,ortoaformofphilosophythatrecognizesboththeadvantagesandnecessityofrelyingonexperienceandobservationtoprovidetheanswerstointellectualquestionsofallkinds.4ThesubtitleofHumesTreatisedescribesitasanattempttointroducetheexperimentalmethodofreasoningintomoralsubjects.5IntheIntroductiontothisworkHumetracesthebegin-ningoftheuseoftheexperimentalmethodinnaturalphilosophytoFrancisBacon(15611626).6Moralphilosophy,Humeargues,andespeciallythefoundationalscienceofhumannaturethatheproposestodevelop,mustalsomakeuseofthismethod:Andasthescienceofmanistheonlysolidfoundationfortheothersciences,sotheonlysolidfoundationwecangivetothisscienceitselfmustbelaidonexperienceandobservation.7Apagelaterheinsiststhat,whilewemusttrytorenderallourprinciplesasuniversalaspossible,bytracingupourexperimentstotheutmost,andexplainingalleffectsfromthesimplestandfewestcauses,tisstillcertainwecannotgobeyondexperience;andanyhypothesis,thatpretendstodiscoverthecollectededitionsofHumesworkspublishedfrom1753to1777,seeTomL.Beauchamp,Introduction,inAnEnquiryconcerningHumanUnderstanding,2ndimpression,corrected(Oxford:ClarendonPress,2006),xxvxxxv.3ForachronologicallistofHumeswritings,seePartIoftheSelectedBibliography.4ForfurtherreadingshowingthewidelydifferentwaysinwhichHumehasbeeninterpreted,seePartsIIIandIVoftheSelectedBibliography.5ThediscussionthatfollowsinPartsIIVIofthisessayfocusesonHumesfirstandmostcomprehensivework,theTreatiseofHumanNature.6InHumestimephilosophyhadtwodistinctivebranches.One,naturalphiloso-phy,includedthosesubjectswenowthinkofasthephysicalandnaturalsciences.Theother,moralphilosophy,focusedonhumansorhumanactivityandincludedthosesubjectswewouldthinkofasthecoreofphilosophy(theoryofknowledge,metaphysics,ethics,andthephilosophyofreligion),aswellassuchsubjectsaspsy-chology,politicalscience,sociology,economics,andaesthetics(touseourterms).7Formoreonthistopic,seeinthisvolumetheessayHumesNewScienceoftheMind.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought5theultimateoriginalqualitiesofhumannature,oughtatfirsttoberejectedaspresumptuousandchimerical(TIntro.78).Findingthatmoralphilosophycouldnot,asnaturalphilosophycan,makeitsexperimentspurposely,withpremeditation,andaftersuchaman-nerastosatisfyitselfconcerningeveryparticulardifficultywhichmayarise,hetellsusthatWemustthereforegleanupourexperimentsinthissciencefromacautiousobservationofhumanlife,andtakethemastheyappearinthecommoncourseoftheworld,bymensbehaviourincompany,inaffairs,andintheirpleasures.Whereexperimentsofthiskindarejudiciouslycollectedandcompard,wemayhopetoestablishonthemascience,whichwillnotbeinferiorincertainty,andwillbemuchsuperiorinutilitytoanyotherofhumancomprehension.(TIntro.10)InhisAbstractoftheTreatiseHumedescribeshimselfashav-ingpromisedtodrawnoconclusionsbutwhereheisauthorizedbyexperience(A2).HeconcludesAnEnquiryconcerningHumanUnderstandingwiththenownotoriousinjunctiontocommittotheflamesanybookofdivinityorschoolmetaphysics,forinstance,thatcontainsneitheranyabstractreasoningconcerningquantityornumbernoranyexperimentalreasoningconcerningmatteroffactandexistence(EHU12.34),butnotbeforehehassubjectedexperimentalreasoningitselftoasevere,experimentalscrutiny(seeEHU4.1423).InOftheOriginalContract,anessayinpoliticaltheoryfirstpublishedin1748,HumetellsusthatAsmalldegreeofexperienceandobservationsufficestoteachus,thatsocietycannotpossiblybemaintainedwithouttheauthorityofmagistrates,andthat,moreover,theobservationofthesegeneralandobviousinter-ests[peaceandpublicorder]isthesourceofallallegiance,andofthatmoralobligation,whichweattributetoit(E-OC25,480).AnEnquiryconcerningthePrinciplesofMoralsundertakestodiscoverthefoundationofethics.Asthis,Humesays,isaquestionoffact,notofabstractscience,wecanonlyexpectsuccess,byfollow-ingtheexperimentalmethod,anddeducinggeneralmaximsfromacomparisonofparticularinstances(EPM1.10).InOftheStandardofTaste,firstpublishedin1756,hetellsusthatitisobviousthattherulesofcompositionarenothingmorethangeneralobserva-tions,concerningwhathasbeenuniversallyfoundtopleaseinallcountriesandinallages,andthatinthisregardtheirfoundationCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n6davidfatenortonisthesamewiththatofallthepracticalsciences,experience(E-ST9,231).8Toappreciatefullytheforceoftheseremarks,wemustkeepinmindthattheycarrywiththemanunspokenbutdeepdistrustoftheapriorireasoningcharacteristicofmuchearlierphilosophy,especiallythatoftheCartesians.Atothertimes,however,Humeisexplicitaboutthelimitationsofourfacultyofreasonandtheshortcomingsofthosephilosophicalsystemsthatgiveitpriority.Hepointedlynotesthat,althoughweallbelievethateveryeventorobjecthasacause,therearenovalidargumentsestablishingthisconclusion(T1.3.3).And,althoughweallbelieveinenduring,exter-nalobjects,reasoncannotestablishthatsuchobjectsexist,andevenifitcould,itwouldbeofnousetothatvastpopulationofpeopleandanimalswho,withouttheuseofasingleargument,believethatsuchobjectsexist.Althoughreasonmayhelpusdeterminehowtoachievesomedesiredgoal,ithasbyitselfabsolutelynomotivat-ingforce.Althoughweallmakemoraldistinctionswetakesomeactsorpersonstobevirtuousorgood,otherstobeviciousormorallywrongitisaspecialkindoffeeling,notreason,thatmakesthispos-sible.Itisthisdistrustofreasoning,coupledwithhiscommitmentto8Humewaslessexplicitabouthiscommitmenttoexperienceandobservationinhisprimarilyhistoricalworks,theNaturalHistoryofReligionandtheHistoryofEngland.Buttheformerworkattemptstodiscovertheoriginofreligioninhumannaturebyextrapolatingfrompresentfacts(religionandhumannatureastheyarepresentlyfoundtobe)andthehistoricalrecordofthebeginningsanddevelopmentofreligion.Thisexerciseisanaturalhistorybecausetheexplanationisconstrainedwithinthelimitsofobservable,naturalphenomena;nosupernaturalbeingsorprin-ciplesareappealedtoorpresupposed.FormoreonthisworkandHumesapproachtoreligion,seeinthisvolumetheessayHumeonReligion.AnalogouscommentcanbemaderegardingTheHistoryofEngland.MotivatedtoaconsiderabledegreebytheexaggeratedclaimsofthetwoleadingpoliticalgroupsinBritain,theWhigsandtheTories,eachofwhominsistedthatthepoliticalinsti-tutionsofeighteenth-centuryBritainreflected,orshouldreflect,aperfectmodelfoundeitherinthemistsoftheirAnglo-Saxonbeginnings(aWhigtendency)orinatimeless,sacredbeginning(aTorytendency),HumeattemptedtowriteanimpartialhistoryofEngland,ahistorythatrecordedthedevelopmentofpoliticalinstitutionsovertime,onethattreatedtheseinstitutionsasthehard-wonandstilldevelopingproductsofcenturiesofexperienceandobservation,andnotassomethingderivedfromaprioriprinciplesingrainedinthehumanmind.Formoreontheseissues,seeinthisvolumetheessaysTheStructureofHumesPoliticalTheoryandDavidHume:TheHistorian.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought7experienceandobservation,thatmakesitnotentirelyinappropriatetothinkofHumeasanearlyempiricist.9Whatisoftenmissed,however,isthefactthatHumescommitmenttoexperienceandobservationisqualifiedinatleastfivesubstantialways.1.Aswehaveseen,Humesupposedthatmoralphilosophycannotmakeitsexperimentspurposely,[and]withpremeditationinthewaythatnaturalphilosophycanbecausesuchartificialorlaboratory-likeexperimentswoulddisturbanddistortthephenomenabeingexamined.Butwealsosawthathewasnotdiscouragedbythislimitationbecausemoralphilosophycancollectitsexperimentaldatafromacarefulexaminationofhumanlife,andhavingdoneso,canhopetoconstructausefulscienceofhumannature.2.Experiencehasintrinsiclimitations.Inourquesttounder-standhumannature,forexample,wemayfollowexperienceasfarasitwilltakeus,butwewillstillremainignorantofthemostfundamentalorultimatefeaturesofournature.AsHumeputsitatthebeginningoftheTreatise,wemaytrytomakeourconclu-sionsasgeneralorasuniversalaspossiblebyexplainingalleffectsfromthesimplestandfewestcauses,butbecauseweknowthatexperiencehaslimitations,wemustrememberthatanytheorythatclaimstohavediscoveredtheultimateoriginalqualitiesofhumannature,oughtatfirsttoberejectedaspresumptuousandchimerical(TIntro.8).3.Hume,asmuchasanyoftheCartesians,insiststhatallsensoryexperienceisindirect.Wedonotexperienceobjectsthemselves.Weexperienceonly,inthelanguageofDescartesandLocke,ideasorwhatsomesupposetobementalrepresentationsofobjects.Hume9HumecontinuedhisdiscussionatEPM1.10(quotedinthepreviousparagraph)bycontrastinghisexperimentalapproachtomoralphilosophywithwhathecalledtheotherscientificalmethod,thatinwhichageneralabstractprincipleisfirstestablished,andisafterwardsbranchedoutintoavarietyofinferencesandconclu-sions.Formsofthislattermethodcameintimetobecalledrationalism,whileformsoftheexperimentalmethodtowhichHumeadheredcame(butnotduringHumeslifetime)tobecalledempiricism.Retrospectively,then,andwhilerecogniz-ingthattherearesignificantlydifferentkindsofempiricism,itisnotamistaketocallHumeanempiricist.Forabriefdiscussionofkindsofempiricism,seeNicholasP.Wolterstorff,Empiricism,inTheCambridgeDictionaryofPhilosophy,ed.R.Audi,2nded.(Cambridge:CambridgeUniversityPress,1999).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n8davidfatenortonusesdifferentterminology,butisfirmlycommittedtotheviewthatourdirectexperienceislimitedtomentalphenomena.EarlyintheTreatise,whilediscussingtheideaofexistence,andofexternalexistence,hereportsthatitisuniversallyallowdbyphiloso-phers,andisbesidesprettyobviousofitself,thatnothingiseverreallypresentwiththemindbutitsperceptionsorimpressionsandideas,andthatexternalobjectsbecomeknowntousonlybythoseperceptionstheyoccasion(T1.2.6.7).HumerepeatsthisclaimineachbookoftheTreatise,intheAbstract,andintheEnquirycon-cerningHumanUnderstanding.104.Giventhefactthatweexperienceonlyperceptions,andfur-therfactsaboutthenatureofourperceptions,Humeconcludesthatourdeepandvirtuallyineradicablebeliefintheexistenceofexternalobjectsisnotduetosenseexperiencealone.Thesensesmayplayanessentialroleintheprocessthatbringsaboutthisfundamentalbelief,butthesensesoperatingalonewouldbeunabletoproduceit.Moreover,thesensesoperatingalonewouldbeunabletoaccountforourbeliefincausalconnectionsorpersonalidentity.5.Experienceisnotthesourceofcertainofourfundamentalpas-sions.Lockehadarguedthattherearenoinnateideas,andhadmadethisconclusiononeofthedefiningfeaturesofhisformofempiri-cism.Locke,accordingtoHume,wasbothconfusedandmistaken.Lockeusedthetermideatoobroadlyandthusfailedtodistinguish,asheoughttohavedone,betweentwokindsofperceptions,impres-sions(especiallyimpressionsofsensation)andtheideasthatderivefromthem.Ifwemakethisneededdistinctionweseethat,whileitmaybetruetosaythattherearenoinnateideas,itisfalsetosaythattherearenoinnateimpressions.AsHumeputsit,itisclearthatsomeofourstrongerperceptionsorimpressionsareinnate,andthatnaturalaffection,loveofvirtue,resentment,andalltheotherpassions,ariseimmediatelyfromnature(A6;seealsoT1.1.1.12,EHU2.9n.1).Aswewillsee,theseperceivedlimitationsofexperiencepro-foundlyinfluencedHumesconclusions.10SeealsoT1.4.2.21and47,2.2.2.22,3.1.1.2;A5;EHU12.9.FormoreonHumesskepticalchallengetoexperimentalreasoning,seeinthisvolumePartIIoftheessayHumesSkepticism.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought9iii.theelementsofhume’sphilosophyFormostofthenearly270yearssincethepublicationofhisTreatise,Humewasroutinelyinterpretedasthephilosopherwhoadvancedhisformofphilosophy(whichhascometobecalledempiricism)toitslogicalandskepticalconclusion.IsuggestthatHumeisbetterunderstoodasapostskepticalphilosopher.BythisImeantosug-gestthatHumesupposed(a)thatearlierphilosophers,andespeciallyNicolasMalebranche,PierreBayle,JohnLocke,andGeorgeBerke-ley,hadalreadytakentraditionalmetaphysicsandepistemologytoitsskepticalconclusions;(b)thattheseskepticalconclusionshadbeensoundlyandvalidlyestablished;and(c)thatthemostimpor-tantremainingtaskofphilosophy,giventhesewell-establishedandobviousconclusions,wastoshowhowwemanagetogetonwithourlives,particularlyourintellectuallives,withouttheknowledgeofultimatecausesandprinciplessoughtbyhispredecessors.Toputthisanotherway,Inotethat,priortoHume,oneoranotherphiloso-pherhad,perhapsunintentionally,thoroughlydiscreditedtheclaimofhumanstohaverationallyorexperientiallyderivedknowledgeoftheexistenceandtruenatureofspace,causalrelations,externalobjects,andmind.ButasHumeputit,eventherabbleoutsidethephilosophicalhalleventhosewhoarenotphilosopherscouldseethatthephilosophicalenterprisewasnotgoingwell.Themosttriv-ialquestionescapesnotourcontroversy,andinthemostmomentouswearenotabletogiveanycertaindecision(TIntro.2).Itistime,surely,tostartanew,toprovidemoralphilosophywithanewfoun-dation,thescienceofhumannature,onwhichalltheotherscienceswillbefounded.ButnoticewhereHumebegins:Theelementsofthisphiloso-phyare,inthemostliteralsense,theimmediateobjectsofthoughtaswellastherelationsbetweenoramongtheseobjectsofthementalworld.Theelementsthemselvesarecalledperceptions,andaredividedintotwokinds,impressionsandideas.Ofthese,impressionsarethemoreforcefulorlively,whileideasarecomple-mentaryinthattheyaresaidtobethefaintimagesofimpressions.Inaddition,Humeclassifiesasimpressionsalloursensations,pas-sionsandemotions,astheymaketheirfirstappearanceinthesoulormind,andthendividesthisclassintotwosubclasses,impres-sionsofsensationandimpressionsofreflection.Thelattersort,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n10davidfatenortonimpressionsofreflection,arederivdinagreatmeasurefromourideas.Ontheotherhand,impressionsofsensation,hesays,ariseinthesouloriginally,fromunknowncauses(emphasisadded).Hethenaddsthattheexaminationofoursensations[ourimpressionsofsensation]belongsmoretoanatomistsandnaturalphilosophersthantomoral;andthereforeshallnotatpresentbeenterdupon(T1.1.4.3,67;1.1.1.1.;1.1.2.1).AsweworkthroughtheTreatisewecometorealizethatthephrasenotatpresentinfactmeansnotinthiswork,foratnotimedoesHumetakeupthetaskthathehasassignedtoanatomistsandnaturalphilosophers.11Indeed,hebeginsBook2oftheTreatisewithmuchthesamedisclaimer:Tiscertain,thatthemind,initsperceptions,mustbeginsomewhere;andthatsincetheimpressionsprecedetheircorrespondentideas,theremustbesomeimpressions,whichwithoutanyintroductionmaketheirappearanceinthesoul.Asthesedependuponnaturalandphysicalcauses,theexam-inationofthemwoudleadmetoofarfrommypresentsubject,intothesciencesofanatomyandnaturalphilosophy.(T2.1.1.2)BetweenthesetworemarksHumetellsusclearlywhyhehaslefttoothersthetaskofexplainingimpressionsofsensation:suchanexplanationisirrelevanttothephilosophicalenterpriseinwhichheisengaged.Asheputsit:Astothoseimpressions,whicharisefromthesenses,theirultimatecauseis,inmyopinion,perfectlyinexplicablebyhumanreason,andtwillalwaysbeimpossibletodecidewithcertainty,whethertheyariseimmediatelyfromtheobject,orareproducdbythecreativepowerofthemind,orarederivdfromtheauthorofourbeing.Norissuchaquestionanywaymaterialtoourpresentpurpose.Wemaydrawinferencesfromthecoherenceofourperceptions,whethertheybetrueorfalse;whethertheyrepresentnaturejustly,orbemereillusionsofthesenses.(T1.3.5.2)1211AlthoughHumewantednothingtodowithaphysicalanatomyattemptingtoexplainsensation,hedoesrepeatedlydescribehimselfasengagedinananatomyofhumannature.SeeT1.4.6.23,3.3.6.6,A2;HL1:323.12ThiscommentismadeinthemidstofHumesattempttoexplainhowwecometohavetheideaof,andtobelievein,necessaryconnection.ButthesuggestionthattheexplanationsofBook1areconfinedtoanexaminationofthecoherenceofelementswithinthementalworldisrepeatedinotherformsinotherplaces.See,forexample,1.4.2(Ofscepticismwithregardtothesenses),wherethediscussionisfocusedonthewayinwhichimpressionsandideascoheretogiveus,notknowledgeof,butonlybeliefin,externalobjects;andtheAppendix(publishedCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought11Butnotice,Isayagain,whereHumebegins:Theelementsofthisphilosophyare,inthemostliteralsense,theimmediateobjectsofthoughtaswellastherelationsbetweenoramongtheseobjectsinthementalworld.Andhisconcernisnottoadvancefromthisbaseinordertodenydogmaticallythattherearecauses,objects,orminds.Hisconcernisnottomakethecaseforskepticismabouttheseentities.Thecaseforskepticismaboutthesemomentousques-tionswaswellknowntoHume.HeknewthosesectionsofBayleandLockethatrevealtheinadequacyofDescartessattemptstoprovebyreasonthatthereisanexternalworld.HeappreciatedtheskepticalforceoftheobjectionsbroughtbyBayle,thensignificantlyampli-fiedbyBerkeley,againsttheprimary-secondaryqualitydistinctionchampionedbyLocke.13Hesawthatphilosophersofallkindswere,inthematterofexplainingtheinteractionofmindandbody,skep-ticsinspiteofthemselves.HesawthattheleadingCartesianoftheday,Malebranche,hadconcludedthattherearenonaturalcausesofanykind,andthatthereisnohumanornaturalknowledgeoftheexistenceofcausesorobjects.Whatwedoknowofthesethingsis,accordingtoMalebranche,theresultofanactofdivinegrace.14in1740asapartofVolume3),whereHumecontraststheoriesofthematerialworldwithhistheoryoftheintellectualworld(App.10).ForafullerdiscussionofHumesaccountofimpressionsandideas,seeinthisvolumetheessayHumeandtheMechanicsofMind.13Lockehadarguedthatcertainideas(thoseofextensionandshape,forexample)causedbywhathecalledtheprimaryqualitiesofobjectsresemblethesequali-tiesinsuchawaythattheyprovideuswithaccurate,reliableinformationaboutthequalitiesthatcausethem.Otherideas(thoseofcolorandtaste,forexample)causedbywhathecalledthesecondaryqualitiesofobjectsfailtoresemblethequalitiescausingthem,andinfactleadustoattributetoobjectscharacteristics(color,taste)whichtheydonotpossess.Baylesuggested,andBerkeleyargued(suc-cessfully,itisgenerallybelieved),thatthisdistinctionisuntenable.SeePierreBayle,HistoricalandCriticalDictionary:Selections,ed.andtrans.R.H.Popkin(Indianapolis:LibraryofLiberalArts,1965),Pyrrho,noteB;GeorgeBerkeley,ATreatiseconcerningthePrinciplesofHumanKnowledge,1.915.ForahelpfulaccountofBerkeleysimpactonHume,seeDavidRaynor,HumeandBerkeleysThreeDialogues,OxfordStudies1:23150.14Humesays,forexample:Butsolittledoesanypowerdiscoveritselftothesensesintheoperationsofmatter,thattheCartesianshavemadenoscrupletoassert,thatmatterisutterlydeprivedofenergy,andthatallitsoperationsareperformedmerelybytheenergyofthesupremeBeing(A26).FormoreonHumesviewsoncausation,seeinthisvolumetheessayHumeonCausation.Forotherswhoarguedthatnocausalconectioniseversensed,seeNortonandNorton,EditorsAnnotations,2:735,annotation55.18.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n12davidfatenortonInshort,Humewassatisfiedthatthebattletoestablishreliablelinksbetweenthoughtandrealityhadbeenfoughtandlost,andhencemadehiscontributionstophilosophyfromapostskepticalperspectivethatincorporatesandbuildsontheskepticalresultsofhispredecessors.15iv.askeptic’sideasandbeliefsTheonce-standardreadingofHumecreditedhimwithseeingtheskepticalimplicationsoftherepresentativetheoryofperception,andwithseizingontheseimplicationsinthecauseofadestruc-tiveskepticism.16ItseemslikelythatHumewasfullyawareoftheskepticalimplicationsofthistheory,but,givenhisexpresseddisin-terestintheconnectionsbetweenimpressionsofsensationandtheirpossiblecauses,wemustconcludeeitherthathedidnotadoptthetheory,orthatheadoptedonlyonepartofit.Humeagreesthattheimmediateobjectsofmindarealwaysperceptions,buthedoesnottakethesetobe,inonecardinalsense,representativeofobjects.Heneverclaimsthatimpressionsorideasresembleobjects.1715ItdoesnotfollowthatHumemadenocontributionstothearsenalofskepticism.Foradiscussionofsomeofthesecontributions,seeinthisvolumetheessaysHumesSkepticismandHumeonReligion.16Thistheorymaintainsthattheimmediateobjectsofthemindareideas(inHumesvocabulary,perceptions,orimpressionsandideas),someofwhicharesupposedtorepresentaccuratelyvariouskindsofentitiesoutsidethemind;seenote13.Theproblemwastodeterminewhichideasdorepresenttheexternalexistences,and,giventhatideasandonlyideasaretheimmediateobjectsofthemind,tofindindependentevidencethatanygivenidearepresentsaccuratelyorevenresemblestheobject(s)supposedtobeitscauseandthatitissupposedtorepresent.ThatHumeexposedtheskepticalimplicationsofthistheorywassuggestedasearlyas1751byhiskinsmanHenryHome,LordKames,andthenrepeatedbysuchothereighteenth-centuryphilosophersasThomasReid.FortherelevantworksbytheseearlycriticsofHume,seetheSelectedBibliography,PartIVA.17Humerepeatedlyinsiststhatideasarederivedfromandrepresentonlyimpres-sions.Impressionsthemselvesareoftwotypes:impressionsofsensationandofreflection.Oursenses,hesays,cannotrepresenttheirimpressionsasdistinctfromus,andhencefailtorepresentacrucialfeatureofexternalobjects.Nor,hesays,cananyofoursenseimpressions,notevenourimpressionsoftouch,representsolidity,noranyrealobject,becausethereisnottheleastresemblancebetweentheseimpressionsandsolidity(T1.4.2.810,1.4.4.13).Apassion,Humesays,con-tainsnotanyrepresentativequality,whichrendersitacopyofanyotherexistenceormodification(T2.3.3.5).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought13Infact,Humegavethewayofideasakindofphenomenolog-icalturn.Thatis,hisprimaryconcerninBook1oftheTreatiseiswithourperceptions,quaperceptions,withperceptionsas,sim-ply,theelementsorobjectsofthemind,andnotasrepresentationsofexternalexistences.Havingfocusedonperceptionsastheonlyobjectsofthemind,HumegoesoninBook1toshowhowsomeoftheseperceptionsareinterrelatedorassociatedtoproducestillfurtherperceptionsthatarethenprojectedontoaworldputativelyoutsidethemind.18Somehowthemindisfurnishedwithimpres-sionsofsensation.Onexaminationwefindthatnotoneoftheseimpressionscanofitselfbetakenasanaccuraterepresentationofspaceortime,causalconnection,anexternalobject,orevenourownmind.Wesimplydonothavesensoryimpressionsoftheseentities.But,notwithstandingthisfact,andthefurtherfactthatallourideasarederivedfromimpressions,wenonethelessdohaveideasofspace,causalconnection,andexternalexistences,etc.,andwearenonethelessirredeemablycommittedtobelievingthattherearerealentitiesthatcorrespondtotheseideas.19Themysterytobeexplained,giventhesuccessofskepticism,ishowwecometohavetheseimportantideasand,moreover,howwecometobelievethattheyrepresent,notmerelyimpressions,butindependent,externalexistencesorrealities.Toputthisdifferently,Humesgreatergoalcanbetakentobeshowinghow,despitethesuccessesofskepticism,wearerescuedfromskepticism.ThefirstbookoftheTreatiseisanefforttoshowhowourper-ceptionscoheretoformideasofthosefundamentalitems(space,causalconnection,externalexistence)inwhich,skepticaldoubts18Husserl,himselfthefounderofphenomenology,recognizedthisfeatureofHumesthought.SeeR.A.Mall,ExperienceandReason:ThePhenomenologyofHusserlandItsRelationtoHume’sPhilosophy(TheHague:Nijhoff,1973),1928.19Weremainirredeemablycommittedtothesebeliefsinthesensethat,whilephilo-sophicalanalysismayonoccasionbringustodoubtthem(see,e.g.,1.4.2.567),thisdoubtcannotbesustained.Evenaskepticmust,withrareexception,believeincausesandobjects.Theskepticmayverywell,however,modifythemannerorintensityoftheseunavoidablebeliefs.Onthislatterpoint,seemyHowaScepticMayLiveScepticism,inFaith,Scepticism,&PersonalIdentity,ed.J.J.MacIn-toshandH.A.Meynell(Calgary:UniversityofCalgaryPress,1993),11939,ortheshorterversionofthispaper,OftheAcademicalorScepticalPhilosophy,inReadingHumeonHumanUnderstanding:EssaysontheFirstEnquiry,ed.P.Millican(Oxford:ClarendonPress,2002),37192.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n14davidfatenortonnotwithstanding,wereposebeliefandonwhich,asHumeputsitintheAbstract,lifeandactionentirelydepend.InBook1,Part2,Humearguesthatwehavenodirectimpressionsofspaceandtime,andyetwehavetheideasofboth.20Heaccountsforourideaofspacebyappealingtoamannerofappearance,inthefollowingway.Bymeansoftwosenses,sightandtouch,wehaveimpressionsthatarraythemselvesassomanypointsrelatedtooneanother.Theseparticularimpressionsarebytheimaginationtransformedintoacompoundimpression,whichrepresentsextensionortheabstractideaofspaceitself.Ourideaoftimeis,mutatismutandis,accountedforinthesameway:Astisfromthedispositionofvis-ibleandtangibleobjectswereceivetheideaofspace,sofromthesuccessionofideasandimpressionsweformtheideaoftime.21Theabstractideaoftime,likeallotherabstractideas,isrepresentedintheimaginationbyaparticularindividualideaofadeterminatequantityandqualityjoinedtoaterm,time,thathasgeneralref-erence(A4,T1.2.3.57,15).Inshort,theimagination,afacultynottypicallyassignedsosignificantarole,achieveswhatneitherthesensesnorreasoncanachieve.Humesaccountofourderivationofandbeliefintheideaofcausalconnection(ofnecessaryconnexion,inhisterms)followsthissamepattern.Heisoftensaidtohavedeniedthatthereisphysicalnecessityandthatwehaveanyideaofnecessaryconnection.Thisinterpretationissignificantlymistaken.HumehadbeenconvincedbytheCartesians,andespeciallybyMalebranche,thatneitherthesensesnorreasoncanestablishthatoneobject(acause)isconnectedtogetherwithanotherobject(aneffect)insuchawaythatthepres-enceofthethefirstnecessarilyentailstheexistenceofthesecond.20ItshouldbeunderstoodthatHumeisconcernedwiththesourceofourmostabstractorgeneralideasofspaceandtimeofspace,forexample,assomethinglikecontinuous,unbounded,orunlimitedextensionineverydirection,regardedasvoidofmatter,orwithoutreferencetomatter(OxfordEnglishDictionary).Ofsuchaspaceweneitherhave,norcouldhave,adirectsensoryimpression,butfromthefactthatwecanintelligiblydiscussthesubject,itfollows,Humeargues,thatwehaveanideaofspacetowhichthewordspacerefers:Nowtiscertainwehaveanideaofextension[orspace];forotherwisewhydowetalkandreasonconcerningit?(T1.2.2.9).21ForadiscussionofHumestheoryofspaceandtime,seeinthisvolumetheessayHumesTheoryofSpaceandTime.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought15Humesownanalysisofwhatwetaketobeexperiencesofcauseandeffectshowsthatobjectsthatwesupposetobecausallyrelatedarecontiguousintimeandspace,thatthecauseispriortotheeffect,andthatsimilarobjectshavebeenconstantlyassociatedinthisway,butnothingmore.Thesearetheonlyperceptiblefeaturesofsuchputativecausalconnections.Andyetthereseemstobemoretothematter.Thereis,hesays,anecessaryconnexiontobetakenintoconsideration,andourbeliefinthatrelationmustbeexplained(T1.3.2.11).Despiteourdemonstratedinabilitytoseeorprovethattherearenecessarycausalconnections,wecontinuetothinkandactasifwehadknowledgeofsuchconnections.Weact,forexample,asthoughthefuturewillnecessarilyresemblethepast,andwoudappearridiculousifweweretosaythattisonlyprobablethesunwillriseto-morrow,orthatallmenmustdye(T1.3.11.2).ToexplainthisphenomenonHumeasksustoimaginewhatlifewouldhavebeenlikeforAdam,suddenlybroughttolifeinthemidstoftheworldandinthefullvigourofunderstanding.Adamwouldhavebeenunabletomakeeventhesimplestpredictionsaboutthefuturebehaviorofobjects.Hewouldnothavebeenabletopredictthatonemovingbilliardball,strikingasecond,wouldcausethissecondballtomove(A11).Andyetwe,endowedwiththesamefaculties,cannotonlymake,butareunabletoresistmaking,thisandcountlessothersuchpredictions.WhatisthedifferencebetweenourselvesandthisputativeAdam?Experience.Wehaveexperiencedtheconstantconjunction(aninvariantsuccessionofcertainpairedobjectsorevents)ofparticularcausesandeffects,and,althoughourexperienceneverincludesevenaglimpseofacausalconnectionperse,itdoesarouseinusanexpectationthataparticularevent(acause)willbefollowedbyanotherevent(aneffect)previ-ouslyandconstantlyassociatedwithit.Regularitiesofexperiencegiverisetothesefeelings,andthusdeterminethemindtotransferitsattentionfromapresentimpressiontotheideaofanabsentbutassociatedobject.Theideaofnecessaryconnectioniscopiedfromthesefeelings(T1.3.14.1423).Theideahasitsfoundationinthemindandisprojectedontotheworld,butthereisnonethelesssuchanidea.Thatthereisanobjectivephysicalnecessitytowhichthisideacorrespondsisanuntestablehypothesis,norwoulddemonstrat-ingthatsuchnecessaryconnectionshadheldinthepastguaranteeCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n16davidfatenortonthattheywillholdinthefuture.FromtheseconsiderationsweseethatHumedoesnotexplicitlyanddogmaticallydenythattherearerealcausalconnections.Wehavenoexperienceofsuchnecessaryconnectionsandhencecanbe,atbest,skepticaloragnosticalabouttheirexistence.Thereis,however,anideaofnecessaryconnection,but,althoughweordinarilyandnaturallybelievethatrealitycorre-spondstothisidea,thecorrectphilosophicalanalysisrevealsthattheideaisderivedfromafeeling,oranimpressionofreflection,anditisthisthattheidearepresents.Inotherwords,Humesanalysisleavesthephilosopherroomtosupposethatourbeliefincausalornecessaryconnections,howevernatural,maybemistaken.Humesaccountofourbeliefinfutureeffectsorabsentcausesoftheprocessofmindthatenablesustoplaneffectivelyisapartofthissameexplanation.Suchbeliefinvolvesanideaorconceptionoftheentitybelievedin,butisclearlydifferentfrommereconceptionwithoutbelief.Thisdifferencecannotbeexplainedbysupposingthatsomefurtheridea,anideaofbeliefitself,ispresentwhenwebelieve,butabsentwhenwemerelyconceive.Thereisnosuchidea.Moreover,givenourabilitytojointogetheranytwoconsistentideaswechoosetosojoin,ifanideaofbeliefwereavailablewecould,con-trarytoexperience,byanactofwillcombinetheideaofbeliefwithanyotheridea,andbysodoingcauseourselvestobelieveanything.Consequently,Humeconcludesthatbeliefcanonlybeadifferentmannerofconceivinganobject.Itisalivelier,firmer,morevividandintenseconception.Beliefincertainmattersoffactthebeliefthatbecausesomeeventorobjectisnowbeingexperienced,someothereventorobjectnotyetavailabletoexperiencewillinthefuturebeexperiencedisbroughtaboutbypreviousexperienceofthecon-stantconjunctionoftwoimpressions.Thesetwoimpressionshavebeenassociatedtogetherinsuchawaythattheexperienceofoneofthemautomaticallygivesrisetoanideaoftheother,andhastheeffectoftransferringtheforceorlivelinessoftheimpressiontotheassociatedidea,therebycausingthisideatobebelieved,ortotakeon,ineffect,thelivelycharacterofanimpression(T1.3.7,A21).Ourbeliefsincontinuingandindependentlyexistingobjectsandinourowncontinuingselvesare,onHumesTreatiseaccountofthesematters,beliefsinentitiesentirelybeyondallexperience.WehaveimpressionsthatwenaturallybutmistakenlysupposetobeCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought17themselvescontinuing,externalobjects,butcarefulanalysisofourexperiencerevealsthattheseimpressionsarebytheirverynaturefleetingandobserver-dependent.Moreover,noneofourimpressionsprovidesuswithadistinctivemarkorevidenceofanexternalorigin(T1.4.2.113;seealsoEHU12.514).Similarly,whenwefocusonourownminds,weexperienceonlyasequenceofimpressionsandideas,andneverencounterthemindorselfinwhichtheseperceptionsaresupposedtoinhere.Toourselvesweappeartobemerelyabundleorcollectionofdifferentperceptions,whichsucceedeachotherwithaninconceivablerapidity,andareinaperpetualfluxandmovement(T1.4.6.4).22Howdowe,then,cometobelieveinexternalobjectsorourownselvesandself-identity?Neitherreasonnorthesenses,workingwithimpressionsandideas,provideanythinglikecompellingproofoftheexistenceofcontinuing,externalobjects,orofacontinuing,unifiedself.Indeed,thesetwofacultiescannotsomuchasaccountforourbeliefinobjectsorselves.Ifwehadonlyreasonandthesenses,thefacultieschampionedbypreviousphilosophers,wewouldbemiredinadebilitatinganddestructiveuncertainty.Sounfortunateanoutcomeisavoidedonlybytheoperationofagenerallyunreliablethirdfaculty,theimagination.Bymeansofwhatappeartobeaseriesofoutrightmistakesandtrivialsuggestionsthisfacultyleadsustobelieveinindependentlyexistingobjectsandinourownselves.Theskepticismofthephilosophersisinthiswaybothconfirmed(wecanprovidenoarguments,forexample,provingtheexistenceoftheexternalworld)andshowntobeoflittlepracticalimport.AsHumesummeduphispoint:Almostallreasoningisthere[intheTreatise]reducedtoexperience;andthebelief,whichattendsexperience,isexplainedtobenothingbutapeculiarsentiment,orlivelyconceptionproducedbyhabit.Noristhisall.Whenwebelieveanythingofexternalexistence,orsupposeanobjecttoexistamomentafteritisnolongerperceived,thisbeliefisnothingbutasentimentofthesamekind.Ourauthorinsistsuponseveralotherscepticaltopics;anduponthewholeconcludes,thatweassenttoourfaculties,andemployourreasononlybecausewecannothelpit.PhilosophywouldrenderusentirelyPyrrhonian,werenotnaturetoostrongforit.(A27)22ForadiscussionofHumeonpersonalidentity,seeinthisvolumetheessayHumeandtheProblemofPersonalIdentity.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n18davidfatenortonv.passionsAttheoutsetofBook2oftheTreatiseHumeturnshisattentiontoanotheroftheelementsofhismentalworld,totheimpressionsofreflection,ortothoseimpressionshecharacterizesassecondaryandreflective.HedoesnotinthisorthefollowingBookdevotesignificantspacetotheprofoundmetaphysicalorepistemologicalquestions(theexistenceofcauses,objects,orminds,forexample)thatarethecentralconcernsofBook1.Havingonceexplainedhowweformideasofthesecoreentities,andthenhowwecometobelieveinthem,Humetakesourbeliefinthemforgrantedandpushesontodiscussourimpressionsofreflection.23Theseturnouttobeourpassions,theprincipalsubjectofBook2,andthemoralsentiments,apairofdistinctivepassionsthat,alongwiththevirtuesandvices,aredealtwithinBook3.Ingeneralterms,Book2oftheTreatisecanbesaidtohavehelpedrescuethepassionsfromthenegativeassessmentsandadhocexpla-nationsfoundinmanyhispredecessors.FromthetimeofPlatoandtheStoicsandwellintotheeighteenthcenturythepassionswereroutinelycharacterizedasnotjustirrational,butasalsounnaturalandmorallydangerousmentalelementsthat,ifgiventheirhead,wouldundermineandenslavereason,thenoblydefining,essentialcharacteristicofhumans.Incontrasttothisoft-repeateddenigrationofthepassions,Humesupposedthepassionstobenaturalphenom-enaandundertooktoexplainthembymeansofthesameexperi-mentalorobservationalmethodusedinBook1.Inshort,Book2extendsHumesscienceofhumannaturetothepassions.2423See,e.g.,T2.1.2.2,2.1.11.48,2.2.1.2,2.3.7.1,whereHumesaccountofthepas-sionstouchesonourideasorimpressionsoftheself,andthebriefbuthelpfulsummaryofhisaccountofcausationandthesourceoftheideaofnecessarycon-nectionincludednearthebeginningofhisdiscussionoflibertyandnecessity(seeT2.3.1.4).24NearthebeginningofBook2HumenotesthatsomeexplanationsofthepassionsresembleastronomybeforethetimeofCopernicus.Theydosobecausetheymakeuseofamonstrousheapofprinciplestoexplainwhatisalmostcertainlysimpleandnatural.Whatisneededisanexplanationofthepassionsthatrespectsthemaximthatnaturedoesnothinginvain(T2.1.3.67).Whennearlytwentyyearslaterhepublishedamuchshorteraccountofthepassions,Humeconcludedthatworkbysaying:Ipretendnottohavehereexhaustedthissubject.Itissufficientformypurpose,ifIhavemadeitappear,that,intheproductionandCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought19Humebeginsthistaskbyremindinghisreadersthatperceptionsareeitherimpressionsorideas.Notingthatimpressionsareoftwokinds,originalandsecondary,aswesawabove,heagainrefusestospeculateaboutthesourceoforiginalimpressions,theimpres-sionsofsensation.Hisinterestisinthesecondaryorreflectiveimpressions,thosethatderivefromsomeoriginalimpressioneitherdirectlyorthroughanideaofthatimpression,andthatmakeupthepassions,andotheremotionsresemblingthem.Thesereflectiveimpressionsorpassionsmaybe,Humesug-gests,dividedintotwokinds,thecalmandtheviolent.Theformerincludesthethesenseofbeautyanddeformityofactions,humancreations,andofnaturalobjects,whilethelatterincludessuchpas-sionsasloveandhatred,griefandjoy,prideandhumility.Butthiscommonlymadedistinctionis,wearetold,farfrombeingexact.Thepleasuresofpoetryandmusicmayrisetotheheightsofrapture,whilesomeoftheso-calledviolentpassionsloveandhatred,griefandjoy,forexamplemaybesogentleastobecomevirtuallyimper-ceptible.Amuchmoreusefulapproachtothepassionsdividesthembetweenthedirectandtheindirect.Amongtheformeraredesire,aversion,grief,joy,hope,fear,despairandsecurity,whilethelatterincludepride,humility,ambition,vanity,love,hatred,envy,pity,malice,generosity(T2.1.1.1,34).Thedirectpassions,towhichHumeturnsonlyneartheendofBook2,areoftwokinds.Therearecertainoriginalinstinct[s]ofthemindthatcauseustoresponddirectlytopainfulorpleasurablestimuliortotheexpectationofsuchstimuli.IfIperceivesomethinggood,somethingthatIbelievewillgivemepleasure,thatexperiencealoneisenoughtoarousethedirectpassionofdesire.IfIamcertaintoobtainthisparticulargood,orifIdoinfactobtainit,Iexperiencejoy.IfitisonlyprobablethatIwillobtainthegoodthing,Iwillfeelhope.Theperceptionofthings(evils)thatIbelievewillcausepainfunctionsinalikenaturalwaytoproduceaversion,grief,orfear.Humeobservesthatthesepassionscontinuetobefeltonlyaslongastheirstimulusorcausecontinuestohaveeffect(T2.3.9.17,911).conductofthepassions,thereisacertainregularmechanism,whichissus-ceptibleofasaccurateadisquisition,asthelawsofmotion,optics,hydrostatics,oranypartofnaturalphilosophy(DP6.19).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n20davidfatenortonInthemidstofhisdiscussionofthisfirstkindofdirectpassionHumenotesthatthereisanotherformofthesepassions.Therearealsoseveralnaturalimpulsesorinstincts:thedesireofpunish-menttoourenemies,andofhappinesstoourfriends;hunger,lust,andafewotherbodilyappetites.Thesepassions,hepointsout,areunliketheotherdirectpassionsinoneimportantrespect.Thesepas-sionsarenotcausedbygoodorevilortheprospectthereof.Onthecontrary,theyproducegoodorevil.Mydesireforthehappinessofmyfriendsleadsmetodogoodthingsforthem.MyhungerleadsmetosubstanceswhichIeatinordertoassuagethathunger;theseediblethingsIdesignateasgoodiftheyaccomplishmydesiredgoal.25Humegivesmuchmoredetailedattentiontotheindirectpas-sions,to,thatis,prideandhumility,loveandhatred(andtheirrelatedcompoundpassions,compassionandmalice,forexample).26Hearguesthatpassionsofthiskindariseastheresultofacomplexdoublerelationofimpressionsandideasthatdependsingreatmea-sureontheassociationofideasandimpressions.Thesepassionshaveastheircausesthequalitiesorpossessions(theirvirtuesortheirfinehouses,forexample)ofindividuals.Ontheotherhand,thesepas-sionshaveastheirobjects(thatis,theyaredirectedtoward)theverypersonswhohavethequalitiesorpossessionsthatcausethepas-sion.Asithappens,theobjectofprideorhumilityisalwaystheverypersonfeelingprideorhumility,whiletheobjectofloveorhatredisalwayssomeotherperson.IfIperceivethatIamvirtuous,theplea-surethisperceptionresembles,andthengivesriseto,isthepleasureofpride,whileatthesametimethefactthatIamthepersonthoughttobevirtuousgivesrisetotheideaofmyself.Asaresult,theimpres-sionorpassionofprideisdirectedtowardtheideaofmyselfinsuchawaythatItakeprideinmyowncharacterorpossessions.Loveoresteemofafriendorrelativeisexplainedanalogously,sothatIloveoresteemavirtuousfriend.ButtheimportantpointinthepresentcontexthasnottodowiththedetailsofHumesaccount,25T2.3.9.111.AttheendofBook2Humeidentifieswhatappearstobeanotherpassionofthissort,namely,curiosityorthatloveoftruththatwasthefirstsourceofallourenquiries(T2.3.10.1).26Farfromconsideringprideavice,Humeinsiststhatjustifiedandproperlydissim-ulatedprideisanormalandevenbeneficialconcomitantofvirtue,aswellasofbeauty,position,orwealth.SeeinparticularT2.1.7.8.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought21butwiththefactthatinprovidingthesedetailshedemonstrateshiscommitmenttotreatingthepassionsasnothingmoreorlessthananintegralpartofthenaturalmentalworld.Thepassions,liketheideasdiscussedinBook1oftheTreatise,arefurtherproductsoftheobservablenaturalprocessesHumeundertooktoanalyzeandexplain.Humesanalysisoftheindirectpassionofprideleadshimtoseethataformofsympathy,sympathyasaprincipleofcommunica-tion,playsafundamentalroleinourmorallives.27Havingaskedwhyitisthattheesteemofothersincreasesourpride,Humerepliesbydrawingattentiontoadeep-seatedhumanpropensitytoshare,unthinkingly,thesentimentsandopinionsofothers.If,forexam-ple,Iencountersomeonewhoisgrieving,theexternalsignsofthatgrief(facialexpressions,tears,words,etc.)arouseinmetheideaofgrief.Thisideaisthenbyastandardprocesstransformedintoanimpressionofgrief,sothatIexperiencetheverypassionbeingexpe-riencedbythepersonIamencountering.Toputthismoregenerally,inthesimplestsituations,theobservationofanotherpersonfeelingagivenpassionwillcausethesamepassiontobefeltbytheobserver(T3.3.1.7).Inothersituations,theeffectontheobservermaybedif-ferent:Humealsofindsthatifthemiseryofanotherisstronglyfeltbyanobserver,thatobserverfeelspity.Butifthatothersmiseryisonlyweaklyfelt,thentheobservermayfeelcontempt(T2.2.9.16).Humesanalysisofthepassionsalsoleadshimtoexaminethehumanwill.Hefindsthatthewillisnotafacultybutonlytheinternalimpressionwefeelandareconsciousof,whenweknow-inglygiverisetoanynewmotionofourbody,ornewperceptionofourmind.Healsofindsthatthecommondistinctionbetweennat-uralandmoralevidencebetweencausalrelationsastheyrelatetophysicalobjectsandastheyrelatetohumanactivityisunfounded.Thesetwokindsofevidence,heconcludes,cementtogether,andformonlyonechainofargumentbetwixtthem,afindingheillus-tratesbypointingoutthataprisoner,whohasneithermoneynor27Onsympathyasaprincipleofcommunication,see,e.g.,T2.1.11.2,2.2.5.1516,2.2.9.1314,2.3.6.8.ForfurtherdiscussionofsympathyseeinthisvolumeHumeandtheMechanicsofMind,PartII,andHumesLaterMoralPhilosophy,PartIII.Forstillmoreonsympathyandthepassions,seeJenniferHerdt,Reli-gionandFactioninHume’sMoralPhilosophy(Cambridge:CambridgeUniversityPress,1997).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n22davidfatenortoninterest,discoverstheimpossibilityofhisescape,aswellfromtheobstinacyofthegoaler[jailer],asfromthewallsandbarswithwhichheissurrounded;andinallattemptsforhisfreedomchoosesrathertoworkuponthestoneandironoftheone,thanupontheinflexiblenatureoftheother(2.3.1.217).Healsofindsthatitisthepassions,andonlythepassions,thatinfluencethewill,andthusthattheage-oldfearthatthereasonmaybecomeenslavedtothepassionsandconsequentlyfailtobeabletodirectthewillandouractions,makesnopracticalsense.Reasonisinertandentirelyunabletodirectthewill,or,asHumeputsitinwhatisperhapsthebest-knownremarkfromthewholeoftheTreatise,Reasonis,andoughtonlytobetheslaveofthepassions,andcanneverpretendtoanyotherofficethantoserveandobeythem(T2.3.3.4).28vi.moralsInthethirdandfinalbookoftheTreatise,OfMorals,Humelookscloselyatmorality,asubjectthatinterestsusaboveallothers.Onceagainhepursueshissubjectbymeansoftheexperimentalmethod,thistimebyfocusingonourmoralpracticesorexperience(byfocusingonmoralityaswefinditintheworld)withtheinten-tionofdiscoveringthosefeaturesofournatureandcircumstancesthatgivetoourmoralexperienceitsdistinctivefeaturesorcharac-teristics.HavingbegunBook3byremindinghisreadersofthemeaningofhiskeyterms,impressionsandideas,Humefirstshowsthatourabilitytomakemoraldistinctions,ourabilitytodistinguishvirtuefromviceormoralgoodfrommoralevil,doesnotderivefromrea-sonandthemanipulationsofideasofwhichthatfacultyiscapable.Moralityisnotaspeculativematter,butapracticalone.Thatrea-sonlacksthenecessarypracticalforcethatmoralitydemandshadbeenestablishedinTreatise2.3.3.Now,in3.1.1(Moraldistinctionsnotderiv’dfromreason)Humeadducesseveralargumentsintendedtoshowthatreasonalonecannothavebeenthesourceofthedis-tinctionbetweenvirtueandvice.Inthefollowingsection(Moral28InBook3wefindthatitisbecauseoftheactivityofsympathythatwecometotreattheartificialvirtuesasgenuinelymoralvirtues(T3.2.2.24,3.3.1.910).Onsympathyandtheartificialvirtues,seeinthisvolumetheessaysTheFoundationofMoralityinHumesTreatise,PartIIIC,andTheStructureofHumesPoliticalTheory,PartIII.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought23distinctionsderiv’dfromamoralsense)Humegoesontoarguethatmoraldistinctionsderivefromanaturalmoralsensethatprovideshumanobserverswithasetofuniqueimpressionsorfeelingswhenthemotives(theintentionsorcharacters)underlyingtheactionsofhumanagentsareimpartiallyobserved.Someofthesemotivesgiverisetouniqueimpressionsofapprovalorapprobation,andasacon-sequencearedenominatedvirtuous.Otheractionsandmotivesgiverisetouniqueimpressionsofdisapprovalordisapprobation,andasaconsequencearedenominatedvicious.Thereis,however,muchmoretobelearned.Humesobserva-tionsalsoconvincehimthattherearetwokindsofvirtue.Therearenaturalvirtues(benevolenceandhumanityareexamples),andartifi-cialvirtues(justiceandallegianceareexamples).Thesetwokindsofvirtuedifferinasmuchasthenaturalvirtuesareinherentfeaturesofhumannature.Theyareinherentnaturalpassionsthathavealwaysmotivatedspecifickindsofhumanbehavior.Inaddition,Humesaysthatthenaturalvirtuesproducemoralgoodoneachoccasionoftheiroperation,andthattheyalsogiverisetoapprovingmoralsentimentswhenevertheyareseentomotivateaction.29Incontrast,theartificialvirtuesarenotinherentfeaturesofhumannature,butmoralizedconventionsthathavedevelopedoverthecourseoftimeinresponsetohumanneeds.Humeconcludesfromtheavailableevidencethathumansinanuncultivatedstate(T3.2.2.4)wouldhavehadnoneedfortheartificialvirtuesbecausetheirnaturalvirtuesordispositionswouldhavebeenadequatetomaintainpeaceandorderinthesmall,kinship-basedunitsintowhichtheyweregathered.30Butashumangroupsbecamelargerandmorecomplex,astheybecamenotfamilies,butsocieties,cir-cumstanceswouldhavechanged,andthesechangeswouldhaveledtoconflictswithinorbetweentheexistingsocialunits,conflictswithwhichthenaturalvirtueswereunabletodealadequately.Asaconsequence,afurtherinherentfeatureofhumannature,self-interest,requiredthedevelopmentofconventions(propertyrightsandrules,ortherulesofjustice)toregulatepossessionsandstill29Humemakesotherclaimsthatsuggestthatperforminganaturallyvirtuousactmaynot,onsomeoccasions,producegood.Forabriefdiscussionofthispoint,seeinthisvolumetheessayTheFoundationofMoralsinHumesTreatise,note40.30HumesupposedhisviewtobeverifiedbytheAmericantribes,wheremenliveinconcordandamityamongthemselveswithoutanyestablishdgovernment,exceptintimeofwarwiththeirneighbors(T3.2.8.2).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n24davidfatenortonotherconventions(theprinciplesofgovernance)toregulatepoliticalbehavior.Giventhattheseconventionsorvirtueswereunknowntohumanslivingintheirfirstoruncultivatedstate,itisobviousthattheycannotalwayshavemotivatedordirectedhumanbehavior.More-over,althoughtheuniformpracticeofanartificialvirtuesuchasjusticemaybeabsolutelynecessaryforthegoodofsocietyasawhole,thepracticeofthisvirtuemayonanygivenoccasionbecontrarytothegoodofbothindividualsandthepublicgenerally(T3.2.2.22,3.3.1.12).Finally,Humepointsoutthattheartificialvirtueshaveadoublefoundation.Justice,hesays,hastwodif-ferentfoundations.Thedistinctionbetweenjusticeandinjusticebeginswithself-interest,whenmenobserve,thattisimpossibletoliveinsocietywithoutrestrainingthemselvesbycertainrules.Butthissamedistinctionalsocomestobefoundedinmorality.Thathappenswhenthisinterestisalsoobservdtobecommontoallmankind,andmenreceiveapleasurefromtheviewofsuchactionsastendtothepeaceofsociety,andanuneasinessfromsuchasarecontrarytoit(T3.2.6.11).vii.recastingsandcontinuations1.Withinamatterofonlyafewyearsafteritsappearance,HumecametoregretthepublicationoftheTreatise.InMyOwnLife,whilereviewingtheyearsimmediatelyfollowingpublicationofthework,hewrote:NeverliteraryattemptwasmoreunfortunatethanmyTreatiseofHumanNature.Itfelldead-bornfromthepress,withoutreachingsuchdistinction,aseventoexciteamurmuramongthezealots.Humesreasonsforthisremarkarenotobvi-ous.From1739to1741,theTreatisewasreviewedbothinBritainandonthecontinent,andsomeofthesereviewsmustfairlybedescribedasexcitingmurmursamongthereligious,whiletheworksskepticalandallegedlyatheisticaltendencieswouldbythemid-1740sbealeadingcauseofHumesfailuretobenamedProfessorofMoralPhilosophyattheUniversityofEdinburgh.31Butdespitethis31ThehistoricalmaterialsketchedhereisdiscussedinmuchgreaterdetailinmyHistoricalAccountofATreatiseofHumanNaturefromItsBeginningstotheTimeofHumesDeath(hereafterHistoricalAccount),inATreatiseofHumanNature,2vols.(Oxford:ClarendonPress,2007),2:451526.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought25notorietyanddespiteHumestwopublishedattemptstoclarifythework(theAbstractoftheTreatise32andtheAppendixtoBook3),theworksoldsopoorlythattherewasneveranopportunityforthesecondeditionHumelookedforwardtoevenbeforevolume3waspublishedinlate1740.33ThusindescribingtheTreatiseasdead-bornfromthepressHumemayhavehadthislackofcommercialsuccessinmind.34Itseemsmorelikely,however,thathisgreatestregretabouttheTreatisehadtodowithhisassessmentofhisownperformanceinthework.Hecametobelievethathehadbungledhisattempttointroduceacomprehensivenewsystemofphilosophy.Hewaslatertosayofthatperformance:IwascarrydawaybytheHeatofYouth&Inventiontopublishtooprecipitately.SovastanUndertaking,plandbeforeIwasoneandtwenty,&composdbeforetwentyfive,mustnecessarilybeverydefective.IhaverepentedmyHasteahundred,&ahundredtimes(HL1:158).35In1748HumepublishedPhilosophicalEssaysconcerningHumanUnderstanding(latertobetitledAnEnquiryconcerningHumanUnderstanding),arecastingofamodestportionofthematerialsfoundinBooks1and2oftheTreatise,alongwithtwoessaysonreligioustopics(OfMiraclesandOfaParticularProvidenceandofaFutureState)andtwosystematicdiscussions:OftheDif-ferentSpeciesofPhilosophyandOftheAcademicalorSceptical32TheAbstractoftheTreatise,ashortworkthatattemptstorenderalargerworkmoreintelligibletoordinarycapacities,byabridgingit,or,asthesubtitleoftheworkputsit,toillustrateandexplainthechiefargumentofthatwork(APreface,titlepage).TheAbstractwaspublishedinMarch1740.Foracom-prehensiveefforttoestablishbeyondallreasonabledoubtthatHumewrotetheAbstract,seemyHistoricalAccount,2:45971.AsitisnowsettledthatHumewrotetheAbstract,thisshortworkcanbeenthusiasticallyrecommendedtothosewhowishtoconsiderHumesownaccountofwhathetooktobethechiefargu-mentofthefirsttwovolumesoftheTreatise,namely,hisargumentabouttheoriginof,andbeliefin,theideaofcausalconnection.33InMarch1740HumewrotetoFrancisHutcheson:IwaitwithsomeImpatienceforasecondEditionprincipallyonAccountofAlterationsIintendtomakeinmyPerformance(HL1:389).34Humescontractwiththebookseller(JohnNoon)whopublishedvols.1and2oftheTreatisepreventedHumefrompublishingasecondeditionoftheTreatiseuntilallcopiesofthefirsteditionweresold.Buttheworksoldverypoorly,sopoorlythat,followingJohnNoonsdeathin1763,290ofthe1,000copiesprintedin1739weresoldtothebooktradefor7penceeach,about6percentoftheestablishedretailpriceof10shillings.Fouryearsearlier,itappearsthat200copiesofvol.3weresoldatasimilarauction.Forfurtherdetails,seemyHistoricalAccount,2:5834.35ForasystematicreviewofHumesassessmentsoftheTreatise,seemyHistoricalAccount,2:57688.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n26davidfatenortonPhilosophy.36OfthisworkhesaidthathethoughtitcontainedeverythingofConsequencerelatingtotheUnderstanding,whichyouwoudmeetwithintheTreatise;&IgiveyoumyAdviceagainstreadingthelatter.Byshortening&simplifyingtheQuestions,Ireallyrenderthemmuchmorecomplete.Addodumminuo[IaddwhileIdecrease].ThephilosophicalPrinciplesarethesameinboth(HL1:158).37ThesubstantiallyrewrittenversionofBook3oftheTreatise,AnEnquiryconcerningthePrinciplesofMorals,theworkthatHumetooktobe,ofallmywritings,historical,philosophical,orliterary,incomparablythebest(MOL10),waspublishedin1751.ADisser-tationonthePassions,clearlyarecastingofBook2intowhatHumedescribedasoneofseveralsmallpieces,waspublishedin1757.VerylateinhislifeHumegrewimpatientwithhiscriticsforfocusingtheirattentionontheTreatiseratherthanhisshorterrestatementsofhisviews,andsohecomposedabriefAdvertisementthatheaskedtohaveincludedintheappropriatevolumeofallexistingandfuturecopiesofhisEssaysandTreatisesonSeveralSubjects.38ThisnoticeasksthattheTreatisebeignored:Mostoftheprinciples,andreasonings,containedinthisvolume,werepub-lishedinaworkinthreevolumes,calledATreatiseofHumanNature:AworkwhichtheAuthorhadprojectedbeforeheleftCollege,andwhichhewroteandpublishednotlongafter.Butnotfindingitsuccessful,hewassen-sibleofhiserroringoingtothepresstooearly,andhecastthewholeanewinthefollowingpieces,wheresomenegligencesinhisformerreasoningand36ThephilosophicalpositionthatHumecallstheAcademicalphilosophyisthatwhichisnowlikelytobereferredtoasacademicskepticism.Itsnamederivesfromthefactthatitwasfirstarticulated,inthethirdcenturyb.c.,attheAcademyfoundedbyPlato.37InMyOwnLife(MOL8)Humewrote:Ihadalwaysentertainedanotion,thatmywantofsuccessinpublishingtheTreatiseofHumanNature,hadproceededmorefromthemannerthanthematter,andthatIhadbeenguiltyofaveryusualindiscretion,ingoingtothepresstooearly.I,therefore,castthefirstpartofthatworkanewintheEnquiryconcerningHumanUnderstanding.38OnHumesEssaysandTreatisesonSeveralSubjects,seenote2above.After1758thiscollectionwasfortherestofHumeslifepublishedineithertwo-orfour-volumeeditions.TheAdvertisementwasintendedtoserveasaPrefacetothevolume(s)ofthisworkcontainingAnEnquiryconcerningHumanUnder-standing,theDissertationonthePassions,andAnEnquiryconcerningthePrin-ciplesofMorals,asitisthesePiecesthatHumetooktorepresenttheTreatisecast...anew.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought27moreintheexpression,are,hehopes,corrected.Yetseveralwriters,whohavehonouredtheAuthorsPhilosophywithanswers,havetakencaretodirectalltheirbatteriesagainstthatjuvenilework,39whichtheAuthorneveracknowledged,andhaveaffectedtotriumphinanyadvantages,which,theyimagined,theyhadobtainedoverit:Apracticeverycontrarytoallrulesofcandourandfair-dealing,andastronginstanceofthosepolemicalartifices,whichabigottedzealthinksitselfauthorizedtoemploy.Henceforth,theAuthordesires,thatthefollowingPiecesmayaloneberegardedascontain-inghisphilosophicalsentimentsandprinciples.40ReasonablethoughhisdesiretodistancehimselffromtheTreatisemayhaveseemedtohim,fewseriousreadersofHumehavebeenabletoconcurwithit.41ForHumescritics,theTreatiseisanirre-sistibletarget;forthosewhobelievehimtohavebeenaprofoundandconstructivestudentofhumannature,theworkistoorichtoignore.2.AbouttheworksthataresaidtorepresenttheTreatisecast...anew,twothingsareobvious.First,aswesawabove,Humescommitmenttotheexperimentalmethodcontinuedunabatedintheselaterworks.Second,Humedoesnotmerely,ashesuggests,improvebyshorteningandsimplifying.HisEnquiryconcerningHumanUnderstanding(thefirstEnquiry)includessomeimpor-tantadditions,mostnotablytwoattention-gettingdiscussionsofreligiousissues.Inanefforttomakehisviewsreligiouslyinnocuoussothattheymightbeconsideredcalmlyandontheirphilosophi-calmerits,hehadcarefullyexcisedfromtheTreatiseanythingthatcouldbetakenasantireligious.Nevertheless,theviewsoftheTrea-tiseandEssays,MoralandPoliticalweretoothoroughlysecularto39ForasurveyofresponsestotheTreatisefrom1739to1777,seemyHistoricalAccount,2:494576.Oftheprincipalresponseswrittenafterthepublicationofthetwoenquiries(EHUin1748,EPMin1751),onlythatofThomasReid,inAnInquiryintotheHumanMindonthePrinciplesofCommonSense(1764),limitsitsdiscussionandcriticismofHumetotheviewsexpressedintheTreatise.40TheAdvertisementisquotedfromAnEnquiryconcerningHumanUnderstand-ing,ed.T.L.Beauchamp,[1].ThesametextisalsofoundintheOxfordPhilosoph-icalTextseditionofthiswork,ed.T.L.Beauchamp(Oxford:OxfordUniversityPress,1999),[83].41AnotablerecentexceptionisthecollectionofessayseditedbyPeterMillicancitedinnote19andintheSelectedBibliography.SeealsoAntonyFlew,Hume’sPhilosophyofBelief:AStudyofhisFirstInquiry(London:Routledge&KeganPaul,1961),andinthisvolumetheessayHumesLaterMoralPhilosophy.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n28davidfatenortonpassunremarkedinareligiousage,and,asnoted,bythemid-1740sHumehadbeenbrandedareligiousskepticwithatheistictenden-cies.Heseemsinconsequencetohavedecidedtochallengeopenlytherationalityofreligiousbelief.Inanyevent,thefirstEnquiry,thefirstofhisrecastings,includedtwoofHumesmostprovoca-tiveforaysintothephilosophyofreligion,OfMiraclesandOfaParticularProvidenceandofaFutureState,whileTheNaturalHistoryofReligionwasdenouncedasatheisticevenbeforeitwaspublished.42ThesetwoessaysestablishedbeyondalldoubtHumescharacterasareligiousskeptic.Takentogether,theychallengethefoundationsofmuchreligiousbeliefandattempttocurbitsexcessesbyundertak-ingtoshowthatthisformofbeliefhasitsbeginningsinsourcesorcausesaboutwhichwemustbedeeplysuspicious.InOfMiracles,forexample,Humearguesthatbeliefinmiracles,akindofputativefactusedtojustifyacommitmenttocertaincreeds,canneverpro-videthesecurefoundationsuchcreedsrequire.Heseesthatthesecommitmentsaretypicallymaintainedwithamind-numbingtenac-ityandadisruptiveintolerancetowardcontraryviews.Tocountertheseobjectionablecommitments,henotesthatthewidelyheldviewthatmiraclesareeventsthatviolateestablishedlawsofnaturemeansthattheevidenceforeventhemostwell-attestedmiraclewillalwaysbecounterbalancedbytheuniformexperience(theproof)thatestablishesthelawofnaturethatthemiracleallegedlyviolates.Moreover,hearguesthattheevidencesupportinganygivenmiracleisnecessarilysuspect.Hisargumentleavesopenthepossibilitythatunusualeventsmayhaveoccurred,butitshowsthatthegroundsforabeliefinanygivenmiracleorsetofmiraclesaremuchweakerthanthereligioussuppose.Thereareandwillbethosewhobelievethatmiracleshaveoccurred,butHumesanalysisshowsthatsuchbeliefs42HumeatonepointthoughtofincludingOfMiraclesintheTreatise,butdecidednottodosoaspartofhisprogramtoeliminatereligiouslyoffensivematerialfromthatwork.Formoreonthissubject,seeinthisvolumetheessayDavidHume:TheHistorian.Humesreputationasareligiousskeptic,andevenanatheist,wasinstrumentalinhisfailure,in1745,tobeappointedtotheChairofMoralPhilosophyattheUniversityofEdinburgh.SeeALetterfromaGentlemantohisFriendinEdinburgh(Edinburgh,1745).OnthecontroversysurroundingthepublicationofTheNaturalHistoryofReligion,seeErnestCampbellMossner,TheLifeofDavidHume(Edinburgh,1954),31935.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought29willalwayslacktheforceofevidenceneededtojustifythearroganceandintolerancethatcharacterizessomanyofthereligious.OfaParticularProvidenceandofaFutureState,Section11ofthesameEnquiry,andtheposthumouslypublishedDialoguescon-cerningNaturalReligionhaveasimilareffect.Philosophersandthe-ologiansoftheeighteenthcenturyoftenargued(theargumentfromdesign,asitisknown)thatthewell-ordereduniverseinwhichwefindourselvescanonlybetheeffectofasupremelyintelligentcause,thateachaspectofthisdivinecreationiswelldesignedtofulfillsomebeneficialend,andthattheseeffectsshowusthattheDeityiscaringandbenevolent.Humeshowsthattheseconclusionsgowellbeyondtheavailableevidence.Thepleasantandwell-designedfeaturesoftheworldarebalancedbyagoodmeasureoftheunpleasantandtheplainlybotched.Ourknowledgeofcausalconnectionsdependsontheexperienceofconstantconjunctions;thesecausethevivacityofapresentimpressiontobetransferredtotheideaassociatedwithit,andleaveusbelievinginthatidea.Butinthiscasetheeffecttobeexplained,theuniverse,isunique,anditscauseunknown.Wesee,then,thatwecannotpossiblyhaveexperientialgroundsforanykindofinferenceaboutthiscause.Onexperientialgroundsthemostwecansayisthatthereisamassive,mixedeffect,and,aswehavethroughexperiencecometobelievethateffectshavecausescommensuratetothem,thiseffectprobablydoeshaveacommensu-ratelylargeandmixedcause.Furthermore,astheeffectisremotelyliketheproductsofhumanmanufacture,wecansaythatthecauseorcausesoforderintheuniverseprobablybearsomeremoteanal-ogytohumanintelligence(DNR12.33,227).Thereisindeedaninferencetobedrawnfromtheuniqueeffectinquestion(theuni-verse)tothecauseofthateffect,butitisnottheargumentofthetheologians,anditprovidesnofoundationforanyformofsectarianpretensionoreventhemildestformsofintolerance.TheNaturalHistoryofReligionfocusesonthequestionoftheoriginofreligioninhumannature.Humeasks,thatis,whatfea-turesofhumannatureaccountforthewidespreadbutnotuniversalbeliefininvisibleandintelligentpower(s).Hedeliversathoroughlydeflationaryandnaturalisticanswer:religiousbeliefspringsnotfromanoriginalinstinctorprimaryimpressionofnature,notfromanyuniversalandfundamentalprincipleofournatures,butfromfea-turesofhumannaturethatarederivativeandwhoseoperationmayCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n30davidfatenortoneasilybepervertedbyvariousaccidentsandcauses[or]...altogetherprevented(NHRIntro.1).Moreover,itisthedarker,lesssalubri-ousfeaturesofournaturethattaketheprincipalpartsinthisstory.Primitivepeoplesdidnotfindnaturetobeanorderlywholepro-ducedbyabeneficentdesigner,butfragmented,arbitrary,andfear-some.Asaconsequence,motivatedbytheirownignoranceandfear,theycametothinkoftheactivitiesofnatureastheeffectofamul-titudeofpettypowerstheeffectsofdeitieswhocould,throughpropitiatingworship,beinfluencedtoamelioratethelivesofthosewhoengagedinthisworship.Subsequentlythesamefearsandper-ceptionstransformedpolytheismintomonotheism,theviewthatasingle,omnipotentbeingcreatedandstillcontrolstheworldandallthattranspiresinit.FromthisconclusionHumegoesontoarguethatmonotheism,seeminglythemoresophisticatedposition,isinfactmorallyretrograde,for,oncehavingestablisheditself,monotheismtendsnaturallytowardzealandintolerance,encouragesdebasing,monkishvirtues,andisitselfadangertosocietybecauseitprovestobeacauseofviolentandimmoralactsdirectedagainstthosewhofailtoactinaccordwithitstenets.Incontrast,polytheismistolerantofdiversityandencouragesthosegenuinevirtuesthatimprovethecircumstancesofhumankind,andthusfromamoralpointofviewissuperiortomonotheism(NHR910).Theimportantpoint,how-ever,isthesuggestionthatallreligiousbeliefderivesinitiallyfromfearandignorance,and,moreover,thatitencouragesthecontinueddevelopmentoftheseundesirablecharacteristics.3.Inanumberofrespects,HumesEssaysandhisHistoryofEnglandconstitutecontinuationsofhisearliestwork.Theyare,ofcourse,furthermanifestationsofhisattempttoextendtheexperi-mentalmethodintomoralsubjects.Theyarealsofurthermanifes-tationsofhisattempttogainunderstandingbymeansofanexam-inationoforiginsorbeginnings.Theirtitlesaloneindicate,oftenenough,theseinterests:ThatPoliticsmaybereducedtoaSci-ence,OftheFirstPrinciplesofGovernment,OftheOriginofGovernment,OftheRiseandProgressoftheArtsandSciences.Others,withlesstelltaletitles,arenonethelessapartofthesameproject.OftheLibertyofthePresstracestheunparalleledlibertyofthepressBritishsubjectsenjoyedintheeighteenthcenturytothemixedformofgovernmentfoundinBritain,andthuslendsCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought31experientialsupporttomixedformsofgovernance(E-LP2,10).InOftheIndependencyofParliamentHumedrawsattentiontothefactthattheHouseofCommonscouldeasilywrestallpowerfromtheKingandLords,butdoesnotdoso.Heexplainsthisunusualsituationbylookingforanexplanationthatisconsistentwithourexperienceofhumannature,andconcludesthatafundamentalfeatureofthisnature,theself-interestoftheindividualmembersoftheCommons,actsasabrakeontheexpansionofthepowerofParliament(E-IP6,445).OfPartiesinGenerallooksforthesources,againinhumannature,ofparties,orthosedetestablefac-tionsthatsubvertgovernment,renderlawsimpotent,andbegetthefiercestanimositiesamongmenofthesamenation,whooughttogivemutualassistanceandprotectiontoeachother(E-PG2,55).OfSuperstitionandEnthusiasmoutlinestheperniciouseffectsongovernmentandsocietyofthetwotypesoffalsereligionalludedtointhetitleoftheessay.43Andsoon.ThereisatleastoneadditionalsenseinwhichtheEssaysandTheHistoryofEnglandrepresentacontinuationoftheprojectthatbeganwithATreatiseofHumanNature:TheworkforwhichHumeisrememberedisallfundamentallyhistorical.Thatis,allthisworkattemptstoexplainsomethingthatwepresentlybelieve,feel,say,thinkordo,toexplainsomepresentstateofaffairs,whetherthatstatebeinthemental,moral,orpoliticalworld,bytracingtheseper-ceptions,actions,orstatesbytracingcertainnoteworthyeffectstothepriorconditionsthatappeartohavebeentheircauses.Aswesawabove,ourexperimentsinthescienceofhumannaturemustbegleanedfromacautiousobservationofhumanlife,fromthecommoncourseoftheworld,bymensbehaviourincompany,inaffairs,andintheirpleasures.Observationsofwhathumanshavedone,howtheirmindswork,howtheirinstitutionshaveformed:theseareallhistoricalobservationsofseveraldifferentkinds.Humerevealssomethingmoreofhisviewofexplanationinoneoftheessaysjustmentioned,OftheRiseandProgressoftheArtsandSciences.Enquiriesintohumanaffairs,hesaysthere,requireustodistinguishbetweenwhatisowingtochance,andwhatproceedsfromcauses.Ifwesaythataneventisowingtochance,weare43ForadiscussionoftheseissuesastheybearonHumespoliticaltheory,seeinthisvolumetheessayTheStructureofHumesPoliticalTheory.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n32davidfatenortonineffectconfessingourignorance,andputtinganendtoattemptsatexplanation.Butifwesupposesomeeventorstateofaffairsistheresultofcauses,weleaveourselvestheopportunityofassign-ingthesecausesanddisplayingourprofoundknowledge.Asageneralrule,hesays,Whatdependsuponafewpersonsis,inagreatmeasure,tobeascribedtochance,orsecretandunknowncauses:Whatarisesfromagreatnumber,mayoftenbeaccountedforbydeterminateandknowncauses(E-RP12,11112).Conse-quently,explanationsof,say,thecourseofdomesticpoliticsortheriseofcommercewillbeeasiertocomebythanexplanationsofartisticdevelopment.Andyetacarefulenquirermayperhapsdetectregularitiesbetweencertainconditionsandtheflourishingoftheartsandsciences.Inthisparticularessay,Humeturnshishandtogivingjustsuchanexplanation.But,moreimportantly,theEssaystakentogether,andTheHistoryofEngland,aretheresultofHumesattemptstopushbackthefrontiersofignoranceormisunderstand-ingbyassigningcausestophenomenapreviouslyattributedtotheworkingsofchance,ortothatwhichinHumesviewisequallydevoidofcontent,theworkingsofprovidence.viii.thereformerInAugust1776,afewdaysbeforehisdeath,HumewasvisitedbyAdamSmith,oneofhisclosestfriends.OnobservingthatHume,whohadbeenseriouslyillforsomemonths,wascheerfulandappar-entlyfullofthespiritoflife,Smithcouldnothelpentertainingsomefainthopesofhisfriendsrecovery.Yourhopesareground-less,Humereplied,andeventuallyturnedtheconversationontoLuciansDialoguesoftheDead,andtheexcusesofferedtoCharontheboatmanfornotenteringhisboattobeferriedtoHades.Noneoftheclassicalexcusesfittedhim,Humenoted.Hehadnohousetofinish,nochildrentoprovidefor,noenemiestodestroy.Hethendivertedhimself,Smithcontinues,withinventingseveraljocularexcuses,whichhesupposedhemightmaketoCharon,andwithimaginingtheverysurlyanswerswhichitmightsuitthecharacterofCharontoreturntothem.Uponfurtherconsideration,said[Hume],IthoughtImightsaytohim,GoodCharon,Ihavebeencorrectingmyworksforanewedition.Allowmealittletime,thatImayseeCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought33howthePublicreceivesthealterations.ButCharonwouldanswer,Whenyouhaveseentheeffectofthese,youwillbeformakingotheralterations.Therewillbenoendofsuchexcuses;so,honestfriend,pleasestepintotheboat.ButImightstillurge,Havealittlepatience,goodCharon,IhavebeenendeavouringtoopentheeyesofthePublic.IfIliveafewyearslonger,Imayhavethesatisfactionofseeingthedownfal[l]ofsomeoftheprevailingsystemsofsuperstition.ButCharonwouldthenlosealltemperanddecency.Youloiteringrogue,thatwillnothappenthesemanyhundredyears.DoyoufancyIwillgrantyoualeaseforsolongaterm?Getintotheboatthisinstant,youlazy,loiteringrogue.44OfthemanyanecdotesaboutHumethathavesurvived,none,Isuggest,betterrevealshischaracter.Thereis,first,thefactthataman,correctlyconvincedofhisimminentdeath,andequallysatisfiedthatdeathissimplyannihilation,wouldtreatthemat-terlightly.45Serioustopicstreatedattimeswithnonchalance:thishasbeenenoughtoleadsomeofhiscriticstosuppose,mistakenly,thatHumelackedseriousnessofpurpose,tosupposethatrhetori-caleffectwastohimmoreimportantthantruth.Humedidattimestreatserioustopicslightly,andhedidhavereservationsaboutclaimstohavefoundultimateprinciplesorTheTruth,butthesefactsareentirelyconsistentwithhismostfundamentalandunmistakablyseriousaim.46IntheconversationwithSmith,forexample,Humeshumorisfocusedontwotopicsofgenuineconcerntohim.Hewas,surely,ashecandidlytellsusinMyOwnLife,concernedwithhisliteraryreputation,andhejustassurelytookgreatpleasureinbeingrecog-nizedasoneofEuropesleadingintellectualfigures.ButitwasnotmerelyfamethatHumesought;itwasalsoreputation.Beforehehadpublishedanythinghesaidthathewoudratherlive&dyeinObscuritythanpublishhisviewsinamaimd&imperfectform(KHL6).WiththeTreatisefinallypublished,hediscouragedafriend44LetterfromAdamSmith,LL.D.toWilliamStrahan,Esq.(DNR2445).45AfewweeksbeforehisdeathHumewasabletosatisfyJamesBoswellthathesincerelybelieveditamostunreasonablefancythattheremightbelifeafterdeath(DNR767).HumesessayonthetopicOftheImmortalityoftheSoul,intendedforpublicationin1755,waspublishedinaFrenchtranslationin1770,andinEnglishin1777,ayearafterHumesdeath(seeE-IS5908).46Forfurtherdiscussionofthispoint,seeinthisvolumetheessayHumesMoralPsychology.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n34davidfatenortonfrompursuingaschemetoincreasesales.Hisfirstconcern,hesaid,wasnotwithcommercialsuccess,butwithearningtheapprobationofthosecapableofjudginghiswritings(NHL4).And,ashisfirstexcusetoCharonindicates,heconstantlyrevisedandalteredhisEssaysandTreatises,andHistoryofEnglandindeed,hedidso,apparently,onhisdeathbedatatime,inotherwords,whenhehadnootherreasonforrevisinghisworkthanhisowninterestinimprovingit.WecanagreethatHumewroteforeffect,butweneednotconcludefromhisoccasionallightnessoftonethathelackedseriouspurpose.47HumessecondexcusetoCharonrevealsmuchaboutthatpur-pose.Hehas,hesays,beenendeavouringtoopentheeyesofthePublicandwouldliketoremainalivelongenoughtohavethesat-isfactionofseeingthedownfal[l]ofsomeoftheprevailingsystemsofsuperstition.Humethereformerisonlyseldomnoticed.48Andyetfromearlydaysreformwastheeffectatwhichheaimed.Inthebeginningitwasreformationthroughthescienceofmanatwhichheaimed,areformationthatwould,ifsuccessful,havetheeffectofreformingalltheothersciences,foralloftheseevenMath-ematics,NaturalPhilosophy,andNaturalReligionaredepen-dentonthescienceofhumannature(TIntro.4).InthePrefacetotheAbstracthedescribeshimselfas,byhisTreatise,suggestingthatwerehisphilosophyreceived,wemustalterfromthefoun-dationthegreatestpartofthesciences.Habit,hisskepticsays,isapowerfulmeansofreformingthemind,andimplantinginitgooddispositionsandinclinations,whilethegreatvalueofphilos-ophyderivesfromthefactthat,properlyundertaken,Itinsensiblyrefinesthetemper,anditpointsouttousthosedispositionswhichweshouldendeavourtoattain,byaconstantbentofmind,andbyrepeatedhabit(E-Sc30,1701).moralphilosophy,hesaysattheverybeginningofAnEnquiryconcerningHumanUnderstanding,47Inthelettercitedinnote44,Smithwentontoadd:Andthatgaietyoftemper,soagreeableinsociety,butwhichissooftenaccompaniedwithfrivolousandsuper-ficialqualities,wasinhimcertainlyattendedwiththemostsevereapplication,themostextensivelearning,thegreatestdepthofthought,andacapacityineveryrespectthemostcomprehensive(DNR,247).48AnimportantexceptionisJohnB.Stewart,OpinionandReforminHume’sPolit-icalPhilosophy(Princeton:PrincetonUniversityPress,1992);seeespecially194317.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought35maycontributetotheentertainment,instruction,andreformationofmankind(EHU1.1).Humehadnothoughtofreformingthefundamentaldispositionsofhumannatureitself.Thesehetooktobesettled,andutopianschemesdependentonachangedconstitutionofhumanityhedis-missedwithoutqualification:Allplansofgovernment,whichsup-posegreatreformationinthemannersofmankind,areplainlyimag-inary(E-IP4,514).Reformation,ifitistotakeplace,willaffectindividuals,andwillbeintheformofthatrefinementofcharacterthatresultsfromnewhabitsofmind,and,mostparticularly,fromnewhabitsofbelief.Itwillbetheeffectofindividualsmelding,asHumemelded,theexperimentalmethodofreasoningintoanupdatedversionOftheAcademicalorScepticalPhilosophy.Thislatterspeciesofphilosophyhas,hesays,aclearadvantageoverallotherkinds:byitsverynatureitprotectsthosewhoadoptitfromtheexcessescharacteristicofotherformsofphilosophy.Theaca-demicskeptic,notingthedangersofhastyanddogmaticjudgment,emphasizescontinuallytheadvantagesofdoubtandsuspenceofjudgment...ofconfiningtoverynarrowboundstheenquiriesoftheunderstanding,andofrenouncingallspeculationswhichlienotwithinthelimitsofcommonlifeandpractice(EHU5.1).Humespostskepticalphilosophydoesnotcounselustosuspendalljudg-ment,belief,andaffirmation.Instead,acceptingthebasiclessonsofskepticism,itattemptstoshowushowtomoderateourbeliefsandattitudes.Thosewhopracticedhisprincipleswould,Humethought,learnhowtoavoidthatcombinationofarrogance,pretension,andcredulitythathefoundsodistastefulandstifling,sodangerousinitstypicalmanifestations,religiousdogmatismandpoliticalfaction.Humedidnotsupposethathewouldeffectchangesinhumannatureitself,buthedidhopethathecouldmoderateindividualbeliefandopinion,and,asaresult,moderatetheopinions,institutions,andbehaviorofthepublicatlarge.Asimplebutprofoundgoal:toopentheeyesofthePublic,withtheaimofundercuttingtheprevailingsystemsofsuperstition.4949PaulRussellarguesthatHumesupposedthereformationofcharactertobedifficultandevenunlikely.SeehisFreedomandMoralSentiment:Hume’sWayofNatu-ralizingResponsibility(NewYork:OxfordUniversityPress,1995),1025,12830.ForadiscussionofHumeslesspessimisticviewsaboutthereformationofbelief,seethepaperscitedinnote19above.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n36davidfatenortonix.fourteenessaysonhume’sthoughtAlthoughperhapsnowbestknownforhiscontributionstothethe-oryofknowledge,metaphysics,andphilosophyofreligion,Humealsomadesubstantialandinfluentialcontributionstomoralthe-oryandmoralpsychology,politicalandeconomictheory,politicalandsocialhistory,and,toalesserextent,aestheticandliterarythe-ory.TheessaysinthisvolumeapproachHumeinthistopicalway.Theyintroducereaderstohiswide-rangingthoughtbyfocusingonfourteenoverlappingareasofinterest.Theessaysthemselvesarearrangedinapatternthatreflects,first,thestructuralorderofATreatiseofHumanNature,Humesmostsystematicphilosophicalpublication,andthenthepatternofhislaterpublications.SomeessaysshowhowHumesthoughtmaybelinkedtothatofhisprede-cessorsandcontemporaries.Othersaremoreconcernedwithlinkstothetwentiethcentury.EachprovidesanaccessibleaccountofsomecentralaspectsofHumesthought.ThefirstessayoutlinesHumesplansforanewscienceofhumannature,asciencethatistoserveasthefoundationofalltheothersciences,moralaswellasnatural.Thisscience,JohnBiroargues,hassignificantaffinitieswithmoderncognitivescienceandoffersinsightsthatwillbeofusetothoseengagedinthiscontemporaryenterprise.DavidOwenpointsouttheinnovationsHumemadetothethe-oryofideasacceptedbyhispredecessors,andthenshowshowhisviewsaboutimpressionsandideasandtheprinciplesofassociationenablehimtoprovideexplanationsofcausalreasoning,belief,sev-eralkeypassions(pride,humility,love,andhatred),andtheprincipleofsympathy,andthusformthecoreofhisscienceofhumannature.DonaldBaxterexplainsHumesperplexingaccountofspaceandtime,andofourideasofthem,inthecontextofhisskepticalrelianceonappearances.TheexplanationshowsthesurprisingforceofHumesargumentsagainstinfinitedivisibility,theexactnessofgeometry,vacuums,andtimewithoutchange,whileilluminatinghisrelationalaccountofspaceandtimeasmannersinwhichcol-oredortangiblepointsexist.MartinBellsessaytakesupHumeswidelydiscussedaccountofcausationornecessaryconnection.TheessayfocusesonHumeseffortstoanswertwoquestionsofgreatimportance:HowdoweCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought37acquiretheideaofcauseandeffect?Andhowareweabletoinfereffectsfromcausesandviceversa?Contrarytowhatwemighthaveexpected,weareshownthatHumesanswertothefirstquestiondependsonhisanswertothesecond.TheessayalsoexaminesHumestwodefinitionsofcauseandsomerecentinterpretationsofHumesviewsoncausation.JaneMcIntyresituatesHumesdiscussionofpersonalidentityinitsbroaderhistoricalcontext,includingtheimportantdebatebetweenAnthonyCollinsandSamuelClarke.ShealsolaysoutthetextualevidencesupportingtheviewthatinhislaterwritingsHumecontinuedtohold,andevenexpandon,hisTreatiseaccountoftheself.NotingthatHumedescribesthephilosophyoftheTreatiseasverysceptical,RobertFogelinattemptstoseewhatthisskep-ticismamountsto,andhowitisrelatedtootheraspectsofhisphilosophicalprogram.HeconcludesthatwhileHumeclearlydidnotrecommendawholesalesuspensionofbelief(hethoughtthisimpossible),heis,insofarashepresentsuswithathoroughgoingcritiqueofourintellectualfaculties,aradical,unreserved,unmiti-gatedskeptic,andthattothinkotherwiseistomissmuchofHumesgenius.OfthefouressaysthattakeHumesmoraltheoryasapointofdeparture,thatbyTerencePenelhumconsidersthoseelementstheself,thepassions,andthewill,forexampleofHumesviewofhumannaturethataremostintimatelyrelatedtohisobjectivesasamoralphilosopher,butnotbeforehehasconsideredHumescharacterandtheimportantquestionsthatsomehaveraisedabouthispsychologicalqualificationsfordoingphilosophy.DavidFateNortonsituatesthemoraltheoryoftheTreatisewithinacentrallyimportantdebateaboutthefoundationsofmorality.AccordingtoHume,wearebynaturemotivatedbybothaltruis-ticandself-interesteddispositions.Wearealsobynaturesensitivetothemotivationsthedesires,intentions,orcharacterofthoseweencounter,andasaconsequencewecometodenominatesomeofthesemotivations,andthosewhohavethem,virtuous,andothersvicious.HealsoreviewsHumessuggestionsabouttheoriginoftheideasofblameandobligation.JacquelineTaylorexaminesHumespost-Treatisewritingsonmorals,drawingattentiontotheemphasistheseplaceontheCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n38davidfatenortonlanguageofmoralityanditsroleinmoralconversationanddebate.ShealsodiscussestheimportanceofHumessentiment-basedethicswhenheconsidersapparentculturaldifferencesofmoraloutlook.KnudHaakonssenarguesthatHumeundertooktoshowthatmostearlymodernviewsofsocietyandpolitics,foundedastheywereontwoformsoffalsereligion,superstitionorenthusiasm,werephilo-sophicallymisconceived,empiricallyuntenable,and,oftenenough,politicallydangerous.Incontrast,Humeprovidedahumanisticaccountofpoliticalmoralityanaccountthatseesourpoliticalinstitutionsashumanconstructsthatdependonhumannatureandhumanexperience.WiththepublicationofhisPoliticalDiscoursesin1752,Humeestablishedhimselfasanimportantpoliticaleconomist.AndrewSkinnersketchesthebackgroundofeconomictheoryinwhichHumesworkappeared,outlinesHumesinsightfulalternativeviews,andconcludesbynotingHumesinfluenceontheeconomicwritingsof,amongothers,hisgoodfriendAdamSmith.In1757,withthepublicationofOftheStandardofTasteandOfTragedy,Humeprovidedhisreaderswithwhatmaybesomeofthesurvivingpiecesofwhatmayoncehavebeenintendedtobeasystematicworkoncriticismacombinationofaesthetics,liter-arytheory,andmoralpsychology.PeterJonessessaybringstogetherHumessomewhatscatteredremarksonthesetopics,thusenablingustoseeandunderstandhisgeneralperspectiveonthearts,andhowitrelatestohisotherviewsabouthumanityandsociety.Becauseofthepopularityofhissix-volumeHistoryofEngland,HumehasoftenbeenidentifiedasDavidHumetheHistorian.DavidWoottonexaminesthemotivationspersonal,moral,andpoliticalthatledtoHumesmonumentalnarrativeofsocialandpoliticalcircumstance,andsuggeststhatitis,toalargeextent,HumesstoryofthedevelopmentoftheuncommonlibertyenjoyedbytheEnglish.ThelastofHumesmajorpublications,hisDialoguesconcerningNaturalReligion,waspublishedonlyin1779,threeyearsafterhisdeath.Inthefinalessayinthisvolume,J.C.A.GaskinreviewsthewholeofHumescritiqueofreligionacritiquethatisatleastimplicitinallofhisworks,andthat,weareshown,issubtle,pro-foundanddamagingtoreligioninwayswhichhavenophilosophicalantecedentsandfewsuccessors.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nAnIntroductiontoHumesThought39AnAppendixsuppliesthereaderwithtwobriefautobiographies.Humewrotethefirstofthesein1734,someyearsafterhehadbegunworkon,butstillfiveyearsbeforehepublished,theTreatise.Thesecondhewroteforty-twoyearslater,onlyafewmonthsbeforehisdeathin1776.ASelectedBibliographysupplementsthesuggestionsforfur-therreadingfoundattheendofeachofthefourteenessaysjustdescribed.50AcomprehensiveIndexcompletesthevolume.50Readersareagainremindedthat,astheworkslistedintheSelectedBibliographyshow,Humehasbeeninterpretedinwidelydifferentways.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\njohnbiro2HumesNewScienceoftheMindHumanNatureistheonlyscienceofman.T1.4.7.14ForHume,understandingtheworkingsofthemindisthekeytounderstandingeverythingelse.Thereisasense,therefore,inwhichtowriteaboutHumesphilosophyofmindistowriteaboutallofhisphilosophy.Withthatsaid,Ishallnonethelessfocushereonthosespecificdoctrinesthatbelongtowhatwetodaycallthephilosophyofmind,givenoursomewhatnarrowerconceptionofthatsubject.ItshouldalsoberememberedthatHumedescribeshisinquiryintothenatureandworkingsofthemindasascience.Thisisanimportantcluetounderstandingboththegoalsandtheresultsofthatinquiry,aswellasthemethodsHumeusesinpursuingit.Aswewillsee,thereisathreadrunningfromHumesprojectoffoundingascienceofthemindtothatoftheso-calledcognitivesciencesofthelatetwentiethcentury.Forboth,thestudyofthemindisinimportantrespectsjustlikethestudyofanyothernaturalphenomenon.WhileitwouldbeanoverstatementtosaythatHumesentireinterestliesintheconstructionofascienceinthissensehehasother,moretraditionallyphilosophical,concernsrecognizingthecentralityofthisscientificaimisessentialforunderstandinghim.11ForadiscussionoftherelationsamongHumesdifferentconcerns,seeJohnBiro,CognitiveScienceandDavidHumesScienceoftheMind,inTruth,Rationality,Cognition,andMusic,ed.K.KortaandJ.Larazabal(Dordrecht:KluwerAcademicPublishers,2004),120.40CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind41i.anewscienceofhumannatureInoneofthebest-knownpassagesinallhiswritings,intheIntro-ductiontoATreatiseofHumanNature,Humedeclareshisaimoffoundingwhathecallsanewscienceofhumannature.Hearguesthatthedevelopmentofsuchascience,basedontheexperimen-talmethodofreasoning,mustprecedeallotherinquiry,sinceonlyitcanservetogroundtherestofourknowledge:Thereisnoquestionofimportance,whosedecisionisnotcomprizdinthescienceofman;andthereisnone,whichcanbedecidedwithanycertainty,beforewebecomeacquaintedwiththatscience.Inpretendingthereforetoexplaintheprinciplesofhumannature,weineffectproposeacompleatsystemofthesciences,builtonafoundationalmostentirelynew,andtheonlyoneuponwhichtheycanstandwithanysecurity.(TIntro.6)Althoughtheprinciplesofhumannaturetobeexplainedmaylieverydeepandabstruse,thenewmethod,modeledafterthatusedwithsuchspectacularsuccessbyGalileo,Boyle,andNewtoninwhatmaybecalledthenewscienceofmatter,holdsoutthehopeofsimilarlyfar-reachingresultshere.2Themethodcallsforcarefulandexactexperiments,toenableustorenderallourprinciplesasuniversalaspossible.This,inturn,requirestracingupourexper-imentstotheutmost,andexplainingalleffectsfromthesimplestandfewestcauses(TIntro.8).Humesexpectationsforhisprojectareatoncegreatandmodest.Heseeshisnewscienceasthekeytoallothers,indeed,toallknowl-edge:HumanNatureistheonlyscienceofman.Yethismodestaimisonlytobringitalittlemoreintofashion,as,inspiteofitsimportance,ithasbeenhithertothemostneglected.Evenmorethantheothersciences,itisstillinitsinfancy:Twothousandyearswithsuchlonginterruptions,andundersuchmightydiscour-agementsareasmallspaceoftimetogiveanytolerableperfection2ThereiscontroversyamongscholarsaboutthepreciseextentandnatureofNew-tonsinfluenceonHume.Forasampling,seeNicholasCapaldi,DavidHume:TheNewtonianPhilosopher(Boston:Twayne,1975),2970;JamesNoxon,HumesPhilosophicalDevelopment(Oxford:ClarendonPress,1973),27123;andMichaelBarfoot,HumeandtheCultureofScienceinEarlyEighteenth-CenturyBritain,inOxfordStudiesintheHistoryofPhilosophy,ed.M.A.Stewart(Oxford:OxfordUniversityPress,1990),15190.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n42johnbirotothesciences;andperhapswearestillintooearlyanageoftheworldtodiscoveranyprinciples,whichwillbeartheexaminationofthelatestposterity(T1.4.7.14).Still,Humeshopeisthatthenewsciencemaydiscover,atleastinsomedegree,thesecretspringsandprinciples,bywhichthehumanmindisactuatedinitsoperations(EHU1.15).Thusweshouldsubstitutethisnewscienceforthemanychimericalsystemsspawnedbythewarmimaginationofmeta-physicians.Werethese[metaphysical]hypothesesonceremovd,wemighthopetoestablishasystemorsetofopinions,whichifnottrue(forthat,perhaps,istoomuchtobehopdfor)mightatleastbesatisfactorytothehumanmind,andmightstandthetestofthemostcriticalexamination(T1.4.7.14).Theonlysolidfoundationwecangivetothisscience...mustbelaidonexperienceandobser-vation.Thisis,ofcourse,trueofallhumanknowledge;forHume,thereisnoothersourceofknowledgebesidesexperience,andnoclaimtoknowledgebasedonanythingelseislegitimate.Wheretheapplicationoftheexperimentalmethodtomoralsubjectsmustdifferfromitsmoreestablisheduseinnaturalphilosophyisintheimpossibilityofmakingexperimentspurposely,withpremedi-tation.HereHumeisspeakingofcontrolledexperiments(typicalofthemodernlaboratorysciences)contrastingthemwithglean[ing]upourexperiments...fromacautiousobservationofhumanlife,andtak[ing]themastheyappearinthecommoncourseoftheworld,bymensbehaviourincompany,inaffairs,andintheirpleasures.Inspiteofthislimitation,thescienceofmanneednotbeinferiorincertainty,andwillbemuchsuperiorinutilitytoanyotherofhumancomprehension(TIntro.78,10).Aswehaveseen,Humeholdsthatallothersciencesdependonthisoneforwhatevercer-taintytheymayhave:itisinthissciencealonethatwecanexpectassuranceandconviction(T1.4.7.14).Suchassuranceandconvictioncannotextendtoanyclaimaboutwhytheprinciplesgoverninghumannaturearethewaytheyare:wecangivenoreasonforourmostgeneralandmostrefindprin-ciples,besideourexperienceoftheirreality(TIntro.9).Humeisadamantonthispoint.Whenheintroduceshisprinciplesofassoci-ation,resemblance,contiguity,andcauseandeffect(thethreeuni-versalprinciplesthatguidetheoperationsoftheimaginationinunitingourideas;seebelow,SectionIII),hesaysthattheirrealityCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind43requiresnospecialproofbeyondrecognizingtheireffectswhichareeverywhereconspicuous.3Bycontrast,whytheycausetheseeffectsismostlyunknown,andmustberesolvdintooriginalqual-itiesofhumannature,whichIpretendnottoexplain(T1.1.4.1,6).Itwouldbeamistake,however,tocomplainaboutthisimpossi-bilityofexplainingultimateprinciplesinthescienceofman,asitisadefectcommontoitwithallthesciences,andallthearts,inwhichwecanemployourselves(TIntro.10).Withrespecttonat-uralphænomena,allwecandoistoresolvethemanyparticulareffectsintoafewgeneralcauses,bymeansofreasoningsfromanal-ogy,experience,andobservation.Butastothecausesofthesegeneralcauses,weshouldinvainattempttheirdiscovery....Theseultimatespringsandprinciplesaretotallyshutupfromhumancuriosityandenquiry(EHU4.12).Suchskepticismaboutthepossibilityofulti-matemetaphysicalexplanationsneednot,however,makeusskepti-calaboutthepossibilityofascienceofthemindthatcontentsitselfwithacarefuldescription,basedonobservation,oftheoperationoftheseprinciples.Humeis,aswehaveseen,fullyexplicitaboutthenatureandsta-tusoftheprojecthewantstoundertake.Yethisdeclarationshavehadremarkablylittleeffectontheinterpretationofthatprojectbychampionsandcriticsalike,fromhisdaytoours.Itisonlyrecentlythathehasbeguntobeseenasengagedinaninquiryatleastcontinuouswithwhatwethinkofasthescientificstudyofthemind.Philosophersofmindtodayoftenseethemselvesasbeingsoengaged,asparticipatinginaninterdisciplinaryinquirytheyarehappytolabelcognitivescience.ButitisanironythatwouldnotbelostonHumethatsomehaveexplicitlycontrastedthatinquirywithhisscienceofman,ratherthanrecognizingitasthelattersdescendant.43Humeregardedhisexplanationofalltheoperationsofthemindbymeansofthethreeprinciplesofassociation(resemblance,contiguity,andcausation)ashismostnoteworthydiscovery,andfamouslydescribedtheseprinciplesasbeingtousthecementoftheuniverse(A35).4JerryA.Fodor,MentalRepresentation:AnIntroduction,inScientificInquiryinPhilosophicalPerspective,ed.N.Rescher(Lanham,MD:UniversityPressofAmerica,1987).MorerecentlyFodorhashadachangeofheart:Humesaccountofthemind...seems,inanumberofrespects,toanticipatetheonethatinformscurrentworkincognitivescience;seehisHumeVariations(Oxford:ClarendonPress,2003),2.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n44johnbiroThisisnottosay,ofcourse,thattherearenodifferencesbetweenthetwoprojects,separatedastheyarebytwoandahalfcenturiesduringwhichbothscienceandphilosophyhavechangedinimpor-tantrespects.ThepassagesIhavecitedfromHumesannouncementofhisnewscienceshouldalertustosomefeaturesofitthatcontrastsharplywiththoseofitstwentieth-centuryoffspring.WhileHumeanticipatedmanyofthedifficultiesandproblemsrecentlydiscov-ered,thesolutionsheoffersor,atleast,hintsatareoftendiffer-entand,Ishallsuggest,comparefavorablywiththosecurrentlyonoffer.OnereasonforthisisthatthemethodhesoclearlyoutlinesintheIntroductiontotheTreatiseismoresuitedtothesubjectmatterofthenewsciencethanisonemodeledonthatoftheso-calledhardsciencesandfavoredbymanyofhismodernsuccessors.(Seebelow,SectionsIIIandVI.)ii.skepticorscientist?Fortwocenturiesafteritsappearance,Humesphilosophywascom-monlyseenasessentially,perhapsentirely,negative.Hisinquirieswerethoughttohavebeenundertakeninaspiritofskepticismandasaimingtoshowhowfarthatoutlookcanandmustbecarriedifsomeseeminglycompellingempiricistprincipleswerefollowedouttotheirinevitableconsequences.ThebarrageofskepticalargumentsinthefirstbookoftheTreatisewereseennotasdirectedagainstvariousphilosophicalaccountsofourknowledgeoftheworldandofourselves(astheyare,increasingly,today),butagainsttheverypos-sibilityofsuchknowledge.ThatsuchskepticismisonthefaceofitincompatiblewiththeprojectHumeannouncedintheIntroductiontotheworkwaseithernotnoticedordismissedasunproblematicbythesimpleexpedientofnottakinghimathisword.Therewas,tobesure,somereasonforsucharesponse.AfterDescartes,epistemologicalquestionsmovedtocenterstageinphi-losophy,andepistemologycametobeseenasprimarilyconsistingincomingtotermswith,inonewayoranother,thekindofskep-ticalthreatposedintheopeningpagesoftheMeditations.ItwasnaturalforHumescontemporariesandforlaterphilosophersuntilfairlyrecentlytoseehimasstrugglingwiththesameproblemsthatpreoccupiedthemandasrespondingtohispredecessorstreatmentCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind45ofthem.Hisextensiveanddevastatingcriticismsofattemptstodealwiththeskepticalthreatbyrelyingeitheronreasontodiscovertruthsabouttheworldapriorioronexperiencetoconveythosetruthstousthroughperceptionseemedtothesecontemporaries,andtomanysince,evidencethathesharedtheirpreoccupationwiththatthreat.YetHumerepeatedlydisclaimssuchaninterestandtellsusclearly,inavarietyofcontextsandways,thatthemainaimofhisenquiriesissomethingquitedifferent.Anexampleishisadmoni-tion,intheopeningparagraphofthesectionintheTreatiseentitledOfscepticismwithregardtothesenses,nottobeconcernedwiththeusualskepticalquestionabouttheexistenceoftheexternalworld:Wemaywellask,Whatcausesinduceustobelieveintheexis-tenceofbody?buttisinvaintoask,Whethertherebebodyornot?Thatisapoint,whichwemusttakeforgrantedinallourreason-ings.Thatsuchaninjunctionshouldappearinthisverysection,nominallyconcernedwithskepticism,issurelynotanaccidentandshouldclinchthecasethat,whateverHumeisdoing,itisneitherpressingnorlookingfor(andfailingtofind)ananswertotheusualskepticalchallenges.Hetellsusexplicitlywhatheisdoing:Thesubject,then,ofourpresentenquiryisconcerningthecauseswhichinduceustobelieveintheexistenceofbody(T1.4.2.2).Asnotedearlier,Humeisskepticalaboutvariousphilosophicalattemptsatjustifyingourbeliefs,especiallywhenitcomestothemostbasicofthese,suchasthebeliefinbodies,intheidentityofourperson,andincausalconnections,beliefsthatevenaskepticcannotseriouslyrejectorlivewithout.Itis,heofteninsists,justaswellthatnaturehasmadesurethat,inspiteofallphilosophy,wetaketheseforgranted,aswithoutthemhumannaturemustimmediatelyperishandgotoruin(T1.4.4.1).Butthisrecognitionofourunreflective,instinctive,andunavoidableacceptanceofcertainbasicbeliefsmustnotbeconfusedwithclaimingtohaveaphilosophicaljustificationofthosebeliefs.Humesskepticalargumentstargetattemptstopro-videsuchjustifications,basedoneitherreasonorexperience:Tisimpossibleuponanysystemtodefendeitherourunderstandingorsenses.Therecognitionofthisfactisskepticismofasortamal-ady,whichcanneverberadicallycurdbutitisdifferentfromtheusualkindofskepticisminacrucialrespect:ourinabilitytojustifyabeliefofthisfundamentalsortisnotseenasareasonforwithholdingCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n46johnbiroassenttoit.Therecannotbeareasontodosomethingwecannotdo.Asforthemalady,Carelessnessandin-attentionalonecanaffordusanyremedy(T1.4.2.57).Inthecrucialcaseofpersonalidentity(tobediscussedatgreaterlengthbelow),5Humemakesitequallyclearthathisinterestliesinexamininghowonecomestoformonesbeliefinonesidentityandinwhataccountsforonesconfidenceinthatbelief,ratherthaninaphilosophicaljustificationofthebelief.Afterdismissingtheclaimofphilosophersthatweareeverymomentintimatelycon-sciousofwhatwecallourself,heasks:Whatthengivesussogreatapropensiontoascribeanidentitytothesesuccessivepercep-tions,andtosupposeourselvespossestofaninvariableanduninter-ruptedexistencethrothewholecourseofourlives?(T1.4.6.1,5).Itshouldbeclearthathispowerfulnegativeargumentsaregivennotintheserviceofapurelyskepticalconclusion,butasaneces-sarypreliminarytorefocusingourattentionongivingananswertotheseotherquestionsinthemannerofadescriptiveandexplanatoryscience.6ThuswhilethereisasenseinwhichHumecanbesaidtobeaskeptic,hisskepticismisbetterunderstoodasoneaboutpretendedsupra-scientificmetaphysicalknowledge,ratherthanaboutscien-tificknowledgeitself.Itisthisthatseparateshimfromthosewhoconceiveofphilosophyasgoingbeyondmerescientificknowledgetogiveusadeeperandmorecertainknowledgeofreality.Hume,bycontrast,thinksofexplanationinathoroughlyscientificspirit.Thetotalalterationorrevolutionheclaimshisnewsciencebringstotheintellectualsceneconsistsinbecomingwhathecallsananatomistofhumannature.Aswesawearlier,Humebelievesthatifweanat-omizehumannatureinaregularmanner,wecandiscoverthelawsgoverningthemind(A2).Thesemaybehiddenorsecretuntiltherightmethodisbroughttobear,but,unlikethesupposedlydeepertruthsofthemetaphysician,theyarenotoccult,destinedforevertoeludediscovery.5SeealsointhisvolumetheessayHumeandtheProblemofPersonalIdentity.6Ontherelationbetweendescriptionandexplanation,seeJoaoPauloMonteiro,HumesConceptionofScience,JournaloftheHistoryofPhilosophy19(1981):32742,andJohnBiro,DescriptionandExplanationinHumesScienceofMan,TransactionsoftheFifthInternationalCongressontheEnlightenment(Oxford:VoltaireFoundation,1979),44957.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind47Thisshiftoffocusfromavainattempttogiveaphilosophicaljus-tificationofourfundamentalbeliefstoascientificaccountoftheiroriginintheoperationsofourminds,iswhatHume,withdeliber-ateparadox,callsaScepticalSolutiontotheskepticalchallenge(EHU5).ThequestionssuchascientificaccountmustanswerareHowdoweformourbeliefs?Howdowemovefromonebelieftoanother?andWhatmechanismsandprinciplesunderlieandgov-ernsuchprocesses?ThesearethequestionstowhichHumealwaysturnsassoonashehasdiscreditedtheclaimsofthosewhothinkthattheyknowwhatentitlesus(bysomenonimmanent,externalstandard)toholdthebeliefsweinfact,andinevitably,dohold.iii.ananatomyofthemindTherawmaterialsfromwhichallmentallifeisconstructedareimpressionsandtheircopies,ideas,bothspeciesofthegenus,per-ception:Alltheperceptionsofthehumanmindresolvethemselvesintotwodistinctkinds,whichIshallcallimpressionsandideas.WhatHumecallshisfirstprinciple(sometimesalsolabeledthecopyprinciple)statesthatalloursimpleideasintheirfirstappear-ancearederivdfromsimpleimpressions,whicharecorrespondenttothem,andwhichtheyexactlyrepresent(T1.1.1.1,7,12).7Manyofthemostskeptical-soundingpassagesoftheTreatiseandtheEnquiryconcerningHumanUnderstandingaredevotedtoshowingthatourstockofthesematerialsismorelimitedthanphilosophershavesupposed.Humeshowsusagainandagainthattheimpressionsfromwhichsomeputativeideapositedbythemetaphysicianwouldhavetoderivearejustnottobefoundinexperience.Buthedoesnotdenytheobvious,andremarkable,factthatfromtheratherlim-itedstockofimpressionsthatcomemyway,Iamabletoconstructanedificeofbeliefsthatgofarbeyondthoseimpressionsandtheideastraceabletothem.First,mycomplexideasarenotconfinedtothecompleximpressionsIhavehad:Icancombinebothsimpleandcompleximpressionsinnovelways,thusformingnewcomplexideas.These,oftencalledbyHumefictions,mayormaynotgiverisetobelief,andoftentheydonot,aswithfictionalideasinthe7Formoreonimpressionsandideasandrelatedtopics,seealsointhisvolumetheessayHumeandtheMechanicsofMind.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n48johnbirousualsense(suchasthoseofEldoradoorSantaClaus).Second,thecourseofmyexperience,thevariousregularitiesamongthepercep-tionsthatmakeitup,isexploitedbythemindinformingbeliefs.Theideaofarockflyingtowardawindowwillgiverisetotheideaofthewindowsshattering,giventheconstantconjunctionoftheseeventsinonesexperience.Animpressionoftherockflyingtowardthewindowwillgiverisetothebeliefthatthewindowwillbreak,which,ofcourse,involvestheideaofthewindowsbreaking,butalsosomethingmore.Thatsomethingmorecomesfromtherela-tionoftheideatoapresentimpression;beingsorelatediswhatmakesanideaintoabelief(T1.3.7.5).Inboththeseways,themindmustbeconceivedasessentiallyactive.Itiswhatitdoeswithwhatitgetsthatmatters,anditisthisactivitythatHumesscienceaimstodescribe.Humesanswertoquestionsabouthowwecometohavethebeliefswehaveisthattheyare,oneandall,theproductofanon-rationalfaculty.Thisfaculty,whichheusuallylabelstheimagina-tion(andsometimesdescribesasaninstinct),producesahabitorcustom,whichconsistsinatendencytomovefromanideaonehastoanotheridealinkedtothefirstideabyoneoftheprinciplesofasso-ciation(resemblance,contiguity,orcauseandeffect).Hecontraststhisfacultywithreasonortheunderstanding,thefacultywhosestandardsandoperationssomephilosophersthinkcanservetopro-videananswertotheskepticschallenge.Thisdistinctionbetweenreason(understoodinoneway,though,aswewillsee,notinanother)andtheimaginationisfundamentaltoHumesanatomyofthemind.Inonesense,reasonisthereflectivefacultyweemploywhenwemakeacomparisonbetweentwoideasconcerningtheirsimilarityordifferenceinsomerespectandwhenweengageinchainsofdemon-strativereasoningconcerningthesesimilaritiesanddifferences.Theserelationsandthusthiskindofreasoningdependexclusivelyontheideascompared:changetheideaandyouchangetherelation.Theideaofathingmaybringtomindotherthingsthoughtofassim-ilar,adjacent,orcausallyconnectedtoit.However,relationsinthissensearemerelyphilosophical.Wecanaskofanytwoobjects,whatisthedistanceor,moregenerally,thedifference,betweenthem,andinsodoing,besaidtobeaskinghowtheyarerelated.Butiftheyareveryfar,orwhollydifferent,fromeachother,thenat-uralthingtosayisthattheyarenotrelated.Humecontrastssuchrelationswithrelationsinamorecommonsense,whichhecallsCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind49naturalrelations.Thesewethinkofasconnectingtwoideasinsuchawaythattheonenaturallyintroducestheother(T1.1.5.1).Theprinciplesofassociationinvolvesuchnaturalrelations.(TheshadeofblueInowseedoesnotmakemethinkofalltheothersIhaveexperienced,butitdoesbringtomindthingsexactly,orverysimilarly,colored.)Theimaginationisthenonreflectivefacultythattakesusfromanideaoranimpressiontoanotherideaconnectedtoitbyoneofthesenaturalrelations.Toquoteagainfromhisdis-cussionofourbeliefinexternalobjects:ourreasonneitherdoes,norisitpossibleitevershoud,uponanysupposition,giveusanassuranceofthecontinudanddistinctexistenceofbody.Thatopin-ionmustbeentirelyowingtotheimagination;whichmustnowbethesubjectofourenquiry(T1.4.2.14).Noristhisparticularkindofbeliefuniqueinthisrespect.Ingeneral,beliefismoreproperlyanactofthesensitive,thanofthecogitativepartofournatures(T1.4.1.8).Ourmostgeneralandmostfundamentalbeliefs(suchasthoseinourownidentity,intheexistenceofanexternalworld,andincausalrelations)areimpervioustotheinfluenceofreason,whichcanneithergroundnordestroythem.Herecogitativepart,andthusreason,refertoourfacultyoftheoreticalreasoning,atworkwhenweconstructdemonstrationsandphilosophicalarguments.Thereis,however,anothersenseofreasoning,applicabletosomeofthenaturalandinstinctivetransitionswemakefromoneperceptiontoanother,fromperceptiontobelief,andthusfromonebelieftoanother.Weareengaged,forexample,inreasoningwhenwemakeacausalinference;indeed,thatiswhatweprimarilymeanbyreason-inginordinary,nontheoretical,contexts:thisinferenceisnotonlyatruespeciesofreasoning,butthestrongestofallothers(T1.3.7n.20).Humecallsthiskindofreasoningexperimentalorproba-blereasoningandinsiststhatweshareitwithinfants,nayevenbrutebeastswhopresumablydonotengageintheoreticalreason-ing.Itisthiskindofreasoningonwhichthewholeconductoflifedepends,[andit]isnothingbutaspeciesofinstinctormechanicalpower,thatactsinusunknowntoourselves;andinitschiefopera-tions,isnotdirectedbyanysuchrelationsorcomparisonsofideas,asaretheproperobjectsofourintellectualfaculties(EHU4.23,9.6).88IntheconcludingparagraphofTreatise1.3.16,Ofthereasonofanimals,Humemakesasimilarclaim:Toconsiderthematteraright,reasonisnothingbutawonderfulandunintelligibleinstinctinoursouls,whichcarriesusalongacertainCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n50johnbiroReasoningofthislattersortisthebusinessoftheimagination,theever-active,sometimesoveractive,nonrationalfacultywhosework-ingsHumesnewsciencesetsouttodescribe.Accordingtothatsci-ence,themindmovesfromoneideatoanotherbywayofthethreeprinciplesofassociationjustmentioned.Responsibleforthesetran-sitionsarecustomorhabitand,mostimportantly,themindsten-dencytocontinueanymotionorprocessonceithasbegunit.ThisisafundamentalpropertyoftheimaginationthatplaysaroleinHumesexplanationsofsomeofthemostremarkablefactsaboutthemind.Theseincludethefactthatintheabsenceofcorrespond-ingimpressionsfromwhichtheideascouldhavebeencopied,onenonethelesscomestobelievethattherearebodiesandthatoneisthesamepersonatonetimeasatanother,andeventhatonecanextend[ones]identitybeyond[ones]memory(T1.4.6.20).InthepivotalsectionOfscepticismwithregardtothesenses,whereheundertakeshisenquiryconcerningthecauseswhichinduceustobelieveintheexistenceofbody(thelegitimatelysci-entificinquiryhedistinguishesfromthevainattempttoanswertheskeptic),Humeremindsusofhisearlierexplanationoftheheargues,mistakenbeliefintheinfinitedivisibilityofspaceandtimeintermsofthisnaturaltendency,whichcanappeartotakeusbeyondwhatisgiveninexperience:Ihavealreadyobservd,inexaminingthefoundationofmathematics,thattheimagination,whensetintoanytrainofthinking,isapttocontinue,evenwhenitsobjectfailsit,andlikeagalleyputinmotionbytheoars,carriesonitscoursewith-outanynewimpulse(T1.4.2.2,22).Thistendency,automaticandnonreflective,iseverywhere:nothing,saysHume,ismoreusual,thanforthemindtoproceedafterthismannerwithanyaction,evenafterthereasonhasceasd,whichfirstdetermindittobegin.Inthesectionsonourideasofspaceandtime,thistendencywasalsoappealedtoinexplaininghowwegenerateanimaginarystandardofequality,ofcorrection[s]beyondwhatwehaveinstrumentsandarttomake,andofnotionsofperfectionbeyondwhat[our]trainofideas,andendowsthemwithparticularqualities,accordingtotheirpartic-ularsituationsandrelations.Thisinstinct,tistrue,arisesfrompastobservationandexperience;butcananyonegivetheultimatereason,whypastexperienceandobservationproducessuchaneffect,anymorethanwhynaturealoneshoudproduceit?CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind51facultiescanjudgeof.This,inturn,enabledHumetoaccountforthefictions,uselessandincomprehensible,ofthemathe-maticianswhoclaimtogiveexactdefinitionsanddemonstrations(T1.2.4.24,29).Humespurposeinthatearlierdiscussionwastoexposethesefictionsasabsurd(T1.2.4.2930,32).Hisrecommendationtherewastoresistthistendencyofthemindandthusavoidtheabsurdity.Inthediscussionofourbeliefinbody,thesametendencyisinvokedintheinterestofadifferentgoal:thatofexplaininghownaturally,evenunavoidably,weformouropinionofthecontinudexistenceofbody.Hewrites:Objectshaveacertaincoherenceevenastheyappeartooursenses;butthiscoherenceismuchgreaterandmoreuniform,ifwesupposetheobjectstohaveacontinudexistence;andasthemindisonceinthetrainofobservinganuniformityamongobjects,itnaturallycontinues,tillitrenderstheuniformityascompleataspossible(T1.4.2.22).9Humedistinguishesbetweenprincipleswhicharepermanent,irresistible,anduniversalandthosewhicharechangeable,weak,andirregular(T1.4.4.1).ThisdistinctionisessentialtothedoubleusemadebyHumeofthistendencyofthemind.Whenthetendencyisguidedbyprinciplesofthefirstsort,asitisintheformationofourfundamentalcommon-sensebeliefs,wehaveaskepticalsolutiontotheskepticsdoubts,whetherthatdoubtbeabouttheexternalworldoraboutpersonalidentity.WhileHumesometimesusesthetermfictiontodenoteafundamentalnaturalbeliefproducedbythispropertyofthemind,wemustbecarefulnottobemisledbythisintothinkingofsuchabeliefassomehowfancifulandarbitrary.Fictionsofthissortarenotoptional:theyareforcedonusbyournature.Distinguishingsuchfictionsfromthoseresultingfromphilosophicalspeculationfloatingfreeofcommonsenseisalarge,arguably,thecentral,partoftheoverallaimofHumesphilosophy.Inthesecasesofwhatwemaycallnaturalfictions,themindsinertialtendencyoperatesinsuchaninsensiblemannerasnever9Foraclassicdiscussionofhowweformourbeliefintheexistenceofbodies,seeH.H.Price,HumesTheoryoftheExternalWorld(Oxford:ClarendonPress,1940).Forabrieferaccount,seeDavidFateNorton,EditorsIntroduction,inHumesTreatiseofHumanNature,OxfordPhilosophicalTextsEdition(Oxford:OxfordUniversityPress,2000),I3842.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n52johnbirotobetakennoticeof,andtheimaginationcandrawinferencesfrompastexperience,withoutreflectingonit;muchmorewithoutforminganyprincipleconcerningit,orreasoninguponthatprinci-ple.Humeaddsthatthistendencymayeveninsomemeasurebeunknowntous.Itisimportanttoseethatbythishemeansonlythatwehavenodirect,introspective,accesstotheprocessesinques-tion.Inmakingcausalinferences,forexample,weobviouslydonotconsciouslyrecallthepreviousinstancesofconstantconjunctionsonwhichtheinferenceisbased:Thecustomoperatesbeforewehavetimeforreflection,andIneveramconsciousofanysuchoperation,sothatindecidingtogivethepreferencetoonesetofargumentsaboveanother,Idonothingbutdecidefrommyfeel-ingconcerningthesuperiorityoftheirinfluence(emphasisadded).Thusitisthatallprobablereasoningisnothingbutaspeciesofsensation(T1.3.8.13,10,12).OfthethreerelationsHumecallsnaturalrelations,causeandeffectistheonlyonebywhichwecangobeyondwhatisimme-diatelypresenttothesenses,sinceitistheonlyone,thatcanbetracdbeyondoursenses,andinformsusofexistencesandobjects,whichwedonotseeorfeel(T1.3.2.3).Thecausalinferencesthatthustakeusbeyondourpresentimpressionsare,forHume,indeedaformofreasoning,eventhoughtheyare,aswehaveseen,auto-maticandnonreflective.Thissortofreasoningisstrongerthanthatwhichinvolvestheseparatingorunitingofdifferentideasbytheinterpositionofothers,whichshowtherelationtheybeartoeachother(T1.3.7).Thelatter,whileconstitutingdemonstra-tions,astheformerdonot,dosoonlywithrespecttophilosophicalrelations.ItisappealstoreasoningofthedemonstrativesorttoestablishfactsabouttheworldthatHumesskepticalargumentsareintendedtoshowtobefutile.Causalreasoning,bycontrast,hasthepowertoyieldbelief.Thedifferencebetweenmereideaandbeliefiseasytoknowbutdifficulttoexplain.WhileitisHumesofficialandoft-expressedviewthatabeliefisnothingbutalivelyidea(T1.3.5,forexample),thematterisrathermorecomplicated.Whenhereflects,intheAppendixtoBook3oftheTreatise,onhisearlierdoctrinethatperceptionscandifferonlyintheirrespectivedegreesofforceandvivacity,herealizesthatthereareotherdifferencesamongCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind53ideas,whichcannotproperlybecomprehendedundertheseterms.HadIsaid,thattwoideasofthesameobjectcanonlybediffer-entbytheirdifferentfeeling,Ishoudhavebeennearerthetruth(TApp.22).10Yet,aswehavejustseen,thespecialfeelingthatmarksoutbeliefthatcertainje-ne-sais-quoi,ofwhichtisimpossibletogiveanydefinitionordescription,butwhicheveryonesufficientlyunder-stands(T1.3.8.16)isnotaltogetherinvoluntaryandbeyondratio-nalcontrol.Thegreatdifference,Humesays,betweenapoeticalenthusiasm,andaseriousconviction...proceedsinsomemeasurefromreflectionandgeneralrules....Alikereflectionongeneralruleskeepsusfromaugmentingourbeliefuponeveryencreaseoftheforceandvivacityofourideas....Tisthustheunderstandingcorrectstheappearancesofthesenses(T1.3.10.1112).Thusnotalltransitionsbetweenideasgroundedinnaturalrelationsdeservethetitlereasoninginthemorehonorificsense.Wemustremem-berthedistinctionbetweenthetwosortsofprinciplethatguidetheimagination,thosepermanent,irresistible,anduniversalandthosechangeable,weak,andirregular.Onlytheformer,andtheyonlywhentemperedbytheproperuseofgeneralrules,aretobereliedoninformingbeliefs(T1.3.13.719,1.3.15.210).Thissec-ondinfluenceofgeneralrulesistheworkofourjudgmentandunderstanding,anditmustbedistinguishedfromthefirst,whichinvolvestheirrashusebytheuncontrolledimaginationorfancy,aswhenwegeneralizehastilyorfromtoosmallasample.ThisfirstinfluenceofgeneralrulesiswhatisatworkinthekindofinertialextrapolationtoanimaginarystandardinmathematicalreasoningthatwesawHumecriticize.ItisalsowhathappenswhenwedeclarethatanIrishmancannothavewit,andaFrenchmancannothavesolidityonthebasisofafewexampleswehaveencountered;andinthisusetheyarethesourceofwhatweproperlycallprejudice(T1.3.13.12,7).Thesecondinfluenceofgeneralrulesconsistsinapplyingsecond-orderprinciplestoobservedfirst-orderregularities,suchastheprincipleDonotjudgeonthebasisofsmallsamples.Suchprinciplesare,ofcourse,themselvesbasedonexperience:in10TheTreatisewasfirstpublishedinthreeseparatevolumes,thefirsttwoin1739,thethirdin1740.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n54johnbirothiscase,ontheexperienceofmistakenjudgmentsresultingfromneglectingtopayattentiontothem.11This,inbrief,isHumesaccountofhowwecomebyourbeliefs,anaccountintendedtoreplacefutileattemptstogivephilosophicaljustificationsofthem.ProvidingsuchanaccountiswhatHumemeanswhenhesayshehasgivenaskepticalsolutiontotheskepticschallenge(EHUSect.5).Itistogiveupbeingametaphysicianandtobecomeascientistananatomistofthemind,ofhumannature.Thisrecommendationbearsastrikingresemblancetothenaturalizingprogramscommoninrecentphilosophyofmindandepistemology.Here,too,theleadingideaistoabandonanapriorimethodperceivedasbankruptinfavorofanempiricalonethatholdsoutthepromiseofgenuineprogress.Manyepistemologistshave,inrecentyears,cometofeelthatthetime-honoredmethodofarmchairconceptualanalysisisunlikelytotellusmuchaboutthenatureofknowledge.12Philosophersofmind,too,interestedinunderstandingreasoning,perception,memory,language,andahostofothermentalphenomena,increasinglylooktothenewdiscipline(orconstellationofdisciplines),cognitivescience,ratherthantotraditionalmethodsofphilosophicalanalysisandargument.13Oneofthemoststrikingfindingsofmoderncognitivesciencehasbeenthatthemechanismsandprocessesinvolved,whetherinper-ception,inlinguisticprocessing,orinreasoninggenerally,arewhat11Forhelpfuldiscussionsofgeneralrules,seeThomasK.Hearn,GeneralRulesinHumesTreatise,JournaloftheHistoryofPhilosophy8(1970):40522;andGen-eralRulesandtheMoralSentimentsinHumesTreatise,ReviewofMetaphysics30(1976):5772.12Theuninspiringhistoryoftheso-calledGettierproblem,involvingincreasinglyarcaneandartificialcounterexamplestoevermorebyzantinedefinitionsofknowl-edge,isoftentakentobeproofofthis.Fordetailsofthishistory,seeRobertK.Shope,TheAnalysisofKnowing:ADecadeofResearch(Princeton,NJ:PrincetonUniversityPress,1983).13Thefountainheadofso-callednaturalizingprogramsinepistemologyandthephi-losophyofmindisW.O.Quine.SeeWordandObject(Cambridge,MA:Technol-ogyPressoftheMassachusettsInstituteofTechnology,1960);andEpistemologyNaturalized,inOntologicalRelativity,andOtherEssays(NewYork:ColumbiaUniversityPress,1969),6990.Forrecentexamples,seeAlvinGoldman,Epis-temologyandCognition(Cambridge,MA:HarvardUniversityPress,1986),andStephenStich,TheFragmentationofReason:PrefacetoaPragmaticTheoryofCognitiveEvaluation(Cambridge,MA:MITPress,1990).Forausefulvolumeofpapersonthesubject,seeNaturalizingEpistemology,ed.H.Kornblith(Cambridge,MA:MITPress,1985).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind55itcallssubdoxastic,modular,andautomatic.Theyaresubdoxasticbecausetheyusuallytakeplacebelowthethresholdofconscious-ness,hence,belowthelevelofbelief.Theyaremodular,becausetheyare,intheoverwhelmingmajorityofcases,task-specific;theydotheirworklargelyinisolationfromeachother,aswellasfromthecognitivestateswewouldattributetothepersonasawhole.Theprocessesunderlyingoneparticularkindofcognitivecapacityorperformancedonotinteractwiththoseinvolvedinothers,andtheirrespectiveoutputsaresimilarlyindependent.(Thinkofthecommoncaseofdifferentsensesdeliveringdifferentverdictsonthepropertiesofoneandthesameobjectoreventasonthestick,half-immersedinwaterseentobebent,yetfelttobestraight.)Theyarealsoinsen-sitivetothethinkersbeliefs,evenifthesearereflectiveandcon-scious,ratherthanmerelytacit,andevenifhemakesanefforttobringthemtobear.(Thinkoftherobustnessofperceptualillusionsknowntobesuchparallelrailroadtracksthatseemtoconvergeastheyrecede.)Finally,theyareautomaticbecausetheyarenotunderthethinkerscontrol.Theyarenotreallyoperationsandprocessesofthethinkersmind,butofsomecomponentsubsystemwecon-strueastheiragent.Asaresultofthesefeatures,thethinkerisnotareliablesourceofinformationabouttheseprocesses.Hencethepreferenceforlaboratoryexperimentsoverarmchairintrospec-tion.(CompareHumespreference,notedearlier,forobservationofthecommoncourseoftheworldoverexperiments[made]withpremeditation.)IhavealreadyremarkedonHumesrecognitionofourtendencytoovergeneralize.Thesamesortofinductiveovergeneralization,sometimesbenign,sometimesnot,hasbeenfoundtobeubiquitousbyrecentempiricalstudiesofourcognitiveprocesses.Weseeitinlanguagelearning,andinvariouskindsofprocessing,phonological,morphological,evensyntacticas,forexample,withgarden-pathsentences,whereweleapaheadtocompleteasentenceinthewrongway.14Weseeitinperception,forexample,inthedetectionoftheedgesandboundariesofobjectsandintheperceptionofthemove-mentofrigidbodies.Weseeitinproblemsolvingandreasoninggenerally,asinourlamentabletendencytomakeclearlyfallacious14SuchsentencesasThehorseracedpastthebarnfell,orThefloristsenttheflowerswasverypleased.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n56johnbiroprobabilisticinferences.Whatalltheseinstancesofthetendencyhaveincommonisthatthemeaningsweassign,thebeliefsweform,andtheinferenceswedraw,whileoftenfaroutrunningtheevidenceavailabletousand,inmanycases,recognizedasdoingso,arenonethelessallbutirresistible.Hencethecommoncharacter-izationofmanyoftheprocessespositedtoexplainourcognitivecapacitiesandperformanceascognitivelyimpenetrableorinfor-mationallyencapsulated.15Thisrecognition,commontoHumeandtomoderncognitivesci-entists,ofthesefeaturesofourcognitivemake-upraisessomedeepmethodologicalissues.Whattheseare,andhowHumesdistinctiveresponsetothemdiffersfromthoseofthelatter,willhelpusseethefullcomplexityofhisapproachtothetaskhehassethimself,aswellasthesourceofsomeofthetensionswemaydetectinit.Butbeforeturningtothesematters,wemustlookinsomedetailatthetopiconwhichdiscussionofHumesphilosophyofmindhastraditionallycentered,thatofpersonalidentity.Here,too,thecontinuitywithsomecentralconcernsofmoderncognitivesciencewillbestriking.Afterhighlightingthese,wewillbeinabetterpositiontoreflectonsomedifferencesbetweenHumesapproachandthatofrecentcognitivescience.iv.personalidentityOneofthecentralphilosophicalquestionsaboutthemindhasalwaysbeenanontologicalone:Whatisit?ButHumeseschewalofspeculativemetaphysicsleadshimtosubstituteforthisquestionotherstowhichhethinkstherearecleareranswers.First,whatkindofthingismybeliefthatIamanidenticalthingovertime,aself,about?Second,howdoIacquiremyideaofsuchathing?Third,howdoIcometobelievethatIamsuchathing?Itisimportantthatthesequestionsbeapproachedfromthefirst-personpointofview,render-ingirrelevanttheeasyanswerthattobelieveinonesidentityisto15DetailsaboutthesemattersmaybefoundinShimonUllman,TheInterpretationofVisualMotion(Cambridge,MA:MITPress,1979);JudgmentunderUncertainty:HeuristicsandBiases,ed.D.Kahneman,P.Slovic,andA.Tversky(Cambridge,MA:CambridgeUniversityPress,1982);andZenonW.Pylyshyn,ComputationandCognition:TowardaFoundationforCognitiveScience(Cambridge,MA:MITPress,1984).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind57believethatoneis,orisatleastassociatedwith,abodythatisthesameovertime.Humeisinterestedintheconceptofidentityingen-eralandinhowwecometoattributeidentitytobodies.Hisaccountofwhatisinvolvedinsuchattributions,however,cannotbeusedtoexplainhowwecometoattributeidentitytoourselves,sincethataccountappealstoactivitiessuchasrememberingandassociatingandinsodoingpresupposestheidentityofamindwhoseactivitiestheyare.Fortalkofaminddoingthingstomakesense,theremustbeatemporallyextendedthingofsomesortdenotedbythetermmind(orthepronounI),onetowhichthepredicatesameattime2asattimecanapply.16Howdoesonecometobelieve,though,thatone1issuchathing?AsHumerecognizes,thefundamentalbeliefstand-inginneedofananalysisandageneticaccountisthebeliefonehasinonesownidentity(T1.4.2.56).Consequently,Humegivesanaccountofwhatonebelieveswhenonehasthatbeliefandofhowonecouldcomebyabeliefofthatsort.Thefactthatinthisaccounthemustmakeuseofthemoregeneralconceptsthatthemselvespresupposehavingsuchabeliefcanobscurethisall-importantpoint,ascan,onoccasion,Humeslan-guage.Butwhenheaskswhetherinpronouncingconcerningtheidentityofaperson,weobservesomerealbondamonghispercep-tions,hemustbetakentobetalkingaboutapersonspronouncingonhisownidentityonthebasisofobservinghisownperceptions(T1.4.6.16).Thereisnoobservinganothersperceptions,asthereisanothersbody.So,forananswertothequestionthatmustbemostbasicHowdoIcometothinkofmyselfasaself?Imustturninward.Imustlooktoseewhatthereisinmyexperiencethatleadsmetothinkofmyselfasthesamepersonormindover16Humesgeneralaccountofidentityinvolvesthinkingofthemindassurveyingobjectsandtrac[ing]thesuccessionoftimesomethingthatseemstopresupposeitsidentityovertime(T1.4.2.29).So,itmayseem,doestalkingofonesbelievingandreidentifying,asIjusthave.ThismayseemtobegthequestionagainsttheveryskepticHumeisoftentakentobe,namely,onewhodoubtsthatthereisaself.Butdoessuchtalknotalreadyimply,byvirtueofitsgrammaralone,thatthereisone?Onemaybetemptedtosaythatiftheskepticwereright,hecouldnotstatehisview.Howeverthismaybe,Humesdiscussionisbetterseen,asIsuggestedabove,asbeingaboutwhatkindofthingtheselfis,ratherthanoneaboutwhetherthereisone.Forfurtherdiscussion,seeJohnBiro,HumeonSelf-IdentityandMemory,ReviewofMetaphysics30(1976):1938,andHumesDifficultieswiththeSelf,HumeStudies5(1979):4554.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n58johnbirotime.ThatIdothinkofmyselfisadatum,afact,onethatHumeisseekingtoexplainwithinthenewscientificframeworkhehasadopted.17Humestreatmentoftheideaoftheselfhastwoparts.Theneg-ativeoneconsistsintheargumenthegivesagainstthewidelyheld(amongphilosophers,atleast)beliefthatonehasanideaofoneselfasasimplesubstance,onethatenduresessentiallyunchangedthroughaccidentalchanges,especiallythroughoneschangingperceptions.Theargumentdependsonthecopyprinciple:fromwhatimpres-sioncoudthisideabederivd?Havingfoundnosuchimpression(whenIentermostintimatelyintowhatIcallmyself,Ialwaysstumbleonsomeparticularperceptionorother),Humeconcludesthatwehavenoideaofsuchanentity(T1.4.6.3).Thismeansthatwecannotbelieveourselvestobesuchathing,belief(atleast,withrespecttomattersoffact)beingnothingbutalivelyidearelatedtoapresentimpression.18ThepositivepartofHumesaccountishisfamousbundlethe-ory,accordingtowhichamind(aselforperson)isacollectionofperceptionsrelatedtoeachotherincertainways,constitutingacom-plexentitytowhichidentityofonesort,thoughnotofanother,maybeintelligiblyandtrulyascribed.ThesortofidentitytheselfhasiswhatHumecallsimperfectidentity,distinguishingthiskindofidentityfromperfect(orstrict)identity,afeaturethatonlythoseentitiesthatundergonochangewhateverhave,or,asheputsit,afeatureofonlythoseentitiesthatarewhollyinvariableanduninterrupted.Theconclusionofhisnegativeargumentisnotthatwehavenoideaoftheself.Itisonlythatwehavenoideaofself,afterthemanneritishereexplaindthatis,noideaoftheselfasasimple,unchangingsubstance(T1.4.6.9,6,5,12).Itdoesnotfollowthatthesameexperiencethatfailstodeliversuchanidea17Therelationbetweenthefirst-personandthesecond-personelementsinHumesmethodisdiscussedinSectionVIandinthepapercitedinnote1.18InthesecondbookoftheTreatise,concernedwiththepassions(withwhatwewouldtodaycallmoralpsychology),Humeinsiststhattheideaofourselvesisalwaysintimatelypresenttous(T2.2.4.7).Theappearanceofinconsistencyhereevaporatesonceweremindourselvesthatthekindofself-awarenessHumerequiresinhisaccountofthepassionsdoesnotentailanythingaboutthenatureoftheobjectweareawareof.Whatourideaoftheselfis,andtheroleitplaysinouremotionallife,aredifferentquestions,largelyindependentofoneanother.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind59oftheselfcannotservetoexplainhowonecomestobelievethatoneisaself,understoodasathinghavingimperfectidentity.Wedohavetherequiredimpressionsforformingtheideaoftheselfasathingofthatsort:acomplexthingthatisnothingbutabundleofperceptionsunitedbycertainrelations.ThebeliefIhaveinmyownidentityovertimeisabeliefaboutsuchathingevenifIdonotrealizethisandneedHumeshelptodoso.19Anditisabeliefimplicatedinallourotherbeliefs,abeliefwithoutwhich,arguably,eventheskepticspositioncouldnotbeunderstood.What,aboveall,unitestheperceptionsthatcollectivelyconsti-tuteamindorselfismemory,andthenaturalrelationofcausationwithwhichmemoryisinextricablyboundup.Memoryisinonewaythemorefundamentalhere,sincewithoutitthenaturalrelationofcauseandeffectitselfcouldnotobtain.Humesaccountofcausa-tionrequiresthatIremembertheconstantconjunctionofpairsofeventsoftwodifferenttypesifexperienceofsuchconjunctionsistoleadmetothinkofeventsofthesetwotypesascauseandeffect.Themereoccurrenceofsuchconstantconjunctionsinmyexpe-riencewouldnotsuffice.SupposethatmyexperiencedidincludesuchrepetitionsofpairsoftypeAandoftypeBevents,butthatitdidnotalsoincludeperceptionsthatarerememberingsofpreviousco-occurrencesofthesepairs.Perhapssuchexperiencecouldgiverise,onafreshobservationofaneventoftypeA,toanexpecta-tionthataneventoftypeBwillfollow.(Therearereasonstodoubt19Wecouldsaythattheideaisanideaofanimperfectlyidenticalself(anideadere,orofathing),evenifitdoesnothavethatasitscontent(dedicto,orofwhatissaid).What,then,isitscontent?Oneansweris,asHumesometimesappearstosay,noth-ing.Another,betterone,istosaythatitistheselfasasimplesubstance.Idohaveanideaofidentity(oneofthesevenphilosophicalrelationsHumethinksexhaustthewaysinwhichanytwoarbitrarilychosenideasmaybecompared),asapplydinitsstrictestsensetoconstantandunchangeableobjects(T1.1.5.4).Toapplythatideatovariableorinterruptedobjectsistoattributeidentity[tothem],inanimpropersense,requiringafiction,eitherofsomethinginvariableanduninter-rupted,orofsomethingmysteriousandinexplicable(T1.4.6.7).Believingofaselfthatitisstrictlyidenticalissimplyamistake,consistingofmisapplyingtheideaofperfectidentity.ButitisamistakeofwhichIcanbefreedbycarefulattentiontoexperience.WhenIentermostintimatelyintowhatIcallmyself,whatIdonotfindisanimpressionofanobjecttowhichthatideacouldbeproperlyapplied(T1.4.6.3).IfIdrawtherightlesson,IshallceaseapplyingthisideaimproperlyandwillcometoseethatthereisanotherideaIhaveofmyself,asanimperfectlyidenticalthing,whosesimplecomponentscanbetracedtoimpressionsinmyexperience.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n60johnbirothis:Whatwouldbethemechanism?What,intheabsenceofmem-ory,woulddistinguishthepresent,nthexperienceofAfromthefirstexperienceofit?)Still,eventhatexpectationwouldnotbeenoughtoexplainthefeltdeterminationofthemindtomovefromthenewA-typeeventtoayet-to-be-observedB-typeevent.Suchdetermina-tionsofthemind,accordingtoHume,resultfromanexposuretotheconstantconjunctionsinquestion,andtheyserveinplaceofthenot-to-be-experiencedimpressionofnecessaryconnection.Itisfromthesedeterminationsofmindthattheideaofnecessaryconnectionderives.20MyexpectationofaneventoftypeB,whilebroughtaboutbytheconstantconjunctionbetweeneventsofthistypeandeventsoftypeAinmyexperience,would,intheabsenceofamemoryofthatconstantconjunction,failtobeanexpectationofaneffect.Beingcausedtoexpectsomethingisnotthesamethingasexpectingsome-thingforareason.WhenIexpectsomethingastheeffectofsomecause,Ireason(howeverunreflectively)fromtherememberedcon-stantconjunctiontotheeventIexpect.Thisreasoningis,ofcourse,matter-of-factorcausalreasoning,notdemonstrativereasoning,andhenceitlacksthekindofcertaintythatcharacterizesdemonstrativereasoning.But,aswehavealreadyseen,Humeinsiststhatcausalreasoningisnotonlyatrueformofreasoning,butis,indeed,thestrongestofall,theformofreasoningonwhichthewholecon-ductoflifedepends(SectionIIIabove).ThusitisthatHumecansay,Hadwenomemory,wenevershoudhaveanynotionofcau-sation,norconsequentlyofthatchainofcausesandeffects,whichconstituteourselforperson(T1.4.6.20).Itisthepresenceofmem-oriesamongmyperceptionsthatistheultimatesourceofmybeliefthatIamatemporallyextendedbeing.2120ForfurtherdiscussionoftheseissuesseeinthisvolumetheessayHumeonCausation.21Thesememoriesneednotbeveridical:whatmattersisthattheyareofeventsIbelievetohaveoccurredearlier.Memory,nolessthanbeliefitself,isamannerofconceiving,ofaspecialfeelfamiliarbutindefinablethatideasmerelyentertained(imagined)lack(T1.3.8.7;seealsoEHU5.12).Inthefullstory,forward-lookingperceptionsanticipationsalsoplayarole,asdoestheinertialtendencywehavealreadyseenatworkelsewhere:Buthavingonceacquirdthisnotionofcausationfromthememory,wecanextendthesamechainofcauses,andconsequentlytheidentityofourpersonsbeyondourmemory,andcancomprehendtimes,andcircumstances,andactions,whichwehaveentirelyforgot,butsupposeingeneraltohaveexisted(T1.4.6.20).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind61Onethingthatemergesfromthisdiscussionofpersonalidentity,withimplicationsthatgofarbeyondthattopic,isthatanentityofthesortHumetakesthemindtobe(complex,dynamic,ever-changing)mustandcanbethoughtofasanactiveagentinthefor-mationofitsbeliefsabouteverything(including,even,aswehaveseen,theformationofthebeliefinitsownidentity).Ageneral-izationofthisinsightunderliesvirtuallyallofHumesanalysesoftheconceptsweemployinthinkingabouttheworldandourrela-tiontoit.Mostimportantly,itdrivesallHumeshypothesesabouthowwecometobelievewhatwebelieve,whateverthecontentandobjectofthebeliefinquestion.FromtheearliestpartsoftheTrea-tise,withitspictureofcomplexideasbeingconstructedfromsimpleones,throughtheaccountofthenatureofbelief,tohisprinciplesofmoralpsychology,andtotheprinciplesofhispracticalphilosophy(hisethics,politics,andaesthetics),itistheconstantactivityofthismindthatdominatesthestoryandtiesittogetherintoaunifiedandcoherentwhole.22ItisnowonderthatKant,whilenevertiringofpointingoutthelimitationsofempiricism,nonethelessownsthatitwasHumewhowokehimfromhisdogmaticslumbers.23v.hume’sselfandsomerecenttheoriesofthemindWiththissketchofHumesaccountoftheselfinplace,wearenowinapositiontoexploresomeparallelsbetweenitandinflu-entialrecenttheories.Doingsowillbothillustratethecontinuingrelevanceofhisthoughtandhelpguardagainstsomecommonmis-understandingsofit.WhilethemoralsIshalldrawapplytootherpartsofthesetheories,Ihavesingledoutthetopicofpersonaliden-tityasmychiefillustration,forthreereasons.First,thisisthetopicthathasreceived,historically,andcontinuestoreceive,today,themostattentionfromthoseinterestedinHumesphilosophyofmind.Second,whilethisemphasisonthesubjectofpersonalidentitymay22Theaccountalsoincludesanaccountofbeliefformation,ofbelieftransitionbywayoftheprinciplesofassociation,andofcausalbeliefsasexpectationsproducedbyexperienceandhabit.23SeeKantsLettertoChristianGarveinhisselectedPhilosophicalCorrespondence,175999,ed.andtrans.A.Zweig(Chicago:UniversityofChicagoPress,1986),252.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n62johnbirohavestoodintheway,atleastuntilrecently,ofachievingasatis-factoryoverallinterpretationofHumesaccountofthemind,itisnonethelesstruethatanyinterpretationofthataccountmustoffer,orpresuppose,answerstothequestionsHumeasksconcerningtheideaoftheself,namely,whatthisideaisanideaof,andhowisitacquired.ForHume,givenhisuniqueaccountoftheworkingsofthemindandhisself-imposedempiricistconstraints,findingasat-isfactoryanswertothesequestionsisespeciallypressing.Third,itisnecessarythatweunderstandtheanswersHumegivestothesequestionsifwearetounderstandtherestofhisscienceofman,particularlythoseaspectsofthatsciencethatcenterontheconstantactivityofthemind,aspectsthatIhavesuggestedarefundamental.Amuch-favoredstrategyinrecentphilosophyofmindhasbeentolookfordecompositions,alongfunctionallines,ofthevariouskindsofbehaviorwethinkofasintelligentof,thatis,behaviorinvolvingrational,andnotmerelycausal,propertiesandprocesses.Myarmsrisingmaybecompletelyexplainedastheeffectofyourliftingit(asattheendoftheboxingmatch),butyourliftingofit,ormyraisingitwhenIwanttoclaimthefloor,cannotbeexplainedintheseterms.Thefirstisaneventrequiringmerelyacausalexpla-nation.Thelatterareactions,and,assuch,requireexplanationintermsofreasons,beliefs,desires,andthelike.Similarlywithblink-ingonemergingintobrightsunlight,incontrasttoputtingonsun-glasses.Thefirstinvolvesnoinferenceorthoughtonmypart,butthesecondrequiresit.Howevermanydifficultiestheremaybeinexplainingthefirstresponse,givingoneofthesecondkindinvolvesmore.Theaimofthesefunctionaldecompositionsistoidentifysim-pleprocesses,which,incombination,wouldexplain,even,perhaps,literallyconstitute,thecomplexrationalbehaviorinquestion.Weassumethattheagentwhosebehaviorwearetryingtoexplainisinvariousmentalstates(sensingthebrightlight,desiringtoavoidit,believingthathecandosobyputtingonsunglasses...)relatedinvariousregularways.Wethinkofoursubjectasgoingthroughvar-ioustransitionsfrommentalstatetomentalstate(fromthepainfulsensationtotheintentiontoacttothecarryingoutoftheacttoendthepainfulsensation)bywayofspecifiablestepsfallingintospeci-fiablepatterns.Withinthisframework,itisthebusinessofnormalempiricalscienceoftheburgeoningfieldofcognitivesciencethatincludescognitivepsychology,linguistics,neurophysiology,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind63andmoretogeneratehypothesesaboutwhatstatesandprocesseswouldbestexplainthesubjectsobservablebehavior.Thesimpleprocesseswehopetoidentifyasunderlying,oreveninsomesenseconstituting,complexintelligentbehavioraresupposedthemselvestobedumb,thatis,merelymechanical.Thechiefreasonforthisisametaphysicalconcern:itisfeltthatonlythenwillthementalhavebeenexplainedinrespectablyphysicalisticterms,showingittobesubjecttothesamelawsastherestofthenaturalworld.Inourdaythisphysicalisticassumptionisnotconsideredtobeinneedofdefence:thehistoricalalternativeofaccepting,withDescartes,aradicaldivisionofnatureintoseparatematerialandmentalrealms,andtheconcomitantbifurcationofourknowledge,isdeemedanonstarter,incompatiblewiththescientificoutlook.Thisso-calledtop-downapproach,whichseesintelligentbehav-iorasthetipofanicebergofunintelligent,mechanical,processessubservingit,hasindeedproventobefruitfulinthecognitivesci-ences,thoughithasnotgoneunchallenged.24Ofcourse,functionaldecompositioncanbeilluminatingandusefulevenwhencouchedinthoroughlymentalisticlanguage,asitoftenisinempiricalpsy-chologyandtheothersocialsciences.(Identifyingtheperceptualprocessesandmentalstatesandeventsthatareinvolvedinacom-plexact,that,say,ofdrivingacar,isanontrivialtask,andsuccessinthetaskyieldsunderstandingofanontrivialsort.)Thedemandthattherebeapointatwhichsuchmentalismprotemisredeemedinphysicalisticcoinisrootedinthemetaphysicalassumptionjustmentioned,oneweneednotmake,andonethatwouldberegardedbyHumeasoneofthemetaphysicalhypotheseshedenounces(TIntro.8).Nonetheless,wecanaskhowvarioustheories,Humesincluded,fareinmeetingthisdemand,supposingittobelegitimateandpressing.Onethingthatseemstostandinthewayofdischargingthedebttophysicalismisthefactthatthementalstatespositedinsuchdecom-positionsaretypicallywhatphilosophers,followingFranzBrentano,havecalledintentionalor,whatismuchthesamething,represen-tational.Simplyput,thismeansthatmentalstatesareaboutthingsotherthanthemselves(notnecessarilyphysicalthingsabelief24Thechiefchallengehascomefromconnectionism;forausefulintroduction,seeWilliamBechtelandAdeleAbrahamsen,ConnectionismandtheMind:AnIntro-ductiontoParallelProcessinginNetworks(Cambridge,MA:Blackwell,1991).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n64johnbirocanalsobeabout,say,anotherbelief).25Itispreciselythisfeaturethatmakesthemusefulinrationalizingexplanations:Iexplainyourreachingfortheglassbyattributingtoyouthebeliefthatitcon-tainsathirst-quenchingliquidandyourdesiretohaveyourthirstquenched.Thatbeliefandthatdesireexplainthatpieceofbehaviorbecauseofwhattheyareabouttheliquidandthethirst,respec-tively.(Aswehavealreadynoted,onlyifitcanbeexplainedinthiswaydoesthebitofbehaviorinquestioncountasanaction.)Butsincearepresentationrepresentsonlytoorforsomeone,astatethatissaidtorepresentmustbethoughtofashavinganinterpreter.Thenaturalthingtothinkinthiscase,ofcourse,isthatthisinterpreteristheagent.Butifwesaythat,wehavenotevenmadeastarttowardexplainingourintelligenceinthewaythedecompositionalstrategyiscommittedtodoing.Thealternativeistopositasubpersonalinter-preter,thatis,thinkofsomecomponentofthepersonasdoingtheinterpretation.(Suchasubpersonalinterpreterisoftenreferredtoasahomunculus.)Unless,however,weareultimatelyabletogetridofthishomunculusbyexplaininghowitsfunctions,includingtheinterpretiveone,canbecarriedoutbydumb,mechanical,compo-nentsofthesystem,weareleftwithanexemptagent,onewhoseownintelligenceisunexplained.Ifso,wewillnothavedischargedourdebtand,fromaphilosophicalpointofview(thoughnot,perhaps,fromascientificone),wemayaswellnothavestarted.DanielDen-nett,oneofthepioneersoftheapproachwearediscussing,explic-itlydubsthisHumesproblem.HeconcedesthatHumewiselyshunnedthenotionofaninnerselfthatwouldintelligentlymanip-ulatetheideasandimpressions.But,Dennettsays,thislefthimwiththenecessityofgettingtheideastothinkforthemselves.Eventhoughthisassociationisticcouplingofideasandimpres-sions,[the]pseudo-chemicalbondingofeachideatoitspredecessorandsuccessor,isanotoriousnon-solutiontotheproblem,DennettthinksthatHumehadnoalternativebuttotakeitseriously.2625Thisisnot,perhaps,trueofallmentalstates:sensations,suchaspains,moods,emotions,andthelikeareperhapsnotatleastnotstraightforwardlyaboutany-thing.Butthekindofstatesmostcentrallyinvolvedinunderstandingintelligentbehaviorbeliefsanddesiresdoseemtobeintentionalinthissense(asensedifferentfrom,thoughrelatedto,theusualoneofintentionalasitmodifiesactions,asenseclosetothatofdeliberate).26SeeDanielC.Dennett,Brainstorms:PhilosophicalEssaysonMindandPsychology(Montgomery,VT:BradfordBooks,1978).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind65IsHumereallyforcedintothepositionDennettattributestohim?Onlyiftherearenooptionsbutahomunculus-self,theverythingHumeclaimsnottofindinexperience,ornoselfatall.Tothinkthatthesearetheonlypossibilitiesis,however,amistake,andasurpris-ingone,giventhefactthatHumesbundletheorymayplausiblybeseenasdesignedpreciselytofindamiddlewaybetweenthesetwounacceptableextremes.WhileHumedoesnotdenythatthereisaself,somethingthatthinks,hasbeliefs,desires,andothercognitivestatesanddispositions,hedoesspendconsiderabletimetellinguswhatitisnot,namely,theunchangingsubstancephilosophershavetypicallytakenittobe.27Indoingthis,hecanappeartobesayingthattheselfisnotanything.Butthisappearanceismisleading,asisthatwhichpromptsDennettsreading.Themistakeis,onemustconcede,aneasyonetomake.Humedoes,indeed,oftenmakeper-ceptions,asopposedtobundlesofthem,thesubjectofcertainverbsthatlookintentional:theyaresaidtoproduceandattracteachother.ThiscanencouragetheattributiontoHumeoftheso-calledNewtonianpicture.(Socalledbecause,inanefforttodescribethebondsamongperceptions,itmakesobviousrelianceonmetaphorsdrawnfromNewtonsaccountofthegravitationalattractionamongbodies.)Takenatfacevalue,suchlanguagesuggeststhatitisthesesubpersonalcomponentsofapersonthatarethefundamentalbear-ersofintentionalproperties,withattributionsatthepersonallevelbeingderivative.Theproblemwiththislineofthoughtisthatwenowhavetomakesenseofperceptionsdoingother,muchmorepuz-zling,things,suchasunderstandingeachotherandthemselves.AsDennettpointsout,thisparallelsthemoderncognitivistsproblemofhavingtomakesenseofrepresentationeitherwithoutanexemptagent,orbyacceptinganunendingregressofcomponents,andthusendingupwithanuninformativetheory.2827IfonedidtakeHumetobereallydenyingthatthereisaself,whatwouldonetakehimtobetalkingaboutinBook2andBook3oftheTreatise,whichunmistakablytakeitforgranted?28FordiscussionoftheNewtonianpictureoftheself,inwhichitisideasthatarethesubjectofintentionalverbs,seeJaneL.McIntyre,IsHumesSelfConsistent?inMcGillHumeStudies,ed.D.F.Norton,N.Capaldi,andW.L.Robison(SanDiego:AustinHillPress,1979),7988,andBiro,HumesDifficultieswiththeSelf.Forfurtherdiscussionofpersonalidentity,seeinthisvolumetheessayHumeandtheProblemofPersonalIdentity.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n66johnbiroIsHumereallyforcedintothepositionDennettdescribes?Infact,thereisevidencethatheismoresuccessfulthansomemoderncognitivistsinavoidingthetrap.Inbrief,theevidenceisthathenevermakesanythingotherthanapersonthesubjectofaseriouslyintentionalasopposedtowhatwemaycallaquasi-intentionalpredication.Whilehetalksofperceptionsattracting,producing,destroying,andinfluencingeachother,henevertalksofthemasthinking,understanding,willing,ordesiring.Expressionsoftheformersortshouldbetakenforthemetaphorstheyare.WealsosaythingslikeCloudsproducerainandMagnetsattractnails,withouttakingourselvestobeattributinggenuineintentionalstatestocloudsandmagnets.Theliteralmeaningofsuchsentencesisthatcloudsormagnetsarethecausesofcertaineffects.WehavealreadyseenthatHumeishappytosaythatanidea(better:theoccurrenceofanidea)cancause(theoccurrenceof)anotherideathatiswhatanideasintroducinganotherideacomesto,andthevariouswaysinwhichthishappensiswhattheprinciplesofassociationdescribe.Theremayindeedbeamysteryabouthowamindconstitutedofperceptions,asHumesaccounthasit,candothethingswesayitdoesinourordinary,nonphilosophical,discourse.Butthereisnomysteryabouthowtheperceptionsthatconstituteamindcandothesesamethings,forthesimplereasonthattheyarenotsaidbyHumetodothem,and,contraDennett,nothinginHumestheorycommitshimtosayingthattheydo.Onthecontrary,Humespointisthatonlyselvesconstitutedinthewayhedescribescanbeintel-ligiblysaidtodothethingspeople(oranyotherintelligentbeings)aresaidtodo.vi.scientistorphilosopher?AsIremarkedearlier,oneofthesurprisingconsequencesofthepic-turebothHumeandmoderncognitivesciencepresentofthecogni-tiveagentisthatheisnotalwaysthebestsourceofinformationandinsightabouthisowncognitivelife.Ifwewanttofindoutwhatpro-cessesareinvolvedintheperformanceofsomecognitivetaskandhowtheywork,askingthecognizerisoftenunhelpfulandsome-timesevenmisleading.Oneofthemostrobustgeneralfindingsinrecentcognitivesciencehasbeenthatweoftengetmoreinteresting,moredetailed,andmorereliableinformationabouttheseprocessesCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind67fromexperimentsthatmeasureresponsetimes,uncovererrorpat-terns,testcomprehension,andsoon.Themoderncognitivescientistcanacceptthislessonwithequa-nimity,evenwithrelish.ButcanHume,inspiteofhisscientificambitions?Ihavesuggestedthathisowngeneralpictureofthescienceofthemind,andsomeofhisspecificinsightsabouthowthemindworks,anticipatesomeoftheseresults.Andhecertainlyemphasizestheneedforexperiments,although,aswehavenoted,whathehasinmindisratherdifferentfromthekindoflaboratoryexperimentonwhichmodernscienceingeneralthrivesandthatis,asIhavejustnoted,commonincognitivescience.ButHumestillhasadeepcommitment,inheritedfromthewayofideasfollowedbyboththeCartesiansandhisBritishempiricistpredecessors,tointrospectionasawayoffindingepistemologicalbedrock.So,forhim,theneedtofindtherightbalancebetweenthesubjective,phe-nomenological,approachsocentraltotheintrospectivetradition,andtheobjective,third-person,experimentalmethodsstandardinscientifictheory,isindeedapressingone.Humescommitmenttotheintrospectivemethodshouldnotbeseenasmerelytheresultofhisinabilitytofreehimselffromaper-vasivepatternofthought,onethatcontributesnothingtothesub-stanceofhisaccountofthemind.Humesaimsare,aswesawearlier,ambitious,andrequirehimtolookforacertainkindofgroundingoftheconceptsandbeliefsthatanonphilosophicalscientistcantakeforgranted.Unlikethelatter,hecannotbeanaiverealistaboutexternalobjects,causation,ortheselfinthewayaworkingscientistcanand,indeed,mustbe.Heneedstogiveanaccountoffundamen-talconceptsandbeliefsthatare,fromapurelyscientificpointofview,unproblematic.Itisnotthatcognitivesciencehasnothingtosayaboutsuchconceptsandbeliefsandabouttheirroleinourmen-tallife.Indeed,itdoes.Butitdoessowithoutraisingphilosophicalquestionsabouttheirlegitimacy,without,inparticular,takingthepossibilityofskepticismseriously.Hume,ofcourse,doestakethispossibilityseriously.AsInotedearlier,oneofthefeaturescommontoHumesscienceofthemindandtheso-callednaturalizationpro-gramsinrecentepistemologyandphilosophyofmindisashiftfromthejustificationofbeliefsinthetraditionalsense,toanexplanationoftheirprovenancethroughanexaminationofourcognitiveendow-mentsandhistory.InbothcasestheshiftleadstodoubtabouttheCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n68johnbirovalue,eventhecoherence,oftraditionalepistemologicalprojects.ButonlyinHumescasedoesitalsoleadtorethinkingtheverynotionofjustification.29ThusthesimilaritybetweenHumesprojectandmoderncogni-tivescienceshouldnotbeoveremphasized.Therearesignificantdif-ferences,aswell.AmongtheseareHumescontinuedadherencetotheintrospectivemethodheinheritedfromhispredecessors,aswellashisrefusaltoabandonthoseelementsofthetraditionalframeworkthatderivefromcommonsenseandoureverydaypractices,ratherthanfromtherarifiedandesotericactivitiesofphilosophersorsci-entists.Thatiswhy,toreturnonceagaintotheclearandexplicitexplanationofhismethodsetoutatthestartofhisproject,theexperimentsinhissciencemustconsistofacautiousobservationofhumanlifeandmustbetakenastheyappearinthecommoncourseoftheworld(emphasisadded).Hume,then,hasamuchmorecomplextaskthanthemoderncognitivescientist,oreventhemodernnaturalizingphilosopher.Hemusttrytofittogetherintoacoherentwholeanumberofelementsthatdonoteasilygotogether:innocentscientifictheorizing,theintrospectivemethodheinheritsfromhispredecessors,philosoph-icalanalysis,andanultimateallegiancetocommonsenseasthetouchstoneforthedeliverancesofeachoftheformer.Humedoesnot,onemustgrant,fullysucceedinweavingtogetherthedifferentstrandsinhisthoughtresponsivetothesedifferentdemands.Itisnowonderthatthedebatesthathavedoggedtheinterpretationofhisworksincehisownday,debatesaboutwhetherheisaphiloso-pherorjustapsychologist,anaturalistoraskeptic,continueunabated.Take,forexample,thethesisthatournaturalbeliefsare29Fodor,whilenowconceding,indeed,emphasizing,thatHumeanticipatessomeofthecentralinsightsofmoderncognitivescience,stillarguesthatHumesempiri-cism,inparticular,hiscommitmenttothecopyprinciple,standsinthewayofhisrecognizingthatanaccountofwhatbeliefsare,whattheyareabout,andhowtheyinteract(theproperconcernsofascientifictheoryofthemind)mustbedis-tinctfrom,andindependentof,atheoryofhowthebeliefsweactuallyhavemaybeproventobetrue,aswellasofoneofhowwecometohavethem,thetwinconcernsofboringoldepistemology.SeeHumeVariations,especiallyChap.2.PerhapsHumespointispreciselytofashionanewnotionofjustification,oneinwhichabeliefisjustifiedbecauseofhowwecometohaveit.Perhapshesupposesthatabeliefisjustifiedifitisunavoidablegivenourcognitivemake-upandourexperience.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesNewScienceoftheMind69irresistible.Wehaveseenthatarecognitionof,indeed,insistenceonthisisacornerstoneofHumesaccountofthemind.Yetevenifnoamountofphilosophicalreflectioncanhalttheprocessesthatyieldthesebeliefs,thatdoesnotmeanthatsuchreflectioncannotleadustoquestiontheirepistemologicalstatus.Whenuncheckedbycommonsense,suchphilosophicalreflectioncanleadtothedeep-estdarknessoftotalskepticism.Fortunately,whencommonsensereassertsitself,asitinevitablydoes,theskepticsspeculationsareseenascold,andstraind,andridiculous(T1.4.7.89).Yetthey,too,areinasense,naturaland,atleastforsomeminds,irresistible.Ifoneisaphilosopher,thedeepestchallengeistofindawayoflivingwiththeseirreconcilabledemandsofonesnature.suggestionsforfurtherreadingInadditiontotheworkscitedinthenotestothisessay,forfurtherreadingthefollowingarerecommended.Baxter,DonaldL.M.HumesDifficulty:TimeandIdentityintheTreatise.NewYork:Routledge,2008.Biro,John.DavidHume.InTheEncyclopediaofCognitiveScience,L.Nadel,ed.(NewYork:Macmillan,2002).Bower,Kennett.Imagery:FromHumetoCognitiveScience.CanadianJournalofPhilosophy14(1984):21734.Bricke,John.HumesPhilosophyofMind.Princeton,NJ:PrincetonUniver-sityPress,1980.74123.Broughton,Janet.WhatDoestheScientistofManObserve?HumeStudies18(1992):15568.Garrett,Don.CognitionandCommitmentinHumesPhilosophy.NewYork:OxfordUniversityPress,1997.Owen,David.HumesReason.Oxford:ClarendonPress,1999.Stroud,Barry.Hume.London:RoutledgeandKeganPaul,1977.Waxman,Wayne.HumesTheoryofConsciousness.Cambridge:CambridgeUniversityPress,1994.Winkler,Kenneth.TheNewHume.PhilosophicalReview100(1991):54179.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\ndavidowen3HumeandtheMechanicsofMindImpressions,Ideas,andAssociationBythetimeHumestartedtoworkonhisTreatise,thenotionofanideaastheprimary,mostgeneralsortofmentalitemdominatedEuropeanphilosophy.AlthoughDescartesnotedthat,strictlyspeak-ing,onlythosethoughtsthatareasitwereimagesofthingswereappropriatelydescribedasideas,inpracticeheusedthewordideatorefertowhateverisimmediatelyperceivedbythemind.1Notonlydowehaveideasoftreesandthesun,butwealsohaveideasofourownactivitiesofthinkingandwilling.LockecharacterizesideaasbeingthatTerm,which,Ithink,servesbesttostandforwhatsoeveristheObjectoftheUnderstandingwhenaManthinks.Lockealsothinksthatwenotonlyhaveideasthatderivefromthingsorobjectsintheworld(ideasofsensation),butalsooftheactivitiesandoperationsofourownminds(ideasofreflection).Ideasofsen-sationareacquiredthroughtheoperationofexternalobjectsonoursenseorgans,whileideasofreflectioncomefromintrospection,fromthinkingaboutwhathappenswithinourownminds.Healsothinksthattheseideasofreflectionareoftwobasicsortsofmentalactivity,perceptionandwilling,thatcorrespondtotwofacultiesofmind:theunderstanding(orthepowerofthinking)andthewill(orthepowerofvolition).2Humeintroducedimportantinnovationsconcerningthetheoryofideas.Thetwomostimportantarethedistinctionbetweenimpres-sionsandideas,andtheusehemadeoftheprinciplesofassociation1ReneDescartes,MeditationsonFirstPhilosophy,inThePhilosophicalWritingsofDescartes,trans.J.Cottingham,R.Stoothoff,andD.Murdoch,3vols.(Cambridge:CambridgeUniversityPress,1984),2:25,127.2JohnLocke,AnEssayconcerningHumanUnderstanding,ed.PeterNidditch(Oxford:ClarendonPress,1975),1.1.8,2.6.2.70CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind71inexplainingmentalphenomena.Humedividedtheperceptionsofthemindintotwoclasses.Themembersofoneclass,impressions,heheldtohaveagreaterdegreeofforceandvivacitythanthemembersoftheotherclass,ideas.Healsoarguedthatideasarecausallydepen-dentcopiesofimpressions.And,unlikeLockeandothers,Humemakespositiveuseoftheprincipleofassociation,bothoftheasso-ciationofideas,and,inamorelimitedway,oftheassociationofimpressions.Suchassociationsarecentraltohisexplanationsofcausalreasoning,belief,theindirectpassions(prideandhumility,loveandhatred),andsympathy.TheseviewsaboutimpressionsandideasandtheprinciplesofassociationformthecoreofHumessci-enceofhumannature.Relyingonthem,heattemptsarigorouslyempiricalinvestigationofhumannature.Theresultingsystemisaremarkablebutcomplexachievement.i.impressionsandideasHumebeginsBook1oftheTreatise,OftheUnderstanding,bysay-ing:Alltheperceptionsofthehumanmindresolvethemselvesintotwodistinctkinds,whichIshallcallimpressionsandideas(T1.1.1.1).InhisEnquiryconcerningHumanUnderstanding(here-afterEnquiry)hesaysmuchthesamething,butaddsanexample:Everyonewillreadilyallow,thatthereisaconsiderabledifferencebetweentheperceptionsofthemind,whenamanfeelsthepainofexcessiveheat,orthepleasureofmoderatewarmth,andwhenheafterwardsrecallstohismemorythesensation,oranticipatesitbyhisimagination(EHU2.1).Inneitherworkdoeshemakeanattempttoexplainwhathemeansbythephraseperceptionsofthemind,butitwouldhavebeenobvioustoanyeighteenth-centuryreaderthatheisusingthatexpressionmuchasDescartesandLockehadusedthetermidea:foranythingthatmindisawareoforexpe-riences.AshehadputitintheTreatise:Tohate,tolove,tothink,tofeel,tosee;allthisisnothingbuttoperceive(T1.2.6.7).HumesinitialstepintheTreatiseistoshowthatperceptionsofthemindmaydividedintotwodistinctkinds,impressionsandideas.Thesetwokindscommonlydiffer,hesays,inthedegreesofforceandliveliness,withwhichtheystrikeuponthemind.Amongourimpressions,thoseperceptionswiththemostforceandvivac-ityaresensations(includingthoseofpainandpleasure)andtheCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n72davidowenpassionsandemotions.Ideasaredescribedasthefaintimagesofimpressionsthatarefoundinthinkingandreasoning.Thedistinc-tionbetweenideasandimpressionsisfurthercharacterizedasthedifferencebetwixtfeelingandthinking.Perceptionsalsodifferinbeingeithersimpleorcomplex.Simpleperceptions,hesays,aresuchasadmitofnodistinctionnorseparation,asinglecolorortaste,forexample.Complexperceptions,incontrast,arethosethatmaybedistinguishdintoparts,forexample,theseveralqualities(color,taste,smell,etc.)unitedtogetherintheperceptionofanapple(T1.1.1.13;cf.EHU2.3).ImpressionsandIdeasofSensationHumesaysthattherearetwokindsofimpressions,thoseofsen-sationandthoseofreflection.Althoughhehasmuchtosayaboutthecausaloriginofimpressionsofreflection,hesaysthatimpressionsofsensationariseinthesouloriginally,fromunknowncauses.3Hesaysalmostnothingaboutthecausesofsenseimpres-sionsandhas,strictlyspeaking,notheoryofperception.Howwecometohaveimpressionsofsensationisaproblemthatheleavestoanatomistsandnaturalphilosophers(T1.1.2.1;cf.T2.1.1.2).HisconcerninBook1oftheTreatiseandthefirstEnquiryislimitedprincipallytotheideasthatarederivedfromsuchimpressions.WhileHumeinitiallydividesimpressionsintosensations,pas-sionsandemotions,ideasarecharacterizedonlyastheimagesofimpressions,andasthematerialsofthinkingandreasoning.ItisimportanttorememberthatHumesinitialdiscussionsinTreatise1.1.1.1andinEnquiry2.1areprovisional,intended,bytheuseofexamples,tointroducethereadertothedistinctionhehasinmind.Onceroughedout,thisdistinctionismadewiththemoreaccuracyasHumeproceedswithamoreaccuratesurvey.Thus,whiletheofficialdistinctionbetweenimpressionsandideasismadeintermsofforceandlivelinessorvivacity,wegetaninitialgriponitonlyasHumegivesusexamplesofimpressions(ofsensations,passions,andemotions),andthentellsusmoreaboutideas,thefaintimagesofimpressions.OnemightthinkthatthecharacterizationofideasasfaintimagesofimpressionsprejudgesHumesimportantcopy3DiscussionofthisimportantpointisdeferreduntilSectionIII.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind73principle.Humetookgreatpainstoargueforthisprinciplethatallsimpleideasarederivedfromsimpleimpressionsonempiricalgrounds.HethensaysthatthisisthefirstprincipleIestablishinthescienceofhumannature(T1.1.1.34,12).Hiseffortstoestab-lishitaremeanttobeaparadigmofempiricalrigor.ButifboththetruthofthisprincipleandHumesmethodofestablishingitaresus-pectbecauseofanapparentprejudgmentoftheissue,thenhisnewscienceisofftoashakystart.4ThecopyprincipleisnotmerelythefirstandoneofthemostimportantresultsofHumesscienceanditsmethod.Inaddition,becauseimpressionseffectivelyconstituteordelineateourexperience,theclaimthatallsimpleideasderivefromsimpleimpressionsgivessubstantialcontenttoHumesmethod-ologicaldeterminationtostaywithintheboundsofexperience.AslongasoneremembersthatHumesinitialdiscussionisdelib-eratelylooseandinexact,thisissuecanberesolvedinHumesfavor.Itistruethatthetermimagesuggestsarelationshipofdependence:animageisdependentonthatofwhichitisanimage.ButallHumeneedsistheweakerconcept,resemblance,ofwhichhemakesexplicituseinthenextfewpages.Theofficialdistinctionbetweenimpressionsandideasisintermsofforceandvivacity.Thereferencetosensations,passionsandemotions,ontheonehand,andfaintimagesofthese,ontheother,isnotpartoftheofficialtheory,atthisstage.TheseexamplesareintroducedearlyoninordertohelpusgraspthedistinctionthatHumegoesontomakeofficially,asitwere,inotherterms.If,inthecourseofgivingtheseexamplestohelpustounderstandthecontentheintendstogivethetermsimpressionsandideas,heusesastrongertermthanheshould,noharmisdone.Noharm,thatis,solongas,inthecourseofestablish-ingtheprecedenceofimpressionsoverideas,therealworkisdonebythenotionofresemblance,notbythatofimage.ThuswhenHumegoesontoconsiderwiththemoreaccuracyhowitisthatimpressionsandideasinteractamongthemselvesandwitheachother,heshiftsfromtalkofimagestothatofresem-blance:Thefirstcircumstance,thatstrikesmyeye,isthegreat4PerhapsthemostimportantaspectofHumesscienceofhumannatureishisdeter-minationnottogobeyondexperience,bothwithrespecttothecontentofourmentalstatesandwithrespecttotheexplanationofmentalphenomena.Ashesays:tisstillcertainwecannotgobeyondexperience(TIntro.8).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n74davidowenresemblancebetwixtourimpressionsandideasineveryotherpar-ticular,excepttheirdegreeofforceandvivacity.Hethenmentionsthatonekindofperceptionseemstobeareflectionoftheother,buthedoesnotsaywhichkindistheoriginalandwhichthereflec-tion.Hefirstestablishesthecorrespondencebetweensimpleideasandimpressions:everysimpleideahasasimpleimpression,whichresemblesit;andeverysimpleimpressionacorrespondentidea.Humeaffirmsthisruleonthebasisofobservation,andfurthersupportsitbyissuingachallengetoanyonewhodoubtstherule:findacounterexample.Hethenturnstoacentraltaskinhissci-enceofhumannature,thatoftracingtheconnections,especiallythecausalconnections,betweenimpressionsandideas.Thefullexaminationofthisquestion,hesays,isthesubjectofthepresenttreatise(T1.1.1.3,5,7).5AtthisearlystageHumelimitshimselftoapreciseenuncia-tionofthecopyprinciple:weshallherecontentourselveswithestablishingonegeneralproposition,thatalloursimpleideasintheirfirstappearancearederiv’dfromsimpleimpressions,whicharecorrespondenttothem,andwhichtheyexactlyrepresent.Heestablishesthisprincipleintwostages.Hefirstreassureshimselfbyanewreview...thateverysimpleimpressionisattendedwithacorrespondentidea,andeverysimpleideawithacorrespondentimpression.Inaforeshadowingofhisanalysisofourideaofcau-sation,hedescribesthisrelationshipasaconstantconjunction.Andhearguesthatthisconstantconjunctionisevidenceofacausalconnectionbetweenimpressionsandideas:Suchaconstantcon-junction,insuchaninfinitenumberofinstances,canneverarisefromchance;butclearlyprovesadependenceoftheimpressionsontheideas,oroftheideasontheimpressions(T1.1.1.78).Thequestionofdependence,Humeargues,canbedecidedbydeterminingwhichofapairofresemblingimpressionsandideasappearsfirst.Hefindsbyconstantexperiencethatsimple5IntheEnquiry,Humeskipsthestepofestablishingthecorrespondenceruleandgoesstraighttoprovingthatallourideasormorefeebleperceptionsarecopiesofourimpressionsormorelivelyones(EHU2.5).ThelackofthecorrespondenceruleintheEnquiryleavesopenaplausibleoptionapparentlyruledoutbythepresenceoftheruleintheTreatise:onemighthaveasimpleimpressionwithoutacorrespondentsimpleidea,thoughonecannothaveasimpleideawithoutasimpleimpression.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind75impressionsarealwaysexperiencedbeforetheircorrespondingideas.Healsofindsthatourpracticeconfirmsthis:whenwewanttointro-duceapersontotheideaoforange,weconveytohertheimpressionbutproceednotsoabsurdly,astoendeavourtoproducetheimpres-sionsbyexcitingtheideas.Hesummarizeshisargumentbysayingthattheconstantconjunctionofourresemblingperceptions,isaconvincingproof,thattheonearethecausesoftheother;andthispriorityoftheimpressionsisanequalproof,thatourimpressionsarethecausesofourideas,notourideasofourimpressions.Healsoprovidesasecondcausalargument,pointingoutthatwheneverbyanyaccidentthefaculties,whichgiverisetoanyimpressions,areobstructedintheiroperations,aswhenoneisbornblindordeaf;notonlytheimpressionsarelost,butalsotheircorrespondentideas;sothatthereneverappearinthemindtheleasttracesofeitherofthem.Noristhisonlytrue,wheretheorgansofsensationareentirelydestroyd,butlikewisewheretheyhaveneverbeenputinactiontoproduceaparticularimpression.Wecannotformtoourselvesajustideaofthetasteofapine-apple,withouthavingactuallytastedit.(T1.1.1.89)6Thisfirstprincipleofthescienceofhumannatureisimportantformanyreasons.PerhapsmostimportantisHumesdeterminationtousetheprincipleasawayoftestingthecontentofideasandthuslimitingmetaphysicalspeculation.Ifideasofsensationarecopiesofimpressions,thenthecontentofsuchanideacannotoutstripthecontentoftheimpressionfromwhichitisderived.IntheEnquirythistestisexplicitlyputforwardasacheckonthemeaningofphilosophicalterms:Whenweentertain,therefore,anysuspicion,thataphilosophicaltermisemployedwithoutanymeaningoridea(asisbuttoofrequent),weneedbutenquire,fromwhatimpressionisthatsupposedideaderived?Andifitbeimpossibletoassignany,thiswillservetoconfirmoursuspicion(EHU2.9;cf.A7).7IntheTreatise,Humefirstusesthischecktotestthenotionofsubstancein1.1.6,butthemostfamousdeploymentoccursinhisdiscussionoftheideaofnecessaryconnection.6SeealsoEHU2.7.7Sincetheclaimsaboutimpressionsandideasareempirical,itis,ofcourse,possibletoconceivethat,inaparticularcase,theordermightbereversed.Themissingshadeofblueexample(seeT1.1.1.10)mightwellbeHumeswayofemphasizingtheempiricalnatureofhisclaims.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n76davidowenIdeasandImpressionsofReflectionInTreatise1.1.2,Humedistinguishesbetweenimpressionsofsen-sationandimpressionsofreflection.Impressionsofreflectionarederivdinagreatmeasurefromourideas.Onthisaccount,wefirstexperienceimpressionsofsensation,includingheatorcold,thirstorhunger,pleasureorpain.Copiesoftheseimpressionsareretainedasideas,andwhenwerecallsuchanideaofpleasureorpainitproducesthenewimpressionsofdesireandaversion,hopeandfear.TheseHumecallsimpressionsofreflectionbecausetheyarederivedfromareflectiononpreviousexperience.Amongimpressionsofreflectionheinitiallyincludesthepassions,desires,andemotions,butthisclassificationisapparentlyprovisional.InBook2hespeaksofsecondaryimpressions,whichcanbecausedbyeitherimpressionsorideas.Onlythelatterareproperlycalledimpressionsofreflection.8Moreover,Humeoftenusesemotiontorefer,nottoanimpressionofreflection,buttothefeelingsuchanimpressionhas.9Healsosometimesusessensationinthesameway,aswhenhesaysofcalmpassionsthattheyproducelittleemotioninthemind,andaremoreknownbytheireffectsthanby[any]immediatefeelingorsensation(T1.1.2.1,2.3.3.8).AtTreatise2.1.1,Humereplacesthedistinctionbetweenimpres-sionsofsensationandreflectionwithamoreaccuratedistinctionbetweenoriginalandsecondaryimpressions.Asatermofclassifi-cation,impressionsofsensationiseithertoonarrowortoobroad.Ifsensationmeanscomesfromthesenses,thenitisnotclearthatimpressionsofsensationincludepleasuresandpains.Butifsensationjustmeansfeeling,thenthereisnocleardifferencebetweenimpressionsofsensationandthoseofreflection.Further-more,thecategoryimpressionsofreflectionismisleading.Notallpassions,emotions,anddesiresarecausedbyreflection,by,thatis,theconsiderationofourideas.Somearecausedimmediatelybyotherimpressions.Originalimpressions,hesays,aresuchaswith-outanyantecedentperceptionariseinthesoul.Theyincludeall8T2.1.2.4;seealsoT2.3.9.2.Humedoesnotseemtohaveanameforthosesecondaryimpressionsthatarenotimpressionsofreflection.9SeethediscussionatT2.2.8.4,andemotionintheGlossaryoftheOxfordPhilo-sophicalTexts(OPT)editionofATreatiseofHumanNature,ed.D.FateNortonandM.J.Norton(Oxford:OxfordUniversityPress,2000),575.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind77theimpressionsofthesenses,andallbodilypainsandpleasures.10WhenHumesaysthattheseoriginalimpressionsmaketheirappear-anceinthesoulwithoutanyintroduction,hedoesnotmeanwithoutcause,butratherwithoutanyprecedingthoughtorper-ception.11Everyimpressionhasacausalhistory,butthecausalhistoryoforiginalimpressionsdoesnot,typically,involveotherper-ceptionsofthemind.Moreover,HumeherereiteratesthepolicyestablishedatthebeginningofBook1.Hewillnotlookforthenat-uralandphysicalcausesoforiginalimpressions.Todoso,hesaysagain,wouldleadhimawayfromhissubject,intothesciencesofanatomyandnaturalphilosophy(T2.1.1.12).FortheremainderoftheTreatise,Humetreatsthepassionsassec-ondaryimpressions.Secondaryimpressionsarecausedbyoriginalimpressionsorbyanideaderivedfromanoriginalimpression.Theoriginalimpressionsareusuallypleasuresorpains.12Consider,forexample,thebodilypainassociatedwithanattackofgout.Thispro-ducesalongtrainofpassions,asgrief,hope,fear.Thesepassionsmayalsobeproduced,notbyanypresentpainfromthiscondition,butbyrememberingthepainitcausedinthepastoranticipatingitsfuturepain.AsHumesays,Bodilypainsandpleasuresarethesourceofmanypassions,bothwhenfeltand[when]considerdbythemind(T2.1.1.2).Humealsodistinguishesbetweendirectandindirectpassions.Amongtheformerhelistsdesire,aversion,grief,joy,hope,fear,despair,security,and,interestingly,volition(T2.1.2.4).Thesedirectpassionsariseimmediatelyfromanimpressionorideaofpainorpleasure,orwhatHumeoftencallsgoodorevil(seeT1.3.10.2,10Humelistsaspossiblecausesoftheseoriginalimpressionstheconstitutionofthebody,theanimalspirits,andtheapplicationofobjectstotheexternalorgans(T2.1.1.1).Notethatanyofthesecausesmightaswellbecausesofbodilypainsandpleasuresascausesofimpressionsthatcomefromthesenses.ThesignificanceofourignoranceofthecausesoforiginalimpressionsisdiscussedinSectionIII.11Inthesameparagraphhesaysthatoriginalimpressionsariseoriginallyinthesoul,orinthebody,which-everyoupleasetocallit.ItaketheextremelycasualnatureofthiscontroversialremarktobefurtherevidencethatHumerefusestobedrawnaboutthecausalbackgroundoforiginalimpressions,includingimpressionsofsensation.12Tiseasytoobserve,thatthepassions,bothdirectandindirect,arefoundedonpainandpleasure,andthatinordertoproduceanaffectionofanykind,tisonlyrequisitetopresentsomegoodorevil(T2.3.9.1).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n78davidowenforexample).Furthermore,themindbyanoriginalinstincttendstouniteitselfwiththegood,andtoavoidtheevil,thotheybeconceivdmerelyinidea,andbeconsiderdastoexistinanyfutureperiodoftime.Byoriginalinstinct,Humemeansabasicfeatureofhumannature,afeaturethatexplainsbehavior,butthatcannotitselfbeexplained.Thesefundamentalconnectionsbetweenplea-sureanddesireorpainandaversionhelpexplainhumanmotivation.AsHumeputsit,thewillexertsitself,wheneitherthegoodortheabsenceoftheevilmaybeattaindbyanyactionofthemindorbody(T2.3.9.12,7).13Inadditiontothosedirectpassionsthatarisefromtheperceptionofpleasureandpain,therearealsothosethatarisefromanatu-ralimpulseorinstinct,whichisperfectlyunaccountable.Theseinstinctivelybasedpassionsincludethedesireofpunishmenttoourenemies,andofhappinesstoourfriends;hunger,lust,andafewotherbodilyappetites.Unliketheotherdirectpassions,thesepassionsarenotcausedbypleasureorpainortheprospectthereof.Onthecontrary,thesepassionsproducepleasureorpain,as,forexample,whenhungerleadsustoeat,anactthatgivespleasure.Giventhecentralrolethatthesetwokindsofdirectpassionplayinhumanmotivation,onewouldexpectHumetohavealottosayaboutthem.But,apartfromtheextensivediscussionofhopeandfearinTreatise2.3.9,hetellsusthatnoneofthedirectaffectionsseemtomeritourparticularattention(T2.3.9.89).Heismuch13Humehasaratherdeflationaryaccountofthewill.AlthoughinT2.3.9.2hecountsvolitionasadirectpassion,inT2.3.1.2hesaysthatthewill,properlyspeakingisnotcomprehendedamongthepassions.Hegoesontodefinethewillastheinternalimpressionwefeelandareconsciousof,whenweknowinglygiverisetoanynewmotionofourbody,ornewperceptionofourmind.Ithinkthesituationisthis.Theterm,thewill,isatermforafaculty;otherfacultiesthatHumetalksaboutincludememory,thesenses,theimagination,judgment,andreason.Facultiesproduceimpressionsorideas.Humedoesnotseemtohaveatermforthefacultythatproducespassions.Lookedatinthisway,ofcourse,thewill,asafaculty,isnotapassion.Butwhatthewillproduces,volitions,arepassions.IdescribethisasadeflationaryaccountofthewillbecauseHumedoesnotthinkthatappealstosuchafacultyareexplanatory.Whatisimportantisnotthepostulationoffaculties,butthetracingofcausesandeffectsamongimpressionsandideas.Anactionisnotexplainedbysayingthatitwascausedbyavolition,whichwasproducedbythefacultyofthewill.Itisexplainedbysayingitwascausedbyavolition,whichwascausedbyadesireoraversion,whichwasinturncausedbytheperceptionofpleasureorpain.Foradditionaldiscussionofthewill,seeinthisvolumetheessayHumesMoralPsychology.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind79moreinterestedinwhathecallstheindirectpassions,discussionofwhichoccupiesroughlytwo-thirdsofBook2oftheTreatise.Humehasanextraordinarilyrichandinterestingstorytotellabouttheindirectpassions,theirnatureandtheircausalorigins.Insomewayshisaccountoftheseisthemosttechnicallysophisticatedpartofhisscienceofhumannature.Thebasicindirectpassionsareprideandhumility,loveandhatred.Othersthatderivefromthesefourareambition,vanity,envy,pity,malice,generosity,andtheirdependants(T2.1.1.4).Severalfeaturesoftheindirectpassionsstandout.1.Theindirectpassions,likeallpassions,aresimpleanduni-formimpressions.Asaresult,tisimpossiblewecanever,byamultitudeofwords,giveajustdefinitionofthem,orindeedofanyofthepassions(T2.1.2.1).2.Thefeelingofthepassionsofloveandprideispleasant,whilehatredandhumilityfeelpainful.Thecausesofloveandprideproduceapleasuredistinctfromthosepassionsthemselves,whilethecausesofhatredandhumilityproduceapaindistinctfromthosepassionsthemselves.3.Theindirectpassionshavebothcausesandobjects,whichmustbedistinguished.Theobjectofprideandhumilityisalwaystheself.Heretheviewalwaysfixeswhenweareactuatedbyeitherofthesepassions.Theobjectofloveandhatredisalwaysanotherperson.Ourloveandhatredarealwaysdirectedtosomesensiblebeingexternaltous(T2.1.2.2,2.2.1.2).4.Theindirectpassions,likeallpassions,donotpurporttorepresentanything.IreturntothisissueinPartIII.Becausethisaccountoftheindirectpassionsreliesheavilyonadoublerelationofimpressionsandideas,furtherdiscussionofitwillbedeferreduntilwehaveconsideredHumestheoryofrelations.ii.relationsInboththeTreatiseandtheEnquiryHumeintroducesthreeprinci-plesofassociation:resemblance,contiguity,andcauseandeffect.Theseareuniversalprinciples,whichrender[theimagination],insomemeasure,uniformwithitselfinalltimesandplaces(T1.1.4.1).IntheEnquiry,hesomewhatmorecautiouslysaysthatCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n80davidowenitisevident,thatthereisaprincipleofconnexionbetweenthedifferentthoughtsorideasofthemind,andthat,intheirappearancetothememoryorimagination,theyintroduceeachotherwithacertaindegreeofmethodandregularity(EHU3.1).IntheTreatise,thesethreeprinciplesofassociationarecallednaturalrelations,andaregiventhisnamebecauseineachsuchrelationoneideainvolun-tarilyornaturallyintroducesanother.Naturalrelationscontrastwithasecondsetofrelations,thephilosophicalrelations,orthoseinwhichoneideaisvoluntarilycomparedwithanother.AsHumesumsuphisview:Thewordrelationiscommonlyusdintwosensesconsiderablydifferentfromeachother.Eitherforthatqual-ity,bywhichtwoideasareconnectedtogetherintheimagination,andtheonenaturallyintroducestheother...orforthatparticularcircumstance,inwhich,evenuponthearbitraryunionoftwoideasinthefancy,wemaythinkpropertocomparethem(T1.1.5.1).Thesetwokindsofrelationrequireourattention.PhilosophicalRelationsHumedrawsattentiontophilosophicalrelationsinTreatise1.1.5andagainin1.3.1.Theearlierdiscussionisessentiallyalist,withabriefdescriptionofeachofthesevenphilosophicalrelations.Inthelaterdiscussion,hedividestheserelationsintotwoclasses.Thefirstclassincludesresemblance,contrariety,degreesinquality,andproportionsinquantityornumber.Humesaysofthesefourrelationsthattheydependentirelyontheideas,whichwecomparetogetherandillustrateshisviewbysaying:Tisfromtheideaofatriangle,thatwediscovertherelationofequality,whichitsthreeanglesbeartotworightones;andthisrelationisinvariable,aslongasouridearemainsthesame(T1.3.1.1).Heissaying,inotherwords,thataslongaswetaketheideaofatriangletobethatofanenclosedplanefiguremadeupofthreestraightlines,wewillfindthatthesumoftheinternalanglesofthisfigurewillbeexactlyequivalenttothesumoftworightangles.Thisisademonstrativejudgment,theconclusion,thatis,ofapieceofdemonstrativereasoningthecontraryofwhichcannotbeconceived,andthatresultsinknowledge.Itistemptingtothinkofthisfirstclassofrelationsasbeingbroadlylogicalincharacterorinstantiatingmetaphysicalnecessity.Butthiscannotberight.Suchrelationsincludedegreesinquality,ofCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind81whichHumegivesthisexample:Andthoitbeimpossibletojudgeexactlyofthedegreesofanyquality,suchascolour,taste,heat,cold,whenthedifferencebetwixtthemisverysmall;yettiseasytodecide,thatanyofthemissuperiororinferiortoanother,whentheirdifferenceisconsiderable(T1.3.1.2).IfIhaveanideaofsomethingbeingprettyhot,andanotherideaofsomethingbeingprettycold,thenaslongasthoseideasremainthesame,therelationbetweenthetwoideaswillalwaysbethesame.Notethatthisrecognitionthatonethingisnoticeablyhotterthananotherthingisanexampleofarelationknown,inHumesterms,withknowledgeandcertainty.Relationsofthesecondclass,thoseinvolvingtimeandplace,identity,andcausation,aredifferent.Theserelationsmaybechangdwithoutanychangeintheideas.Humesays,forexample,thattherelationsofcontiguityanddistancebetwixttwoobjectsmaybechangdmerelybyanalterationoftheirplace,withoutanychangeontheobjectsthemselvesorontheirideas;andtheplacedependsonahundreddifferentaccidents,whichcannotbeforeseenbythemind(T1.3.1.1).Considerthisexampleoftwoobjectssub-jecttosuchchanges.SupposeIhavetwocars,aJaguarandaBMW.Thesecouldbeside-by-sideinmydriveway.Thesamecarscouldequallywellbeseparatedbyacontinent,theJaguarinNewYorkandtheBMWinLA.Whatevertheirlocation,myideasofthesecarsremainexactlythesame,butthespatialrelationshipbetweenthemmayhavechanged.Or,supposeIthinkthatthegolfballIjusthitdownthefairwayisthesameonethatIusedwhenImadeahole-in-onelastweek.ButthenIrememberthatIputthehole-in-oneballawayasasouvenir,andsoitandtheonejuststruckcannotbeidenti-cal.Theideasofthesegolfballsareindistinguishabletheideasareeffectivelythesamebuttherelationshipbetweenthemisnotoneofidentity.Judgmentsregardingthissecondsetofrelationsarenot,accordingtoHume,demonstrativeandcertain,butonlyprobable.14HumesdistinctionbetweenthesetwoclassesofphilosophicalrelationsisreplacedintheEnquirywiththemorefamousdistinc-tionbetweenrelationsofideasandmattersoffact.Theformer14Forfurtherdiscussionofthisdifficultdistinctionbetweenthetwoclassesofphilo-sophicalrelations,seeDavidFateNorton,EditorsIntroductiontotheTreatise,I247,andtheAnnotationstoTreatise1.3,4467,bothintheOPTeditionoftheTreatise.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n82davidowenincludeseveryaffirmation,whichiseitherintuitivelyordemon-strativelycertainandwhichisdiscoverablebythemereoperationofthought,withoutdependenceonwhatisanywhereexistentintheuniverse.Anypropositionthatisintuitivelyordemonstrativelytrueissuchthatitsdenialwouldimplyacontradiction,andcouldneverbedistinctlyconceivedbythemind(EHU4.12).ThisgivesHumeacriterion:anypropositionthefalsehoodofwhichwecanconceivewithoutcontradictionisneitherintuitivelynordemon-strativelycertain.Itmustthereforefallintotheotherclass,matteroffact.15InboththeTreatiseandtheEnquiryHumealsodistinguishesbetweenintuitionanddemonstration,betweenthoserelationsthatwecanperceiveimmediately,andthoserelationsthatweperceiveastheresultofareasoningprocess.Threerelations,resemblance,contrariety,anddegreesinquality,are,Humesays,discoverableatfirstsight,andfallmoreproperlyundertheprovinceofintuitionthandemonstration.Weintuitarelationwhenwecanjustsee,withouttheaidofanyotherideas,thattwoideasstandinacer-tainrelationshipwitheachother.Thinkoftheideasofahotcoalandanicecube.Weunderstand,withoutfurtherthought,thatthequalitiesoftemperaturerepresentedbythoseideasaresignificantlydifferent.Nowconsiderarelationinvolvingproportionsinquantityornumber.Humeillustrateshispositionbymeansofageometricalexample,therelationshipoftheinternalanglesofatriangleandtworightangles.Thisisarelationshipofequality,butmostofuscannotintuitorseethatthisistrue.Wehavetodemonstratethistruthtoourselves,orproceed,asHumeputsit,inamoreartificialmanner(T1.3.1.23).Thisprocessofdemonstrationisamatterofconstruct-ingachainofideas,thefirstofwhichistheideaoftheinternalanglesofatriangle,andthelastofwhichistheideaoftworightangles.Eachoftheintermediateideasmustbeseentobeintuitively15HumesdistinctioninEHUbetweenrelationsofideasandmattersoffactseemsclosertoourdistinctionbetweennecessaryandcontingentthandoesthedistinc-tionfoundintheTreatise.ItisextremelyunclearhowandtowhatextentHumesthinkinghaschangedanddevelopedhere.ItisworthnotingthatHumeoftenusesthephrasematteroffacttomeanmattersoffactthatarenotobservedbythesenseormemory,i.e.,mattersoffactthataretheconclusionsofcausalinference.SeethequotationfromtheEnquiryinnote17below.Forfurtherdiscussionandreferences,seeNorton,EditorsIntroductiontotheTreatise,I33andnote38.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind83connectedtotheideaswithwhichitisadjacentinthechain.Weineffectseethattwothingsequaltoathirdthingareequaltooneanother.Inthiscasethewaytoproceedistodiscoverananglethatwecanjustseeisequaltoboththeinternalanglesofatriangleandtotworightangles.16Humealsodistinguishesbetweentheimmediateandtheinferen-tialinmattersoffact.Therelationsofidentityandoftimeandplacearegrasped,hesays,astheresultofamerepassiveadmissionoftheimpressionsthrotheorgansofsensation.Consequently,weoughtnottoreceiveasreasoninganyoftheobservationswemaymakeconcerningthem(T1.3.2.2).Inotherwords,weimmediatelysensethatoneimpressionorideaisorisnotidenticalwithanother,orwasorwasnotpriortoanother.Itisonlytherelationofcauseandeffectthatenables,andevenforcesus,tomovefromtheexpe-rienceofsomethingpresentlyobservedtotheideaofsomethingnotpresentlyobserved.17Butitisthenaturalrelationofcausationthathasthisall-importantfeature:Thusthocausationbeaphilo-sophicalrelation,asimplyingcontiguity,succession,andconstantconjunction,yettisonlysofarasitisanaturalrelation,andpro-ducesanunionamongourideas,thatweareabletoreasonuponit,ordrawanyinferencefromit(T1.3.6.16).NaturalRelations,orAssociationAllideasarederivedfromimpressions,andideascanstandinphilo-sophicalrelationstoeachother.ButmoreimportantlyforHume,ideasandimpressionscanstandinnaturalrelationstoeachother.Thecentralroleplayedbyassociationisoneofthemostdistinctive16InspiteofthefactthatthisgeometricalexampleisHumesown,afewpageslater,Humelimitsdemonstrativereasoningtoalgebraandarithmetic.Thisisbecauseinthosesubjectswehaveaprecisestandard,bywhichwecanjudgeoftheequalityandproportionofnumbers.Thisstandardistheunite,sothatwecanalwaystell,nomatterhowlongourdemonstrativechainofideas,whenonenumberhasalwaysauniteansweringtoeveryuniteoftheother(T1.3.1.5).HumeappearstohavedroppedthislimitationintheEnquiry.17ThepointiswellmadeintheEnquiry:Itmay,therefore,beasubjectworthyofcuriosity,toenquirewhatisthenatureofthatevidence,whichassuresusofanyrealexistenceandmatteroffact,beyondthepresenttestimonyofoursenses,ortherecordsofourmemory....AllreasoningsconcerningmatteroffactseemtobefoundedontherelationofCauseandEffect(EHU4.34).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n84davidowenaspectsofHumesscienceofhumannature,anaspectofwhichhewasproud,astheAbstractshows:Throthiswholebook,therearegreatpretensionstonewdiscoveriesinphilosophy;butifanythingcanentitletheauthortosogloriousanameasthatofaninventor,tistheusehemakesoftheprincipleoftheassociationofideas,whichentersintomostofhisphilosophy...thereisasecrettieorunionamongparticularideas,whichcausesthemindtoconjointhemmorefrequentlytogether,andmakestheone,uponitsappearance,introducetheother....Twillbeeasytoconceiveofwhatvastconsequencetheseprinciplesmustbeinthescienceofhumannature,ifweconsider,thatsofarasregardsthemind,thesearetheonlylinksthatbindthepartsoftheuniversetogether...theyarereallytousthecementoftheuniverse,andalltheoperationsofthemindmust,inagreatmeasure,dependonthem.(A35)WenotedHumescarefulempiricalestablishmentofthecopyprinciple(allsimpleideasarederivedfromsimpleimpressions),andsawthatthisisanaspectofhisdeterminationnottogobeyondexperienceandwhatcanbederivedfromexperience.Hisaccountoftheassociationofideasissimilarlyempiricalbutproceedsinadifferentfashion.Humesproofofthecopyprincipledependedonasurveythatledtoageneralization.Incontrast,theprinciplesofassociationarepostulatedtoexplainuniformitiesinphenomena.AcompletedefenceoftheseprinciplescannotbegivenwhentheyarefirstenunciatedinTreatise1.1.4.Infact,theadequacyofHumesaccountoftheprinciplesofassociationcanbejudgedonlybythesuccessorfailureoftheTreatiseasawhole.Humeobservesthatitiscommonthatthesamesimpleideas...fallregularlyintocom-plexones.Suchpatternswouldbeimpossiblewereideasentirelylooseandunconnected.Thustheremustbesomeuniversalprin-ciples,whichrender[theimagination],insomemeasure,uniformwithitselfinalltimesandplaces.Consistentlywithhavingfoundthattheimaginationisfreetojoinanytwosimpleideasitchooses,theprinciplesofassociationlinkideaswithagentleforce,whichcommonlyprevails,ratherthanbyaninseparableconnexion(T1.1.4.1).Humeliststhreequalities,fromwhichthisassociationarises,andbywhichthemindisafterthismannerconveydfromoneideaCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind85toanother(T1.1.4.1).Theseareresemblance,contiguityintimeandplace,andcauseandeffect.Hemakesnoattemptatthisstagetoshowthattherearenootherprinciplesofassociation.Hesimplyclaimsthatitisplainandevidentthatthesequalitiespro-duceanassociationamongideas,andupontheappearanceofoneideanaturallyintroduceanother(T1.1.4.12).InthefirstEnquiry,hedoesclaimthatthisenumerationiscompleatwhileadmittingthedifficultyofprovinghisclaim:Allwecando,insuchcases,istorunoverseveralinstances,andexaminecarefullytheprinci-ple,whichbindsthedifferentthoughtstoeachother,neverstoppingtillwerendertheprincipleasgeneralaspossible(EHU3.3).ThisisatypicalfeatureofHumesscienceofhumannature.Regulari-tiesatonelevelaretobeexplainedbyoverarchingprinciples,butwecannotexpectthatthoseprinciplesthemselvescanalwaysbeexplained.18TheUsesofAssociationAlthoughHumesaysintheAbstractthatitistheusehemakesoftheprinciplesofassociationofideasthatisinnovative,twoofhismostimportantusesofassociationinvolveimpressions.Humespecifi-callyspeaksoftheassociationofimpressions,notingthisremark-abledifference,thatideasareassociatedbyresemblance,contiguity,andcausation;andimpressionsonlybyresemblance(T2.1.4.3).Thissecondsortofassociationiscentralinexplainingtheindi-rectpassions.TheotheruseofassociationinvolvingimpressionsisperhapsthemostimportantuseofassociationintheTreatise.ThecentralargumentofTreatise1.3concernsprobableorcausalreason-ing,andtheproductionofbeliefsinwhatispresentlyunobservedorunobservableonthebasisofwhatispresentlyobserved,of,thatis,apresentimpression.1918AlthoughHumewantshismostbasicprinciplestobeasuniversalaspossible,herecognizestheimpossibilityofexplainingultimateprinciples(TIntro.8;TIntro.10).19HumesfirstexampleoftheassociationofideasintheEnquiryisaninstanceoftheassociationbetweenanimpressionandanidea:Apicturenaturallyleadsourthoughtstotheoriginal(EHU3.3).Animpressionofapictureleadsus,bymeansoftherelationofresemblance,tothinkofwhatthepictureisapictureof.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n86davidowenBeliefandCausalReasoningTheactofbelieving,forHume,istheactofassentingorjudging.ButHumescentralcaseofbeliefisnottheassentwemightattachtoanyproposition,whetherknownormerelybelieved,butratherthebeliefwehaveinunobservedexistents.Beliefintheseiscon-trastedwithknowledge,knowledgeofmathematicalpropositions,forexample.ThecentralissueofTreatise1.3ishow,onthebasisofpastexperience,wecometobelieveintheexistenceofobjectsoreventsthatwearenotcurrentlyexperiencing.Hisnegativeaccountisthatincomingtohavesuchbeliefs,wearenotdeterminedbyrea-son.20Hispositiveaccountinvolvesbothhistheoryofassociationandhisaccountofbelief.Pastexperienceshowsusthatonesortofobjectorevent,B,hasalwaysfollowedanothercontiguousobjectorevent,A.Thissetsupanassociation,sothatwheneverwehaveanimpressionofA,weareledtohaveanideaofBand,furthermore,tobelievethatBwillagainoccur.ButwhatisittobelievethatBwilloccur,asopposedtomerelythinkingorconceivingB?HumesansweristhatwebelievethatBwilloccurinthecomingexistenceofBwhentheforceorvivacityofourideaofBapproximatesthatwhichanimpressionofBwouldhave.Abeliefinsuchamatteroffactissimplyamoreforcefulorlivelyidea.Whatdistinguishesbeliefsfrommereconceptionsisthesamethingthatdistinguishesideasfromimpressions:forceandvivacity,orliveliness.2120Thishascometobeknownastheproblemofinduction.Thereis,ofcourse,ahugeliteratureonthesubject.SeeDonGarrett,CognitionandCommitmentinHume’sPhilosophy(NewYork:OxfordUniversityPress,1997),Chap.4;PeterMillican,HumesScepticalDoubtsconcerningInduction,inReadingHumeonHumanUnderstanding,ed.PeterMillican(Oxford:ClarendonPress,2002);andKennethWinkler,HumesInductiveScepticism,inTheEmpiricists:CriticalEssaysonLocke,Berkeley,andHume,ed.MargaretAtherton(Lanham,MD:Row-manandLittlefield,1999),183212.21ItisimportanttorealizethatonHumesview,asthatofDescartes,wecanmakejudgmentsorformbeliefswithonlyoneidea.Lockeheldthemoretraditionalview,whichHumedescribesastheseparatingorunitingofdifferentideas.Humethinksthisisamistakebecausetisfarfrombeingtrue,thatineveryjudgment,whichweform,weunitetwoideas;sinceinthatproposition,Godis,orindeedanyother,whichregardsexistence,theideaofexistenceisnodistinctidea,whichweunitewiththatoftheobject(T1.3.7,n.20).SeeDescartessdiscussionoftheideaofGodandtheexistenceofGodinMeditation3,andLockesaccountofbelief,opinionorassentinEssay4.1416.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind87Associationworksintwowaysincausalreasoningandthepro-ductionofbelief.Thefirstexplainswhyitisthatontheappearanceofonesortofimpressionthemindcomestohaveanothersortofidea.Humehereappealsexplicitlytotheprinciplesofassociation:Whenthemind,therefore,passesfromtheideaorimpressionofoneobjecttotheideaorbeliefofanother,itisnotdetermindbyreason,butbycertainprinciples,whichassociatetogethertheideasoftheseobjects,andunitethemintheimagination.Hadideasnomoreunioninthefancythanobjectsseemtohavetotheunderstanding,wecoudneverdrawanyinferencefromcausestoeffects,norreposebeliefinanymatteroffact.Theinference,therefore,dependssolelyontheunionofideas.(T1.3.6.12)Thisexplainswhythemindmakesthetransitionfromafirstimpressionorideatoanotheridea,butitdoesnotexplainwhythatideaisbelieved,andnotmerelyconceived.Forexample,ifIseeapatchofscorchedearth,Imightthenbelievethatthispatchwasrecentlyburned,forfireandscorchedearthhaveinmypastexpe-riencebeenassociatedas,respectively,causeandeffect.Butsofar,Humehasonlyaccountedfortheoccurrenceoftheideafireandnotformybeliefthattherehasbeenafire.Foranideatobeabelief,itmusthaveagreaterdegreeofforceandvivacity;itmustinthisrespectapproximatetoanimpression.22Butwhatisthesourceofextraforceandvivacitythatconstitutemybeliefthatafirehasscorchedtheearth?Thisforceandvivacitycomefromtheimpres-sionwithwhichtheideaisassociated.Again,Humeisexplicit:Iwoudwillinglyestablishitasageneralmaximinthescienceofhumannature,thatwhenanyimpressionbecomespresenttous,itnotonlytransportsthemindtosuchideasasarerelatedtoit,butlikewisecommunicatestothemashareofitsforceandvivacity(T1.3.8.2).Thethreeprinciplesofassociationarefeaturesofhumannature,andsoisthefactthatforceandvivacityistransmittedfromoneassociatedperceptionofthemindtoanother.BotharecrucialtoHumesaccountofcausalreasoningandtheproductionofbelief.2322Humedoesnotdistinguishthebelieffromwhatisbelievedorwhatwemightcallthecontentofthebelief.23ItisextremelydifficultnottoreadthisaccountintoHumesdiscussionofthesametopicintheEnquiry.ButinfacthenevermentionseitherofthesetwousesofassociationinEHU5,ScepticalSolutionoftheseDoubts.Nor,forthatmatter,doeshedescribewhatturnsanideaintoabeliefastheverysamethingasCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n88davidowenHumesaccountofthosebeliefsreachedbycausalorprobablerea-soningdoesnotexplaintheassentwegivetoknowledge.Concerningpropositionsknownbyintuitionordemonstration,Humesaysthattheperson,whoassents,notonlyconceivestheideasaccordingtotheproposition,butisnecessarilydetermindtoconceivetheminthatparticularmanner.IfoneunderstandsthatAisequaltoB,andthatBisequaltoC,onehasnochoicebuttosupposethatAisalsoequaltoC.Inotherwords,theassentthatattachestoknowledgecomesfromourinabilitytoconceiveanitemofknowledgetobefalse.Incontrast,sofarasprobableorcausalbeliefisconcerned,theimaginationisfreetoconceivebothsidesofthequestion,andbelief,aswehaveseen,dependsontheforceandvivacityoftheideaassociatedwithsomepresentimpression(T1.3.7.3).DoubleRelationHumesaccountoftheindirectpassionsissummarizedinTreatise2.1.Thebasicindirectpassionsareprideandhumility,loveandhatred.Althoughthecauseofeachoccurrenceofanyofthesepas-sionsvaries(Icanbeproudofmynewcar,ofmydaughtersgrades,andevenofthequalityofalocalrestaurant),Humethoughtitabsurdtothinkthatthereisadistinctorfundamentalfeatureofhumannaturethatexplainseachsuchinstanceofpride.Itismoreinkeepingwithascienceofhumannaturetothinkthateachoftheseitemscanproducepridebypartakingofsomegeneralquality,thatnaturallyoperatesonthemind(T2.1.3.5).Therelevantqualities,itturnsout,arethosethatproduceapleasureresemblingbutdistinctfromloveandpride,orapainresemblingbutdistinctfromhatredandhumility.Andthesequalitiesnaturallyoperateonthemindthroughasso-ciation.Considermyprideinmynewcar.WhenIsurveythiscarIseeanexpensiveandcolorfulobjectcapableofsafelytransportingmeandmyfamily.Thissurveycausesmetofeelpleasure.Butthefeelingofprideisalsopleasant,and,byresemblance(theonlyprin-cipleofassociationthatlinksimpressions),thefeelingofpleasurewhatdistinguishesanideafromanimpression,viz.,forceandvivacity.ThereisnoscholarlyconsensusastowhetherHumehaschangedhismind,orwhetherhehasjustsuppressedtechnicaldetailsforeaseofpresentation.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind89arisinginsuchcircumstancesisassociatedtoadistinctfeeling,thepassionofpride.Furthermore,asthecarismine,thesurveyofit,bytheassociationofideas,bringstomindtheideaofmyself,theideaofthepersonwhoisthenaturalobjectofmypride.Inasecondway,then,associationfacilitatestheproductionofpride.UsingHumestexttosumup:Thatcause[somequalityofmynewcar],whichexcitesthepassion[pride],isrelatedtotheobject[ofpride,myself],whichnaturehasattributedtothepassion;thesensation[pleasure],whichthecauseseparatelyproduces,isrelatedtothesensation[anotherpleasure]ofthepassion:Fromthisdoublerelationofideasandimpressions,thepassionisderivd.(T2.1.5.5)ThisdoublerelationofimpressionsandideasisanotherexampleoftheimportanceofassociationinHumesphilosophy.Ashesays:Anything,thatgivesapleasantsensation,andisrelatedtoself,excitesthepassionofpride,whichisalsoagreeable,andhasselfforitsobject.Itistheassociationofapleasantsensationwithpride,andthefurtherassociationoftheideaofwhatproducesthatpleasantsensationwiththeideaofself,thatexplaintheoccurrenceofpride.Wehaveseenthatassociationiscrucialinexplainingtheformationofcausalbeliefs.Humedrawsattentiontothesimilaritiesbetweenthataccountandhisaccountoftheindirectpassions.HeevenusesthesameexpressionakindofattractionthatheusedwhenhefirstdiscussedassociationatTreatise1.1.4.Intheformationofcausalbeliefs,hesays,thepresentimpressiongivesavivacitytothefancy,andtherelation[ofassociation]conveysthisvivacity,byaneasytransition,totherelatedidea....Thereisevidentlyagreatanalogybetwixtthathypothesis,andourpresentoneofanimpressionandidea,thattransfusethemselvesintoanotherimpressionandideabymeansoftheirdoublerelation:Whichanal-ogymustbeallowdtobenodespicableproofofbothhypotheses.(T2.1.5.8,1011)SympathyAnotherimportantconceptintroducedinBook2isthatofsympa-thy.Ittooishandledwithtechnicalfacility,andexplainedintermsofHumesaccountofimpressions,ideas,andassociation.IntheCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n90davidowenTreatise,atleast,sympathyhasaspecialmeaningforHume.Wesympathizewithotherswhenwereceivebycommunicationtheirinclinationsandsentiments,howeverdifferentfrom,orevencon-trarytoourown.Humeexplainsthisprocessintwostages.Wearefirstmadeawareofthesentimentoropinionofanotheronlybyitseffects,andbythoseexternalsignsinthecountenanceandconversa-tion,whichconveyanideaofit.Butthereismoretosympathythanhavinganideaofanothersfeeling.Bysympathy,therelevantideaisconvertedintoanimpression,andacquiressuchadegreeofforceandvivacity,astobecometheverypassionitself,andproduceanequalemotion,asanyoriginalaffection.Humehastoexplainboththesourceofourideaofanothersfeelingandhowthisideacomestohavetheextraforceandvivacitythattransformsitfromanideaofafeelingintotheveryfeelingitrepresents.Hisexplanationissimple:alivelyidea,indeedanimpression,ofourselfisalwayspresenttous,24andthus:Whateverobject...isrelatedtoourselvesmustbeconceivdwithalikevivacityofconception,accordingtothefore-goingprinciples,namely,theprinciplesofassociation.Thegreatertheresemblanceorcontiguitytousofthepersonfeelingtheoriginalsentiment,themorelikelyourideaofthatfeelingistobeconvertedintothefeelingitself:Thestrongertherelationisbetwixtour-selvesandanyobject,themoreeasilydoestheimaginationmakethetransition,andconveytotherelatedideathevivacityofcon-ception,withwhichwealwaysformtheideaofourownperson(T2.1.11.25).Resemblancefunctionshereintwoways.Thereisthegeneralresemblanceofournatures,sothatwehavesometendencytosym-pathizewithanyotherhumanbeing.Furthermore,wherethereisanypeculiarsimilarityinourmanners,orcharacter,orcountry,orlanguage,itfacilitatesthesympathy.Thegreatertheresem-blancebetweenourselvesandothers,themorelikelywearetosympathizewiththem.Contiguityandcausationalsoplayarole:Thesentimentsofothershavelittleinfluence,whenfarremovdfromus,andrequiretherelationofcontiguity,tomakethem24ThisclaimcreatesdifficultiesforHume.SeeJaneMcIntyre,PersonalIdentityandthePassions,JournaloftheHistoryofPhilosophy27(1989):54557,andDonaldAinslie,ScepticismaboutPersonsinBookIIofHumesTreatise,JournaloftheHistoryofPhilosophy37(1999):33361.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind91communicatethemselvesentirely.Therelationsofblood,beingaspeciesofcausation,maysometimescontributetothesameeffect(T2.1.11.56).25Asinhisaccountofbelief,thisexplanationofhowanideacanbeturnedintoanimpressionworksonlybecauseofHumesdoctrineofimpressionsandideas.Ifanideadiffersfromitscorrespondentimpressiononlyintermsofforceandvivacity,thenwecanexplaintheoperationofsympathysimplybyaccountingforthesourceoftheadditionalforceandvivacitythataccountsfortheconversionofarelevantideaintoaresemblingimpressionorpassion.Ingeneralterms,Humesaccountofsympathyrunsparalleltohisaccountofbelief.Ashesays:Letuscompareallthesecircumstances,andweshallfind,thatsympathyisexactlycorrespondenttotheoperationsofourunderstanding;andevencontainssomethingmoresurpriz-ingandextraordinary(T2.1.11.8).Theassociationofideasplaysacrucialroleineach,andineachcaseanideaisenlivenedbytheadditionofextraforceandvivacity.Inthecaseofbelief,thesourceoftheextraforceandvivacityisanassociatedimpression.Inthecaseofsympathythesourceisanever-presentimpressionoftheself,andbecauseofthestrengthofthisidea,sympathyisevenmoreextraordinarythanbelief,for,whilebeliefenlivensanideasothatitapproximatestoanimpression,sympathyactuallyturnsanideaintoanimpression.26Sympathycanenlivenanideasothatitbecomesan25InTreatise3.3.1,Oftheoriginofthenaturalvirtuesandvices,Humearguesthatsympathyisthesourceoftheesteem,whichwepaytoalltheartificialvirtues(T3.3.1.9).Becausesympathyisfacilitatedbyresemblance,contiguity,andcau-sation,theesteemwefeelwillhaveatendencytofavorthosewhoareclosesttous,mostresembleus,orarelinkedtousbytiesofblood.Sometimesthisisappropriate.Itispropertofavoronesownfamily,tosomeextent.Butingeneral,wedonotwantourjudgmentsofcharactertodependonthesefactors.Sowemustcorrectourmoralsentiments,injustthewaywecorrectoursenseimpres-sionsandtheinfluenceofthepassionsandourbeliefs.SeePartIIIofthisessay,andJ.L.Mackie,Hume’sMoralTheory(London:RoutledgeandKeganPaul,1980),Chap.5,aswellasRachelCohon,TheCommonPointofViewinHumesEthics,PhilosophyandPhenomenologicalResearch57(1997):82750.Seealsointhisvol-umetheessayHumesLaterMoralPhilosophy.26Thisdifferenceisimportant.Ifasensoryideaactuallyturnedintoasenseimpres-sion,itwouldpresumablybeahallucination.Partofthepointofbeliefistoenableustodealwithobjectsoreventsthatwearenotcurrentlyexperiencing,sobeliefscouldnotperformtheirroleiftheywereimpressions.Buttheymustimitateimpressionsinthesensethatbothinvolvejudgmentsaboutexistence.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n92davidowenactualpassion.Butsympathycanalsoenlivenotherkindsofideas.Itcan,forexample,enlivenbeliefssothattheytooaretransferredfromindividualtoindividual.Becauseofsympathy,Humesays,menofthegreatestjudgmentandunderstanding...finditverydifficulttofollowtheirownreasonorinclination,inoppositiontothatoftheirfriendsanddailycompanions(T2.1.11.2).27iii.representation,intentionality,andcorrectionSensationandRepresentationHumesfirstprincipleinthescienceofhumannatureisthatallsimpleideasarederivedfromsimpleimpressions,whicharecor-respondenttothem,andwhichtheyexactlyrepresent(T1.1.1.7).Ideasrepresentimpressions,andtheydosoinvirtueofresemblingthem,andbeingcausedbythem.28Becauseideasarecausallydepen-dentonimpressionsanddifferfromthemonlyinthedegreeoftheirforceandvivacity,theycaninanotherimportantrespectresem-bleimpressions:thecontentofthetwokindsofperceptioncanbeexactlyidentical.Butthisleadsustoaquestionofthegreatestimpor-tance.Ifideasrepresentimpressionsbecauseofcausaldependenceandresemblance,howandwhatdoimpressionsrepresent?Wewantourideaofthetelephoneinthenextroom,forexample,tobeanideaofanactual,physicalobjectinthenextroom,andnotmerelytheideaofanimpression,whichafterallisjustanotherperceptionofthemind.Butitlooksasifthiscanbesoonlyiftheimpressionwehadofatelephonewhenwewereinthenextroomrepresentedanactual,physicaltelephone.ThequestionWhatdoimpressionsrepresent?turnsouttobeextremelydifficulttoanswer.Inordertoaddressit,wemustfirstremembertodistinguishbetweenimpressionsofsensationandimpressionsofreflectionorsecondaryimpressions.Thequestion27Notethesuggestionthatthistendencyshouldberesisted.Hereagainwehaveacasewherecorrectionmaybenecessary.SeethediscussioninTreatise2.1.11.9,andlaterinthenextsectionofthisessay.28Humeisexplicitaboutthis:Ourideasarecopydfromourimpressions,andrepre-senttheminalltheirparts;Allideasarederivdfrom,andrepresentimpressions(T1.3.7.5,1.3.14.11).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind93aboutwhatandhowimpressionsrepresentdoesnotariseforimpres-sionsofreflection.Humeisclearaboutthis:apassion,theparadigmimpressionofreflection,is,hesays,anoriginalexistence,or,ifyouwill,modificationofexistence,andcontainsnotanyrepresentativequality,whichrendersitacopyofanyotherexistenceormodifica-tion(T2.3.3.5).29Sotheissueconcerningtherepresentativenatureofimpressionsislimitedtoimpressionsofsensation.30ButaswesawaboveinPartI,Humesupposesthatthesourcesorcausesofimpressionsofsensationarebeyondourreach.Suchimpressions,hesays,ariseinthesouloriginally,fromunknowncauses.Thisisenoughtoruleoutanyinvestigationofthecausesofsuchimpres-sions,andisinadditionconsistentwithhislimitationofenquirytoexperience.Ifexperienceitselfisconstitutedbyimpressions,thenitisclearthatthecauseofimpressionscannotitselfbeasubjectofempiricalinquiry.31Wecannot,forexample,sensiblyaskwhetherourimpressionofatreeiscausedbyandrepresentsatree.Wecaninquireintothecausesofourbeliefsabouttrees,asHumedoeswithgreatsubtletyin1.4.2(Ofscepticismwithregardtothesenses),butthatisadifferentmatter.Wecanalsoexaminethecharacterofoursensationsandhowtheyareinterrelatedtoeachotherandtootherperceptionsofthemind,butbeyondthatwecannotgo.Wemay,hesays,drawinferencesfromthecoherenceofourperceptions,butastothoseimpressions,whicharisefromthesenses,theirultimatecauseis,inmyopinion,perfectlyinexplicablebyhumanreason,andtwillalwaysbeimpossibletodecidewithcertainty,whethertheyariseimmediatelyfromtheobject,orareproducdbythecreativepowerofthemind,orarederivdfromtheauthorofourbeing(T1.3.5.2).3229Thisisnottosaythatnoquestionsofintentionalityarisewithrespecttothepassions.Whenwelove,ourloveisdirectedtowardanotherperson,andhungerisoftendirectedtowardacertainitemoffood.Wewillreturntothisissuebelow.30Weshouldnotethatthequestionofrepresentationdoesnotariseforallimpres-sionsofsensation.Humeclassifiesthepainsandpleasures,thatarisefromtheapplicationofobjectstoourbodiesasimpressionsofsensation(T1.4.2.12).31Thisignoranceislimitedtoimpressionsofsensation;wedo,ofcourse,knowsomethingofthecausaloriginsofimpressionsofreflection.Thedifferenceseemstobethatthecausesofimpressionsofreflectionarethemselvesperceptionsofthemind.32SeealsoT1.2.6.79;EHU12.1114.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n94davidowenConsequentlyoneanswertothequestionWhatdoimpressionsofsensationrepresent?is:Asfarasweknow,nothing!Impressionsofsensationmaynomorerepresentthandoimpressionsofreflection.33Ideasrepresentimpressions,becausetheyarecausedbyandresembleimpressions.ButHumesaysmorethanoncethatwecannotknowthecausesofimpressionsofsensation.Thismaybewhyhecallsthemoriginalimpressions.34Andifwecannotknowthecausesofoursenseimpressions,wecanneverbeinapositiontoknowwhethertheseimpressionsresem-bletheircauses.Ontheotherhand,wecertainlytalkasifoursenseimpressionsrepresentobjectsintheworldanddistinctfromourminds,andHumeowesusanaccountofthis.Thisheprovidesin1.4.2.Humeiswellawarethatwhathecallsthephilosophicalsystemdoessupposeexternalobjectstoresembleinternalpercep-tions.Butheclaimstohavealreadyshown,inargumentsmadeearlierinthesection,thattherelationofcauseandeffectcanneveraffordusanyjustconclusionfromtheexistenceorqualitiesofourperceptionstotheexistenceofexternalcontinudobjects.Hegoesontoaddthateventhotheycoudaffordsuchaconclusion,weshoudneverhaveanyreasontoinfer,thatourobjectsresembleourperceptions(T1.4.2.54).3533Thisisacontroversialclaim.Forafurtherdefense,seeRachelCohonandDavidOwen,Representation,ReasonandMotivation,Manuscrito20(1997):4776.Foracontraryview,seeDonGarrett,HumesNaturalisticTheoryofRepresentation,Synthese152(2006):30119.34Aswesawearlier,Humesometimesusesoriginaltodescribesomethingabouthumannatureforwhichwehavenoexplanation.Forfurtherdiscussion,seeMiriamMcCormick,HumeonNaturalBeliefandOriginalPrinciples,HumeStudies19(1993):10316.35Humethinksthathisdistinctionbetweenimpressionsandideas,andhisclaimthatwecannotknowthecausesofsensation,bringsclaritytothedebateaboutinnateideas.Onanaturalreadingofinnateness,thefactthatallideasarederivedfromimpressionsshowsthatnoideasareinnate.Butwhataboutimpressions?IntheEnquirythefactthatimpressionsarenotcopiesofothermentalitemsisenoughtoshowthatallimpressionsareinnate(EHU2,n.1).InthebodyoftheTreatise,Humeismorecautious,apparentlynotcommittinghimself(T1.1.1.12).TheAbstractisambiguous:itisevidentourstrongerperceptionsorimpressionsareinnate,andthatnaturalaffection,loveofvirtue,resentment,andalltheotherpassions,ariseimmediatelyfromnature(A6).Humecouldbesayinghereeitherthatallimpressionsareinnate(areadinginaccordwiththeEnquiry),orjustthatimpressionsofreflectionare.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind95ThefactthatHumedoesnotdiscussindetailwhatandwhetherimpressionsofsensationrepresentuntillateinBook1explainsanotherfact.ItisprettyclearthatHumetakessenseimpressionstobenotjustcontentbutalsojudgments,judgmentsabouttheexis-tenceofthings.ThenatureofthosejudgmentsisnotclarifieduntilOfscepticismwithregardtothesenses,butevenintheearlystagesoftheTreatiseHumeseemscommittedtosomethinglikethefol-lowingview:whenwehaveanimpressionofanapple,wedonotsimplyhaveanapple-likeexperience;wealsojudgeanappletoexist.Ofcourse,Humeisreluctantatthisstagetosaymuchaboutthis,becauseoftheproblematicnatureofourbeliefsintheworldweexperience.Nonetheless,hedoestalkaboutthebelieforassent,whichalwaysattendsthememoryandsensesandgoesontosaythatthisbeliefisnothingbutthevivacityofthoseperceptionstheypresent....Tobelieveisinthiscasetofeelanimmediateimpressionofthesenses,orarepetitionofthatimpressioninthememory.Tismerelytheforceandlivelinessoftheperception,whichconstitutesthefirstactofthejudgment(T1.3.5.7).Senseimpressionsarenotmeresensedata.Tohaveanimpressionistomakeajudgment.Ifitwerenotso,Humecouldnotexplainthenatureofbeliefsreachedbycausalreasoningbyshowingthattheseapproximatetoimpressions.IntentionalityandthePassionsHumesdiscussionoftheoriginsoftheindirectpassionsastheresultofadoublerelationofideasandimpressionsshowsgreattechnicalsophistication,andfurtherillustrateshiscommitmenttothefor-mationofascienceofhumannature.36AttentiontothisaspectofHumesaccountcanhelpsoften,perhapsdisarm,somefairlyobvi-ouscriticismsofhisaccountofthepassions,especiallywithrespecttointentionality.37WenotedthatHumethinksofallthepassionsassimpleanduniformimpressions,[sothat]tisimpossiblewecanever,byamultitudeofwords,giveajustdefinitionofthem36See,e.g.,Treatise2.2.2,Experimentstoconfirmthissystem.37Intentionalityisthatcharacteristicofamentalstateinvirtueofwhichitisoforaboutsomething.So,typically,wethinkthatourvisualrepresentationofatreeisoforaboutthetree.Butastatecanbeintentionalwithoutbeingrepresentational.WhenIamangrywithyou,myangerdoesnotrepresentyou,butitisdirectedatyou.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n96davidowen(T2.1.2.1).Whentakenoutofcontext,thismayseemimplausi-ble.Howcouldsomethingasrichandcomplexaslovebeasim-pleimpressionofreflection?Simpleimpressionsmayhavenologi-calrelations,buttheloveDesdemonabearsforOthellonecessarilyinvolvesOthello.Humecapturesthisfeaturewhenhesaysthattheobjectofloveisanotherperson.Itistruethat,onHumesaccount,therelationbetweenDesdemonaslove(animpression)andtheideaofOthelloiscausalandnotlogical.38Buttheindirectpassionsaredeterminedtohavetheirobjects,notonlybyanaturalbutalsobyanoriginalpropertyofthehumansoul.Nowthesequalitites,whichwemustconsiderasoriginal,aresuchasaremostinseparablefromthesoul,andcanberesolvdintonoother:Andsuchisthequality,whichdeterminestheobjectof[theindirectpassions](T2.1.3.23).39Love,asabasicfactofhumannature,justdoestakeanotherpersonasitsobject.AndDesdemonasloveforOthellowouldnotbetheloveitisifittooksomeotherperson,forexample,Cassio,asitsobject.Althoughstrictlyspeakingapassionisasimpleimpression,incapableofdefinition,eachpassionisboundinarelationalnexusofadoublerelationofimpressionsandideas,andcannotbeunderstoodindependentlyofthatcontext.40Humenotonlysaysthatpassionsaresimpleimpressions,healsosaysthattheydonotpurporttorepresentanything.Aswehaveseen,heinsiststhatapassioncontainsnorepresentativequality,whichrendersitacopyofanyotherexistenceormodification(T2.3.3.5).Thismaylooklikeanimplausibledenialoftheinten-tionalityofpassionsanddesires.41WeshouldfirstnotewhatHume38Herethenisapassionplacdbetwixttwoideas,ofwhichtheoneproducesit,andtheotherisproducdbyit.Thefirstidea,therefore,representsthecause,thesecondtheobjectofthepassion(T2.1.2.4).39Humeisactuallytalkingaboutprideandhumilityhere,butitisclearheintendsthesametobetrueoftheobjectsofloveandhatred.Noteagaintheuseoforiginaltodescribesomethingasabasic,unexplainedpartofhumannature.40Thatwemayunderstandthefullforceofthisdoublerelation,wemustconsider,thattisnotthepresentsensationaloneormomentarypainorpleasure,whichdeterminesthecharacterofanypassion,butthewholebentortendencyofitfromthebeginningtotheend(T2.2.9.2).41ItseemssoimplausiblethatAnnetteBaierregardsthisclaimofHumesasanisolatedmistakethatshouldbeignored.SeeAnnetteC.Baier,AProgressofSen-timents:ReflectionsonHume’sSentiments(Cambridge,MA.:HarvardUniversityPress,1991),Chap.6.Forfurtherdiscussionofthisissue,seeinthisvolumetheessayHumesMoralPsychology.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind97isconcernedtodenyinthispassage.Aswehaveseen,impressionsofsensationareasmuchjudgmentsastheyarepuresensations.Wenormallythinkofthisasmeaningthatanimpressionrepresentsanexternalobject,andthat,iftherepresentationisaccurate,ourjudg-mentistrue.42WhatHumeisclaimingaboutthepassionsisthattheyaredifferentlystructured.Havingapassionisnotakintomak-ingajudgment.Passionsorsecondaryimpressionsdonotcarrywiththemthisrepresentationalcharacter.Astheydonotpurporttocopyorrepresentanything,theycannotbetrueorfalse,reasonableorunreasonable.43Todenythatpassionsarejudgmentsisnottoanswerallques-tionsabouttheirintentionality.Theymaynotexpresstruthsorfalsehoods,buttheyarenonethelessdirected.IfDesdemonalovesOthello,thenthereisanalternativesenseinwhichOthelloisrepre-sentedbythepassion.OthelloistheobjectofDesdemonaslove.Thepassionofloveisasimpleimpression,sotheideaofOthellocan-notbepartofthatimpression.ButtheideaofOthelloistheobjectofthatpassion,andisbroughttomindbyit.Whenthepassionsareconsideredintheirrelationalcontext,theiralternativeformofintentionalitybecomesapparent.Thedirectpassionsarealsointentionalinthislimitedsense.Angeristypicallydirectedtowardanindividual,asislust.Butthesearedirect,notindirectpassions,andwethuscannotappealtothedoublerelationofideasandimpressionstoaccountfortheirinten-tionality.Nonetheless,directpassionsoftenbringtomindideas,justasthepassionofpride,forexample,bringstomindtheideaofself.Humesays,Thesensationsoflustandhungeralwayspro-duceinustheideaofthosepeculiarobjects,whicharesuitabletoeachappetite(T2.1.5.6).Andheexplicitlycomparesthiswiththeproductionoftheideaofselfbypride.Similarly,desireandaver-sionariseinresponsetoanimpressionorideaofpleasureandpain,andproduceinturnideasofvariousthingsthatcanproducethe42Humeisusuallyperfectlyhappytospeakthisway,inspiteofhisqualmsaboutimpressionsofsensationrepresentingsomethingexternaltothemind.43AlthoughHumeclaimsthatitisingeneralimpossible,thatreasonandpassioncaneveropposeeachother,heiscarefultoaddthattherearetwowaysinwhichapassionmaybecalldunreasonable...whenfoundedonafalsesupposition,orwhenitchoosesmeansinsufficientforthedesigndend(T2.3.3.7).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n98davidowenenvisagedpleasureorhelptoavoidpain.44SoHumeexplainstheintentionalityofallpassions,notjusttheindirectones,intermsofthecomplexnexusofrelationspassionsbeartovariousimpressionsandideas.And,ofcourse,passionsmuststandinsuchrelations,ortheywouldbeunabletointeractwithbeliefsinthedeterminationofthewillandtheproductionofbehavior.Beliefsaboutsourcesofpleasureandpainproducedesires,whichinturnproduceideasofwhatwillsatisfythosedesires.CorrectionHumenotesthatthepassionshavecommonlybeendividedintothecalmandviolent.Anexampleoftheformeristhesenseofbeautyanddeformityinaction,composition,andexternalobjects.Thelatterseemtoincludesuchpassionsasloveandhatred,griefandjoy,prideandhumility.ButHumethinksthatthisdivisionisfarfrombeingexact;indeeditisvulgarandspecious.Therapturesofpoetryandmusicfrequentlyrisetothegreatestheight;whilethoseotherimpressions,properlycalldpassions,maydecayintososoftanemotion,astobecome,inamanner,imperceptible(T2.1.1.3).Thedistinctionbetweencalmandviolentoccursagainlater,notasameansofclassifyingimpressionsofreflection,butasamannerofdescribingthedifferentwaysthatpassionscanbeexperienced.Thisisimportantforatleasttworeasons.First,reasonexertsitselfwithoutproducinganysensibleemotion.Therearealsocertaincalmdesiresandtendencies,which,thotheyberealpassions,producelittleemotioninthemind,andaremoreknownbytheireffectsthanbytheimmediatefeelingorsensation.Sinceitisnatural,butmistaken,toimagine,thatthoseactionsofthemindareentirelythesame,whichproducenotadifferentsensation,itiscommontomistaketheinfluenceofthecalmpassionsonourbehaviorfortheinfluenceofreason.Second,althoughboththecalmandviolentpassionsinfluencethewill,andhenceourbehavior,itisnaturaltothinkthatthemoreviolentthepassion,themoreeffectitwillhaveonus.Butthisisnotso.Humesaysthatwhatwecall44Themindbyanoriginalinstincttendstouniteitselfwiththegood,andtoavoidtheevil,thotheybeconceivdmerelyinidea,andbeconsiderdastoexistinanyfutureperiodoftime(T2.3.9.2).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind99strengthofmind,impliestheprevalenceofthecalmpassionsabovetheviolent(T2.3.3.8,10).45Atfirstglance,itlooksasifthereisadisanalogybetween,ontheonehand,theassentweattachtoourbeliefs(thatistotheforceandvivacityorlivelinessofideas)andthesubsequenteffectofbeliefsonourbehavior,and,ontheotherhand,apassionsviolenceandthesubsequenteffectofthatpassiononourbehavior.Abeliefinfluencesthewillbecauseitapproximatestoanimpression:theeffectofbeliefistoraiseupasimpleideatoanequalitywithourimpressions,andbestowonitalikeinfluenceonthepassions.Thiseffectitcanonlyhavebymakinganideaapproachanimpressioninforceandvivacity.Forasthedifferentdegreesofforcemakealltheoriginaldifferencebetwixtanimpressionandanidea,theymustofconsequencebethesourceofallthedifferencesintheeffects(T1.3.10.3).Sotheassentweattachtobeliefs,anditseffectonbehavior,justisequivalenttoanideasforceandvivacity,andHumeishappytospeakofthisforceandvivacityasafeelingorsentiment.46Iftheeffectofapassiononthewillwasexactlyparalleltothatofabelief,thenthestrongerinfeelingthepassion,themoreitwouldinfluencebehavior.Butthisisjusttosaythatapassionsinfluenceisdirectlyproportionaltoitsviolence.ButwehaveseenthatHumedeniesthisconclusion.Thisapparentdisanalogyiseasytooverstate.Inbothcasesthereisaninitialcorrelationbetweenthestrengthinfeelingoftheper-ceptionanditseffectonthewill.But,aswehaveseen,towardtheendofhisimportantdiscussion,Oftheinfluencingmotivesofthewill,Humeallowsthatthiscorrelationcanbecorrected,atleastwithrespecttothepassions.Itisnotalwaysthecasethatthemoreviolentapassionis,themoreitaffectsbehavior.Thereisavirtue,strengthofmind,thatcomeswithmaturityandreflection.Althoughweareinclinedtoactonourmostviolentimpulses,welearnnotto.Humefrequentlyspeaksofcorrectingourfeelingsorsentiments,andevenofcorrectingoursenseimpressions.Tisthustheunderstandingcorrectstheappearancesofthesenses,andmakesusimagine,that45Humeadmits,quiterightly,thatthereisnomansoconstantlypossessdofthisvirtue,asneveronanyoccasiontoyieldtothesollicitationsofpassionanddesire.Nonetheless,calmpassionscananddocounteractmoreviolentpassions.Humesaysthatitisnotthereforethepresentuneasinessalone,whichdeterminesus(T2.3.3.10).46See,e.g.,App.23.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n100davidowenanobjectattwentyfootdistanceseemseventotheeyeaslargeasoneofthesamedimensionsatten(T1.3.10.12).47Healsospeaksofcor-rectingthemoralsentiments,andthensays:Suchcorrectionsarecommonwithregardtoallthesenses;andindeedtwereimpossiblewecoudevermakeuseoflanguage,orcommunicateoursentimentstooneanother,didwenotcorrectthemomentaryappearancesofthings,andoverlookourpresentsituation(T3.3.1.16).48Ifwecancorrectourmoralsentiments,andevenoursenseimpres-sions,thensurelywecancorrectourbeliefs.ThisHumesuggestswecandoonatleasttwooccasions.First,hepointsoutthatwheneverwemakeacausaljudgment,therearebothessentialandsuperflu-ouscircumstances.Wherethereisagreatsimilarityofsuperfluouscircumstances,butfewofessentialones,wearetemptedtomakefalsecausaljudgments.Humesaysthatwemaycorrectthispropen-sitybyareflectiononthenatureofthosecircumstances.Thesec-ondcaseoccursintheverysameparagraphinwhichHumetalksaboutcorrectingtheappearancesofthesenses.Therehesays:Alikereflectionongeneralruleskeepsusfromaugmentingourbeliefuponeveryencreaseoftheforceandvivacityofourideas.Whereanopinionadmitsofnodoubt,oroppositeprobability,weattributetoitafullconviction;tho...itsforce[be]inferiortothatofotheropinions(T1.3.13.9,1.3.10.12).HereHumeexplicitlydifferentiatesbetweenthedegreeofbelieforassent,andtheamountofforceandvivacity,andthusitlooksasifbeliefscanbecorrectedinthesamewaythatsenseimpressions,passions,andmoralsentimentscanbecorrected.4947SeealsoT3.3.3.2.48Seenote25above.49Humesdistinguishingherebetweendegreeofassentandamountofforceandvivacityisproblematicforhistheoryofbelief.Accordingtotheofficialtheory,thedegreeofforceandvivacityjustisthedegreeofbelief,sonosuchdistinctionispossible.Itisnothardtothinkofwaystomaketheoverallaccountconsistent.Hecoulddistinguishfirst-andsecond-orderbeliefs:afirst-orderbeliefmaybeinitiallystrong,butwemighthaveasecond-orderbeliefthatbeliefsofthefirstsorttendtobemisleading.Thissecond-orderbeliefmightactuallycausetheforceandvivacity,thatis,degreeofassent,ofthefirst-orderbelieftodecline.IamnotsuggestingHumeactuallythoughtthis,thoughitisanotimplausiblewayofreadingWemaycorrectthispropensitybyareflectiononthenatureofthosecircumstances.Thewholeissueofthecorrectionofbelief,especiallyinthecontextofgeneralrules,includingtherulesbywhichtojudgeofcausesandeffects,isextremelyinterestinganddeservesfurtherstudy.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind101iv.conclusion:impressions,ideas,andthescienceofhumannatureHumesscienceofhumannaturewasmodeledonthephysicalsciences,whichachievedsuchstunningsuccessesintheseven-teenthandeighteenthcenturies.Theanalogybetweenhisintendedmethodologyforhisscienceofhumannatureandthephysicalsci-encesisrevealedinthefulltitleoftheTreatise:ATreatiseofHumanNature:BeinganAttempttoIntroducetheExperimentalMethodofReasoningintoMoralSubjects.Humethinksthathumanshaveanatureandthatitcanbeempiricallyinvestigated.Althoughprac-tionersofthescienceofhumannaturecannothopetoperformthecarefulandexactexperimentsofthephysicalsciences,theycangleanupourexperimentsinthissciencefromacautiousobserva-tionofhumanlife,andtakethemastheyappearinthecommoncourseoftheworld,bymensbehaviourincompany,inaffairs,andintheirpleasures(TIntro.8,10).Suchcautiousobservationisnotsimplyamatterofpayingattentiontowhatisgoingonaroundone;italsoinvolvesthestudyofhistoryandliterature,ancientandmodern:Wouldyouknowthesentiments,inclinations,andcourseoflifeoftheGreeksandRomans?StudywellthetemperandactionsoftheFrenchandEnglish:Youcannotbemuchmistakenintransferringtotheformermostoftheobservations,whichyouhavemadewithregardtothelat-ter.Mankindaresomuchthesame,inalltimesandplaces,thathistoryinformsusofnothingneworstrangeinthisparticular.Itschiefuseisonlytodiscovertheconstantanduniversalprinciplesofhumannature.(EHU8.7)TherelianceoncarefulobservationisthecentralfeatureofHumesmethodology,andithastwoconsequences.Oneishisdeter-minationnottogobeyondexperience,andtheotheristhelimitationofexplanation.Humestheoryofideasandimpressions,especiallythecopyprinciple,playsacrucialroleinbothofthese.Ifallideasareultimatelyderivedfromimpressions,from,thatis,whatweexpe-rience,theverycontentofourthoughtislimitedbywhatwehaveexperienced:wecangivenoreasonforourmostgeneralandmostrefindprinciples,besideourexperienceoftheirreality(TIntro.9).Itisimpossibletoexplaintheultimateprinciplesofhumannature;CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n102davidowenwejustobservethem.Humesthreeprinciplesofassociationarelikethis,asisthegeneraldesireofthegoodorthepleasantandthegen-eralaversiontothebadorthepainful.Histheoryofimpressionsandideasisthecentralpartof,andpartiallyinstantiates,theempiricalmethodologyofhisscienceofhumannature.Describedatthislevelofgenerality,HumesmethodologyandtheroleplayedbythetheoryofimpressionsandideasiscommonbothtotheTreatiseandtheEnquiry.ThisessayisastudyofhowHumestheoryofimpressionsandideasandhisprinciplesofassociationfeatureintheempiri-calmethodologyheusestoestablishsomeofhismostimportantpositions,first,inBooks1and2oftheTreatise,andthenlaterintheEnquiry.Simpleideasarederivedfromandexactlyresemblesim-pleimpressions,andthisfactiscrucialtoHumesempiricalestab-lishmentofthefirstprinciple...inthescienceofhumannature.Humehaslittletosayaboutsenseimpressionsandtheircauses,thoughitisclearthattheyarejudgments,andnotsimplysensedata.Beliefsandtheirproductioninprobablereasoningareexplained,notbyappealtoatraditionalfacultyofreason,butbytheprinciplesofassociationandthetransferofvivacityfromimpressionstoideas.Abeliefjustisaforceful,vivaciousidea.SooneofthemostimportantstoriesHumehastotellinBook1oftheTreatisereliescruciallyinallitsaspectsonthetheoryofimpressionsandideasandtheprinciplesofassociation.Inalikemanner,twoofthemostimportantaccountsinBook2centerroundbothassociationandthetheoryofimpressionsandideas.First,thesecomponentsareusedextensivelyinHumesthe-oryoftheindirectpassions,whichHumeexplicitlycomparestohisaccountofcausalbeliefs.Second,themechanismofsympathyturnsideasintoimpressions.Ifideasdidnotresembleandrepre-senttheimpressionsfromwhichtheywerederived,thisaccountwouldnotwork.Furthermore,associationisusedtoexplainjusthowthroughtheoperationofsympathyanideacanturnintoanimpression.Humealsocomparestheoperationsofsympathywiththeoperationsoftheunderstanding.Intheprocessofestablishingthecopyprinciple,thefirstprin-cipleinthescienceofhumannature,Humesaysthatthesub-jectoftheTreatiseisthefullexaminationofideasandimpres-sionsandofhowthesestandwithrespecttoeachother,especiallyCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheMechanicsofMind103causally.50Thisisaremarkableclaim,butthewayHumedevelopshisargumentsandpositionsbearsitout.AlltheissueswehaveexploredinBooks1and2comedowntotracingtherelationsideasbeartoimpressionsandtooneanother.Thishastheeffectofless-eningtheimportanceoffacultiesinHumesaccountoftheunder-standingandthewill.Aswehaveseen,Humearguesthatavolitionissimplyanimpressionwefeelwhenwemoveourbodyorgiverisetoanewperceptionofthemind.51Thewillisthefacultythatproducesvolitions,butappealtosuchafacultydoesnoworkforHume.Theworkisdonebytracingconnectionsamongsuchimpres-sionsandideasasdesires,aversions,andperceptionsofpleasureandpain.Humesaccountofreason,especiallycausalreason,issimilarlydeflationary.Reasonisthefacultythatgivesrisetobeliefs,butitisnotanappealtoreasonthatexplainswhywehavethebeliefswehaveortheirnature.Thatworkisdonebyappealtotheassociationofideassetupbypastexperience,andbytheenliveningofcertainideasrelatedtopresentimpressions.AlthoughHumeishappytotalkaboutthesenses,memory,andreason,justasheishappytotalkaboutthewill,alltherealworkofthemindisdonebythecon-nections,especiallytheassociativeandcausalconnections,amongideasandimpressions.suggestionsforfutherreadingInadditiontotheworkscitedinthenotestothisessay,forfurtherreadingthefollowingarerecommended.OntheScienceofHumanNatureBiro,John.HumesNewScienceofHumanNature.Thisvolume.OnIdeas,Impressions,andtheCopyPrincipleEverson,Stephen.TheDifferencebetweenFeelingandThinking.Mind97(1988):40113.Stroud,Barry.Hume.London:Routledge&KeganPaul,1977.Chap.2.50Letusconsiderhowtheystandwithregardtotheirexistence,andwhichoftheimpressionsandideasarecauses,andwhicheffects.Thefullexaminationofthisquestionisthesubjectofthepresenttreatise(T1.1.1.7).51Seenote13above.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n104davidowenOntheIndirectandDirectPassionsArdal,P´allS.´PassionandValueinHume’sTreatise,2nded.Edinburgh:EdinburghUniversityPress,1989.Chap.1.McIntyre,JaneL.HumesNewandExtraordinaryAccountofthePas-sions.InTheBlackwellGuidetoHume’sTreatise,editedbySaulTraiger.Malden,MA:BlackwellPublishing,2006.199215.OnBeliefBell,Martin.BeliefandInstinctinHumesFirstEnquiry.InReadingHumeonHumanUnderstanding:EssaysontheFirstEnquiry,editedbyP.Millican.Oxford:ClarendonPress,2002.17585.Norton,DavidFate.OftheAcademicalorScepticalPhilosophy.InRead-ingHumeonHumanUnderstanding:EssaysontheFirstEnquiry,editedbyP.Millican.Oxford:ClarendonPress,2002.37192.Owen,David.LockeandHumeonBelief,JudgmentandAssent.Topoi22(2003):1528.OnRepresentationandtheIntentionalityofthePassionsAlanen,Lilli.ReflectionandIdeasinHumesAccountofthePassions.InPersonsandPassions:EssaysinHonorofAnnetteBaier,editedbyJ.Jenkins,J.Whiting,andC.Williams.NotreDame:UniversityofNotreDamePress,2005.11742.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\ndonaldl.m.baxter4HumesTheoryofSpaceandTimeinItsSkepticalContextInTreatise1.2,Oftheideasofspaceandtime,Humeexaminesourideasofspatialextensionandtemporalduration,ourideasofgeometricequality,straightness,flatness,andmathematicalpoint,andourideasofavacuumandoftimewithoutchange.Humedoesnot,however,restricthisattentiontotheseideas;healsodrawsconclusionsaboutspaceandtimethemselves.Hearguesthatspaceandtimearenotinfinitelydivisible,thattheirsmallestpartsmustbeoccupied,andthatasaconsequencethereisnovacuumorintervaloftimewithoutchange.Histreatmentsofmattersbeyondthescopeofthesectionstitlehavereceivedharshcriticism.1Hisconclusionshaveseemedcontrarytomathematicsandphysics.Hismethodofarguingapplyingfeaturesofourmereideasofspaceandtimetospaceandtimethemselveshasseemedphilosophicallyinept.TheapparentsuccessofthesecriticismshasledtowidespreadneglectofthispartofHumeswork.Theneglectisunfortunate.InOftheideasofspaceandtimeHumegivesimportantcharacterizationsoftheskepticalapproachthatwillbedevelopedintherestoftheTreatise.Whenthatapproachisbetterunderstood,theforceofHumesargumentsconcerning1SeeNormanKempSmith,ThePhilosophyofDavidHume(London:Macmillan,1941),287;C.D.Broad,HumesDoctrineofSpace,ProceedingsoftheBritishAcademy47(1961):16176;AntonyFlew,InfiniteDivisibilityinHumesTrea-tise,inHume:ARe-evaluation,ed.D.W.LivingstonandJ.T.King(NewYork:FordhamUniversityPress,1976),25769;andRobertFogelin,HumeandBerkeleyontheProofsofInfiniteDivisibility,PhilosophicalReview97(1988):4769,andHumesSkepticismintheTreatiseofHumanNature(London:RoutledgeandKeganPaul,1985),2537.105CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n106donaldl.m.baxterspaceandtimecanbeappreciated,andtheinfluentialcriticismsofthemcanbeseentomissthemark.Humessystemconcerningspaceandtimeconsistsoftwoparts,whichareintimatelyconnectedtogether.First,heconcludesthatfiniteportionsofspaceandtimearenotinfinitelydivisible,butarecomposedofafinitenumberofsimpleandindivisibleparts.Second,heconcludesthattheseindivisiblepartsareinconceivableunlessoccupiedbysomethingrealandexistent,andsospaceandtimemusteachsimplybethemannerororder,inwhichobjectsexist.Inconsequenceitisimpossiblebothtoconceiveofemptyspaceandtoconceiveoftimewithoutanythingchanging(T1.2.4.12).TheargumentscomposingthefirstpartofHumessystemaretheonesmostderided.Notmanyreadersbotherwiththesecondpartofhissystem,sinceitis,byHumesowncharacterization,aconsequenceofthefirstandthusdependentonitssuccess.AnewevaluationofHumeonspaceandtimeshouldbegin,then,withanewinterpretationanddefenseofhismainargumentsagainstinfinitedivisibility.i.thefirstpartofhume’ssystemAgainstInfiniteDivisibilityHumemeanstoshowthatnoteverypartofspacehasparts.2TheargumentreliesonaclaimestablishedinSection1:thattheidea,whichweformofanyfinitequality,isnotinfinitelydivisible.IdeasforHumearelikeimages.3Justastheimageofachessboardinamirrorcanbeseentohaveparts,soanyideaofanythingthattakesupspacewillhaveparts.Noideawillhaveaninfinitenumberofparts,however.Thisconclusionfollowsfromtwopremises.Thefirstisthatthecapacityofthemindislimited,andcanneverattainafullandadequateconceptionofinfinity.Humemeanssimplythatthemindcannothaveaninfinitenumberofideas.Thesecondpremise2HedoesnotstatethisexplicitlyintheTreatise,butseeEHU12.18.3Here,followingHume,Iuseimageinasensethatextendsbeyondvisualimages.Cf.ThomasHobbes,HumanNature,inTheEnglishWorksofThomasHobbes,ed.W.Molesworth,11vols.(London,183945),4:39(Chap.2).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime107isthatwhateveriscapableofbeingdividedininfinitum,mustconsistofaninfinitenumberofparts.Fromthesetwopremisesitimmediatelyfollowsthatnoideaisinfinitelydivisible.Afterall,aninfinitelydivisibleideawouldhaveaninfinitenumberofparts,andsothemindwouldhaveaninfinitenumberofideas(T1.2.1.2).ThesecondpremisecontradictsalongtraditiongoingbacktoAristotle,whoconcludedthatsomethingdivisibleneednothavepartsthatexistactually,butneedonlyhavepartsthatexistpoten-tially.4However,arejoindercanbemadetotheAristoteliansonHumesbehalf.Somebutnotallofawholewouldbecomeitslefthalf,werethewholedividedinhalf.Therestwouldbecometherighthalf.Theformerisactuallyontheleft,andthelatterisactu-allyontheright.Theyactuallydiffer.Sincesomethingcantdifferfromitself,theyareactuallynumericallydistinct.Sincetheselesseramountsofthewholeareactuallydistinctfromeachother,thentheyareactuallypartsofthewhole.Soanythingdivisibleactuallyhasthepartsitisdivisibleinto.5Fromtheconclusionthatnoideaisinfinitelydivisible,Humeconcludesthatwehavesomeminimalideasideasthatcannotbedivided.Ashesays,Tisthereforecertain,thattheimaginationreachesaminimum,andmayraiseuptoitselfanidea,ofwhichitcannotconceiveanysub-division,andwhichcannotbedimin-ishdwithoutatotalannihilation.Likewisethereareminimalimpressions,whereimpressionsarethevividimageswegetinsen-soryorintrospectiveexperiences.Humedescribesanexperimenttoallowonetoseeaminimalvisualimpression:Putaspotofinkuponpaper,fixyoureyeuponthatspot,andretiretosuchadis-tance,thatatlastyoulosesightofit;tisplain,thatthemomentbeforeitvanishdtheimageorimpressionwasperfectlyindivisible(T1.2.1.34).NotethatHumeisnotsayingthatthespotofinkis4Aristotle,PhysicsBookIII,Chap.6,andBookVIII,Chap.8;andOnGenerationandCorruptionBookI,Chap.2.TheBasicWorksofAristotle,ed.R.McKeon(NewYork:RandomHouse,1941).5BaylegivesasimilarargumentagainstEpicureanatoms:PierreBayle,Histori-calandCriticalDictionary:Selections,trans.R.H.Popkin(Indianapolis:HackettPublishing,1991),s.v.ZenoofElea,noteG,p.360.Therewaswidespreadsup-portforHumespositiononthisissue.SeeThomasHolden,InfiniteDivisibilityandActualPartsinHumesTreatise,HumeStudies28(2002):pp.325,andTheArchitectureofMatter:GalileotoKant(Oxford:ClarendonPress,2004).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n108donaldl.m.baxterindivisible.Withasmalltelescopeonecouldcertainlydiscernitsparts.Rather,Humeistalkingaboutthevividimagecausedinthemindbythespotofink.Whenthespotofinkisclose,theimageofitinthemindhasmanyparts.Asthespotmovesfurtherawayitcausesasmallerimagewithfewerparts.Justbeforethespotistoofarawaytocauseanyimageatall,itcausesanimagethatcannotbefurtherdiminishedonewithnoparts.Thislastimageisindivisi-ble;itisaminimalimpression.Minimalideasaresimplylessvividcopiesofsuchminimalimpressions.ArmedwithminimalideasHumeproceedsinTreatise1.2toargueconfidentlythatspace,or(ashealsocallsit)extension,isnotinfinitelydivisible.Moreprecisely,hearguesthatnofiniteexten-sion,noportionofspacewithfinitelength,isinfinitelydivisible.Thebasisforhisargumentishisclaimthatminimalideasareade-quaterepresentationsofthemostminutepartsofextension.For,hesays,Whereverideasareadequaterepresentationsofobjects,therelations,contradictionsandagreementsoftheideasareallappli-cabletotheobjects(T1.2.2.1).Heisabouttoarguethatbecausetherearesmallest,indivisibleideasofthepartsofspace,therearesmallest,indivisiblepartsofspace.Theadequacyoftheideasissup-posedtojustifythisinference.6Whatmakesthemadequateisthatnothingcanbemoreminutethantheyare,sincetheseareideasandimagesperfectlysimpleandindivisible(T1.2.1.5).Beingtoosmalltobedivided,theyareperfectrepresentationsofanythingtoosmalltobedivided.Theargumentbeginswiththeassumptionusedpreviously,Everythingcapableofbeinginfinitelydividedcontainsaninfinitenumberofparts(T1.2.2.2).Soanyportionofspace,ifitisinfinitelydivisible,hasaninfinitenumberofparts.Soevenaportionofspacewithfinitelength,ifitisinfinitelydivisible,hasaninfinitenumberofparts.However,Humethinkshecanshowthatanyportionofspacewithaninfinitenumberofpartsmusthaveinfinitelength.Itwouldfollowthatitiscontradictorytotalkofaninfinitelydivisible6HumelikelyborrowsthisnotionoftheadequacyofanideafromLocke,whosays,Those[ideas]IcallAdequate,whichperfectlyrepresentthoseArchetypeswhichtheMindsupposesthemtakenfrom.AnEssayconcerningHumanUnderstanding,ed.P.H.Nidditch(Oxford:ClarendonPress,1975),2.31.1.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime109portionofspacewithfinitelength.Sotherecanbenosuchthing.HeusestheminimalideasfromSection1toexplainthecontradic-tion.Heformsaminimalideainhisimagination.Ithenrepeatthisideaoncethatis,heimaginesasecondminimalidearightnexttothefirst.Nowhehasanideaofthesmallestpossibleexten-sion,thesmallestpossiblelength.Notethat,becauseextensionhasparts,asingleminimalideaisnotanideaofextensionatall(seeT1.4.4.8).Onemustresistthetemptationtothinkthattwominimalideasaretwiceaslongasone.Oneminimalideahasnolengthatall;itissimplyasinglepoint,whereassomethingwithlengthmusthavetwodistinctendpoints.Thetwoformalengthsimplybecausetheyareattheirclosestapproachwhilestillbeingdistinguishablefromeachother,whichisafactaboutthemind,notaboutanybulktheminimalideashave.So,again,withthefirstrepetitionHumeformstheideaofthesmallestpossiblelength.NowIthenrepeatthisidea...twicethatis,heimaginesathirdminimalidearightnexttothesecond.Withtheadditionofathirdminimalidea,theoriginallengthdoubles.Inotherwords,totheoriginallengthcom-posedofthefirstandsecondminimalideasisaddedanequallengthcomposedofthesecondandthirdminimalideas.Onthethirdrepe-titionoftheoriginalidea,thatis,theadditionofafourthminimalidea,theoriginallengthtriples.Andsoon.OrasHumeputit,Ithenrepeatthisideaonce,twice,thrice,&c.andfindthecompoundideaofextension,arisingfromitsrepetition,alwaystoaugment,andbecomedouble,triple,quadruple,&c.tillatlastitswellsuptoaconsiderablebulk,greaterorsmaller,inproportionasIrepeatmoreorlessthesameidea(T1.2.2.2).FromhereHumemovesswiftlytohisconclusion.Heseesthatonlyanideacomposedofafinitenumberofminimalideaswillhavefinitelength.Sincethelengthisproportionaltothenumberofparts,anideacomposedofaninfinitenumberofminimalideaswouldbeinfinitelylong.Sincetheminimalideasareadequate,whatistrueofthemistrueofthesmallestpartsofextensionitself.Heconcludesthatnofiniteextensioniscapableofcontaininganinfinitenum-berofparts;andconsequentlythatnofiniteextensionisinfinitelydivisible(T1.2.2.2).Thisconclusionseemstooverlooktheobviousmathematicalpos-sibilitythatthepartsthatareaddedcouldbeproportionatelysmallerCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n110donaldl.m.baxtereachtime.Taketheadditionofproportionatelysmallerfractions,forinstance,1/2+1/4+1/8+1/16+1/32+···,andsoon.Witheachadditionthesumapproaches1.Nomatterhowcloseto1youmightchoosetobe,thesumofthewholeinfiniteseriesisevencloser.Forthisreasonthesumoftheserieshasbeendefinedas1.However,1isafinitenumber,nowherenearaninfinitenumber.Similarly,toapartoflength1/2couldbeaddedanonoverlappingpartoflength1/4andthenoneoflength1/8andthen...,andsoon,toarriveatawholeoflength1,adecidedlyfinitelength.Humeconsideredthisobjection,butrefusedtoacknowledgeitsforce.Inafootnotehesays,Ithasbeenobjectedtome,thatinfinitedivisibilitysupposesonlyaninfinitenumberofproportionalnotofaliquotparts,andthataninfinitenumberofproportionalpartsdoesnotformaninfiniteextension.Butthisdistinctionisentirelyfrivolous.Whetherthesepartsbecalldaliquotorproportional,theycannotbeinferiortothoseminutepartsweconceive;andthereforecannotformalessextensionbytheirconjunction.(T1.2.2.2n.6)Aliquotpartsareofuniformsize,whereasproportionalpartsareofsuccessivelysmallersize.7Humeissayingthatnopartsofextension,proportionaloraliquot,will,whenaddedtogether,yieldalengthsmallerthanthelengthyieldedbytheadditionofthesamenumberofminimalideas.Why?Becauseourminimalideasareadequaterepresentationsofthemostminutepartsofextension(T1.2.2.1).Perhapseverythinginspaceisbiggerthantheseminimalparts,per-hapssomethingsarethesamesize,butatleastnothingcanbesmallerthantheyare.OnemightworrythatHumeisassumingthattherearemostminutepartsofextensioninordertoprovethatthereare.Butheisnot.Heisonlyassumingthathisminimalideasareatleastassmallasanypartofextension.Humesproofthatfiniteextensionisnotinfinitelydivisibledependscrucially,evenexcruciatingly,ontheassumptionthatourminimalideasareadequate.Onthisassumptiondependstheinfer-encefromfeaturesofminimalideas,andfeaturesofcompoundsof7Bayledistinguishesthetwokindsofparts,thendefinesaliquotpartsaspartsofacertainmagnitudeandofthesametype.Bayle,HistoricalandCriticalDictionary,362,367.FormoreonthedistinctionseeMarinaFrasca-Spada,SpaceandtheSelfinHumesTreatise(Cambridge:CambridgeUniversityPress,1998),338.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime111them,tofeaturesofspaceorextensionitself.Itisthisassumption,however,thatseemstheleastdefensible.8Ideasaresimplyimagesinthemindthatmayormaynotaccu-ratelyreflectreality.HowcanHumeassumewithsolittleargu-mentthatinthecaseofminimalideasthereflectionisaccurate?Hisassumptionisespeciallytroublingsincethereflectionseemssoclearlyinaccurate,asappearseasytoshow.First,distinguish(asHumedoesnot)betweenimaginingandcon-ceiving.Descartesillustratesthedistinctionbypointingoutthatalthoughitisimpossibleaccuratelytoimagineachiliagonathousand-sidedclosedplanefigureitiseasytoconceiveofone.Ageometercanevenprovesomeofthepropertiesofsuchafigure,forinstance,thatthesumoftheinterioranglesis179,640degrees.9Second,grantthatwecannotimaginesomethingashavinganinfin-ityofparts.However,surelywecanconceiveit.Wecan,forinstance,conceiveofalineasasetofdimensionlesspointsasGeorgCantordid.10Third,grantthatwecanonlyimaginethesmallestpartsofalinebeingorderedsequentially,onenexttoanother.However,wecanconceiveofdifferentwayspointscanbeordered.Theymightbeordereddenselybetweenanytwopointsisanotherasaretherationalnumbers.11Orpointsmightevenbeorderedthewayacontinuumis,suchastherealnumbers,wherethereareevenmoreofthemthantherearerationalnumbers.12Fourth,notethatoftenourbestnaturalscienceusesmathematicssuchasgeometryorcal-culusforsuccessfulexplanationandprediction.Thissuccessfuluseofmathematicspresupposesthatspaceisnotasweimagineit,butisacontinuumthewayweconceiveittobe.Thus,ourbestsci-encetellsusthatspaceisnotthewayourimaginationsrepresentit.8BesidesFogelin,HumeandBerkeley,54,seeJamesFranklin,AchievementsandFallaciesinHumesAccountofInfiniteDivisibility,HumeStudies20(1994):85101.9OrasHumeputsit1996rightangles(T1.3.1.6).Thesumoftheinterioranglesofapolygonisequaltothenumberofsidesminus2,multipliedby180degrees.10SeeAdolfGrunbaum,ModernScienceandRefutationoftheParadoxesofZeno,¨inW.C.Salmon,ed.,ZenosParadoxes(Indianapolis:Bobbs-Merrill,1970),16570.11Therationalnumbersarethosethatcanbeexpressedasaratiooftwowholeintegers,e.g.,2/3.12Therealnumbersaretherationalnumbersalongwiththealgebraicirrationalnumberssuchasthesquarerootof2andthetranscendentalirrationalnumberssuchaspi(π).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n112donaldl.m.baxterThereforethereisnoreasontoacceptHumesassumptionthatourminimalideasareadequate.Theyarenotperfectrepresentationsofthedimensionlesspointsorderedasacontinuum,thewayscientistspresupposespacetobe.13ThetroublewiththiscriticismofHumesadequacyassumptionisthatitoverlookstheskepticalcontextofHumestreatmentofspaceandtime.Inthatcontextthiscriticism,powerfulasitis,remainsbesidethepoint.HumesSkepticismFromamodern,narrowlyepistemologicalstandpoint,HumeisaPyr-rhonianskepticwhoemulatesalateAcademicskeptic.14Whetherbyaccidentordesign,heisaPyrrhonianmoreinthewaydescribedbySextusEmpiricusthaninthewayHumedescribesPyrrhonism.PerhapsHumesinterpretationofPyrrhonismwassomewhatdis-tortedbythecommentatorshewasfamiliarwith,orperhapshefounditusefuljusttoappealtoawidelyextantinterpretation.15Inanyevent,tounderstandthecomplexitiesofHumesskepticism,ithelpstodisregardhisowncharacterizationsandtoapplythenuanceddescriptionsofancientskepticismgivenbytheancientsourcesthemselvesandsomeexactingcommentatorsofourown13ForstrandsofthislineofthoughtseeFlewandFogelincitedinnote1above.14HereIsupplementtheaccountofHumesskepticismgivenbyFogelinintheessayHumesSkepticisminthisvolume.HegivesthedetailsofHumesspecificskep-ticalarguments,whileIgiveamoregeneralcharacterization.Ourtreatmentsareconsistent,Ithink,exceptthatwhereasFogelinsaysthatHumestheoryofbeliefexplainshisPyrrhonism,IaminclinedtosaythatHumesPyrrhonismexplainshistheoryofbelief.Thisdisagreementmaywellbemerelyapparent,however,sinceFogelinseemstousePyrrhonisminHumesownsense.AnexpandedversionofthissectionappearsinChap.1ofmyHumesDifficulty:TimeandIdentityintheTreatise(NewYork:Routledge,2008).15SeeFredesextendedargumentthatthePyrrhoniansandearlierAcademicsweresubsequentlyinterpretedtobemoredogmaticthantheyreallywere.MichaelFrede,TheSkepticsTwoKindsofAssentandtheQuestionofthePossibilityofKnowledge,inEssaysinAncientPhilosophy(Minneapolis:UniversityofMin-nesotaPress,1987),20122.PopkinthinksHumescharacterizationofPyrrhonismfitsthisdescription.RichardH.Popkin,DavidHume:HisPyrrhonismandHisCritiqueofPyrrhonism,inHume:ACollectionofCriticalEssays,ed.V.C.Chap-pell(GardenCity,NY:AnchorBooks,1966),556.CommentatorsonwhichHumereliedwere,e.g.,Cicero,DiogenesLaertius,Montaigne,andBayle.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime113day.16Suchdescriptionsgiveussomekeyfeaturestolookforinaskepticspositionthatwemightotherwiseoverlook.ThePyrrhoniansdistinguishedtwokindsofassent(1)activeendorsementofaviewastruebasedonanappropriatereason,and(2)passiveacquiescenceinaviewforcedononebyappearances.Inseekingaftertruththeyfoundthatanyreasonsforendorsingaviewastruecouldbecounterbalancedbyreasonsfornotsoendorsingit.Theyfoundthemselvessuspended,unabletoendorseanyviewortheoppositeofanyview.Thissuspensionofjudgmentwasnot,contrarytothecontentionsofHumeandmanyofhispredecessors,supposedtobeasuspensionofallassentwhatsoever.ThePyrrhonians,whilecontinuingtoseekthetruthaboutreality,allowedthemselvestoacquiesceinwhateverviewhappenedtobeforcedonthembytheappearancesofthings.Iftheyappearedtobeattheedgeofacliffindangeroffalling,theywouldtrynottofall.Theywoulddoso,however,withoutendorsingastruetheviewsthatweremotivatingthem.ThelateAcademicswereabitmoredogmaticthanthePyrrho-nians.Theyfoundthattheyretainedcertainbeliefsafterweigh-ingtheconflictingarguments,andregardedthisfactasevidencethatsomebeliefsweremorelikelytruemoreprobablethanothers.Thustheyfounditplausiblethattruthcouldatleastbeapproached.1716IamrelyingformyremarksmainlyonSextus,Frede,andPopkin.SeeSextusEmpir-icus,OutlinesofScepticism,ed.J.AnnasandJ.Barnes(Cambridge:CambridgeUniversityPress,2000).SeealsoDavidFateNorton,DavidHume:Common-SenseMoralist,ScepticalMetaphysician(Princeton:PrincetonUniversityPress,1982),25579.AnnashintsatanaccountliketheoneIwillgive,butdoesnotdevelopitandendsupreadingHumeasadogmatist.JuliaAnnas,HumeandAncientScepticism,ActaPhilosophicaFennica66(2000):27185,especially276and279.MycharacterizationofPyrrhonismisselectiveinaccordancewithmodernepiste-mologicalconcerns,sothatIignore,e.g.,tranquillityasitsaim.17HumewouldhavebeenfamiliarwiththemodernrevivalofsuchaviewinMersenneandGassendi,itsdevelopmentbyAnglicantheologiansandtheRoyalSociety,anditsexpressioninBoyle,Newton,andLocke.SeeRichardH.Pop-kin,ConstructiveorMitigatedScepticism,inTheHistoryofScepticismfromSavonarolatoBayle(Oxford:OxfordUniversityPress,2003),11227;andHenryvanLeeuwen,TheProblemofCertaintyinEnglishThought:16301690(TheHague:MartinusNijhoff,1963).Forawide-ranginghistoryofreasoningwithoutcertaintyseeJamesFranklin,TheScienceofConjecture:EvidenceandProbabilitybeforePascal(Baltimore:JohnsHopkinsUniversityPress,2001).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n114donaldl.m.baxterThePyrrhoniansundercutphilosophy,science,andmathematics.ThelateAcademics,withtheirrelianceonprobablebeliefs,providedfortheseendeavors.HumeadmiredthemodernfloweringofsuchanapproachinNewton.18However,theAcademicsintheirdegreeofdogmatismwerestillvulnerabletoaPyrrhonian-typechallenge.HumegivesitinOfscepticismwithregardtoreason,wherehecritiquesanAcademicrelianceontheprobable.HearguesineffectthattheAcademicscannotjustifyanydegreeofbeliefinanycon-clusion(T1.4.1).However,HumeinventsaPyrrhonianfacsimileofAcademicprobability,asfollows.19LikethePyrrhonians,Humefindsnofinalreasonactivelytoendorseanyviewsastrue,orevenasprobable.Heiscontentpas-sivelytoacquiesceinwhateverviewsareforcedonhimbyappear-ances.Hesays,Afterthemostaccurateandexactofmyreason-ings,IcangivenoreasonwhyIshoudassenttoit;andfeelnothingbutastrongpropensitytoconsiderobjectsstronglyinthatview,underwhichtheyappeartome(T1.4.7.3).Hismaininnovationistomakedistinctionsbetweentherelativestrengthorweaknessoftheforcing.Itisbyfeelingthattheminddistinguishesthedegreetowhichanideaisforcedonit:Anideaassentedtofeelsdiffer-entfromafictitiousidea,thatthefancyalonepresentstous:AndthisdifferentfeelingIendeavourtoexplainbycallingitasuperiorforce,orvivacity,orsolidity,orfirmness,orsteadiness(T1.3.7.7).Becausetheyconcernfeeling,thedistinctionsHumemakesarelikeaestheticdistinctions:Tisnotsolelyinpoetryandmusic,wemustfollowourtasteandsentiment,butlikewiseinphilosophy.WhenIamconvincdofanyprinciple,tisonlyanidea,whichstrikesmorestronglyuponme.WhenIgivethepreferencetoonesetofargumentsaboveanother,Idonothingbutdecidefrommyfeelingconcerningthesuperiorityoftheirinfluence(T1.3.8.12).20Some-timesideasareimposedonusbyprinciplesofreasoningthatare18DavidHume,TheHistoryofEngland,fromtheInvasionofJuliusCaesartotheRevolutionin1688,Vol.6(Indianapolis:LibertyClassics,1983),Chap.70.19ItmaybethatHumewasfollowinginthefootstepsoftheearlyAcademicCarneades,ifCarneadesisreadasaclassicalskepticinFredessense;seeTheSkepticsTwoKindsofAssent,201.20Humesrestrictioninthispassagetobeliefasaresultofprobablereasoningisnorealrestrictionsincehethinksallknowledgeresolvesitselfintoprobability(T1.4.1.4).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime115changeable,weak,andirregular.Theirinfluencecanbeundercutbyaduecontrastwithideasimposedonusbyprinciplesthatarepermanent,irresistible,anduniversal(T1.4.4.1).Inthiswaywemighthopetoestablishasystemorsetofopinions,whichifnottrue(forthat,perhaps,istoomuchtobehopdfor)mightatleastbesatisfactorytothehumanmind,andmightstandthetestofthemostcriticalexamination(T1.4.7.14).ThusHumeisabletodistinguishbetween,ontheonehand,viewsthatwouldremainstablethroughtimeandfromplacetoplace,and,ontheotherhand,viewsthatwouldvarybytimeorplace.Thelatterwouldincludesuperstitions,myths,thefictionsoftheancientphilosophers.Someofourstableviewswillbefundamentalcommonsensebeliefs,suchasthoseintheunitaryselfandtheexternalworld,butthereisalsoroomforviewsinphilosophy,science,andmathematics.AndsoHumemakesroomforrefindreasoningandthemostelaboratephilosophicalresearchesinhisskepticalapproach.21Likeacriticintheartswhotriestodistinguishclassicworksfrompassingfancies,Humetriestodistinguishthemoststableoftheviewsforcedonusbyappear-ances.HeextendsthePyrrhonianapproachbeyondactivedailylifeintotheoreticalmatters.ThusHumeisaPyrrhonianskepticwhosediscriminationconcerningtheforceandstabilityofpassiveacquies-cenceinviewsgiveshimaprocedurefordoingphilosophy,science,andmathematicswithresultsakintothoseofthelaterAcademicskepticsandtheirmodernheirs.ThedifferenceisthatHumemakesnoconnectionbetweenthedegreetowhichaviewisforcedonhimandthedegreetowhichitislikelytobetrue.Muchsuccessinarrivingatstabilityistobehopedforwhenourviewsconcernsensoryappearancesofobjects.Thereareviewspro-vokedbyothersortsofappearances,butfewasstable:Aslongasweconfineourspeculationstotheappearancesofobjectstooursenses,withoutenteringintodisquisitionsconcerningtheirrealnatureandoperations,wearesafefromalldifficulties,andcanneverbeembar-rassdbyanyquestion(T1.2.5.26n.12).Notetheappealtoobjectsastheyappeartothesenses.Humesskepticismexplainshisempiri-cism.Thatis,hisassentonlytoviewsforcedonhimbyappearancesexplainshisrelianceonexperience.Hesaysthatwecannever21ThephraserefindreasoningisfromT1.4.7.7andelaboratephilosophicalresearchesfromT1.4.7.15.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n116donaldl.m.baxterpretendtoknowbodyotherwisethanbythoseexternalproperties,whichdiscoverthemselvestothesenses.Extensionanddurationwillturnouttobetwosuchproperties.Whendiscussingthemhecontentshimselfwithknowingperfectlythemannerinwhichobjectsaffectmysenses,andtheirconnexionswitheachother,asfarasexperienceinformsmeofthem(T1.2.5.26).Thisapproachistheonehepromisesinintroducinghisscienceofman.Thushesaysthattheonlysolidfoundationforhistheoryofhumannaturemustbelaidonexperienceandobservation(TIntro.7).Thisempiricism,whichhelearnedfromBoyle,perhapsNewton,andothers,HumetakessimplytobeanaspectofhisPyrrhonism:Andthowemustendeavourtorenderallourprinciplesasuniversalaspossible,bytracingupourexperimentstotheutmost,andexplainingalleffectsfromthesimplestandfewestcauses,tisstillcertainwecannotgobeyondexperience;andanyhypothesis,thatpretendstodiscovertheultimateoriginalqualitiesofhumannature,oughtatfirsttoberejectedaspresumptuousandchimerical.(TIntro.8)Aswiththestudyofhumannature,sowithallsciences.Noneofthemcangobeyondexperience,orestablishanyprincipleswhicharenotfoundedonthatauthority(TIntro.10).HumesPyrrhonianEmpiricismisthusconcernedonlywithobjectsastheyappeartousinexperience.22Someincautioususesofthewordsrealandreallymightbethoughttobeliethisclaim.23WhencharacterizinghisskepticismHumecontraststheappearancesofobjectswiththeirunknowablerealnatures,asonewouldexpect.However,oncehetakesittobeunderstoodthatheisconfininghisattentiontotheseappearances,hesometimesfeelsfreetoreapplythecontrastbetweenappearanceandreality.Ineffectheisthendistinguishinghowanobjectreallyappearsfromhowitapparentlyappearsacontrastallowedbyhisdistinctionbetweenmorestableandlessstableviewsofthings.Atothertimeshecontrastsbeingrealwithbeinganonentity,or,in22Cf.Berkeleysdistinctionbetween,forinstance,soundasitisperceivedbyus,andasitisinitself,andhisexclusiveconcernwiththeformer.GeorgeBerkeley,ThreeDialoguesbetweenHylasandPhilonous,inTheWorksofGeorgeBerkeley,BishopofCloyne,ed.A.A.LuceandT.E.Jessop,9vols.(London:ThomasNelsonandSons,194857),2:1745,1802.23Iwillnotetheseinthecourseoftheessay.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime117otherwords,withbeingsomethingthatappearsasopposedtobeingnothing.Sometimesheusesinrealitytodistinguishaviewforcedonthemindfromtheviewbeingcriticized.Sometimeshecontrastsbeingrealwithbeingonlyinthemind,thoughagainthiscontrastiswithintheworldasitappears.Suchusesofrealimplynodogmaticalspiritanymorethanphrasessuchastisevident,tiscertain,tisundeniable.Thepropensitytousesuchexpressionsis,asHumesays,sonaturalthattheywereextortedfrommebythepresentviewoftheobject(T1.4.7.15).Butdespitesuchexpressions,heisstillconfininghisattentiontoobjectsastheyappeartousinexperience.OneofHumesinnovationswastofindoutaboutobjectsastheyappeartousbyexaminationoftheideasweusetorepresentthem.Forinstance,hecomestotheconclusionthatamathematicalpointmusthavecolororsolidityinordertobeanentitythatcanbyitsconjunctionwithothersformarealexistence(T1.2.4.3).24Inthecourseofthisinvestigationheasks,Whatisourideaofasimpleandindivisiblepoint?andcomments,Nowonderifmyanswerappearsomewhatnew,sincethequestionitselfhasscarceeveryetbeenthoughtof.Wearewonttodisputeconcerningthenatureofmathematicalpoints,butseldomconcerningthenatureoftheirideas(T1.2.3.14).Examiningourideascanhelpinfindingouthowobjectsappeartous.Whenourideasareobscure,wecansettlecontroversiesaboutthembyexaminingtheimpressionsfromwhichthepartsoftheideasarecopied(T1.2.3.1).Humesfamouscopyprinciplethatallsimpleideasarecopiedfromimpressionsispartofhisskepticalviewthattheultimatesourceofallourviewsisappearances(seeT1.1.1.7).AsPyrrhonian,Humedoesnotevenactivelyendorsethephilo-sophicalframeworkheusestocharacterizehisskepticalapproach.Thatthereareexternalphysicalobjects(whatHumecallsingeneralbody)isnotsomethingHumefindsanyjustificationtobelieve.Itisabeliefforcedonus.Ashesays,tisinvaintoask,Whethertherebebodyornot?Thatisapoint,whichwemusttakeforgrantedinallourreasonings(T1.4.2.1).Further,thatthereisaninternalworldofperceptionscausedbytheexternalworldofobjectsisabeliefphilosophersfindthemselveswithwhentheyreflectonexperience.24HereHumecontrastsbeingrealwithbeinganonentity.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n118donaldl.m.baxterThisbelieflacksjustificationaswell,yetitisdurable.OnemightfocusonpassagesinwhichHumeisquestioningthisphilosophicalframework,oristhinkingalongthelinesofordinarynonphilosoph-icalpeople,toconcludethatbytalkabouttheexternalworldhemeanstobetalkingaboutourimpressionsofit.Butwhendoingphi-losophyonecannotshaketheframeworkforlong(T1.4.2.4157).25WhenHumeisoperatingwithinit,hemeansreallytobetalkingabouttheexternalworldasitisexperiencedviaourimpressionsandasitisrepresentedbyourmoststableideas.Thisframeworkmayseemforamomenttobelessjustifiedthantheconclusionthatthereareonlyperceptions,buttheframeworkisactuallymorestable.WemightbemisledbyHumesclaimthathisphilosophypre-tendsonlytoexplainthenatureandcausesofourperceptions,orimpressionsandideas(T1.2.5.26).Wearemoreeasilymisledifweconfuseobjectsastheyappearwiththeimpressionstheyoccasion.CertainlyHumedevotesattentionandcaretocharacterizingourperceptions.ButthecontextoftheaboveremarkisaparagraphinwhichHumeiscontrastingwhatheisnottryingtodopenetrateintothenatureofbodieswithwhatheistryingtodotoknowbody...bythoseexternalproperties,whichdiscoverthemselvestothesenses.Partofthetaskinexaminingperceptionsistodiscoverthepropertiesoftheobjectsofthoseperceptions,thatis,objectsastheyappeartous.Inotherwords,asPyrrhonian,heisconcernedtocharacterizetheexternalworldbythoseappearancesthatforceviewsonus.TheSkepticalContextofHumesArgumentHumesassumptionthatourminimalspatialideasareadequaterep-resentationsoftheleastpartofextensionfallsoutofthisskepticalapproach.Ourideasoftheexternalworldareimagesoftheworldasitappearstous.Beingcopiedfromsenseimpressions,theircharac-teristicsaredeterminedbythewayexternalobjectsdiscoverthem-selvestothesenses(T1.2.5.26).Notallcomplexideasarefaithful25Theconsiderationsagainstaframeworkofbothanexternalworldofobjectsandaninternalworldofperceptionsadmitofnoanswerandproducenoconviction,asHumefamouslysaysofBerkeleysphilosophy.SeeEHU12.15n.32.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime119totheworldasitappearsbecauseofthelibertyoftheimaginationtotransposeandchangeitsideas(T1.1.3.4).Byrearrangementoftheirpartswecaninventnewcomplexideasdifferingfromanycompleximpressionswehavehad.Butsimpleideasarecompletelyforcedonus.Theygiveustheelementsoftheworldasitappears.Theyareadequaterepresentationsoftheseelements.Andsoourminimalspatialideasareadequaterepresentationsoftheleastpartofextensionasitappears.Iftheycanonlybeorderedsequentially,onenexttotheother,thenthesameholdsofextensionasitappears.Ifnoextendedideaoffinitelengthcancontainaninfinitenumberofadequateminimalideas,thennoextensionasitappearscancontainaninfinitenumberofparts.TalkbyHumescriticsofinfinitelydivisiblespaceasasetofpointsorderedasacontinuumwouldforhimamountmerelytoemptywords.Giventhecopyprinciple,sincetherecouldbenoimpressionofsuchasetwithitselementsorderedinsuchaway,therecouldbenoideaofit.Beyondthat,thisconceitofamathemat-icalcontinuumisnotforcedonusbyappearances,andthereisnowaytoknowifitistrueofreality.26Ifwetrytoapplyittotheworldasitappears,weendupincontradiction.Notonlyisassenttotheconceitnotforced,dissentfromitisforced:Fortisevident,thatasnoideaofquantityisinfinitelydivisible,therecannotbeimagindamoreglaringabsurdity,thantoendeavourtoprove,thatquantityitself[asitappearstothesenses]admitsofsuchadivision;andtoprovethisbymeansofideas,whicharedirectlyoppositeinthatparticular(T1.2.4.32).Ihaveaddedthephraseasitappearstothesensesasareminderoftheskepticalcontextofallsuchinferencesfromideatoobject.ThephrasehelpsbringouttheforceofHumesremarkjustquoted:onecannotuseideasthatareadequaterepresen-tationsofquantityasitappears,inordertoprovethatquantityasitappearsisdifferentthanitappears.Humesincautioususeofthephrasesreallyimpossibleandcontradictoryandrealqualityofextensionintheargumentat1.2.2.12mustnotmisleadusaboutHumesskepticism.Heismerelyexpressingtheviewsforcedon26Eveniftakingspacetobeacontinuumyieldsthebestexplanationforthesuccessfuluseofgeometryandthecalculus,thereisnoconclusivereasonwhythebestexplanationmustbetrue.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n120donaldl.m.baxterhimbyappearances,whateverincautiousexpressionshappentobeextortedfromhim.27IfoneisstilldubiousaboutHumesappealtoideastoprovehispoint,thenoneshouldreturntoimpressions.Lookatgrainsofsandfromthefurthestdistanceatwhichtheyarestillvisible,perhapsarmslength.28Movethemaroundwithaneedle.Onestooclosetogethercannotbedistinguishedfromasingleone.Pairsthatcanbedistinguishedfromasinglegrainformatinyextension.Theapparentlengthoftheextensiongrowswiththenumberdiscerniblyadded,justasHumesays.Thelengthisproportionaltothenumberdis-cerniblyadded,justasHumesays.Itisofnousetolookmorecloselytoseehowthegrainsarereallyarrangedandifthereisreallyanyspacebetweenthemoriftheyreallytouch.Thegrainsarenottheconcern;ofconcernareonlytheimpressionscausedbythematthemaximumdistanceatwhichthegrainsarevisible.Fromtheseimpressionstheconclusionisforcedononethatanextensionofmerelyfinitelengthcannotbebuiltupfromaninfinitenumberofsuchparts.29Itseemsobviousthatspaceasitappearsisnotinfinitelydivisible.Aswithspace,sowithtime,Humethinks.Headdsoneargu-mentpeculiartotimethat,again,isbestunderstoodasanargumentconcerningtimeasitappears.Inourexperience,allmomentsoccursuccessively,oneimmediatelyaftertheother.Sotheyexistoneatatimeinthepresent.Soonlythepresentmomentexists.Thusifthepresentmomentweredivisibleintofurthermomentsitspartswouldeachbepresent.Sotheywouldcoexist.Butsinceallmomentsaresuccessive,nomomentscoexist.Soifallmomentsweredivisible27WhenHumesays,Thatwhateverappearsimpossibleandcontradictoryuponthecomparisonofthese[adequate]ideas,mustbereallyimpossibleandcontradictory,withoutanyfurtherexcuseorevasion,heistalkingaboutwhatisreallytrueofwhattheideasareadequateof,inotherwords,whatisreallytrueoftheworldasitappears.Whenhespeaksofarealqualityofextensionheisspeakingofextensionasitreallyappears(T1.2.2.12).ForfurthersupportoftheclaimthatHumeisconcernedwithspaceandtimeasapprehendedbyus,seeMichaelAyers,BerkeleyandHume:AQuestionofInfluence,inPhilosophyinHistory:EssaysontheHistoriographyofPhilosophy,ed.R.Rorty,J.B.Schneewind,andQ.Skinner(Cambridge:CambridgeUniversityPress,1984),31314.28Formeitwasarmslength.Thedistancewilldependongrainsizeandcolor,amountoflight,backgroundcolor,onesvisualacuity,etc.29Acarefullydesignedexperimentwouldbeneededtoconclusivelyestablishtheseconclusions.Iwriteonlyfrommyowncrudeexperiment.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime121intofurthermoments,thepresentmomentwouldbeaswell,soitspartswouldbecoexistingmomentsthatdonotcoexistanarrantcontradiction(T1.2.2.4).HumesPyrrhonianEmpiricismispresupposedbyhismainargu-mentsconcerningspaceandtime.Allhisconclusionsaboutspaceandtimeshouldbereadasclaimsaboutspaceandtimeastheyappear.IfoneisgoingtodisagreewithHumeoninfinitedivisibility,thenoneisgoingtohavetotakeissuewithhisPyrrhonianEmpiri-cismbyshowing,forexample,thatknowledgeispossible,orthatthereisanotherwaytobeaskepticthanbyacquiescingtoviewsforcedononebyappearances,orbyshowingthattheappearancesoftheexternalworldarenotalwayssocloselytiedtowhatiscon-veyedbythesenses.Onecannot,however,simplyaccuseHumeofineptlyreasoningfromideatorealityorofmakingmathematicalblunders.Suchaccusationsoverlookhisscepticalapproachtoissuesconcerningspaceandtime,andareirrelevanttojudgingitssuccess.TheGeneralArgumentagainstGeometricProofsofInfiniteDivisibilityHumegivesvariousotherargumentsthatspacehasindivisibleparts,thatis,partswithoutparts.30Hesupportsthesewithacrucialgen-eralargumentagainstanysupposedgeometricproofoftheinfinitedivisibilityofspace.Humearguesthatnosuchproofcansucceed(T1.2.4.1731).Hedoessobydenyinggeometrytobeascienceexactenoughtoadmitofconclusionssosubtile[subtle]asthosewhichregardinfinitedivisibility(A29).TheimportancehegivestothisargumentisevidencedbyitsmentionintheAbstractapiecewrit-tentopromotetheTreatise.Besidesthelongsummaryofhisviewsoncausationandcausalinference,theonlyotherhighlightsofBook1oftheTreatisementionedtherearehisbundletheoryoftheselfandhisgeneralargumentagainstgeometricproofsofinfinitedivisibility.Allsuchproofspretendtobedemonstrationsbasedonprecisedefinitionsandexceptionlessgeometricaxioms.Humearguestothecontrarythattheyarenotproperlydemonstrations,beingbuiltonideas,whicharenotexact,andmaxims,whicharenotpreciselytrue.Theprinciplesofgeometrydonotapplybeyondacertain30AtT1.2.2.3,1.2.2.5,1.2.2.9,1.2.4.14.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n122donaldl.m.baxterlevelofprecisionandcertainlynottosuchminuteobjectsasareenvisionedbythedefendersofinfinitedivisibility(T1.2.4.17).Humetakesforgrantedthatwhatisatissueisthestatusofgeom-etryasitappliestothespacewhichweexplorebymeansofoursenses.Giventhecopyprinciple,apuregeometricalspacewouldnotbesomethingofwhichwecouldhaveideas.Butbeyondthat,asbefore,theconceitofpuregeometricalspaceisnotforcedonusbyappearancesandthereisnowaytoknowifitistrueofrealitybeyondappearances.Geometryisinexact,Humesays,becausetheultimatestandardofequalityforgeometricfigures,ofstraightnessforlines,andofflatnessforplanesismerelyappearancetothesensesortheimagination.Thusourideasofthesegeometricpropertiesaremerelyideasofappearingequal,appearingstraight,andappearingflat.Thinkingthatthereisamoreexact,yetstilluseful,standardthanmereappearanceisanaturalfiction,butafictionnonetheless.Forwhatcouldthatstandardbe?Humeshowsthatnothingbesidesappearancewilldo.Considerequality.Humesowntheoryaffordsaprecisestandardofequality.Twofiguresareequalwhenthenumbersofpointsineachareequal.However,thisproposedstandardisuselessbecausewecannotdiscerntheindivisiblepartsofthingsoutsidethemind.Theyaretoosmall,orasHumesays,sominuteandsoconfoundedwitheachother,thattisutterlyimpossibleforthemindtocomputetheirnumber.FurtherthisproposedstandardisnotavailabletothoseHumeisarguingagainst,viz.,thosewhocontendthateveryparthasparts,since,accordingtotheirhypothesis,theleastaswellasgreatestfigurescontainaninfinitenumberofparts(T1.2.4.1920).Theonlyotheralternativeisthatthestandardofequalityfortwofiguresissimplytheirappearingequal.Actually,firstappearancescansometimesbemisleading.Wehavelearnedtocorrectthembydeterminingwhetherthefigurescontinuetoappearequalevenaftertheyhavebeenjuxtaposedorcomparedinsizetosomemovablethirdfigure(thecommonmeasure).Theresult,asHumeputsit,isthattheveryideaofequalityisthatofsuchaparticularappearancecorrectedbyjuxta-positionoracommonmeasure(T1.2.4.24).Thusthenotionofequalityingeometryisbasedonthisstandarddeterminedbyappearance.Wearewellawarethatthestandardisnotexact.Additionorremovalofasingleminutepartwouldmakenodifferencetoequal-ityaccordingtothisstandard,sinceitwouldbenotdiscernibleCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime123eitherintheappearanceormeasuring.Yetweimaginethatsuchactionrendersequalsunequal,sowesupposesomeimaginarystan-dardofequality,bywhichtheappearancesandmeasuringareexactlycorrected.Therelativelycrudeprocessofcorrectionwearecapableofimpelsthemindtocontinue,byakindofinertia,toimaginethatthereisthepossibilityofcorrectionbeyondthatofwhichweareca-pable.Weimaginetheretobefactsabouthowthingswouldappearthingstoosmalltomakeanydifferencetoanypossibleappearanceifonlytherightconditionsenabledustodiscernthem.Butthisstandardofequalityisamerefictionofthemind,anduselessaswellasincomprehensible,Humesays.Thereisnoreasontobelievethatwhatcannotappearwouldappearacertainwayincertainconditions.Sinceournotionofequalityisderivedonlyfromappearance,thereisnoapplyingitbeyondpossibleappearance.Thusappearance(oratleastappearancecorrectedbyjuxtapositionoruseofacommonmea-sure)isourbest,thoughinexact,standardofequality(T1.2.4.24).3131ThisimprecisionofgeometricequalityenablesHumetoanswersomestandardobjectionstohisdiscretegeometry.Thefirstobjectionisthatforanyunitofextension,nomatterhowminimal,thereisalinesegmentnotcomposedofawholenumberoftheseunits.Simplylettherebearighttrianglewithsidesof√nunits.Thehypotenusewillben2units.Humesansweristhattherewillalwaysbealinesegment,forfigureslargeenoughtoyieldanappearance,thatis√equalinhissenseton2units.TheotherobjectionisthatonHumesdiscretegeometrysomelinesegmentshavenomidpoint,viz.,thosecomposedofanevennumberofminima,socannotbebisectedbyanotherline.Humesansweristhatforlinesegmentslargeenoughtoyieldanappearance,thehalveswillbeequal.TheseanswersrequiretwoamendmentstoHumesgeometryofminusculeimprecision.First,hewouldhavetoexpandhisideaofacommonmeasuretoincludenotonlyfiguresthesamesizeorsmallerthantheonesbeingtested,butalsooneslarger.ThedangeristhatsomefiguresequalonHumesdefinition,repeatedthesamenumberoftimes,wouldcomposeunequalfigures.Thisproblemwouldoccuriftherewereanindiscernibledifferencebetweentheoriginalfiguresthat,aftersufficientrepetitionofthem,summedtoadiscernibledifferencebetweentheresultingcompositefigures.Humewouldhavetosaythatinsuchacasetheoriginalfiguresfailedthetestforequality.Additionally,hewouldhavetosaythatthereisamaximumsizebeyondwhichgeometrydoesnotapply,justasthereisaminimumsize.Otherwiseevenadifferenceofasingleminimumwouldcountagainstequality,forahugenumberofrepetitionscouldmakethatdifferencediscernible.BoththeseamendmentswouldbeinthespiritofHumesempiricalapproach.Thefirstgivesawayofmakingthenonapparentapparent;thesecondrulesoutfigurestoolargetoyieldanappearance.Fortheobjectionssee,forinstance,DonGarrett,CognitionandCommitmentinHumesPhilosophy(Oxford:OxfordUniversityPress,1997),745.Foradditionaldiscussion,seeDaleJacquette,DavidHumesCritiqueofInfinity(Leiden:Brill,2001).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n124donaldl.m.baxterWehaveonlythesamesortofstandardforstraightness.Noth-ingismoreapparenttothesenses,thanthedistinctionbetwixtacurveandaright[straight]line.However,wecannotdefineeitherintermsofhowtheirtiniestpartsarearranged,becausethisorderisperfectlyunknown,andnothingisobservdbuttheunitedappear-ance.Geometersmaytrytodefineastraightlineastheshort-estwaybetwixttwopoints,butthisoverlooksthefactthattheconnectionbetweenstraightnessandshortnessisadiscovery,notsomethingtruebydefinition(T1.2.4.256).32Thesituationisthesameforflatness.Theideaofaplanesurfaceisaslittlesusceptibleofaprecisestandardasthatofarightline;norhaveweanyothermeansofdistinguishingsuchasurface,thanitsgeneralappearance.Geometersmaytrytodefineaplaneaswhatisproducdbytheflowingofarightline,butthatwouldworkonlyifthelineflowedalongaplane,sothedefinitioninordertobetruewouldhavetobecircular(T1.2.4.28).Analogoustothecaseoftheequalityofgeometricfigures,werealizethattherecouldbecurvingsofanapparentlystraightlineorwarpingsofanapparentlyflatplanethatarenotnoticedatfirstbutthatcanbedetectedbyjuxtapositionoruseofacommonmeasure.Andanalogouslyourpracticesofcorrectinginitialappearancesleadsustoformthelooseideaofaperfectstandardtothesefigures,withoutbeingabletoexplainorcomprehendit(T1.2.4.25).Humeconcludesthatforcasesinwhichitisinanydegreedoubt-fulwhetherfiguresareequalornot,whetheralineisstraightornot,andwhetheraplaneisflatornot,thereisnostandardthatdetermineswhichoftheoptionsisinfacttrue.Theideasmostessentialtogeometryarenotexactanddeterminate.Thatis,again,becausetheonlystandardfortheapplicationoftheseideasistheweakandfalliblejudgment,whichwemakefromtheappear-anceoftheobjects,andcorrectbyacompassorcommonmeasure(T1.2.4.29).Humecontinues:Nowsincetheseideasaresolooseanduncer-tain,Iwoudfainaskanymathematicianwhatinfallibleassurance32Theworrythatmimimacannotbearrangedinatrianglehavingsidesincom-mensurablewithitshypotenusewithoutleavingasubminimalgap,isassuagedbyrealizingthattheworrypresupposesaprecisestandardofstraightness.SeeGarrett,CognitionandCommitment,75.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime125hehas,notonlyofthemoreintricateandobscurepropositionsofhisscience,butofthemostvulgarandobviousprinciples?Forexamplesupposetwolineshavealinesegmentincommon.Iftheyformanobviousanglewitheachother,itisclearthatatleastoneofthelinesisnotstraight.Atleastoneofthemmustbendorcurveinorderforthelinestooverlapforabit.Butwhatiftheangleisvery,verysmall?Whatifthelinesapproachattherateofaninchintwentyleagues?Eveniftheybecomeoneatsomepoint,theywillgiveeveryappearanceofbeingstraight.Thereisnofurtherruleorstandardthatmakesittrueofeitherlinethatitisnotstraight.ThustheseeminglyobviousassumptionofEuclidsgeometrythatdistinctstraightlinescannotshareacommonsegmentfailsinsuchacase,becausethereisnomoretobeingstraightthangivingeveryappearanceofbeingstraight:Theoriginalstandardofarightlineisinrealitynothingbutacertaingeneralappearance;andtisevi-dentrightlinesmaybemadetoconcurwitheachother,andyetcorrespondtothisstandard,thocorrectedbyallthemeanseitherpracticableorimaginable(T1.2.4.30).33Thusgeometricalaxiomsareliabletofailincasesinwhichtheirsuccesswouldrequiredifferencestoosubtletomakeadifferencetoappearance.Reliableapplicationofourgeometricprinciplesisrestrictedtoclearappearances.34Thisrestrictionentailsthatitisillegitimatetoapplytheseprinciplestofigurestoosmalltoyieldanypossibleappearance.Yetsuchillegitimateapplicationiswhatthegeometersdointheirproofsofinfinitedivisibility:iftheyemploy,asisusual,theinaccuratestandard,derivdfromacompar-isonofobjects,upontheirgeneralappearance,correctedbymeasuringandjuxta-position;theirfirstprinciples,thocertainandinfallible,aretoocoarsetoaffordanysuchsubtileinferencesastheycommonlydrawfromthem.Thefirstprinciplesarefoundedontheimaginationandsenses:Theconclu-sion,therefore,cannevergobeyond,muchlesscontradictthesefaculties.(T1.2.4.31)33HerewiththephraseinrealityHumeindicatestheviewforcedonthemindasopposedtotheviewbeingcriticized.34Humeallowsthatbecausethefundamentalprinciplesofgeometrydependontheeasiestandleastdeceitfulappearances,theyenableustodrawsomeconclu-sionsmoreexactthantheeyeortouchcoulddeterminebythemselves.Geometryallowsahighdegreeofexactness.Itjustfallsshortofthatperfectprecisionandcertainty,whicharepeculiartoarithmeticandalgebra(T1.3.1.6).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n126donaldl.m.baxterThegeneralproofagainstargumentsforinfinitedivisibilityex-plainsthecautionagainstapplyinggeometricalprinciplesbeyondclearappearance.ThecautionagainstcontradictingappearanceharksbacktoHumesinitialargumentthatourideaofextensionconsistsofindivisibleparts.Thusmathematicianswhoarguethateverypartofextensionhaspartsnotonlyillegitimatelygobeyondappearance,theyactuallycontradictit.However,itisabsurdtotrytouseideasthatcopyappearanceinordertoprovethatappearancediffersfromtheseideas(T1.2.4.32).Humegivesanexampleofthefailureoftryingtouseideasderivedfromexperiencetogobeyondandcontradictexperience.Heconsid-erstheclassofargumentsthateverypartofextensionhaspartsderivdfromthepointofcontact(T1.2.4.33).HumegivesmoredetailaboutsuchargumentsinthefirstEnquiry.Forexample,con-sideracircleandastraightlinetangenttoit.Foranyangleformedbytwostraightlines,nomatterhowsmall,theangleofcontactbetweenacircleanditstangentisless(EHU12.18).Thinkoftheangleformedbythetangentandsomechordfromthepointoftan-gency.Nomatterhowsmalltheangle,thespacebetweenthetwolinesisdividedbythearcofthecircle.Thus,Humesopponentsargue,foreveryportionofspacethereisasmallerportion.Farfromhelpinghisopponents,however,thecircleandtangentexamplebringsupadilemmatheycannotescape.SoHumeargues.Considerthepointoftangency.Ifitisanindivisiblepart,thenHumesopponentsarewrongthateveryparthasparts.Ifthepointoftangencyisapartwithparts,thenastraightlineandacurvedlinecoincideforadistance,whichestablishesthatatacertainlevelofminutenessthereisnostandardfordistinguishingastraightlinefromacurvedone.Which-eversidehechooses,herunshimselfintoequaldifficulties(T1.2.4.33).Proponentsofinfinitedivisibil-itymusteitheracceptindivisiblepointsormustaccepthisgeneralargumentagainsttheirsupposedgeometricproofsoftheirposition.35ThegeneralargumentsupportsHumesconclusionthatthereareindivisiblepartsofextension.Inaddition,showingtheideasmost35DisciplesofmoderngeometrywhotakealinetobeaninfinitesetofdimensionlesspointswillbeuntroubledbyHumesdilemma.However,theywouldstilloweaPyrrhonianEmpiricistanaccountofthesourceoftheirbasicconcepts,andanexplanationwhysomeoneshouldbelieveinthepurespacetheirgeometrypurportstodescribe.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime127essentialtogeometrytobederivedsimplyfromappearanceispartofHumesskepticalviewthatindoinggeometrywearesimplyacquiescinginviewsforcedonusbyappearances.ii.thesecondpartofhume’ssystemColoredorTangiblePointsThefirstportionofHumessystemconcerningspaceandtimecon-cludesthattheyarenotinfinitelydivisible,butconsistofindivisibleparts.Inthesecondportionheconcludesthatthesepartsareincon-ceivableunlessoccupiedbysomethingreal,andsospaceandtimemustsimplybemannersinwhichrealthingsarearranged.36Incon-sequenceitisimpossibletoconceivebothofemptyspaceandoftimewithoutanythingchanging(T1.2.4.12).HereHumemightseemtobeopposingtheviewendorsedbyNewtonthatspaceandtimeareabsoluteinthesenseofexistingindependentlyofthethingsthatoccupythem.37However,Humesapproachissubtlerthanmereopposition.Newtoncontendedthatspaceasitreallyis,isabsolute;spaceasitappearsisrelativetoperceivedobjects:38Idonotdefinetime,space,place,andmotion,asbeingwellknowntoall.OnlyImustobserve,thatthecommonpeopleconceivethosequantitiesundernoothernotionsbutfromtherelationtheybeartosensibleobjects....II.Absolutespace,initsownnature,withoutrelationtoanythingexternal,remainsalwayssimilarandimmovable.Relativespaceissomemovabledimensionormeasureoftheabsolutespaces;whichoursensesdeterminebyitspositiontobodies.3936Herebeingrealcontrastswithnotexisting.37HumemightfurtherbeseenasopposingNewtonsviewthatspaceandtimeareabsoluteinthesenseofprovidingnonrelativeframesofreferenceforinertialmotionandthepassageoftime.38TheapparenttensionbetweenNewtonsskepticalempiricalmethodandhispro-nouncementsaboutabsolutespaceandtimewasnoticedearlyon.SeeMaryShawKuypers,StudiesintheEighteenthCenturyBackgroundofHumesEmpiricism(Minneapolis:UniversityofMinnesotaPress,1930),1920.39IsaacNewton,ScholiumtotheDefinitions,inMathematicalPrinciplesofNaturalPhilosophy,Vol.1,trans.A.MotteandF.Cajori(Berkeley:UniversityofCaliforniaPress,1962),6.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n128donaldl.m.baxterHumefollowsthislead.AsskeptichesuspendsendorsementofNewtonsfirstcontentionthatspaceasitreallyis,isabsolute,whileacquiescinginthesecondthatspaceasitappearsisrelativetoper-ceivedobjects.ThethinkerwhoseviewsHumehadmostprominentlyinmind,however,wasBayle.Humessystemasawholewasmostlikelywrit-tenasaresponsetothegeneralproblemBayleposedforthedivis-ibilityofspaceandtime:Extensioncannotbemadeupofeithermathematicalpoints,atoms,orparticlesthataredivisibletoinfin-ity;thereforeitsexistenceisimpossible.40Inotherwords,thereareonlythreecandidatesforthepartsofspaceunextendedindivisibleparts(whatBayleandHumebothcallmathematicalpointsandsometimesHumecallsatoms),extended,indivisibleparts(whatBaylecallsatomsandwhatHumecallsphysicalpoints),andextended,divisibleparts(thusinfinitelydivisiblepartssinceeverypart,beingdivisible,woulditselfhaveparts).41Baylegivesargu-mentsagainsteachalternative.Heintendstoshowthatreasoncan-notmakesenseofthecompositionofextension.Asafideist,hisoverallgoalistohumblereasonsoitwillnotpresumetochallengereligiousfaith.Baylegivesseveralargumentsagainstinfinitelydivisibleparts.However,hethinkstheabsurdityoftheviewisasclearandevidentasthesuntotheunprejudiced:Aninfinitenumberofpartsofextension,eachofwhichisextendedanddistinctfromalltheothers,bothwithregardtoitsbeingandtotheplacethatitoccupies,cannotbecontainedinaspaceonehundredmilliontimessmallerthanthehundredththousandthpartofagrainofbarley.42Humeagrees,ofcourse,andhashisownargumentsinsupportoftheabsurdityofinfinitedivisibility.Baylesargumentagainstphysicalpointsisasfollows:Anythingextendedhasdistinctparts,forinstance,alefthalf40Bayle,HistoricalandCriticalDictionary,359.AsourceforBaylestrilemmaislikelyAristotle,OnGenerationandCorruption,BookI,Chap.2,316a17317a13.ItislikelythatBerkeley,whowasalsoinfluencedbyBayle,wasanotherinfluenceonHumehere.SeeDavidRaynor,MinimaSensibiliainBerkeleyandHume,Dialogue19(1980):196200;andAyers,BerkeleyandHume,30614.41Bayleletsitgowithoutsayingthatwhatmightseemtobeafourthoption,namely,unextended,divisibleparts,makesnosense.Presumablytheexplanationisthatsomethingthattakesupnospacecannothavespatialparts.42Bayle,HistoricalandCriticalDictionary,362.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime129andarighthalf.Havingdistinctpartsentailsbeingdivisibleintherelevantsenseofdivisible.Sotherecannotbeextended,indivisiblepoints.Humeagreeswiththisargumentaswell,notingthatthesystemofphysicalpoints...istooabsurdtoneedarefutation(T1.2.4.3).BaylesargumentagainstmathematicalpointsderivesfromZenosparadoxofextension.Mathematicalpointsareunextended.Thustheyhavezerolength.Addingpartsofzerolengthtogetheryieldsawholeofzerolengthnomatterhowmanypartsareinvolved.Sonofiniteintervalcanbecomposedofmathematicalpoints.AsHumeputsBaylesargument,amathematicalpointisanon-entity,andconsequentlycanneverbyitsconjunctionwithothersformarealexistence(T1.2.4.3).43HereHumedisagrees.Wedohave,afterall,anideaofextensionasconsistingofaconjunctionofunextended,indivisiblepointstheveryideaappealedtoinHumesfirstargumentagainstinfinitedivis-ibility.Sincewehavetheidea,extensioncomposedthatwaymustbepossible.AsHumesays,Tisanestablishdmaximinmeta-physics...thatnothingweimagineisabsolutelyimpossible(T1.2.2.8).Theextenttowhichaskepticisentitledtouseestablishedmaximsinmetaphysicsmightbedebatable,butinthiscaseHumesrelianceonappearanceleaveslittleroomfordebate.Asmysandexperimentshows,theideaHumeappealstocanbedirectlycopiedfromanimpressionofextension.Sowecanbesurethatextensionasitappearscanconformtotheidea.Baylegoeswrongbyassumingthatthereisnowaytoconjoinpartsofnolengthtoformawholeofsomelength.44Tofindaway,Humethinks,wemerelyhavetouseourimaginations.Alterna-tivelywecouldusethesandexperiment.Thekeyisthatwedonottrytoconceivehowthingsareinsomepureconceptualspace.Westickwithappearance.Thatamountstobestowingacolourorsolidityonthesepoints,inotherwords,itamountstoimaginingor43ThiscontrastwithbeinganonentityiswhatHumehasinmindwhenhesaysthesmallestpartsofspacearefilledwithsomethingrealandexistentandformarealexistence(T1.2.4.23).44BaylesassumptionhasbeenquestionedalsobysupportersofGeorgCantorscon-ceptionofalineasasetofunextendedpoints.SeeAdolfGrunbaum,¨ModernSci-enceandZenosParadoxes(Middletown,CT:WesleyanUniversityPress,1967),Chap.3.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n130donaldl.m.baxtersensingthemtobeoccupiedbythingsdetectablebysightortouch(T1.2.4.3).Baylewasrightconcerningunoccupiedmathematicalpointsthattheyarenobetterthannonentitiesandcannotbecon-joinedtoformanentity.Suchpointsareinconceivablewhennotfilldwithsomethingrealandexistent(T1.2.4.2).ButfilledonesavoidBaylesconclusion:Tisnotonlyrequisite,thattheseatomsshoudbecolourdortangible,inordertodiscoverthemselvestooursenses;tisalsonecessaryweshoudpre-servetheideaoftheircolourortangibilityinordertocomprehendthembyourimagination.Thereisnothingbuttheideaoftheircolourortangibility,whichcanrenderthemconceivablebythemind.Upontheremovaloftheideasofthesesensiblequalities,theyareutterlyannihilatedtothethoughtorimagination.(T1.2.3.15)Onemightobjectthatcolorandtangibilitydonothelp.Theindi-visiblepointsmusthaveedgesthateithertouchornot,andtheymusthaveshape.Iftheyhaveedgesorshape,thentheymusthavedifferentiablepartsandsocanbefurthersubdivided,afterall.Fur-ther,iftheytouch,theymusttotallypenetrateeachother,sincetheywouldhavenopartsthatdonottouch(T1.2.4.4).Alterna-tively,iftheydonottouchthenthereismoretospacethanHumespoints.Suchobjectionsseemforceful,butmisunderstandHumesconcernwithappearance.Theimpressionreceivedfromabarelyvisiblegrainofsandaccuratelyreflectsacoloredortangiblemathe-maticalpoint.Thatimpressionhasnoedges,noshape.Itcantouchanotherimpression,butonlyinthesenseoftherebeingnoindivisi-bleimpressionbetweenthemasensethatneitherentailspenetra-tionnortherebeingsomethingelsebetweenthem.45TounderstandHume,wemustnotmakethemistakeofthinkingthatsuchimpres-sionshavegeometriccharacteristicsatalevelbelowourabilitytodiscernthem.4645Cf.RichardSorabji,AtomsandTimeAtoms,inN.Kretzman,ed.,InfinityandContinuityinAncientandMedievalThought(Ithaca,NY:CornellUniversityPress,1982),67.Inlightofthediscussionofvacuumstofollow,thedefinitionoftouchcouldbeemendedtoreadtherebeingnovisibleortangiblenorinvisibleandintangibledistancebetween.46AnadditionalobjectiontoanatomistictheoryofspaceisgivenbyH.Weyl.Ifasquareisbuiltupofminiaturetiles,thenthereareasmanytilesalongthediagonalastherearealongtheside;thusthediagonalshouldbeequalinlengthtotheside.AgainthisobjectionassumesthatHumeanpointshavepreciseshapesandCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime131Itisnotjustthepointsthatmustbeconceivedascoloredortangible.AsHumesays,Nowsuchasthepartsare,suchisthewhole.Wecanonlyconceiveofextensionbyconceivingofitspartsandwecanonlyconceiveofthembyconceivingofthemascoloredortangible.Wehavethereforenoideaofspaceorextension,butwhenweregarditasanobjecteitherofoursightorfeeling(T1.2.3.16).MannersinWhichObjectsExistTheideaofspaceistheideaofawholecomposedofindivisibleparts.Itisanabstractorgeneralidea,however,notaparticularidea(T1.1.7.1,1.2.3.5).Inotherwords,theideaofspaceisnotanideaofaparticularvastcontainerordimension,asonemightcommonlysuppose.Rather,inaccordancewithHumestheoryofabstractorgeneralideas,theideaofspaceisanideaofsomearbitraryextendedwhole,suchasatabletoporawall,withaneyetothewayitresem-blesotherextendedwholes.Thewayextendedwholesresembleisspecificallythemannerinwhichtheirindivisiblepartsarearranged.ThusspaceissynonymouswithextensionforHume,inthesenseofextendedness.Humesviewbelongstoafamilyofviewsaboutspacetermedrelationalorrelationist.Contrastingviewsaretermedabso-lute(inthesensegivenabove)orsubstantival.Inafamouscorre-spondence,LeibnizdefendedarelationalviewagainsttheabsoluteviewoftheNewtonian,SamuelClarke.47ThebasisofHumesrela-tionalviewofspaceishisrelianceonthecopyprinciple,theprinciplethatalloursimpleideasintheirfirstappearancearederivdfromsimpleimpressions,whicharecorrespondenttothem,andwhichtheyexactlyrepresent(T1.1.1.7).NotethatHumesskepticism,hisexclusiveconcernwithappearances,helpsmotivatethecopyprin-ciple.Anideaofspaceconcernsspaceasitappearstooursenses,orarrangementsbelowourabilitytodetectthemanassumptionHumedoesnotgrant.Heneedonlysaythatwefindtheretobemorepointsalongthediagonalofasquarethanalongaside.Thearrangementthatallowsthistohappenisbeyondustodiscover.Thissituationwouldbeanalogoustothewaywewouldfindmoregrainsofsandalongthediagonalofasquareshadowonthebeach,whenlookingdownfromaboardwalkperhaps.HermannWeyl,PhilosophyofMathematicsandNaturalScience(Princeton:PrincetonUniversityPress,1949),43.47H.G.Alexander,ed.,TheLeibniz-ClarkeCorrespondence(Manchester:Manches-terUniversityPress,1956).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n132donaldl.m.baxterinotherwords,asitappearsinoursenseimpressions.Notonlydoesthecopyprinciplehelpcodifyhisrelianceonhowthingsappearinourexperience,itgiveshimamethodofmakingideasclear.Nodiscoverycoudhavebeenmademorehappilyfordecidingallcon-troversiesconcerningideas,thanthatabove-mentiond,viz.,thecopyprinciple.Ourideasareoftenobscurebutourimpressionsareallsoclearandevident,thattheyadmitofnocontroversy(T1.2.3.1).Theideaofspacemustbederivedinsomewayfromsomeofourimpressions.Itcannotbederivedfrominternalimpressionsourpassions,emotions,desiresandaversions(T1.2.3.3).Suchfeelingsarenotspatial.Theycannotbeconjoinedtoformsome-thingthattakesupspace(T1.4.5.9).Sotheideamustcomefromimpressionsconveyedbythesenses.Therelevantimpressionsaresensoryimpressionsofthingsinspace,suchasatabletop.Butmysensesconveytomeonlytheimpressionsofcolourdpoints,dis-posdinacertainmanner.Itfollowsthattheideaofextensionisnothingbutacopyofthesecolourdpoints,andofthemanneroftheirappearance(T1.2.3.4).48Withoutthepreparationaffordedbytheinkspotexperimentandwithoutaconsiderationandrejectionofinfinitedivisibility(espe-ciallyinthelightoftheanswertoBayleschallenge),wereadersofHumemighteasilyhaveoverlookedthecoloredpointsinourexpe-rience.Afterall,becauseeachissosmallandisrightnexttoitsnearestneighbor,wetooeasilyruntogethersuchpointsinpercep-tion(T1.2.4.19).However,giventhatpreparation,Humescharac-terizationisanaturalone.Whenweperceiveatableweperceiveanexpansewithpartswecandistinguisheveniftheyareseamlesslyconnected.Suchpartshavediscernibleparts,whichthemselveshavediscernibleparts,andsoon,downtothelevelofindivisible,part-lessparts.Theselastarehardtodistinguishprecisely,butwiththesuitablepreparationwecannothelpbutbelievetheyarethere.4948KempSmithssuggestionthatappealtoideasofmannersisinconsistentwithHumesempiricismisdecisivelyrefutedinLorneFalkenstein,HumeonMan-nersofDispositionandtheIdeasofSpaceandTime,ArchivfurGeschichteder¨Philosophie79(1997):179201.49BroadscriticismofHumehereseemstoconfusecontinuousinthesenseofnothavingdiscerniblegapsbetweenparts,withcontinuousinthemathematicalsense(HumesDoctrineofSpace,1667).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime133Thusseeingatabletopgivesusanimpression,andsoanidea,ofcoloredpoints,perhapsofapurplecolour,arrangedinacertainmanner.Seeingexpansesofothercolors,violet,green,red,white,black,andofallthedifferentcompositionsofthese,givesustheideaofdifferentcoloredpointswithsimilararrangement.Merelyfeelingexpansessuchasmarblefloors,pillows,ovendoors,andsoon,thatarehardorsoft,hotorcold,givesusfurtherideasofpoints,tangibleoneswithoutcoloryetwithasimilararrangement.Itisbynoticingthesimilarmannerofbeingarrangedthatwegettheideaofspace(T1.2.3.5).HereHumeappealstohisaccountofabstractideas,betterthoughtofasgeneralideas.Anabstractideaisanideaofsomecharacteris-tic,somerespectinwhichparticularthingsresemble.Forinstance,theabstractideaofcolorwouldbeanideaoftherespectofresem-blancebetweencoloredthings;thatofgreenwouldbeanideaoftherespectofresemblancebetweengreenthings.Wecannothaveanideaofarespectofresemblancewithoutthinkingofthethingsthatresemble,accordingtoHume.Wecannotthinkseparatelyofredness,say;wecanonlythinkgenerallyofredthings.Sowecometohaveabstractideasasfollows:Anideaofaparticularthingnatu-rallybringstomindideasofparticularsthatresembleitinacertainway.Forexample,anideaofatulipmayremindoneofasimilarlycoloredrose.Whichrespectofresemblanceisinplayatagiventimeisatfirstarbitraryandinexplicable.Butafterawhilesomesuchassociationsbecomehabitualandwe,byanotherinexplicableprocess,begintoassociateawordwiththeparticularideasjoinedbyagivenhabit.Forinstance,thewordsaltymaybringtomindtheideaofthetasteofolivesandengagethehabitofbringingtomindtheideasofothersaltytastessuchasthatofpickles(T1.1.7).AsHumesays,Allabstractideasarereallynothingbutparticularones,considerdinacertainlight;butbeingannexdtogeneralterms,theyareabletorepresentavastvariety,andtocomprehendobjects,which,astheyarealikeinsomeparticulars,areinothersvastlywideofeachother(T1.2.3.5).50Thusanabstractideaistheideaofsomeparticularthinginsofarasitresemblesinacertainwayvari-ousotherparticularthings.Theparticularideaisusedasageneral50Herereallyindicatestheviewforcedonthemindasopposedtotheviewbeingcriticized.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n134donaldl.m.baxterrepresentativeorproxyforthevariousideasofthevariousparticu-larsthatsharethecharacteristic.51Theideaofextensionisanabstractorgeneralideainjustthewaydescribed.Theideasofawholecomposedofpurplepoints,awholecomposedofcoldpoints,andsoon,allresembleinacertainrespect,namely,inthemannerinwhichtheirindivisiblepartsarearranged.Theabstractideaofextensionisanideaofsomeparticularextendedwhole,insofarasitresemblesotherextendedwholes.Inotherwords,theabstractideaisanideaofanextendedwholeingeneral.NowitisclearerwhyHumeusesthetermsspaceandextensioninter-changeably.Forusthefirstwordsuggestsacontainerofextendedthingswhilethesecondsuggestsacharacteristicofthem.ForHume,however,bothtermsaretermsforextendedthingsingeneral.52Humedoesnotsaymuchofthemannerofarrangementexceptthatthepartscoexist.Forthatqualityoftheco-existenceofpartsbelongstoextension,andiswhatdistinguishesitfromduration(T1.2.3.8).Thereismoretothearrangementthancoexistence,how-ever,becausetastesandsmellsqualitiesthatHumethinkscannothavespatiallocationcancoexist(T1.4.5.1012).Fromhisdis-cussionofinfinitedivisibilityitfollowsthateachindivisiblepartofextensionisrightnexttoitsnearestneighbors;thereisnodis-tancebetweenthem.However,thisarrangementcouldbetrueofsimultaneous,barelydistinguishablesounds.ThebestHumecandotocharacterizetherelevantarrangementofpartsistorefertothearrangementinthecontextoftalkingaboutsomethingextended.Thisisfine.Heisnotdefiningthearrangement;heisonlycallingourattentiontoit.Humehasbeenchargedwithcircularityinhisaccountofacquir-ingtheideaofspace,foritseemsasifwemustalreadyhavetheideaofthemannerofbeingarrangedinordertonoticeit.53However,51ForthemedievalbackgroundofsuchanaccountseeJuliusR.Weinberg,TheNominalismofBerkeleyandHume,inAbstraction,Relation,andInduction(Madison:UniversityofWisconsinPress,1965).52TheCartesiansalsousedthetermsinterchangeably.SeeDescartes,PrinciplesofPhilosophy,inThePhilosophicalWritingsofDescartes,trans.J.Cottingham,R.Stoothoff,andD.Murdoch,2vols.(Cambridge:CambridgeUniversityPress,1984),1:227(Pt.II,Sect.10).53Forinstance,inparagraph259ofT.H.GreensGeneralIntroductiontoHumesATreatiseofHumanNature(London:Longmans,1878).ThechargemaybeinspiredbyKantsclaimthatwemusthavetheconceptofspacetheabilitytorepresentCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime135thechargeresultsfrommisunderstanding.Themannerinwhichthepointsarearrangedmerelyservestobringtomindotherpointsarrayedinasimilarmannerbytheinexplicableprocessmentionedabove.Suchassociationofparticularideasisaquasi-mechanicalcausalprocess.Anynoticingofaresemblanceisnomorethanthecomingtomindofanassociatedparticularidea.Itiscertainlynottheapplicationofageneralidea.Thegeneralideaofspaceisnothadbeforethehabitofassociatingtheparticularideaswitheachotherisestablished.Theaccountoftheideaoftimeisverysimilartothatoftheideaofspace:Theideaoftime,beingderivdfromthesuccessionofourperceptionsofeverykind,ideasaswellasimpressions,andimpressionsofreflectionaswellasofsensation,willaffordusaninstanceofanabstractidea,whichcomprehendsastillgreatervarietythanthatofspace,andyetisrepresentedinthefancybysomeparticularindividualideaofadeterminatequantityandquality.(T1.2.3.6)Theideaoftimeisanideaofaparticularsuccessionofobjects,insofarasitresemblesothersuccessionsofobjects.Thustheideaoftimeistheideaofsuccessiveness,orbetter,ofasuccessioningeneral.CasesofsuccessionforHumeareanycasesofreplacement,oralteration,ormovement.Hegivestheexampleoffivesucces-sivenotesplayedonaflute(T1.2.3.10).ThisaccountexplainswhyHumeusestimeanddurationinterchangeably.Forus,thefirstwordsuggestssomesortofcontainerforobjectswithduration,whilethesecondsuggestsacharacteristicofthem.ForHume,however,theyarebothtermsforenduringthingsthatis,thingswithdurationingeneral.Humeconcludes,Theideasofspaceandtimearethereforenoseparateordistinctideas,butmerelythoseofthemannerororder,inwhichobjectsexist(T1.2.4.2).Theideaofamannerororderisnotseparablefromideasofthethingsordered.Anideaofamannerthingsinspatialarrayslogicallypriortoexperiencingspace.Thisclaim,however,isnotsomethingHumewoulddisagreewith.HisideasarenotlikeKantsformalconcepts.ImmanuelKant,CritiqueofPureReason,trans.N.K.Smith(NewYork:St.MartinsPress,1965),TranscendentalAesthetic,Sect.2;seealsoSect.4abouttheconceptoftime.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n136donaldl.m.baxterororderisageneralideaofobjectsinsofarastheyareorderedinthatmanner.Asinthecaseofindivisiblepoints,Humeinvestigatesthenatureofspaceandtimebyinvestigatingtheirideas.Andagainthismethodofinvestigationisjustifiedbythefactthathisconcerniswithspaceandtimeastheyappear.AgainsttheIdeaofaVacuumHumesviewisthatwecanonlyconceiveofspaceandtimebymeansofabstract,thatis,general,ideas.Inotherwordswecanonlycon-ceivethembymeansofparticularideasideasofparticularthingsusedgenerallytorepresentresemblingparticularthings.Aconse-quenceofthisviewisthatwecannotconceiveofavacuum,oremptyspace.Norcanweconceiveoftimepassingwithnothingchanging.Humesviewseemsopentotheobjectionthatofcoursewecancon-ceiveofbothavacuumandaperiodoftimewithoutchange.ThepossibilityofsuchideasisprovidedforbyNewtonsauthoritativeconceptionsofabsolutespaceandabsolutetime.Humeresponds,however,thatwedonotactuallyhavetheseideas.Weonlyimagine(insomesense)thatwedo.TheseideasofavacuumandchangelesstimewouldbepossibleonlyiftheideasoftimeandspacewereliterallyseparablefromtheideasofthingsthatoccupythemseparableasinLockesaccountofabstraction.54However,Humehastakencaretoshowthattheyarenotseparable.Hedetailshispointfortime:Inordertoknowwhetheranyobjects,whicharejoindinimpression,beseparableinidea,weneedonlyconsider,iftheybedifferentfromeachother;inwhichcase,tisplaintheymaybeconceivdapart.Everything,thatisdifferent,isdistinguishable;andeverything,thatisdistinguishable,maybeseparated,accordingtothemaximsabove-explaind.55Ifonthecontrarytheybenotdifferent,theyarenotdistinguishable;andiftheybenotdistinguishable,theycannotbeseparated.Butthisispreciselythecasewithrespecttotime,compardwithoursuccessiveperceptions.Theideaoftimeisnotderivdfromaparticularimpressionmixdupwithothers,andplainlydistinguishablefromthem;butarisesaltogetherfromthemanner,in54SeeLocke2.11.9,2.12.1,3.3.69,3.6.32.55ThesemaximsarefirstmentionedatT1.1.3.4,1.1.7.3.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime137whichimpressionsappeartothemind,withoutmakingoneofthenumber.(T1.2.3.10)Noristheideaofspacederivedfromaparticular,separableimpres-sionmixedupwiththeimpressionsofcoloredandtangiblepoints.AsHumeputstheinferenceforspace:Ifthesecondpartofmysystembetrue,thattheideaofspaceorextensionisnothingbuttheideaofvisibleortangiblepointsdistributedinacertainorder;itfollows,thatwecanformnoideaofavacuum,orspace,wherethereisnothingvisibleortangible(T1.2.5.1).Humerestrictshisattentionheretotheimpossibilityoftheideaofavacuum,anddoesnotexplicitlysaythatavacuumisimpossi-ble.Thereisnoneedto.Hehasalreadysaidthatspaceconsistsofindivisiblepartsthatwouldbenonexistentunlesstheywerecoloredortangible.Theabsurdityofalternativeviewsdemonstratesthetruthandrealityofhisown.Thusspacenotonlymustbecon-ceivedtobefilldwithsomethingrealandexistent,italsomustreallybeso(1.2.4.13).56Section1.2.4isanextendeddefenseoftheclaimthatspaceconsistsofcoloredorsolidpoints.Laterhecallsextension,acompositionofvisibleandtangibleobjects(1.2.5.14).InthecaseoftimeHumeexplicitlyswitchesfromtalkoftherebeingnoideaofchangelessdurationtotalkoftherebeingnochangelessduration.Sincewhatisatissue,givenHumesPyrrhonianEmpiri-cism,istimeandspaceastheyappear,itfollowsthatneitheremptyspacenorchangelesstimecanexist.Neithertimenorspacecanappeartousexceptasindivisible,partlessthingsarrangedintheappropriatemanner.Havingshownthatspaceisfilledwithcoloredortangiblethings,Humeisconcernedtoshowthatwedonotevenhaveanideaofavacuum,despiteseeminglystrongargumentstothecontrary.Humeputshispointinapotentiallyconfusingway:Eventhoughwefalslyimaginewecanformsuchanideaofavacuum,wecannotreallyhavesuchanidea(T1.2.5.14).Howcouldweliterallyimaginethatwehaveanideawithouthavinganideaoftheimaginedidea?OnHumesaccount,ideasarelikepictures.Apictureofapictureofacentaurcanbeusedasapictureofacentaur.Likewiseanideaofanideaofavacuumcould56HereagainHumeistalkingaboutwhatisreallytrueoftheworldasitappears.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n138donaldl.m.baxterbeusedasanideaofavacuum.Soitdoesnotmakesensetosaythatweliterallyimaginewehaveideasthatwecannotreallyhave.Humemustbeusingthephrasefalslyimaginewecanformsuchanideainanotherlegitimatebutnotabsolutelyliteralsense.Hemustmeanthatwefalselytalkasifwehavesuchanidea.Weusethephrasesemptyspaceorvacuumasiftherewereacoherentideacorrespondingtotheseterms.Thereisnotreallyone,heargues.Insteadofacoherentideawehaveaconfoundingofdistinctideas(seealsoT1.2.5.21).InexplainingHumesthinkingIwilllimitmyselftoconsideringvision,thoughhegivesanalogousdetailsforthesenseoftouch.Onemightthinkwecouldgettheideaofemptyspacefromanexperienceofdarkness.But,Humeargues,darknessfailstogiveusanyvisualimpressionfromwhichanideacouldbecopied.Beinginalightlesscave,forinstance,givesonenovisualexperience,ratherthanavisualexperienceofblack.Whenoneseessomethingblack,onecandiscernpartsofit,Humethinks.Howeverwhen,asaresultofdarkness,onelacksvisualexperience,therearenopartstodiscern.Onegetsnoperceptiondifferentfromwhatablindmanreceivesfromhiseyes(T1.2.5.11).Thustherearenovisualideasofthelackofsomethingvisible.Thereisonlythelackofvisualideas.Thisconclusionholdstrueevenifdistantvisiblethingsappearinthedarkness,suchasstarsinanightsky.Thestarsgivetheviewervisualimpressionsandsoideas,butthedarknessdoesnot.Insuchacasethestarsareexperiencedasdistantfromeachotheryetnotasspatiallyseparated.Theyarenotexperiencedasspatiallyseparatedbecausebetweentheimpressionsofthestarsthereisnoextendedimpressionthatcouldgiverisetoanideaofspace.Thereisnoimpressionatall,muchlessonewithparts.So,hesaysoftheverydistance,whichisinterposdbetwixtthem,itisnothingbutdarkness,orthenegationoflight;withoutparts,withoutcomposi-tion,invariableandindivisible(T1.2.5.11).Suchadistancecannotgiverisetotheabstractideaofspace,sotheideacannotproperlybeappliedtoit(seeT1.2.3.11).Nonethelessthestarsareperceivedasdistantfromeachotherinsomesense,becausetheyaffectdifferentpartsoftheeyesothattheimpressionsofthemarenotdirectlyadja-cent.Humeusestheworddistanceofthisseparation,butcallsitaninvisibleandintangibledistancetocontrastwithextensionCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime139thatisvisibleortangibledistance.57Thushedistinguishestwokindsofdistance(1.2.5.1617).Suchadistinctionmightwellseemstrange,butitisappropriateifweconfineourattentiontotheworldasitappears.Wecandistin-guishdistantthingsbetweenwhichotherthingsappear,fromdistantthingsbetweenwhichnootherthingsappear.Theformerdistanceismarkdoutbycompoundedandsensibleobjects,whereasthelatterisknownonlybythemanner,inwhichthedistantobjectsaffectthesenses(T1.2.5.17).Thedifferingwaysinwhichthingsareexperiencedcanrenderimportantdifferencesintheworldasitappears.Inthiscasewhatmakesthedifferenceiswhetherornotthestarsareexperiencedbymeansofimpressionsthathaveimpressionsinbetween.Intheworldasitappears,thestarsareeitherseparatedbyvisibleortangibledistanceorbyinvisibleandintangibledistance,depending,forexample,onwhetherornottheskyisdark.Inacaseofstarsinthedarkness,althoughweperceivenointer-veninglocations,nopartsofspace,weassumetherearesuch.Why?Becausetheideaofinvisibleandintangibledistanceiscloselyrelatedtotheideaofvisibleortangibledistance,andsoweconfusethem.First,theangleatwhichlightraysfromthedistantthingsstriketheeyecanbethesamewhetherthedistanceisinvisibleandintan-gibleorisvisibleortangible:twovisibleobjectsappearinginthemidstofutterdarkness,affectthesensesinthesamemanner,andformthesameanglebytherays,whichflowfromthem,andmeetintheeye,asifthedistancebetwixtthemwerefilldwithvisibleobjects,thatgiveusatrueideaofextension(T1.2.5.15).Humelaterretractshisclaimthatweknowtheangle,perhapsunderfur-therconsiderationofBerkeleysNewTheoryofVision(A22).58Butthefactisthatthesensationsofmovingtheeyefromdirectcon-templationofonethingtodirectcontemplationofanotherarethesamewhetherthedistanceisinvisibleandintangibleorisvisibleortangible(T1.2.5.15).57Broadgivesanattractivebutlesstextuallyboundaccountofthecontrastbetweenthetwosortsofdistance(HumesDoctrineofSpace,173).Humealsocallsinvisibleandintangibledistanceimaginarywhenarrivedatpurelybykinestheticsensations(T1.2.5.13)andfictitiouswhenconfusedwithextension(T1.2.5.23).58GeorgeBerkeley,AnEssaytowardsaNewTheoryofVision,inTheWorksofGeorgeBerkeley,1:1714(Sects.415).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n140donaldl.m.baxterSecond,invisibleandintangibledistancecanbeconvertedintovisibleortangibledistancemerelybyputtingsomethingbetweenthethings.Forinstance,asdawnapproachesandtheskygainssomecolor,thedistancebetweentwostarscanshifttoavisibledistance.GivenHumesconcernsolelywiththeworldasitappears,itisirrelevantthatbeyondouratmospheretherealspacebetweenthestarsremainsuncolored.Heisconcernedonlywiththewaythedistanceisexperienced.Infact,asfarasHumeisconcernedonecouldconverttheinvisibleandintangibledistancebetweenthestarsintovisibleortangibledistancemerelybyholdingastripofwhitepapersothatthedistantstarsappearadjacenttoeachendofthestrip.Hisdistinctionbetweenvisibleortangibledistanceandinvisibleandintangibledistanceispurelyamatterofhowthingsappeartothesenses.Theimpressionsofpointsofthepaperarebetweentheimpressionsofthestars,hethinks,evenifthepaperisnotliterallybetweenthestars.Humeassumes,asdoesforinstanceLocke,59thatourvisualfieldislikeaflatplane:Tiscommonlyallowdbyphilosophers,thatallbodies,whichdiscoverthemselvestotheeye,appearasifpaintedonaplanesurface,andthattheirdifferentdegreesofremotenessfromourselvesarediscoverdmorebyreasonthanbythesenses(T1.2.5.8;seealsoT1.2.5.16).Inanyeventinvisibleandintangibledistancehasthecapabilityofbeingconvertedintovisibleortangibledistance.Third,bothkindsofdistancediminishtheforceofotherqualitiesofadistantobject:Forasallqualities,suchasheat,cold,light,attraction,&c.diminishinproportiontothedistance;thereisbutlittledifferenceobservd,whetherthisdistancebemarkdoutbycompoundedandsensibleobjects,orbeknownonlybythemanner,inwhichdistantobjectsaffectthesenses(T1.2.5.17).Thefurtheryourfriendsflashlightisfromanobject,thedimmeritshinesonthatobject.Thisissowhetheryouseetheflashlightandtheobjectattheshore,againstthebackdropofalightlesssky,orinthewoods,againstthebackdropofmistandtrees.Thefirstandthirdofthesecloserelationsareinstancesofresem-blance.Thesecond,Humesays,isaninstanceofcauseandeffect.Presumablyheisassumingthatcapability,likepower,isunderstoodintermsofcauseandeffect(T2.1.10).Sothecapabilityofbeing59Locke,EssayconcerningHumanUnderstanding,2.9.8.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime141convertedintovisibleortangibledistanceisunderstoodintermsofcauseandeffect.Thesecloserelationsmakeiteasyforthemindtouseoneideaintheplaceoftheother.Whenwetalkaboutacaseinwhichthedistanceisproperlyspeakinginvisibleandintangible,anideaofvisibleortangibledistanceislikelytocometomindinstead.(Thementalslipdoesnotseemtogointheotherdirection,perhapsbecausetheperceptionofvisibleortangibledistanceissomuchmorecommon.)Thuswhenwethinkoftwodistantstarsinthenightskywehavetwoideaswithoutanyideasbetweenthem.Immediatelyandunawaresweaddbetweenthemanarrayofspa-tiallyminimalideasendowed,say,withthecolorblackanarrayofideasthatcanserveasthegeneralideaofspace.Atfirst,thereisnothingbetweenthestarideas,sowecharacterizethedistanceasempty.Thenimmediatelythereareideasinbetween,sowegoontocharacterizethedistanceasspace.Wedonotnotice,however,thatwehaveshiftedfromanideaofaninvisibleandintangibledis-tancetoanideaofavisibleortangibledistance.Thuswetalkofemptyspacealthoughwehavenocoherentideaofsuchathing.Anideathatwouldliterallybeanideaofemptyspacewouldhavetobeanideawithbothnothingandsomethingbetweentheideasofthestars.Itwouldbeanideaofinvisibleandintangibledistancethatisvisibleortangible.Oblivioustotheincoherencewecome,astimegoeson,touseotherexpressions,suchasvacuum,interchangeablywithemptyspace.Humeusesthisaccounttoanswerthethreeseeminglystrongargumentsfortheclaimthatwehaveanideaofavacuum.Thefirstisthatwemusthaveanideaofvacuumssinceweargueaboutthem.Humesresponseisjustthatweimagine,inthewayjustexplained,thatwehavetheidea.Thesecondargumentisthatwecanconceivethematterwithinacubetobeannihilatedwithoutmotionofthesidesofthecube,andinsodoingwearehavinganideaofavacuum.Notso,Humereplies,weareratherhavinganideaofwhathehascalledinvisibleandintangibledistance.Thethirdargumentisthatwemustconceivetheretobevacuumsintheworldinordertoconceivetheretoberoomforthingstomove.Iftheworldwereaplenumtherewouldalwaysbesomethinginthewaytopreventmotion.Invisibleandintangibledistancecandothesamework,saysHume.Itiseasilyconvertedintovisibleortan-gibledistancebythingsmovinginbetweenthebodiesthatborderCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n142donaldl.m.baxtertheinvisibleandintangibledistance.Invisibleandintangibledis-tance,notavacuum,iswhatwehaveanideaof(T1.2.5.224).OnemightobjectthatHumesclaimtohavenoideaofemptyspaceisjustanartifactofhisdecisiontousethewordspacetorefertoonekindofdistance.Whynotuseittorefertodistanceingeneral?Theninvisibleandintangibledistancecouldbecalledemptyspacewithoutcontradiction.Humesreplywouldbethatgeometrygivesthepropertiesofspace.Therecanbenogeometry,however,ofinvisibleandintangibledistance.Therewouldbenopointstocomposelinesandplanes.Aspatialpointwithneithercolornorsolidityisanon-entity(T1.2.4.3).Sincetherecanbenogeometryofinvisibleandintangibledistance,itisnotspace.Insum,spaceasdealtwithingeometryhasparts;invisibleandintangibledistancelacksparts.OnemightfurtherobjectthatHumeissimplydiscussingappear-ancesofthingsbutnottheirrealnatures:Twillprobablybesaid,thatmyreasoningmakesnothingtothematterinhand,andthatIexplainonlythemannerinwhichobjectsaffectthesenses,withoutendeavouringtoaccountfortheirrealnatureandoperations.Thusthechargeisthatheismerelyfinessingargumentsthattheremustreallybevacuumsintheworld.Humeblithelyresponds,Ianswerthisobjection,bypleadingguilty,andbyconfessingthatmyinten-tionneverwastopenetrateintothenatureofbodies,orexplainthesecretcausesoftheiroperations.TheworldasitappearsistheonlylegitimateconcernofthePyrrhonianEmpiricist(T1.2.5.256).AgainsttheIdeaofTimewithoutChangeJustasonecannotconceiveofspaceexceptbyconceivingofobjectscoexisting,soonecannotconceiveoftimeexceptbyconceivingofobjectsinsuccession.Humesdoctrineisthattimeisnothingbutthemanner,inwhichsomerealobjectsexist(T1.2.5.28).60Fromthisdoctrineheconcludesthatwecannotconceiveofastedfastandunchangeableobjectenduringthroughtime.Weonlyimaginethatwecanconceiveit.Humepromisestoconsider,Bywhatfictionweapplytheideaoftime,eventowhatisunchangeable,andsuppose,asiscommon,thatdurationisameasureofrestaswellasofmotion(T1.2.3.11).60Hereagainbeingrealcontrastswithnotexisting;seenotes24,36.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime143TheconnectionbetweenthedoctrineandtheconclusionsisHumesdenialthatasteadfastobjectenduresthroughtime.HereHumeisroughlyconcernedwiththegeneralissueraisedbyAristo-tle,whethertherecanbetimewithoutchange.61Hisconcernisnotwhethertherecanbetimeinasteadfast,changelessuniverse,orinanemptyuniverse.Clearly,fortimeasitappearshewouldthinknot;ifthereisnosuccession,thenthereisnomannerinwhichsuccessiveobjectsexist.AsforNewtonsclaimthatabsolutetrue,andmathematicaltime,ofitself,andfromitsownnature,flowsequablywithoutrelationtoanythingexternal,hewouldsuspendjudgment.62Inanyevent,hisconcernisratherwiththemoreimme-diatecasesofparticularsteadfastobjects,suchasstonesandhouses.Hisclaimthatthesedonotenduremightseemimplausible,andsohehastoexplainhimself:Iknowtherearesomewhopretend,thattheideaofdurationisapplicableinapropersensetoobjects,whichareperfectlyunchangeable;andthisItaketobethecommonopinionofphilosophersaswellasofthevulgar.Buttobeconvincdofitsfalshoodweneedbutreflectontheforegoingconclusion,thattheideaofdurationisalwaysderivdfromasuccessionofchangeableobjects,andcanneverbeconveydtothemindbyanythingstedfastandunchangeable.(T1.2.3.11)Becausetheideaoftimeordurationisageneralideaofsucces-sions,nonsuccessionscannotconveyittothemind.Changeableobjectsareonesthatalter,move,orarereplacedbyothers.Steadfastandunchangeableobjectsareonesthatneitheralternormove,andthatarenotreplacedforawhile.Theflamesinafireplaceareexam-plesofchangeableobjects,whereasthemantleabovethefireplaceisanexampleofasteadfastone.Theexamplesdonothavetobeexter-nalobjects,however.Whenapersonisstronglyoccupydwithonethought,thatthoughtisasteadfastobject.Humespecificallycon-trastssteadfastobjectswithsuccessions.Steadfastobjectsarenotsuccessions,socannotconveytheideaofduration:anunchange-ableobject,sinceitproducesnonebutco-existentimpressions,pro-ducesnonethatcangiveustheideaoftime;andconsequentlythatideamustbederivdfromasuccessionofchangeableobjects,andtimeinitsfirstappearancecanneverbeseverdfromsucha61PhysicsIV,11.62Newton,Principia,1:6.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n144donaldl.m.baxtersuccession(T1.2.3.78).Humegoesontoconcludethatsteadfastobjectsdonotendure,thatis,donothaveduration:Foritinevitablyfollowsfromthence,thatsincetheideaofdurationcannotbederivdfromsuchanobject,itcanneverinanyproprietyorexactnessbeapplydtoit,norcananythingunchangeablebeeversaidtohavedura-tion.Ideasalwaysrepresenttheobjectsorimpressions,fromwhichtheyarederivd,andcanneverwithoutafictionrepresentorbeapplydtoanyother.(T1.2.3.11)NotethatHumehasnotarguedthatsteadfastobjectsarenotintime.Theyareintime,insofarastheyaremembersoflargersucces-sions.Itisjustthattheydonothavedurationbecausetheyarenotsuccessionsthemselves.Onceitisclearthatsteadfastobjectsarenotsuccessions,itbecomesobviouswhyHumesayswecanhavenoideaofastead-fastobjectwithduration.Durationissuccessiveness.Noimpressioncouldgiverisetotheideainquestionunlessitwereanimpressionofasuccessionwithoutsuccessiveness.Butifyoucannotpointoutanysuchimpression,youmaybecertainyouaremistaken,whenyouimagineyouhaveanysuchidea(T1.2.5.28).Despitenotbeingsuccessionsandsolackingduration,individ-ualsteadfastobjectscancoexistwithsuccessions.Forexample,thesinglethoughtwithwhichthepersonisstronglyoccupiedcancoex-istwitharealsuccessionintheobjects.63Thatpersonwouldbeinsensibleoftimethoughitwouldbepassing,becausetherewouldnotbeasuccessionofimpressionsreflectingthesuccessionsintheworld(T1.2.3.7).Ifasteadfastobjectwereawholemadeupofasuccessionofparts,thenitwouldendure.Itwouldbethesuccessionofitsparts.Sinceitdoesnotendure,ithasnoparts.Yetitcoexistswithsomesuccessions.Sobigger,temporallypartlessthingscancoexistwithsmallertemporallypartlessthings.Itisbecauseallsteadfastobjectscoexistwithsuccessionsthatwetend,falsely,toimagineinaconfusedwaythattheyhavedurationaswell:Butthoitbeimpossibletoshowtheimpression,fromwhich63HerebyrealHumemeansthatthesuccessionofobjectsisoutsidethemindintheworldasitappears,insteadofmerelybeingasuccessionofperceptionsinthemind.(Ofcourse,thereismoretotheworldasitappearsthanjustwhatappearstothepersonstronglyoccupydwithonethought.)CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesTheoryofSpaceandTime145theideaoftimewithoutachangeableexistenceisderivd;yetwecaneasilypointoutthoseappearances,whichmakeusfancywehavethatidea(T1.2.5.29;seealsoT1.2.3.11,1.2.5.28,1.4.2.29).Justastherearetwokindsofspatialdistancesotherearetwokindsoftemporalseparation.Objectsintimecanbeseparatedbyaninterven-ingsteadfastobject(temporalseparationbysomethingunchanging,thatis,bysomethingwithoutduration),aswellasbyaninterveningsuccession(temporalseparationbysomethingchanging,thatis,bysomethingwithduration).Becauseofthecloserelationsbetweenthesetwoseparations,wetendinadvertentlytosubstitutethesec-ondwhenthinkingaboutthefirst.Justasweimaginethatwehaveanideaofavacuum,weimaginewehavetheideaofanenduringsteadfastobject.Therelationsbetweenthetwokindsoftemporalseparationpar-allelthosebetweenthetwokindsofdistance,andsoHumeonlybrieflyexplainstheprocessofarrivingatthefalselyimaginedideaofchangelessduration.Thereaderisclearlymeanttofillinthedetailsbasedonthepreviousdiscussion.First,wecanconsiderastedfastobjectatfive-a-clock,andregardthesameatsix,justaswecanconsiderthechimingoftheclockat5:00andthenat6:00.Betweenthesenoticingstheactionofthemindaboutitsotherbusi-nesscanbethesame.Thefirstandsecondappearancesoftheobject,beingcompardwiththesuccessionofourperceptions,seemequallyremovdasiftheobjecthadreallychangd.64Thisisanalogoustothemotionoftheeyebeingthesamewhetherattendingtothingsatinvisibleandintangibledistanceoratvisibleortangibledistance.Second,theobjectwassusceptibleofsuchanumberofchangesbetwixttheseappearances.Itmighthavechangedevenifitdidnot.Itmighthavebeenmovedorpaintedorwhatever.Sothepotentialforchangeinacaseofchangelesstemporalseparationputsoneinmindofchangeinaduration.Thisisanalogoustothewaythecapac-ityofinvisibleandintangibledistancetobeconvertedintovisibleortangibledistancerelatesthosetwointhemind.Third,qualitiesofobjectsseparatedbyaperceivedchangelesstemporalseparationundergothesamechangesasthoseseparatedbyaperceivedduration:theunchangeableorratherfictitiousdurationhasthesameeffect64Herereallyindicateshowappearancereallyisasopposedtohowitisfalselysupposedtobe.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n146donaldl.m.baxteruponeveryquality,byencreasingordiminishingit,asthatsucces-sion,whichisobvioustothesenses.Whenonereturnsonesgazetothefireaftercontemplatingtheunchangingviewoutthewindowforawhile,thefireisperceivedtohaveburneddownasmuchasifonehadbeencontinuouslywatchingitdance.Analogously,invisi-bleandintangibledistancehasthesameeffectonthequalitiesofdistantobjectsasvisibleortangibledistance(T1.2.5.29).Thesecloserelationsmakeusfancywehavethatideaofdura-tioninwhichnothingchanges.Theideaofthesteadfastobjectaccompanyingthephrasenothingchangesisreplacedbyanideaappropriatetothewordduration.Perhapsthereplacementideaisofaclosesuccessionofobjectseachresemblingthesteadfastobject.Thusonetalksasifthephrasedurationinwhichnothingchangesstandsforacoherentidea.Wethencometouseotherphrasesasequivalent,suchastimewithoutachangeableexistence(T1.2.5.29).conclusionHumessystemconcerningspaceandtimecannotbeunderstoodwithoutseeingitasaconsequenceofhisPyrrhonianEmpiricism.Hisexclusiveconcernwithviewsforcedonhimbyappearancesleadshimtoclaimthatspaceandtimeconsistofindivisiblepointsandcannotbeconceivedwithoutthem.Verylikelyheexpectedthatagreementfromhisreaderswouldcomerathereasily.Afterallhewasarguing,ashesawit,onthesideofcommonsenseconcerningthecoreissueofinfinitedivisibility,anissueoverwhichphilosophyandcommonsense...havewagedmostcruelwarswitheachother(A29).Ironically,hisdiscussionofspaceandtimehasmetwithdeterminedresistance.Theresistancehasbeenmisdirected,how-ever.IfonewantstoengageHumeconcerninghisviewsonspaceandtime,onemustatthesametimegrapplewiththeskepticismonwhichtheydepend.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nmartinbell5HumeonCausationHumestheoryofcausationisoneofthemostfamousandinflu-entialpartsofhisphilosophy.WhencomparedwiththeaccountsprovidedbyearlierphilosopherswhomHumestudied,suchasRene´Descartes(15961650),JohnLocke(16321704),andNicolasMale-branche(16381715),histheoryisrevolutionary.Itisalsocontro-versial,andhasbeeninterpretedinanumberofdifferentways.Thisisnotsurprising,becauseHumesideasaboutcausationarenotonlychallenginginthemselves,butalsolieattheheartofmuchoftherestofhisthought.Asaresult,interpretationsofHumeoncausa-tioninfluence,andareinfluencedby,interpretationsofhisgeneralphilosophicalaims,methods,andpurposes.Humesaccountofcausationispartofhistheoryofhumannatureorscienceofman.Thetheoryisapartlyepistemological,partlypsychologicalinvestigationofhowhumanbeingsacquirebeliefsandknowledge,makemoral,political,andaestheticjudgments,andactinandreacttothenaturalandsocialworld.Humesapproachtothesequestionsisgenetic.Hesetsouttoidentifytheoriginsofthoughts,feelings,judgments,andpatternsofbehaviorandresponse.Histheoryofcausation,therefore,isnotprimarilyametaphysicalaccountofwhatcausalityconsistsin,althoughithasimplicationsforthat.Rather,itisaninvestigationoftwomainquestions.First,howdohumanbeingscometohavetheideaofcausation?Second,howdohumanbeingscometobeabletoinfereffectsfromcausesandcausesfromeffects?Humesanswertothefirstquestionisthattheideaofcausalconnectionsbetweenobjectsandeventsdependsgeneticallyonthecapacitytoinfercausesfromeffectsandviceversa.Hisanswertothesecondquestionisthatthecapacitytomakecausalinferencesdoesnotoriginateinreasonoraprioriknowledgebutin147CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n148martinbellthewayinwhichexperienceofrepeatedregularsequencesofeventsaffectstheimagination.ThusinHumestheorytheanswertothefirstquestiondependsontheanswertothesecond,andnot,asonemightexpect,theotherwayround.WhileHumesanswerstothesequestionsarethecoreofhisaccountofcausation,thereareotherwaysinwhichcausationiscentraltohisphilosophy.Thetheoryofhumannature,orscienceofman,asageneticinvestigationoftheoriginsofhumancognitiveandemotionallife,isitselfacausaltheory.Itis,hesays,anattempttointroducetheexperimentalmethodofreasoningintomoralsubjects.1BytheuseoftheexperimentalmethodHumemeanstheattempttodiscovercausallawsonthebasisofobservationandexperienceratherthanonthebasisofsupposedrationalinsightintotheessenceornatureofthings.2Thedistinctionbetweentheexper-imentalmethodontheonehandandsupposedrationalinsightintothenatureofthingsontheother,andHumesadoptionoftheoneandrejectionoftheother,isitselfsupportedbyhiscentralargumentsabouttheideaofcausationandthenatureofcausalinference.Theargumentshaveconsequencesforthemethodologyofbothnaturalandmoralphilosophy.3Theapplicationoftheexperimentalmethodtohumannatureispredicatedontheassumptionthattherecanbenoknowledgeoftheessenceofthemind,oroftheultimateoriginalqualitiesofhumannature.Knowledgeoftheprinciplesofhumannature,likeknowledgeoftheprinciplesofnatureingeneral,isbuiltupgraduallyfromobservationofparticularevents,theforma-tionofbeliefsabouttheircauses,thebringingofparticularcausalconnectionsundermoregeneralcausallaws,andtheendeavourtorenderallourprinciplesasuniversalaspossible,bytracingupourexperimentstotheutmost,andexplainingalleffectsfromthesimplestandfewestcauses(TIntro.8).Causation,itisclear,iscentraltoHumesmethodology.1FromthesubtitleoftheTreatise.2SeeTIntro.7.3InHumestimenaturalphilosophymeantroughlywhatismeanttodaybynaturalscience,althoughwhatsciencesthesetermscoverandhowtheyaredifferentiatedhaschanged.Moralphilosophymeantthestudyofwhattodayweregardashistory,economicandpoliticaltheory,logic,epistemologyandmetaphysics,aesthetics,andethics.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation149Humealsouseshistheoryofcausationincriticaldiscussionsofphilosophicalquestionsinmoralphilosophyandnaturaltheology.InboththeTreatise(T2.3.12)andthefirstEnquiry(Sect.8)heexaminesthequestionoflibertyandnecessity,thatis,theques-tionwhetherhumanactionsarecausallydeterminedandwhether,ifso,thisshouldleadustoreviseorabandonthewaysinwhichweascribeandacceptmoralresponsibilityforthem.ExactlywhatHumearguesinthesesectionsisnotsomethingthatcanbeconsid-eredinthisessay.4ButcertainlypartofHumesargumentsinbothbooksisthathisrevolutionarytheoryofcausationenableshimtothrownewlightonaperennialproblemand,hethinks,toresolveit.Anotherequallyfamousapplicationofhistheorycomesinhiscriticalattacks,inthefirstEnquiry(Sect.11)andtheDialoguescon-cerningNaturalReligion,onthedesignargumentfortheexistenceofGod.IntheseattacksHumearguesthatanyinferencefromwhatwecanobserveofthenatureoftheworldconsideredasaneffect,toanomnipotentandwhollybenevolentDeityconsideredasitscause,violatesarangeofprinciplesthatfollowfromhistheoryofcausalinference.5Inshort,Humestheoryofcausationisarevolutionaryandinflu-entialdiscussionthatplaysacentralroleinhisoverallphilosophicalprojectandisusedbothconstructivelyandcriticallyinexaminingotherphilosophicalissues.Thesectionsthatfollowexaminethetheoryindetailandconsidersomeoftheinterpretativeproblemsthatitposes.4FordiscussionofthesesectionsseeinthisvolumetheessayHumesMoralPsy-chology.OtherdiscussionsincludePaulRussell,FreedomandMoralSentiment(NewYork:OxfordUniversityPress,1995);DonGarrett,CognitionandCommit-mentinHumesPhilosophy(NewYork:OxfordUniversityPress,1997),11836;GeorgeBotterill,HumeonLibertyandNecessity,inReadingHumeonHumanUnderstanding:EssaysontheFirstEnquiry,ed.P.Millican(Oxford:ClarendonPress,2002),277300;JamesHarris,OfLibertyandNecessity:TheFreeWillDebateinEighteenth-CenturyBritishPhilosophy(Oxford:ClarendonPress,2005),6487.5See,forexample,J.C.A.Gaskin,HumesPhilosophyofReligion,2nded.(London:Macmillan,1988);DavidOConnor,RoutledgePhilosophyGuidebooktoHumeonReligion(London:Routledge,2001).ItshouldbenotedthatinboththeEnquiryandtheDialoguescriticismsofthedesignargumentareputforwardbyfictionalcharactersratherthanbytheauthoroftheworkinhisownvoice.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n150martinbelli.causalinferenceOnetaskforthetheoryofhumannatureistoexplainhowbeliefandknowledgearise.Empiricalbeliefsaboutmattersoffactmayresultfromsenseperceptionandmemoryalone,butagreatmanybeliefsofthiskindconcernmattersoffactthatliebeyondtheevi-denceprovidedbyperceptionandmemory.Wheresuchbeliefsareproductsofreasoningtheyareconclusionsinferredfromotherfac-tualbeliefs,andultimatelyfrombeliefsproducedbyperceptionandmemory.Totakeonefactualbeliefasevidenceforanother,toinferonefromtheother,Humeclaims,istoassumethatthereisacausalconnectionbetweenthestatesofaffairsthatarebelievedtoobtain.Forexample,totakepuddlesonthepavementasevidencethatithasbeenrainingistoassumethatraincausedthepuddles.Some-onewhoseassumptionaboutthecauseofthepuddleswasdifferent,forexample,someonewhosupposedthattheywerecausedbyawatersprinkler,woulddrawadifferentconclusionfromobservingthepuddles.Inthekindofreasoninginwhichhumanbeingsdrawconclusionsabouttheexistenceofsomethingtheydonotobservefromtheexistenceofsomethingtheydoobserve,theytracetherelationofcauseandeffect(T1.3.5.7).Humestheoryisthatrea-soningfromfactualevidenceisreasoningfromcausestoeffects,orfromeffectstocauses.Humesaccountofcausalreasoningis,aspartofhistheoryofhumannature,aninvestigationintohowhumanbeingsacquirethecapacitytoinfercausesfromeffectsandviceversa.Hearguesthatinnocasecansuchinferencesbemadeapriori.ThisisanimportantpartofHumestheory,andhisreasonforitplaysamajorrolenotonlyintheexplanationofcausalinference,butalsointheaccountoftheoriginoftheideaofcausation.Hemeansthatwhenweobserveanobjectorevent,wecannotinfer,merelyfromtheideaweacquirefromthisimpression,anyotherobjectoreventasthecauseortheeffectofwhatweobserve:Tiseasytoobserve,thatintracing[therelationofcauseandeffect],theinferencewedrawfromcausetoeffect,isnotderivdmerelyfromasurveyoftheseparticularobjects,andfromsuchapenetrationintotheiressencesasmaydiscoverthedependanceoftheoneupontheother.Thereisnoobject,whichimpliestheexistenceofanyotherifweconsidertheseobjectsinthemselves,andneverlookbeyondtheideaswhichweformofthem.SuchCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation151aninferencewoudamounttoknowledge,andwoudimplytheabsolutecontradictionandimpossibilityofconceivinganythingdifferent.Butasalldistinctideasareseparable,tisevidenttherecanbenoimpossibilityofthatkind.Whenwepassfromapresentimpressiontotheideaofanyobject,wemightpossiblyhaveseparatedtheideafromtheimpression,andhavesubstitutedanyotherideainitsroom[orplace].(T1.3.6.1)InthefirstEnquiry,Humearguesthateveniftwoobjectsoreventsarecausallyconnected,theyare,nevertheless,distinctanddifferentfromeachother.Theirideasintheimaginationareseparable,andso,ifweareproceedingaprioriandnotonthebasisofpreviousexperience,whenwethinkofoneobjectwecouldalwaysthinkofanyotherobjectwelikeasapossiblecauseoreffect.Anychoiceisasarbitraryasanyother.Theimaginationisfreetoputtogetheranytwodifferentandseparableideas.Nothingabouttheideaofaparticularobjectoreventfixes,or,touseHumesterminology,determinestheimaginationtoputthemtogether:6Inaword,then,everyeffectisadistincteventfromitscause.Itcouldnot,therefore,bediscoveredinthecause,andthefirstinventionorconceptionofit,apriori,mustbeentirelyarbitrary.Andevenafteritissuggested,theconjunctionofitwiththecausemustappearequallyarbitrary;sincetherearealwaysmanyothereffects,which,toreason,mustseemfullyasconsistentandnatural.Invain,therefore,shouldwepretendtodetermineanysingleevent,orinferanycauseoreffect,withouttheassistanceofobservationandexperience.(EHU4.11)Causalinference,Humeconcludes,isimpossibleaprioriandisdependentonobservationandexperience.Thefeatureofexperiencethatheidentifiesascrucialisrepetition.ApersonwhoexperiencesforthefirsttimetheoccurrenceofaneventofkindAsucceededbytheoccurrenceofaneventofkindBwillnotthinkofA-typeeventsascausesofB-typeevents,nor,ifandwhenaneventofeithertypeisnextexperienced,willthepersonnormallyinferthataneventofthe6ForHumesnotionofdeterminationanditsimportanceinhisthought,seeAnnetteC.Baier,AProgressofSentiments:ReflectionsonHumesTreatise(Cambridge,MA:HarvardUniversityPress,1991),78100.ForHumesnotionofaprioriinfer-ence,andthecontrastbetweenarbitraryanddetermined,seePeterMillican,HumesScepticalDoubtsconcerningInductioninReadingHumeonHumanUnderstanding,10769.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n152martinbellothertypehasoccurredorwilloccur.7However,Humeclaims,ifABsequencesoccurrepeatedlyinapersonsexperiencethepersonwillcometoinferthataB-typeeventwilloccurwhenheorsheexperi-encesanA-typeevent,andtoinferthatanA-typeeventhasoccurredwhenaB-typeeventisexperienced.AtthesametimethepersonwillcometobelievethatA-typeeventscauseB-typeevents.Inbrief,onthistheoryhumanbeingsrequireexperiencebeforetheycanmakecausalinferences,andthisexperiencetakestheformofrepetitionofregularsequencesofevents.Experiencingrepeatedregularsequencesofeventsisacausallynecessaryconditionforacquiringthecapacitytomakecausalinfer-ences.Aftersuchexperiencetheconjunctionofthecauseandtheeffectnolongerappearstothemindasarbitrary.Rather,afterrepeatedexperienceofaregularsequenceofeventsoftypesAandB,whenatoken(anexample)oftheonetypeisnextexperiencedthemindisdeterminedtoinfertheoccurrenceofatokenoftheothertype.Wemakecausalinferences,then,inconformitytoourpastexperience(T1.3.6.2).Humenowaskswhatelse,besidesexperi-enceofwhathecallstheconstantconjunctionofevents,isneededinordertomakecausalinferences.Howdoestheexperienceofregu-larityresultinthedeterminationofthemindtoexpecttheregularitytocontinue?Whydoweexpectthatfurtherinstancesofthesametypesofeventwilloccurinconformitytoourpastexperience?Causalinferencesarealwaysmadeinawaythatmanifeststheassumptionthatcausalconnectionsareuniforminalltimesandplaces.Inoneofhismostfamousarguments,Humeclaimstoprovethatthisassumptionoftheuniformityofcausalconnectionscan-notbeestablishedastheconclusionofanyformofreasonedargu-ment.Thereare,hesays,nodemonstrativeargumentstoprovetheprinciplethatinstances,ofwhichwehavehadnoexperi-ence,mustresemblethose,ofwhichwehavehadexperience,andthatthecourseofnaturecontinuesalwaysuniformlythesame(T1.3.6.45).ByademonstrativeargumentHumemeansthekindofproofofapropositionthatshowsthatitcouldnotpossiblybefalse,thatitisnecessarilytrueinwhathecallstheabsolutesensethat7However,Humedoesgrantthatoncethehabitofcausalinferencehasbeenestab-lished,asingle,carefullydesignedexperimentmaybeenoughtoproduceinference.See,forexample,T1.3.8.14,1.3.15.6.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation153characterizesmathematicaltruthssuchas2+2=4.8Wheresomethingcanbedemonstrated,Humethinks,themindisdeter-minedtoconceivethepropositionbyreasonalone:Whenademon-strationconvincesmeofanyproposition,itnotonlymakesmecon-ceivetheproposition,butalsomakesmesensible,thattisimpos-sibletoconceiveanythingcontrary.Whatisdemonstrativelyfalseimpliesacontradiction;andwhatimpliesacontradictioncannotbeconceived(A18).Butwecanconceivethattheconstantconjunc-tionsofeventsthatwehaveexperiencedsofarwillnotcontinueinthefuture;wecanconceiveachangeinthecourseofnature.Iftheprincipleoftheuniformityofnatureisnotprovableapri-ori,byademonstrativeargument,canitbeinferredonthebasisofevidence?Humearguesthatitcannot.Anyproofofitbyinferencefromwhatwehavesofarexperiencedwouldbecircular,becauseinferencefrompastexperiencewouldalreadyassumethepointinquestion.Therefore,itisnotpossibletoexplainhowexperienceenableshumanbeingstomakecausalinferencesandacquirecausalbeliefsbyappealingtotheircapacityforreason.Thisfamousargument,commonlyreferredtoasHumesskep-ticismaboutinduction,iscentraltohisaccountoftheoriginoftheabilitytomakecausalinferences.Becausememoryandreasonalonecannotbethebasisonwhichwemakecausalinferencesanddrawonourpreviousexperiencetoformourexpectationsofwhatwillhappeninnewcases,Humeconcludesthatthereisanotherprincipleofhumannaturewhoseeffectonthemindaccountsforourcausalinferences.Thisistheprincipleofcustomorhabit(T1.3.8.12,1.3.12.9).Whenthereisrepeatedexperienceofthecon-junctionoftypesofevents,theideasofthesetypesofeventsbecomeassociatedintheimaginationbycustom.Whatthismeansisthattheimaginationcomestobestructuredinsuchawaythatwithinittheideasoftypesofeventsthathavebeenrepeatedlyexperiencedasconjoinedareconnectedtogether.Theimaginationbecomeshabit-uatedtomovefromoneideatoanotherwithoutanyintermediatestep.9RepeatedexperienceoftheconjunctionofA-typeeventswith8T1.3.7.3.9AlthoughHumesometimesusesthetermscustomandhabitasiftheydesignatedthesamephenomenon,itseemsbettertothinkofcustomasdenotingthewayinwhichrepetitionsofexperienceleadtoideasbeingassociatedintheimagina-tion,andhabitasdenotingaresultingtendencyfortheimaginationtotransferitsCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n154martinbellB-typeeventsproducesabeliefintheoccurrenceofatokeneventoftheonetypewhenatokeneventoftheothertypeisexperienced.Hume,then,givesacausalexplanationoftheabilitytomakecausalinferences,anexplanationthatpresentsthisabilityastheeffectofafundamentalprincipleofhumannature,custom,orhabit.AfterexperiencingeventsoftypesAandBconstantlyconjoined,whenanotherinstanceofanA-typeeventisobserveditisnotreasonthatdeterminesthemindtoinferthataB-typeeventwilloccur.Whatdeterminesthemindiscustomandhabit.10ii.theideaofnecessaryconnectionInarguingforhistheoryofhowwecometobeabletomakecausalinferencesHumeclaims(1)thatnosuchinferencecanbemadeapriori,(2)thatexperienceoftheconstantconjunctionoftypesofeventsisnecessarybeforewecaninferonefromtheother,and(3)thatevenafterwehavesuchexperienceitisnotreasonbutcus-tomthatdeterminesthemindtomaketheinference.Onemightbeinclinedtoobjectthatthemovefrom(2)to(3)restsontoothinanotionofwhatexperienceofregularsequencesofeventsreveals.Aswehaveseen,Humespeaksoftheexperienceofconstantconjunc-tions.Hemeansbythisexperienceofeventsoftwotypes,AandB,alwaysoccurringinaregularorderofcontiguityandsuccession:thatis,A-typeandB-typeeventsalwaysoccurwithoutanyspatialortemporalintervalbetweenthem,andA-typeeventsalwayspre-cedeB-typeevents(T1.3.6.2).11However,asHumehimselfpointsout,theideathatAcausesBisnotjusttheideathatA-typeeventsandB-typeeventsareconstantlyrelatedbypriorityandcontiguity.TheideathatAcausesBincludesasamostimportantelementtheideathatBmustoccurifAoccurs.Thisidea,thataB-typeeventattentionandthevivacityofitsconceptionfromoneideatoanothercustomarilyassociatedwithit.Humereferstothehabitualmovementofthemindfromoneideatoanotherasatransition,andhesaysthatthetransitionofthemindisessentialtorelations(see,e.g.,T1.3.5.1,1.4.2.34).10ForHumesownsummaryaccountoftheprocessbywhichwecometomakecausalinferencesandbelieveincausalconnections,seeA826.11IntheTreatiseHumearguesthattherecannotbeaneventthatisthedirecteffectofacausethatisspatiallydistantfromit;causeandeffectmustbecontiguous.InthefirstEnquiryhenolongertreatsthecontiguityofcauseandeffectasessentialtodirectcausation.Inbothworkshesays,however,thatcausesmustbetemporallypriortotheireffects.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation155necessarilyoccurswheneveranA-typeeventoccursiswhatHumecallstheideaofnecessaryconnexion(T1.3.2.11).ButifexperienceshowsthatA-typeeventsarenecessarilyconnectedtoB-typeevents,wouldthatnotimply,oncetherehasbeensufficientexperiencetorevealthisconnection,thattheinferencefromtheobservationofanA-typeeventtoaB-typeeventis,afterall,determinedbyreason?ForHume,thislineofthoughtismisconceived.Heseekstoshowthattheideathatcausesandeffectsarenecessarilyconnectedisonethatcouldnotarisesimplyfromtheexperienceoftheconstantcon-junctionsofevents.Infact,oncetherehasbeenexperienceofsuchaconjunction,itisonlybecausethemindisdeterminedbycustomtoinferfutureeventsinconformitytopastexperiencethattheideaofanecessaryconnectionbetweencausesandeffectsarises.Theexpe-rienceofmereconstantconjunctionsalone,withouttheeffectofcustomandhabitontheimagination,wouldnotproducethisidea.ThatiswhyinboththeTreatiseandthefirstEnquirythesectionsentitledOftheideaofnecessaryconnexion(T1.3.14,EHU7)comeafterHumehasarguedthatinferencesfromcausesandeffectsdependnotonreasonbutoncustomandhabit.ThesesectionsprovideananswertothequestionWhatisourideaofnecessity,whenwesaythattwoobjectsarenecessarilyconnectedtogether?(T1.3.14.1).Theanswerappealstothecustomaryassociationsbetweenideasofobjectsthathavebeenrepeatedlyexperiencedasoccurringtogetherinsuccession,andtheresultinghabitualtransitionsofthemindfromtheonetotheother.IntheTreatisethesectionontheideaofnecessaryconnectionbrieflyoutlinesHumestheoryoftheoriginofthisidearightatthestart,inthefirstparagraph.Followinghisprinciplethatideasarederivedfromimpressions,Humeconsiderswhatrelevantimpres-sionswehaveofanytwoobjectsthatwetaketoberelatedascauseandeffect.Examiningasinglecase,hesays,revealsonlythatimpres-sionsofthecausearetemporallypriortotheeffect,andthatthetwoimpressionsarespatiallyandtemporallycontiguous.Thereisnoimpressionofathirdrelationthatcouldgiverisetotheideaofanecessaryconnection.12Ifweproceedtoconsidermultipleinstancesofthesamekindofcauseandeffect,thesesamerelationsofpriority12SeeT1.3.2.12andtheEditorsAnnotationtothistextinATreatiseofHumanNature,ed.D.F.NortonandM.J.Norton,2vols.(Oxford:ClarendonPress,2007),2:735.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n156martinbellandcontiguityarerepeated.Atfirstsightitappearsthatrepetitiondoesnotproduceanynewimpression.Humesays,however,thatrepetitiondoesinfactproduceanewimpression,foritchangesthewaythemindrespondstoanexperienceofoneoftheconstantlyconjoinedobjects:Forafterafrequentrepetition,Ifind,thatupontheappearanceofoneoftheobjects,themindisdetermindbycus-tomtoconsideritsusualattendant,andtoconsideritinastrongerlightuponaccountofitsrelationtothefirstobject.Tisthisimpres-sion,then,ordetermination,whichaffordsmetheideaofnecessity(T1.3.14.1).Afterthisbriefstatementofhispositivetheory,Humeproceedsbydrawingattentiontoitssignificance.Followinghisempiricistprinciplethattracingtheoriginofanideatoitssourceinimpres-sionsservestoclarifytheideaandmakemoreprecisethemeaningoftermsthatareusedtoexpressit,heclaimstohavejustnowexamindoneofthemostsublimequestionsinphilosophy,viz.thatconcerningthepowerandefficacyofcauses(T1.3.14.2).Therele-vanceofdiscoveringtheoriginoftheideaofnecessaryconnectiontometaphysicalaccountsofcausalityisalsoemphasizedatthestartofthecorrespondingsectionofthefirstEnquiry:Therearenoideas,whichoccurinmetaphysics,moreobscureanduncertain,thanthoseofpower,force,energy,ornecessaryconnexion,ofwhichitiseverymomentnecessaryforustotreatinallourdisquisitions.Weshall,therefore,endeavour,inthissection,tofix,ifpossible,theprecisemeaningoftheseterms,andtherebyremovesomepartofthatobscurity,whichissomuchcomplainedofinthisspeciesofphilosophy.(EHU7.3)ItcanbeseenfromthesequotationsthatHumesupposesthattheideaofanecessaryconnectionbetweenacauseandeffectisprettywellthesameastheideathatthecausehasapowerorforceorenergythatproducestheeffect.Indeed,hesays,thetermsofeffi-cacy,agency,power,force,energy,necessity,connexion,andpro-ductivequality,areallnearlysynonimous(T1.3.14.4).Tounder-standhisdiscussionsoftheideaofnecessaryconnectiononemustkeepthisassumptioninmindthroughout.Philosophicalaccountsofcausationhavebeengivenaccordingtowhichthesenseinwhichacauseisnecessarilyconnectedtoitseffectistobeexplainedbysaying,veryroughly,thattheoccurrenceoftheeffectislogicallydeduciblefromtheoccurrenceofthecause,assumingthatasetofCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation157initialconditionsandanonnecessarylawofnatureobtain.13Athe-oryofcausalnecessityofthiskindneednotemployconceptssuchaspowerandefficacyatall.Humesaccountoftheoriginoftheideaofnecessaryconnectionis,however,giveninaphilosophicalandhistoricalcontextinwhichtheideasofcausalnecessityareboundupwithideasaboutcausalpowersandforces.Hedescribesacausalpowerasaquality,whichbindstheeffecttothecause,andrenderstheoneaninfallibleconsequenceoftheother(EHU7.6).Similarly,hedescribestheefficacyofcausesasthatqualitywhichmakesthembefollowdbytheireffects(T1.3.14.3).ThusHumeassumesthattothinkthatacauseisnecessarilyconnectedtoitseffectistothinkthatthecausehasapowerorefficacywhoseoperationinproducingtheeffectisinfallibleorunfailing.Giventhiswayofunderstandingcausalnecessity,itisnaturaltoassumethattheideaofanecessaryconnectionmustarisefromexperienceofcausalpowers.Indeed,Humethinksthatbeforewestarttothinkphilosophicallyaboutthematterweusuallytakeitthatwedodirectlyperceivecausalpowersoperating:Thegeneralityofmankindneverfindanydifficultyinaccountingforthemorecommonandfamiliaroperationsofnature;suchasthedescentofheavybodies,thegrowthofplants,thegenerationofanimals,orthenour-ishmentofbodiesbyfood:Butsuppose,that,inallthesecases,theyperceivetheveryforceorenergyofthecause,bywhichitisconnectedwithitseffect,andisforeverinfallibleinitsoperation.(EHU7.21)13Itisinteresting(butmistaken)thattheoriesofthiskind,socallednomologicalaccountsofcausation,areoftenclaimedtohavetheirorigininHumeswork.Forexample,ErnestSosa,arguingthatnomologicalaccountscannotcaptureallkindsofcausation,says:ItisanessentialfeatureofnomologicalaccountsofcausationthataccordingtothemaneventorstateofaffairsP(partially)causes(orisacauseorcausalfactorof)another[stateofaffairs]Qonlyifthereareactual(initial)conditionsIandalawofnatureLsuchthat,bynecessity,ifPandIandLallobtainthenQmustobtain,whereLisessentialinthatPandIalonedonotnecessitateQ.Itisfurthercommonlyassumed,especiallybyHumeans(regularists),(i)thatlawsmakenoessentialreferencetoanyparticulars:thatlawsarepurelygeneral,and(ii)thatlawsarenotnecessarytruths(VarietiesofCausationinCausation,ed.E.SosaandM.Tooley[Oxford:OxfordUniversityPress,1993],23442).ThisillustratesthewidespreadassumptionthatHume,indenyingthatwehaveanideaofcausalpowerasapropertyofobjects,deniedthatcausesnecessitatetheireffectsinanysenseotherthanthatexpressedbysuchnomologicaltheories.Forexample,A.H.BassonclaimsthatHumesaysthatcausalityisnothingmorethanuniformityofsequence,DavidHume(London:Pelican,1958),75.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n158martinbellItisHumesview,however,thatthiscommonsuppositionismis-taken.Weneverdo,andinfactwenevercan,observeanyqualityorpropertyofcausesthatconstitutescausalpower.Inobservingasin-gleinstanceofacausalrelation,heargues,noneoftheperceivablequalitiesoftheobjectsinvolvedconstitutespowerornecessarycon-nectionbecausenoneofthemisaquality,whichbindstheeffecttothecause,andrenderstheoneaninfallibleconsequenceoftheother.Perceivableorsensiblequalitiesarealwayscompleatinthemselves,andneverpointoutanyothereventwhichmayresultfromthem.Butforaperceivablequalitytocountasacausalpoweritwouldneedtobesuchastopointouttheeffectitproduces.Indeed,ifcausalpowersweresensiblequalities,thenknowledgeoftheeffectsofanobjectwouldbecomepossibleapriori:Butwerethepowerorenergyofanycausediscoverablebythemind,wecouldforeseetheeffect,evenwithoutexperience;andmight,atfirst,pro-nouncewithcertaintyconcerningit,bythemeredintofthoughtandreasoning(EHU7.68).Humearguesforthesameconclusioninconsideringthecausalrelationbetweenvolitionandbodilymovementormentalactions.HediscussesthisbrieflyintheTreatise(T1.3.14.12)andatgreaterlengthinthefirstEnquiry.Weknowfromexperiencethatvolitioncanresultinbodilymotion,as,forexample,thevoluntarymove-mentofanarm.Butthereisnoimpressionofthepowerofthewill.Humegivesthreeargumentsinsupportofthisconclusion.Thefirstisthatitisamysteryhowthesoulandthebodyareunited:Butifbyconsciousnessweperceivedanypowerorenergyinthewill,wemustknowthispower;wemustknowitsconnexionwiththeeffect;wemustknowthesecretunionofsoulandbody.Thesecondisthatwecanvoluntarilymovesomepartsofthebodybutnotothers,andwediscoverthisonlybyexperience;whereas,werewecon-sciousofthepowerofthewillintheonecaseanditsimpotenceintheother,weshouldthenperceive,independentofexperience,whytheauthorityofwillovertheorgansofthebodyiscircum-scribedwithinsuchparticularlimits(EHU7.1112).Thethirdisthatifwehadanimpressionofthepowerofthewill,wewouldthenknowtheeffectofthewill;butinfactwedonotknowtheeffectsofthewillwhenwemoveanarm,forexample,becausetheimmediateobjectofpowerinvoluntarymotion,isnotthememberitselfwhichismoved,butcertainmuscles,andnerves,andanimalspirits,and,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation159perhaps,somethingstillmoreminuteandmoreunknown,throughwhichthemotionissuccessivelypropagated,ereitreachthememberitselfwhosemotionistheimmediateobjectofvolition.Cantherebeamorecertainproof,thatthepower,bywhichthiswholeoperationisperformed,sofarfrombeingdirectlyandfullyknownbyaninwardsentimentorconscious-ness,is,tothelastdegree,mysteriousandunintelligible?(EHU7.14)Humegivesthreesimilarargumentstoshowthatthereisnoimpressionofpowerornecessaryconnectionwhenwevoluntarilyperformmentalactions.Asbefore,theargumentsturnontheclaimthatifthereweresuchanimpression,therewouldbeknowledgeofakindthatwedonotactuallypossess.Weknowthatwecandirectourthoughts,exercisingsomecontroloverourideas.Butwehavenoknowledgeofhowthemindproducesideas:themanner,inwhichthisoperationisperformed;thepower,bywhichitisproduced;isentirelybeyondourcomprehension.Again,asinthecaseofvolitionandbodilymotion,therearelimitstothecontrolofthewilloverideasandpassionsthatcanbediscoveredonlyfromexperience.Theyarenotknownbyacquaintancewiththenatureofcauseandeffect,orby,thatis,acquaintancewiththenatureofthepowerofthewilloverthoughtsandpassions.Finally,thefactthatthemindscommandofthoughtsandfeelingsvariesindifferentcircumstances,suchasinsicknessandinhealth,suggeststhatthereissomesecretmechanismorstructureofparts,uponwhichtheeffectdepends,andwhich,beingentirelyunknowntous,rendersthepowerorenergyofthewillequallyunknownandincomprehensible(EHU7.1719).Inconsideringwhethertheideaofnecessaryconnectionarisesfromobservationofthebehaviorofmaterialbodies,fromexperi-enceoftheinteractionofhumanmindsandbodies,orfromintro-spectionintotheworkingsofourminds,HumesargumentsareclosetothosegivenbyMalebranche.14HumerejectsMalebranchesconclusionsaboutcausation,butseeingtheextentoftheagreementbetweenthetwophilosophers,andthepointsofdifference,helpstohighlightwhatisrevolutionaryaboutHumestheory.LikeHume,Malebrancheseesacloserelationbetweennecessaryconnectionandcausalpower.Forhim,atruecauseissomethingthathasanactive14AtT1.3.14.7,n.29,HumereferstoMalebranchesSearchafterTruth6.2.3andtoElucidation15ofthiswork.SeeNicolasMalebranche,TheSearchafterTruthandElucidationsoftheSearchafterTruth,trans.T.M.LennonandP.J.Olscamp(Columbus:OhioStateUniversityPress,1980).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n160martinbellpowertobringabouttheeffect,andtruecausesnecessitatetheireffects.15Itisimpossiblethatthepowerofacauseacts,butthattherelevanteffectdoesnotfollowfromthatact.Consequentlyheholdsthatifthemindhasanideaofatruecause,therewillbeaconcep-tualconnectionbetweentheideaofthecauseandthatoftheeffect.Themindwillperceiveorseeanecessaryconnectionbetweenthecauseandtheeffect,meaningbythatthatitwillbeinconceivablethatthecauseshouldoperatewithouttheeffectresulting.Withthisunderstandingofnecessaryconnectionandcausalpowerinplace,Malebrancheseekstoshowthatthereisonlyonecausalpower,Godswill,andonlyonecaseinwhichwecanperceivenecessaryconnection:betweenthewillofGodandeverythingthathappens.Accordingtothisdoctrine,whenamovingbilliardballcollideswithastationaryone,itisnotthefirstballthatcausesthesecondtomove.TheimpactisnomorethantheoccasiononwhichGodwillsthatthesecondballshouldmove,whileitisjustthisactofwillingthatcausesittomove.16InMalebranchesoccasionalisttheory,theregularitiesweobserveinnaturaleventslikethecollisionsofmovingbodiesaretheresultofthefactthatGodsactsofwillarethemselvesregularanduniform.Onanyoccasionwhereaparticulartypeofeventoccurs,Godalwayswills,miraclesaside,thataparticulartypeofeventshouldfollow.Thuswecometoexperienceconstantconjunctionsofevents.GivenexperienceoftheconstantconjunctionofeventsoftypesAandB,hesays,weassumethatAcausesB,andformthehabitofexpectingBwhenweobserveA.Unfortunately,inhisview,wearealsopronetoassumethatAhasapowerornaturetoproduceB.Butthisisaprejudicefromwhichitisalmostimpossibletodeliveroneselfwithouttheaidthatcanbedrawnfromtheprinciplesofaphilosophythathasnotalwaysbeensufficientlyknown.17TheprejudicethatBisaneffectofAspowerblindsustotherealizationthatBisin15Malebranchedistinguishesbetweenapowertoact,whichisatruecause,andapassivecapacitytoundergochange.SeeNicolasMalebranche,DialoguesonMetaphysics,trans.W.Doney(NewYork:ArbarisBooks,1980),147,151.16ForthisreasonMalebranchestheoryisreferredtoasoccasionalism.Themotionofthefirstballisalsosaidtobeasecondarycauseofthemotionofthesecondball,becausethetrue,orprimarycauseisthedivinewill.Humeusesthetermi-nologyofsecondarycausesatT1.3.14.11.17Malebranche,SearchafterTruth,Elucidation15,6578.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation161facttheeffectofdivinepower,andthatAisnotthetruecauseofB,butonlytheoccasionforGodtoproduceBinaccordancewithhisgeneralwill.Thus,forMalebranche,beliefinnaturalforce,efficacy,orpoweriscontrarytosoundphilosophyandtruereligion.WhenMalebranchearguesthatsecondarycausesarenottruecauses,forthereisnonecessaryconnectionbetweenthemandanyothernaturalevent,hesoundsverylikeHume.Accordingtoocca-sionalisttheory,whenthewhitebilliardballstrikestheredone,thereisnonecessaryconnectionbetweenthisimpactandtheensu-ingmovementoftheredball,foritisconceivablethattheredballwillreactinsomewaydifferentfromusual.Itmayshatter,ordisap-pear,ordoanythingimaginable.Thereisneveranecessaryconnec-tionbetweenthemotionofonebodyandthemotionofanother.AndwhenIwillmyarmtomove,itisconceivablethatmyarmdoesnotmove.Thereisneveranecessaryconnectionbetweenvolitionandbodilymovement.WhenIwilltothinkofsomeobject,thereisneveranecessaryconnectionbetweenmyvolitionandtheoccurrenceoftheideaofthatobjectinmymind.Furthermore,noonereallyhasanyideaofwhatheorshemeansbytalkingofcausalpowersinbodiesorinfiniteminds:Therearemanyreasonspreventingmefromattributingtosecondaryornaturalcausesaforce,apower,anefficacytoproduceanything.Buttheprincipaloneisthatthisopiniondoesnotevenseemconceivabletome.WhatevereffortImakeinordertounderstandit,Icannotfindinmeanyidearepresentingtomewhatmightbetheforceorthepowertheyattributetocreatures.AndIdonoteventhinkitatemerariousjudgmenttoassertthatthosewhomaintainthatcreatureshaveforceandpowerinthemselvesadvancewhattheydonotclearlyconceive.Forinshort,ifphilosophersclearlyconceivedthatsecondarycauseshaveatrueforcetoactandproducethingslikethem,thenbeingamanasmuchastheyandparticipatingliketheminsovereignReason,Ishouldclearlybeabletodiscovertheideathatrepresentsthisforcetothem.ButwhatevereffortofmindImake,Icanfindforce,efficacy,orpoweronlyinthewilloftheinfinitelyperfectBeing.18Despitethesimilaritiesbetweensomeoftheargumentsusedbythetwophilosophers,Humesconclusionissignificantlydifferent18Malebranche,SearchafterTruth,Elucidation15,658.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n162martinbellfromthatofMalebranche.First,Humedeniesthatitfollowsfromthefactthatthereisnoconceptualconnectionbetweenideasofnaturalcausesandeffectsthatnaturalcausesarenottruecausesandhavenopowerstoproducetheireffects.Second,herejectsMalebranchesclaimtofindforce,efficacy,orpowerinthewilloftheinfinitelyperfectBeing.Third,hedeniesthatwecanthinkthatthereisanecessaryconnectionbetweenacauseanditseffectonlyifwecannotconceiveofthecauseoccurringwithouttheeffect(T1.4.5.31,1.3.14.10,33).AttheheartofHumescritiqueofMalebrancheishisempiricistaccountofthenatureofideas.ForHume,allideasareultimatelyderivedfromimpressionsofsenseorreflection.Andimpressionsarewhatideasrepresent.Ideasalwaysrepresenttheirobjectsorimpressions;andviceversa,therearesomeobjectsnecessarytogiverisetoeveryidea(T1.3.14.6).19Ideasdonot,therefore,rep-resenttheessencesofthings,wherethismeansalevelofrealityunderlyingtheappearancesofthesethings;onthecontrary,ideasrepresenttheappearancesofthingsandnothingmore.20ForMale-branche,however,thereareclearideasthatarenotderivedfromimpressionsofsenseorfeeling,andthese,heclaims,dorepresenttheessencesofthings.ThisnotionisusedbyMalebrancheinstatingaprincipleofcertaintyandevidence,namely,that,inassertingofathingwhateverisincludedintheclearideathatrepresentsit,onecanbecertain.21AsAndrewPyleexplains,MalebrancheregardsthisprincipleaslicensinginferencefromtheclearideaofXdoesnotcontainFtoXisnon-F.22HenceMalebranchesupposesthatthe19FordiscussionofHumestheoryofideasandimpressionsseeinthisvolumetheessayHumeandtheMechanicsofMind.InthequotedsentenceHumeisusingobjecttomeanobjectofthemind.Thusobjectandimpressionareinterchangeableinthiscontext.SeeMarjorieGrene,TheObjectsofHumesTreatise,HumeStudies20,no.2(1994):16377.20Myintentionneverwastopenetrateintothenatureofbodies,orexplainthesecretcausesoftheiroperations.Forbesidesthatthisbelongsnottomypresentpurpose,Iamafraid,thatsuchanenterprizeisbeyondthereachofhumanunderstanding,andthatwecanneverpretendtoknowbodyotherwisethanbythoseexternalproperties,whichdiscoverthemselvestothesenses(T1.2.5.26).21Fortextualreferencesanddiscussionofthisprinciple,seeJean-ChristopheBardout,MetaphysicsandPhilosophy,inTheCambridgeCompaniontoMalebranche,ed.S.Nadler(Cambridge:CambridgeUniversityPress,2000),Chap.6.22AndrewPyle,Malebranche(London:Routledge,2003),102.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation163factthatideasofsecondarycausesnevercontainanythingrepresent-ingcausalpowersjustifiestheconclusionthatsecondarycausesarecausallyimpotent.Incontrast,Humedoesnotinferfromtherearenoideasofcausalpowersinobjectsthattherearenocausalpow-ersinobjects.Certainly,therearenoappearancesofcausalpowersinobjectsorthemind.Butifcausalpowersbelongtotheessencesornaturesofthings,thentheproperthingforanempiricisttosayisthatthey,likeanythingelsetodowiththeessenceornatureofthings,areunknown(orare,asheoftensays,inconceivable,unintelligible,hidden,secret,andsoon).HumeuseshisempiricisttheoryofideastorefuteMalebranchesviewthatGodistheonetruecause.Thedemolitionisswift.Male-branchehasassertedthatsecondarycausesarenottruecausesbecauseourideasofthemdonotcontainrepresentationsofcausalpowers.Butthen,Humeobjects,becauseallideasarederivedfromimpressionsandtherearenoimpressionsofpower,neithercantheideaofGod,whichitselfmustderivefromimpressions,containarepresentationofpower.Itfollows,then,thatonecannomoreper-ceiveanecessaryconnectionbetweenthedivinewillandnaturaleventsthanonecanperceiveanecessaryconnectionbetweentheeventsthemselves:Forifeveryideabederivdfromanimpression,theideaofadeityproceedsfromthesameorigin;andifnoimpression,eitherofsensationorreflection,impliesanyforceorefficacy,tisequallyimpossibletodiscoverorevenimagineanysuchactiveprincipleinthedeity.Sincethesephilosophers[theoccasionalists],therefore,haveconcluded,thatmattercannotbeendowdwithanyefficaciousprinciple,becausetisimpossibletodiscoverinitsuchaprinciple;thesamecourseofreasoningshouddeterminethemtoexcludeitfromthesupremebeing.(T1.3.14.10)2323MalebranchealsoarguesthatGodistheonlytruecausebecauseHeisinfinitelypowerfulandthusthatwhateverhewillsmustresult.Godswillisnecessarilyconnectedtoeveryevent.Thereis,hesays,acontradictioninthethoughtthatGodwillsthatXoccursandyetXdoesnotoccur.HumerevealsthemuddlethatMalebrancheisin.Wehavenoideaofabeingendowdwithanypower,muchlessofoneendowdwithinfinitepower,andsoweareunabletousetheconceivabilitytestfornecessaryconnection.Ontheotherhand,ifwesimplydefinetheinfinitepowerofGodswillintermsoftheimpossibilitythatanythinghappenscontrarytothiswill,theclaimthatthedivinewillisthetruecauseofeverythingbecomestautological,forthenwereallydonomorethanassert,thatabeing,whosevolitionisconnectedwitheveryeffect,isconnectedwitheveryeffect(T1.4.5.31).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n164martinbellMalebrancheinsiststhatnothingisatruecauseunlessitspowerisknowableinsuchawaythatthenecessaryconnectionbetweencauseandeffectisperceivablebythemind.Humeshowsthatthereisnocaseinwhichthisrequirementissatisfied.Allcausesandeffectsareobjectsoreventsdistinctanddifferentfromeachother.Theyneverhavediscoverable,knowablepropertiesorqualitiesthatrelatethemtogethersothatconceptualconnectionsofthekindthatMale-brancherequirescanbeseen.ThisappealofMalebranchetotheideaofGodis,onempiricistprinciples,afailure.Consequently,Humeconcludes,iftruecausationrequiresknowablenecessaryconnec-tions,ifitrequiresideasthatrepresentimpressionsofcausalpowersinobjects,thenthereisnotruecausationatall.24ThewayoutHumeofferstheoccasionalistsisthecentralfeatureofhisownpositivethe-ory.25Becausethereisnoideaofpowerornecessaryconnectionasapropertyorqualityofobjects,allthattheexperienceofcausesandeffectsprovidesusisthoseregularitiesofcontiguityandsucces-sionthatHumecallsconstantconjunctionsandthatMalebrancheidentifiesascharacteristicofthebehaviorofwhathesupposesaresecondarycausesandeffects.WhatHumespositivetheorydoesismakeexperienceofconstantconjunction,togetherwiththeprinci-pleofcustomandhabit,sufficientinthecircumstancesofhumannaturetogiverisetotheideaofcausalityasanecessaryrelation.ItispreciselybecausetheideaofnecessaryconnectionthatHumeidentifiesisnotarepresentationofanypropertyorqualityofobjectsthatsecondarycausesandeffectscanbethoughtofasnecessarilyconnected.Whatisrevolutionaryinhistheoryisthathesolvestheproblembyturningfromtheobservationofobjectstotheeffectofthatobservationinthemind,andpresentstheideaofcauseandeffectasacomplexphenomenonthatincludesanideathatdoesnotrepresentanythingintheobjects:Theseveralinstancesofresemblingconjunctionsleadusintothenotionofpowerandnecessity.Theseinstancesareinthemselvestotallydistinctfromeachother,andhavenounionbutinthemind,whichobservesthem,andcollectstheirideas.Necessity,then,istheeffectofthisobservation,andisnothingbutaninternalimpressionofthemind,oradeterminationtocarryourthoughtsfromoneobjecttoanother.Withoutconsideringitin24SeeT1.4.5.31.25T1.3.14.10.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation165thisview,wecanneverarriveatthemostdistantnotionofit,orbeabletoattributeiteithertoexternalorinternalobjects,tospiritorbody,tocausesoreffects.(T1.3.14.20)Whenwesay,therefore,thatoneobjectisconnectedwithanother,wemeanonly,thattheyhaveacquiredaconnexioninourthought,andgiverisetothisinference,bywhichtheybecomeproofsofeachothersexistence(EHU7.28).Therearetwoaspectstothewayinwhich,onthisempiricisttheory,experienceproducestheideaofcauseandeffect.Ontheonehand,thereisexperienceoftheconstantandinvariableconjunctionsoftypesofobjectsorevents.Ontheotherhand,thereistheresultinghabitualassociationoftheideasofthesethingsandthefeelingofthecustomarytransitionofthoughtfromonetotheother.FromthesetwoperspectivesHumeformstwodefinitionsofcause.Ontheonehandacauseisanobject,followedbyanother,andwherealltheobjects,similartothefirst,arefollowedbyobjectssimilartothesec-ond.Ontheotherhandacauseisanobjectfollowedbyanother,andwhoseappearancealwaysconveysthethoughttothatother(EHU7.29).26Therearetwopointstonoteaboutthesedefinitions.First,theyaredefinitionsinHumessense,namely,identificationsoftheimpressionsfromwhichaparticularideaarises.Therelationsbetweenobjectsmentionedinthedefinitions,namely,succession,resemblance,andhabitualassociation,areallobservableinthesensethattheycanbeperceivedorfelt.Second,theyare,asHumelaterpointsout,asmuchhisproposeddefinitions(in,again,hissenseofidentifyingtheoriginofanideainimpressions)ofnecessityastheyaredefinitionsofcause.27Theconcludingsectionofthisessaydis-cussesthesedefinitionsandsomeoftheproblemsofinterpretationthatariseinseekingtounderstandHumestheoryofcausation.iii.thedefinitionsofcauseandtheirimplicationsThefirstdefinitionofcause(callthisC1)referstoconstantcon-junctionsofobjectsandevents.Acauseisanobject,followedbyanother,andwherealltheobjects,similartothefirst,arefollowed26SimilardefinitionsaregivenintheTreatiseat1.3.14.31.27EHU8.5.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n166martinbellbyobjectssimilartothesecond.Theseconddefinition(callthisC2)referstotheeffectontheimaginationofobservingconstantconjunctions.Acauseisanobjectfollowedbyanother,andwhoseappearancealwaysconveysthethoughttothatother(EHU7.29).Ithasbeenobjectedthatthetwodefinitionsareneitherlogicallynorextensionallyequivalent.28IfthereferenceinC1toalltheobjects,similartothefirstisreadasmeaningallthesimilarobjectsthathaveexisted,doexistandwillexist,thentherecouldbemanycauses(byC1)thatarenotcauses(byC2)becausenoonehasobservedanyorenoughresemblingcasesforideasoftheseobjectstobecomeassociatedintheimagination.Conversely,therecouldbecauses(byC2)thatarenotcauses(byC1).Someonemayhaveexperiencedsufficientresemblinginstancesoftheconjunctionoftwotypesofobjectsoreventsfortheideasoftheseobjectstobecomeassociatedintheimagination,althoughthecasesthathavebeenobserveddonotconstitutearepresentativesampleofthesetypesofobjectsorevents.Theseconsiderationsmightsuggestthatthetwodefinitionsshouldbereadasdefiningdifferentsensesofcause.C1mightbereadasanobjectivesense,expressingwhatitisforanobjectoreventtobeacauseofaneffectinreality,independentlyoftheexperienceandresultinginferentialtendenciesofobservers.C2mightthenbereadasasubjectivesense,expressingwhatitisforanobservertoassumeorbelievethatanobjectoreventisacauseofanother.Ifthedef-initionsaresounderstood,thereseemstobeafurtherimplicationabouttheideaofnecessity.Althoughneitherdefinitionexplicitlymentionsanecessaryconnectionbetweencausesandeffects,Humetracestheoriginofthatideatothefeelingofthedeterminationoftheimaginationinhabitualcausalinferences.ThatseemstomeanthatnecessityisproperlyspeakingafeatureofcausesonlyinthesenseofC2,andthatobjectivelyspeaking,inthesenseofC1,causesdonotnecessitatetheireffects.ThiswayofreadingHume,asholdingwhatiscalledaregularitytheoryofcausationintheobjectivesense,iswidespread.Forexample,ithasbeenclaimedthataboutcausa-tionasitreallyisintheobjectsHumeusuallysaysthatitisregular28Forexample,J.A.Robinson,HumesTwoDefinitionsofCause,PhilosophicalQuarterly12(1962):16271,reprintedinHume:ACollectionofCriticalEssays,ed.V.C.Chappell(GardenCity,NY:Doubleday,1966),12947.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation167successionandnothingmore.29Inthistradition,C1issupposedtobethecompletedefinitionofcausation,asitexistsobjectivelyinnature.30Furthermore,becauseHumesaccountoftheoriginoftheideaofnecessitytiesittothementalhabitsreferredtoonlyinthedefinitionofthesubjectivesenseofcause,itwouldseemthathemustruleoutasmeaninglesstalkofcausesasobjectivelyandnec-essarilyconnectedtotheireffects,orofobjectsasreallypossessingcausalpowers.TherearepassagesintheTreatisethatseemtosupporttheviewthatHumeregardstalkofobjectivecausalnecessityorcausalpowerasmeaninglessorconfused.31Inthesepassageshedescribeshisthe-sisthattheideaofcausalnecessityarisesfromourawarenessofthedeterminationofthemindincausalinferencesasaparadox.Buthedoesnotmeanthatthedoctrineinvolvessomekindofcon-tradiction.32Hemeans,rather,thatreaderswillbepronetothinkthathisviewdoesinvolveacontradiction,andwilllikelyrejectitasagrossabsurdity.Thereasonwhythisreactionistobeexpectedis,hesays,thatthemindhasagreatpropensitytospreaditselfonexternalobjects,andtoconjoinwiththemanyinternalimpressions,whichtheyoccasion(T1.3.14.256).Hereturnstothisnotion,thenotionthatthemindconjoinstheinternalimpression,thatis,thefeelingofdeterminationintheimagination,withtheimpres-sionsofexternalobjects,intheconclusionofBook1.Theviewthathistheoryisagrossabsurdityistheexpectedreactionofthosewhofailtorealizethatinthemostusualconjunctionsofcauseandeffectweareasignorantoftheultimateprinciple,whichbindsthemtogether,asinthemostunusualandextraordinary.Thepropensityofthemindaboutwhichheisspeakingis,consequently,thatwhichleadspeopletoimaginethattheinternalimpressionisinfactasenseimpression,animpressionofapropertyorqualityofanobject.Becausetheinternalimpressionistheoriginoftheideaofaneces-saryconnectionbetweenobjects,thosewhomanifestthepropensitymistakenlythinkthattheyperceive,andsoareacquaintedwith,29J.L.Mackie,TheCementoftheUniverse:AStudyofCausation(Oxford:Claren-donPress,1974),20.30GeorgesDicker,HumesEpistemologyandMetaphysics(London:Routledge,1998),115.31Forexample,T1.3.14.14,228.32ByparadoxHumemeansastrangeorunusualopinion.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n168martinbellthatenergyinthecause,bywhichitoperatesonitseffect;thattie,whichconnectsthemtogether;andthatefficaciousquality,onwhichthetiedepends.Itisthebeliefthatexperienceacquaintsuswithcausalpowersinobjectsthatisanillusionoftheimagina-tion.Thecorrectconclusiontodraw,Humethinks,isthatweareignorantofcausalpowersornecessaryconnectionsinobjects.Thereisadeficiencyinourideas,anabsenceoragap,which,becauseofthepropensityofthemindtoconjoininternalwithexternalimpres-sions,wecommonlyfailtoperceive(T1.4.7.56).ThethemeofthedeficiencyorinadequacyofourideasofcausalpowersandnecessaryconnectionsoccursalsoinwhatHumesaysaboutthetwodefinitionsofcause.IntheTreatisehetalksaboutthepossibilitythatthedefinitionsmaybejudgedtobedefective,becausedrawnfromobjectsforeigntothecause(T1.3.14.31).Itisnotimmediatelyclearwhatismeantbyobjectsforeigntothecause,norwhyreferencetothemmakesthedefinitionsdefective.InthecorrespondingpassageinthefirstEnquiry,however,Humemakesitclearerwhysomeonemightmaketheseobjections.Humetherespeaksoftheideaofcauseandnecessitythatweactuallyhave,theideathatthetwodefinitionsreferto,asimperfect.Theideaweactuallyhaveisnot,aswehaveseenatlength,anideathatrepresentsthatcircumstanceinthecause,whichgivesitaconnexionwithitseffect.Wehavenoideaofthisconnexion;norevenanydistinctnotionwhatitiswedesiretoknow,whenweendeavouratacon-ceptionofit(EHU7.29).Humedoesnotagreethathisdefinitionsaredefective.Inhisviewtheycapturetheideaweactuallyhaveofcauseandeffectandsodefinethetermcausecorrectlybyindicatingtheideaitexpresses.Theydonot,ofcourse,indicateanideawedonothave,anideaofpowerornecessaryconnectionasaqualityof,orrelationbetweenobjects.Comparedtothatwhichperhapswewouldwishtomean,theideaweactuallyhaveisimperfect,whilethedefinitionsofcausecanberegardedasdrawnfromobjectsforeigntothecausebecausetheyinvokeideasofentitiesthatarecausallyconnected,butnotanideaofsomethingthatdoestheconnecting.Inthesediscussionsofthemeaningofcausallanguage,Humepointstoatensionbetweenwhathearguesweactuallymean,theideasweactuallyhaveandexpress,andwhatwehaveastrongten-dencytothinkwemean.Wethinkwehaveideasthatinrealitywedonothave,andwethinkthatourwordsexpressthem.Inshort,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation169oureverydaythinkingaboutcausationisconfused.Thisconfusionseemstobeaprimecandidateforillustratingthemethodofclarify-ingideasbytracingthemtotheirorigininimpressions.ButHumeseemsconsciousthatthetensionsandconfusionshehasidentifiedcontinuetooperateevenaswetrytounderstandhistheory.AttheendofhisaccountinthefirstEnquiry,hewrites:Iknownot,whetherthereaderwillreadilyapprehendthisreasoning.Iamafraid,that,shouldImultiplywordsaboutit,orthrowitintoagreatervarietyoflights,itwouldonlybecomemoreobscureandintricate.Inallabstractreasonings,thereisonepointofview,which,ifwecanhappilyhit,weshallgofarthertowardsillustratingthesubject,thanbyalltheeloquenceandcopiousexpressionintheworld.Thispointofviewweshouldendeavourtoreach,andreservetheflowersofrhetoricforsubjectswhicharemoreadaptedtothem.(EHU7.30)ItisdifficulttobesurethatonehassucceededinreachingHumespointofview.Histheoryofcausationisstillthesubjectofcom-petinginterpretations.JohnWright,forexample,hasarguedthatHumehimselfregardedhisdefinitionsasimperfect.33AccordingtoWright,Humeheldthatcausationinrealityinvolvesrealphys-icalforce,andthatsuchforcesdonecessitatetheireffects.Humespurpose,heargues,istoshowthatwehavenoideaofsuchthings,noideaofthecausalpowersthatconstitutetheultimatereasonswhythingshappenthewaytheydo.Hepointsoutthat,aswehavealreadyseen,Humeclaimsthatifwehadideasofultimatecauseswewouldbeabletoinfereffectsfromcausesapriori.Thatwecannotmakesuchinferencesisbecauseourideasofcausesandeffectsarealwaysdistinctfromeachotherandseparable.Asaresult,Wrightsays,Humeregardsourideasasinadequate.Humespurpose,hesays,wasnottodenythattherearesuchultimatecausesonwhichdependtheregularitiesofnature,butonlytoshowthatwehavenoknowledgeofthem:Humeswholeargumentassumestheexistenceofforceornecessaryconnectioninthematerialuniverse.34Con-sequently,itmustbestressedthatHumesdefinitionsofcausality33JohnP.Wright,HumesCausalRealism:RecoveringaTraditionalInterpretationinTheNewHumeDebate,ed.R.ReadandK.A.Richman(London:Routledge,2000),8899.See,bythesameauthor,TheScepticalRealismofDavidHume(Manchester:ManchesterUniversityPress,1983),12386.34Wright,ScepticalRealism,147,144.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n170martinbellareclearlypresentedasdefinitionswhicharewhollyinadequatetowhattheypurporttodefine.35Wrightsreadingdiffers,therefore,fromthatgivenabove,accordingtowhichinadequateorimperfectappliestotheideaswehave,nottothedefinitions.Theideaswehave,Ihaveargued,areinadequateorimperfectcomparedtotheideawedonothave,thatofqualitiesoforrelationsbetweenobjectsthatconstitutetheultimateprinciplethatbindsacausetoitseffect(T1.4.7.6).Humecanrefertotheideawedonothave,theideaofanultimateprinciple,consistentlywithsayingthatallideasoriginateinimpressions,becauseheexplainstheideaofanultimatecauseintermsofhumancognitivepowers(intermsofwhatwecouldinferifwehadsuchanidea),asindeedWrighthasgranted.36ThusHumecangivecontenttothethoughtofwhatanideaofcausalpowerorforceornecessaryconnectionintheobjectswouldbe,althoughwelackit,andinthesetermscanspeakoftheideawedohaveasinadequate.Wright,however,saysthatthedefinitionspurporttodefineultimatecausesandthatHumerecognizesthattheydonotdoso.ThisispartofWrightsevidenceforhisinterpretationofHumeasascepticalrealist.HereadsHumeasholdingthattobelievethatAcausesB,properlyspeaking,istobelievethatthereexistssomefeatureofAsuchthatifwewereawareofitwewouldbeabletoinferBfromAapriori.ThebeliefthatsuchfeaturesexistisHumesrealismaboutcauses,whilehisdenialthatweeverhaveanyideaofsuchfeaturesishisskepticism.Thisentailsthat,forHume,ourexperienceofcausesandeffectsasdistinctanddifferentfromeachotherdoesnotrepresenttheworldasitreallyis.AsWrightsays,theconsequenceofthepositionheascribestoHumeisthat,inreality,causesandeffectsarenotdistinctanddifferentevents:[Custom]leadsustoascribeanabsolutenecessityrelatingthoseobjectswhichwecallcauseandeffect.Anditleadsustosupposethatthereisnoabsoluteexistentialdistinctionbetweenthem.3735Wright,HumesCausalRealism,91.36Humeusuallyspeaksofanultimatecauseassomethingsuchthat,wereweacquaintedwithit,wecouldinfertheeffectapriori.Hedoesnottreatthisepiste-mologicallydefinednotionasanideaofnecessaryconnectionbetweenobjects.FordiscussionofthispointseeKenLevy,Hume,theNewHume,andCausalCon-nections,HumeStudies26(2000):4175;PeterKail,ProjectionandNecessityinHume,EuropeanJournalofPhilosophy9(2001):2454.37Wright,ScepticalRealism,161.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation171Wrightsskepticalrealistreadingisoneofthosethathavebeendescribed,byKenWinkler,asNewHumereadings.38ThesearereadingsaccordingtowhichHumestheoryofcausationincludesthethesisthatthereexistcausalpowersthatnecessitatetheireffectseventhoughwecanhavenoknowledgeofthem.However,theserevisionistreadingsdifferamongthemselves.GalenStrawson,forexample,arguesthatHumesfrequentreferencestosecretpowersandprinciplesareevidencethathetakestheexistenceofsome-thinglikenaturalnecessityorcausalpowerforgrantednotonlyincommonlifebutalsoasaphilosopher.ButStrawsondoesnotthinkthatthisbeliefisabeliefincausalconnectionsofthekindWrightascribestoHume.Rather,Humesbeliefincausalpowerisdescribed,essentially,asadenialoftheregularitytheoryofobjec-tivecausation.Itisthebeliefthattheremustbeandissomethingaboutrealitygivenwhichitisorderedandregularinthewaythatitis.39Tobelievethattheremustbeanunderlyinggroundofnaturesregularity,implies,however,somekindofevidenceorargu-menttothiseffect,butnosuchargumenteverappearsinHumestexts.40ThedenialthatHumeheldtheregularitytheoryofcausationdoesnotentailascribingtohimcausalrealismineitherWrightsorStraw-sonssense.Forexample,bothAnnetteBaierandDonGarrettattacktheascriptiontoHumeoftheregularitytheory,andbotharguethatdefinitionC1ismisinterpretedwhenitisreadassupportingthisascription.41GarrettarguesthatC1andC2donotdefinedifferentsensesofcause,butaretwodifferentwaysofindicatingthesameidea.Therearegoodreasonsforthisinterpretation.AtEHU8.5,Humerefersinthesingulartoouridea...ofnecessityandcausa-tion,nottoideas.Thisideaarisesentirelyfromtheuniformity,observableintheoperationsofnature.Itisthisobservationofuni-formitythathastwoaspects,thetwocircumstancesthattakentogetherformthewholeofthatnecessity,whichweascribeto38KenWinkler,TheNewHume,PhilosophicalReview50(1991):54179,reprintedwithanewpostscriptinTheNewHumeDebate,5287.39GalenStrawson,TheSecretConnexion:Causation,RealismandDavidHume(Oxford:ClarendonPress,1989),12;seealso,bythesameauthor,DavidHume:ObjectsandPowers,inTheNewHumeDebate,3151.40Levy,Hume,theNewHume,andCausalConnections,42.41Baier,AProgressofSentiments,92:Garrett,CognitionandCommitment,111.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n172martinbellmatter.42ForHume,definitionsoftermsaregivenbyinvokingtheideasthatthetermsexpress,andinthiscasetheideaisofarela-tionbetweenobjectsorevents.IntheTreatisehesaysthathecangivetwodefinitions,notbecausetherearetwodifferentideas(asitwere,objectiveandsubjective),butbecausetherelationofcauseandeffectisbothaphilosophicalrelationandanaturalrelation(T1.3.14.31).Thetwodefinitionspresenttheideaoftherelationeitherasacomparisonoftwoideas(aphilosophicalrelation),orasanassociationbetwixtthem(anaturalrelation).TounderstandwhyHumegivestwodefinitionsthatseemnottobeequivalentitisnecessarytounderstandhowthedefinitionofthephilosophicalrelation,C1,andthedefinitionofthenaturalrelation,C2,providedifferentperspectivesonthesameobject.AsGarrettsays,Humesinitialdefinitionoftherelationofcauseandeffectisintermsofthethreecomponentrelationsofpriority,contiguity,andnecessaryconnection.43Itisonlybecausetheideaoftherelationofnecessaryconnectionisobscurethatthisisnotyetasatisfactorydefinition.Garrettexplainsthat,forHume,anabstractidea,suchasthatofcauseandeffect,isaparticularidea(inthiscaseaparticularideaofsomepairofobjectsorevents)associatedtoaterm(cause)whoseuseisconnectedtoacustomofcallingtomindotherparticularideas(inthiscaseofpairsofobjects)thatresembleeachotherinacertainrespect.WhenHumefirstdiscussesideasofphilosophicalrelationsintheTreatise,hesaysthattheyarisefromcomparison,andthatcomparisonrequiressomerespectinwhichthethingscomparedresembleeachother.Speakingoftherelationofcauseandeffect,hestatesthatthisisatoncebothaphilosophicalandanaturalrelation,andthencommentsthattheresemblanceimplydinthisrelation,shallbeexplaindafterwards(T1.1.5.9).WhatHumestheoryeventuallyshowsiswhatitisthatallthepairsofobjectsoreventsthatarebroughtundertheabstractideaofthecausalrelationhaveincommon:theyresembleeachotherinthatineachpairthefirstelementisrelatedtothesecondelementbypriority42Infact,HumearguesinthissectionofthefirstEnquirythatwefindthatthesametwocircumstancestakeplaceinthevoluntaryactionsofmen,andintheoperationsofthemind,andthuswecanascribecausalnecessitytohumanactioninthesamesensethatwecanascribeittomatter.43Garrett,CognitionandCommitment,99.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation173andcontiguity,whileallobjectsthatresemblethefirstelementstandinthesamerelationsofpriorityandcontiguitytoobjectsthatresem-blethesecondelement.Therefore,Garrettexplains,thepairsofobjectswhoseideasareincludedinthesetofideasassociatedwiththetermcauseandeffectarepairsofobjectsthatareperceivedasresemblingeachotherinthisrathercomplexway.ButinthatcaseonHumestheorytheideasoftheseobjectswillalsoberelatedbythenaturalrelationofcausation,fortheywillbehabituallyasso-ciatedintheimagination,andtherewillbeasettleddispositiontoinferonefromtheother.Theconnectionbetweencauseandeffectasaphilosophicalrelationbasedoncomparisonandasanaturalrelationbasedoncustomalsoworkstheotherwayround.Alltheideasofpairsofobjectsoreventsthatfallunderthenaturalrelationbybeingideasbetweenwhichtheimaginationisdisposedtomakehabitualtransitionswillalsobeideasofpairsofobjectsoreventsthatfallunderthephilosophicalrelation.TheconnectionbetweenthetwodefinitionsthatjustifiesHumesclaimthattheysimplypresentdifferentviewsofthesameobjectis,ofcourse,thatonekindofexperience(thatreferredtoinC1)causesanotherkindofexperi-ence(thatreferredtoinC2).AsAnnetteBaierhasshown,Humesdefinitionsarearrivedatbyacausalinvestigationintoourconceptofcause.44Thehabitualassociationofideasthatgivesrisetothefeelingofbeingdeterminedtoinferonefromtheotheristheeffectoftheexperienceoftheconstantconjunctionofobjectsthatcausesthemindtothinkoftheseobjectsasrelatedbythephilosophicalrelationofcauseandeffect.Eitherdefinitionis,asHumeclaims,adefinitionofcauseandnecessity.ThewayinwhichGarrettinterpretsthetwodefinitionsandtheirrelationenableshimtoofferaninterestingsolutiontotheproblemthatC1seemstobeanobjectivedefinitionwhileC2seemstobeasubjectiveone.HecomparesthetwodefinitionsofcausetothetwodefinitionsofvirtueHumegivesinthesecondEnquiry.Inthecaseofcausation,wehaveconstantconjunctionsofobjects,andafeelingofdetermination.Inthecaseofvirtue,wehavequalitiesofmindthatareusefuloragreeabletothepossessorortoothers,andafeelingofapprobation.Garrettarguesthat,inthecaseofthe44Baier,AProgressofSentiments,91.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n174martinbelldefinitionsofvirtue,itmakessensetodistinguishbetweenasub-jectivereading,inwhichthefeelingofapprobationisonethatarisesinsomeparticularempiricalobserver,andanobjectivereadinginwhichthefeelingarisesinanidealspectatorwhohasaproperlydevelopedmoralsenseandisfreefrombias,prejudice,orotherdis-tortinginfluences.Similarly,Garrettsuggests,wecouldreadbothdefinitionsofcauseashavingasubjectivesense,oneinwhichtheconstantconjunctionsarejustthosethathappentohaveoccurredintheexperienceofsomeparticularempiricalobserver,whiletheideasthatareassociatedare,similarly,justthoseassociatedinthatpersonsmind.Alternatively,wecouldreadbothdefinitionsashav-inganobjectivesense,oneinwhichtheconstantconjunctionsaresuchastoholdatalltimesandinallplaces,whilethehabitualtran-sitionsofthemindarethosethatwouldoccurinanidealobserverwhohasaproperlydevelopedcapacityformakingcausalinferencesfreedfromprejudice,bias,orotherdistortinginfluences.Inthiscon-textitisalsoimportanttorecognize,asBaierstresses,thatintheTreatise,aftergivingthedefinitionsofcause,Humeformulatesasetofrulesforjudgingwhatreallycauseswhat.ThatHumederivesnormativerulesfromthetheorythatledtothedefinitionsprovidessomesupportforGarrettsinterpretation.45TheproposalthatHumesaccountofcausationisbestinterpretedbyconsideringitssimilaritywithhisaccountoftheideasofviceandvirtueisamajorpartoftheso-calledquasi-realistreadinggivenbySimonBlackburn.BlackburnagreesthatHumedidnotholdtheregularitytheory.Hedidnotintendtoreducecausationtoregularity.But,Blackburnsays,itiswrongtothinkthattheonlyalternativeisrealismofthekindsascribedtoHumebyWrightandStrawson.ItispossiblethatHumewasarealistaboutcausalpowersinsofarashewasconvincedthatreferencetothemcouldnotbeeliminatedfromeverydaywaysofthinkingandspeakingaboutcausation.Butatthesametime,Humecouldhavebeenanantirealistaboutcausalpowersinsofarasheheldthatatheoreticalaccountofwhatwearedoinginthinkingandspeakingaswedodoesnotmeanthatwe45AsGarrettputsit,theserulesbecome,inpart,rulesformakingoneselfmorelikeanidealizedmind(CognitionandCommitment,112).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonCausation175arerespondingtothepresenceofcausalpowersasentitiesintheworld.46Thefinalsectionofthisessayhasshownthat,intheattempttounderstandHumestheoryofcausationandarriveatthatpointofviewthat,hehimselfconfessed,mightcontinuetoeludehisread-ers,therivalinterpretationsofcontemporaryscholarstestifytothecontinuinginterestandsignificanceofHumesideasoncausation.ItalsodemonstratesthatinterpretationsofHumeoncausationmustalsobeinterpretationsofHumesphilosophicalworkasawhole.suggestionsforfurtherreadingInadditiontotheworkscitedinthenotestothisessay,forfurtherreadingthefollowingarerecommended.OntheWillandBodilyandMentalActsHarris,James.HumesReconcilingProjectandTheCommonDistinctionbetwixtMoralandPhysicalNecessity.BritishJournalfortheHistoryofPhilosophy11(2003):45171.Pitson,Tony.Liberty,Necessity,andtheWill.InTheBlackwellGuidetoHumesTreatise,editedbyS.Traiger.Oxford:BlackwellPublishing,2006.21631.Norton,DavidFate.EditorsIntroduction.InDavidHume,ATreatiseofHumanNature,editedbyD.F.NortonandM.J.Norton.Oxford:OxfordUniversityPress,2000.I6572.OnRelations,ConceivabilityandPossibility,andDemonstrationCosta,Michael.HumeontheVeryIdeaofaRelation.HumeStudies24(1998):7194.Falkenstein,Lorne.SpaceandTime.InTheBlackwellGuidetoHumesTreatise,editedbyS.Traiger.Oxford:BlackwellPublishing,2006.679.Lightner,D.Tycerium.HumeonConceivabilityandInconceivability.HumeStudies23,no.1(1997):11332.Norton,DavidFate.EditorsIntroduction.InDavidHume,ATreatiseofHumanNature,1246.46SimonBlackburn,HumeandThickConnexions,PhilosophyandPhenomeno-logicalResearch50(supplement)(1990),23750,reprintedwithapostscriptinTheNewHumeDebate,10012.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n176martinbellOwen,David.HumesReason.Oxford:OxfordUniversityPress,1999.83112.OnProjectivistandRealistInterpretationsofHumeonCausalityBeebee,Helen.HumeonCausation.London:Routledge,2006.142225.Kail,Peter.ProjectionandRealisminHumesPhilosophy.Oxford:Claren-donPress,2007.75124.Millican,Peter.HumesOldandNew:FourFashionableFalsehoods,andOneUnfashionableTruth.ProceedingsoftheAristotelianSocietysup-plementaryvolume81(2007):16399.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\njanel.mcintyre6HumeandtheProblemofPersonalIdentityi.backgroundTheproblemofpersonalidentity,asphilosophersunderstandittoday,emergedfromthediscussionofidentitythatLockeaddedtothesecondeditionofTheEssayconcerningHumanUnderstanding,publishedin1694.Intheforty-fiveyearsbetweenthepublicationofthatworkandthepublicationoftheTreatise,theliteratureontheproblemofpersonalidentitymushroomed,promptingHumetoobservewryly:“Wenowproceedtoexplainthenatureofpersonalidentity,whichhasbecomesogreataquestioninphilosophy,espe-ciallyoflateyearsinEngland,wherealltheabstrusersciencesarestudy’dwithapeculiarardourandapplication”(T1.4.6.15).Hume’sownexplanationofthenatureofpersonalidentitydrewontheresourcesofhisaccountsoftheimaginationandthepassions,andwasthereforeuniqueinmanyrespects.Nevertheless,thedebatesoftheprecedingdecadeshadcoveredconsiderableground,andthedistinctivefeaturesofHume’sownviewemergemoreclearlywhenseeninthecontextofwhathadcomebefore.LockeaddedthechapterOfIdentityandDiversitytothesecondeditionofhisEssaytoaddressissuesaboutpersonalidentitythatarosefromhisargumentsagainsttheCartesianviewthatthesoulalwaysthinks.1Locke’sbasicassumptionwasthatquestionsaboutidentityhadtobeunderstoodinrelationtothekindofobjectunderconsideration:oneandthesameobjectcouldinstantiatedifferentkinds,eachofwhichwouldhavedifferentconditionsofidentity.1JohnLocke,AnEssayconcerningHumanUnderstanding,ed.P.H.Nidditch(Oxford:ClarendonPress,1975).Referencestothisworkwillcitethebook,chapter,andsectionnumbersinthisedition.177CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n178janel.mcintyreHearguedthatatree,consideredasanorganizedlivingbody,couldbedistinguishedfromagivenmassofparticles.Itmaybecorrecttosay,Lockeheld,thatthetreeremainsthesametreethroughtimeeventhoughallofitsparticlesmayhavegraduallychanged:itmaybethesametree,butitisnotthesamemassofmatter.ForLocke,therefore,thecrucialquestionaboutpersonalidentitywasthequestionofwhatkindofthingpersonsare:hisansweroutragedsomeofhiscontemporaries.Lockearguedthatthetermpersondoesnotstandforthesameconceptasimmaterialsubstance,orsoul,orspirit.Lockealsodistinguishedtheconceptofapersonfromthatofaman,whichheusuallyunderstoodtomean“nothingelsebut...anAnimalofsuchacertainForm.”2AccordingtoLocke,apersonis“athinkingintelligentBeing,thathasreasonandreflection,andcanconsideritselfasitself,thesamethinkingthingindifferenttimesandplaces;whichitdoesonlybythatconsciousness,whichisinseparablefromthinking,andasitseemstomeessentialtoit:Itbeingimpossibleforanyonetoperceive,withoutperceiving,thathedoesperceive.”3And,forLocke,theidentityofpersonsdependsonlyonconsciousness.Heargued:ForitisbytheconsciousnessithasofitspresentThoughtsandActions,thatitisselftoitselfnow,andsowillbethesameselfasfarasthesameconsciousnesscanextendtoActionspastortocome;andwouldbebyadistanceofTime,orchangeofSubstance,nomoretwoPersonsthanaManbetwoMen,bywearingotherCloathstoDaythanhedidYesterday,withalongorshortsleepbetween:ThesameconsciousnessunitingthosedistantActionsintothesamePerson,whateverSubstancescontributedtotheirProduction.4Lockeexplicitlyarguedthatpersonalidentitycouldpersistthroughachangeofsubstanceandthatthesamenessofanunderlyingsub-stancedidnotguaranteethatpersonalidentitywaspreserved.5Sub-stantialidentity,therefore,wasneithernecessarynorsufficientforpersonalidentity.ForLocke,consciousnessaloneconstitutespersonalidentity.Locke’stheoryofpersonalidentitythusbrokenotonlywiththeCartesianviewthattheselfisanimmaterial2Locke,Essay2.27.4,8.3Locke,Essay,2.27.9.4Locke,Essay,2.27.10.5Locke,Essay,2.27.13–14.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity179substance,butalsowithanyviewthattookpersonalidentitytobe,orbedependenton,theidentityofthehumanbody.Lockedescribedtheconceptofapersonasa“Forensick”term,thatis,asonetiedtoascriptionsofmoralandlegalresponsibility.Consciousnessalone,Lockeargued,makesusconcernedandac-countableforourpastactions:“ForsupposingaManpunish’dnow,forwhathehaddoneinanotherLife,whereofhecouldbemadetohavenoconsciousnessatall,whatdifferenceistherebetweenthatPunishment,andbeingcreatedmiserable?”AlthoughLockehimselfallowedthat“themoreprobableOpinionis,thatthisconsciousnessisannexedto,andtheAffectionofoneindividualimmaterialSubstance,”hedidnotholdthattherewasanythingaboutthoughtorconsciousnessitselfthatrequiredthis.Lockearguedthatitis“notmuchmoreremotefromourComprehensiontoconceive,thatGODcan,ifhepleases,superaddtoMatteraFacultyofthinking,thanthatheshouldsuperaddtoitanotherSubstance,withaFacultyofThinking.”6Lockewasnotunawareoftheimplicationsthatmightbedrawnfromhisview,andaddedthatAllthegreatEndsofMoralityandReligion,arewellenoughsecured,withoutphilosophicalProofsoftheSoul’sImmateriality;sinceitisevident,thathewhomadeusatfirstbegintosubsisthere,sensibleintelligentBeings,andforseveralyearscontinuedusinsuchastate,canandwillrestoreustothelikestateofSensibilityinanotherWorld,andmakeuscapabletheretoreceivetheRetributionhehasdesignedtoMen,accordingtotheirdoingsinthisLife.7Wecansee,then,thatLocke’spositiononpersonalidentityinvolvedthreecontroversialclaims:first,thatpersonalidentityisconstitutedbyconsciousness;second,thattheidentityofanunderlyingsub-stance(whethermaterialorimmaterial)isneithernecessarynorsufficientfortheexistenceofthesameconsciousness;andthird,thatasaconsequencetheselfisnotnecessarilyanimmaterialsubstance.Inthefirsthalfoftheeighteenthcentury,philosophicaldebateabouttheseclaimswasintense,andLocke’snovelaccountof6Locke,Essay,2.27.25–6,4.3.6.7Locke,Essay,4.3.6.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n180janel.mcintyrepersonalidentitywonfewfollowers.Asearlyas1702HenryLeeattackedLocke’scentralthesisthatconsciousnessconstitutesper-sonalidentity.Leearguedthatapersoncouldbe“decay’dinhisMemoryandIntellectuals”withoutlosinghisidentity.Heappealedprimarilytowhathetooktobeourcommonnotionofaperson,whichhethoughtLockefailedtoadhereto,buthealsoheldthatLocke’spositioncouldnotgiveanadequateaccountofcommonlyacceptedmoraljudgments.AccordingtoLee,whenaperson’smem-oryandintellectfails,wenonethelesshaveanobligationtotreatthatindividualasthepersonheorshewasbefore.8ThenatureoftheselfwasalsocentraltoShaftesbury’swidelyreaddialogueTheMoralists,publishedin1709andregularlyreprintedinhispopu-larCharacteristics,athree-volumeworkfirstpublishedin1711.Shaftesburymaintainedthattheidentityoftheselfpersists“whenneitheroneatomofbody,onepassion,noronethoughtremainsthesame.”9PerhapsthemostinfluentialcriticoftheLockeanaccountofper-sonalidentity,however,wastheNewtonianphilosopherandtheolo-gianSamuelClarke.ClarkehaddefendedtheviewthatthesoulisnecessarilyimmaterialandimmortalinhisBoylelecturesof1704and1705.AlivelypublishedexchangewithAnthonyCollins,apro-ponentofLocke’sposition,extendedoverthenextseveralyearsandexploredindetailtheproblemsassociatedwithaLockeanaccountofpersonalidentity.10Clarkeattackedtheviewthatconsciousnesscouldexistinany-thingbutasimpleimmaterialsubstance.Hisargumentsfocusedonwhathetooktobetheessentialfeaturesofmatterand8HenryLee,Anti-scepticism(London,1702;fac.ed.,NewYork:GarlandPublishing,1984),124–30.9AnthonyAshleyCooper,ThirdEarlofShaftesbury,CharacteristicsofMen,Man-ners,Opinions,Times,ed.LawrenceE.Klein(Cambridge:CambridgeUniversityPress,1999),301.Humepurchasedacopyofthisworkin1726andreferstoTheMoralistsinthediscussionofpersonalidentityinafootnotetoT1.4.6.6.FordetailsconcerningHume’spurchase,seeDavidFateNortonandMaryJ.Norton,TheDavidHumeLibrary(Edinburgh:EdinburghBibliographicalSociety,1996),16.10ThepublishedexchangeoflettersbetweenClarkeandCollinsisreprintedinitsentiretyinSamuelClarke,TheWorksofSamuelClarke,4vols.(London,1739;fac.ed.,NewYork:GarlandPublishing,1978).CitationstobothClarkeandCollinswillbetoClarke’sWorks.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity181consciousness–thatmatterisdivisible,ormadeupofparts,andthatconsciousnessitselfhasaunitythatisincompatiblewithbeingmadeupofparts.Ifhundredsofparticlesofmattersetmilesfromeachothercouldneverconstituteone“individualConsciousBeing,”Clarkeargued,thenbringingtheseparticlestogetherintoasinglesystem,howeverorganized,couldnevermakethemintoone.For,ifthepartsthemselveswereconscious,thesystemtheycreatedwouldbenomorethanacollectionofdistinctconsciousbeings.Andifthepartswerenotessentiallyconscious,Clarkeheld,itwouldbeimpos-siblethatconsciousnesscouldresultfromjoiningthemtogether.11CollinsrepliedthatClarkehadbeggedthequestion:systemsofmattercould,anddid,haveindividualpowersthattheirpartsdidnotpossessseparately.Consciousnesscouldthereforebeaqualityofacompoundmaterialsubstancesuchasthebrain.12Inresponse,Clarkeraisedafurtherproblem:“TheSpiritsandParticlesoftheBrain,”heargued,“beinglooseandinperpetualFlux”couldnotbetheseatofconsciousnessofpastactions“bywhichaMannotonlyremembersthingsdonemanyYearssince:butalsoisConsciousthatHehimself,thesameIndividualConsciousBeing,wasthedoerofthem.”13AccordingtoClarke,theexperienceofbeingconsciousthatonedidsomethinginthepastdependsontheexistenceofasimple,unchangingsubstance.Collinscounteredthatinfact“wearenotconscious,thatwecontinueaMomentthesameindivid-ualnumericalBeing.”Rather,the“continualintermediaterepeat-ing”ofideasbeforetheyarelostpreservesmemoriesofthepastintheever-changingbrain.14Notsurprisingly,Clarkemaintainedthatcontinuityofconsciousnessofthiskindwouldbe“mereDeceitandDelusion,”destructiveofmoralagency.Personswouldhavenorealconnectiontothepast,andthereforenogroundsfortakingrespon-sibilityforpastactions:norwouldtheyhaveabasisforthinkingabouttheconsequencesoftheiractionsinthefuture,sincetheythemselveswillnotexisttosufferthem.Similarly,ClarkearguedthatontheLocke-Collinsviewthejusticeofrewardandpunish-mentisundermined,sincethepersonrewardedorpunishedwould11Clarke,Works,III:730.12Collins,inClarke,Works,III:751–2.13Clarke,Works,III,:798.14Collins,inClarke,Works,III:819–20,870.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n182janel.mcintyrenotbethesameastheonewhocommittedtheaction.15Collins,likeLocke,arguedthatcontinuityofconsciousnessalonemakesusaccountableforpastactions,butheaddedthatthedistinctbeingsthatareunitedbyconsciousnessintoasingleselfarealsoboundtoeachotherby“SympathyandConcern.”16ContemporarycognitivescientistsmayfindCollins’sargumentstobesuggestive,butinthephilosophicaldiscussionofpersonaliden-titypriortotheTreatiseitwasClarke’sview,notthoseofLockeorCollins,thattookonthestatusofthereceivedopinion.Neverthe-less,thedebatebetweenClarkeandCollinshadprovidedtheirread-erswithacomprehensivereviewofquestionsfacedbyanyaccountofthenatureoftheself,orpersonalidentity.JosephButler’s“Disser-tationonPersonalIdentity”(1736),oneofthebest-knowncritiquesofLocke’saccount,citedClarkeanddeployedmanyofthesamearguments.Afewyearslater,manyofthepositionsCollinshaddefendedfoundacongenialhomeinHume’sTreatise.ii.treatise1.4.5andhume’saccountoftheselfHume’sextendedaccountofthenatureoftheselfbeginsinTrea-tise1.4.5,“Oftheimmaterialityofthesoul.”Asthetitleindicates,thissectionanalyzesoneofthecentralquestionsdebatedbyClarkeandCollins.Ofcourse,evenbeforetheyreached1.4.5,readersoftheTreatisecouldnotbeinmuchdoubtaboutthebroadoutlineofHume’sposition.Treatise1.4.2,Ofscepticismwithregardtothesenses,hadprovidedaclearprecisofwhatwastofollow:“whatwe´callamind,isnothingbutaheaporcollectionofdifferentpercep-tions,unitedtogetherbycertainrelations,andsuppos’d,tho’falsly,tobeendow’dwithaperfectsimplicityandidentity”(T1.4.2.39).Hume’sopeningargumentinTreatise1.4.5continuesthecritiqueoftheconceptofsubstanceinitiatedinearliersectionsofTrea-tise1.4.Everysimpleideamustbederivedfromaprecedentsim-pleimpression(T1.1.1.7).However,the“curiousreasoners”whodebatewhetherourperceptionsinhereinamaterialorimmate-rialsubstancecannotpointtoanimpressionthatproducestheideaofsubstance–aprimafacieproofthatthetermsubstancelacks15Clarke,Works,III:845,851.16Collins,inClarke,Works,III:877.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity183significance.Appealstothetraditionaldefinitionofsubstanceas“somethingwhichmayexistbyitself”providenohelp,sinceeveryperceptionisdistinguishableandseparablefromeveryother,andthereforesatisfiesthisdefinition.Humeconcludesthattheques-tion“Whetherperceptionsinhereinamaterialorimmaterialsub-stance”isdevoidofmeaning(T1.4.5.2,5–6).Somewhatsurprisingly,Humeisnonethelessinterestedinpro-vidingamoredetailedanalysisofoneoftheargumentspresentedbythedefendersoftheimmaterialityofthesoul:theargumentchampi-onedbyClarkethatthought,orperceptions,cannotexistinabeingmadeupofparts.Hume’sstrategyhererequiressomecomment.AsHumeunderstoodit,becausethemetaphysicalconceptofsubstanceismeaningless,therecouldbenoseriousphilosophicaldebateoverthequestionofwhetherthesubstanceofthesoulismaterialorimmaterial.Whythenshouldhecontinuetoparticipateinadis-putehebelievedtobemeaningless?PartoftheanswermaybethatHumecouldnotresistthetemptationtoofferacleverargumentagainstaformidableopponent,eventhoughitwassomewhatsuper-fluous.Hume’saudaciousclaimthatdefendersoftheimmaterialityofthesoulholdatheoryindistinguishablefromthatofSpinoza(seeT1.4.5.17–25),whomtheyreviled,certainlysuggeststhis;butother,philosophicallydeeper,reasonsforpursuingthisstrategywillemergeinthediscussionbelow.Humebeginshisnewanalysisoftheissuewithareviewoftheargumentthatthoughtandextensionare“qualitieswhollyincom-patible,andnevercanincorporatetogetherintoonesubject”(T1.4.5.7).Inthelightofhispreviousconclusionaboutthemeaning-lessnessoftheconceptsofsubstanceandinhesion,Humemaintainsthatthequestionatissueismoreaccuratelywhetherthoughtandmatteraresusceptibleoflocalconjunction(of,thatis,conjunctioninspace–dothoughtsorsensationshavelocations?),ratherthanabouttheunderlyingsubstanceofthesoul.Humearguesthatsuchperceptionsaspassions,desires,moralreflections,smells,sounds,andtastescanexist“andyetbenowhere”(T1.4.5.10).Thesepercep-tionshavenosizeorfigure,andthuscannotbelocatedinspace.Noraretheylikemathematicalpointsthatcanbemanipulatedtocreateafigurewithspatialdimensions.Perceptionsofthiskindcannothaveanyspatialrelationtomatter.Theycannotbepartofmatter,andinthatsensecannotliterallyinhereinmatter:“inthisviewofthings,”CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n184janel.mcintyreHumeobserved,“wecannotrefusetocondemnthematerialists,whoconjoinallthoughtwithextension.”But,hecontinued“alittlereflectionwillshowusequalreasonforblamingtheirantagonists,whoconjoinallthoughtwithasimpleandindivisiblesubstance.”Therearesomeperceptions,namely,thosederivedfromsightandtouch,thatareextended.Theperceptionofatable,forexample,haspartsthatarearrangedtoformafigurehavingallthefeaturesofextension.“Tocutshortalldisputes,theveryideaofextensioniscopy’dfromnothingbutanimpression,andconsequentlymustper-fectlyagreetoit.Tosaytheideaofextensionagreestoanything,istosayitisextended”(T1.4.5.15).Sincethereareextendedaswellasunextendedperceptions,thosewhoarguethatperceptionsinhereinanimmaterialsubstancefacethesameproblemasthematerial-ists.Justasthematerialistcannotexplainhowperceptionsthatlackextensioncanexistinanextendedsubstance,sotheimmaterialistcannotexplainhowextendedperceptionsexistinimmaterialsub-stance.Neitherview,accordingtoHume,canbecorrect.Defendersofmaterialismanddefendersoftheimmaterialityofthesoulbothrunafoulofthediversenatureofperceptions.Thisshedsfurtherlightonwhy,havingconcludedthatthedebateoverthesubstanceofthesoulwasmeaningless,Humedidnotjustlettheargumentdropandmoveon.Hisanalysisisintendedtoshowthateveniftheconceptofsubstancewerenotmeaningless,accountsofthemindthatattempttogroundperceptionsinanunderlyingsubstancewouldfail.Thefactoftheirreduciblediversityofourperceptionsisthestart-ingpointofHume’sownaccountoftheself.Themind,accordingtoHume,isinherentlycomplexandcomposite.HumethereforegoesbeyondLocke’scontroversialconclusionthatthemindisnotnec-essarilyanimmaterialsubstance,tothestrongerpropositionthatitcouldnotbeanimmaterialsubstance.And,whileHumeiscriticalofbothmaterialismandimmaterialism,onthiscentralpointhisownaccountofthemindissimilartothatofCollins,andfundamentallyatoddswiththeologianslikeClarke.OftheimmaterialityofthesoulconcludeswithadiscussionofanotherissuedebatedbyClarkeandCollins,whethermatterandmotioncouldbethecauseofthought.Thisquestion,Humeargues,canberesolvedbyappealtohisownaccountofthenatureofcausalityinTreatise1.3.Althoughthereisnoapparentconnectionbetweenmotionandthought,thesameistrueforallcausalrelations.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity185Itisonlythroughtheexperienceoftheirconstantconjunctionthatweconcludethattwoobjectsarerelatedascauseandeffect:“wefindbycomparingtheirideas,thatthoughtandmotionaredifferentfromeachother,andbyexperience,thattheyareconstantlyunited;whichbeingallthecircumstances,thatenterintotheideaofcauseandeffect,whenapply’dtotheoperationsofmatter,wemaycer-tainlyconclude,thatmotionmaybe,andactuallyis,thecauseofthoughtandperception”(T1.4.5.29–30).Humeacknowledgedthatthisargument“givestheadvantagetothematerialistsabovetheirantagonists.”But,whilehedoesnotseemreluctanttoacknowledgethisassociation,hispreviousgroundsforrejectingmaterialismstillstand:hereiteratesthat“allourperceptionsarenotsusceptibleofalocalunion,eitherwithwhatisextendedorunextended;therebeingsomeofthemoftheonekind,andsomeoftheother”(T1.4.5.32–3).AsHumehasjustargued,however,thediversenatureofpercep-tionsisnoobstacletotheirstandingincausalrelationstoeachother.ThusTreatise1.4.5endshavingestablishedthreecentralele-mentsofHume’saccountofthemind:thereisnosubstanceinwhichperceptionsinhere;theselfiscomprisedofperceptionsofdifferenttypes;and,inspiteoftheirdiversenatures,theseperceptionsmaybelinkedtogetherbytherelationofcauseandeffect.Thesemetaphysi-calconclusionsprovidethefoundationforthediscussionofpersonalidentitythatimmediatelyfollows.iii.treatise1.4.6:personalidentityasitregardstheimaginationOftheimmaterialityofthesouladdressedthosephilosopherswhoclaim“thatwehaveanideaofthesubstanceofourminds”(T1.4.5.4),withoutdirectdiscussionoftheirviewsconcerningtheexis-tenceofthemindovertime.TheopeningsentenceofTreatise1.4.6signalsanexpandedfocus.Thereare,Humegrants,“somephiloso-pherswhoimagineweareeverymomentintimatelyconsciousofwhatwecallourself;thatwefeelitsexistenceanditscontinuanceinexistence;andarecertain,beyondtheevidenceofademonstra-tion,bothofitsperfectidentityandsimplicity.”But,Humeargues,wehavenoideaoftheself“afterthemanneritishereexplain’d,”thatis,assomethingsimpleandidenticalthroughtime.Aswehaveseen,everysimpleideamustbederivedfromasimpleimpression.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n186janel.mcintyreIfanyimpressionweretogiverisetotheideaoftheselfasabeingthatexistsovertime,“thatimpressionmustcontinueinvariablythesame,thro’thewholecourseofourlives;sinceselfissuppos’dtoexistafterthatmanner.”ButHumefindsnosuchimpressioninhimself,anddoubtsthatanyone,after“seriousandunprejudic’dreflection,”willbeabletodosoeither.17“Painandpleasure,griefandjoy,passionsandsensationssucceedeachother,andneverallexistatthesametime.Itcannot,therefore,befromanyoftheseimpressions,orfromanyother,thattheideaofselfisderiv’d;andconsequentlythereisnosuchidea.”18Themindis“nothingbutabundleorcollectionofdifferentperceptions,whichsucceedeachotherwithaninconceivablerapidity,andareinaperpetualfluxandmovement....Thereisproperlynosimplicityinitatonetime,noridentityindifferent;whatevernaturalpropensionwemayhavetoimaginethatsimplicityandidentity”(T1.4.6.1–4).19Thislastsen-tenceindicateshowHumewillbringtheprincipleshehasdevelopedthroughoutBook1oftheTreatisetobearontheproblemofpersonalidentity.Humeisnotuninterestedinthemetaphysicalquestionofthenatureoftheself:hehadalreadyarguedforthecomplexityofthemind,andwoulddevelophisaccounttoexplaintherelationsthatconnecttheselfovertime.ButinOfpersonalidentityHume’sfocusisonexplainingourtendencytobelieveintheperfectidentityoftheselfovertime,giventhatwehavenoimpressionorideaonwhichtogroundthatbelief.ThedistinctionHumeintroducesatthisjuncture,betweenper-sonalidentity“asitregardsourthoughtorimagination”andper-sonalidentity“asitregardsourpassionsortheconcernwetakeinourselves,”reflectsthewayinwhichhehadreconceptualizedtheproblemofpersonalidentity.Atheoryofpersonalidentityasit17EarliercommentsintheTreatisereflectthisviewaswell.Humearguedthatthevulgar“havenothinglikewhatwecancallabeliefoftheeternaldurationoftheirsouls.”Humealsoarguedthat“incommonlife’tisevidenttheseideasofselfandpersonareneververyfix’dnordeterminate”(T1.3.9.13,1.4.2.6).18Asthepassagequotedindicates,Humehereisdenyingthatwehaveanideaofasimpleself.Wecan,anddo,haveacomplexideaoftheself,oftheselfasacollectionofperceptions.19WhenintroducingthephilosophicalrelationofidentityHumesaid:“ThisrelationIhereconsiderasapply’dinitsstrictestsensetoconstantandunchangeableobjects;withoutexaminingthenatureandfoundationofpersonalidentity,whichshallfinditsplaceafterwards”(T1.1.5.4).Ofcourse,Humealsosaidthatthemindisfalselysupposed“tobeendow’dwithaperfectsimplicityandidentity”(T1.4.2.39).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity187regardstheimaginationshouldexplainourtendencytosuppose–incorrectly,accordingtoHume–thatweare“possestofaninvari-ableanduninterruptedexistencethro’thewholecourseofourlives”(T1.4.6.5).Humethusrecaststheunderlyingphilosophicalprob-lemofpersonalidentityasoneaboutthecausesofaparticularandmistakenbelief.GiventhestructureoftheTreatise,HumelimitshimselfinBook1tothediscussionofthisaspectofpersonaliden-tity.Atheoryofpersonalidentityasitregardsourpassionsfocusesnotonourbelief,butonourconcernforourselves.Aswesawear-lier,Clarke(andothers)hadraisedatroublingobjectiontoLocke’saccountofpersonalidentity,onethatwouldapplytoanytheoryonwhichtheselfisacompoundentity.How,Clarkeasked,couldaselfthatwasnotstrictlyidenticalthroughtimebearesponsiblemoralagent?Whatbasiscouldtherebeforconcernwithpastacts,orcon-sequencesforthefuture?Humeacknowledgedthatthisisapartoftheproblemofpersonalidentity,buthedoesnotin1.4.6addresstheproblematanylength.Hume’sapproachtotheproblemofpersonalidentitycanthere-forebesummedupinthefollowingway:theargumentsofTrea-tise1.4.5showthatthemindisacollectionofperceptionswithoutanunderlyingsubstance.Whenweconsidertheexistenceofthiskindofentityovertime,thereisnosingleperceptionthatcontinuesunchangedthroughoutaperson’slife.Nevertheless,wetendtothinkofourselvesasthesameperson,andtofeelconcernedwithdistantportionsofourlives.Theproblemofpersonalidentity,asHumeunderstoodit,istoaccountforthesebeliefsandfeelingsaboutour-selves.Thereis,first,personalidentity“asitregardsourthoughtorimagination.”DiscussionofthedetailsofthispartofHume’sthe-oryofpersonalidentitycontinuesbelow.Andthenthereispersonalidentity“asitregardsthepassions,”thesubjectofthenextsectionofthepresentdiscussion.Humebeginswithhisfirsttask–with,thatis,ageneralaccountoftheideaofidentityasthatofanobjectthatremains“invari-ableanduninterrupted”overtime.Whenwethinkcarefullyandaccurately,werecognizethatanysuccessionofobjectsexhibitingvariation,evenwhentheyarecloselyrelated,isnotanexampleofidentity.However,theexperiencethemindhaswhenitconsidersanunchangingobjectisverysimilartotheexperienceithaswhenitconsidersacloselyconnectedseriesofobjects.Resemblanceisoneoftheprinciplesbywhichperceptionsareassociated,andthusCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n188janel.mcintyrethesetwoexperiencesbecomeassociatedwitheachother.Theideaofidentityconsequentlybecomeslinkedintheimaginationwiththeexperienceofasuccessionofrelatedobjects,andwetendtoattributeidentitytotherelatedsequences.Thisisamistake,butonewecanunderstandinlightoftheprinciplesofassociation.Wefallintothiserrorsoeasilyandreadily,that“tojustifytoourselvesthisabsurdity,weoftenfeignsomenewandunintelligibleprinciple,thatconnectstheobjectstogether,”orweare“apttoimaginesome-thingunknownandmysterious,connectingtheparts,besidetheirrelation”(T1.4.6.6).Humeanalyzesthesituationsinwhichweattributeidentitytophysicalobjects,plants,andanimalstosupportthisgeneralaccount.Aphysicalbody–amassofmatterwithcontiguousandconnectedparts–remainsstrictlythesameovertimeonlyifallitspartsremainunchanged.Butwhensmallchangesaremadegradually,wescarcelynoticethem,andthemind“feelsaneasypassage”asitsurveystheobjectfromonemomenttothenext.Wecontinuetoascribeiden-titytotheobjectbecauseofthissmoothtransitionofthought.Butstrictlyspeaking,asmallchangedestroystheidentityoftheobjectjustasmuchasalargechangewould.Theonlydifferenceisthatalargechangedoesnotusuallygounnoticed.Evenaconsiderabledegreeofvariationcanbecompatiblewithacceptingthecontin-uedidentityofanobject,ifthereissomething,acommonendorafunctionalorganization,thatfacilitatesthetransitionoftheimag-inationwhenitconsiderstheobjectatdifferenttimes.Inanimalsandplants,thepartsarealsocausallyrelatedtoeachother,makingitmorelikelythattheobjectwillactontheimaginationasifitwerestrictlyidenticalthroughtime.Totheimagination,ashipremainsthesameshipaftermajorrepairs,anoakisthesameasitgrowsfromasmallplanttoamaturetree,andan“infantbecomesaman,andissometimesfat,sometimeslean,withoutanychangeinhisidentity”(T1.4.6.8–12).Personalidentity“asitregardsourthoughtorimagination”isnodifferent:“Theidentity,whichweascribetothemindofman,isonlyafictitiousone,andofalikekindwiththatwhichweascribetovegetablesandanimalbodies.Itcannot,therefore,haveadifferentorigin,butmustproceedfromalikeoperationoftheimaginationuponlikeobjects.”Themindatanyonetimeisacol-lectionofperceptions:overtime,itisaseriesofsuchcollections,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity189interconnectedbutnotidentical.Theinterconnectionsmakepossi-blean“uninterruptedprogress”ofthoughtasapresentselfreflectsonthesuccessiveexperiencesthathaveprecededitandleduptoit.Thisfeelingofeasytransitioncreates“ournotionsofpersonalidentity”(T1.4.6.15–17).Ofthethreerelationsthatassociateideas,onlytwo,resemblanceandcausation,havearoleincreatingthefictionoftheself’siden-tityovertime.Memoryisinvolvedinbothoftheseassociations.Withinthefluxandvariationofcontinuingexperience,mostpeopleretainmemoriesofpastthoughtsandexperiences.Sincememoriesresembletheperceptionsthatcausedthem,theexistenceofmemo-riescreatesnumerouslinksofresemblancewithinthesuccessionofperceptionsthatconstitutesamindovertime.Inthisrespect,Humeargues,memoryproducespersonalidentity“byproducingtherela-tionofresemblanceamongtheperceptions”(T1.4.6.18).Memory,however,isonlyoneofmanycausalrelationsthatcon-necttheperceptionsofthemind:impressionsgiverisetoideas,whichmayinturnproduceimpressionsofreflection;beliefsinflu-encethepassions,andpassionsfigureamongthesourcesofotherbeliefs(T1.1.2.1,1.3.10.3–4).ThecenterpieceofBook2oftheTrea-tise,theaccountoftheindirectpassions,isbuiltonadoubleasso-ciationofimpressionsandideasthatinvolvesbothcausationandresemblance(T2.1.5).ForHume,“thetrueideaofthehumanmind,istoconsideritasasystemofdifferentperceptionsordifferentexis-tences,whicharelink’dtogetherbytherelationofcauseandeffect,andmutuallyproduce,destroy,influence,andmodifyeachother.”Memoryisnecessaryifoneistoexperiencetheideaofcausation.Similarly,onecannothavetheideaoftheself,aseriesofcausallyrelatedperceptions,withoutmemory.However,oncewehavethisidea,weextendtheideaofouridentitytoourentirecausalhistory,whetherrememberedornot.“Inthisview,therefore,memorydoesnotsomuchproduceasdiscoverpersonalidentity,byshowingustherelationofcauseandeffectamongourdifferentperceptions”(T1.4.6.19–20).Hume’sdiscussionoftheroleofmemoryinpersonalidentityismotivatedbyhisdesiretoavoidoneoftheproblemsassoci-atedwithLocke’searlieraccount.Lockehadarguedthatpersonalidentitydependsonlyoncontinuityofconsciousness.Onhisview,consciousnessofpastthoughtsandactionsmakesapresentselfCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n190janel.mcintyreidenticalwiththepersonwhohadthosethoughtsandperformedthoseactions.Thisaccountwascriticizedonthegroundsthat,inmanyordinaryjudgments,includingsituationswhereweassignmoralresponsibilityorevaluatemoralobligations,personaliden-tityisassumedtobemoreextensivethanthisaccountsuggests.Indeed,intheeyesofthecritics,personalidentitymustbesome-thingmaintainedfromthestartoflifethroughtheLastJudgment.Hume’stheoryislikeLocke’singivingmemoryaroleinactuallypro-ducingpersonalidentity,andnotmerelygivingevidenceoftheexis-tenceofsomethingelse,suchasasoulorimmaterialsubstancethatisindependentofmemory.However,onHume’saccount,thetrueideaofthemindisthatofasystemofcausallyrelatedperceptions.Thisenableshistheorytoextendpersonalidentitybeyondmemorywithoutappealingtothepersistenceofsubstance.Thethoughtsandemotionsoftheperson“decay’dinhisMemoryandIntellectuals,”asHenryLeeputit,arestillcausallydependentonearlierimpressionsandideas,andthehabitsandassociationsbuiltupovertime.Asaconsequence,thereisnoneedforHumetodenythatthelaterselfisthesamepersonastheearlieronewhenmemoryfails.Nonetheless,itisunlikelythatHume’sappealtocausalitywouldhavesatisfiedmany(orany)ofthecriticsofLocke’sview.Humeacknowledgedwithoutreluctancethatrelationslikeresemblanceandcausalityoccurindegrees.Memory,andthecognitivecapacitiesthatrelyonmemory,oftenfadeorfail:thereisno“juststandard,”accordingtoHume,todeterminepreciselywhenidentityceases.Questionsofthiskindaboutidentity,andevenaboutpersonalidentity,areexamplesofwhatHumecalls“verbal”disputes.Complexrelationsofresemblanceandcausationconnectperceptionsovertime:theseexplainthetendencytospeakofaself,andattributeidentitytoit,eventhoughthereisnothinginvariableanduninterrupted.Theoneissueaboutpersonalidentitythatisnotverbal,andthatcanbedefinitivelyresolved,concernsthesubstancethatispurportedtoconnectourperceptionsthroughtime:nosuchsubstanceexists(T1.4.6.21).iv.personalidentityasitregardsourpassionsWereturnnowtothedistinctionHumemadebetweenpersonaliden-tityasitregardstheimaginationandpersonalidentityasitregardsourpassions.Astheprevioustwosectionshaveshown,HumearguedCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity191thattheself–eitheratagiventimeorovertime–isnothingmorethanacomplexcollectionofperceptions:thereisnounderlyingsubstanceinwhichtheperceptionsinhere,andnoperceptionthatremainsthroughthewholecourseofaperson’slife.Personaliden-tityisthusafiction.AndasHumeconceiveditthetwofoldproblemofpersonalidentityis,first,toexplainwhywetendtobelieveintheidentityoftheself,and,second,toexplainwhyapresentselfwhoisnot,infact,identicalwithaselfwhoexistedinthepastorwillexistinthefutureisnonethelessconcernedwithasetof“pastorfuturepainsorpleasures”(T1.4.6.19).HumelimitedthediscussioninTreatise1.4.6totheformerquestion,whichisdirectlyconnectedtohistheoryoftheimagination.ThusbyHume’sowncharacteriza-tion,thetreatmentoftheproblemofpersonalidentityinBook1oftheTreatiseisincomplete.ThehistoryofthediscussionofpersonalidentityafterLockemakesitclearwhyHumewouldhaveseenthelatterproblem,theexplanationofself-concern,asanimportantissue.ForbothClarkeandButlerhadarguedthatwithoutasubstantialselfthatremainsidenticalthroughtimeapersonwouldhavenoreasontofeelrespon-sibleforpastactions,noranyreasontobeconcernedaboutthefutureconsequencesofpresentactions.BothphilosopherstookthistobeareductioadabsurdemofLocke’sview.Thesameargumentcould,ofcourse,beurgedagainstHume.ItmightbethoughtthataHumeanaccountofthemindhasnoavenueofresponsetothiscriticism.SinceHumedeniesthatanysubstanceidenticaltome,now,existedinthepast,orwillexistinthefuture,theremightappeartobenothingformypresentselftobeconcernedabout:nopastactionsIamresponsiblefor,nofutureconsequencesthatwillaffectme.Thisargumentbegsthequestion,however.Aswehaveseen,Humesupposesthattheselfisextendedthroughtime,butitisnotsoextendedbymeansofthepersistenceofsubstance.OnHume’sview,mypastencompassesthoseperceptions,thoughts,andactionsrelatedtoapresentselfbyresemblanceandcausation.WhatHume’saccountofthepassionsneedstoexplainishowthosepastperceptions,thoughts,andactionsaffectmypresentfeelings,andthereforewhytheyareofinterestandimportancetome.And,sim-ilarly,hisaccountmustexplainwhyIsometimesactoutofcon-cernforafutureself–afuturecollectionofperceptions–thatwillbeartome,now,therelationIbeartomypast.Hume’stheoryofthepassionsdoes,infact,accomplishbothofthesetasks.ItisbyCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n192janel.mcintyreprovidinganaccountofself-concernthatBook2oftheTreatisecompletesHume’saccountofpersonalidentity.LetusfirstconsiderwhatHume’saccountofthepassionshastosayabouttherelationshipofapresentselftoitspast.AthemethatiscommontoHume’sexplanationofthepassionsandhismoralthe-oryisthatcertainactionsandcharactersnaturallyevokepleasureorpaininthosewhoobserveorcontemplatethem.(See,forexam-ple,T2.1.5.2,2.1.7.3–5,3.1.2.3.)Assourcesofpleasureandpain,actionsandcharactershavearoletoplayinthegenesisofwhatHumecallstheindirectpassions.Hume’sinnovativetheoryoftheindirectpassionshasmanysubtleties,butthecentralconceptcanbebrieflysummarized.Althoughsomepassions,desires,oraversionsarisedirectlyfromtheobservationorcontemplationofsomethingpleasurableorpainful,otherfeelingshaveadifferentorigin.Theselatterfeelingsorpassionsoccuronlywhenthesourceofpleasureorpainisrelatedeithertooneself,ortoanotherperson.Prideandhumility,loveandhatred,areinthissenseindirectpassions.Hume’sexplanationofaperson’sconcernwithherpastisrootedspecificallyinhisaccountoftheindirectpassionsofprideandhumil-ity.Whensomethingrelevantlyrelatedtomeisalsoasourceofplea-sureorpain,theresultisafeelingofprideorhumility,respectively.Thesepassionshavetheselffortheirobject,thatis,theyreintroducetheideaoftheself(understoodasaconnectedsuccessionofpercep-tions)intothemind.Humedescribestheunderlyingmechanismasadoublerelation(orsometimesasan“association”)ofimpressionsandideas(T2.1.2.3,2.1.5.5,10).Thepleasurecausedbysomethingcloselyrelatedtome(myvirtueormyhouse,forexample)isasso-ciatedwiththepleasurablefeelingofpride;thisistheassociationoftwoimpressions.Theideaofthecauseisassociatedwiththeideaoftheself.Thesetwoassociationsreinforceeachother,simultaneouslyturningmyattentiontomyselfandgeneratingthefeelingofpride.2020Asanillustration:awellplannedgardenisasourceofpleasurenomatterwhoplannedit,orwhoownsit.Ifthegardenismine,ordesignedbyme,however,Iwillfeelnotmerelypleasure,butpride.Duetotheassociationofideas,lookingatthegardencausesmetothinktomyself(“Imadethat”or“Iownthat”).Thisenhancesanotherassociation,betweenthepleasureoflookingatthegardenandthepleasurableglowofpride.Thecauseofprideisthefactthatsomethingpleasurableisrelatedtome.Thisisnottheresultofaprocessofreasoning,butofmutuallyreinforcingassociations.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity193Analogously,whensomethingrelatedtomeisasourceofpain,theresultingfeelingishumility.Hume’sanalysisofprideandhumilityprovidesanexplanationofaconcernwiththepastthatisanaturalextensionoftheaccountofpersonalidentityinTreatise1.4.6.InlightofthisaccountHumecanexplainwhy,forexample,mypastactionscontinuetoconcernme,eventhoughtheyarenottheactionsofasinglesubstance,andeventhoughtheselfisnotstrictlyidenticalthroughtime.Mypresentselfstandsinrelationsofresemblanceandcausationtovariouspastperceptions,thoughts,andactions:thisiswhatmakesthemmine.Thissamefactofrelatednesstomypresentselfgivesapastoccurrencearoletoplayinthegenerationofmypassions.Insofarasitisindependentlyasourceofpleasureorpain,apastactionwillaffectmewithprideorhumility.Myshameinapastfailureistheresultoftwocircumstances:thepastactisrelevantlyrelatedtome,anditisofakindthatevokesfeelingsofpainordispleasure.Concernwiththepastthereforedependsontherelationsthatholdamongperceptions–specificallytheappropriatedoubleassociationofideasandimpressions–andnotonidentityofsubstance.Thepresentselfextendsitsconcerntothefutureaswellasthepast.Eventhemostordinaryplanscanrequirethatourpresentactionsaregovernedbyafutureinterestthatweseeasourown.Wethinkofourselvesinthefuture,andaremotivated(atleastsometimes)bytheprospectoffuturepainorpleasure,punishmentorreward(T2.3.2.5,2.3.3.3,3.3.12).But,onHume’sview,whatisinvolvedinthinkingofoneselfinthefuture?Howdosuchthoughtsaffectourpassionsandservetomotivateourpresentactions?Book2oftheTreatise,likeBook1,takescausationtobetheprimaryconnectionbetweentheperceptionsthatconstitutetheselfthroughtime.ButBook2introducesanothercausalrelationintotheaccountoftheself:therelationbetweenintentionandaction.Humearguedthatalthoughsomemightrefusetoapplythetermscauseorneces-sitytothatrelation,theconstantconjunctionbetweenintentionandaction,andtheinferencefromtheonetotheother,is“universally,tho’tacitly”allowed(T2.3.2.4).Evenincommonlife,therefore,myfutureisbelievedtobecausallyrelatedtomypresentandmypast.Inthinkingofmyselfinthefuture,Iamthinkingoftheactionsthatfollowfrommymotives,intentions,andcharacter,ratherthanofsomesubstanceinwhichtheseactionsinhere.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n194janel.mcintyreTounderstandwhythepassionsareaffectedbyourideasofthefuturerequiresatleastabriefexaminationoftherelationshipbe-tweentheimaginationandthepassions,andtheinfluenceonthemofspatialandtemporalrelations.Theimaginationtendstofocusonthevividideasofwhatispresentandnear:it“passeseasilyfromobscuretolivelyideas,butwithdifficultyfromlivelytoobscure.”Ideasofthingsthataredistantineitherspaceortimetendtoberel-ativelyfaint,andassuchmightbeexpectedtohavelittleinfluenceonthepassions.Butthepassionshaveadynamicoftheirownthatcounteractsthis:theytendtospreadthemselvesout.ThefeelingofloveorhatredIhaveforanotherpersontransfersreadilytotheirrela-tions(to,thatis,membersoftheirfamily)andotheracquaintances(T2.2.2.15,19).Furtherexperiencemaychangeorcorrectthesetransferredfeelings,butthepassionsarenaturallyexpansive.Oneofthemostfundamentaltraitsofhumannature,themind’sabilitytoturn,asitdoesinsympathy,fromtheideaofmyselftotheideaofanotherpersonpresupposesthisexpansivetendency.Andtheoper-ationofsympathyexplainstheinfluenceofourideasofthefutureonourpresentactions.Humeinitiallypresentssympathyasthepropensityto“receivebycommunication”thesentimentsofothers.Wefirstformanideaofanotherperson’sfeelingbyperceivingitseffectson“countenanceandconversation.”Theresemblancebetweenotherpersonsandour-selvesenlivensthatidea,transferringtoitthevivacityofourownpassions.Asaresultofthisprocess,wearenotmerelyconvinced,byapieceofcausalreasoning,oftherealityoftheotherperson’sfeeling,buttheideaofthatfeelingisactuallyconvertedintoanimpressionthatcopiesthefeelingofthatperson.Inotherwords,theoperationofsympathyallowsustoshare(andthustoknow)thefeelingsofothers(T2.1.11.2–8).Thebasicfunctioningofsympathyisinterper-sonal,butitcanalsobeextendedbeyondthepresenttopainsandpleasuresthatareonlyanticipated.Humewrote:Sympathybeingnothingbutalivelyideaconvertedintoanimpression,’tisevident,that,inconsideringthefuturepossibleorprobableconditionofanyperson,wemayenterintoitwithsovividaconceptionastomakeitourownconcern;andbythatmeansbesensibleofpainsandpleasures,whichneitherbelongtoourselves,noratthepresentinstanthaveanyrealexistence.(T2.2.9.13)CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity195Theidentificationofmyinterestwiththeinterestsofafutureperson–theidentificationofmyselfwiththatperson–istheresultoftheextendedoperationofsympathy.WhenIthinkofmyselfinthefutureIthinkoftheactionsthatfollowfrommypresentintention,motives,andcharacter:Ithinkoftheconsequencesofthoseactions,andthecircumstancesinwhichtheywilltakeplace.Thiscausalconnectionbetweensomethingcentraltomypresentself(to,thatis,myintentions,motives,andcharacter)andtheactionsandcir-cumstancesofafuturepersonfacilitatestheoperationofsympathy.The“possibleorprobablecondition”ofthatfuturepersonbecomesmyownpresentconcern.Thisaccountexplainswhy,eventhoughthereisnostrictidentityoftheselfthroughtime,weidentifyafutureinterestasourown.Amerefictionoftheimaginationwouldhavenoinfluenceonthepassions,accordingtoHume,butsympathywithafutureselfcanserveasthemotivationforaction.Thisdoesnotentail,however,thatwealwaysactonthebasisofthatfutureinterest.TothinkofmyselfatapointoftimeinthefuturerequiresthatItraceacausalchainbetweenmyselfandthatfuturetime(T2.3.6.10,2.3.7.2).Themoreremotethetime,theharderitistotracetheconnection,thuslesseningtheeffectofsympathy.21v.thediscussionofpersonalidentityintheappendixtothetreatiseTheargumentoftheprevioussectionsrevealsaremarkableconsis-tencyinHume’saccountoftheselfthroughoutBooks1and2oftheTreatise.Themind,orself,isacompoundentity–acollectionofperceptionsofdifferenttypesconnectedtoeachotherbycausa-tionandresemblance.Asaconsequenceofthisview,theproblemofpersonalidentityhastwodimensions.First,toexplainwhywetendtobelieveintheidentityoftheself,and,second,toexplainwhyapresentselfwhoisnot,infact,identicalwithaselfinthepastorthefutureisnonethelessconcernedwithapastandafutureitidentifiesasitsown.Humeaddressedthefirstpartoftheproblemof21Onthisaccount,self-concernandconcernforothersdifferindegree,ratherthaninkind.ToadoptthedisinterestedperspectiveHumetooktobenecessaryformoralevaluationistorecognizetheessentialsimilarityoftheserelations.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n196janel.mcintyrepersonalidentityinTreatise1.4.6,andthesecondpartthroughtheaccountoftheindirectpassionsandsympathyinBook2.Takenasawhole,then,theTreatisepresentsacompleteanalysisofthenatureoftheselfanditsidentity–anaccountthatwascraftedtoberespon-sivetotheissuesaboutpersonalidentityraisedinthefiftyyearsfollowingLocke’sdiscussioninhisEssay.Nevertheless,intwelveparagraphsoftheAppendixtotheTreatise(TApp.10–21)HumefamouslyexpresseddissatisfactionwiththecentralargumentshehadgiveninOfpersonalidentity.Theseparagraphsconfessamis-take,buttheargumentbehindthemiscryptic,andscholarshavenotbeenabletoagreeonthenatureofthemistaketowhichHumerefers.22ThissectionaimstoclarifytheissueraisedbytheAppendix.TheopeningsentencesofHume’sdiscussionofpersonalidentityintheAppendixsetthestageforthediscussionthatfollows:Ihadentertain’dsomehopes,thathoweverdeficientourtheoryoftheintel-lectualworldmightbe,itwou’dbefreefromthosecontradictions,andabsurdities,whichseemtoattendeveryexplication,thathumanreasoncangiveofthematerialworld.Butuponamorestrictreviewofthesectioncon-cerningpersonalidentity,Ifindmyselfinvolv’dinsuchalabyrinth,that,Imustconfess,Ineitherknowhowtocorrectmyformeropinions,norhowtorenderthemconsistent.(TApp.10)Humegoesontosaythathewillreviewthearguments“onbothsides”oftheissue,beginningwiththosethatleadhimtodenythesimplicityandidentityoftheself.Heproceedstorepeat,briefly,thatwehavenoideaofasimpleandindividualselforsubstancebecausewehavenoimpressionofsuchasubstance;thatallpercep-tionsmaybeconceivedasexistingseparately,andthusmayinfactexistseparately;thatthereisnocontradictioninvolvedinholdingthatperceptionsdonotinhereinasimplesubstance;thatreflectiononthemindrevealsnothingbutperceptions,sothat“’Tisthecom-positionofthese,therefore,whichformstheself”(TApp.15).Humenotesthattheseconclusionsaresimilartothosethatphilosophers22ForreviewsofvariousproposalsconcerningHume’sargumentintheAppendix,seeDonGarrett,CognitionandCommitment(NewYork:OxfordUniversityPress,1997),167–80;andCorlissSwain,“PersonalIdentityandtheSkepticalSystemofPhilosophy,”inTheBlackwellGuidetoHumesTreatise,ed.S.Traiger(Malden,MA:BlackwellPublishing,2006),133–50.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity197havealreadydrawnwithrespecttoexternalsubstances(TApp.11–15,19).AtthisjunctureHumeremarks:“SofarIseemtobeattendedwithsufficientevidence”(TApp.20).Whatthenpromptshimtohavereservationsabouttheaccountofpersonalidentitypre-sentedinTreatise1.4.6?WhenweturntotheremainderofHume’sdiscussion–tothecritiqueofhisownview–thesourceofhisdissat-isfactionisdifficulttomakeout.Hume’sfocusinAppendix20and21isontheprinciplesthatbindthedistinctperceptionstogetherandmake“usattributetothemarealsimplicityandidentity.”TheanalysisofcausalityinTreatise1.3ledtotheconclusionthattheconnectionsbetweendistinctexistencesareonlythe“determina-tionofthethought,topassfromoneobjecttoanother.”Applyingthatconclusiontotheself,itfollowsthat“thoughtalonefindsper-sonalidentity,whenreflectingonthetrainofpastperceptions,thatcomposeamind.”Thatis,thoughtfindsthattheideasofpastpercep-tionsarefelttobeconnectedtogether,andnaturallyintroduceeachother.Humedrawsaparallelbetweenthisviewandthepositionthatpersonalidentityarisesfromconsciousness,whichhesurprisinglysays“mostphilosophers”areinclinedtoaccept.23However,hetakesthistobeafurther“promisingaspect”ofhisownaccount,notaneg-ativeone.So,althoughphilosopherssuchasClarkeandButlerwouldhaveviewedthisasadevastatingcriticismofHume’saccountoftheself,Humeendorsesitasaconsequenceofhisaccountofcausalrela-tions.Inspiteofthis,paragraph20concludeswithHume’sassertionthathecannotdiscoveranytheorythatprovidesasatisfactoryexpla-nationoftheprinciplesthat“uniteoursuccessiveperceptionsinourthoughtorconsciousness.”Paragraph21oftheAppendixisevenmoreperplexing,forthereHumesaysthattherearetwoprinciplesthathecanneitherrenderconsistentnorrenounce,namely,“thatallourdistinctperceptionsaredistinctexistences,andthatthemindneverperceivesanyrealconnexionamongdistinctexistences.”Virtuallyeverycommenta-torhasacknowledgedthattheseprinciplesarenotinconsistent,andthattheyareessentialnotmerelytoHume’saccountoftheself,buttomanyofthecentralargumentsofBook1oftheTreatise.What23AswehaveseeninSectionIofthisessay,thisLockeanviewwasnotoriouslycontroversialfromthemomentofitspublication.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n198janel.mcintyrethenisHume’sspecificdisappointmentwithhisaccountofpersonalidentity?AnanswertothisquestioncanbefoundbyreexaminingAppendix10,theparagraphquotedaboveinwhichHumesaidthathehadhopedthathistheoryoftheintellectualworldwouldbe“freefromthosecontradictions,andabsurdities,whichseemtoattendeveryexplication,thathumanreasoncangiveofthematerialworld.”24ThissentencereferstothefirstparagraphofTreatise1.4.5,Oftheimmaterialityofthesoul,whereHumehadclaimedthattheaccountoftheintellectualworld“isnotperplex’dwithanysuchcontradic-tions,asthosewehavediscoveredinthenatural”world.Throughthatcomment,italsorefersindirectlytotheargumentsof1.4.2,Ofscepticismwithregardtothesenses,and1.4.4,Ofthemodernphilosophy,andtothesummariesandcommentsonthoseargu-mentsatT1.4.4.15and1.4.7.3–5.Eachofthesepassagesdiscussesacontradictionwithintheimagination.Inthem,Humearguesthattwooperationsoftheimaginationleadtocontradictoryresults.TheassociationsofideasdescribedinTreatise1.4.2leadtobeliefinthecontinuedandindependentexistenceofobjects.Ontheotherhand,reasoningfromcausesandeffects,anotherassociationofideas,leadstotheconclusionthattherearenoobjectswithcontinuedandinde-pendentexistence.TheaccountoftheintellectualworldinTrea-tise1.4.5,Oftheimmaterialityofthesoul,wasbyitselfable(asitclaimedto)toavoidthisinconsistency.Itdidsobecauseitdidnotconsiderbeliefinthecontinuedexistenceofanysubstance.Itarguedthatthemindisnothingbutperceptions,withoutanunder-lyingsubstance.Thereisnoneed,then,in1.4.5topositafeigned“doubleexistence”resultingfromadherencetotwocontraryprin-ciples.Infact,1.4.5presentsacleverargumentshowingwhytherecouldnotbeadoubleexistenceofperceptionsandanunderlyingsubstance.Itarguedthatsomeperceptionsareextended,andsomeunextended,andasaresultitisclearthattheycannotallinhereineitheramaterialoranimmaterialsubstratum.TheargumentofTreatise1.4.6,Ofpersonalidentity,however,reopenstheproblemofinconsistency.Inthissection,asin1.4.2,24ThisgeneralapproachisalsotakenbyA.E.PitsoninHume’sPhilosophyoftheSelf(London:Routledge,2002),Chap.4;andbyCorlissSwain,“PersonalIdentityandtheSkepticalSystemofPhilosophy.”PitsonandSwaindeveloptheiranalysesinwaysdifferentfromthatfollowedhere.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity199theassociationofideasisappealedtoasthesourceofthebeliefthatsomethingcontinuestoexist(inthiscasesoulormentalsub-stance)inspiteofallthevariationinourperceptionsovertime.ThestructureoftheargumentinT1.4.6.6isthesameasthatoftheargu-mentinT1.4.2.31–43.Consequently,theaccountoftheintellectualworld,liketheaccountofthematerialworld,revealsinconsistenttendenciesorprinciplesintheimagination.Bothaccountsfacethesameproblem:theimaginationproducesbeliefinthecontinuedexistenceofsomethingotherthanperceptions(asubstratum),whileatthesametimerevealingthatnothingofthatkindexists.OfpersonalidentitypresentsaviewthatHumeexplicitlycalled“thetrueideaofthehumanmind,”theviewofthemindasasys-temofdifferentperceptionsorexistenceslinkedtogetherbytherelationofcauseandeffect.Wearriveatthisideainthefollowingway:(1)Memoryacquaintsuswiththecontinuanceandextentofasuccessionofperceptions.(2)Frommemorywealsogettheideaofcausation.And(3)oncewehavetheideaofcausation,weextendthechainofperceptionsbeyondmemory(T1.4.6.19–20).Thisprogres-sionisthenaturaldevelopmentofthepositionarrivedatin1.4.5,namely,thattherecouldnotbeasubstanceinwhichperceptionsinhere,andhencethatthemindcanbenothingbutperceptionsandtheircausalrelations.In1.4.5and1.4.6Humeneverexpressesanyreservationsaboutthisviewofthemind,whichisgroundedinhisownaccountofcausality.Thisisapointofcontrastwithhisdiscus-sionofthebeliefintheexistenceofbody,wherenoviewissingledoutasthe“true”view.Humeheldthatthebeliefthattheselfisidenticalovertimeistheresultofanotheroperationoftheimagination–thetendencyoftheimaginationtoassociatebyresemblance,andtoconfusetheexperienceofcloselyrelatedsuccessionwiththeexperienceofanunchangingthing.Thebeliefintheidentityoftheselfisfalse,butitisproducedbyoperationsoftheimaginationthatareasentrenchedascausalreasoning.Theimaginationthereforeleadsbothtothetrueviewofthemindasacausalsystemofperceptions,andthefalseviewthattheselfisasubstancethatisidenticalthroughtime.InthiswayHume’sargumentabouttheselfrecapitulatesthedilemmahearticulatedconcerningbeliefintheexistenceofbody.Asaconsequence,thetheoryoftheintellectualworldisnotfreefromthecontradictionsthatbedeviledthetheoryofthematerialworld.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n200janel.mcintyreWecannowseewhyHumesaidintheAppendixthathedidnotknowhowtocorrecthisformeropinions,ortomakethemcon-sistent.Hehadformerlyheldthathisaccountoftheintellectualworldwasfreefromthecontradictionsthatfacedhisaccountofthematerialworld.Henowseesthatbothaccountsfacethesameprob-lem,namely,thatcertainfundamentaltendenciesoftheimaginationleadtoinconsistentconclusions.Furthermore,thereisnoway(ornowayacceptabletoHume)togiveanaccountofwhywebelieveintheidentityoftheselfovertimethatavoidsthisproblem.Beliefistheresultoftheenliveningofideas,anenliveningthatisaproductofassociation.OnHume’saccount,theonlyconnectionsthatexistamongperceptionsarefeltconnectionsthataretheproductoftheassociationofideas.Thefeelingproducedbytheseconnectionsisthesourceofthebeliefinthecontinuedexistenceofbothmindandbody.WecanseewhyHumemightnothaverecognizedthisproblemearlier.First,Hume’sbasictheoryoftheintellectualworldisinter-nallyconsistent,howeverunsatisfactoryitwastotheologiansonothergrounds.Heholdsthat1.Themindisnothingbutperceptions2.Thereisnosubstance(materialorimmaterial)underlyingthoseperceptions3.Theexistenceoftheselfthroughtimecanbeexplained,withoutcontradiction,asasuccessionofcausallyrelatedperceptions4.Nothingmorethanthisisnecessarytoexplaintheoperationofthepassions,particularlytheconcernwehaveforourfutureselves.Furthermore,Humedidnotrejectthebeliefinacontinuallyexist-ingselfsolongasthatwasnottakentoincludethefalsebeliefincontinuallyexistingsubstanceorsoul.AlthoughHumewasambiva-lentaboutthestrengthofthegeneraltendencytoendorsethisfalsebelief,heclearlyfounditpossibletocorrectitinhimself.ElsewhereintheTreatiseHumearguedthateducationandindoctrinationarti-ficiallypromotethebeliefinafuturestateofthesoulafterdeath(T1.3.9.13).Initially,therefore,theredidnotappeartobeaproblemwiththeaccountembeddedin1.4.5and1.4.6.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity201Unfortunately,ifHume’slaterreflectionsarecorrect,theviewbeforetheAppendixunderestimatesthetendencyoftheimagina-tiontoarriveatthefalsebeliefinacontinuallyexistingmentalsubstanceorself.Ifwelookbacktotheaccountofthebeliefintheexistenceofbody,andthecontradictionsthatoccurinthetheoryofthematerialworld,wecanseesomeofthedifficultiesemerginginhisaccountoftheself.Hume’stheoryofthematerialworlddepictssuchastrongtendencytotakeresemblingperceptionstobeidenticalovertimethatitisimpossiblereallytoeradicateit.Wealsocannotgiveuptheopinionthatresemblingimpressionsareinterruptedanddifferentfromeachother(T1.4.2.51–2).Thisleadstotheunsatisfac-toryphilosophicaltheoryofthedoubleexistenceofperceptionsandobjects.Theoperationoftheimaginationshouldbethesamewhencon-sideringtheperceptionsthatconstitutethemind.Thiswouldleadtotwoopposingbeliefs:anineradicable(thoughstillfalse)beliefintheidentityofselforsubstanceovertime;andtheopinion,wellestablishedbyreflection,thatallourperceptionsaredistinctexis-tences.Theanalogousunsatisfactoryphilosophicalpositionwouldbethatthereisasubstance,differentfromperceptions,thatcontin-uesidenticallythesame,inwhichthevariableperceptionsinhere.Asanaccountofadilemmainearlymodernphilosophyofmind,thisisnotimplausible.Infact,itseemstoreflecttheviewofper-sonalidentityheldbynearlyallofLocke’scritics.Butthereisnolongeracompletetheoryof“theintellectualworld”thatisfreeofcontradictions.Theinternalworldisbesetbythesamecontradic-tiondiscoveredinthetheoryofthematerialworld.Agreatercausefordespair,however,isthatitdoesnotleavemuchhopethatthe“trueideaofthehumanmind”willbeonethatwecanbelieve.vi.“maturereflection”:hume’sviewoftheselfafterthetreatiseOfpersonalidentitypresentedthetendencytobelieveinaselfthatwasstrictlyidenticalthroughtime(andthereforewasnotabundleofperceptions)asonethatcouldbeexplainedthroughtheoper-ationoftheimaginationandpotentiallycorrectedthroughreflec-tion.Hume’slaterclaimintheAppendixthatthetheoriesoftheintellectualandmaterialworldsaresubjecttothesameinherentCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n202janel.mcintyrecontradictionsindicatedthathehadlosthisconfidencethatphilo-sophicalargumentslikethoseinTreatise1.4.5–6couldcounterbal-ancethefalsebeliefinasubstantialself.TheAppendix,however,doesnotretractthepositiveaccountoftheselfarguedforintheTreatise.And,althoughHumedidnotdirectlyaddresstheproblemofpersonalidentityagain,bothhispublishedandhisunpublishedworkaftertheTreatiserevealaviewoftheselfstrikinglysimilartotheearlieraccount.AftertheTreatise,Hume’sexpressionofhisviewsontheselfwasmoreguardedinhispublishedworksthaninthoseheleftunpub-lished.Nevertheless,theEnquiryconcerningHumanUnderstand-ingprovidesexamplesofthepersistenceofmanyofthemaintenetsHumesoughttoestablishinT1.4.5–6,eveninpassagesnottrans-plantedfromtheTreatise.Enquiry7,inwhichHumediscussestheideaofnecessaryconnection,providesanimportantexampleofthis.TheTreatisecontainedonlyabriefmention,addedasanoteintheAppendix,oftheimpossibilityofderivingtheideaofpowerfromthefeelingofvolition(T1.3.14.12).Bycontrast,theargumentagainstthisoriginoftheideaofpowerismoreextensiveintheEnquiry,incorporatingdiscussionsofboththeinfluenceofmindonbody,andtherelationofthemindtoitsownideasandpassions.InbothworksHumedeniedthatwehaveanyideaofaconnectionotherthanobservedconjunction.Evenwithinthemindthereisnostrongertietobediscoveredamongourperceptions.BothT1.3.14.12andtheEnquiryagreewiththeviewexpressedinOftheimmaterialityofthesoul,theviewthat“tho’thereappearnomannerofconnex-ionbetwixtmotionorthought,thecaseisthesamewithallothercausesandeffects.”ThiswasacentralpartoftheargumentHumegaveinsupportofhisviewofthecompoundnatureoftheself.Fur-ther,Enquiry7continuestorejecttheconceptofimmaterialsub-stance,referringtothemysteriousunionofthesoulandthebody“bywhichasupposedspiritualsubstance”acquiresinfluenceoveramaterialone(EHU7.11).ThustheargumentofthissectionoftheEnquirysupportsmanyoftheconclusionsHumearguedforinT1.4.5–6.Everythingweknowabouttheconnectionbetweenmindandbody,ortherelationshipofamindtoitsideas,comesthroughexperience.Experiencerevealsthattherearesuchrelations,butnothowtheyoperate.Humeconcludesbysaying:“Sothat,uponthewhole,thereappearsnot,throughoutallnature,anyoneinstanceCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity203ofconnexion,whichisconceivablebyus.Alleventsseementirelylooseandseparate.Oneeventfollowsanother;butwenevercanobserveanytyebetweenthem”(EHU7.26).Thenatureofthemindprovidesnoexceptiontothisprinciple.UnliketheTreatise,theEnquiryincludesextendeddiscussionsoftheologicalissues.ThischangeofemphasisisalsoreflectedinHume’streatmentofquestionsabouttheself.WhileEnquiry7showsthatHumecontinuedtorejecttheviewthatthereareneces-saryconnectionsamongthecomponentsofthemind,inSection11,“OfaParticularProvidenceandofaFutureState,”Humeendorsestheradicaltheologicalconsequencesofhisviewoftheself.AswasnotedinPartIofthisessay,Butler’s“DissertationonPer-sonalIdentity”wasoneofthebest-knowncritiquesoftheLockeanaccount.Butleropenedwiththefollowingstatement:WhetherwearetoliveinafutureState,asitisthemostimportantQues-tionwhichcanpossiblybeasked,soitisthemostintelligibleonewhichcanbeexpressedinLanguage.YetstrangePerplexitieshavebeenraisedabouttheMeaningofThatIdentityorSamenessofPerson,whichisimpliedintheNotionofourlivingNowandHereafter,orinanytwosuccessiveMoments.25LikeClarke,Butlerarguedthatthesoul’sexistenceinafuturestate,oritsrewardandpunishmentinthatstate,requiredasubstantialself,strictlyidenticalthroughtime.WecanonlyspeculatewhetherHumehadreadButler’saccountofpersonalidentitybeforewritinghisown,butitwouldbeverysurprisingifhehadnotreaditbythetimehecomposedtheEnquiry.AlthoughtheaccountofpersonalidentitypresentedintheTreatiseclearlydidnotsatisfythecondi-tionsheldbytheologianstobenecessaryforafutureexistence,theTreatisecontainsonlyabriefcriticismofthebeliefinthesoul’sstateafterdeath(T1.3.9.13).Section11oftheEnquiry,however,presentsadirectattackonamajorargumentforthisbelief.Writ-tenintheformofadialogue,itdevelopsacritiqueoftheargumentfromdesign.InthiscontextHumeconsidersandrejectstheclaimthatdistributivejusticerequirestheexistenceofafuturestateinwhichsoulsarerewardedorpunished.Ourexperienceofjusticeintheworldprovidesnogroundforexpectingafuturestateofthe25JosephButler,DissertationIinTheAnalogyofReligion(London,1736),301.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n204janel.mcintyresoulinwhichjusticewillbemoreperfectlysatisfied(EHU11.22).RatherthanretreatingfromtheanalysisoftheselfintheTreatise,theEnquiryembracesoneofitsmostcontroversialimplications.OfHume’sunpublishedworks,“OftheImmortalityoftheSoul”shedsthemostlightonhisviewoftheselfaftertheTreatise.Thisessaywasoneoftwointendedforpublicationin1755,butsuppressedbyHumebeforeitwasdistributed.26Itincludessomeoftheargu-mentsfromOftheimmaterialityofthesoulandSection11oftheEnquiry,butitsscopeisconsiderablybroader.“OftheImmortalityoftheSoul”touchesonthefullrangeofproblemsabouttheselffoundintheTreatise,includingtheconcernoftheselfwiththepastandthefuture,andtherelationshipofself-concerntofactsaboutourbodilyexistence.TheorganizationofthisessayindicatessomeofitsoverlapwithHume’srelatedpublishedwork.Humeconsiders,inturn,metaphys-ical,moral,andphysicalargumentsastheybearonthequestionoftheimmortalityofthesoul.Themetaphysicalargument,hecon-tends,isthatthesoulisimmaterial,andthereforeimmortal.HumerepeatstheargumentsexplicitlypresentedintheTreatise(andonlyslightlylessexplicitlyintheEnquiry)thattheessenceofmatterandspiritareequallyunknown;wecannotruleout,apriori,thatmattermightbethecauseofthought(E-IS4,591).Atthispoint,however,Humeintroducesanargumentnotfoundinhispreviousworks.Evenifimmaterialsubstancedoesexist,thiswouldnotprovetheimmortalityofindividualpersonalsouls:“Asthesamematerialsubstancemaysuccessivelycomposethebodyofallanimals,thesamespiritualsubstancemaycomposetheirminds:Theirconsciousness,orthatsystemofthought,whichtheyformedduringlife,maybecontinuallydissolvedbydeath;andnoth-inginteresttheminthenewmodification”(E-IS5,591–2).Twoaspectsofthispassagearesignificant.First,contrarytoHume’sclaiminAppendix21,theinherenceofideasinasubstancewouldnot,accordingtothisargument,adequatelyaccountforpersonalidentity.Second,Humearguesthatwewouldnotbeinterestedintheunderlyingsubstanceonceourownconscioussystemofthought26E.C.Mossner,TheLifeofDavidHume,2nded.(Oxford:ClarendonPress,1980),Chap.24;J.C.A.Gaskin,HumesPhilosophyofReligion,2nded.(NewJersey:HumanitiesPress,1988),Chap.9.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity205isdissolved.Ourconcerniswiththeorganizedconscioussystem,notwithasubstancesupportingit.Inthisrespect,HumereiteratestheviewofTreatise1anddevelopedinhisaccountofthepassions,theviewthatthecausalconnectionsamongthecomponentsoftheselfmake“ourdistantperceptionsinfluenceeachother,”andgiveusapresentconcernforour“pastorfuturepainsorpleasures”(T1.4.6.19).Underscoringthis,Humewrote:“whatisincorruptiblemustalsobeingenerable.Thesoul,therefore,ifimmortal,existedbeforeourbirth:Andiftheformerstateofexistencenowisecon-cernedus,neitherwillthelatter”(E-IS6,592).Themetaphysicalargumentfails,accordingtoHume,preciselybecauseitfailstoprovetheexistenceofanobjectwhosefatewewouldbeinterestedin,oraboutwhichwewouldbeconcerned.Theissueofself-concernplaysanimportantpartaswellinHume’sreformulationofhiscritiqueofthemoralargumentsfortheimmortalityofthesoul.Thesemaintainthatthesoulmustbeimmortalsothattherewillbeanarenainwhichvirtuewillunfail-inglyberewardedandvicepunished.AlthoughHumerepeatstheargumentconcerningdistributivejusticegiveninSection11oftheEnquiry,hisemphasishereisonarguingthataperson’sconcernsareall“limitedtothepresentlife.”Humeasked:“Withhowweakacon-cern,fromtheoriginal,inherentstructureofthemindandpassions,doesheeverlookfarther?Whatcomparison,eitherforsteddinessorefficacy,betweensofloatinganidea,andthemostdoubtfulpersua-sionofanymatteroffact,thatoccursincommonlife.”Ratherthanassuringjustice,theimmortalityofthesoulwouldinfactconflictwithit.Moralideasarederivedfromthereflectionontheinterestsofhumansocietyandrelatetothepresentlife.Eternalrewardsandpunishmentsservenopurpose,“afterthewholescene[ofthepresentlife]isclosed,”andareoutofproportiontothelimitedintereststheysupposedlyguard(E-IS11,19–22,28,592,594–5).Theargumentsinthefinalsectionoftheessay,whichHumedescribesas“physicalargumentsfromtheanalogyofnature,”haveanimportanttietoHume’sreferencestomindandbodyinBook2oftheTreatise(E-IS30,596).IntheTreatisethepassionsofprideandhumilityareamongtheprimaryexpressionsoftheconcernwetakeinourselves.Humeidentifiedqualitiesofbothmindandbodyaspartsoftheself,andascausesofprideandhumility(T2.1.9.1).Fromthepointofviewofthepassions,mindandbodyareonaCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n206janel.mcintyrepar.Inthefinalsectionof“OftheImmortalityoftheSoul”thisperspectiveonthepassions,inwhichthebodyisfullyintegratedintotheconceptoftheself,becomesthedominantview.Inthepresentlife,whichHumearguesisouronlyconcern,“Everythingisincommonbetweensoulandbody.Theorgansoftheoneareallofthemtheorgansoftheother.Theexistencethereforeoftheonemustbedependentonthatoftheother”(E-IS36,596).Hume’sconclusionthatthesoulisdependentonthebody,how-ever,goesbeyondeventheenhancedviewoftheselfinBook2oftheTreatise.Thebodyisnolongerdepictedasapartofwhatweconsiderasourselves.Onthecontrary,ourmentalqualitiesdependfortheirexistenceontheexistenceofourbodies.BoththeTreatiseandtheEnquiryacceptedtheexistenceofcausalconnectionbetweenmindandbodybecauseoftheexperiencedconstantconjunctionsbetweenthem.In“OftheImmortalityoftheSoul”Humeappealstoanalogi-calargumentstosupportthestrongerconclusionthatchangesinthebodyarealwaysaccompaniedbychangesinthemind.Frominfancytooldage,thepowersofthebodyandthemindare“exactlypro-portioned.”Humecontinued:“Thestepfartherseemsunavoidable;theircommondissolutionindeath”(E-IS33,596).vii.conclusion“OftheImmortalityoftheSoul”providesanaccountofthebeliefinthefutureexistenceofthesoulthatdrawsmoreonthepassionsthantheimagination.Althoughourconcernsallrelatetothislife,andwecannothelpbutrecognizethatthedissolutionofthebodyresultsintheannihilationofthesoul,thatthoughtterrifiesus.Thehorrorthatwefeelhastheutilityofhelpingtopreservethehumanspecies,butthisinitselfmakesthebeliefinimmortalityproblematic,for“Alldoctrinesaretobesuspected,whicharefavouredbyourpassions”(E-IS43,598).“OftheImmortalityoftheSoul”andtherelatedessay“OfSui-cide”alsoreviveandextendtheargumentHumepresentsonlybrieflyintheTreatise,theargumentthateducationandindoctri-nationcontributetothestrengthandentrenchmentofthebeliefintheexistenceofthesoul,andacontinuedlifeafterthedeathofthebody.Intheseessays,beliefinthesoulisnottheoutcomeCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeandtheProblemofPersonalIdentity207oftheworkingsoftheimagination,buttheresultofartificiallyfos-teredsuperstitionthatexploitsournaturalhorrorofannihilation.AsHumesaysin“OftheImmortalityoftheSoul”:“Therearise,indeed,insomeminds,someunaccountableterrorswithregardtofuturity:Butthesewouldquicklyvanish,weretheynotartificiallyfosteredbypreceptandeducation.Andthose,whofosterthem;whatistheirmotive?Onlytogainalivelihood,andtoacquirepowerandrichesinthisworld”(E-IS12,593).“OfSuicide”reiteratesthisaccountoftheinfluenceofsuperstitionandappealsto“truephilosophy”tocounterthefalseopinionthatsuicideiscriminal(E-Su1,579).And,inasimilarvein,inPartXIIoftheDialoguesconcerningNaturalReligion,Philoclassifiesbeliefinafuturestateasavulgarsuper-stition.Religiousmotives,suchastheconcernforeternalpunish-mentsandrewards,“operateonlybystartsandbounds”and“mustberousedbycontinualefforts”(DNR12.13,17,221,222).InHume’slaterworks,therefore,itisnottheimaginationalonethatisthesourceofanineradicablebeliefinasimpleselfthatremainsstrictlythesamethroughtime.Thepassions,assistedbyeducationandindoctrination,alsoplayapartinenliveningthisbelief.Inthisrespect,wefindsomeresolutiontotheproblemthatworriedHumeintheAppendix,fortheimaginationaloneisnolongerimplicatedinalabyrinthofcontradictioninitsaccountoftheintellectualworld.ThefinalevidencethatHumemaintainshisviewoftheselfinhislaterphilosophyisrevealedintheDialoguesconcerningNat-uralReligion.TherewefindbothDemeaandCleanthesexplicitlyendorsingtheanalysisHumeputforthintheTreatise.AttheendofPartIIIDemeaeloquentlyproclaims:“Ourthoughtisfluctuating,uncertain,fleeting,successive,andcompounded;andwerewetoremovethesecircumstances,weabsolutelyannihilateitsessence”(DNR3.13,156–7).HecontinuesatthebeginningofPartIV:Whatisthesoulofman?Acompositionofvariousfaculties,passions,sen-timents,ideas;united,indeed,intooneselforperson,butstilldistinctfromeachother.Whenitreasons,theideas,whicharethepartsofitsdiscourse,arrangethemselvesinacertainformororder;whichisnotpreservedentireforamoment,butimmediatelygivesplacetoanotherarrangement.Newopinions,newpassions,newaffections,newfeelingsarise,whichcontinu-allydiversifythementalscene,andproduceinitthegreatestvarietyandCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n208janel.mcintyremostrapidsuccessionimaginable.Howisthiscompatiblewiththatperfectimmutabilityandsimplicity,whichalltruetheistsascribetotheDeity?(DNR4.2,159)Cleanthesassertsslightlylater:“Amind,whoseactsandsentimentsandideasarenotdistinctandsuccessive;one,thatiswhollysimple,andtotallyimmutable;isamindwhichhasnothought,noreason,nowill,nosentiment,nolove,nohatred;orinaword,isnomindatall”(DNR4.3,159).Philotakesnoexceptiontothesedescriptionsofthemind,andenjoyspointingouttoCleanthesthat,onhisview,allthe“orthodoxdivines”areatheists–anargumentstronglyreminiscentofHume’sinOftheimmaterialityofthesoulintheTreatise.Humewasobviouslyhavingfunhere,forneitherClarkenorButler,thesupposedmodelsforDemeaandCleanthes,wouldhaveembracedthoseHumeansentiments.YetbyhavingallthreeparticipantsintheDialoguesagreeonthisviewofthehumanmindHumeleftlittlepossibilityfordoubtabouthisfinalviewoftheself.Afterthirty-sevenyearsofmaturereflection,theaccountoftheselfintheTreatiseemergedagain,unscathed.suggestionsforfutherreadingInadditiontotheworkscitedinthenotestothisessay,forfurtherreadingthefollowingarerecommended.Baier,Annette.“HumeonHeapsandBundles.”AmericanPhilosophicalQuarterly16(1979):285–95.AProgressofSentiments:ReflectionsonHumesTreatise.Cambridge,MA:HarvardUniversityPress,1991.Chap.6.Garrett,Don.“Hume’sSelf-DoubtsaboutPersonalIdentity.”PhilosophicalReview90(1981):337–58.Penelhum,Terence.ThemesinHume:TheSelf,TheWill,Religion.Oxford:ClarendonPress,2000.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nrobertj.fogelin7HumesSkepticismByallthathasbeensaidthereaderwilleasilyperceive,thatthephilosophycontainedinthisbookisverysceptical,andtendstogiveusanotionoftheimperfectionsandnarrowlimitsofhumanunderstanding.Almostallreasoningistherereducedtoexperience;andthebelief,whichattendsexperience,isexplainedtobenothingbutapeculiarsentiment,orlivelyconceptionproducedbyhabit.Noristhisall.Whenwebelieveanythingofexternalexistence,orsupposeanobjecttoexistamomentafteritisnolongerperceived,thisbeliefisnothingbutasentimentofthesamekind.Ourauthorinsistsuponseveralotherscepticaltopics;anduponthewholeconcludes,thatweassenttoourfaculties,andemployourreasononlybecausewecannothelpit.PhilosophywouldrenderusentirelyPyrrhonian,werenotnaturetoostrongforit.A27TheabovepassagecomesfromapamphletwrittenbyDavidHumetosecureareadershipforhislargelyunappreciatedTreatiseofHumanNature.Althoughnotsuccessfulinthisregard,theAbstractremainsavaluableguidetoHumesTreatise,throughofferinghisownassess-mentofthesignificanceofthatwork.Here,atleast,Humeisunequivocalindescribinghisphilosophyasverysceptical.ButevenifHumedescribeshisphilosophyinthisway,andevenif,atthetime,hisphilosophywasalmostuniversallytakeninthislight,itremainsunclear,first,whatthisskepticismamountstoand,second,howthisskepticismisrelatedtootheraspectsofhisphilosophicalprogram.Thegoalofthisessayistoanswerbothofthesequestions.IbeginbygivingabroadsketchoftheroleofskepticisminHumes209CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n210robertj.fogelinphilosophyandthen,insucceedingsections,offeradetailedanalysisofthecentralskepticalarguments.i.skepticismandbeliefOnecluetothenatureofHumesskepticismisgiveninthesen-tencethatimmediatelyfollowshisclaimthatthephilosophyfoundintheTreatiseisverysceptical,andtendstogiveusanotionoftheimperfectionsandnarrowlimitsofhumanunderstanding:Almostallreasoningistherereducedtoexperience;andthebelief,whichattendsexperience,isexplainedtobenothingbutapeculiarsen-timent,orlivelyconceptionproducedbyhabit(A27).Now,thereductionofallreasoningtoexperience(empiricism)doesnot,byitself,yieldskepticalconsequences,atleastofthestrong(Pyrrho-nian)kindreferredtoattheendofthepassage.Empiricismcanleadtoamildversionofskepticismifweinsist(perhapsincorrectly)thatknowledgemustinvolvecertainty,andthenfurtherinsist(per-hapsincorrectly)thatempiricalclaimsthatgobeyondreportsofimmediateexperiencealwaysfallshortofcertainty.Skepticismofthiskindmightbetterbecalledfallibilism,notskepticism.Infact,athorough-goingempiricisttypicallyabandonsclaimstocertaintyoverawiderangeofcaseswheremostpeoplethinktheypossesscer-tainty,buttraditionalempiricistsdidnotthinkthattheirpositionforcedawholesalesuspensionofbelief.Withanimportantexcep-tiontobenotedlater,1itisnotHumesempiricismbutprimarilyhistheoryofbeliefthatpusheshisphilosophyinthedirectionofextreme(orPyrrhonian)skepticism.Thestory,broadlysketched,isthis:acentralpartofHumesprojectofintroducingtheexperimentalmethodofreasoningintomoralsubjectsinvolvedgivinganaturalisticaccountofhowhumanbeingscometobelievecertainthingsabouttheworldthat(theysuppose)surroundsthem.Asingleexamplewillserveourpurposes.Ashumanbeingswenaturallysupposethatwearedirectlyawareofaworldthatisindependentofusandcontinuestoexistwhenwearenotawareofit.Whatisthesourceofthisbelief?Itcannotbetheresultofsoundargument,for,first,thegreatbulkofmankind1Thisoccursinhisskepticismconcerningthesenses,discussedinPartIVofthisessay.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism211iswhollyunacquaintedwithanyargumentsonthesematters.Theybelieve,butdosoinatotalabsenceofjustifyingarguments.2Fur-thermore,thoseargumentsintendedtoprovetheexistenceofanenduringexternalworldareeasilyshowntobeirreparablynogood.Thus,forHume,thecommonbeliefinanexternalworldisnotbasedonanysortofreasoningtobeginwith,andcannotbesup-portedbysoundreasoningafterthefact.ThisisonesideofHumesskepticism.AsecondsideofHumesskepticismemergeswhenhelaysbarewhathetakestobethemechanismsthatdo,infact,governtheformationofbeliefsonthesematters.ThewordinginthepassagefromtheAbstractisrevelatory:thebelief,whichattendsexperi-ence,isexplainedtobenothingbutapeculiarsentiment,orlivelyconceptionproducedbyhabit.Noristhisall.Whenwebelieveanythingofexternalexistence,orsupposeanobjecttoexistamomentafteritisnolongerperceived,thisbeliefisnothingbutasentimentofthesamekind(A27).Now,indescribingabeliefasnothingbutapeculiarsentimentproducedbyhabit,Humeisobviouslycon-trastinghispositionwiththatofotherswhoholdthattheremustbemoretobeliefformationthanthis.Thatview,crudelyput,isthatbeliefistheresultofreasoning,andsoundbeliefsaretheresultofsoundreasoning.OveragainstthisrationalistorCartesiancon-ceptionofbeliefformation,Humeholdsthatreasoning,byitself,isgenerallyincapableoffixingbeliefand,inthisparticularcase,incapableofestablishingabeliefintheexistenceofanexternalworld.TheseskepticalmotifsarefurtherdevelopedbythedetailsofHumesexplanationofhowthisfundamentalbeliefisformed.Pre-sentedwithHumescausalaccountoftheactualmechanismsthatleadustobelievethatweareawareofanindependentexternalworld,wearesimplyappalledthatourbeliefsshouldbeformedonsuchanarbitrarybasis.Furthermore,whenthisarbitrarybasisforourfun-damentalbeliefsisrevealedtous,then,foratimeatleast,belief2Humemakesthispointexplicitly:Andindeed,whateverconvincingargumentsphilosophersmayfancytheycanproducetoestablishthebeliefofobjectsinde-pendentofthemind,tisobvioustheseargumentsareknownbuttoveryfew,andthattisnotbythem,thatchildren,peasants,andthegreatestpartofmankindareinducdtoattributeobjectstosomeimpressions,anddenythemtoothers(T1.4.2.14).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n212robertj.fogelinitselfevaporates.IntheEnquiryconcerningHumanUnderstand-ing,Humedescribesskepticismgeneratedinthiswayasfollows:Thereisanotherspeciesofscepticism,consequenttoscienceandenquiry,whenmenaresupposedtohavediscovered,eithertheabsolutefallacious-nessoftheirmentalfaculties,ortheirunfitnesstoreachanyfixeddeter-minationinallthosecurioussubjectsofspeculation,aboutwhichtheyarecommonlyemployed.Evenourverysensesarebroughtintodispute,byacertainspeciesofphilosophers;andthemaximsofcommonlifearesub-jectedtothesamedoubtasthemostprofoundprinciplesorconclusionsofmetaphysicsandtheology.(EHU12.5)3FromallthisitappearsthatHumeswritingscontaintwoskep-ticalstrategies.Thefirstwemightcalltheargumentativestrategy;thesecondthegeneticstrategy.Whenusingtheargumentativestrat-egy,Humeadoptsthecommonskepticalployofpresentingargu-mentsintendedtoshowthatsomeclassofbeliefsisnotcapableofrationaljustification.Inthisclasswefindmanyoftheendur-ingfeaturesofHumesphilosophy,4mostimportantly,hisskepti-cismconcerninginduction,hisskepticismconcerningtheexternalworld(T1.4.2;EHU12,Part2),and,moreexotically,hisskepticismwithregardtoreason(T1.4.1).HiscriticismoftheargumentfromdesignfoundintheDialoguesconcerningNaturalReligionandhisexaminationofargumentsinvolvingmiraclesfoundinSection10oftheEnquirycanalsobeplacedinthiscategoryofargumentativeskepticism.WhatIhavecalledHumesgeneticstrategyreflectshisideaofaskepticismthatisconsequentonscienceandenquiry.Asystemofbeliefscanbediscreditedbyrevealingitsdisreputableprovenance.Thus,inhisdiscussionofScepticismwithregardtothesenses,Humeoffersadetailedaccountofthemannerinwhichfictionsare3HumesEnquiryconcerningthePrinciplesofMoralsisnotdiscussedinthisessay,althoughitsopeningtwosectionsdodiscussskepticalmotifs.IdiscussHumeseth-icalskepticisminHumesSkepticismintheTreatiseofHumanNature(London:Routledge&KeganPaul,1985).ThereareexcellentdiscussionsofHumesethicalskepticisminJ.L.Mackie,Ethics:InventingRightandWrong(Harmondsworth:Penguin,1977);HumesMoralTheory(London:Routledge&KeganPaul1980);andDavidFateNorton,DavidHume:Common-senseMoralist,ScepticalMetaphysi-cian(Princeton:PrincetonUniversityPress,1982).4ThisisatleastforeshadowedinTreatise1.3,thenstatedexplicitlyintheAbstract,andinSection4oftheEnquiry.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism213piledonfictionsinawaythatleadsustoadoptwhathecallssoextraordinaryanopinionthattheobjectsofourawareness(which,forHume,areperceptions)canenjoyacontinuedanddistinctexis-tence(T1.4.2.20).Here,then,isadoublemovementinthedevel-opmentofHumesskepticalposition.First,reasoningshowsusthatourbeliefinanexternalworldisnotbasedonsoundargument,fornosuchsoundargumentonthismatterexists,and,second,whenempiricalinvestigationlaysbaretheactualmechanismsthatleadustoembracethisbelief,weareimmediatelystruckbytheirinadequacy.Thiscontrastbetweenargument-basedandgenetic-basedskepti-cismhasanotherside.If,asisnottrue,ourmostgeneralbeliefsabouttheworldrestedonarguments,thensoundskepticalarguments,onceencountered,woulddepriveusofthesebeliefs.Butthisdoesnothappen.Skepticalargumentsmayconfoundusforthemoment,butlacklastingeffects.Humemakesthispointnicelyincomment-ingonthenatureandforceofsomeofBerkeleysarguments:Butthatallhisarguments,thoughotherwiseintended,are,inreality,merelysceptical,appearsfromthis,thattheyadmitofnoanswerandproducenoconviction.Theironlyeffectistocausethatmomen-taryamazementandirresolutionandconfusion,whichistheresultofscepticism(EHU12.15,n.32).Ithinkthatwecannowunderstandwhy,onHumesterms,skep-ticalargumentsproducenoconviction.Anexaminationoftheactualmechanismsofbeliefformationshowsthatbeliefsarerarelybasedonratiocination.Forthisreason,askepticalargument,evenifcorrect,removesnothingthatpreviouslysupportedbeliefs.Itismoredeeplydisturbingtocomefacetofacewiththeactualmech-anismsthatdogeneratebeliefs,forthenwecannothelpbutbeingstruckbytheirinadequacy.Withthesemechanismsexplicitlydis-playedbeforeus,wedo,infact,findourselvesinastateofradicaldoubt.Butasourthoughtsreturntothecommonconcernsoflife,theauthorityofthesenormalmechanismsisrestored,andwefindourselvesbelievinglargelyaswehadbeforewebeganourinquiries.Thisisoursoledefenseagainstradicalskepticism,forphilosophy,asHumetellsusintheAbstract,wouldrenderusentirelyPyrrho-nian,werenotnaturetoostrongforit(A27).Theironyisthatthewaysofnature,whenrevealed,hardlyfilluswithconfidenceorwithasenseofhumandignity.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n214robertj.fogelinii.hume’sinductiveskepticism1.Fromhislifetimedowntothepresent,noaspectofHumesphi-losophyhasattractedmoreattentionthanthethingsthathesaysabouttherelatednotionsofcausality,necessity,andinduction.Thelimitationsofthepresentessayprecludeacloseexaminationofhisimportantideasonnecessityandcausality,5but,verybroadly,forHume,causalityandinductivereasoningarerelatedinthefollowingway.InboththeTreatiseandtheEnquiryHumearguesthatcausalconnectionscannotbeestablishedbyanyformofapriorireason-ing.Norcanacausalrelationshipbeascertainedthroughimmediateexperience,forinspectionofthecauserevealsnoconnectinglinkbetweenitanditseffect.Simplifying,itisonlyourexperienceofaconstantconjunctionbetweentwosortsofeventsthatleadsustosupposethatoneisthecauseoftheother.Wereachtheproblemofinductionbyraisingthefollowingquestion:Howdoestheexpe-rienceofeventsbeingconsistentlyconjoinedinthepastlicenseaninferencetotheclaimthattheywillcontinuetobesoconjoinedinthefuture?That,asitturnsout,raisesaquestionthatprovesverydifficulttoanswer.InHumeswords:Butifwestillcarryonoursiftinghumour,andask,Whatisthefoundationofallconclusionsfromexperience?thisimpliesanewquestion,whichmaybeofmoredifficultsolutionandexplication(EHU4.14).HumeposeshisdifficultquestionthreetimesfirstintheTrea-tise,thenintheAbstract,andfinallyintheEnquiryandthoughthereareimportantdifferencesindetail,thebasicmoveisthesameineach.Ourrelianceonpastexperiencerests,hetellsus,ontheprinciplethatthoseinstances,ofwhichwehavehadnoexperi-ence,resemblethose,ofwhichwehavehadexperience,and,withrespecttothefuture,thisamountstotheassumptiontherewillnotbeachangeinthecourseofnature.Onwhatbasis,Humeasks,canwejustifythisassumption?Hisclaimandthisishiscore5MyHumesSkepticism,Chap.4,providesadetaileddiscussionofHumestreat-mentoftheinterrelatednotionsofcausality,necessity,andinduction.DetailedexaminationsofHumesdefinitionofcausalitycanbefoundinTomL.BeauchampandAlexanderRosenberg,HumeandtheProblemofCausation(NewYork:OxfordUniversityPress,1981),andinJ.L.Mackie,CausesandConditions,AmericanPhilosophicalQuarterly2(1965):24564,andTheCementoftheUniverse(Oxford:ClarendonPress,1974).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism215thesisisthatnoargumentcanjustifythisassumption.Therecanbenodemonstrativeargumenttoproveitforitisatleastconceiv-ablethatthecourseofnaturemightchange:whatisconceivableispossible;whatispossiblecannotbedemonstratedtobefalse;there-fore,itcannotbedemonstratedthatthecourseofnaturewillnotchange(T1.3.6.5).ForHume,theonlyalternativetodemonstrativereasoningisrea-soninginvolvingprobability.IntheTreatiseHumedismissesthisalternativequickly,andsomewhatobscurely:probabilityisfoundedonthepresumptionofaresemblancebetwixtthoseobjects,ofwhichwehavehadexperience,andthose,ofwhichwehavehadnone;andthereforetisimpossiblethispresumptioncanarisefromproba-bility.Thesameprinciplecannotbeboththecauseandeffectofanother;andthisis,perhaps,theonlypropositionconcerningthatrelation,whichiseitherintuitivelyordemonstrativelycertain.(T1.3.6.7)Byprobability(asopposedtodemonstration),Humeseemstomeananyformofinductivereasoningbasedonpastexperience.Hisbasicpoint,whichheputsratherquaintly,isthatsuchreasoningitselfpresupposesthatthecourseofnaturewillnotchange,andthuscannotbeused,withoutcircularity,toproveit.HumemakesthispointmorecleanlyintheAbstract,wherehetellsusthatitisnotpossibletoprovebyanyprobablearguments,thatthefuturemustbeconformabletothepast.Allprobableargumentsarebuiltonthesupposition,thatthereisthisconformitybetwixtthefutureandthepast,andthereforecanneverproveit(A14).Giventhatneitherdemonstrativenorprobableargumentscanprovethatthefuturemustbeconformabletothepast,itseemsthatnothingcouldprovethis.66JanetBroughtonhasarguedthattheTreatisedoesnotcontainaskepticalargu-mentconcerninginduction,butthatHumeissimplyarguingthatitispastexpe-rienceplustheimagination,ratherthanpastexperienceplusreason,thatcauseustoprojectpastregularitiesintothefuture.SeeHumesSkepticismaboutCasualInferences,PacificPhilosophicalQuarterly64(1983):318.IthinkthatBroughtonisrightinidentifyingthiscontestbetweenthefacultiesofimaginationandreasonasthemajorthemeofTreatise1.3.Ithinksheisalsorighttosaythattheexami-nationofthecausesofourcausalreasoningisthecentralthemeofTreatise1.3.6.Allthesame,itseemstomethatpartofHumesattackagainsttheclaimsofrea-sonisthatreasoncannotaccountforourtendencytoprojectpastregularitiesintothefuturesimplybecausenoargumentderivedfromreasoncanjustifysuchpro-jections.HumesskepticismconcerninginductionispresentintheTreatise,butCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n216robertj.fogelinHumesbasicargumentforinductiveskepticismgetsitsmostelaboratestatementintheEnquiry.IntheAbstracthebroughtthecoreargumentintosharpfocus;intheEnquiryhemadeitacenter-pieceofhisphilosophy.TheargumentintheEnquiryhasthesameunderlyingstructureasthoseintheTreatiseandtheAbstract,butitemploysanessentiallynewargumentativedevice:thedistinctionbetweenrelationsofideasandmattersoffactforargumentativepurposes.7Presentedwithanyclaimonanysubject,wecanalwaysask:canthisclaimbeestablishedasarelationofideas?Ifnot,canitbeestablishedasamatteroffact?Ifitstruthcanbeestablishedinneitherway,thenitstruthcannotbeestablishedatall.AntonyFlewcallsthisargumentativedeviceHumesFork.8UnfortunatelyHumesdistinctionbetweenrelationsofideasandmattersoffactraisesproblemsofitsown.First,thedistinctionishastily,andperhapsincoherently,drawn.Second,byrestinghisargu-mentonthisdistinction,Humeopenshimselftoseriousobjectionsconcerningthedistinctionitself,objectionsthatdonotbeardirectlyontheproblemofinduction.Iwilltakeupthesepointsoneatatime.Humeintroduceshisdistinctionbetweenrelationsofideasandmattersoffactasfollows:alltheobjectsofhumanreasonorenquirymaynaturallybedividedintotwokinds,towit,RelationsofIdeasandMattersofFact.OfthefirstkindarethesciencesofGeometry,Algebra,andArithmetic;andinshort,everyaffirma-tion,whichiseitherintuitivelyordemonstrativelycertain....Propositionsofthiskindarediscoverablebythemereoperationofthought,withoutdependenceonwhatisanywhereexistentintheuniverse....Mattersoffact,whicharethesecondobjectsofhumanreason,arenotascertainedinthesamemanner;norisourevidenceoftheirtruth,howevergreat,ofalikenaturewiththeforegoing.Thecontraryofeverymatteroffactisstillpossible;becauseitcanneverimplyacontradiction,andisconceivedbythemindwiththesamefacilityanddistinctness,asifeversoconformabletoreality.(EHU4.12)deeplyembeddedinalargerprogram.Incontrast,inboththeAbstractandtheEnquiryconcerningHumanUnderstandingthisargumentisgivenprominenceasafreestandingphilosophicalmove.7ThedistinctionbetweenrelationsofideasandmattersoffactappearsintheTreatiseaswell,butitsuseasadialecticalweaponemergesclearlyonlyintheEnquiry.8AntonyFlew,HumesPhilosophyofBelief(London:Routledge&KeganPaul,1961),53.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism217Thoughitisnotspelledoutfully,inthispassageHumedividesrela-tionsofideasandmattersoffactalongtwolines:onelogical,theotherepistemological.Hiscriterionforarelationofideasisepiste-mological,thatis,thecriterionisdrawnintermsofhowwecometoknowsuchrelations:Propositionsofthiskindarediscoverablebythemereoperationofthought.Thusstatementsexpressingrela-tionsofideascanbeknowntobetrueapriori.9Incontrast,Humescriterionforamatteroffactisbothepistemologicalandlogical.First,mattersoffactdifferepistemicallyfromrelationsofideasinthattheyarenotascertainedinthesamemanner;thatis,theyarenotascertainedbythemereoperationsofthought.Second,theydifferlogicallyfromrelationsofideas,inthatthecontraryofeverymatteroffactisstillpossible;becauseitcanneverimplyacontra-diction.Now,givenHumesinitialclaimthatAlltheobjectsofhumanreasonorenquirymaynaturallybedividedintotwokinds,hisuseofdualcriteriafordistinguishingrelationsofideasfrommattersoffacthasthefollowingconsequence:nopropositionthatisnotarelationofideascanbeknowntobetrueapriori.Thus,intheguiseofmerelyclassifyingtheobjectsofhumanreasonorenquiry,Humehasembracedaverystrongthesiswithoutofferinganyargumentinitsbehalf.Furthermore,bybasinghisargumentonthedistinctionbetweenrelationsofideasandmattersoffact,Humehasopenedhimselftocriticismsthathavenothingtodowiththeissueathand,namely,theproblemofinduction.Inparticular,propositionsexistthatseemnottofitintoeitherofHumescategories,forexample,thatthewestwallofabuildingcannotbesimultaneouslybothentirelywhiteandentirelygreen.Ofcourse,ifpropositionsexistthatcannotbeaccom-modatedwithinHumesclassification,thenthatclassificationisnolongerexhaustive,andtheargumentativestrategyknownasHumesForkfails.Itseems,then,thatbeforewecanevaluatetheargumentintheEnquiryinbehalfofinductiveskepticism,wewillhaveto9Inpassing,wecannotethatHumeseemstoforgetthatthefalsehoodofcertainpropositionscanalsobediscoverablebythemereoperationofthought.Asaresult,ifwetakeHumesstatementliterally,itseemsthathewouldhavetoclassifyapropositionlike2+2=5asamatteroffact.AsimilarconfusionoccursinKantsclassificationofjudgmentsintheIntroductiontohisCritiqueofPureReason.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n218robertj.fogelinenterintoamoregeneralinvestigationofthekindsofpropositionsthatexistandthemethodsofjustificationappropriatetothem.Thisisanexcursionfromwhichwemightneverreturn.PerhapsthereisashorterroutebacktoHumesoriginalconcernwithinductiveskepticism.Thedrawingofadistinctionbetweenrelationsofideasandmattersoffactcanbeviewedastheargu-mentativecounterpartofatacticusedinboththeTreatiseandintheAbstract,namely,thatofholding(orjustassuming)thatallargumentsfallintotwodistinctcategories:demonstrativeandprob-able.Hume,ofcourse,canbechallengedonjustthispoint.ButwecangetbacktothecenterofHumesargumentifwerecallthathewasinterestedinthepossibilityofanargumentestablishingthetruthofquiteaspecificclaim,namely,thatthefuturemustbeconformabletothepast.Onitsface,thisseemstobeasubstan-tiveclaimabouthowthefuturewillunfoldandthusnotsome-thingsusceptibletoanyformofapriorijustification.10Further-more,thesecondpartofHumesargumentseemspersuasiveaswell:anyattemptedinductivejustificationofthisclaimwillbequestion-begging.ElsewhereIhaveclaimedthatHumeputforwardwhatIcalledano-argumentargumentconcerninginduction,namely,anargu-mentintendedtoshowthatnoargumentcouldpossiblyjustifytheclaimthatthefuturewillbeconformabletothepast.11Thetext,Ibelieve,clearlyshowsthatthiswashisintention.ItnowalsoseemscleartomethatHumesno-argumentargumentfails.InboththeTreatise/AbstractversionandtheEnquiryversionanimportantstepismissing.Intheearlyversion,weneedaproofshowingthatallargumentsmaybedividedintodemonstrativeargumentsandprobableargumentsinthesenseinwhichHumedescribesthem.Inthelaterversion,weneedaproofshowingthat:Alltheobjectsofhumanreasonorenquirymaynaturallybedividedintotwokinds,towit,RelationsofIdeasandMattersofFact.Infact,HumeseemstofaceamoredifficultproblemwiththeEnquiryversionofhis10BarryStroudemphasizesthispoint,for,amongotherthings,ithelpstoclearHumeofthechargeofbeingadeductivechauvinist,thatis,ofholdingorassumingthattheonlyformofproofisasounddeductiveargument.SeehisHume(London:Routledge&KeganPaul,1977),5667.11SeemyHumesSkepticism,Chap.4andAppendixA.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism219argumentthanwiththeearlierversions,forthereseemtobeanum-berofclearexamplesofpropositionsthatfallintoneitherofhistwocategories.ButperhapsarelatedcomplaintmightbemadeagainsttheTreatise/Abstractversionoftheargument,namely,thattherearelegitimatemodesofargumentationthatHumehasnotconsid-ered.Again,becauseHumehasnoteliminatedthispossibility,hisargumentfailsasano-argumentargument.2.InthefirstpartofthisessayIindicatedthatHumesskepticismhadtwochiefsources,onebasedonarguments,theotherbasedonaccountsofhowhumanbeingsactuallyformbeliefs.ThesecondthemewillplayacentralroleinthediscussionofHumesskepticismwithregardtoreasonandhisskepticismwithregardtothesenses,buthisaccountofhowweactuallycometoprojectpastregularitiesintothefuturehasskepticalconsequencesaswell.Asweshallsee,however,thesearenotasdramaticasthosefoundinhisdiscussionofreasonandthesenses.IntheTreatiseHumesskepticalargumentconcerninginductionisembeddedinapsychologicalaccountofthecomponentpartsofourreasoningconcerningcausesandeffects.12IntheEnquirythetwodiscussionsareneatlypartitionedintotwosuccessivesections.Section4oftheEnquiryisentitledScepticalDoubtsconcerningtheOperationsoftheUnderstanding.Itsintendedresultsareessen-tiallynegative:noargumentcanjustifyinferencefrompasttofutureexperience.Section5hasthecurioustitleScepticalSolutionoftheseDoubts.ThefollowingpassagegivessomeideaofwhatHumehasinmindinspeakingofascepticalsolution:Ifthemindbenotengagedbyargumenttomakethisstep,itmustbeinducedbysomeotherprincipleofequalweightandauthority;andthatprinciplewillpreserveitsinfluenceaslongashumannatureremainsthesame.Whatthatprincipleis,maywellbeworththepainsofenquiry(EHU5.2).PresumablyanonskepticalsolutiontothedoubtsraisedinSection4wouldbesomesortofargumentthatwouldjustifythestepinquestion.Inthatsenseofsolution,askepticalsolutionisnosolutionatall;instead,itisameredescriptionofthemechanisms12Thisdiscussionstretchesover1.3.414.Theskepticalargumentappears(oratleastisadumbrated)inT1.3.6.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n220robertj.fogelinthatleadthemindtooperateasitdoes.Thedescriptionofthesemechanismswillnotresolveskepticaldoubts,and,totheextentthattheiroperationsstrikeusasarbitrary,ourskepticaldoubtmaybeheightenedbytheirdiscovery.Whatprincipleleadsustomakethistransitiontoabeliefinamatteroffactbeyondthepresenttestimonyofthesensesgiventhatnoargumentcanvindicateit?Thisprincipleiscustomorhabit.Forwherevertherepetitionofanyparticularactoroperationpro-ducesapropensitytorenewthesameactoroperation,withoutbeingimpelledbyanyreasoningorprocessoftheunderstanding;wealwayssay,thatthispropensityistheeffectofCustom.Morespecifically,afterexperiencingtheconstantconjunctionoftwoobjects,heatandflame,forinstance,weightandsolidity,wearedeterminedbycustomalonetoexpecttheonefromtheappearanceoftheother(EHU5.5).13Arecognitionthatallourinferencesbeyondpresentorpastexperi-encederivefromthissourcemayormaynotmakeusmorescepticalconcerningthem,butthisdiscovery,attheveryleast,deflatesourintellectualpretensionsbyrevealingthatsomeofourmostimpor-tantmodesofinferencearemadeinthecompleteabsenceofrationalinsight.Asnaturehastaughtustheuseofourlimbs,withoutgivingustheknowl-edgeofthemusclesandnerves,bywhichtheyareactuated;sohassheimplantedinusaninstinct,whichcarriesforwardthethoughtinacorre-spondentcoursetothatwhichshehasestablishedamongexternalobjects;thoughweareignorantofthosepowersandforces,onwhichthisregularcourseandsuccessionofobjectstotallydepends.(EHU5.22)Wearesoconstructedthatundercertaincircumstancesourmindsirresistiblymaketransitionsfromoneideatoanother.InthisregardwedonotdifferinanyessentialwayfromanimalswhoalsolearnfromexperienceandwhoalsodosowithoutanycomprehensionoftheunderlyingmechanismsthatbringthisaboutapointthatHumedwellsoninboththeTreatiseandtheEnquiry(T1.3.14.1523;EHU9).13Humeseemstohaveforgottenliquidmercury.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism221Humegivestheargumentaniceturnbycommentingonthewonderwefeelconcerningthecomplexinstinctualendowmentpos-sessedbyanimals:Butourwonderwill,perhaps,ceaseordiminish;whenweconsider,thattheexperimentalreasoningitself,whichwepossessincommonwithbeasts,andonwhichthewholeconductoflifedepends,isnothingbutaspeciesofinstinctormechanicalpower,thatactsinusunknowntoourselves;andinitschiefoperations,isnotdirectedbyanysuchrelationsorcomparisonsofideas,asaretheproperobjectsofourintellectualfaculties.(EHU9.6)Wecanthinkofskepticismasasetofargumentsintendedtoundercutclaimsforknowledgeorevenrationalbelief.Section4,withitsskepticaldoubtsconcerningthehumanunderstanding,illustratesthisfirststrategy.Wecanalsothinkofthecentralaimofskepticismasanattempttodestroythepretensionsofreason.Sec-tion5,withitsskepticalasopposedtorationalsolutiontothesedoubts,illustratesthissecondstrategy.iii.hume’sskepticismwithregardtoreason1.ThetargetofHumesskepticismisnotsimplythewritingsofphilosophers,butthefacultiesofthemindthatgeneratethesewrit-ings.Humedoes,ofcourse,discussthephilosophicalpositionsofothers,andallusionstootherphilosophicalstandpointsoccurthroughouthiswritings,but,moreoftenthannot,suchreferencesaremadeintheserviceofdevelopinghisscienceofman.Bad,evennonsensical,philosophicalargumentsarerevelatoryoftheunderly-ingfacultiesthatgeneratethem.14AlthoughHumeisnotcarefulinhisuseofterminology,thefirstbookoftheTreatiseislargelyconcernedwithfourfaculties:under-standing,reason,thesenses,andtheimagination.Bytheunderstand-ing,Humeusuallyhasinmindreasoningfromexperience,notablycausalreasoning.Byreason,Humeusuallyhasinminddemonstra-tiveandintuitivereasoning.15Bythesenses,Humehasinmindthat14ThisisthecentralthemeofPartVofthisessay.15InanumberofplacesHumedoesnothonorthiscontrastbetweenreasonandunderstanding,butthis,atleast,ishisgeneraltendency.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n222robertj.fogelinfacultythat(seemingly)givesusinformationaboutasurroundingworld.Bytheimagination,Humehasinmindafacultythatgener-atesnewideasfromoldbymeansofprinciplesofassociation.Humesgeneralstrategyistoarguethattheoperationsofthefirstthreefac-ultiesareultimatelygroundedintheoperationsofthefourth:theimaginationor,ashesometimescallsit,thefancy.Humesstandardstrategyinfurtheringthisprojectistoproduceskepticalargumentsintendedtoshowthatbeliefsgeneratedbythefirstthreefacultiescannotbegroundedinanyformofratiocination.Hethenattemptstoshowhowtheyaregeneratedbytheinstinctivemechanismsoftheimagination.16WehavealreadyseenthisdoublestrategyatworkinHumestreatmentofourreasoningfrompastexperience,butitsmoststrikingoccurrenceappearsinthesectionoftheTreatiseenti-tledOfscepticismwithregardtoreason(T1.4.1).Humesskepticismwithregardtoreasonhasnotfaredwell.MostwritersonHumesaylittleornothingaboutit.Humedidnotrepeatitinhislaterwritings.Thisalmostuniversalneglectprobablyspringsfromoneoftwosources:(1)abeliefthatthebasicskepticalargumentisnogood,or(2)arevulsionagainstthetotalskepticismthatitwouldentailifitwerecorrect.Howeverthismaybe,itisclearintheTreatisethatHumeacceptedtheskepticalargumentheputforwardandexplicitlyembracedtheradicalskepticalconsequencesitentailed.HumesoverallargumentdependsontwosubargumentsthatIwillcall,respectively,theregressionargumentandthediminutionargument.Thiswiderargument,presentedlargelyinHumesownwords,hasthefollowingform.TheRegressionArgumenta.Ineveryjudgment,whichwecanformconcerningprob-ability,aswellasconcerningknowledge,weoughtalwaystocorrectthefirstjudgment,derivdfromthenatureoftheobject,byanotherjudgment,derivdfromthenatureoftheunderstanding.16Formoredetailsonthis,seemyHumesSkepticism,Chap.5.ThissamethemeisdevelopedbyAnnetteC.Baier,AProgressoftheSentiments:ReflectionsonHumesTreatise(Cambridge,MA:HarvardUniversityPress,1991),Chaps.23.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism223b.Asdemonstrationissubjecttothecontroulofprobability,soisprobabilityliabletoanewcorrectionbyareflexactofthemind,whereinthenatureofourunderstanding,andourreasoningfromthefirstprobabilitybecomeourobjects(T1.4.1.5).TheDiminutionArgumenta.Havingthusfoundineveryprobability,besidetheorigi-naluncertaintyinherentinthesubject,anewuncertaintyderivdfromtheweaknessofthatfaculty,whichjudges,andhavingadjustedthesetwotogether,weareobligdbyourrea-sontoaddanewdoubtderivdfromthepossibilityoferrorintheestimationwemakeofthetruthandfidelityofourfaculties.b.Nofiniteobjectcansubsistunderadecreaserepeatedininfinitum;andeventhevastestquantity,whichcanenterintohumanimagination,mustinthismannerbereducdtonothing.c.Thus,alltherulesoflogicrequireacontinualdiminu-tion,andatlastatotalextinctionofbeliefandevidence(T1.4.1.6).Ihavecalledthefirststeptheregressionargumentbecauseittellsusthatinourjudgmentswemustnotonlyattendtotheobjectunderconsideration,butwemustalsostepbackandaskthepriorquestionwhetherortowhatextentthoseproceduresweuseindealingwiththeobjectarereliable.Forexample,thatsomeonehasbeenverycarefulincastingahoroscopeshouldnotleadustotrusthispredictionsuntilwesatisfyourselvesonthepriorques-tionwhetherhoroscopescanbetrusted.Similarly,Humetellsusthatweshouldrelyonourfacultiesonlytotheextentthattheyhaveshownthemselvestobetrustworthy:Wemust,therefore,ineveryreasoningformanewjudgment,asacheckorcontroulonourfirstjudgmentorbelief;andmustenlargeourviewtocomprehendakindofhistoryofalltheinstances,whereinourunderstandinghasdeceivdus,compardwiththose,whereinitstestimonywasjustandtrue(T1.4.1.1).ForHume,allfacultiesaresubjecttothisrestraint,includingreasonthesourceofdemonstrativeandintuitiveCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n224robertj.fogelinknowledge.Evenwithreason,beforetrustingit,wemuststepbackandaskhowreliableithasproventobe.Theupshotofthis,Humetellsus,isthatallknowledgedegeneratesintoprobability;andthisprobabilityisgreaterorless,accordingtoourexperienceoftheveracityordeceitfulnessofourunderstanding,andaccordingtothesimplicityorintricacyofthequestion(T1.4.1.1).Humeishereprobablywronginsayingthatknowledgedegeneratesintoproba-bility,forthefactthattheremaybesomechancethatademonstra-tiveargumentisinvaliddoesnotchangeitintoadifferentkindofargument.17ButforHumespurposes,itmightbesufficienttoclaimthateveryclaimforknowledgeinevitablyleadsustoapriorclaimconcerningprobabilitythatmustbeansweredbeforeweassesstheknowledgeclaim.Part(b)oftheregressionargumenttellsusthatjustaseveryknowledgeclaimmustbecheckedbyregressingtoaprobabilityclaim,sotoomusteveryprobabilityclaimbecheckedagainstafur-therprobabilityclaim.Thisleadstoaninfiniteregressofprobabilityjudgmentsconcerningprobabilityjudgments,formingastackofthefollowingkind:........(4)(3)hasaprobabilityofn4.(3)(2)hasaprobabilityofn3.(2)(1)hasaprobabilityofn2.(1)17+39=56hasaprobabilityof1.Hereprobabilityclaimsarebeingnestedinsideoneanother.What(4)saysisthis:Thatthatthat17+39=56hastheprobability1hastheproba-bilityn2hastheprobabilityn3hastheprobabilityn4.Thehumanmindbucklesunderthecomplexityofsuchapropo-sition:afactthatHumewillexploitinofferingwhatmightbecalledhisskepticalsolutiontotheproblemhehereraises.AtthispointHumecouldhavemoveddirectlytoatraditionalskepticalconclusionbypointingoutthatasrationalcreatureswearecommittedtoanunstoppableregressofhigher-orderprobability17Theexpressioninvaliddemonstrativeargumentisnotasolecism.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism225assessments.That,itwouldseem,wouldbesufficientforhispur-poses.Instead,hegivesthistraditionalskepticalargumentaturnofhisownbyarguingthatwitheachascenttoahigherprobabilityassessment,thebaseproposition(17+39=56)losessomemeasureofitsprobability.Finally,sincewemustperforminfinitelymanysuchassessmentseachdiminishingtheinitialprobabilityatleasttosomeextentHumeconcludesthatalltherulesoflogicrequireacontinualdiminution,andatlastatotalextinctionofbeliefandevidence(T1.4.1.6).ThisisnottheplacetoexaminethetechnicaldetailsofHumesprobabilisticarguments,butitisimportanttonoteboththebreadthandthedepthofHumesskepticalconclusion.18Withrespecttobreadth,Humesskepticismseemsverynearlyallencompassing.Hisoriginaltargetwasdemonstrativereasoning,but,havingreduceddemonstrativereasoning,ashethought,toprobabilisticreasoning,hethenappliesthesameargumenttoprobabilisticreasoning,finallydeprivingusofallthosethingswebelieveonthatbasisaswell.Theonlythingswemaybeleftwithasobjectsofbeliefareimmediatereportsofexperience,and,perhaps,certainsimpleintuitivetruths.Thisisnotamerefallibilismacautionaryreminderthatwelackcertaintyinareaswherepeoplecommonlysupposewepossessit.IfHumesargumentiscorrect,wefindourselvesinthedeepskepticismtraditionallyassociatedwithPyrrhonism.19WhenIreflectonthenaturalfallibilityofmyjudgment,Ihavelesscon-fidenceinmyopinions,thanwhenIonlyconsidertheobjectsconcerningwhichIreason;andwhenIproceedstillfarther,toturnthescrutinyagainsteverysuccessiveestimationImakeofmyfaculties,alltherulesoflogicrequireacontinualdiminution,andatlastatotalextinctionofbeliefandevidence(T1.4.1.6).18Formoreonthis,includingreferencestootherswhohavediscussedthistopic,seemyHumesSkepticism,Chap.2,andIanHacking,HumesSpeciesofProbability,PhilosophicalStudies33(1978):2137.19AgoodintroductiontotheeffectsofPyrrhonismonearlymodernphilosophyisRichardH.Popkin,TheHistoryofScepticismfromSavonarolatoBayle(NewYork:OxfordUniversityPress,2003),andTheHighRoadtoPyrrhonism(SanDiego:AustinHillPress,1980).ForahelpfulintroductiontothecharacterofancientPyrrhonism,seeMylesBurnyeat,CantheScepticLiveHisScepticism?inDoubtandDogmatism:StudiesinHellenisticEpistemology,ed.M.Schofield,M.Burnyeat,andJ.Barnes(Oxford:ClarendonPress,1980),2053.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n226robertj.fogelin2.YetforallofHumesskepticalarguments,experienceshowsthatthistotalextinctionofbeliefdoesnottakeplace.Ofcourse,mostpeopleindeed,mostphilosophershaveneverheardofHumesskepticismwithregardtoreason.Others,whohaveheardofit,holdthatitisincorrectandarethusimmunetoitsforce.ButHume,whopropoundedtheargument,thinksthatitisirrefutablewhileyetcontinuingtobelievemanythingsonmanytopics.How,onhisownterms,isthispossible?Inresponsetothisquestion,Humeofferswhatagainamountstoascepticalsolutiontohisdoubts:sincerationalmechanismscannotsustainourbeliefs,and,indeed,leadtotheirextinction,theremustbenonrationalmecha-nismsthatdothisforus.Humedescribesthemasfollows:Ianswer,thatafterthefirstandseconddecision;astheactionofthemindbecomesforcdandunnatural,andtheideasfaintandobscure;thotheprinciplesofjudgment,andtheballancingofoppositecausesbethesameasattheverybeginning;yettheirinfluenceontheimagination,andthevigourtheyaddto,ordiminishfromthethought,isbynomeansequal....Theattentionisonthestretch:Thepostureofthemindisuneasy;andthespiritsbeingdivertedfromtheirnaturalcourse,arenotgoverndintheirmovementsbythesamelaws,atleastnottothesamedegree,aswhentheyflowintheirusualchannel.(T1.4.1.10)Thesolereasonthatwearenottotalskepticsisthatwelackthementalcapacitytopursueourreflectionstothis,theirpredeterminedend.TowardtheendoftheconcludingsectionofBook1,Humeputsthematterthisway:Wesaveourselvesfromthistotalscepticismonlybymeansofthatsingularandseeminglytrivialpropertyofthefancy,bywhichweenterwithdifficultyintoremoteviewsofthings,andarenotabletoaccompanythemwithsosensibleanimpression,aswedothose,whicharemoreeasyandnatural(T1.4.7.7).Thusitistheweaknessofthemind,notitsstrength,thatsavesreasonfromtheskepticaldestinyimplicitinit.iv.hume’sskepticismwithregardtothesenses1.Humesexaminationofthesensesbeginswithacomparisonbetweentheskepticalproblemconcerningreasonandtheskepticalproblemsconcerningthesenses:Thusthescepticstillcontinuestoreasonandbelieve,eventhoheasserts,thathecannotdefendCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism227hisreasonbyreason;andbythesamerulehemustassenttotheprincipleconcerningtheexistenceofbody,thohecannotpretendbyanyargumentsofphilosophytomaintainitsveracity(T1.4.2.1).Humeholdsthatthereareunanswerableskepticalargumentsagainstthepretensionsofbothofthesefaculties,buthismodeofexpositionisdifferentinthetwocases.Aswehavejustseen,inhisdiscussionofskepticismwithregardtoreason,Humebeginsbystatinghisskepticalargumentandthen,verybriefly,describesthosenonrationalmechanismsthatpreservebeliefdespitetheexistenceofacontraryskepticalargument.20Inhisdiscussionofthesenses,Humereversesthisorder.Hebeginsbymerelyalludingtoaskepticalargumentconcerningthesensesandthenannouncesthathismaintaskwillbetoexaminethecauseswhichinduceustobelieveintheexistenceofbody(T1.4.2.2).Whatfollowsisalong,complex,andratherperplexingexaminationofthosecausalmechanismsthatleadhumanbeingstoadoptthefalsebeliefthatourinnerperceptionscanenjoyanexistencedistinctfromourminds,andcancontinuetoexistevenwhenunperceived.Thestandardskepticalargumentconcerningtheexternalworldappearsonlyafterthiscausalaccountofthecommonbeliefiscompleted.Theskepticalargument,whenitdoesappearintheTreatise,hastwoparts.Thefirstisintendedtoshowthatourperceptions[thosethings,thatis,ofwhichweareaware]arenotpossestofanyinde-pendentexistence.HereHumeusesstandardargumentsfromper-ceptualvariability:Whenwepressoneeyewithafinger,weimmediatelyperceivealltheobjectstobecomedouble,andonehalfofthemtoberemovdfromtheircommonandnaturalposition.Butaswedonotattributeacontinudexistencetoboththeseperceptions,andastheyarebothofthesamenature,weclearlyperceive,thatallourperceptionsaredependentonourorgans,andthedis-positionofournervesandanimalspirits.(T1.4.2.45)Convinced,perhapswrongly,thatweareonlyawareofourownprivateperceptions,thephilosopherstepsinandsuggeststhatsomeoftheseperceptionsareimagesorrepresentationsofexternalobjects.Thistheory,sometimescalledrepresentationalrealism,holdsthat20Later,intheEnquiry,Humeadoptedthissameorderofexpositionindiscussinginductiveinference.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n228robertj.fogelinwearenotdirectlyawareofexternalobjects,butweareawareofperceptionsthatserveastheirrepresentations.HereHumespeaksoftheopinionofadoubleexistenceandrepresentation,aviewheobviouslyassociateswiththephilosophyofJohnLocke(T1.4.2.31).ThesecondstepinHumesskepticalargumentisaimedatsuchdoubleexistencetheoriesandisintendedtoshowthatnoargumentcanestablishtheexistenceofexternalobjectsresemblingourper-ceptions.IntheTreatise,Humestatesthebasicargumentinonlyafewsentences:Theonlyconclusionwecandrawfromtheexistenceofonethingtothatofanother,isbymeansoftherelationofcauseandeffect,whichshows,thatthereisaconnexionbetwixtthem,andthattheexistenceofoneisdependentonthatoftheother.Theideaofthisrelationisderivdfrompastexperience,bywhichwefind,thattwobeingsareconstantlyconjoindtogether,andarealwayspresentatoncetothemind.Butasnobeingsareeverpresenttothemindbutperceptions;itfollowsthatwemayobserveaconjunctionorarelationofcauseandeffectbetwixtdifferentperceptions,butcanneverobserveitbetwixtperceptionsandobjects.Tisimpossible,therefore,thatfromtheexistenceoranyofthequalitiesoftheformer,wecaneverformanyconclusionconcerningtheexistenceofthelatter,oreversatisfyourreasoninthisparticular.(T1.4.2.47)ThecounterpartargumentintheEnquiryisequallysuccinct:Itisaquestionoffact,whethertheperceptionsofthesensesbeproducedbyexternalobjects,resemblingthem:Howshallthisquestionbedetermined?Byexperiencesurely;asallotherquestionsofalikenature.Buthereexpe-rienceis,andmustbeentirelysilent.Themindhasneveranythingpresenttoitbuttheperceptions,andcannotpossiblyreachanyexperienceoftheirconnexionwithobjects.Thesuppositionofsuchaconnexionis,therefore,withoutanyfoundationinreasoning.(EHU12.12)Humethoughtthatthisskepticalargumentwascompletelyunan-swerable,tellingusthatThisisatopic,therefore,inwhichthepro-founderandmorephilosophicalscepticswillalwaystriumph,whentheyendeavourtointroduceanuniversaldoubtintoallsubjectsofhumanknowledgeandenquiry(EHU12.14).Itseemsthatsoundreasoningleadsustoabandonournaivebeliefinadirectawarenessofanexternalworld,andthenfurthersoundreasoningleadsustoaskepticismthatcastsdoubtontheveryexistenceofsuchaworld.ForHume,thingsgetworsethemorewereason.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism229Thereisanotherimportantsideofthestorythatcannotbepur-suedindetailhere.WehaveseenbeforethatHumesskepticismisstrengthenedbyhisaccountoftheactualmechanismsthatfixbelief.IntheTreatise,thoughnotasmuchintheEnquiry,Humeempha-sizesthesheerarbitrarinessofthementalmechanismsthatleadustobelieve,quitefalsely,thatwearedirectlyawareofanexternalworld.Theupshotofthismove,combinedwiththeskepticalargu-mentthatforeclosesanyhelpfromthephilosopher,isoneofHumesdeepestexpressionsofskepticism:Havingthusgivenanaccountofallthesystemsbothpopularandphilo-sophical,withregardtoexternalexistences,Icannotforbeargivingventtoacertainsentiment,whicharisesuponreviewingthosesystems.Ibegunthissubjectwithpremising,thatweoughttohaveanimplicitfaithinoursenses,andthatthiswoudbetheconclusion,Ishouddrawfromthewholeofmyreasoning.Buttobeingenuous,Ifeelmyselfatpresentofaquitecon-trarysentiment,andammoreinclindtoreposenofaithatallinmysenses,orratherimagination,thantoplaceinitsuchanimplicitconfidence.Ican-notconceivehowsuchtrivialqualitiesofthefancy,conductedbysuchfalsesuppositions,caneverleadtoanysolidandrationalsystem.(T1.4.2.56)2.Theliteratureontheso-calledproblemoftheexternalworldonthequestionsWhethertheexternalworldexists,andHowonewouldprovethatitdoesistoolargetosummarizeinanydetail.HerethereisonlyroomtonotethattheformofHumesargumentsuggeststwopossiblestrategiesforrespondingtothesequestions.First,wecanattempttoblocktheargumentintendedtoshowthatweareonlydirectlyawareofourownperceptionsandnotdirectlyawareofexternalobjects;second,wecangrantthismuchoftheargu-mentandthenattempttofindsomeformofinferencethatwilltakeusfrombeliefsconcerningourprivateperceptionstowell-foundedbeliefsconcerningobjectsexternaltothem.ThefirststrategywaschampionedinHumesdaybyThomasReidandinthelastcen-turymostnotablybyJ.L.Austin;thesecondisthemoretraditionalwayofrespondingtoskepticismconcerningtheexternalworld.2121SeeEssayIinThomasReid,EssaysontheIntellectualPowersofMan,ed.D.R.BrookesandK.Haakonssen(UniversityPark:PennStateUniversityPress,2002),aworkfirstpublishedin1785;andJ.L.Austin,SenseandSensibilia(Oxford:Claren-donPress,1963).AmongHumespredecessors,LockeandDescartesattemptedthesecondstrategy.SincetheearlytwentiethcenturyithasbeenadoptedbyCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n230robertj.fogelinAlthoughtremendousefforthasbeenexpendedonthesematters,noconsensushasemergedthateitheroftheseapproachesissuccessfulinmeetingHumeschallenge.v.thewordsofphilosophersIntheprevioussectionwesawthatHumepresentedaskepticalchallengetothosewhoheldwhathecalledadoubleexistencetheoryofperception.22HumecannotbecreditedwithagreatdealoforiginalityinpresentingthischallengeforhereheislargelycastingintohisownvocabularyargumentsfoundinthewritingofGeorgeBerkeley.23Butinthissamecontext,Humemakesamovethatseemstobeentirelyoriginalandofgreatimportanceforunderstandinghisattitudetowardphilosophicalreflection.AccordingtoHumeandsurelyhewasrightinthisthethe-oryofdoubleexistence(orrepresentationalrealism)wasintroducedbyphilosophersasareplacementforthenaivelyrealisticviewofperceptionheldbytheplainman.WehavealreadyseenthatHumerejectedthisreplacementsinceitwassubjectorsohethoughttoadecisiveskepticalrefutation.HereHumefollowsBerkeley.ButHumegoesbeyondBerkeleyinmakingthefollowingfurthercrit-icismofdoubleexistencetheories:Therearenoprincipleseitheroftheunderstandingorfancy,whichleadusdirectlytoembracethisopinionofthedoubleexistenceofperceptionsandobjects,norcanwearriveatitbutbypassingthrothecommonhypothesisoftheidentityandcontinuanceofourinterruptedperceptions(T1.4.2.46,emphasisadded).Humesremarkablesuggestionisthatthetheoryofdoubleexistenceisnotsimplytheresultofrationalmanyphilosophers,includingBertrandRussellinProblemsofPhilosophy(London:OxfordUniversityPress,1912).22MuchofwhatissaidinthissectionisstatedinmoredetailinmyHumesSkepti-cism,Chap.7.23See,forexample,BerkeleysPrinciplesofHumanKnowledge,PartI,Sects.915,andthefirstdialogueofThreeDialoguesbetweenHylasandPhilonous.InTreatise1.4.4,Ofthemodernphilosophy,HumefollowsBerkeleyinderivingskepticalconsequencesfromthedistinctionbetweenprimaryandsecondaryqualities,buthedoesnotsayso.OnBerkeleyandHumeseeDavidRaynor,HumeandBerkeleysThreeDialogues,inStudiesinthePhilosophyoftheScottishEnlightenment,ed.M.A.Stewart(Oxford:ClarendonPress,1990),23150.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism231reflection,butisthecausalproductofcompetingforcesoperatinginthemind:Theimaginationtellsus,thatourresemblingperceptionshaveacontinudanduninterruptedexistence,andarenotannihilatedbytheirabsence.Reflectiontellsus,thatevenourresemblingperceptionsareinterruptedintheirexistence,anddifferentfromeachother.Thecontradictionbetwixttheseopinionsweeludebyanewfiction,whichisconformabletothehypothesesbothofreflectionandfancy,byascribingthesecontraryqualitiestodifferentexistences;theinterruptiontoperceptions,andthecontinuancetoobjects.(T1.4.2.52)Thusthedoubleexistencetheoryisnotsimplyarationalreplace-mentforthenaiveview;instead,itisapositionthatnaturallypressesitselfonphilosophersbecausethenaiveviewstillexertsaforceonthem.Thedoctrineofdoubleexistenceisnotsimplyahypothesisthatphilosophersconjureup;itissomethingtheyfindthemselvesconstrainedtobelieve.Paradoxically,thesourceofthisbeliefisthenaivepositiontheyclaimtohaveovercome:Natureisobstinate,andwillnotquitthefield,howeverstronglyattackdbyreason;andatthesametimereasonissoclearinthepoint,thatthereisnopossibilityofdisguisingher.Notbeingabletoreconcilethesetwoene-mies,weendeavourtosetourselvesateaseasmuchaspossible,bysucces-sivelygrantingtoeachwhateveritdemands,andbyfeigningadoubleexis-tence,whereeachmayfindsomething,thathasalltheconditionsitdesires.(T1.4.2.52)Humetellsasimilarstoryconcerningthetraditionalideaofsub-stance.Briefly,providedthatthechangesinanobjectaregradual,theeasytransitionfromoneperceptionofittothenextwillleadustobelievethatanobjecthasremainedtheselfsamethingeventhoughithasundergoneconsiderablealteration:Thesmoothanduninterruptedprogressofthethought...readilydeceivesthemind,andmakesusascribeanidentitytothechangeablesuccessionofconnectedqualities.Yetashiftinperspectivecanleadustothereverseopinion:Butwhenwealterourmethodofconsideringthesuccession,andinsteadoftracingitgraduallythrothesuccessivepointsoftime,surveyatonceanytwodistinctperiodsofitsduration,andcomparethedifferentconditionsofCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n232robertj.fogelinthesuccessivequalities;inthatcasethevariations,whichwereinsensiblewhentheyarosegradually,donowappearofconsequence,andseementirelytodestroytheidentity.(T1.4.3.4)Sowhetherachangingobjectwillseemtopreserveitsidentityorloseitdependsontheperspectivewetakeonit.Furthermore,sincebothperspectivesarereadilyavailableandseementirelynatural,themind,followingitsownprinciples,seemstobedriventowardwhatHumecallsacontradiction.Fromoneperspective,wearenat-urallyinclinedtoascribeidentitytoanobjectgraduallychangingovertime;fromanotherperspectiveweareinclinedtowithdrawthisascription.24Again,themindseemstobeatoddswithitself,andagainittriestoextricateitselffromthisdifficultythroughtheintroductionofafiction.Inordertoreconcilewhichcontradictionstheimaginationisapttofeignsomethingunknownandinvisible,whichitsupposestocontinuethesameunderallthesevariations;andthisunintelli-giblesomethingitcallsasubstance,ororiginalandfirstmatter(T1.4.3.4).IdonotthinkthatHumesupposesthatthisfictionofsubstance,ororiginalandfirstmatter,ispartoftheordinarypersonsconcep-tualapparatus.ForHume,ordinaryhumanbeings(thevulgarinhiseighteenth-centuryvocabulary)liveblissfullyinnocentofthefactthatthegreaterpartoftheirbeliefsiseitherfalseorunfounded.Itisthephilosopherswho,havinglosttheirinnocence,standinneedofthenotionofsubstance,ororiginalandfirstmatter.Theyneedthisnotionpreciselybecausetheycannotfullystifletheirnaturalinclinationtosupposethatchangingobjectspreservetheiridentityovertime,whileyetholdingthatweareonlyawareoffleetinginter-nalperceptions.Thenotionofsubstanceisasurrogatefortheselostbeliefsofthevulgar.Thisphilosophicalfictionofsubstancehasafeaturethatordi-naryfictionslack:strictlyspeakingitisunintelligible.Examinationrevealsthatthetermsubstance,atleastasemployedbythephiloso-pher,hasnoidea,eithersimpleorcomplex,associatedwithit.All24Humefollowsthisdiscussionoftheidentityofanobjectovertimewithaparalleldiscussionofthesimplicityorunityofanobjectatagiventime(T1.4.3.58).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism233thesame,ifwepressourenquiriesfarenough,atacertainstagewearenaturallyledtoembracethisdoctrine.Thewholesystem,there-fore,isentirelyincomprehensible,andyetisderivdfromprinciplesasnaturalasanyofthoseabove-explaind(T1.4.3.8).Buthowcanasystemthatisentirelyincomprehensiblebecomeanobjectofbelief?Whatwouldtheobjectofsuchabeliefbe?Humeanswersthesequestionsinapassagethatanticipatesdevelop-mentsintwentieth-centurylinguisticphilosophy.Sinceithasbeenneglected,itisworthcitinginitsentirety:Butasnatureseemstohaveobservdakindofjusticeandcompensationineverything,shehasnotneglectedphilosophersmorethantherestofthecre-ation;buthasreservdthemaconsolationamidstalltheirdisappointmentsandafflictions.Thisconsolationprincipallyconsistsintheirinventionofthewordsfacultyandoccultquality.Foritbeingusual,afterthefrequentuseofterms,whicharereallysignificantandintelligible,toomittheidea,whichwewoudexpressbythem,andtopreserveonlythecustom,bywhichwerecaltheideaatpleasure;soitnaturallyhappens,thatafterthefrequentuseofterms,whicharewhollyinsignificantandunintelligible,wefancythemtobeonthesamefootingwiththeprecedent,andtohaveasecretmeaning,whichwemightdiscoverbyreflection.(T1.4.3.10)Broadlyspeaking,thisisHumesassessmentofmansintellectualcondition:forthemostpart,thefundamentalbeliefsofordinarypersonsareeitherfalseorunfounded.Thephilosophersattemptstoputsomethingbetterintheirplaceiswhollyinsignificantorunintelligible.vi.hume’sskepticismTowhatextentwasHumeaskeptic?Thisquestiondoesnotadmitofadirectanswerfortworeasons.First,describingaphilosopherasaskepticcanmeanavarietyofthings,and,dependingonwhatismeant,ourassessmentofHumesskepticismcanvary.Second,andmoredeeply,Humesownphilosophicalpositionprecludesanysimpleattributionofdoctrinestohim.Concerningthefirstpoint,variousthingscanbemeantbydescrib-ingaphilosopherasaskeptic.Skepticismisoftenassociatedwithdoubtorwiththesuspensionofbelief.ThePyrrhonists,atleastasCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n234robertj.fogelinHumeunderstoodthem,recommendedsomethingclosetoatotalsuspensionofbelief.25ClearlyHumedidnotrecommendawhole-salesuspensionofbelief,forheheld,first,thatitwouldbedisastroustohumanlife,and,second,thatitisnotsomethingwearecapableofachieving.Wearenaturallydeterminedhardwired,asitweretoformcertainbeliefsincertaincircumstances.IfskepticismisequatedwithPyrrhonism(asHumeunderstoodit),thenHumewasnotaskepticbutheatseveralplacessaysthatheisaskeptic.26Skepticismcanalsobeunderstoodasacritiqueofthecapacitiesofourintellectualfaculties.Takenthisway,Humeisaradical,unre-served,unmitigatedskeptic.ThedoctrineoftheTreatiseisthatourrationalfaculties,lefttothemselves,arewhollydestructiveofbelief.Torepeatapassagealreadycited:scepticaldoubtarisesnaturallyfromaprofoundandintensereflectiononthosesubjects,italwaysencreases,thefartherwecarryourreflections,whetherinoppositionorconformitytoit(T1.4.2.57).Aswehaveseen,Humesupportsthisclaimintwoways:byproducingwhathetakestobeirrefutableskepticalargumentsandthroughdisplayingthearbitrarinessofouractual,nonrationalmodesofbeliefformation.Hisfundamentalideaisthatwearesavedfromtotalskepticismonlybecausethenon-rationalaspectsofournatureoverwhelmthedoubtsthatreasonattemptstoforceonus,butwemustnotlosesightofthefact,assomehave,thatthisisaskepticalconclusion.Thesecond,deeper,reasonitisdifficulttodecidewhether,ortowhatextent,Humewasaskepticisthathisownaccountofbeliefformationprecludessimpleascriptionsofbeliefs.Onhisowntheory,apersonsbeliefs,includingaphilosophersbeliefs,willbeafunc-tionofthelevelofinquiryatwhichtheyareformed.ThisappliestoHumesexpressionofhisownbeliefs,eveninhistheoreticalwriting.25HumesunderstandingofPyrrhonismwasprobablyhistoricallyinaccurate,butthisisratheracomplextopic,sincescholarstodaydisagreeontheproperinter-pretationofPyrrhonism.DiscussionsrevealingthecompetingviewsonthenatureofPyrrhonismaretobefoundinJonathanBarnes,TheBeliefsofaPyrrhon-ist,inProceedingsoftheCambridgePhilologicalSocietyNS28(1982):129;MylesBurnyeat,CantheScepticLiveHisScepticism;andMichaelFrede,TheSkepticsBeliefs,inEssaysinAncientPhilosophy(Minneapolis:UniversityofMinnesotaPress,1987),20122.26Inadditiontotheseveralpassagesalreadycited,see,forexample,theclosingparagraphofTreatise,Book1.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism235Inthesewritingsheoftenexpresseshimselfinamannerthatsug-geststhatheacceptsacompletelynaivenotionofperception.Therearepassagesthatsuggestacommitmenttoacausaltheoryofpercep-tionwhatHumecalledadoubleexistencetheory.Therearealsopassagesintendedtoshowtheinadequaciesofboththesepositions.Finally,therearepassagesthatreflectanear-Pyrrhoniandespairofbasinganybeliefonthesenses.WhichistherealHume?ThemostappropriateansweronHumesowntermsisthathiswritingssimplyexhibitthatpropensity,whichinclinesustobepositiveandcer-taininparticularpoints,accordingtothelight,inwhichwesurveytheminanyparticularinstant(T1.4.7.15).Thisradicalperspecti-valismhadhistoricalprecedentsperhapsinProtagoras,certainlyinSextusEmpiricusbutHumewasoneofveryfewphilosopherstounderstanditandtraceoutitsimplications.IntheclosingsectionoftheEnquiry,Humerecommendsamod-erateormitigatedskepticismasamiddlewaybetweennaiveaccep-tanceandPyrrhonism.Thismaysuggestthathewasasensiblefel-lowafterall.Hisskepticismmightbenothingmorethanaversionoffallibilism,theappropriatelycautiousattitudeofahard-workingsocialscientistattemptingtointroducetheexperimentalmethodofreasoningintomoralsubjects.Thisreadingdimin-ishesHumesgenius.Hisownaccountofhowonearrivesatamoderateormitigatedskepticismisofapiecewithhisaccountofhowotherphilosophicalpositionsemerge:theycomeintoexis-tencebecauseofaclashbetweenbruteirresistiblecommonbeliefsandphilosophicalreflectionthatshowsthesebeliefstobeground-less:[A]speciesofmitigatedscepticism,whichmaybeofadvantagetomankind,andwhichmaybethenaturalresultofthePyrrhoniandoubtsandscruples,isthelimitationofourenquiriestosuchsubjectsasarebestadaptedtothenarrowcapacityofhumanunderstanding.Theimaginationofmanisnat-urallysublime,delightedwithwhateverisremoteandextraordinary....AcorrectJudgmentobservesacontrarymethod,andavoidingalldistantandhighenquiries,confinesitselftocommonlife,andtosuchsubjectsasfallunderdailypracticeandexperience....Tobringustososalutaryadeter-mination,nothingcanbemoreserviceable,thantobeoncethoroughlyconvincedoftheforceofthePyrrhoniandoubt,andoftheimpossibility,thatanything,butthestrongpowerofnaturalinstinct,couldfreeusfromit.(EHU12.25)CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n236robertj.fogelinLikeotherphilosophicalpositions,mitigatedskepticismistheproductofaconflictbetweenphilosophicaldoubtsandinstinctualbeliefs.Here,asinothercaseswherephilosophyandinstinctclash,itisinstinct,notphilosophicalreflection,thatmaintainsbelief.suggestionsforfurtherreadingInadditiontotheworkscitedinthenotestothisessay,theeditorsrecom-mendthefollowing.27Bell,Martin,andMarieMcGinn.NaturalismandScepticism.Philosophy65(1990):399418.Broughton,Janet.HumesNaturalismandScepticism.InACompaniontoHumesTreatise,editedbyE.S.Radcliffe.Oxford:BlackwellPublishing,2008.Cummins,PhillipD.HumesDiffidentSkepticism.HumeStudies25(1999):4365.Falkenstein,Lorne.Naturalism,Normativity,andScepticisminHumesAccountofBelief.HumeStudies23(1997):2972.Fogelin,RobertJ.ADefenseofHumeonMiracles.Princeton:PrincetonUniversityPress,2003.Garrett,Don.ASmallTinctureofPyrrhonism:SkepticismandNatural-isminHumesScienceofMan.InPyrrhonianSkepticism,editedbyW.Sinnott-Armstrong.NewYork:OxfordUniversityPress,2004.6898.McCormick,Miriam.AChangeinManner:HumesScepticismintheTreatiseandtheFirstEnquiry.CanadianJournalofPhilosophy29(1999):43147.Hume,WittgensteinandtheImpactofSkepticism.HistoryofPhiloso-phyQuarterly21(2004):41734.Millican,Peter.HumesScepticalDoubtsconcerningInduction.InRead-ingHumeonHumanUnderstanding,editedbyP.Millican.Oxford:ClarendonPress,2002.10773.Morris,WilliamEdward.HumesConclusion.PhilosophicalStudies99(2000):89110.Norton,DavidFate.HowaSkepticMayLiveSkepticism.InFaith,Scepti-cism,andPersonalIdentity,editedbyJ.J.MacIntoshandH.A.Meynell.27Fortheirassistanceincompilingthislist,theEditorsgratefullyacknowledgeWilliamEdwardMorrisandDavidOwen.SeealsotherelevantsectionsofthebookslistedintheSelectedBibliographyby,amongothers,AnnetteBaier,DonaldBaxter,StephenBuckle,DonGarrett,JohnLaird,LouisLoeb,DavidFateNorton,DavidOwen,ThomasReid,NormanKempSmith,andJohnWright.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesSkepticism237Calgary:UniversityofCalgaryPress,1994.11939.ForTerencePenel-humsresponsetothisessay,seeinthesamevolume26771.Owen,David.ScepticismwithRegardtoReason.InTheCambridgeCompaniontoHumesTreatise,editedbyD.Ainslie.Cambridge:Cam-bridgeUniversityPress,forthcoming.Apre-publicationversionofthispapermaybeseenathttp://phil.web.arizona.edu/faculty/extra/dowen/dowen_swrtr.htm.HumeandtheIrrelevanceofWarrant.InEnsaiossobreHume,editedbyLiviaGuimaraes.BeloHorizonte:EditoraSegrac,2005.2358.http://phil.web.arizona.edu/faculty/extra/dowen/dowen_hiw.htm.Read,Rupert,andKennethA.Richman,eds.TheNewHumeDebate.London:Routledge,2000.Roth,Abe.Causation.InTheBlackwellGuidetoHumesTreatise,editedbyS.Traiger.Oxford:BlackwellPublishing,2006.95113.Singer,Ira.NatureBreaksDown:HumesProblematicNaturalisminTrea-tiseIiv.HumeStudies26(2000):22543.Stroud,Barry.HumesScepticism:NaturalInstinctsandPhilosophicalReflection.PhilosophicalTopics19(1991):27191.Winkler,Kenneth.HumesInductiveSkepticism.InTheEmpiricists:CriticalEssaysonLocke,Berkeley,andHume,editedbyM.Atherton.Lanham,MD:Rowan&Littlefield,1999.183212.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nterencepenelhum8HumesMoralPsychologyIn1927A.E.TaylorconcludedhisLeslieStephenLectureonDavidHumeandtheMiraculouswithajudgmentofHumesattitudetohisphilosophicalworkthathasbeenheldbymanyotherreadersofHume:Whatkindofresponseonemakestolifewill,nodoubt,forbetterorworse,dependonthesortofmanoneisforgoodorbad....Butwecanallmakeitourpurposethatourphilosophy,ifwehaveone,shallbenomereaffairofsurfaceopinions,butthegenuineexpressionofawholepersonality.BecauseIcanneverfeelthatHumesownphilosophywasthat,IhavetoowntoahauntinguncertaintywhetherHumewasreallyagreatphilosopher,oronlyaverycleverman.TaylorishereexpressinganattitudetoHumethatmanyofushavefelt:thathisphilosophydoesnotdeservetobetakentoomuchtoheart,becauseforallhisintellectualvitalityandthedisturbingchar-acterofmuchthathesays,thereisastreakoffrivolityinhimthatleadshimtofollowargumentstooutrageousconclusionswithoutseriousconsiderationoftheeffectsuchconclusionsmayhaveonthosewhoaredriventothem;andthattheloveofliteraryreputa-tionthatheopenlyexpressedwasoffargreaterpersonalimportancetohimthanphilosophicaltruth.1ThisestimateofHumeisadeeplymistakenone,anditinvolvesamisconstructionofelementsinhiswritingsandhispersonalitythathaveaverydifferentexplanation.1A.E.Taylor,DavidHumeandtheMiraculous(Cambridge:CambridgeUniversityPress,1927),534.238CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology239i.humeandhisphilosophicalsystemThereisnodoubtthatHumewriteswithalightnessoftouch,anironichumor,andadegreeofself-depreciationthatarerareamonggreatphilosophers.Heisnothardenoughtoreadforajudgmentofgreatnesstocomereadilytoourminds,infact.Heisalsoabletodealwiththeissueimmediatelybeforehimwithoutbelaboringitsconnectionswiththoseotherpartsofhissystemnotpresentlybeingconsidered;this,too,toreadersinanerawhensystembuildingisunfashionablemakesithardertosupposeheistryingtoconstructoneinthewaygreatphilosophersdo.Andnothinkerwhoissofrequentlysuccessfulintheartofphilosophicalcriticismcanescapeseemingtocarefirstandforemostaboutscoringpoints.Suchfeaturesaremosteasilyexplainedastheresultofatemperamentalimmunitytophilosophicalanxieties.Buttheevidenceisclearlyagainstthis,andanotherexplanationiscalledfor.Thelightnessisdeliberatelyassumedforphilosophi-calreasonsbysomeonewhoisnotimmunetophilosophicalanx-ieties,butknowsverywell,andsays,whatitisliketobetheirvictim.Therearetwowell-knownplaceswherehetellsusaboutthis.One,notoriginallydestinedforoureyes,istheletterhewrotetoanunnamedphysicianin1734,yetdidnot(itseems)send,butratherpreserved(KHL4).Inthisheoutlines,withremarkableacu-ity,thesymptomsofbreakdownthathehadsufferedasaresultofhisphilosophicalexertionsintheperiodpriortothecompositionoftheTreatisesymptomssuchasScurvySpotsonmyFingers,Watrynessinthemouth,andacompulsiveappetite,whichheinterpretedassignsoftheDiseaseoftheLearned.TheotheristhefamousconcludingsectionofBook1oftheTreatiseitself,wherehetellsusoftheeffectsthathisresearcheshaveuponhim.Hefancieshimselftobesomestrangeuncouthmonster,tobeinthemostdeplorableconditionimaginable,invirondwiththedeepestdark-ness,andutterlydeprivdoftheuseofeverymemberandfaculty(T1.4.7.2,8).Onbothoccasionsheseeksreleasefromtheseanxi-eties,whicharethedarkundersideoftheintellectualexhilarationthatsofrequentlyburststhroughinthetextoftheTreatise;andthisreleaseissomethinghethinkstobeavailabletohimonlyifhemakeshimselfbalancetheexcessesofhisphilosophicalreflectionswithCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n240terencepenelhumdeliberateabsorptioninbusinessorsocialactivities.Theseallowtheresourcesofhisnaturetoovercomethedebilitatingeffectsofoverindulgenceinphilosophicalreasoning.ThisclearevidenceshowsusthatHumewasnotsomeoneforwhomphilosophywasanactivityofminorconsequence,butsome-onewhosawhimselfaslikelytobethrownoffbalancebyhispredilectionforit.Sotheaffableandcorpulentgentlemanlyloiterer(touseTaylorsphrase)2whomsomeseeasthehistoricalHumeis,atmost,adeliberatelyassumedpersona,beneathwhichamuchmorecomplexandseriousrealityisatwork.Thepersonaisnotamereduplicateofthereality,butaproductofexperienceandtheory:experienceofwhatphilosophyleadstowhenpractisedinawaythatdoesviolencetoournature,andatheorythatputsphilosophyinitsproperplace.Whatsortoftheoryisit?Anytheorythatsuggestslimitsbeplacedonphilosophyitselfhasanappearanceofinconsistencyifitisitselfaphilosophicaltheory;andthefactthatHumebelongssomewhereintheskepticaltraditionmightseemtoaccentuatethisrisk.ToalargeextentHumestheoryofhumannatureisnot,inourterms,philosophical,butpsychological,eventhoughoneofitskeypur-posesistodeterminetheproperlimitsofphilosophicalthought.Hecertainlythinksthatphilosophicalactivity,properlypursued,sus-tainspersonalequilibriumandcankeepthreatstoitincheckaswhenitprotectsusfromthefarmoredangerousrisksthatarisefromsuperstition(T1.4.7.13).Buttoknowwhentopursuephilosophyandwhennot,onehastounderstandhumanneedsandweaknesses,andmakephilosophytakeaccountofthem.Humedoesnotconfusephi-losophyandpsychology,assomesuppose;buthedoesmixthem,inaspecialblendofhisown.HumeisaSocraticthinker.Hebelievesthatinordertoavoidbeingplaguedbyanxietywemustachieveself-knowledge.Thephiloso-pherstandsinneedofitasmuchashisorherfellowsdo.Socrateswouldhaveagreed;butheappearedtothinkthatself-knowledgewastocomethroughthepursuitofthedialecticalquestioninginwhichthephilosopherisexpert,andHumedoesnotthinkthis.Humethinksthathehasavailableascientificmodeofunderstandingthatilluminatesournatureforus,andthatthephilosophermustturn2Taylor,DavidHumeandtheMiraculous,53.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology241tothistosavehimself.Ournatureisintelligible;andoncewehavelearneditskeyfeatures,wecanavoidthoseinfluencesinphilosophy(andinreligion)thatwouldleadustodoviolencetoit.Theunder-standingofhumannaturethatHumeurgesonusisdifferentindeedfromthatderivingfromSocrates,atleastasPlatopresentshimtous.ii.humannature,theself,andthepassionsHumeconfidentlyproclaimstheimportanceofhistheoryofhumannatureintheintroductiontotheTreatise:Herethenistheonlyexpedient,fromwhichwecanhopeforsuccessinourphilosophicalresearches,toleavethetediouslingringmethod,whichwehavehithertofollowed,andinsteadoftakingnowandthenacastleorvillageonthefrontier,tomarchupdirectlytothecapitalorcenterofthesesciences,tohumannatureitself....Thereisnoquestionofimportance,whosedecisionisnotcomprizdinthescienceofman;andthereisnone,whichcanbedecidedwithanycertainty,beforewebecomeacquaintedwiththatscience.Inpretendingthereforetoexplaintheprinciplesofhumannature,weineffectproposeacompleatsystemofthesciences,builtonafoundationalmostentirelynew,andtheonlyoneuponwhichtheycanstandwithanysecurity.(TIntro.6)Thisisambitiouslanguage,fullycomparabletotheclaimsDescarteshadmadeinthepreviouscenturytoberebuildingallknowledgeafresh.Butthebasesthetwothinkersofferforthisrebuildingareverydifferent.ThedifferenceshelpustounderstandwhyHumehasalwayshadthereputationofbeingaspoilerratherthanabuilder,inspiteofthepositivethrustofthisprogrammaticproclamation.InDescartessreconstructionofhumanknowledge,themetaphys-icalseparationofthementalandthephysicaldictateslimitstosci-ence:scienceisaccordedtheautonomyitdeserves(andwhichthechurchhaddenieditincondemningGalileo)becauseitisconfinedinitssubjectmattertothephysicalworld;thesoulisexemptedfromitsscrutinybecauseofitssimplicity,itsfreedom,anditsself-consciousness.TheessenceofHumesreconstructionistobefoundintheinsistencethattherecanindeedbeascienceofmind,andthatitisexperimental,orobservational.ThescientificidealHumehasisoftendescribedasNewtonian,andtheevidenceforthisishisCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n242terencepenelhumproclamationofthetheoryoftheassociationofideas.ThisseemstoduplicateNewtonianexplanationinthephysicalrealm.Itdoesthisbyidentifying,first,theultimatecorpuscularunitsthatourobserva-tionofmentalliferevealstous;Humecallstheseperceptions,anddividesthemintoimpressionsandideas.Itthenprovidesaprincipleroughlycorrespondingtothatofgravitationtoaccountforthecon-stantinnermovementandchangethatcharacterizesthementallifeweareabletointrospect.Thisanaloguetogravitationisassociation,whichdeterminesoneperceptiontocallup,orleadonto,another.Inspiteofawiseandcautionarystatementthatweareonlytoregarditasagentleforce,whichcommonlyprevails(T1.1.4.1),thegravitationalanalogyisofferedwithpride,alongwithasimi-larlyNewtonianreticenceaboutwhatmayliebeneaththatgentleforce:Hereisakindofattraction,whichinthementalworldwillbefoundtohaveasextraordinaryeffectsasinthenatural,andtoshowitselfinasmanyandasvariousforms.Itseffectsareeverywhereconspicuous;butastoitscauses,theyaremostlyunknown,andmustberesolvdintooriginalqualitiesofhumannature,whichIpretendnottoexplain.(T1.1.4.6)IntheAbstract,hisownanonymousreviewoftheTreatise,inwhichwecansupposehewoulddrawparticularattentiontofeaturesoftheworkhethinkscentral,hesaysthatifanythingjustifiescallingtheauthoraninventor,itistheusehemakesoftheprincipleofassociation.PeterJoneshasarguedthattheinfluenceofNewtononHumehasbeenoverrated,andthatHumesdirectacquaintancewithNewtonswritingswasprobablylimited.3Evenifthisisso,andeventhoughthedoctrineofassociationislessprominentinhislaterwritingsthanitisintheTreatise(andeventhoughtheTreatiseitself,asweshallsee,leansheavilyonpsychologicaltheoriesthatdonotcombinewithitwithoutdifficulty),IthinkthattheimpactofsomethinglikeaNewtonianpictureofthescienceofmindlingeredinHumessystemlongafterthedetailsofassociationismceasedtointeresthim.Therearetwoplaceswherethiscanbeseenmostclearly.Oneisinhisviewoftheself.Theotherisinhisfamousclaimthatreason3PeterJones,HumesSentiments:TheirCiceronianandFrenchContext(Edinburgh:EdinburghUniversityPress,1982),1119.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology243is,andoughtonlytobe,theslaveofthepassions.Inboththeseplaceswefindourselvesattheheartofhismoralpsychology.Tosaytherecanbeascienceofthemental,asHumeseesthematter,istosaythatwhatwethink,feel,orwillcanbeexplainedastheeffectofacauseandtheinstanceofanaturallaw.Humanmindsarenotstrangersinnature,butinextricablypartofit.HumetriestodemonstratethisindetailintheTreatisebyshowinghowourbeliefsandouremotiveandconativecommitmentsarise.Theaccountsareintendedtotreatthoughtsandfeelingsandvolitions(allperceptions,inhisvocabulary)astheunitsofexplanation,andtoshowhowtheygiverisetooneanother.Thisformofexplanation,atleastnominally,givestheminditselfnoroletoplay.Ifthenever-endingchangesinthephysicalworldarealltobeexplainedintermsoftheattractionofmaterialparticlestooneanother,thereisnoroomforthesuggestionthattheworlditself,whichmerelycontainsthem,exertsaforceofitsown.Itisjusttheplacewheretheeventsbeingdescribedoccur.Similarly,ifthecourseofmymentalhistoryisdeterminedbytheassociativeattractionofmyperceptions,sothattheycauseoneanothertoarise,thereseemsnoplace,perhapsevennoclearsense,tothesuggestionthatI,themindorsoulthathasthem,canexertanyinfluenceovertheircourse.Allitdoesisincludethem.Theself,orego,ashesays,isjustakindoftheatre,whereseveralperceptionssuccessivelymaketheirappearance.ThedenialofanindependentrealselfisnotanawkwardconsequenceofHumestheoryofknowledge,whichrequiresustosaythatitisnottherebecausewecannotfinditwhenwelookforit(althoughthisistrue);itisacornerstoneofhissystem,requiredbythesupposedfactofascienceofmanconceivedinquasi-Newtonianterms.Thisscienceisdeterministic,sincementaleventsoccurasaresultoflawsthatsupposedlygovernthesequencesofsucheventsalone;andifitmentionsmindsoragentsthemselves,theseareconstruedtobemerebundles,collections,orsequencesofsuchevents.Theyarethesuccessiveperceptionsonly,thatconstitutethemind;norhavewethemostdistantnotionoftheplace,wherethesescenesarerepresented,orofthematerials,ofwhichitiscomposd(T1.4.6.4).Thisunderstandingofhumannaturestandsinsharpcontrasttoanother,whichforconvenienceIshallcalltherationalistmodel.Thisderives,historically,fromPlatosPhaedo,inwhichSocratesispresentedasteachingthatthehumansoulisnotpartofnature,butisCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n244terencepenelhumalientoit.Itcanchoosehowfaritalliesitselfwiththealienforcesofitspresentenvironment,andhowfaritassertsitsindependencefromthem.Thesealienforcesmakeinroadsonitthroughthepassionsanddesires,towhichthesoulcansayyesorno.Theimplicationofthisisthatsomeoftheelementsofourinnerlife,namely,thepassionsanddesires,arenottrulypartsofourselvesatall;whatistobeidentifiedwiththetrueselfisthereasonwhichsaysyesornotothem.ThisPlatonicviewofthesoulhastakendeeprootinourcultureinmanypopular,andsophisticated,doctrinesthatarenotovertlyascribedtohim.Thereisthecommoncontrastbetweenreasonandthepassions,acontrastthatyieldstheassumptionthatwhenoneactsfrompassiononeactsinpassivity,sothatwhatonedoesisnotfullyanactatall,andthatoneisnotfullyoneselfindoingit.Thereisthecorrelativeassumption,philosophicallyexpressedinmoderntimesintheCartesiantradition,thattheselfistobeequatedwiththerationalfaculty,andthatoneisfullyoneselfwhenitdictateswhatonebelievesandwhatonechooses.Descartesindeedcarriedthistotheextentofholdingthatonehasfullfreedomwhethertosayyesorno,notonlytothepassions,buttothepresentationsofsense,sothatwecanalwayssuspendjudgmentonthesewhenourgroundsareinconclusive.4Thistheoryistheepistemologicalaspectofthegeneralviewthattheuniquedignityofthehumansoulconsistsinitspossessionofaspecialkindoffreedomtoassentto,ortoreject,thepromptingsofthesenses,theemotions,andtheinstincts.Wecanreadilywonderwhetheralltheelementsinthisviewofourselvesarenecessarilyconnected,andevenwhethertheyareconsistent,buttheyareallpowerfullypresentinbothpopularcultureandrationalistphilosophicaltheory.Humesunderstandingofhumannatureisatoddswiththisratio-nalistpictureofitateveryimportantpoint,andheseesitsmaincontentionsasinconsistentwiththeverypossibilityofascienceofman.Soheassaultsitineverypossibleway,andinassaultingithasacquiredadestructivereputationamongphilosopherswhofeelthedignityofhumannatureandthedignityoftheirownprofessionarebothlinkedtothetruthoftherationalistpicture.OnewayHumeassaultsthatpictureisbymakingstatementsofhighshockvalue4SeehisFourthMeditation.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology245forthoseimbuedwithit.ThemostfamousishisdictumthatRea-sonis,andoughtonlytobetheslaveofthepassions,andcanneverpretendtoanyotherofficethantoserveandobeythem(T2.3.3.4).Thisisfundamentallyaninsistencethattherecanbeascienceofhumannatureinawaytherationalistpicturewould(inHumesopinion)makeimpossible.Itis,ofcourse,morethanthis:itisalsoaclaimthatwhenwelookandsee,weshallfindthathumanbeingsarecreaturesofinstinctandfeelingwhoserationalpowerscannot,oratleastshouldnot,beusedinanywayatoddswiththedominanceofinstinctandfeeling.NormanKempSmithandothershavemadeclearthatHumesthe-oryofknowledgeisitselfanapplicationofthisclaimabouthumannature.5Humeseesourmostfundamentalbeliefsasproductsofinstinct;andhethinksweareluckythattheyare.Therationalqueriesofthephilosophicalskepticwouldhavetheeffect,iftherationalistviewofthemindweretrue,ofreducingustoaconditionofchronicanxietyandindecisionthroughourinabilitytojustifytheclaimsofoursensesortheexpectationofregularityinnatureortheidentityoftheself.Theskepticisquiterightaboutwhatwecannotrationallyjustify,butisalso,fortunately,quitewrongaboutwhatweareabletodisbelieve.Skepticaldoubtsareintellectuallycorrect,butarevainorimpotentdoubts.Humeishimselfaskepticinhisestimateofthesoundnessofskepticalarguments,butsideswiththemosttruculentoftheCommonSensethinkersindenyingthattheseargumentscandisturbusformorethanbriefperiods.6Thesebriefperiods,however,areanxiousones,tobeavoidedbydistraction,socialorintellectual.HumerejectsthecontentionoftheScepticsofantiquitythattherecognitionofreasonsinabilitytosupportthecommitmentsofCommonSenseleadsofitselftoinnerpeace.Onthecontrary,ashemakesclearintheconcludingSection7ofBook1oftheTreatise,suchrecognitionwouldleadtodespairifnotovercomebytheresourcesofinstinct.Humedoesseeournatureascreative:ingeneratingourfunda-mentalbeliefs,itinvestsourperceptionswithmeaning.Butitisinstinctandnotreasonthatdoesthis.5NormanKempSmith,TheNaturalismofHume,Mind14(1905):14953,33547,andThePhilosophyofDavidHume(London:Macmillan,1941).6SeeEssay9inTerencePenelhum,ThemesinHume(Oxford:ClarendonPress,2000).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n246terencepenelhumWhyisitthatourinstinctsmanagetoinvestourperceptionswithmeaningsthataresousefulandadaptive?Humedoesnotprofesstoknow,andcontentshimselfwithanironicalsuggestionthattheremustbeapreestablishedharmonyatwork(EHU5.21).Heneversaysthelifeworldourinstinctscreateforusisoneweknowtobethetrueone.7HisviewofourbeliefsisessentiallyaDarwinianview.IturnnowtoamoredetailedaccountofthewayHumesviewofhumannatureunderlieshisaccountofourconductandourmorality,leavingasidehisepistemologywiththecommentthat,asKempSmithmadecleartous,Humesviewsontheinterrelationbetweenreasonandpassionrunparallelinthetwoareas.iii.humanchoiceandthepassionsEpistemologyhasneverhadmuchofaplaceinpopularculture.Buttherationalistunderstandingofhumannaturehasastrongholdonthecommonunderstandingofourchoices.Weprideourselvesonthesupposedfactthatweareablesometimestochoosecoursesofactionthatoverrideourpassionsanddesiresinthelightofagreatergood.Weprideourselvesonthesupposedfactthatwhenwedothis,weexercisethepowertobefreefromtheinfluencesandtemptationsthatwouldotherwisecondemnustowhatKantcalledheteronomy.Andweparticularlyprideourselvesonthesupposedfactthatweareabletopursuetheausteredemandsofdutyandsofunctionaspillarsofsocietybyputtinginclinationsaside.Humedeniesnoneoftheexperiencesonwhichthesepopularself-estimatesdepend.Wecan,anddo,choosethegoodovertheattractive,andresistmanyofthepassionsthatagitateus.Weareindeedentitledtotalkofourselvesasactingfreelyonmanysuchoccasionsaswellasequallyonthosewhenweyieldtopassions,andchoosetheattractiveratherthanthegood.Andwedo,indeed,choosemanyactionsbecausetheyareourduty,eventhoughtheydonotappealtous,andoursocietydependsforitshealthonthefactthatwedothis.Buthemaintainsthatnoneofthesefamiliarexperiencesistobeinterpretedinthewayrationalistsinterpretthem.Ishall7Thisiswhatseparateshimsoclearlyfromthecommon-senseschool.SeeDavidFateNorton,DavidHume:Common-SenseMoralist,ScepticalMetaphysician(Prince-ton:PrincetonUniversityPress,1982),Chap.5.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology247takeeachofthesethreepopularviewsinorder,andtrytoshowhowHumeoffersanalternativeaccountoftherelevantphenomena.Ibeginwiththoseoccasionswhenwepursueourgoodinthefaceofinclination.TherationalistholdsthatwhenIdothis,reasontriumphsoverpassion.Humesalternativeaccountofthisfamiliarexperiencedependsonhisanalysisofthepassions,whichhedevelopsatlengthinthelargelyneglectedsecondbookoftheTreatise.8Thetwotechnicalclassificationsthatareessentialforunder-standingHumesanalysisofconflictandchoicearehisdistinctionsbetweendirectandindirectpassions,andbetweencalmandvio-lentpassions.BothdistinctionsareintroducedinthefirstsectionofBook2(T2.1.1).Everypassion,inHumesview,isaunique,sim-plesecondaryimpression.Whatmakesitthepassionitis,ratherthansomeother,isthereforethefeltqualityithas.Moresimply,eachpassionisadistinctfeeling;andquestionsabouthowitarisesandhowitleadstootherexperiencesortoactionsareconstruedbyHumeascausalquestionstobedealtwithwithinhisNewtonianmentalscience.Incallingthemsecondaryimpressions,Humeseekstodistinguishthemfromthesensoryimpressions,whichhecallsoriginalatermindicating(hereatleast)thattheydonotoccurinusinconsequenceofpriorperceptions,asthesecondaryonesdo.Passions,then,alwaysariseinusfrommentalcauses:fromsensoryimpressions,fromideas,orfromotherpassions.Whentheyarisefromotherpassions,theydosobyassociation.Thereis,therefore,anassociationofimpressions(basedonresemblance)aswellasanassociationofideas.8WhatfollowshereisnotanattemptattheimpossiblefeatofsummarizingBook2inafewparagraphs,butmerelyanattempttoindicatethepartsofitsargumentthatareofmostimportanceinassessingHumesalternativetorationalisminmoralpsy-chology.IgiveamoredetailedtreatmentinChap.5ofPenelhum,Hume(London:Macmillan,1975),andinanessayTheIndirectPassions,Myself,andOthers,inTheCambridgeCompaniontoHumesTreatiseofHumanNature,ed.D.Ainslie(NewYork:CambridgeUniversityPress,forthcoming).ImportantaccountsofBook2aretobefoundinPall´Ardal,´PassionandValueinHumesTreatise(Edinburgh:EdinburghUniversityPress,1966);NicholasCapaldi,HumesPlaceinMoralPhi-losophy(NewYork:PeterLang1989),Chap.5;andAnnetteC.Baier,AProgressofSentiments:ReflectionsonHumesTreatise(Cambridge,MA:HarvardUniversityPress,1991),Chaps.6and7.ButthebestplaceforanyreadertogofirstinaseriousstudyofBook2isDavidNortonsEditorsIntroductiontotheOxfordPhilosophicalTextsEditionoftheTreatise(2000),I4672.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n248terencepenelhumThedistinctionbetweendirectandindirectpassionsisadistinc-tionbetweentwowaysinwhichpassionsarise.Directpassionsariseimmediatelyfromgoodorevil,frompainorpleasure.Thisseemstomeanthattheyarisewhensomethinghasgivenusplea-sureorpain,orisbelievedtoofferustheprospectofoneortheother.ThisatleastiswhathesaysattheoutsetofBook2;butwhenhediscussesthedirectpassionsinmoredetailinSection9ofPart3,headdsthatsomeofthemfrequentlyarisefromanaturalimpulseorinstinct,whichisperfectlyunaccountablearemarkthatcomesclosetomakingthemoriginalafterall.9Theindirectpas-sionsproceedfromthesameprinciples,butbytheconjunctionofotherqualities(T2.1.1.4,2.3.9.8).ThisconjunctionisdescribedinmuchdetailinParts1and2ofBook2;butthekeyelementinitisthefactthattheindirectpassionsrequireadistinctionbetweentheircausesandtheirobjects:10roughly,thequalitiesthatoccasionthemandthepersons(thatisoneselforanotherorothers)whohavethesequalities,andtowhomtheindirectpassionsaredirected.Thefun-damentalindirectpassionsarethoseofprideandhumility(thatis,shame),wheretheobjectisoneself,andloveandhatred,wheretheobjectisanotherpersonorpersons.Ineachcasethepassionarisesonlywhenweareconsciousbothofthequalitythatcausesitandofthefactthatthisqualityispossessedbyoneselforbyanotherthatis,bythepersonwhoistheobject,towhich[thepassion]isdirected(T2.1.3.4).Thedirectpassionsareamixedgroupindeed;butthecriticalfactaboutthemforpresentpurposesisthattheynotonlyincludesuch9Thisremarkisprobablyintendedtoavoidtheappearanceofpsychologicalhedo-nismthatcouldbeleftbytheearlierclassification.KempSmithandArdalhavesaid´thatthepassionsHumereferstohereshouldbeclassifiedseparatelyasprimaryratherthanasdirect.Nortondividesthepassionsintoproductiveandresponsive,theformerbeingdesiresthatmayyieldpleasuresbutarenotpromptedbythem,andassumesadistinctioninthepresentpassagebetweentheinstinctsandthepas-sionstowhichtheygiverise.TheseclassificationsamendHumesowndivisions,whicharecertainlymademoreproblematicbytheremarkIquotehereandbywhatHumesaysat2.3.3.8.IcannotexplorethesemattersherebeyondnotingthatHumesregardforthefactsofouremotionallifehaswonoutoverhisarchitectonictendencies.10Thisshouldnotbeequatedtooreadilywithwhatanalyticalphilosphersinourtimehavemeantbytheseterms.IhaveattemptedtospelloutthedifferencesinChapter5ofHume(1975)andinTheIndirectPassions,Myself,andOthers,inTheCambridgeCompaniontoHumesTreatiseofHumanNature.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology249reactiveemotionsasjoyorgriefordespair,butalsoincludesomeofthemostfundamentaldeterminantsofhumanconduct,namely,thedesires.Humedoesnotonlyincludedesiresforperceivedobjectslikeclothes,orforbodilysatisfactionslikefoodorsex,butmen-tionsthedesireofpunishmenttoourenemies,andofhappinesstoourfriendsandeventhegeneralappetitetogood,andaversiontoevil,considerdmerelyassuch(T2.3.9.8,2.3.3.8).Itdoesnotseemnaturaltowriteofdesiresaspassions,unlesstheyareveryagitat-ingandoverwhelmingones,butHumespsychologydependsonhisbeingabletocounterourresistancetohisdoingthis.Heattemptstocounteritthroughhisveryimportantdistinctionbetweencalmandviolentpassions.Whenintroducingit,Humesaysthatitiscommonforustodistinguishbetweengentleandintenseemotions,andtousethewordpassiononlyofthelatter,buthecallsthisavulgarandspeciousdivision.Oneandthesamepassioncanbebothmildandintense,thoughonanyoccasionagivenpassionwillusuallybeoneortheother.Itiscriticallyimportantthatwhenapassionhasbecomethepredominantinclinationofthesoul,itcommonlyproducesnolongeranysensibleagitation(T2.1.1.3,2.3.4.1).Wemustthereforedistinguishbetweentheviolenceofapassion,whichisamatterofitsfeltintensity,anditsstrength,whichisamatterofitsdegreeofinfluenceonourchoicesandconduct.Apassioncanbestrongbutcalm;andsuchapassionmayovercomeamoreviolentoragitatingone.Thisispresumablywhathappenswhenwechoosethegoodoverthealluringsothattheachinglongingforthedessertlosesouttothewishtostayslim,whichagitatesnotatall.Sothoseoccasionswhenwethinkourreasonhaswonoutoverpassionareactuallycaseswhereacalmpassionhasshownmorestrengththanaviolentone.11ThedoctrineofcalmpassionsisHumesmaincardinthegameagainstrationalistpsychology.Itsmaininternaldifficultyisthefactthatitrequireshimtosaythatpassionscanbeinamanner,imper-ceptible,whileclassingthemasimpressions;whenhehasearlierdistinguishedimpressionsfromideasonthebasisoftheirforceandvivacity,evenusingtheverywordviolenceindoingso(T2.1.1.3,1.1.1.1).11WeowetheclearunderstandingofHumesdistinctionbetweencalmandviolentpassionstoArdals´PassionandValueinHumesTreatise,95ff.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n250terencepenelhumHesupportshispositiveanalysisofchoicebysomefamousnega-tiveargumentsagainstrationalism.TheyaretobefoundinSection3ofPart3ofBook2,entitledOftheinfluencingmotivesofthewill.Theseargumentsareintendedtoshowthatreasonalonecanneverbeamotivetoanyactionofthewillandthatitcanneveropposepassioninthedirectionofthewill(T2.3.3.1).Humearguesforthefirstcontentionintwoways:hesaysthatreasonhastwofunctionsonly,namely,thediscoveryofrelationsofideas,asinmathematics,andthedescriptionofmattersoffact,asintheempiricalsciencesandcommonlife.12Reasonintheformerfunctionhaspracticalimportonlywhencalculationplaysaroleinempiricalinvestigation;andinitsempiricalfunctionreasoncanaffectpracticeonlybyshowingusthecausesoreffectsofobjectsthatwealreadydesireorshun.Inotherwords,itisourdesiresthatpromptustopursueorfleefromtheobjectsofourchoice.Reasonmerelyshowsuswhatleadsto,orawayfrom,thatinwhichourdesiresmakeustakeinterest.Itisneveritselfthesourceofsuchinterest.Ifreasonisthusshowntobeincapableoforiginatingourchoicesandinclinations,thenonthoseoccasionswhenwemakechoicesinoppositiontoapassion,itcannotbereasonthatmovesus:itcannotprovidethenecessarycontraryimpulseitself.Atmostitcanservesomedesireoraversionthatistherealcounterforcetothepassionthatlosesthecontest.Humetriestoclinchtheseargumentsbydrawingonafunda-mentalfeatureofhistheoryofthepassions:thattheyaresecondaryimpressions,andnotideas.Onlyideashavereferencetoanyotherobject,becausetheyarecopies,whereaspassions,asimpressions,donothaveanysuchrepresentativequality.Theycannot,there-fore,becontradictorytotruthandreason,sincesuchcontradic-tionentailsadefectinthatveryrepresentativequality.Thisself-containedness,orlackofreference,thatsupposedlycharacterizesallpassionsisafeatureofthemevenwhentheyaredesires.Humegivestheexampleofanger,whichonhisviewisadesireforharmtoanother(whatwewouldcallhostility).WhenIamangry,hesays,Iamactuallypossestwiththepassion,andinthatemotionhavenomoreareferencetoanyotherobject,thanwhenIamthirsty,orsick,ormorethanfivefoothigh(T2.3.3.5).12IamhereusingtheterminologyHumeintroducedlaterinEHU4.ItisclearthatthesamedistinctionisintendedinT2.3.3.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology251Asaconsequenceofthisapparentdenialoftheintentionalityofpassionsanddesires,Humemaintainsthattheycannotproperlybecalledunreasonable.13Thisterm,thoughoftenappliedtothem,should,hesays,beappliedonlytothejudgmentsthataccompanythem:Inshort,apassionmustbeaccompanydwithsomefalsejudgment,inordertoitsbeingunreasonable;andeventhentisnotthepassion,properlyspeaking,whichisunreasonable,butthejudg-ment(T2.3.3.6).Hencethereisnounreasonablenessinpreferringthedestructionofthewholeworldtothescratchingofmyfingerorinchoosingmytotalruin,topreventtheleastuneasinessofanIndianorpersonwhollyunknowntome,ortoprefermylessergoodtomygreater.Noneofthesepreferencesrequireanyfalsejudgments,andcouldonlybeunreasonableiftheydid.Ifweputasidetheattention-drawingrhetoric,wecanseethatHumedoesnotdenyreasonanessentialroleinhumanconduct.Itshowsushowtosatisfyourdesires,andinenablingustorecognizethatwhichwethencometowant,itcanevenpromptthem,althoughhedoesnotconcedethisexplicitly.Whatreasoncannotdoistomotivateusofitself.Itistheslaveofthepassions.Buttherearemanythingsthatwecandowiththehelpofaslavethatwecouldnotdoifwedidnothavehim,andforalltheairofparadoxwithwhichHumepronounceshistheories,hedoesnotdenythis.iv.freedomHumebelievesthatifthereistobeascienceofhumannature,ouractionsandchoicesmustshowthesamesortofregularitythatwefindinthephysicalworld.Intracingourchoicestotheworkingsof13AdefenderofHumemightsayherethatinspiteoftherhetoricheismerelyinsist-ingthatpassionsarenotthemselvesrepresentativeoftheirobjects,andthatsuchrepresentationisconfined,inhistheory,totheideasthatgeneratethepassionsoraregeneratedbythem.Histheoryentailsthattheintentionalityofapassion(itsbeingaboutapersonoraquality)isafactaboutitsidealaccompanimentsoritseffectsandisnotafeatureofthefeelinginwhichthepassionitselfconsists.Thisiswhatenableshimtoarguethatthepassionitselfisimmunetoappraisalsofitsrationality.Iargueagainstthislaterinthisessay.Forfurtherdiscussionsofthesecomplicatedquestions,seeChapter3ofDavidNorton,DavidHume:CommonSenseMoralist,ScepticalMetaphysician;Chapter7ofAnnetteBaier,AProgressofSentiments;andChapter1ofJohnBricke,MindandMorality:AnExaminationofHumesMoralPsychology(Oxford:ClarendonPress,1996).AnolderbutstillrewardingtreatmentistobefoundinRachelM.Kydd,ReasonandConductinHumesTreatise(Oxford:ClarendonPress,1946).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n252terencepenelhumthepassionsthatariseinusthroughthemechanismsofassociation,hehastriedtoshowthattheseregularitiesdoindeedgovernthem.Suchaprogramseemstoimplyadenialofthefreedomthatwethinkdistinguishesusfromotherbeings,andthatisassociatedinratio-nalisttheorywiththeassertionofthesupposedauthorityofreason.Humeseekstoshowthathishumansciencecanaccommodateourfreedomwithoutexemptinghumanchoicefromtheregularityandpredictabilitythathefindsinournatures.Hencehisphilosophicalsystemcontainsthebest-knownclassicalstatementofwhatisnowknownasCompatibilism.14Compatibilismisthethesisthatthereisnoinconsistencyinhold-ingthathumanactionsarecausedandyetarefree.Thisisalogi-calthesis,normallycombinedwiththesubstantiveclaimthatouractionsalwaysarecaused,andthattheyaresometimesfreeaswell.Ishallusethetitletocomprisethecombinationofallthreeproposi-tions.IshallusethecommontermLibertarianismtonametheviewthatitisindeedinconsistenttoholdthathumanactionscanbefreeyetalwayscaused;thatsomeofthemareindeedfree;andthatsomearetherefore,insomemanner,exemptfromcausation.HumespositionispresentedmostclearlyinSection8ofthefirstEnquiry,thoughmostofwhathesaysthereisanticipatedinSec-tions1and2ofPart3ofBook2oftheTreatise.TheTreatiseversionismoreaggressive,andintheEnquiryhedescribeshisargumentasareconcilingproject.Thisphrasemightsuggestthathethinkshispositionisfullyinaccordwithcommonsense,butitclearlyisnot,andHumedoesnotseriouslypretenditis.Whathethinksheisrec-oncilingaretheneedsofahumanscienceandtheneedsofourordi-narymoraldiscourse,andhearguesthatcommonopinionisinerroraboutthelatterneeds.Popularopinionholdsthatweneedonesortoffreedomthatwedonothave,insteadofanotherthatwedohave.IntheTreatiseheusesscholasticterminologytonamethesetwokindsoffreedom:hedistinguishesbetweenlibertyofspontaneity14HewasanticipatedbyHobbesinChap.21ofLeviathan(1655),andbyAnthonyCollinsinhisPhilosophicalInquiryConcerningHumanLiberty(1717).Collinssworkisoftenoverlooked(e.g.,bymyself),eventhoughitsexistencewaspointedoutbyT.H.HuxleyinhisHume(London:Macmillan,1886).ItisconvenientlyavailableinDeterminismandFreewill,ed.J.OHiggins(TheHague:MartinusNijhoff,1976).IhaveattemptedtoexamineHumesviewsonfreedommorefullyinChapter8ofmyThemesinHume.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology253andlibertyofindifference(T2.3.2.1).Libertyofspontaneitycon-sistsintheabsenceofhindrancestotheexecutionofonesdecisions.HedescribesitintheEnquirythus:apowerofacting,ornotact-ing,accordingtothedeterminationsofthewill;thatis,ifwechoosetoremainatrest,wemay;ifwechoosetomove,wealsomay.Heimmediatelyaddsthatitispossessedbyeveryonewhoisnotapris-onerandinchains(EHU8.23).Hethinks,correctly,thatthislastclaimisnotcontroversial.Heisalsocorrectinthinkingthatlibertyofspontaneity,sodefined,iscompatiblewithuniversalcausation;foritismerelytheabsenceofinterferencewiththeexerciseofoneschoices,nottheabsenceofcausaldeterminationinthemakingofthosechoices.Whereheiscontroversialisinwhathesaysabouttheothersortoffreedomthatwethinkwehave,butinhisviewdonothave.Wethinkthatsometimes,whenwechooseoneway,wecouldequallyhavechosenanotherway.InHumeslanguagewebelievethatsometimes,whenwechoosetoremainatrest,wemight(eventhoughwedonot)chooseinsteadtomove;andthatifwechoosetomove,wemight(eventhoughwedonot)chooseinsteadtoremainatrest.Webelieveintherealityofunexercisedpowersofchoice,andseethisrealityasessentialtoourfreedomasagents.Humecallsthissortoffreedomlibertyofindifferenceandinterpretsitasadenialoftheuniversalityofcausationinhumanaffairs,andinsistsweneitherhaveitnorneedit.Indeed,hebelievestherequirementsofourmoralthinkinganddecisionmakingareinconsistentwithitsexistence.Heattackslibertyofindifferenceinthreeways.First,heassertstheuniversalityofcausation,andtheunrealityofchance,andemphasizesthathumanaffairsdonotdifferintheserespectsfromthenaturalworld.Forexample:itisuniversallyallowed,thatnoth-ingexistswithoutacauseofitsexistence,andthatchance,whenstrictlyexamined,isamerenegativeword,andmeansnotanyrealpower,whichhas,anywhere,abeinginnature(EHU8.25).Tothisdogmaticmetaphysicalargument,headdsthatwecaninferandpredicthumanactionsfromthemotivesandcharactersofhumanagentsinawaythatisfullycomparabletoourabilitytoexplainandpredictnaturalphenomena;andwhenpeopleseemtoactinbizarreorunpredictableways,wecanpostulateanddiscoverhid-dencausesthataccountforthisagain,asweareabletodoforsurprisingphysicalevents.SowemustacknowledgenecessityinCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n254terencepenelhumhumanaffairsaswellasinphysicalnaturethistermbeingunder-stood,ashestresses,inthesamewayashehasinterpreteditinhisearlieranalysisofcausalinferences.ItisimportanttorecallthatwhenheoutlineswhathecallssomecorollariesofthatanalysisintheTreatiseheremarks,withastonishingcasualness,thatthedistinction,whichweoftenmakebetwixtpowerandtheexerciseofit,is...withoutfoundation(T1.3.14.32,34).Oneofthewaysinwhichweoftenmakethisdistinctionis,ofcourse,inthepopularascriptiontoagentsoftheunexercisedpowerofchoice.Humessecondlineofattackonlibertyofindifferenceisthemorepracticalonethatweneedpredictabilityinhumanaffairsinordertomakeourdecisions.Hegivesthemelancholyexampleoftheprisonercondemnedtothescaffold,whorecognizeshewillgetnohelpinescapingfromhisgaolerorhisguardsafterobservingtheircharacters,anddecideshewouldbebetteremployedintryingtoweakenthebarsofhiscellthanintryingtochangetheirresolution(T2.3.1.17).Themultitudeofexamplesthathumanexperienceoffersusofregularconnectionsbetweencharacterandactionwouldnotbeopentousiflibertyofindifferencewereareality.Humesthirdargumentagainstlibertyofindifferenceconsistsinrefutationsofthenatural,butinhisviewmisguided,suggestionthatwecanintrospectitsreality(T2.3.2.2).Whathesayshereparallelsthemanyimportantthingshesaysinoppositiontotheclaimthatwecandetectwithinourselvestheexperienceofthepowerthatweascribetophysicalcauses(see,forexample,EHU7.920).Humedoesnotdenytherearevolitions,assomehave;15heseesthemasareadilydetectablecomponentinthemechanismofhumanchoice.16Buthedeniesthatwecaneverdetectthattheyarethemselvessubjecttonothing(EHU8.22,n.18).Libertyofindifference,then,isamyth;butwehaveneverhadanyneedofit,andinfactpresupposeitsabsenceinpracticalreflection.Itsrealitywouldbeinconsistentwiththepossibilityofascienceofman,asconceivedbyHume.Itisimpossibleheretoexplorethequestionoftherelationshipbetweenhumanscienceanddetermin-ismthatisraisedbyHumesstance.Instead,Imentionanimportantimplicationofhisviewforhismoralpsychology.15Thebest-knowncaseisChapter3ofGilbertRyle,TheConceptofMind(London:Hutchinson,1949).16SeePenelhum,Hume(1975),11117,andBricke,MindandMorality,4959.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology255IfHumeisright,weareofteninapositiontoenactthechoiceswemake,andalsotoenactthealternativechoicesthatwedonotmake.Butweareneverinapositiontochooseinawayotherthanthewaywedochoose.Hebelievesintherealityofunexploitedopportuni-ties;butnotintherealityofunexercisedpowersofchoice.Thisentails,however,thatmoralpraiseorblamecanneverbeappliedonthegroundthatsomeonehaschosenacourseofactionthatheorsheneednothavechosen.Commonopinionfollowsrationalisminthinkingthatthisisinfactthebasisofmuchpraiseorblame;andHumemustdenyit.Hedoesinfactdenyit,andoffersanaccountofmoralvirtuethatconnectsitwiththeverypredictabilitythatheinsistswecanfindinhumanaffairs,notwiththelibertyofindifferencethathesaysdoesnotexist.v.obligationandvirtueWehaveseenthatHumetracesallchoicestothepassions,andrejectstherationalistunderstandingofhumanfreedom.Butthisleadsustowhatheseemstoseeasthemajorproblemofhismoralphilosophy.Rationalistsmightconcedethemainfeaturesofhisaccountofprudentialchoice,butstillsaythatwhenIchoosewhatIthinkisgoodformeinsteadofwhatIamnowinclinedtodesire,Iremaintheservantofmydesires.IdonotceasetoservethemwhenImerelypostponetheirsatisfactiontothefuture.Wedo,however,sometimesmanagetoactinthefaceofallourdesires,shorttermorlongterm.Wedothiswhenweactfromduty.Whenwedothis,reasondoesindeedtriumphoverpassion.Thebest-knownversionofthisviewfromHumestimeisthatofJosephButler,whoinsistsonthesupremacyofconscienceinhumannature.17Heaccordsitsupremacyoverallotherspringsofaction,includingself-love,benevolence,andparticulardesires.Humesaccountofourregardfordutyisonethatconcedesitsreal-ity,butstillderivesitfromouremotionalnaturesashisscienceofmandepictsthem.17Butleravoidscommitmentonwhetherconscienceisarationalpoweroramoralsense.SeeTheWorksofBishopButler,ed.J.H.Bernard,2vols.(London:Macmil-lan,1900),2:287.ButtheroleheascribestoitisonetowhichHumemustfindanalternativewithinhisownhumanscience.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n256terencepenelhumHisaccountdependsonaprincipleheenunciatesasanun-doubtedmaxim,namely,thatnoactioncanbevirtuous,ormorallygood,unlesstherebeinhumannaturesomemotivetoproduceit,distinctfromthesenseofitsmorality(T3.2.1.7).Herecognizesthatthisclaimlookstobeatoddswiththefactthatwesometimesactfromasenseofdutyalone.Hisattempttoshowthathisclaimcanaccommodatethisfactisattheheartofhisaccountofjustice.Wemustbeginwithhisaccountoftheroleofthepassions,orsentiments,ofapprovalanddisapproval,sinceheviewsthesenseofdutyasaderivativeofthese.Humeholdsthatmoraljudgments,inwhichwedescribebehaviorasvirtuousorvicious,expressthesesen-timents.Likeallotherpassions,theyareuniquesecondaryimpres-sions,andcannotthereforebeanalyzed;butwecansayhowtheyariseandwhattheireffectsare.Thestoryiscomplex,butwecanseeattheoutsetthatifindeedthesenseofdutyisaproductofthesenti-mentsofapprovalanddisapproval,itisaproductofsentimentsthatarisewhenwepassjudgmentonhumanbehaviorthatmustalreadybeproducedbysomethingotherthantheapprovalanddisapprovaltowhichitgivesrise.IdrawinwhatfollowsonSections1,2,and3ofPart3ofBook3oftheTreatise,andSections5through7ofthesecondEnquiry.Humemaintainsthatmoralapprovalanddisapprovalhavehumancharacters,ratherthanindividualactions,astheirobjects.Itissig-nificantthathetakesthetermsvirtuousandviciousastheparadigmsofmorallanguage,thusmakingiteasiertopersuadeusthatevaluationsaredirectedtowardpersonsratherthantheirdeeds:Ifanyactionbeeithervirtuousorvicious,tisonlyasasignofsomequalityorcharacter.Hesaysthatactionsthatdonotreflectsettledstatesofcharacterintheiragentsareneverconsiderdinmoral-ity.Reasonassistsinthegenerationofapprovalanddisapprovalbyshowingustheeffectsthatcertainstatesofcharacterhave.If,byadisinterestedexamination(anexaminationconductedwithoutref-erencetoourparticularinterest;T3.3.1.4,3.1.2.4),wefindthataparticularcharactertraitisagreeableoruseful,ordisagreeableorharmful,totheagentwhohasit,ortoothers,thenthemechanismthatgeneratesapprovalordisapprovalcancommence.Themechanismiscomplex,andinvolvestheworkingsofsym-pathy.Thispsychologicalprincipleisnottobeconfusedwiththesentimentofcompassion,whichismerelyoneofitsproducts.TheCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology257principleistheonethatenablesustoparticipateintheemotionallife,andthepleasuresandpains,ofothers.HumeintroducesitinTreatise2.1.11.18Accordingtohisaccountofitthere,Ibecomeawareofthepassionofanotherbyobservingitsmanifestationsinhisorherbehavior;Ihave,therefore,anideaofit.Sofar,however,Iamnotmovedbytheotherspassion.Forthistohappen,myideahastobeenlivened:thenitwillturnintoanimpression,andIshallhavetheverypassionIhaveinferredtobepresentintheotherperson.Humesays,tothesurpriseofthereaderswhoencounterthissoearlyinBook2,withtheirmemoriesofTreatise1.4.6intheirminds,thatwhatenlivenstheideaIhaveoftheotherspassionistheidea,orratherimpressionofmyself.HecannothererefertotheimpressionofthepureegothathewassoemphaticinBook1thathedidnothave,butmustrefertothatsuccessionofrelatedideasandimpres-sions,ofwhichwehaveanintimatememoryandconsciousness(T2.2.11.4,2.1.2.2).Thisissolivelyandvividthatitslivelinessiscommunicatedtotheideaoftheotherspassion,whichIthencometohavemyself.Itcanthenleadontootherpassionsthroughtheprincipleofassociation.Thesympatheticmechanismenablesmetoshareinthepleasuresandpainsthataretheeffects,intheagentorothers,ofthosecharac-tertraitsIamdisinterestedlysurveying.Theassociationofimpres-sionscausesmethentoexperienceapproval(whentheseeffectsarepleasant)ordisapproval(whentheyarepainful).Iexpressthesesen-timentsinmymoraljudgments,andIcallthecharactertraitsIhaveassessedinthiswayvirtuesorvices,respectively.Theirvirtuousnessorviciousnessconsistsintheircapacitytoarousethesesentimentsinobservers;butthesesentimentshavenot,ofcourse,causedthesecharactertraitstobepresentintheobservedagentsinthefirstplace.Humedescribesapprovalanddisapprovalascalmformsoftheindirectpassionsofloveandhatred(T3.3.5.1).19Loveandhatredare18SympathyseemstodropoutofsightinthesecondEnquiry,andithasbeenamatterofcontroversywhetherthisshowsHumetohaveabandoneditornot.Forthenegativeview,seetheappendixtoJ.B.Stewart,TheMoralandPoliticalPhilosophyofDavidHume(NewYork:ColumbiaUniversityPress,1963);forthepositive,seeCapaldi,HumesPlaceinMoralPhilosophy,Chap.7.WhateveritsplaceinthesecondEnquiry,itsroleintheTreatiseismuchgreaterthanIhavedescribedhere.SeealsointhisvolumetheessayHumesLaterMoralPhilosophy.19ThispassageiskeytoArdalsinterpretationoftherelationbetweenHumestheory´ofthepassionsandhismoralphilosophy.InthisessayIhavefollowedhim,asthepassageseemsquiteexplicit.Itsimportancehasbeenquestioned,however,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n258terencepenelhumcausedbythequalitiesoractionsofpersons,buthavethepersonsthemselvesastheirobjects.Approvalanddisapprovalarearousedbythequalitiesagentsdisplay,butaredirectedtowardtheagentsthemselvesasthebearersofthecharacterstheymanifest.Wehaveyettoaccountforthesenseofduty,however.Theaccountcomesintwoparts.ThefirstisHumesexplanationofhowitisthatwesometimesperformactsfromasenseofdutythatoth-ersperformfrom(say)benevolence.Hesaysthatsomeonemaybeconsciousofthefactthathelacksacharactertrait(suchaskindnesstochildren)thatcausesustoapproveofthosewhohaveit.Hemaythencometohatehimselfuponthataccountandmayperformtheactionfromacertainsenseofduty,inordertoacquirebypractice,thatvirtuousprinciple(T3.2.1.8).Onthisview,thesenseofdutyisaconscioussubstituteformorenaturalmotives,andisaproductofself-hatred.Tofeelitistofeelthedisapprovalofyourownlackofavirtuousinclination.Thesephenomenaoccur,thoughIthinkwemaydoubtwhethertheyarethekeytotheoriginofthesenseofduty.Buteveniftheyare,theydonotincludeamuchlargerrangeofcases:thoseoccasionswhenweseemwillingtoactfromdutyevenwhenthereisnopriornaturalmotive,eitherinourselvesorinothers.Thesearethecaseswhenweactfromjustice.Thereisnonaturalinclination(suchasbenevolence)toexplainourwillingnesstopayourtaxes,ortoreturnmoneywehaveborrowedfromourbankers.Yetjusticeisesteemedasavirtue,anditsdenialisjudgedvicious.Humesunderstandingofthisfactisofprimaryimportanceinhispsychologyofduty.InHumessystemjusticeisnotanaturalvirtuebutanartificialone:thatis,itisnotasettledstateofcharacterthatisduetoinnatecauseswithinus,butaconditionweacquirebecauseoftheinfluenceonusofsocialinstitutions.WedohavesomesociallyunifyingmotivesinbyDonaldAinslie,inhisessayScepticismaboutPersonsinBook2ofHumesTreatise,JournaloftheHistoryofPhilosophy27(1999):46992.IfArdalisright,´Humesaccountofobligationisadirectapplicationofhistheoryofthepassions;ifheisnot,andweneedtodistinguishmorecarefullythanthispassageseemstobetweenpersonsandtheircharactersormentalqualitiesasobjectsofapprovalanddisapproval,thegreatemphasisthatHumeplacesonhisaccountoftheindirectpassionsneedsfurtherexplanation.Ainslieofferssuchanexplanation.IinclinetothinkthattheroleAinsliegivestothesepassionsisonetheyindeedhave;butIdonotseethatthisshowsArdaltobemistaken.´CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology259ournaturalbenevolenceandloveoffamily;butthesemotivesaretoorestrictivetosustainlargesocialgroupings.Weareable,however,toseethevalueofconventionsthatwouldsafeguardsuchthingsaspropertyrights,andweadoptthemthroughanimplicitrecognitionofcommoninterests.BothintheTreatiseandinthesecondEnquiryHumeusestheanalogyofoarsmenwhorowtogetherwithoutanyexplicitmutualundertakingtodoso.Suchconventionsoftenentailinconvenienceforus,butwesustainthemthroughself-interest.Oncetheyareestablished,itiseasytounderstandhowtheyacquiretheextrastatusgiventhemthroughtheoperationofapprovalanddisapproval.Eachofusisable,throughsympathy,tobecon-sciousoftheunpleasantresultsofunjustactionsforthosewhosufferfromthem.Wemaysufferfromthemourselves.Weexpressourdis-pleasureattheseeffectsbysayingthatjustactionsareourduty,andavoidinnerdiscomfortbydoingourdutyourselves.Hencejusticebecomesvirtuouswithoutfirstbeingattractive.Humesmostsuc-cinctsummaryofhisaccountofthegenesisofthesenseofdutyisperhapsthis:Allmoralitydependsuponoursentiments;andwhenanyaction,orqualityofthemind,pleasesusafteracertainmanner,wesayitisvirtuous;andwhentheneglect,ornon-performanceofit,displeasesusafteralikemanner,wesaythatwelieunderanobligationtoperformit(T3.2.5.4).vi.humeandcommonopinionForallhiswillingnesstoexpresshimselfparadoxically,Humesmoralpsychologyisdesignedtoaccommodatethephenomenaofourdailymoralexperience,andonlytorejectarationalistinterpre-tationofthem.Hedoesnotseektooverturnthemoralconventionsofcommonsensebut,onthecontrary,seekstosupportthemanewonfoundationsofexperimentandobservation,freeofmisleadinganddisruptivetheory.20Itisthereforeimportant,inassessinghissuccessesandfailures,todeterminehowfarhisopinionsconformtocommonopinion,andhowfarnot.Ibeginwithacommentonhistheoryofobligation.Itsveryinge-nuitypresentsanimmediatedifficulty.Isitsoobviousthatthesense20IaminagreementherewiththepositioninNorton,DavidHume:CommonSenseMoralist,ScepticalMetaphysician.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n260terencepenelhumofdutyisderivative?Humeisfreeoftheworldlywisecynicismofpsychologicalegoism.InthesecondappendixtothesecondEnquiryhearguesagainstit,muchinthemannerofJosephButler,andmain-tainsthatthosewhoholdit(likeHobbes)areforcingatheoryontheobservablefactsofconduct.ButwhynotfollowButlerfurtherandsaythattheobservablefactsalsoshowwehaveanaturaltendencytofeelandactonasenseofobligation?21Thereasonisprobablytobefoundnotonlyinthedeterminationtoundermineethicalratio-nalism,butalsointheequallystrongdetermination,inHume,toavoidanytheorythatmightseemtorequire,orinvite,theologicalunderpinnings,andtoofferinsteadapurelysecularaccountofallthephenomenaheexplains.Butinseekingtoofferanexplanationofconscienceatall,insteadoftakingthefactofitasadatumashetakesbenevolencetobe,heisforcedtointerpretitasaproductoftheinstitutionsofsocialjustice,whenthelatterareprobablyregardedbymostasderivingsomeoftheirholdonusfromthepowerofoursenseofobligation,nottheotherwayabout.Thefactthatmanyotherphilosopherstrytoexplainthemasderivingfromself-interest,muchasHumedoes,putsthematoddswithcommonopinionalso.ThereisanotherplacewhereHumesaccountofmoralvirtueputshimatoddswithcommonsense,andwherehehimselfshowssignsofgreaterdiscomfortatthefact.Inhisstoryofthewayswecometofeelmoralapproval,hetellsusthatitisdirectedtowardestablishedcharactertraitsinournatures,andariseswhenwedisinterestedlyrecognizethatthesecharactertraitsareusefuloragreeabletoour-selvesorothers:thattheyhaveutility,inthelanguageofthesecondEnquiry.Thisaccountpromptsaquestion:therearemanyhumancharacteristicsthathaveutilityinthiswaythatwedelightin,butarenotobjectsofmoralapproval.Similarly,manyhumantraitsthatareharmfulordisagreeabledonotelicitmoraldisapproval.Wepraisecharm,wit,oreloquence,butnotinthemannerofbenevolence,industry,ortemperance.Whynot?HumeaddressesthispotentiallyvexingquestioninTreatise3.3.45,andinthefourthappendixtothesecondEnquiry.HetriestodismissitasnotverymaterialandinentitlingtheEnquiryappendixOfSomeVerbalDisputes21ForButlersarguments,seethefirst,second,andthirdSermonsinTheWorksofBishopButler,Vol.1,2557.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology261evincesalamentableandatypicalinclinationtodismissaseriousconceptualissueaswhatmisguidedtheoriststodaysometimescallamerequestionofsemantics.Butitisaproblem;andheshowsadegreeofrecognitionofthesortofproblemthatitisbytryingtofendoffonepossibleexpla-nationofthedistinctionwedoindeedmakebetweenvirtuesontheonehandandtalentsontheother.Thisisthesuggestionthatvirtuesarevoluntarilyacquiredandtalentsarenot.Hesays,perhapscorrectly,thatthereisnogroundformaintainingthis,andsuggestsinsteadthattherelevantconsiderationisthatvirtues(andvices)canbechangedbylawsandbyeducation,whereastalentscannot.Thisisinteresting,butseemswrong:onethinksoftheworkofremediallanguageinstructors,long-sufferingpianoteachers,orphysiothera-pists,whoallseemtobeinthethanklessbutnotwhollyineffectualbusinessofmodifyingourtalentsbytraining.What,then,isthegroundofourdistinction?Wecanapproachitbynoticingthatinordertoassimilatetalentstovirtues,Humehastoassumethatthetalentsareusedwellorwisely.Avirtuecannot(necessarilycannot)beusedbadlybyitspossessor,butatalentcan.22Avirtueis,inpart,thepredictabletendencytousesometalentwell,ratherthanbadly.Butusingatalentwellinvolvesusingitattherighttimesandnotusingitatthewrongtimes.Wepraisesomeonewhocanbepredictedtodothis(bycallinghimvirtuous),becausethatpersonchoosestousethattalentwhenitisgoodto,andnottouseitwhenitwouldbebadto.Heorsheispraiseworthybecausetheyuseitingoodwayswhentheycoulduseitinbadwaysinstead.Wepraisethepredictabilityofvirtuousactionpreciselybecausewethinkitcouldbedoneotherwise.OnHumesviewoffreedomthisiswhatwecanneversayaboutanyoneschoices.Humesscienceofhumannature,then,seemstohavetheunattractiveconsequencethatweaccordmoralapprovalanddis-approvaltopatternsofchoicethatcouldnotbeotherthanwhattheyare.Agoodcharacterisjustapieceofgoodfortune.Whilepop-ularethicalthinkingisfrequentlyforcedtogiveeartothisview,itisstillseenasparadoxical.Goodcharacteris,forthemostpart,still22OnerecallsherethedefinitionofavirtueinAquinas:agooddispositionofthemind,bywhichweliverighteously,andofwhichnoonecanmakebaduse.SummaTheologiae1a2ae55,4.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n262terencepenelhumregardedastheregulartendencytomakefreechoicesthataregood,notmerelytoperformpleasingactshabitually.ThisbringsustothebedrockofHumesunderstandingofwhatascienceofhumannaturehastobelike.Ihavesuggestedthatthecommondistinctionbetweenvirtuesandtalents,whichhefindsasourceofdifficulty,existsbecausethepopularascriptionofvirtuetosomeoneinvolvesascribingsomedegreeofwhatHumecallslibertyofindifferencetothatperson.ButHumewouldrespondthatthisentailsthedenialoftheverypredictabilityofhumanconductthatourethicalthinkingrequires,andisinconsistentwiththescientificstatusofthestudyofmankind.CriticsofalibertarianturnofmindwouldsaythatHumesdifficultiesmerelyshowwemustjettisontheNewtonianmodelofthehumansciences.Wemust,theywouldsay,acceptthatthesocialsciencesareabletopredicthumanbehavior(suchasvotingpatterns)aswellastheydobecause,infact,mostpeopledochooseinroughlythesamewaysinsimilarsituations,eventhoughtheycould,iftheychose,notdoso.Butsomepeopledo,nowandthen,surpriseus(whentheycouldhavechosennotto!)andwehavetobecontentwithstatisticalpredictionsinconsequence.SofarwehavefoundaspectsofHumesmoralpsychologythatareatoddswithcommonopinioninwaysthatseeminevitableconse-quencesofhisunderstandingofthescienceofhumannature.Thereisanotherwell-knownclaimthathemakesthatisindeedatoddswithcommonopinion,butinawayhecouldhaveavoided.Thisishisclaimthaterroneousorbizarreemotionsarenotcontrarytoreason.Herecognizesthattheunderstandingcangiverisetopassionbyproducingopinionsthatgiverisetosuchstatesasgrieforjoyorresentment,orbypromptingdesiresorvolitionswhenweseethatsomecourseofactionwillleadtowhatwealreadywantorthinkgood.Butheinsiststhatthisdoesnoteverentitleustocallthepas-sionsordesiresunreasonable,ortoholdthatreasonandpassioncaneveropposeeachother,ordisputeforthegovernmentofthewillandactions(T2.3.3.7).WhatHumehasdonehereisemphasizetheimportanceofpassionand/ordesireinthegenesisofchoiceandconduct,whilecontinuingtoaccept,indeedtostress,therational-istinsistenceonthesharpseparationofreasonandemotion.Humeteachesthea-rationalityofpassionwheretherationalistteachestheir-rationalityofpassion.Both,infact,misinterpretcommonmoralCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology263opinion,whichiscommittedtoneitherview,butacceptsthatemo-tion,aswellasopinion,canbebothreasonableandunreasonable.Humeseemstothinkthattheonlycaseswherethemoralevalu-ationsofcommonsenserequiretheascriptionofirrationalitytothepassionsarecaseswherethesearedeemedtobetheresultoffalsejudgments.Butthisisnotso.Onthecontrary:ifIpursueanobjec-tivethatisharmfultome,becauseImistakenlythinkitwillbegoodforme,thenmydesireforitmaybejudgedtobeerroneous,sincemyjudgmentis;butitisnottherebyjudgedtobeunreasonable.IfcommonsenseagreesthatthecourseIamfollowingwillleadtotheobjectiveIampursuing,butholdsmetobemistakeninthinkingitwillbegoodforme;orifitholdsmetoberightinthinkingitwouldbegoodformebutwronginthinkingthecourseIamfollowingwillhelpmeattainit,itisstilllikelytocallmychoiceareasonableone.Thefalsityofmyjudgmentistheverythingthatmakesmyactionreasonableincasesofthissort.IfIgrieveatthesupposedlossofalovedonewhoisinfactaliveandwell,mygriefismistaken,butnotunreasonable.Weapplythetermunreasonabletoanemotionortoadesirewherethatemotionordesireisthoughttobeinsomewayinappropriatetothesituationthattheagentconsidershimself,orherself,tobein:whenitisthewrongwaytorespond,emotion-allyorconatively,toasituationofthatsort.Ifthesituationisnotofthatsort,theresponseismistakenaswell.Butitcanbequitefreeoferror,andstillbeeitherreasonableorunreasonable:bybeingmoderateorexcessive,helpfulorunhelpful,saneorsilly.Thesearealldimensionsofrationalitythatcanbemanifestedbythepassionsthemselves.Humehasperceivedtheimportanceofthepassionsforallourchoiceandconduct,buthasmistakenlyfeltobligedtodenytheirrationalityinordertoaccommodatethisimportance.Inthisrespecthesharesanestimateofthemwiththerationalistswhosetheorieshecontests.Theestimateisonefromwhichcommonsenseisalreadyfree.vii.moralpsychologyandtheselfWehaveseenthatHumesconceptionofascienceofhumannaturereducesmentallifetotheinterplayofimpressionsandideas,andtreatstheminditselfasthetheaterwherethisinterplayoccurs,notasaparticipantinit.ThescholarlyliteraturecontainsmanyCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n264terencepenelhumcriticismsandreappraisalsofwhatHumesaysabouttheself,almostalldirectedtohistreatmentofitinBook1oftheTreatise.Twoofthecriticismsprominentinthisliteratureareofparticularimportance.ThefirstcriticismisthatinspiteoftheNewtonianismofpercep-tionsthatHumeproclaimsattheoutsetoftheTreatise,andagaininthefirstEnquiry,hisaccountsoftheoriginsofourbeliefsleanheavilyontheascriptiontousofpropensities,tendencies,orhabits.Thisleadssometosuggestthatheiscommittedtoacrypto-Kantianpsychologyinwhichthesubjectofexplanationsisthemindanditsdispositions,ratherthantheperceptionsitcontains.23Thesec-ondcriticismisthattheascriptionofapropensity(inthiscasethepropensitytoconfuseonesortofsuccessionwithanother)isessen-tialtoHumesaccountofthegenesisofthebeliefintheunityoftheminditselfthusopeninghimtotheobjectionthathecannotexplainhowwecometohavethebeliefhecriticizeswithoutfirstassumingitstruth.ItispossibletorespondonHumesbehalftothefirstcriticismbysuggestingthattalkofthemindspropensitiesshouldbecon-struedaspopularshorthandforagenuinelyNewtonianaccountthatspeaksinsteadofhowimpressionsandideasgiverisetooneanotherinthemind.Itispossibletorespondsimilarlytothesec-ondbysayingthattheperceptionsthemindhascanwellincludeperceptionsoftheseriesthatconstituteit,withouttherehavingalsotobeanysupervenientsubjectbeyondtheseriessuccessivemembers.Suchresponsesseemtosavehimfromchargesofformalinconsistency.24ButthetransitiontothepassagesabouttheselfinBook2isstillasurprisingoneforthereaderofBook1.Humehastriedtoprepareusforitbytellingustodistinguishbetwixtpersonalidentity,asitregardsourthoughtorimagination,andasitregardsourpassionsortheconcernwetakeinourselves(T1.4.6.5).Healsotriestoeasethetransitionbyclarifyinghisuseofthetermselfinitsfirst23ThetwofundamentalandclassicessaysonthisthemeareRobertPaulWolff,HumesTheoryofMentalActivity,PhilosophicalReview69,no.3(July1960):289310;andFredWilsonsessayofthesametitleinMcGillHumeStudies,ed.D.Norton,N.Capaldi,andW.Robison(SanDiego:AustinHillPress,1979),10120.24SeeNelsonPike,HumesBundleTheoryoftheSelf:ALimitedDefense,Ameri-canPhilosophicalQuarterly4(1967),andEssay3inPenelhum,ThemesinHume,4060.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology265introductioninBook2asthenameoftheobjectoftheindirectpas-sionofpride:Thisobjectisself,orthatsuccessionofrelatedideasandimpressions,ofwhichwehaveanintimatememoryandcon-sciousness(T2.1.2.2).Thismakesitclearthatheisnotrevertingtothepureowner-selfwhoseexistenceherejectssobrusquelyinBook1.Butthisdoesnotprepareusfortheclaimthattheidea,orratherimpressionofourselvesisalwaysintimatelypresentwithus,andthatourconsciousnessgivesussolivelyaconceptionofourownperson,thattisnotpossibletoimagine,thatanythingcaninthisparticulargobeyondit(T2.1.11.4).Moreserious,perhaps,isthefactthattheaccountoftheaetiologyoftheindirectpassionsrequirestheuseoftheideaoftheselfasdistinctfromothers;andtheaccountoftheoriginsofourbeliefinself-identityinBook1isconfinedtoourbeliefintheselfsowninnerunityovertime,andtellsusnothingofhowwecometobeawareoftheexistenceofotherminds.Thisisaseriousgapinhissystem,butperhapsnotamanifestinconsistency.Letusturninsteadtotheroleheascribestothislivelynotionoftheselfinouremotionallife.Whateverthisroleis,hedoesnotthinkitundermineshisNewto-nianmentalscience.Thereisnoplaceinhissystemforthesugges-tionthatchoicesaretheproductofanythingotherthantheseriesofpassionsandcognitionsthatleadtothem.Hisdenialoflibertyofindifferencepermitsnoconsiderationofwhathasbeencalledagentcausation:thetheorythatinfreeactionitistheagent,ratherthantheagentsdesiresorvolitions,thatisthelocusofcausality.25Thisdenialiscoupledwithgreatstressontheclaimthatourunderstand-ingandevaluationofhumanagencydependsonourrecognitionofsettledstatesofcharacter.Thisraises,inthesphereofaction,aperplexityparalleltothatraisedbyhiscriticsinthesphereofepiste-mology:thathisviewseemstorequireacontinuingselfthathasthecharactertraitshefeelsnecessaryforpredictionandevaluation.Wecanperhapsofferasimilaranswer:thattalkofanagentscharacterisshorthandfortalkofthatagentsemotionsanddesires.Howeverwerespondtothesedifficultiesofinterpretation,thereisavitaldimensiontoHumestheoryoftheselfinBook2that25Foraclassictreatmentofthisnotion,seeR.M.Chisholm,FreedomandAction,inFreedomandDeterminism,ed.KeithLehrer(NewYork:RandomHouse,1966),1144.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n266terencepenelhumisonlylatelyrecognizedascentraltohismoralpsychology.26Itpermeateshiswholevisionofthehumancondition.Wefinditsclearestexpressionintheintroductionoftheprincipleofsympathy,inTreatise2.1.11.Scholarshaveinterpretedsympathyasamecha-nismtoexplainmyconcernforothers,whichemergesthroughmyhavingmyselftheveryfeelingsIdiscerninthem.Thisiscorrect,butseriouslyincomplete.TheprincipleisintroducedbyHumeasasecondarysourceoftheself-regardingindirectpassionsofprideandhumility.PridedoesnotmerelycomeaboutthroughmytakingpleasureinqualitiesthatIrecognizetobelongtome;italsocomesaboutthroughmysympatheticallysharingtheadmiration(thatis,inHumesview,thelove)thatothershavetowardmewhenthey,too,discernthesepleasingqualities.Somyownprideisinparttheproductofthementalityofothers,notonlyofmyown.AndsinceIamloved,oradmired,forqualitiesIhaveorobjectsIpossess,myemotionallifeissuchthatIshallpridemyselfonthosequalitiesorobjectsforwhichothersadmireme,andbeashamedofthosequalitiesorobjectsforwhichtheyhate(ordespise)me.Theyarethecocreatorsofmyself-image,andtounderstandthecharacterofmyself-concernitisnecessarytotakethemeasureofthesocietyofwhichIamamember.AsBaierpointsout,manyofthefeaturesothersthusmakepartofmyself-imagewillbephysicalones,sotheselfofthepassionsisaphysicalizedconstruct,andnotthequasi-solipsistmonsterofBook1.27Oncethisisrecognized,itisalsoevidentthatIsometimescometohaveprideorhumilityinsomecharacteristicIascribetomyselfonlyafterothersadmireordespiseit:theirevaluationofitandofmemaynotonlyaugmentmyown,butactuallyengenderit.AndImay,ofcourse,cometosimulate,oractuallydevelop,somecharactertraittheywouldpraiseinordertopreventtheirblamingme(andhencemyblamingmyself)foritsabsence:this,aswehavealreadyseen,ispartofHumesaccountoftheoriginofthesenseofduty(seeagainT3.2.1.8),anaccountthatseekstoturntherationalistskeyethicalendowmentintoaninternalizedsocialproduct.26ItisgivenitsdueplaceinBaiersAProgressofSentiments;seeespeciallyChap.7.SeealsointhisvolumetheessayHumeandtheProblemofPersonalIdentity,Sect.IV.27Baier,AProgressofSentiments,136.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology267ThesortofstorythistellsusabouttheselfassocialconstructionisonewehaveheardsincefromFreud,Marx,andtheexistentialists,alwayswithideologicalaccretionswhollyforeigntoHumesnatural-ism.Hisownsummarystatementisasfollows:Ingeneralwemayremark,thatthemindsofmenaremirrorstooneanother,notonlybecausetheyreflecteachothersemotions,butalsobecausethoseraysofpassions,sentimentsandopinionsmaybeoftenreverberated,andmaydecayawaybyinsensibledegrees(T2.2.5.21).Itiseasytoseefromthisinsistencethattheselfisnotdis-cerniblewithinbutlargelyascribedbytransferencefromwithoutwhyHumehassuchdeephostilitytoallsystemsthatviewper-sonsasalientothesocialworldtheyinhabit.Hisnegativitytowardrationalismanditscravingforautonomyistheresultofitsbeingatheoreticalforcethatcanonlyencourageself-distancingfromthesourcesofemotionalnourishmentthatmakeuswhatweare.Andhisintemperaterejectionofthereligiousausteritiesofthemonkishvirtuescanbeseenashavingthesametheoreticalsource.28Eachislife-denying,andinaquiteliteralsenseself-destructive.Humannaturedoesnotneedtobemastered,nordoesitneedtoberedeemed.Itneedssocialnurture.Bothreasonandtruereligionaretheslavesofthepassions.viii.conclusionIhavearguedthatHumeisaneo-HellenisticthinkerwhofollowstheStoics,Epicureans,andSkepticsinmaintainingthatweshouldavoidanxietybyfollowingnature.Thisprescriptionisnotoriousamongphilosophersforcombiningdescriptiveandnormativeele-ments.Humeisnot,inanygeneralway,confusedbetweendescrip-tiveandnormativeclaims:thereisnothinginprincipleconfusedaboutseeinganunderstandingofournatureasaguidetooneswayoflife,oreventotheproperpracticeofphilosophy.Thereismorethanonewayofgettingandusingsuchguidance.Humethinksaphilosophermust,firstandforemost,learntoaccepthisorhernatureforwhatitis.Thismeansrecognizingthatitissoprogrammedthatourinstinctsfurnishuswithbeliefsthatwecannotsurvivewith-out,orsupplyindependently,orseriouslyquestion.Facedwiththis28EPM9.2.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n268terencepenelhumfact,thephilosophicalenterprisesofskepticaldoubtandrationalistreconstructionaredoomedtofailureonpsychologicalgroundsalone,andtheattemptstopursuethemcanonlygenerateandexacerbateanxiety.WhenweturntoHumesmoralthought,wefindtheparallelinsistencethatwemustrecognizethedominanceofthepassionsinournature,andnotriskmiserybyattemptingtofolloweccentricprogramsofchoicethatfrustratetheminthesupposedinterestsofreason,orthemortificationsofreligion.Hereagain,wehavetoacceptournature,notviolateit.HereHumerisksconfusioninafundamentalrespect:whilethereisnothingincoherentindescribingournatureandthensayingwemustacceptitandnotviolateit,thisisincoherentifweareunabletoviolateit.Tocombinethedescriptivewiththenormativewithoutincoherence,itisnecessarytopermitfreedomofchoiceinaformforwhichHumesownaccountoflibertyallowsnospace.Thepriceofusingthestudyofhumannatureasaguidetochoiceisthepriceofrecognizingthatitispartofournaturetobeabletochoose.Butifthisisadmitted,wecanthenfollowhiminsayingthatifwemakecertainkindsofchoices,wemayruinourselves,andendupanxious,orincapacitated,orotherwisemiserable,byfrustratingourbasicneeds.Readthisway,hissystemtellsusthatthepolitesocietyhumanbeingshaddevelopedinproperty-owningWesternEuropebyhisday,withallitsprotectiveartifices,meetstheneedsofhumannaturebetterthanitsalternatives.Whilethismaybejudgedbysometobecomplacentorenervating,theexperienceofmoreradicalprogramsthatarebasedonideologiesthatattendlesstothedetailsofhumannatureshouldmakeushesitatetodismisshisadvicetooreadily.suggestionsforfurtherreadingInadditiontotheworkscitedinthenotestothisessay,forfurtherreadingthefollowingarerecommended.Harris,James.OfLibertyandNecessity:TheFreeWillDebateinEighteenth-CenturyBritishPhilosophy.Oxford:ClarendonPress,2005.Kane,Robert,ed.TheOxfordHandbookofFreeWill.Oxford:OxfordUni-versityPress,2002.Kenny,Anthony.Action,EmotionandWill.London:RoutledgeandKeganPaul,1963.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesMoralPsychology269McIntyre,Jane.PersonalIdentityandthePassions.JournaloftheHistoryofPhilosophy27(1989):54557.Mercer,Philip.SympathyandEthics.Oxford:ClarendonPress,1972.Russell,Paul.FreedomandMoralSentiment:HumesWayofNaturalizingResponsibility.NewYork:OxfordUniversityPress,1995.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\ndavidfatenorton9TheFoundationsofMoralityinHumesTreatiseIfoundthatthemoralPhilosophytransmittedtousbyAntiq-uity,labordunderthesameInconveniencethathasbeenfoundintheirnaturalPhilosophy,[namely,]ofbeingentirelyHypotheti-cal,&dependingmoreuponInventionthanExperience.EveryoneconsultedhisFancyinerectingSchemesofVirtue&ofHappiness,withoutregardinghumanNature,uponwhicheverymoralCon-clusionmustdepend.ThisthereforeIresolvedtomakemyprin-cipalStudy,&theSourcefromwhichIwoudderiveeveryTruthin...Morality.KHL6InBook2oftheTreatiseofHumanNatureHumereportsthataques-tionconcerningthefoundationofmoraldistinctionshadoflateyearsbeenofgreatpublicinterest.Thequestioniswhethermoraldistinctions(distinctionsbetweenvirtueandvice)arefoundedonnaturalandoriginalprinciples,orarisefrominterestandedu-cation.Hethensuggeststhatthosewhotracedthedistinctionbetweenvirtueandvicetoself-interestortheprejudicesofedu-cationsupposedthatmoralityhasnofoundationinnature.Incontrast,thosewhosaidthatmoraldistinctionsarefoundedonnat-uralandoriginalprinciplessupposedthatmoralityissomethingreal,essential,andfoundedonnature(T2.1.7.23,5).Thisdebate,aswewillsee,raisedbothanontologicalquestion(whichfeaturesoftheworld,ifany,doourmoraljudgmentsreflect?)andanepistemo-logicalquestion(whichofourfaculties,reasonorsense,enablesustograspmoraldistinctions?).1Thefirstsectionofthisessayfocuses1TheseissuesarebothaddressedinTreatise3.1.12,butaremoreexplicitlystatedinHumeslaterEnquiryconcerningthePrinciplesofMorals;seeEPM1.3.Foran270CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise271ontheviewsoffouroftheprincipalearlyeighteenth-centurypartic-ipantsinthisdebateinordertounderstandthecontroversy,whichHumehopedhisownmoraltheorywouldresolve.TheremainderoftheessayoutlinessomekeyfeaturesofHumesresponse,inBook3oftheTreatise,totheseandotherquestionsaboutthefoundationsofmorality.2i.thefoundationsdebate:someprincipalpositionsTwoturn-of-the-centuryphilosopherswhosewritingscarriedthedebateregardingthefoundationsofmoralityintotheeighteenthcen-turywereSamuelClarkeandAnthonyAshleyCooper,LordShaftes-bury.BernardMandevilleandFrancisHutchesonwerecentralfig-uresintheearlyeighteenth-centurydebate.3TheworksofallfourwriterswerewellknowntoHume.SamuelClarke(16751729)Inthefirstoftwodiscoursespublishedin17056,4SamuelClarkearguedthattherehasnecessarilyexistedfrometernityaunified,unchangeable,andself-existingBeingwhoisnotonlyinfinite,omnipotent,andthecauseofeverything,butalsoaBeingofinfi-nitegoodnessandjusticeandalltheothermoralcharacteristicsappropriatetotheSupremeGovernorandJudgeoftheworld.IntheaccountofHumeslaterviewsonmorals,seeinthisvolumetheessayHumesLaterMoralPhilosophy.2Humesmoraltheoryisinterpretedinwidelydifferentways.Thebooksandarticlesrecommendedattheendofthisessayrepresentsomeofthesedifferentinterpreta-tions.3Fortheviewsofadditionaleighteenth-centurywriterswhoengagedinthisdebate,seeDavidFateNortonandManfredKuehn,TheFoundationsofMorality,inTheCambridgeHistoryofEighteenth-CenturyPhilosophy,ed.K.Haakonssen,2vols.(Cambridge:CambridgeUniversityPress,2006),2:93986.Fordiscussionofseventeenth-centurycontributionstothecontroversy,seeinthisvolumetheessaybyKnudHaakonssen,or,forgreaterdetail,see,bythesameauthor,NaturalLawandMoralPhilosophy:FromGrotiustotheScottishEnlightenment(Cambridge:CambridgeUniversityPress,1996).4TheseworksareADemonstrationoftheBeingandAttributesofGod(London,1705)andADiscourseconcerningtheUnchangeableObligationsofNaturalReli-gion(London,1706).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n272davidfatenortonseconddiscourseClarkearguedthatthereareeternalandnecessarydifferencesbetweencreatedthings,andthatthesedifferencesarealonesufficienttomakeitmorallyrightthatcreaturesshouldactincertainways,andmorallywrongthattheyshouldactinotherways.PriortoanycommandoftheDeity,andpriortoanyconventionorpositivelaw,thereare,intheverynatureofthingsthemselves,char-acteristicsandrelations(andespeciallydifferences)thattranslateimmediatelyintorealmoraldifferences.Becausetherearecertainnecessaryandeternaldifferencesofthings,humanscanseethatmoraldifferencesarefoundedunchangeablyinthenatureandrea-sonofthings,andunavoidably[arise]fromthedifferencesofthethingsthemselves,withtheconsequencethatsomethingsareintheirownnatureGoodandReasonableandFittobedone.5ThatClarkemeanstosaythattheserealandenduringfactualdif-ferencesprovidebothanecessaryandasufficientgroundformoralityisconfirmedbywhathehastosayaboutmoralknowingandmoralobligation.Thedifferencesandrelationsheisspeakingofare,heargues,absolutelyfixedandunalterable,andnomoreopentowill-fulchangethanarethedifferencesbetweenLightandDarknessorMathematicalorArithmeticalTruths.6Furthermore,thediffer-encesthatconstitutemoralityareknown,bymeansoftheactivitiesofreason,injustthewaytheseotherdifferencesareknown,eitherimmediatelyorbydemonstration.Foranyrationalbeingtodenythattherearesuchdifferenceswouldbeequivalenttodenyingthattwicetwoisnotequaltofour,orthatawholeisnotlargerthananyofitsconstituentparts.7Clarkegrantsthatthereisonenotabledifferencebetweenourresponsetononmoraltruthsandourresponsetoperceivedmoraldifferences.Assumingweunderstandtheterms,seeingtheproposi-tion2+2=4leavesuswithnochoicebuttoassent.Incontrast,theperceptionofamoraltruthleavesusfreetoactinamannerthatiscontrarytowhatwehaveperceived.8But,heargues,actinginthiscontrarywayisnotonlyabsurd;itisalsoimmoral.Weper-ceivethatthereisaninfiniteDeityandthatwearefinitecreatures5Discourse,TheWorksofSamuelClarke,ed.B.Hoadly,4vols.(London,1738),2:61112.6Discourse,Works,2:626.7Discourse,Works,2:609,613.8Discourse,Works,2:615.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise273dependentonHim.Wherethereissuchadifferencebetweenbeings,wealsoperceivethatweastheinferiorcreaturesoughttohonor,imitate,andobeythevastlysuperiorbeing.Tofailtoactconsis-tentlywiththeseperceptionsisabsurdbecauseitisineffecttodenytheverytruthswehaveperceived;itisabsurdinthesamesensethatitisabsurdtodenythattwicetwoisfour.Inaddition,failingtoactconsistentlywiththeseperceptionsisalsoblameworthy.Oncewehaveseentherealandeternaldifferencesbetweenthesebeings,ourminds,oftheirownaccord,arecompelledtoownandacknowl-edgethattherereallyaresuchobligationsunavoidablybindingustoaparticularcourseofaction.9TheoriginalObligationsofMoral-ity,asClarkecallsthem,followimmediatelyfromourrecognitionofthenecessaryandeternalrelationsofthingsthemselves.Thegroundorfoundationofmoralityisnothingmorethanasetofrealandperceivablerelationsbetweenexistingthings,relationsthemereperceptionofwhichissufficienttoinformusofwhatisrightandwrong,andtoobligeustotherightorvirtuouscourseofaction.10AnthonyAshleyCooper,thethirdEarlofShaftesbury(16711713)AtthebeginningofhisInquiryconcerningVirtueorMerit,Shaftes-buryobservesthatreligionandvirtueseemsocloselyrelatedthattheyappearinseparable.11Toknowhowvirtuerelatestoreligionwemustknow,ashesays,whatvirtueis,consideredbyitself.Tothisend,hedrawsourattentiontowhathecallstheframeofnature,thesystematicinterconnectionsofnature.Althoughthereismuchwedonotknowaboutnature,andparticularlyabouttherolesomespeciesareintendedtoplayinthelargerschemeofthings,thereisalsomuchthatwedoknow.Weknow,forexample,thateachcrea-tureisbetteroffinsomeconditionsthaninothers,andthusthatthereisinrealityarightandawrongstateofeverycreatureandacertainendtowhicheverythingin[its]constitutionmustnaturally9Discourse,Works,2:61314;seealso2:618.10Discourse,Works,2:630.11ShaftesburysInquirywasfirstpublishedfromanImperfectCopyin1699.Anauthorizedversionwaspublishedin1711asapartofCharacteristicksofMen,Manners,Opinions,Times,andisherequotedfromtheeditionofJohnRobertson,2vols.(London,1900;reprinted,Indianapolis:Bobbs-Merrill,1964).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n274davidfatenortonrefer.Wecansee,too,thatwhatisgoodforindividualsisgoodfortheirspecies,andthusthatwhatwemightcallantisocialbehaviorisagainsttheinterestofeachindividual.Wecanalsoseethat,asthewingoftheflyissuitedtothewebofthespider,soaremanyspeciessuitedtotheexistenceandwell-beingofother,verydifferentspecies.Inconsequenceofmanysuchperceivedinterdependencies,itisreasonabletoconcludenotonlythatallanimalsformasystem,butalsothateverythingfoundintheuniverseispartofageneralsys-tem.Anditisalsoreasonabletoconcludethatwhatevercontributespositivelytothissystemisgood,andwhateverisdestructiveofitisillorbad.Thetermsgoodandill,inotherwords,canbeusedtorefertorealorobjectivedifferencesbetweenexistingorpossiblestatesofaffairs.12Shaftesburygoesontosaythatwedonotconsideracreaturevir-tuousmerelybecauseitcontributespositivelytothegoodofitself,itsspecies,ortheuniverseingeneral.Tobevirtuous,thiscreaturemustalsosatisfyotherrequirements.Shemustfirstbeasensible,reflectivecreaturewhoisawareofwhatshedoes.Thatis,shemustnotactmerelyfrominstinctinthemannerofathoughtlessandunreasoningautomaton.Shemustalsohaveanotionofthepublicinterestandasenseofrightorwrongshemustgraspthemoralcharacterofsituationsandofwhatshedoes.Andfinally,shemustactfromaself-determinedmotivetodogoodoravoidevil.Acrea-turewholacksself-consciousnessortheabilitytograspthemoralcharacterofsituationscannotbevirtuous.Apersonwhocontributestothepublicgoodmerelyasaconsequenceofselfishmotivesisnotcountedamongthevirtuous.13Thevirtuousindividualistheindi-vidualwho,awareofwhatcontributestoordetractsfromthepublicgood,undertakes,byconsciousintentorfromsettledcharacter,toaddtothestoreofgoodortoavoidincreasingthesupplyofevil.Thesefeaturesofvirtueestablished,Shaftesburyreturnstothequestionoftherelationofreligionandvirtue.Itisclear,heargues,thatneithervirtuenorwhatcomestothesamething,thepracticeofvirtue,isdependentonreligion.Itistruethatbeliefinaprovidential,judgingDeityandafuturestateofrewardorpunishmentmaypro-videanincentivetoactinwaysthatareconsistentwithvirtue.And12Inquiry,Characteristics,1:238,2436.13Inquiry,Characteristics,1:24758.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise275yetvirtuousbehaviordoesnotdependonholdingreligiousbeliefs,andmayevenbehinderedbysuchbeliefs.Manyreligionsteachthattreachery,ingratitude,orcrueltyhavebeengivenadivinesanc-tion,orcallontheirfollowerstopersecutefriends,toofferhumansacrifices,ortoabuseandtormentthemselvesoutofreligiouszeal.Butnothing,Shaftesburyinsists,notevenreligion,canjustifybrutal-ityorbarbarityormakethembeneficial.Nothing,neithercustom,law,norreligioncaneveraltertheeternalmeasuresandimmutableindependentnatureofworthandvirtue.14Whetheranactisjudgedtobemoralorimmoral,accordingtoShaftesbury,dependsonthedifferentmotivesavailabletorationalagents.Thathesupposedthisdistinctionofmotivesaneffective,practicalfoundationofmoralityismadeclearinhisattacksonwhathetooktobethemoralskepticismofThomasHobbesandJohnLocke.Shaftesburyknew,ofcourse,thatbothoftheseearlierphilosophershadproposedmoraltheories.Henonethelessconsid-eredthemtobemoralskepticsbecausetheirtheorieshavetheeffectofquestioningordenyingtherealityofmoraldistinctions.15Hobbeshadattemptedtoexplaineveryhumanactbytheoneprincipleofself-interestorself-regard,16anexplanationthathastheconsequenceofmakingapparentmoraldistinctionsmeaningless.IfHobbesisright,thensuchwordsasfriendship,love,publicinterest,allthosewordsthatappeartomakereferencetoaltruisticactsormotives,haveultimatelythesamemeaningastheircontrariesbecausetherearenodifferencesinmotivation.IfHobbesisright,thentherecanbe,accordingtoShaftesbury,nosuchthinginrealityasvirtue.ToShaftesbury,whoalludestohimselfasarealistinmorality,reducingallmotivationstoselfishonesistointroduceageneralscepticismaboutmorality,askepticismthatdeniesthatthereisanyrealvirtueormoralgoodandnaturaljustice.17Lockesmoraltheory,becausemoresubtlethanthatofHobbes,isevenworse.Lockearguedthatourmoralideasaremereinventions,constructsbasedonnorealmodelorarchetypeintheworld,andhecredu-louslyrepeatedstoriesofculturesthathavenoideaofvirtue.He14Inquiry,Characteristics,1:255.15SeeLife,Letters,andPhilosophicalRegimen,ed.B.Rand(NewYork,1900),378.16Thisviewisnowcalledpsychologicalegoism.Inthisessaythetermegoismalwaysreferstothisformofegoism.17Characteristics,1:615,79;2:53.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n276davidfatenortonthenconcludedthatvirtue...hasnoothermeasure,law,orrule,thanfashionandcustom;[that]morality,justice,equity,dependonlyonlawandwill....Andthusneitherrightnorwrong,virtuenorvice,areanythinginthemselves;noristhereanytraceorideaofthemnaturallyimprintedonhumanminds.18SuchviewsShaftesburytooktobedangerous.BecauseHobbesandLockeexplicitlydenyeitherthatmoraldistinctionsarereal,observer-independentdistinctions,orthattheyderivefromrealdif-ferencesinthenatureofthings,theirviewsencourageindividualstodisregardmoralconsiderationsandasaconsequencethreatentounderminemoralityandsocietyitself.Tocounterthisdanger,Shaftesburyrejectsspeculationandapriorireasoninginfavorofanobservationallybasedstudyofhumannature,astudythatleadshimtoconcludethathumansareinherentlymoralandinherentlycapa-bleofrecognizingmoraldistinctions.Justasanimalshaveinstinctsordispositionsthatenablethemnotonlytosurvive,butalsotothrive,sotoodohumanshavethedispositionstheyneedtothrive.Humanscannotsurvivewithoutsociety.Itisnotsurprising,then,thathumansaremotivatednotonlybytheselfishnessthatHobbesdescribes,butalsobypublicspiritoraconcernforthegeneralgood.Humansalsodeveloptheabilitytodistinguishbenevolencefromindifferenceormalice.Wefind,tosumup,thathumanshaveakindofmoralinstinct,anaturalmoralsensethatenablesthemtobejustthekindofcreatureHobbessaystheyneverare,andthatpro-videsforusanarchetype,anobjectivemodel,ofthosemoralideasthatLockesaysareonlyarbitraryconstructs.Inaddition,toshowthatmoralityisnotfoundedonreligiousdecreeandthatithasarealfoundationinnature,Shaftesburyarguesthatmoralityrestsonthedistinctionsofmotiveavailabletorationalhumanagents.Theseagentsareabletoactnotonlyfromself-regarding,butalsofromother-regardingoraltruistic,motivesanddooftenenoughactfromtheseother-regardingmotives.Whentheydoso,theyareactingvirtuously.BernardMandeville(16701733)ThedecadefollowingtheappearanceofShaftesburysCharacteri-stickswasarelativelyquietoneforthefoundationsdebate,but18Life,Letters,andPhilosophicalRegimen,4035.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise277thepublicationin1723ofanexpandedversionofBernardMandev-illesTheFableoftheBeeschangedthat.MandevilletookdirectaimatShaftesbury,andparticularlyathisoptimisticviewofhumannatureandhisaccountofthefoundationofthemoraldistinctionswemake.ButMandevilleschallengemayeasilybemisunderstood.Shaftesburyhadtracedmoralitytodistinctionsofmotive,arguingthatvirtuousindividualsarethoseandonlythosewhoconsciouslyactfromother-regardingmotives.Mandevilleacceptedthisconclu-sionandmadeitakeypremiseinanargumentleadingtotheconclu-sionthatthedistinctionbetweenvirtueandvicelacksanadequatefoundationinmorals.Moraldistinctionsaremerelyconventionscre-atedbyaclassofinventivehumansandfoistedontotherestofus.Mandevillesanalysistakesthisform.Shaftesburysclaimaboutmotivationandmoraldistinctionsisacceptedasafirstpremise:theonlyvirtuousactsarethosemotivatedbyother-regardingmotives.Itis,Mandevillesays,impossibletojudgeofaMansPerformance,unlessweareth[o]roughlyacquaintedwiththePrincipleandMotivefromwhichheacts.19MandevillessecondpremiseisstatedintheopeningwordsofhisEnquiryintotheOriginofMoralVirtue:allanimals,includinghumans,hesays,areonlysolicitousofpleasingthemselves,andnaturallyfollowthebentoftheirownInclinations,withoutconsideringthegoodorharmthatfromtheirbeingpleasedwillaccruetoothers.20Butifonlyotherregardingactsare(gen-uinely)virtuous,andiftherearenoother-regardingactsbecauseallofusaremotivatedonlybyself-interest,thenitfollowsthattherearenovirtuousacts.Mandevilledoesnotdenythatwe(appearto)makemoraldistinctions.Hegrantsthatwedescribesomeactsorpersonsasvirtuousandothersasvicious,buthedeniesthatthisfactofourexperienceestablishesthatthismorallanguagerepresentsrealmoraldifferences.Mandevillebringshomehispointwithastrikinganalysisofpity,apassionsupposedtoderivefrompurekindness.Pity,saysMan-deville,isthemostgentleandtheleastmischievousofallourPassions,butitisstillasmuchapassionasAnger,Pride,orFear.Asaconsequence,hegoeson,whoeveractsfrompity,nomatter19AnEnquiryintotheOriginofMoralVirtue,TheFableoftheBees:or,PrivateVices,PublickBenefits,ed.F.B.Kaye,2vols.(Oxford:ClarendonPress,1966),1:56.20Enquiry,Fable,1:41.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n278davidfatenortonhowmuchgoodhehappenstodo,reallyactsfromaself-interestedimpulseandhasnothingtoboastofbutthathehasindulgedaPassionthathashappenedtobebeneficialtothePublick.ThereisnoMeritinsavinganinnocentBabereadytodropintotheFire:TheActionisneithergoodnorbad,andwhatBenefitsoevertheInfantreceived,weonlyobligedourselves;fortohaveseenitfall,andnotstrovetohinderit,wouldhavecausedaPain,whichSelf-preservationcompelldustoprevent.21SupposeMandevilleiscorrectaboutthesebasicfacts.Moraldis-tinctionsdependonaspecificdistinctionofmotives,whicharesaidtobeeitherother-regardingorself-regarding(theviewheshareswithShaftesbury).Hefinds,however,thatnosuchdistinctionofmotivesexists,andthuswemustconcludethattherearenorealmoraldis-tinctions.Howdoesithappen,then,thatwecommonlydenominatesomeactsandpersonsvirtuousandotheractsandpersonsvicious?Man-devillesanswertothatquestiontakestheformofaninventorsstory.Noanimal,Mandevilletellsus,issoheadstrong,selfish,cun-ning,anddifficulttocontrolorgovernashumanbeingsare.Indeed,humansaresodifficulttogovernthat,ifwearetobegoverned,itwillneverbebyforcealone.Fromthisbeginning,hegoesontoexplainthathumanshavebecomegovernablethroughtheinven-tionofmorality.Thosewhohaveundertakentoestablishsociety(toestablishawell-orderedgroupofhumans)havebeenengagedinamassiveandlong-standingdeceit.Knowingthathumansareallselfishandself-willed,theseinventorshavenonethelesstoldusthatwewilleachbebetteroffifeachofuseffectivelysubjugateshisorherappetitesandsacrificespersonalintereststothegeneralgoodofthepublic.InMandevillesview,theclaimthatourself-interestisbestservedbysublimatingittothegeneralgoodisnothingmorethananout-rightlie.Inorderforthislietobeeffective,ithadtobeattractivelypackaged.Thatis,ifthisliewastoservetheendforwhichitwas21Enquiry,Fable,1:56.OnthefollowingpageMandevillegrantsthatsomemayappeartoperformgoodactionsfromnootherMotivebuttheirLovetoGoodness,buthethenarguesthattheRewardofaVirtuousAction,whichistheSatisfactionthatensuesuponit,consistsinacertainPleasure[theagent]procurestohimselfbyContemplatingonhisownWorth:WhichPleasure,togetherwiththeOccasionofit,are...certainSignsofPride.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise279(andis)told,thosewhocurtailtheirnaturallyselfishinclinations(theirself-interest)becauseofithad(andhave)toberewarded,whilethosewhofailtocurtailtheseselfishinclinationsmustbepun-ished.Therewardsofferedincludetangibleormaterialbenefits,buttheyarenotlimitedtobenefitsofthatkind.Therearenot,foronething,enoughtangiblegoodstosatisfyeveryone,and,eveniftherewereenough,therewouldbeproblemswiththeirdistribution.Asithappens,themostimportantofthenontangiblerewardsofferedhasbeenfoundtobeaparticularformofmoralpraiseorflattery.Havingfoundsuchpraiseaneffectivemeanstotheirend,theinven-torsalsoconvincedtheindividualsmakinguponepartofhumanitythattheywereaclasssuperiortotheremainingpart.Thatis,somehumansweremadetofeelmorallysuperiortothoselow-mindedtypeswho,alwayshuntingafterimmediateEnjoyment...yieldedwithoutResistancetoeverygrossdesire,andmadenouseoftheirRationalFacultiesbuttoheightentheirSensualPleasure.Char-acteristically,individualsofthesuperiorClass,supposingthem-selvestobeofaloftiersortaltogether,makewaronthemselves(theyfightagainsttheirnaturalbutallegedlyirrationalinclinations)andseekthegeneralgoodthroughtheConquestoftheirownPassion.Inshort,theseindividualsweredupedbytheirownprideandthemachinationsoftheinventors,andthisinducedmanyofthem,espe-ciallythefiercest,mostresolute,andbestamongthem,toendureathousandInconveniences,andundergoasmanyHardships,thattheymayhavethepleasureofcountingthemselvesMenofthe[superior]Class.Inthisway,Mandevillesays,SavageManwasbroke,or,inotherwords,trainedinpseudo-morality.Moreover,inthecourseoftime,eventhesensuallowerclasscametobelievethattheirindi-vidualinterestswouldbestbeservedbymoderatingtheirnaturalinclinations.Takengenerally,thestoryleadstotheconclusionthatmoralityisbestdescribedasthePoliticalOffspringwhichFlatterybegotuponPride.22FrancisHutcheson(16941746)TheInquiryintotheOriginalofourIdeasofBeautyandVirtueofFrancisHutchesonwasfirstpublishedin1725.ThesubtitleofthefirsteditiondescribestheworkasadefenseofShaftesburys22Enquiry,Fable,1:4351.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n280davidfatenortonprinciplesagainsttheAuthoroftheFableoftheBees.InhisPref-aceHutchesonsaysthathisaimistoshowthatHumanNaturewasnotleftquiteindifferentintheAffairofVirtue,23andtheworkcontinuallyreverberateswithexplicitconcernwithmoralfounda-tions.Notingthatwehumansdistinguishbetweenmoralgoodandevil,HutchesonaskswhatgeneralFoundationthereisinNatureforthisDifference.AfterexaminingtheSpringsoftheActionswhichwecallvirtuous,asfarasitisnecessarytosettlethegeneralFoundationoftheMoralSense,heundertakestoshowthatnei-theresteemnorbenevolenceisorcanbefoundedonSelf-Love,orViewsofInterest,andthattheuniversalFoundationofourSenseofmoralGoodisbenevolence,whilethefoundationofoursenseofmoralevilisMalice,orevenIndolence,andUnconcernednessabout...manifestpublickEvil.24GivenhisstatedaimofdefendingShaftesburysviews,Hutche-sonsassociationofvirtuewithother-regardingmotivescomesasnosurprise.AndgiventhatMandevillehadalsoacceptedtheprinciplethatmoraldifferencesderivefromdifferencesofmotive,itisclearthatheandHutchesonreachedtheircruciallydifferentconclusionsabouttherealityofvirtuebecausetheyaffirmedcontraryviewsaboutcertainfactualmatters.Mandevillessurveyofhumanbehav-iorledhimtotheviewthatindividualsneveractfrombenevolentorother-regardingmotives.HutchesoncounterswithasurveyshowingthatMandevilleismistakenabouttherelevantfacts:atleastsomehumanactionsaremotivatedbyaregardorconcernforothers.Con-sequently,Hutchesoncanvalidlyconcludethattherearegenuinelyvirtuousindividualsandthatvirtueissomethingwithareal,objec-tivefoundationinthenatureofthings,in,morespecifically,mindsofhumanagents.25Hutchesonsconclusiondependsheavilyontheresultsofhisnewand,heargues,morecarefulsurveyofourmoralapprovalsordis-approvals,ofthecircumstancesinwhichwesayindividualsarevirtuousorvicious.Amongotherthings,thissurveyreveals23AnInquiryintotheOriginalofourIdeasofBeautyandVirtue,4thed.(London,1738;facsimilereprint,Westmead:GreggInternationalPublishers,1969),xiii.24Inquiry,105,132,154,172.Hutchesonsplentifulitalicshavebeeneliminated.25ForHutchesonsconcerntoprovetheRealityofVirtue,andthatnoActionofanyPersonwaseverapprovdbyus,butupontheApprehension...ofsomereallygoodmoralQuality,seeInquiry,xi,201.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise281Thatofmanylong-deadindividualswhocannolongercon-tributetoourinterestorpleasure,somearemorallyap-proved,somearemorallydisapproved,andaboutsomewearemorallyindifferent.That,althoughthegenerousactsofanagentandtheproductsofaplotoflandmayequallybeofbenefittousandtherebygainourapproval,itisonlytheagentwhoisthoughttobevirtuous,and,althoughthefraudulentactsofapartnerandafallingbeammaybothinjureus,itisonlythepartnerwhoisthoughttobevicious.Thatiftwoindividualscontributeinsimilarwaystoourwell-being,theonefromanintenttobenefitus,theotherfromapurelyself-interestedmotive,wecountasvirtuousonlytheindividualwhointendedtobenefitus;moreover,becauseourmoralassessmentsofindividualsdependuponthemotivesfromwhichindividualsact,wefindthatwemaymorallydisapproveanagentwhosebehaviorcausesnoactualinjurytoothers,andmorallyapproveagentswhosebehaviorhasaccidentallycausedthemharm.Thatalthoughwecanbebribedtoperformanactionthatwethinktobemorallywrong,wecannotbebribedtofeelthatthissameactionismorallyrightorthatwewererighttoundertakeit.Wecanbribeanenemytobetrayhiscountryandbenefitours,therebyfurtheringourowninterest,butwenonethelessfeelmoraldisapprovaloftheenemywhohasbeenbribed,takinghimtobeatraitorandvicious.Moregenerally,wefindthatwecannotbyanyactofourwillalterourmoralapprovalsanddisapprovals.26Factsofthissort,Hutchesonconcludes,establishbeyonddoubtthatMandevilleandotheregoistssuchasHobbesaremistaken.Factsofthissortshowthattherearenaturalorunlearneddifferencesinourresponsestoactionsorevents,differencesthatwouldnotariseiftheegoistshadcorrectlydescribedhumannature.Buttheyhavenotdoneso.Theegoistshavebadlymisdescribedourmoralexperience.Whenwelookcarefullyatourmoralapprovalsanddisapprovalswefindthatwequitenaturallymakesomeimportantdistinctions:26Inquiry,11115,1237.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n282davidfatenortonWithouttrainingorindoctrinationwedistinguishbetweennaturalandmoralgoods,andwealsodistinguishbetweenmoralgoodorvirtueandmoralevilorvice.Moralgood,Hutchesonargues,isacharacteristicof,oratleastattributableto,onlyrationalagentsasagents,whilenaturalgoodmaybeacharacteristicofmanydifferentclassesofthings.Hutch-esonscrucialclaimisthatwerecognizethisdistinction,andshowthatwerecognizeit,becausewerespondinonewaytothehon-esty,kindness,orgenerosityofanagent,andinasubstantiallydif-ferentwaytothebeneficialqualitiesofaninanimateobject,ortothewealth,houses,lands,health,sagacity,orstrengthpossessedbyhumanbeings.Wenecessarilyholdinhighesteemthosewhopossesssuchqualitiesasgenerosity,butwemayverywellenvyorholdindisesteemthosewhopossesswealthorpower,andwesimplydonottakepossessionsorbodilyendowmentstobeortohavemoralchar-acteristics.Inaddition,ourideaofmoralgoodistheIdeaofsomeQualityapprehendedinActions,aqualitythatproducesAppro-bation,attendedwithDesireoftheAgentsHappiness,whileourideaofmoralevilisofaqualitythatproducesdisapprobation,andisattendedwithadesirefortheagentsmisery.Hegoesontosug-gestthatapprobationanddisapprobationprobablycannotbefurtherexplainedbecausetheyaresimpleorprimitiveideas,andeventhesedescriptionsofmoralgoodandevilareofferedprovisionally.Theyarethebestwecandountilwefindoutwhetherwereallydohavesuchideasandwhatgeneralandnaturalfoundationthereisforthisdistinctionbetweenthemorallygoodandthemorallyevil.27WehavebeenreviewingHutchesonsattempttoshow,contrarytoMandeville,thatsomeactionsareperformedfromother-regardingmotives,andthusthatthereisareal,well-foundeddistinctionbetweenvirtueandvice.AssumingthatHutchesonsdescriptionofourapprovingexperiencesisaccurate,heisontracktoaccom-plishhisgoal.Ifitistruethatourresponsestodifferentkindsofgoodandtodifferentkindsofactionarethemselvessignificantlydifferent,andifitistruethatwhatconstitutestherelevantdiffer-encesbetweentheseactionsisrecognitionthatthereisasignifi-cantdifferenceinthemotivationsthatgiverisetothem,sothatwefeelaspecialkindofapproval(approbation)onlyinresponseto27Inquiry,1056.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise283other-regardingactions,thenitisclearthatsomeactionsaretheresultofsuchother-regardingmotives.Ourexperienceissignifi-cantlymorecomplexthanMandevillehadsupposed.Giventhesefacts,itfollows,contrarytoMandeville,thattherearegenuinelyvir-tuousindividuals,andthatvirtueandvicearerealandfoundedinthemindsofagents.Giventheseimportantfacts,thequestionbecomes,asHutchesontypicallyframesit,whatfeaturesofhumannaturearepresupposedbythefactthatwecananddomakethesemoraldistinc-tions?ElaboratingonShaftesburyssuggestion,HutchesondevotesasectionofhisInquirytoshowingthatwehaveimplantedinourNatureacomplexmoraldisposition,amoralsense,whosepresenceandoperationprovideafoundationformorality.28Themoralsenseprovidesthisfoundationbymakingitpossibleforustobemoralagentsandmoralobservers.Butevenbeforehereachesthispartofhisaccount,HutchesonhasarguedthatthemoralsensecomprisesbothaninherentbenevolenceabletodirectourActions,andaninnatedisinterestedultimateDesireoftheHappinessofothers.29Itisthisaspectofthemoralsensethatmotivatesustoparticipateinsociety,topursuethepublicgood,andtotakepleasureinthereal-izationofthatgood.Hadwelackedthisdispositionwemighthavedevelopedanabstractorspeculativeideaofvirtue,butourconcernforourowninterestwouldhavecausedus,contrarytofact,tobeconcernedonlywiththisinterestandtoapproveonlythoseagentsandactionsthatservethisinterest.Hutchesonalsofindsthatthemoralsenseincludesacognitivefunction,apowerthatenablesustoresponddifferentlytobenev-olenceandself-interestandtheactionstheymotivate.Thehumanmindisformedinsuchawaythatitcanandoftendoesapproveorcondemnactionsoragentswithoutconcernforitsownpleasureorinterest.Thusiftwoindividualscontributeinsimilarwaystoourwell-being,buttheoneactsfromanultimateDesireofourHap-piness,orGood-willtowardus;andtheotherfromViewsofSelf-Interest,orbyConstraint,weresponddifferently.30Inresponsetothefirstwefeelgratitudeandapprobation;tothesecondweareindifferent.Or,ifweknowthatanindividualhasbenevolent28Inquiry,titletoPartII,Sect.V,218.29Inquiry,129,152.30Inquiry,113.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n284davidfatenortondispositions,buthasbeenpreventedfromexercisingthese,weapproveofherandcountherasmorallygoodeventhoughshehasnotbeenabletoact,eventhoughshehasdonenothingtobene-fitus.Thenatureandcomplexityoftheseresponsesshowthatwehaveaperceptualpower,asenseofmoraldiscrimination,forwith-outsuchasensewewouldassessfieldsandagents,orpatriotsandtraitors,inthesamewayandonlywithregardtoourowninterestsandwell-being.BecauseHutchesonheldthatthemoralsensediscriminatesbetweenvirtueandvicebymeansoffeelingsofapprovalanddisap-proval,someofhiscriticssupposedthathehadreducedvirtueandvicetothesefeelingsorsentiments.Hutchesonexplicitlydeniedsuchanintentandsucharesult.Themoralsensereliesonfeel-ingstodistinguishvirtueorvice,butmoralqualitiesarethemselvesindependentoftheobserverwhofeelsapprobationordisapprobationofthem.TheadmiredQuality,hesays,isaqualityoftheagentjudged,andentirelydistinctfromanyapprobationorpleasurefeltbyeithertheapprovingobserverortheagenthimself.Themoralper-ceptionsorapprobationexperiencedplainlyrepresentssomethingquitedistinctfromthisPleasure.31Feelingsplaybothamotivatingandacognitiverole,butvirtueisconstitutedbythebenevolentdis-positionthatgivesrisetoapprobation,whileviceisconstitutedbythemalevolentorindifferentdispositionsthatgiverisetodisappro-bation.Virtueandvice,althoughknownbymeansofthefeelingstheyarouse,arereal,observer-independentqualitiesofagents.ii.hume,humannature,andthefoundationsofmoralityAswehaveseen,Humewasawareofthecontroversyaboutthefoun-dationsofmorals.Andalthoughhevoicedsubstantialdisagreementwithhispredecessors,hedidagreethatmoralityhasitsfoundationinhumannatureandindifferencesofmotivation.Inthespringof1734,hewroteintheautobiographicalletterincludedattheendofthisvolumethathehadfoundthemoralphilosophyoftheancientstoospeculativetobeofanyvalue.Thesephilosophersformedmoral31Inquiry,1301.Hutchesonaddedthiscommenttothethirdedition(1729)ofhisInquiryinordertopreventwhathetooktobemisunderstandingsofhisposition.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise285systemsbyrelyingontheirinventiveimaginations,andwithoutregardforhumanNature,thefoundationuponwhicheverymoralConclusionmustdepend.Consequently,heresolvedtomakehumannaturemyprincipalStudy,&theSourcefromwhichIwouldderiveeveryTruthin...Morality(KHL6).Whenin173940HumepublishedtheTreatiseofHumanNature,hedescribeditbyitssubtitleasanattempttointroducetheexperimentalmethodofreasoningintomoralsubjects.IntheIntroductiontothisworkhesaidthatallthesciences,butespeciallythehumansciencesofLogic,Morals,Criticism,andPolitics,arefundamentallydependentontheknowledgeofman.Sofarasthesesciencesareconcernedthereisnoquestionofimpor-tance,whosedecisionisnotcomprisdinthescienceofman;andthereisnone,whichcanbedecidedwithanycertainty,beforewebecomeacquaintedwiththatscience(TIntro.46).Humealsosaidthat,justasthescienceofmanistheonlysolidfoundationfortheothersciences,sotheonlysolidfoundationwecangivetothisscienceitselfisexperienceandobservation.Healsosupposedthatanyscienceofhumannaturethatattemptedtocarrytheexplanationofthatnaturetounobservedprinciplesorcausesallegedlymoreultimatethanthisnatureitselfasitisobservedwouldbeapoorscienceindeed.32Tosaythatmoralityisfoundedonhumannatureistosuggestthat,withrespecttomorals,humannatureisaprimitiveelement,anultimatefact,beyondwhichexpla-nationcannotgo.Hesaw,however,thathecouldnotgainasoundknowledgeofhumannaturebyundertakingthekindofcarefullyplannedexperimentsmadefamiliarbynaturalphilosophy.Hesawthatwemustcollectourexperimentsinthescienceofhumannaturethroughacautiousobservationofhumanlife,by,thatis,observ-inghumanbehaviorinthecommoncourseoftheworld.Whenexperimentsofthiskindhavebeencarefullycollectedandcom-pard,heargues,wemayhopetoestablishonthemascience,whichwillnotbeinferiorincertainty,andwillbemuchsuperiorin32LaterintheTreatiseHumesaysthatthecausesofeffectsinthementalworldaremostlyunknown,andmustberesolvdintooriginalqualitiesofhumannature,whichIpretendnottoexplain,andalsomentionstheparticularoriginalprinci-plesofhumannature,whichcannotbeaccountedfor(T1.1.4.6,3.3.1.27;seealsoEPM,n.19).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n286davidfatenortonutilitytoanyotherofhumancomprehension(TIntro.710).But,isolatedashewasintheScottishcountrysideortheonlymarginallywidersocietyofferedbyEdinburgh,Scotlandssmall,provincialcap-ital,andthesmallFrenchtown(LaFleche)wherehewrotemuchof`theTreatise,itwasbooks,booksofphilosophy,literature,history,andpoliticsthatgaveHumeawiderwindowonhumannature.33ThroughthiswindowhesawthatItrequiresbutverylittleknow-ledgeofhumanaffairstoperceive,thatasenseofmoralsisaprincipleinherentinthesoul,andoneofthemostpowerfulthatentersintothecomposition,andhesetaboutdescribingtheprincipalfeaturesofthispowerfulsense(3.3.6.3).MoralDistinctionsNotDerivedfromReasonAshehadpromisedinTreatise2.1.7,HumetakesupthedisputeregardingtheoriginofmoraldistinctionsinOfMorals,thethirdandfinalbookoftheTreatise.Hebeginsbyremindinghisreadersthatperceptionsarefoundinonlytwoforms,impressionsorideas,andthenaskingwhichofthesetwoformsenablesustorecognizemoraldifferencesandtomakemoraljudgments(T3.1.1.2).34Inrais-ingthisquestion,Humedirectlyconfrontstherationalistsaccountofmorality.Henotesfirstthatthosewho,likeClarke,claimthatmoralityrestsonakindofconformitybetweenactionsandcertainunchangingrelationsofthingsineffectclaimthatmoraldistinctionscanbetracedtoideasandtheirrelations,andthusimplicitlyclaimthatitisreasonalonethatenablesustomakemoraldistinctions.Drawingonimportantfactsthatcharacterizemoralityasitisprac-ticed,Humechallengesthisaccountwithabatteryofargumentsintendedtoshowthatmoraldistinctionsareneithertheimmediateeffectofreason,norknownbyreasonalone,andthatasaconse-quencewemustconcludethatmoraldistinctionsarenotderivedfromreason(T3.1.1).Ofthesearguments,thefollowingarerepre-sentative.1.Theclaimsoftherationalistsareinconsistentwiththefactthattheveryactofmakingamoraldistinctioninfluencesourbehavior.33OnHumesearlyeducationandreading,seemyintroductoryessay,n.1.34ForanaccountofHumesviewsonperceptions,impressions,andideas,seeinthisvolumetheessayHumeandtheMechanicsoftheMind.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise287Moraldistinctionsdirectlyandbythemselvesarousepassionsandmotivateustoaction.ButHumehadearlierarguedthatreasoniseffectivelyinertandthatanyinfluenceithasonthewillisindirectandmerelyinstrumental.35Reasoncaninformusofthenatureofthingsandhowtoobtainthem,butitcanneitherarousenorextin-guishdesireforthem.Inshort,reasonisunabletobringabouttheeffectsattributedtoitbytherationalists.Consequently,moralitycannotbefoundedonreasonalone(T3.1.1.48,10).2.Theclaimsoftherationalistsareinconsistentwiththefactthatmoralassessmentsadmitofdegrees.Sofarasrationalisttheoryisconcerned,thetheftofapieceoffruitandthetheftofakingdom,actionsequallyinconsistentwiththerelationssaidtoholdbetweenownersandtheirproperty,shouldbejudgedequallyimmoral.Infact,wetakethesecondtobesignificantlymorereprehensiblethanthefirst(T3.1.1.13).3.Theclaimthatreasonaloneiscapableofmakingmoraldis-tinctionsentailsthatsuchdistinctionsareeitherderivedfromsomedistinctiverelationofideasorfromsomematteroffactdiscoverablebytheunderstanding.Ourmoralexperienceshowsthatmoraldis-tinctionsarenotderivedineitheroftheseways.First,suchdistinc-tiverelationsofideascannotbefound.Preciselythesamerelationthatcharacterizeshumaningratitudeintheformofparenticide,ofallhumancrimes...themosthorridandunnatural,istakentobemorallyinnocentwhenfoundamongotherlivingspecies.Incest,totakeasecondexample,istakentobeviciouswhenengagedinbyhumans,yetthesamerelationscharacterizeanimalinstantiationsofthispractice,andthesearefoundtobemorallyinnocent.Inaddition,therationalistsinsistthatmorality,likemathematics,iscapableofdemonstration.Thattheyholdsuchaviewshowstheydonottakemoralitytobeafactualmatter,foritisuniversallyagreedthatitisprobablearguments,notdemonstrations,thatestablishmattersoffact.Moreover,howevercloselyourreasonorunderstandingana-lyzesanactiontakentobevirtuousorvicious,weneverdiscovervirtueorviceamongthefeaturesofthataction.Usingthesenses,aprincipaltooloftheunderstandingtoexamineacaseofwillfulmurder,weneverdiscovervirtueorviceamongthefeaturesassoci-atedwiththisact.Wemayinitiallylearnaboutthisactionbymeans35SeeT2.3.3,Oftheinfluencingmotivesofthewill.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n288davidfatenortonofimpressionsofsensation,andtheremaybesomewaysinwhichviceandvirtueresemblecolororheatorcertainideasofsensation,butHumeexplicitlydeniesthatvirtueandviceareknownthroughimpressionsorideasofsensation(T3.1.1.1727).MoralDistinctionsDerivedfromaMoralSenseIf,asHumesays,perceptionsareofonlytwoforms,impressionsandideas,andifanalysisorcomparisonsofideasarenotthesourceofthedistinctionwemakebetweenvirtueandvice,thenitmustbebymeansofimpressionsthatwemakethisdistinction.But,fromHumesbriefanalysisofthemorallysignificantactofwilfulmurder,wealreadyknowthatviceandvirtuedonotproduceimpressionsofsensation.Consequentlythesedistinctivemoralqualitiesmustmakethemselvesknownbymeansofimpressionsofreflectionorfeelings.Morality,therefore,Humesays,ismoreproperlyfeltthanjudgdof(T3.1.2.1).Treatise3.1.2,Moraldistinctionsderivdfromamoralsense,explainshowthishappens.Itexplains,inotherwords,howcertainimpressionsofreflectionorfeelingsenableustomakemoraldistinctions.Treatise3.1.2.beginsbydrawingourattentiontothefactthatthosethingsthatwecallvirtuousproduceinthosewhoobservethemanagreeableimpressionorfeeling,apleasure,whilethosethingscalledviciousproduceinthosewhoobservethemadisagree-ableimpressionorfeeling,apain.Humethenpointsoutthatitisnotthecasethateveryitemdesiredbecauseitgivespleasureistakentobemorallygood,orthateveryitemthatproducespainistakentobemorallyevil.Therelevantfeelingsareparticularordistinctivepleasuresandpains.TheyarethosepleasuresandpainsthatHutch-esonandHumeoftencallapprobationanddisapprobation,andtheyareonlyarousedinuniquecircumstances.Thatthisissomeans,Humesays,thatexplaininghowwecometofeeltheseuniquemoralsentimentswillbeequivalenttounderstandingmoraldistinctionsthemselves.Ifwecallamotiveoractionvirtuousorviciousbecauseitsviewcausesapleasureoruneasinessofaparticularkind,thenbyexplainingthosedistinctivepleasuresandpainswewillalso,hesays,sufficientlyexplaintheviceorvirtuethatcausesthem(T3.1.2.3).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise289Inthecourseofexplainingthemoralsentiments,Humemakesthreeimportantclaims:1.Therearemanydifferentkindsofpleasureandpain.Theplea-sureproducedbygoodmusic,forexample,isdifferentfromthatproducedbygoodwine.Thepleasuresandpains(themoralsenti-ments)producedbythemotivesoractionsofhumansaredifferentfromthoseproducedbyotherkindsofthing.2.Itisonlythequalitiesoractionsofpersonsthatgiverisetotheparticularpainsorpleasuresknownasthemoralsentiments.Humealsosaysthatvirtueandviceconsistentlyexciteoneofthefourindirectpassions(pride,humility,love,orhatred)thathaveoneselforotherpersonsfortheirobjects.Drawingonhisearlierdiscussionofthepassions,hefurtherrefineshisview,sayingthatitisonlycertaindurableprinciplesofthemind(motives,includingcharacterandintentions)thatgiverisetothemoralsentiments,andthatactionsthemselvesareneithervirtuousnorvicious,butonlysignsofthevirtueorvice(thesignsofvirtuousorviciousqualities)inthosewhoperformthem(T3.1.2.3,5;3.3.1.35).363.Theseuniquemoralsentimentsarearousedonlyinveryspecialcircumstances.Wedonot,forexample,experiencedisapprobationeverytimesomeindividualsdurableprinciplesofmindcausesomeformofpain.Theuniquesentimentofdisapprobationisfeltonlyifweabstractfromorignoreourowninterest:Tisonlywhenacharacterisconsiderdingeneral,withoutreferencetoourparticularinterest,thatitcausessuchafeelingorsentiment,asdenominatesitmorallygoodorevil.Thus,forexample,althoughthegoodqualitiesofanenemy,hisvalororcourage,maycauseusharm,thesequalitiesmaystillcommandouresteemandrespect,andwe36HumescleareststatementoftheviewthatactionsaresignsofmotivesisperhapsthatmadeatT3.2.1.2:Tisevident,thatwhenwepraiseanyactions,weregardonlythemotivesthatproducdthem,andconsidertheactionsassignsorindica-tionsofcertainprinciplesinthemindandtemper.Theexternalperformancehasnomerit.Wemustlookwithin[theagent]tofindthemoralquality.Thiswecan-notdodirectly;andthereforefixourattentiononactions,asonexternalsigns.Buttheseactionsarestillconsiderdassigns;andtheultimateobjectofourpraiseandapprobationisthemotive,thatproducdthem.Seealso,e.g.,T2.1.2.2;2.2.1.2;2.2.3.38;2.3.2.6;3.2.1.4,8.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n290davidfatenortonmaystilljudgethisenemytobehonorableandvirtuous(T3.1.2.4;seealso3.3.1.3).Imagine,then,thatyouobserveaspecificcaseofintentionalkilling.OnHumesviewofthematter,asaconsequenceofexpe-riencingthisaction,youwillfirstexperiencecertainimpressionsofsensationthatinformyouofthephysicalfeaturesoftheevent.Supposethatthesesensationsinformyouthatoneperson,A,haspointedapistolatanotherperson,B,demandedBswallet,and,whenBresisted,shothim.FurtherimpressionsofsensationinformyouthatBhasfallenlifelesstotheground.Noneoftheseimpressionsofsensationhasamoraldimension.NoneofthemevensuggeststhatAsactionhasmoralconnotations.Butinresponsetothisextendedactionyoucanalsoexpecttohave,amonganumberofimpressionsofreflection(shock,pity,andfear,forexample),aparticularimpres-sionofreflection.Youcanalsoexpecttoexperiencethedistinctiveimpressionofmoraldisapprovalordisapprobation.Ifthatimpres-siondoesariseitwillmeanthatyouhavedeterminedthatA,thepersonwhocarriedoutthekillingaction,isviciousormorallyevil.Moreover,youwillhavemadethisdeterminationsimplybecauseyouhavehadthisdistinctivemoralsentiment.Youwillhavenoneedtomakeuseofreasoninordertocompare,step-by-step,theactiontoaruleofbehavior,findthattheactionviolatesthatrule,andthenconcludethatasaconsequencemoralevilhasbeenpresentandobserved.37Shoulditbeaskedwhy,incontrast,aleopard,seentohavedeliberatelystalkedandkilledagazelle,isnotsaidtobevicious,Humewouldsaythatthissecondaction,becauseitdoesnotinvolveahumanagent,doesnotarousedisapprobation,thedis-tinctivemoralsentiment.Hedoesnotattempttoexplainwhythisshouldbethecase.Hesimplytakesittobeafactofournature(afactillustratedbyourmoralpractice)thatourmoralsenseisarousedbycertainfeaturesoractionsofhumans,butisnotarousedbytheapparentlyanalogousfeaturesoractionsofanimals,plants,orotherexternalobjects.3837WehaveseenthatHumedeniesthatvirtueandviceareknownbymeansofthesenses,thetoolbywhichtheunderstandingattemptstodiscovermattersoffact.38SeeT3.1.2.4.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise291iii.thevirtuesA.AComparisonofKindsWesawabovethatwhenHumefirstmentionedthedebateoverthefoundationsofmoralityhedescribeditasadebatebetweenthosewhosupposethatmoraldistinctionsarebasedonnaturalandorig-inalprinciples,andthosewhosaythattheyarisefrominterestandeducation.AlthoughinBook3Humearguesthatmoralityhasasubstantivefoundationinhumannature,healsoarguesthattherearetwokindsofvirtue,naturalandartificial,andthatwhileinimportantsensesallmoraldistinctionsderivefromnaturalandoriginalprinciplesofhumannature,theartificialvirtues(justice,fidelity,allegiance,treatykeeping,andchastityarediscussed)areatfirsttheproductofself-interestandeducation.Treatise3.2focusesontheartificialvirtues,while3.3focusesonthenaturalvirtuesandvices.ButbeforeturningtoHumesextendeddiscussionsofeachofthesekindsofvirtue,itwillbeusefultosketchthesekindsandtheprincipaldifferencesbetweenthem.Thenaturalvirtues:a.Areinherentfeaturesofhumannature(T3.3.1.1)b.Arespecificpassionsthathavealwaysmotivatedspecifickindsofhumanbehavior;examplesincludegenerosity,humanity,loveofchildren,greatnessofmind,compassion,gratitude,friendship,anddisinterestedness39c.Aremotivatingpassionsthataresaidtoproducegoodoneachoccasionthattheymotivatebehavior(T3.3.1.12)4039SeeT3.2.1.5,3.2.5.6,3.3.2,3.3.3.3.Therearealsonaturalvices.Theseincludeingratitude,cruelty,villainyorbaseness,andinhumanity;seeT3.1.1.24;3.1.2.2,4;3.3.3.89.Foranotherdiscussionofthetwokindsofvirtue,seeinthisvolumetheessayTheStructureofHumesPoliticalTheory,PartII.40GiventhatHumein3.3.1.12goesontosaythatoneofthemostimportantnaturalvirtues,humanity,maycomeintoconflictwiththeestablishedschemeofjus-tice,itappearsthatonhisaccountitcouldhappenthatpursuinganaturalvirtuewouldundercuttheschemeofjusticeandwouldthusnot,allthingsconsidered,producegood.Afathersloveofhischildrencouldalsohavethiseffect.Thisprob-lemismoredirectlyaddressedinthesecondEnquiry;App.3.26(SomeFartherConsiderationswithRegardtoJustice).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n292davidfatenortond.Produceinobserverspositivesentimentsofmoralapprovalwhenevertheyareseentomotivatethebehaviorofanotherperson(T2.1.7.5,3.1.2.35).Incontrast,theartificialvirtues:a.Derivefromhumannature,andthusareinthatattenuatedsensenatural,butthesevirtueswereunknowntohumanslivingintheirfirstrudeandmorenaturalcondition(T3.2.1.9)b.Derivefromournaturalself-interest,asithasbeenmodi-fiedbycontingentcircumstancesandnecessities,andhavedevelopedoverthecourseoftimeinresponsetojustsuchcircumstancesandnecessities(T3.2.2.122)c.Constituteasystemofrulesorconventionstowhichitisnecessarythatweconformforthepublicgood,butwhichmayyetonanygivenoccasionrequireactionsorbehaviorcontrarytobothindividualandpublicgood;insuchcases,therelevantvirtuousactsproduceonlyweaksentimentsofapproval(T3.3.1.12)d.Are,onceestablished,naturallyattendedwithastrongsen-timentofmoralsowingtooursympathywiththeinterestsofsociety(T3.3.1.12).41Forhumansintheiroriginaloruncultivatedstate,theartifi-cialvirtuesandtheconstraintstheyintroducewereunnecessary.Inthatoriginalstatethenaturalvirtueswereadequatetomain-tainorderinthesmall,kinship-basedgroupsthathumansformed.Ashumansocietybecamelargerandmorecomplex,circumstanceschanged(somematerialgoodsbecamescarce,forexample),andthesechangesledtoconflictswithinorbetweentheexistingsocialunits,conflictsthatthenaturalvirtueswereunabletoresolve.Asacon-sequence,conventionsregulatingfirstproperty,thenpromises,andthengovernmentwere(gradually)developed.41Therearealsocorrespondingorparallelartificialvices.Theseincludethecon-trariesoftheartificialvirtues,e.g.,injustice,infidelity,treason,licentiousness,andcowardice,aswellasill-breedingandoverweeningpride(seeT3.2,passim,3.3.2.10).HumeoffersasimilarcomparisonbetweennaturalandartificialdutiesinOftheOriginalContract,anessayfirstpublishedin1748;seeE-OC335,47980.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise293Humesaccountoftheartificialvirtuesundertakestoexplainthreerelatedmattershavingtodowiththisdevelopment:theoriginoftheseconventionsinself-interest;thedevelopmentofmoralmotivesadequatetoproduceactionsconformingtotheseconventions;andthefactthatwenowhavesentimentsofappro-bationinresponsetoactionsthatconformtotheseconventions,anddisapprobationinresponsetothosethatfailtoconformtothem.B.MotivesandMoralQualitiesHumebeginshisexplanationoftheartificialvirtueswithadiscus-sionoftherelationshipbetweenthemotivesofanagentsactionsandthemoralcharacterofthatagent.Aswesawabove,Humesup-posesthatthemoralsentimentsariseinresponsetotheactionsandmotivesofhumanagents.In3.2.1(Justice,whetheranaturalorartificialvirtue?)heagainarguesthat,althoughactionsmayappeartobevirtuousorvicious,theyareinrealityonlysignsofcertainprinciplesinthemindandtemper.Theseprinciples,theunder-lyingfeaturesofmindofwhichactionsaresigns,arethemotivesthatcauseagentstoperformtheactionsinquestion.Giventhatitisthesemotivesthataretheobjectsofmoralassessment,itfollowsthattosayofagivenindividualthatheorsheisvirtuousistosaythisindividualhasmotives(passionsanddesiresaswellasdispo-sitions,intentions,andcharacter)ofadistinctivekind(T3.2.1.2,quotedinfull,note36).Inmakingthisclaim,Humeisagreeingwithmanyotherphiloso-phers.WehaveseenthatShaftesbury,Mandeville,andHutchesonagreethatthemotivesofagentsdeterminethemoralcharacteroftheactionsperformedbythoseagents.42Thesemoralistsagreedthatanactionismorallygoodonlyifithasbeenproducedbyanother-regardingmotive,withtheobviousconsequencethatanaction42HumetracedthisviewtoCicero,whomhefoundtohavearguedthatitisontheGoodnessorBadnessoftheMotivesthattheVirtueoftheActiondepends(LettertoFrancisHutcheson,17Sept.1739;Letters1:35).ForCicerosdiscussion,seeDefinibusbonorumetmalorum[AbouttheEndsofGoodsandEvils]4.1617.438.Forsomeofthemanyotherswhotookthisview,seeEditorsAnnotations,inATreatiseofHumanNature,ed.D.F.NortonandM.J.Norton,2vols.(Oxford:ClarendonPress,2007),2:845,8967,949(annotations226.1,307.7,373.19).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n294davidfatenortonmotivatedbyself-interestisnevervirtuous.Suchanactionmayben-efitsomeoneorthepublicgenerally,butbenefitaloneisnotenoughtomaketheagentwhoperformedtheactionvirtuous.GiventhatHumemaintainsthatactionsinthemselvesarealwaysmorallyneu-tral,andthatvirtuousactionsderivetheirvirtueonlyfromvirtuousmotives,andareconsiderdmerelyassignsofthosemotives,itisclearthathealsosupposesthatactionsaremorallygoodonlywhenmotivatedbyaregardforothers(T3.2.1.2,4).43ButHumespositiononthisissueissubstantiallydifferentfromthatofhispredecessors.Althoughheneverabandonstheviewthatthemoralcharacterofactionsisdeterminedbytheirmotives,wewillseethathemaintainsthattheartificialvirtuesareinitiallymotivatedbyself-interest.Itisonlyoverthecourseoftimethatthesevirtuescometobemotivatedbyaregardforothers,andthustomeetHumesstandardforamoralvirtue.C.JusticeHumebeginshisdiscussionoftheartificialvirtueswithjustice,theconventionsgoverningproperty.44Imagine,hebegins,thatsomeonehasborrowedasumofmoney,andsaidthathewillreturntheequiv-alentamountbysomespecifiedlaterdate.Thisdatehasnowbeenreached,buttheborrowerasks,WhatreasonormotivehaveItorestorethemoney?We,farremovedfromthetimewhenjusticeoriginated,wouldsaythatjusticeobligesonetorepayaloanwhenitisdue,andapersonaskingthequestionnowwouldunderstandthatanswer.Humesexample,however,isintendedtodirectouratten-tiontoanearliertime,atimewhenhumanitywasinthatrudeandmorenaturalconditiontowhichherefers.Thereplywewouldnowmake,ifmadetoaborrowerstillinthatrudestate,wouldmakenosensetothatperson.Thathypotheticalpersonwouldreply,Hume43Humeobservesthatactionsmaybethoughttohavemoralcharacterbecausewehaveatendencytofocusonthemratherthanonthemotivesfromwhichtheyderive.Asaconsequence,moralassessmentsaredirectedtotheactions.AsHumeputsit:Actionsareatfirstonlyconsideredassignsofmotives:buttisusual,inthiscase,asinallothers,tofixourattentiononthesigns,andneglect,insomemeasure,thethingsignifyd(T3.2.1.8).44InHumesaccount,therulesofjusticehavetodoonlywithexternalgoodsorproperty(seeT2.1.10.1,3.2.2.7).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise295suggests:Whatdoyoumeanbyjustice?Andwhatdoesthisjustice,ofwhichIhaveneverheard,havetodowithme?InthiswayHumeforcesustoaskstillanotherquestion:Insuchamorallyundevel-opedenvironment,whatmotivecouldtherebetorepaytheloan?Hisanswertothislastquestion:None.Inthosecircumstancestherecouldbenomotivetorepaytheloan(T3.2.1.9).Humeundertakestoestablishthispointbyarguingthatthereareexactlythreeoriginalmotivestoaction:self-love,oraconcernforonesownprivateinterest;generalbenevolence,oraconcernforthewell-beingofothersgenerally;andprivatebenevolence,oraconcerntodowhatbenefitssomeotherindividualorlimitedsetofindividuals.Humemaintainsthatnoneofthesemotivesisanatural,universalmotivetobejustorafoundationofthevirtueofjusticeaswenowknowit.Insupportofthisconclusionhepointsoutthatself-loveappearstomotivatemoreunjustactionsthanjustones,45thatthereisinhumannaturenogeneralregardforthewell-beingofothers(wearebynaturegenerous,butthescopeofthisgenerosityissignificantlylimited),norissucharegardessentialtoactsofjustice,andthatprivatebenevolenceissometimesbestservedbyviolatingtherulesofjustice.Thereis,then,noeffectiveanswertothefirstquestionposedbytheuncultivatedindividual.Intherudeoruncultivatedconditionenvisaged,thereisnomorallycompellingreasonormotivetorepaytheloan.Toconcludethatamorallyuncultivatedpersonwouldneitherunderstandwhatjusticeis,norhaveaneffectivemotivationtoactjustly,istoconcludethatjustice,giventhatitisnowavirtue,isanartificialvirtue.Itistoconcludethatoursenseofjusticeandinjustice(oursensethatactingjustlyistherightwaytoact)isnotderivdfromnature,butarisesartificially,thonecessarilyfromeducation,andhumanconventions(T3.2.1.917).45Humealsoarguesthat,beforetheconventionsofjusticeweredeveloped,actinginaccordancewiththeimpartialitydemandedbytheseconventionswouldhavebeenseentobevicious.Inouroriginalstate,self-interestmotivatesustoactpartiallyorselfishly.Moreover,thepartialitywefeelinthatoriginalstateleadsustoexpectsuchpartialbehavior,andtotreatanysignificantdeviationfromthisexpectationaswrong:suchpartiality,hesays,hasaninfluencenotonlyonourbehaviourandconductinsociety,butevenonourideasofviceandvirtue;soastomakeus[wheninthatstate]regardanyremarkabletransgressionofsuchadegreeofpartiality...asviciousandimmoral(T3.2.2.8).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n296davidfatenortonHavingreachedthisimportantconclusion,Humeturnshisatten-tiontotworelatedquestions.Howisitthathumanshavedevelopedtheconventionsofjustice?Andwhyisit,givenwhatwehavelearnedabouttheoriginoftheseconventions,thatwetreattheirobservanceorneglectasamoralmatter?InresponsetohisfirstquestionHumeemphasizeshumanitysnativelyperilouscondition.Ofallanimals,individualhumansap-peartohavethefewestnaturaladvantagesinproportiontotheirneedsanddesires.Individually,humansareweak,inept,andinconstantdangeroflosingwhatevermaterialgoodstheymayhaveacquired.Itwasonlybyjoiningforcesthathumanscouldremedythesedeficiencies.Thatis,itwasonlybyformingsocietiesthathumanscouldenhance,astheyneededtodo,theirstrength,abilities,andsecurity.Humeassumesthatourmostremoteancestorshadnoexperienceofsocieties,noexperienceof,thatis,convention-orrule-governedsocialunits.Fortunately,thedevelopmentandrecognitionoftheadvantagesofsuchunitsdependsonanineradicablefeatureofhumannature,sexualappetite.Rule-governedsocialgroupswerenotapartofourrudeanduncultivatedbeginnings,butsuchgroupsemergednaturallyasaconsequenceoftheprocessofsocializationthatbeginswithsexualappetiteandleadstofamilies,themselvesminiaturesocialunitsorproto-societies.Inthebeginning,then,apowerfulandentirelynaturalbutamoralmotivegaverisetosmallproto-societiescharacterizedbyasignificantdegreeofcooperation.46Thisinitialdevelopmentthenledsomeofourancestorstorealizethattheleadingsourceofconflictbetweentheirproto-societieswasdisputesaboutexternalgoods.Somealsorealizedthatitwouldbeintheinterestofeachindividualandeachfamilytoreducetheseconflicts,andthatthisdesirableendcouldbeachievedbyadoptingconventionsorrulesthatstabilizedthepossessionofsuchgoods.46Humesuggeststhatintherudeandnaturalconditionofwhichhespeaks,noteventherudimentsofjusticewerepresent,needed,orknown.Hecomparesthesituationtothatofclosefriendsandmarriedcouplesortocircumstancesinwhichthereisagreatplentyofsomeitem:Tiseasytoremark,thatacordialaffectionrendersallthingscommonamongfriends;andthatmarrydpeopleinparticularmutuallylosetheirproperty,andareunacquaintedwiththemineandthine,whicharesonecessary,andyetcausesuchdisturbanceinhumansociety.Thesameeffectarisesfromanyalterationinthecircumstancesofmankind;aswhenthereissuchaplentyofanythingassatisfiesallthedesiresofmen:Inwhichcasethedistinctionofpropertyisentirelylost,andeverythingremainsincommon(T3.2.2.17).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise297OnHumesaccount,theunreflectiveself-interestthatcharacterizeshumanityswilduncultivatedstateandthatinhibitsthedevelop-mentofgenuinesocietieswasgraduallyrestrainedandredirected.Butitwasself-interestitself,apartiallyenlightenedself-interest,thatbroughtaboutthisimportantchangeandmadesocietypossi-ble.Theinitialconventionsofjusticeweretheresultofindividualstacitlyrecognizingthattheywouldeachgainbyincludingintheirroutinebehaviorcertainrestraintsregardingproperty,providedonlythatotherswoulddothesamething(Irefrainfromtakingyourgoodsaslongasyourefrainfromtakingmine).47Infact,therestraintsandexpectationsthatproducedthepeaceandorderofsocietywereartic-ulatedasexplicitrulesofpropertyonlyaftertheyhadbeguntohaveeffect(T3.2.2.122).Ifthisaccountoftheoriginsofjusticeiscorrect,whathavewelearnedfromit?First,Humehasshownthatjusticeistheprod-uctofahistoricalprocessthatdependedonunwittingcontrivance.Second,hehasshownthattheconventionsofjusticearosebecausecertaincontingentconditionsmadehumanwell-beingorsurvivaldependentonthem.Hadthoseconditionsbeendifferent(hadnotnecessarymaterialgoodsbeeninrelativelyshortsupply,forexam-ple),rulesofpropertywouldnothavebeennecessaryforpeaceandorder.Evennow,weretheretobeabountifulsupplyofthesemate-rialgoodsorwerehumansallgenerouswithoutlimit,justicewouldalterorwitheraway.Humehasalsoconfirmedthatjusticeisnotfoundedonideasortheirallegedlyeternalandimmutablerelations.Itwasonlyafterthepracticeofmutualrestrainthadresultedintherelevantconventionsthattheideasofjusticeandinjustice,aswellastheideasofproperty,right,andobligation,wereformed.Theideasandtherelevantrelationsdonotprecede,butfollowafter,experience.Theportionofhistheoryjustoutlinedexplains,Humesays,ournaturalobligationtojustice,viz.interest.Humeissatis-fiedthathehasshownwhy,astheresultofthegradualprocesshehasdescribed,individualhumanshavegainedaneffectivemotive,enlightenedself-interest,toestablishtheconventionsofjustice,amotivethattheuncultivatedmembersofourspecieslacked.Hehas47IndirectoppositiontoHobbesandLocke,Humeinsiststhatthisdevelopmentwasnottheresultofapromiseorsocialcontract;seeT3.2.2.10.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n298davidfatenortonalsoshownwhythesesameindividualshavecometofeelthattheyoughttoactjustly.Hehasexplainedhowweasaspeciesdevelopedconventionstowhichwearenowmotivatedtoconform,andthushowwecametofeel,whenwefindourselvesincertaincircum-stances,that(forself-interestedorprudentialreasons)weoughttoconformtothoseconventions,ortofeelthatweoughttoactjustlywherejustlymeansinourownenlightenedself-interest.Wehavedevelopedconventionsthatserveourinterestsandthatalsocallforeachindividualtokeepuphisorherpartinwhatisatleastatacitagreementtorespectpropertyrights.ButHumegoesontosaythathemuststillexplainwhyitisthatwetreattheobservanceorneglectoftheconventionsofjusticeasamoralmatter.Givenhiscommitmenttotheviewthatmotivesdeterminemoralmerit,andthatonlyother-regardingmotivessecuremoralapprobation,hestillmustexplain,Whyweannextheideaofvirtuetojustice,andofvicetoinjustice?orhowithappensthat,inmatterspertainingtojusticewedevelopamoralobligation,orthesentimentofrightandwrong(T3.2.2.23).WhenHumeequatesthemoralobligationtojusticewiththesentimentofrightandwrong,heremindsusthatoneoftheaimsofTreatise3istodiscoverthefoundationsofmorality,andthattheprecedingaccountofthedevelopmentofjusticeomitsanessentialpartofthatstory.Humesawthedevelopmentofthevirtueofjusticeaspassingthroughthreestages.Heisnowabouttosketchthethirdandfinalstage.WhatHumedescribesasouroriginalorwilduncultivatedstatewouldhavebeenthefirststageinthedevelopmentofjustice.InthatportionofTreatise3.2.2justoutlined,hehasexplainedourmovementfromaninitialstageinwhichtherewerenoconven-tionsofjusticeandinwhichanyreferencetosuchconventionswouldhavebeenunintelligible,toasecondstage.Aswehaveseen,thissecondstageischaracterizedbythefactthattheconventionsdefiningandprotectingpropertyhavebeendeveloped.Tosaythisistosaythatwhengroupsofindividualshavereachedthisstagewhengenuinesocietieshavebeendevelopedindividualscompos-ingthesesocietiesroutinelyrespecteachotherspossessions.Theyalsohaveaworkingunderstandingoftheconceptsofjustice,injus-tice,andproperty,aswellasthesensethattheyasindividualshaveanobligationtorespectthepropertyrightsofothers.ThesenewlyCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise299developedconventionshavealsogainedanaction-guidingforce.AsHumedescribesthesituation,Afterthisconvention,concerningabstinencefromthepossessionsofothers,isenterdinto,andeveryonehasacquirdastabilityinhispossessions,thereimmediatelyarisetheideasofjusticeandinjustice;asalsothoseofproperty,right,andobligation(T3.2.2.11).Thus,whileindividualsinthefirststagewereunabletounderstandtalkaboutjusticeorpropertyrightsandconsequentlycouldnotknowwhatitmeanttorespectsuchrights,someoftheirdescendantswerecapableofunderstand-ingthesekeynotions.Theseindividualsalsounderstoodthatbeingjustmeansthatoneconformsonesbehaviortotheconventionsofjustice,thatconformingtotheseconventionsisinonesbestinter-est,and,consequently,thatthisistheprudentthingtodo.Inshort,individualswhohavereachedthesecondstageofthisdevelopmenthaveacquiredimportantnewcapabilities.Theyhavedeveloped,andthustheyunderstand,theconventionsorrulesofjustice.Theyhavealsodiscoveredamotive(enlightenedself-interest)formaintainingtheserules.Finally,theseindividualsexperienceaformofapprovalinresponsetoactionsthatconformtotheserules,andaformofdisapprovalinresponsetothosethatdonot.Stillremainingtobeanswered,however,isHumessecondquestion:Whydoactsofjus-ticeandinjusticearousethesesentimentsofapprovalordisapproval,andwhatcircumstanceseventuallycausetheseapprovalstotaketheformofthedistinctivemoralsentiments?Theinitialmotivetojus-ticeisself-interest,notthatregardforothersthatwasearliersaidtobeanessentialcomponentofourmoralapprovals.Why,then,doweinthethirdandfinalstageofthedevelopmentofjusticegivemoralapprovaltoactsofjustice?Or,topressevenfurther,howandwhyisthereathirdstage,astageinwhichwesupposewehaveamoralobligationtobejust?(T3.2.2.23).NeartheendofTreatise3.2.2HumesaysthatacompleteanswertothisquestiondependsondiscussionsfoundonlyinTreatise3.3,buthenonethelessprovidesanenlighteningsketchofhowhesup-posesthatwecometoattachmoralsignificancetowhatisinitiallyonlyself-regardingconcernthattheconventionsofjusticebemain-tained.Inthenormalandslowlychangingcourseofevents,thesocietiesmadepossiblebytheconventionsofjusticegrewsignifi-cantlylargerandmorecomplex.Asaresult,itbecamemoredifficultforindividualstoseehowtheirprivateinterestswerebeingservedCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n300davidfatenortonbyadherencetotheconventionsofjusticethathadbeendeveloped.Asaresult,someindividualsbegantodisregardtheseconventions.Theybegantoactunjustly,perhapswithoutevennoticingthattheyweredoingso.Otherindividuals,however,didnoticewhentheseconventionswereviolatedbecausetheywereharmedbytheactionstakingplace.Moreover,evenwhensuchunjustactionswereremotefromtheseobserversandconsequentlydidnotharmthem,thesedistantobserversnonethelessdisapprovedoftheseactions.Theydidso,accordingtoHume,becausetheyfoundsuchunjustbehav-iorprejudicialtohumansociety,andpernicioustoeveryonethatapproache[d]thepersonguiltyofit(T3.2.2.24).Infact,eventhoseindividualswhoactedunjustlyweremadetofeeluneasybytheirownunjustactions.Somehow,afurtherstageofdevelopmenthadbeenreached.Whatinthesecondstagehadbeenaself-regardingconcernthattheconventionsofjusticebemaintained,becameinadditionanother-regardingconcernthattheseconventionsbefollowed.48Humefindsthattwofeaturesofhumannaturemadethisimpor-tantmoraldevelopmentpossible.Thefirstfeatureisourtendencytoformgeneralrules,andtogivetotheserulesaninflexibilitythatcanwithstandeventhepressuresofself-interest.49Oncewehave48LaterinhisdiscussionofjusticeHumesays,Uponthewhole,then,wearetoconsiderthisdistinctionbetwixtjusticeandinjustice,ashavingtwodifferentfoundations,viz.thatofself-interest,whenmenobserve,thattisimpossibletoliveinsocietywithoutrestrainingthemselvesbycertainrules;andthatofmoral-ity,whenthisinterestisonceobservdtobecommontoallmankind,andmenreceiveapleasurefromtheviewofsuchactionsastendtothepeaceofsociety,andanuneasinessfromsuchasarecontrarytoit.Tisthevoluntaryconventionandartificeofmen,whichmakesthefirstinteresttakeplace;andthereforethoselawsofjusticearesofartobeconsiderdasartificial.Afterthatinterestisonceestablishdandacknowledgd,thesenseofmoralityintheobservanceoftheserulesfollowsnaturally,andofitself(T3.2.6.11).NotethatHumehereandelse-wheregrantsthattheinstructionsofpoliticiansandofparentscontributetothedevelopmentofjustice,buthealsosays,surelywithMandevilleinmind,thatthisaccountofthematterhasbeencarrydtoofarbycertainwritersonmorals,whoseemtohaveemploydtheirutmosteffortstoextirpateallsenseofvirtuefromamongmankind,andthatthisaccountiserroneous(T3.2.2.25,3.3.1.11).49Humesfirstaccountofgeneralrulessuggeststhattheyarewhatwouldnowbecalledgeneralizations,andindicatesthattheymaybewell-orill-founded.Buttherelevantgeneralruleinthepresentcaseisadefinitionthatcharacterizespropertyasastablepossession,derivdfromtherulesofjustice,ortheconventionsofmen,or,asHumehadearliersaid,anunderstandingofpropertyassucharelationCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise301establishedtheconventionsthataretogovernthepossessionandexchangeofproperty,oursentimentsmaybeinfluencedbytheseconventionsevenwhentheyortheiruseconflictswiththeself-interestthathasproducedthem.Conventionshavingthatkindofcontinuingforceexerciseatleastapartialcheckonself-interest.Thesecondfeatureissympathy.Humerecognizesthatanypartic-ularactofjusticemaybecontrarytobothprivateandpublicgood:asingleactofjustice,considerdinitself,mayoftenbecontrarytothepublicgood;andtisonlytheconcurrenceofmankind,inageneralschemeorsystemofaction,whichisadvantageous...ifweexamineallthequestions,thatcomebeforeanytribunalofjustice,weshallfind,that,consideringeachcaseapart,itwoudasoftenbeaninstanceofhumanitytodecidecontrarytothelawsofjusticeasconformabletothem.Judgestakefromapoormantogivetoarich;theybestowonthedissolutethelabouroftheindustrious;andputintothehandsoftheviciousthemeansofharmingboththemselvesandothers.Thewholescheme,however,oflawandjusticeisadvantageoustothesocietyandtoeveryindividual.(T3.3.1.12)Humetellsus,inotherwords,thatonlyanunremittingcom-mitmenttothesystemofjusticeisbeneficialtoallconcerned,andyetfewwouldseemtohavesufficientmotivetomaintainsuchacommitment.Itisaninherentprincipleofcommunication,sympa-thy,thatmakesthiscommitmentpossible.Sympathyenablesustotranscendournarrowlyselfishinterestsandtofeelapprobationinresponsetoactionsthatmaintainthesystemofjustice,anddisap-probationinresponsetothosethatfailtogivesuchsupport.Itdoessobyenablingobserverstofeelthepleasuresandpainsproducedinothersaffectedbyjustorunjustactsthathavenodirectbear-ingonthoseobservers.Humearguesatlengththatself-interestwastheoriginalmotivetotheestablishmentofjustice,buthethengoesontoinsistthatwefeelthemoralsentimentsinresponsetojustandunjustactionsonlybecauseofsympathy.If,forexample,IobservethatDoeispainedbytheunjustactionofRoe,thenthebetwixtapersonandanobjectaspermitshim,butforbidsanyother,thefreeuseandpossessionofit,withoutviolatingthelawsofjusticeandmoralequity(T1.3.13.713,3.2.3.7,2.1.10.1).Formoreonthesegeneralrules,seeThomasK.Hearn,GeneralRulesinHumesTreatise,JournaloftheHistoryofPhilosophy8(1970):40522;andGeneralRulesandtheMoralSentimentsinHumesTreatise,ReviewofMetaphysics30(1976):5772.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n302davidfatenortonoperationofsympathycausesmetofeelnotonlyDoespain,butalsothedisesteemthispainhascausedhertofeelforRoe.Moregenerally,asHumelatersumsupthematter,oncetheconventionsofjusticeareestablished,justactionsarenaturallyattendedwithastrongsentimentofmorals;whichcanproceedfromnothingbutoursympathywiththeinterestsofsociety(T3.2.2.24,3.3.1.12).50Ourdispositiontoformandfollowgeneralrulesandourabilitytosharesentimentswithourfellowhumanshaveenabledmoralitytoevolvetothepointthatatleastsomeofusfeelthedistinctivemoralsentimentsinresponsetojustorunjustactions.51D.TheNaturalVirtuesHumeannouncesatthebeginningofTreatise3.3.1,Oftheoriginofthenaturalvirtuesandvices,thatheisreadyfortheexami-nationofsuchvirtuesandvicesasareentirelynatural,andhavenodependanceontheartificeandcontrivanceofmen.Laterinthissectionhenotesthatgreatmenarepraisedfortwokindsofqualities,thosethatenablethemtoperformtheirpartinsoci-ety,namely,generosityandhumanity,andthosethatmakethemserviceabletothemselves,andenablethemtopromotetheir50Humealsosaidofsympathythatitproduces,inmanyinstances,oursentimentsofmorals....Novirtueismoreesteemdthanjustice,andnovicemoredetestedthaninjustice;norarethereanyqualities,whichgofarthertothefixingthecharacter,eitherasamiableorodious.Nowjusticeisamoralvirtue,merelybecauseithasthattendencytothegoodofmankind;and,indeed,isnothingbutanartificialinventiontothatpurpose.Thesamemaybesaidofallegiance,ofthelawsofnations,ofmodesty,andofgood-manners.Allthesearemerehumancontrivancesfortheinterestofsociety.Theinventorsofthemhadchieflyinviewtheirowninterest.Butwecarryourapprobationofthemintothemostdistantcountriesandages,andmuchbeyondourowninterest.Andsincethereisaverystrongsentimentofmorals,whichhasalwaysattendedthem,wemustallow,thatthereflectingonthetendencyofcharactersandmentalqualities,issufficienttogiveusthesentimentsofapprobationandblame.Nowasthemeanstoanendcanonlybeagreeable,wheretheendisagreeable;andasthegoodofsociety,whereourowninterestisnotconcernd,orthatofourfriends,pleasesonlybysympathy:Itfollows,thatsympathyisthesourceoftheesteem,whichwepaytoalltheartificialvirtues(T3.3.1.9;seealso3.3.6.12).Forfurtherdiscussionofsympathy,seeinthisvolumeHumesMoralPsychology,PartV,andHumesLaterMoralPhilosophy,PartIV.51ForfurtherdiscussionofHumesaccountofjustice,seeinthisvolumeTheStruc-tureofHumesPoliticalTheory,PartIII.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise303owninterest,namely,prudence,temperance,frugality,industry,assiduity,enterprize,[and]dexterity.In3.3.3,Ofgoodnessandbenevolence,hedescribesgenerosity,humanity,compassion,grat-itude,friendship,fidelity,zeal,disinterestedness,[and]liberalityasthequalities,whichformthecharacterofgoodandbenevolent,buttheinclusionoffidelity,elsewheredescribedasanartificialvirtue,requiresustoaskifallthequalitieslistedherearenatu-ralvirtues.Later,whilediscussingnaturalabilities,henotesthatindustry,perseverance,patience,activity,vigilance,application,[and]constancy,aswellastemperance,frugality,conomy,[and]resolution,areusefulvirtues,whileprodigality,luxury,irreso-lution[and]uncertainty,arevicious.Ifthesequalitiesarealltobeclassedamongthenaturalvirtues,theyconstituteawidespectrumofmotivatingqualities(T3.3.1.1,3.3.1.24,3.3.3.3,3.3.4.7).AlthoughtheTreatiseneverprovidesuswithacomprehensiveandunambiguouslistofthenaturalvirtues,wecanbesurethatgenerosityandhumanityaretwosuchvirtues.Knowingthis,know-ingthatatleastthesetwoinstinctivedesiresormotivationsareanintrinsicpartofhumannature,meansthatwehave,asHumeputsit,aproof,thatourapprobationhas,in[some]cases,anorigindifferentfromtheprospectofutilityandadvantage,eithertoour-selvesorothers.Wehave,inotherwords,proofthattheaccountsofhumannaturegivenbyMandevilleorotheregoistsaremistaken.Thedescriptionsofourselfishnessthatthesephilosophersgive,Humesaid,areaswideofnatureasanyaccountsofmonsters,whichwemeetwithinfablesandromances.Itmayberare,hegrants,toencounteranyonewholovesanysinglepersonbetterthanhimself,butitisjustasraretomeetwithone,inwhomallthekindaffections,takentogether,donotover-ballancealltheselfish(T3.3.3.4,3.2.2.5).E.DutyandObligationAlthoughHumefollowsthedevelopmentoftheartificialvirtuestothepointofexplaininghowtheseconventionsbecomemoralvirtuesandthatmanyhumansfeelamoralobligationtoactjustlyorkeeptheirpromises,hedoesnotintheTreatise(orelsewhere)provideuswithasustaineddiscussionofhowitiswecometofeelmoralobligation,ofhowwecometofeelthatweoughttoCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n304davidfatenortonbehavehumanelyorjustly.Moreover,inoneofthemostwidelydiscussedparagraphsintheTreatise,Humereportsthatineverysystemofmoralitythathehasencounteredtheauthorbeginsintheordinarywayofreasoningandinthiswaypresentsreaderswithpropositionsthatconnecttheirpartswithisorisnot(ThereisaGod,forexample).Soonenough,however,theseauthorsgoontopresentuswithpropositionsthatconnecttheirpartswithoughtoroughtnot(WeoughttoobeythecommandsofGod,forexample).Humedescribesthisshiftasasubtleorimperceptibleone,butconsidersittobeofthegreatestconsequence:Forasthisought,oroughtnot,expressessomenewrelationoraffirmation,tisnecessarythatitshoudbeobservdandexplaind;andatthesametimethatareasonshoudbegiven,forwhatseemsaltogetherinconceivable,howthisnewrelationcanbeadeductionfromothers,whichareentirelydifferentfromit(T3.1.1.27).ThisparagraphhasledsometosupposethatHumeisarguingthatallmoralimperatives(allpropositionsoftheform,Xoughttodo/oughttohavedoneY)areunfounded,andthatheisinconsistentwhenhelatersuggeststhathumansdoinfacthavebothnaturalandmoraldutiesorobligations,andthattheseobligationsderivefromrelevantfacts.Suchinterpretationsfailtonoticethat,althoughthisparagraphbeginswithareferencetoeverysystemofmorality,itisineffectanaddendumtoasectionoftheTreatisedevotedtoshow-ingthatmoraldistinctionsarenotderivedfromreason,andthatitconcludeswiththecommentthatmerelynoticingthatthisimpor-tantshiftoflanguageinfectsmanymoralsystemswillbeenoughtosubvertallthevulgarorcommonsystemsofmorality,andletussee,thatthedistinctionofviceandvirtueisnotfoundedmerelyontherelationsofobjects,norisperceivdbyreason(T3.1.1.27).52WecanbegintounderstandtheaccountoftheoriginsofmoralobligationordutyfoundintheTreatisebyobservingthatwhenHumesaysthatnoactioncanbelaudableorblameable,without52Fromthefifteenthintothenineteenthcentury,themeaningsoftheadjectivevulgarincludedIncommonorgeneraluse;common,customary,orordinary,asamatterofuseorpractice(OxfordEnglishDictionary,vulgar,a.2).Circa1740,whenHumewrotethissentence,themostcommonsystemofmoralswasmostlikelythatofClarkeandhismanyrationalistdisciples.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise305somemotivesorimpellingpassions,distinctfromthesenseofmorals,hegoesontosaythatthesedistinctandmotivatingpas-sionsnotonlyhaveagreatinfluenceonoursenseofmorals,butalsothatweblameorpraiseaccordingtothegeneralforceinhumannaturethesepassionshave.Healsosaysthatwealwaysconsiderthenaturalandusualforceofthepassions,whenwedeter-mineconcerningviceandvirtue,andthatoursenseofdutyalwaysfollowsthecommonandnaturalcourseofourpassions.Andhelatersummarizeshisviewofmoralitysaying:Allmoralitydependsuponoursentiments;andwhenanyaction,orqualityofthemind,pleasesusafteracertainmanner,wesayitisvirtuous;andwhentheneglect,ornon-performanceofit,displeasesusafteralikemanner,wesaythatwelieunderanobligationtoperformit(T3.2.1.18,3.2.5.4).ThesecommentssuggestthatHumetakestheoriginoftheideaofdutyorobligationtoresembletheoriginoftheideaofcausalornecessaryconnection.Thislatteridea,theideathatcertaineventsoftypeAnecessarilycausecertainothereventsoftypeBderives,heexplains,fromanimpressionofexpectation(adeterminationofthemind)feltwhen,aftereventsoftypeBhavebeenrepeatedlyexperiencedtofollowcloselytheexperienceofeventsoftypeA,weagainexperienceaparticulareventoftypeA.Whenthislatestexpe-rienceofanA-typeeventoccurs,weexpectaB-typeeventtofollow.If,forexample,wearefamiliarwithlightningandthethunderthatfollowsit,then,ifweseelightning,weexpectthundertofollow.53Theremarksquotedinthelastparagraphbutonesuggestthatanimpressionorfeelingofexpectationplaysananalogousroleintheformationofboththesenseofobligationandourideaofthisphe-nomenon.Theyalsosuggestthatanotherimpressionofexpectation53TheTreatiseaccountoftheoriginofourideaofthecausalrelationandnecessaryconnectionissetoutindetailinSections1.3.38,1415.Humelatersummarizeshisconclusionsbysaying,Ifobjectshadnotanuniformandregularconjunctionwitheachother,weshoudneverarriveatanyideaofcauseandeffect;andevenafterall,thenecessity,whichentersintothatidea,isnothingbutadeterminationofthemindtopassfromoneobjecttoitsusualattendant,andinfertheexistenceofonefromthatoftheother(T2.3.1.4).At1.3.15.8Humespeaksoftheexpec-tationproducedbyauniformexperience,andalsoofthedisappointmentfeltifthatexpectationisnotmet.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n306davidfatenortonplaysthesamekindofroleintheproductionoftheimpressionofblameandofatleastsomeformsofmoralapproval.Thatis,fromthecommoncourseofactionsmotivatedbythecommonandnat-uralcourseofourpassionstherearisesinustheexpectationthat,inspecifiablecircumstances,certainactionswillbeperformed.Iftheseactionsareperformed,weapproveorfeelapprobation.Iftheyarenotperformed,wedisapproveorfeeldisapprobation.Feelingsorsentimentsofdisapprovalofthisparticulartype(thosefeelingsthathavethecausesandfeaturescharacteristicofmoralsentiments),become,whencopied,theideaofblame.Theybecome,thatis,theideathatanindividualoughttohaveactedinaparticularway,butfailedtodoso.Thisideaofblame,understoodasapplyingtoallthosewhofailtoperforminaccordancewiththecommonandnat-uralcourseofthepassions,isitselftheideaofdutyorobligation.Inshort,theideaofobligationisderivedfromauniqueimpressionofexpectation,animpressionthatisuniqueinsofarasitisaresponsetospecificfeaturesofourexperience.Dutyorobligationcannotbededucedfromfactualpremises.Butanimpressionofobligationafeelingthatsomethingoughttohavebeendonecanbederivedfromourexperienceofthoseactionsthathavemoralrelevanceinsofarastheyserveassignsoftherelevantvirtuousorviciousmotives(seenote36).ThisinterpretivehypothesisissupportedbyHumesbriefcom-mentsabouttheobligationsassociatedwithanaturalvirtue,theaffectionofparentsforchildren.Hesays:Weblameafatherforneglectinghischild.Why?becauseitshowsawantofnaturalaffec-tion,whichisthedutyofeveryparent.Hethengoesontosaythat,werenotnaturalaffectionaduty,thecareofchildrencoudnotbeaduty,thussuggestingthat,becauseaffectionforoneschildrenisastandardfeatureofhumannature,thecareofchildrenisnotonlyexpectedofparents,butasaconsequencebecomestheirduty(T3.2.1.5).Helaterreturnstothisexampleandlendsfurthersupporttothesuggestionthathesupposesthatmoralobligationarisesfromaprocessofthesortjustoutlined.Hesays:Noactioncanberequirdofusasourduty,unlesstherebeimplantedinhumannaturesomeactuatingpassionormotive,capableofproducingtheaction.Thismotivecannotbethesenseofduty.AsenseofdutysupposesanCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise307antecedentobligation:Andwhereanactionisnotrequirdbyanynaturalpassion,itcannotberequirdbyanynaturalobligation;sinceitmaybeomittedwithoutprovinganydefectorimperfectioninthemindandtemper,andconsequentlywithoutanyvice.Then,afteracommentregardingourobligationtokeeppromises,anartificialvirtue,hegoesontospeakagainofthenaturalvirtues,andsaysthatThotherewasnoobligationtorelievethemiserable,ourhumanitywoudleadustoit;andwhenweomitthatduty,theimmoralityoftheomis-sionarisesfromitsbeingaproof,thatwewantthenaturalsentimentsofhumanity.Afatherknowsittobehisdutytotakecareofhischildren:Buthehasalsoanaturalinclinationtoit.Andifnohumancreaturehadthatinclination,noonecoudlieunderanysuchobligation.(T3.2.5.6)Dutyorobligation,Humesays,cannotbededucedfromfactualpremises,butheindicatesthatanaturalobligationmaybederivedfromanaturalpassionornaturalsentimentsfromanincli-nationtohumanityortocareforoneschildren.Theseimplantedorinnatenaturalvirtuesleadhumanstobehave,ordinarily,inpre-dictablewaysthatbenefitothers.Whenoneofourspeciesfailstobehaveasweexpect,wefeelblameormoraldisapproval,andfromthisimpressionwegoontoformtheideaofmoralobligation.54HereagainHumetracesacentralaspectofmoralitytohumannature.iv.thefoundationinhumannatureWehaveseenseveralwaysinwhichHumeintheTreatisemakesgoodonhiscommitmenttomakehumanNaturehisprinci-palStudy,&theSourcefromwhich[he]woudderiveeveryTruthin...Morality.Wehaveseenthat,havingdirectedhisattentiontohumannature,hehasfoundinthatnaturefeatures(principles,54Humeallowsforunavoidableexceptionstotheexpectedbehavior:Whereapersonispossessdofacharacter,thatinitsnaturaltendencyisbeneficialtosociety,weesteemhimvirtuous,andaredelightedwiththeviewofhischaracter,eventhoparticularaccidentspreventitsoperation,andincapacitatehimfrombeingserviceabletohisfriendsandcountry.Virtueinragsisstillvirtue;andthelove,whichitprocures,attendsamanintoadungeonordesart,wherethevirtuecannolongerbeexertedinaction,andislosttoalltheworld(T3.3.1.19;seealso2.2.3.34).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n308davidfatenortonheoftencallsthem)thatexplaintheoriginandbasicworkingsofmorality.Centralamongthefeaturesorprinciplesdiscoveredarethosethatwehavebrieflyreviewedinthisessay:1.Humansaremotivatedbythepassions,notbyreason.2.Humanshaveanaturalmoralsensitivity,sothattheyrespondtoactionsseentohavebeenmotivatedbyaregardforthewelfareofotherswithasentimentofapprobation,andtothoseactionsmotivatedbymaliceoroverweeningself-concern,theyrespondwithasentimentofdisapproba-tion.3.Humannatureincludesimportantnaturalvirtues:naturaldesiresordispositionstoactinhumane,generous,orcom-passionateways,butthescopeorstrengthofthesevirtuesislimited.4.Humannaturealsoencompassesineachindividualthestrongestofallhumanpassions,self-interest.5.Therudeself-interestnaturaltohumanscanwithtimeandreflectionbechangedfromunprincipledselfishnesstoanenlightenedself-interestthatenablesgroupsofustodevelopsuchvaluableconventionsorartificialvirtuesasjustice(anunderstandingofandrespectforproperty),fidelity(anunderstandingofandrespectforpromisesorcontracts),andallegiance(anunderstandingofandrespectforpoliticalsociety).6.Humannaturealsoincludesaverypowerfulprinciple,sympathy,whichisthechiefsourceofmoraldistinctionsandwhich(amongotherthings)servestotransformthefun-damentallyself-regardingartificialvirtuesofjustice,fidelity,andallegianceintoformsofbehavior(beingjust,faithful,andloyal)motivatedbyaregardorconcernforthewell-beingofothers,evenofotherswithwhomwehavenoacquaintanceorconnection(T3.3.6.1).7.Humannatureincludesafurtherdispositiontorespondtocustomaryhumanbehaviorinmuchthewaythatwerespondtothecustomarybehaviorofnaturalobjects:byformingexpectations,expectationsinmoralmattersthat,whentheyaredisappointed,giverisetotheimpressionsandideasofblameandobligation.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nTheFoundationsofMoralityinHumesTreatise309AlthoughHumesaccountofmoralityissignificantlydifferentfromthatofHutcheson,hecouldclearlyjoinHutchesoninsayingthatHumanNaturewasnotleftquiteindifferentintheAffairofVirtue.suggestionsforfurtherreadingForalternativepointsofviewontheissuesraisedinthisessay,see,inadditiontothethebooksandessayscitedtherein:AnthologiesChappell,V.C.,ed.Hume:ACollectionofCriticalEssays.NewYork:Dou-bleday,1966.Seeesp.thecontributionsonHumestheoryofobligationbyA.C.McIntyre,R.F.Atkinson,W.D.Hudson,andBernardWand.Cohon,Rachel,ed.Hume:MoralandPoliticalPhilosophy.Aldershot,Eng-land:DartmouthPublishing,2001.ThisvolumeincludesrelativelyrecentpapersonvirtuallyalltheissuestakenupinPartsIIIVofthisessay.MonographsandPapersArdal,P´all.´PassionandValueinHumesTreatise,2nded.Edinburgh:Edin-burghUniversityPress,1989.4179,10989.Baier,AnnetteC.AProgressofSentiments:ReflectionsonHumesTreatise.Cambridge,MA:HarvardUniversityPress,1991.127259.Darwall,Stephen.HumeandtheInventionofUtilitarianism.InHumeandHumesConnexions,editedbyM.A.StewartandJ.P.Wright.Edin-burgh:EdinburghUniversityPress,1994.5882.Ferreira,M.Jamie.HumeandImagination:SympathyandtheOther.InternationalPhilosophicalQuarterly34(1994):3957.Green,T.H.Introduction.InDavidHume:ATreatiseofHumanNature,ThePhilosophicalWorksofDavidHume,editedbyT.H.GreenandT.H.Grose,4vols.London,1875,1882.2:171.Herdt,JenniferA.ReligionandFactioninHumesMoralPhilosophy.Cam-bridge:CambridgeUniversityPress,1997.Korsgaard,ChristineM.TheSourcesofNormativity.Cambridge:CambridgeUniversityPress,1996.5166,869.Mackie,J.L.HumesMoralTheory.London:Routledge&KeganPaul,1980.Martin,MarieA.HutchesonandHumeonExplainingtheNatureofMoral-ity:WhyItIsMistakentoSupposeHumeEverRaisedtheIs-OughtQues-tion.HistoryofPhilosophyQuarterly8(1991):27789.McGilvary,E.B.AltruisminHumesTreatise.PhilosophicalReview12(1903):27298.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n310davidfatenortonMcIntyre,JaneL.HumesNewandExtraordinaryAccountofthePas-sions.InTheBlackwellGuidetoHumesTreatise,editedbyS.Traiger.Oxford:BlackwellPublishing,2006.199215.Mounce,H.O.HumesNaturalism.London:Routledge,1999.6298.Norton,DavidFate.EditorsIntroduction.InDavidHume:ATreatiseofHumanNature,editedbyD.F.NortonandM.J.Norton.Oxford:OxfordUniversityPress,2000,reprintedwithcorrections,2005.I4797.Stroud,Barry.GildingorStainingtheWorldwithSentimentsandPhantasms.HumeStudies19(1993):25372.Taylor,Jacqueline.JusticeandtheFoundationsofSocialMoralityinHumesTreatise.HumeStudies24(1998):530.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\njacquelinetaylor10HumesLaterMoralPhilosophyHumewritesinhisautobiographicalessay,MyOwnLife,thatheregardedhisEnquiryconcerningthePrinciplesofMoralsasofallmywritings,historical,philosophical,orliterary,incomparablythebest.1Thiswork,liketheEnquiryconcerningHumanUnder-standing,wasapartofhisTreatiseonHumanNaturethathehaddecidedtocastanew(MOL910).2Inanadvertisementcomposedlateinhislife,Humewrotethatthevolumeofessaysthatincludedthesetwoworksaffordedhimanopportunitytocorrectsomeneg-ligencesinhisformerreasoningandmoreintheexpressionoftheTreatise,andthathewantedthelaterwork,andnottheTreatise,toberegardedascontaininghisphilosophicalsentimentsandprinci-ples.3ThesecondEnquiryisoftenconsideredamoreeloquentworkthantheTreatise.ButHumescorrectionofsomenegligencesinreasoning,theadditionofnewarguments,andarestructuringofmainthemesgivethelaterworkaphilosophicalsignificanceinitsownright.1Inaletterof1755totheAbbeleBlanc,Humewrotethatthe´EnquiryconcerningthePrinciplesofMoralsismyfavoritePerformance(HLI:227).2Humesearliestreviewswereprimarilynegative.ForanextensivediscussionandthefulltextandanEnglishtranslationoftheonlyreviewofBook3oftheTreatise,seeTheBibliothequeraisonn`ee´ReviewofVolume3oftheTreatise:Authorship,Text,andTranslation,byDavidFateNortonandDarioPerinetti,HumeStudies32(April2006):352.SeealsoJamesFieser,ed.,EarlyResponsestoHume,Vols.3and4(Bristol,England:ThoemmesPress,2000).3SeetheAdvertisementprefacingtheOPTeditionofHume’sEnquiryconcerningHumanUnderstanding,ed.TomL.Beauchamp.HumementionstheadvertisementinaletterofOctober1775totheprinterWilliamStrahan;seeHL2:301.TheadvertisementwasfirstprintedinJanuary1776andwastobeincludedinallunsoldandfuturecopiesofthevolumeofessayscontainingthetwoEnquiries.311CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n312jacquelinetaylorThisessayexaminesHumeslatermoralphilosophy.IalsolookatsomedifferencesbetweentheTreatiseandhislaterworks.ButthemainaimoftheessayistodrawattentiontowhatmakesthelaterworkphilosophicallyinterestingandimportantincomparisonwithHumesearlierwork.Inparticular,IshowtheimportanceofthematureHumesattention,inthesecondEnquiry,ADialogue,andsomeofhisessays,totheroleoflanguageandreasoninmorality.Healsohasimportantthingstosayabouttheissueoftheuniversalityandrelativityofmorals.IbeginwithabriefoverviewofthesecondEnquirytodrawattentiontoHumesmethodandthemainthemesofthework.Ithendiscussinmoredetailtherelationbetweensym-pathy,reason,anddiscourseorlanguage,ontheonehand,andmoralsentiment,ontheother.Intheconcludingsections,IreconstructHumesaccountofhowweassessmoralcommunitieswhentheirvaluesdifferfromourown,drawingattentiononcemoretotheimportanceofmoraldiscourse.i.briefsurveyofthemethodandthemesofthesecondenquiryThesecondEnquiryopenswithadiscussionofmoralcontroversies.Humefirstidentifiestwosortsofphilosophernotworthengagingwith,thedogmatistandthedisingenuousdisputantwhodeniestherealityofmoraldistinctions.Bothsortsadhereblindlytotheirownviews,andshowcontemptforwhoeveropposesthem.Thus,asreasoningisnotthesource,whenceeitherdisputantderiveshistenets;itisinvaintoexpect,thatanylogic,whichspeaksnottotheaffections,willeverengagehimtoembracesounderprinciples.Abettercontroversytotakeupconcernsthefoundationofmorality:doesitlieinreasonorsentiment?Somethinkthatmoralityisamatteroftruth,arrivedatthroughaconsiderationofthefactsandarguments.Othersinsistthatmoralitymustengageourfeelings,forotherwisewewouldbeindifferenttoit.Humesayshewillnotbeginbyexaminingindepththeseeminglyconvincingargumentsthatmightbemadeoneithersideofthiscontroversy.Instead,heproposesfirsttocollecttogetherallthosementalqualitiesthatmakeupwhat,incommonlife,wecallpersonalmerit,bywhichhemeansthosecharactertraitswefindpraiseworthyorblameworthy(EPM1.12,10).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy313Thestartingpointisthuscommonlifeandtheeverydaytermi-nologyweusetotalkaboutcharacterandotherqualitiesofhumannature.Byattendingtothenamesweusefortheattributesofthemindwewillfindthattheverynatureoflanguageguidesusalmostinfalliblyinformingourmoraljudgments,andthattheleastacquaintancewiththeidiomsuffices,withoutanyreasoning,todirectusincollectingandarrangingtheestimableorblameablequalitiesofmen.Wecansimplyaskourselveswhichqualitieswewanttohaveattributedtous:forexample,wouldweratherhaveothersfindusbenevolentandhonest,orcruelanddisloyal?Withthevirtuesandvicesappropriatelycataloged,thetaskofreasonistoexaminethecircumstancescommontothevirtuesandthosecommontothevices.Throughthisexperimentalapproachwecanfinallyreachthefoundationofethics,andfindthoseuniversalprinciplesthatexplainwhywepraisesomementalqualitiesandblameothers.Determiningtheseprinciplesisaquestionoffact.TheartfulconstructionoftheEnquiryallowsHumetosetoutargu-mentsagainstarangeofopponents,includingdogmaticdivinesanddisingenuousselfishtheorists,inthecourseoffollowingtheexper-imentalmethod(EPM1.10).Thecarefulsurveyofmentalqualitiesleads,inSection9,tothedefinitionofvirtueasanymentalqualitythatisusefuloragreeabletothemoralagentherselfortoothers.Humeimplicitlyreliesonthisdefinitiontostructurethediscussionofthequalitiesthatmakeuppersonalmerit.Insteadofdistinguishing,ashedidintheTreatise,betweennaturalandartificialvirtues,hebeginswithbenevolenceandjustice,treatingthemtogetherasthetwofundamentalkindsofsocialvirtue(thatis,virtuesusefulforsociety).InSection2,Humenotesthatitisalmostasuperfluoustasktoshowthatweadmirebenevolence.Thetermsthatnamebenevolentqualities,humane,merciful,generous,andsoon,universallyexpressthehighestmerit,whichhumannatureiscapableofattaining(EPM2.1).Oneimportantreasonweapproveofbenevolenceasavirtueliesinitsusefulnessbothforthepossessorandforsociety.Itcanguidethedevelopmentandexerciseofsuchotherqualitiesascourageorambi-tion,anditpromotesthegeneralinterestaswellasthehappinessofthoseforwhomwecaremost.Humearguesthatinrecommendingsomethingasuseful,whetherafertilefield,acomfortablehouse,oravirtuouscharacter,weexpressourapprovalofit.Usefulness,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n314jacquelinetaylororutility,isasourceofmeritorvalue.Usefulcharactertraitshavemoralmerit,andearnourmoraladmiration.Justiceisasecondkindofsociallyusefulvirtue.HumesaiminSection3istoshowthatutilityisthesoleoriginofjustice,andhencethesolefoundationofitsmerit,andtheadmirationwefeelforit(EPM3.1).Hemakeshiscasebyshowingthatthehumancon-ditionisamediumbetweentheextremesofmaterialabundanceandabsolutescarcity,aswellasbetweenextremesoftemperament.Wearenotself-sufficient,andwenaturallyliveinfamilysettingswhereweformfriendshipsandenjoyconversation.Butourbenevolence,includingourlove,compassion,andgenerosity,isnaturallylimited,typicallytothosetowhomwearemostpartial,suchasfamilyandfriends.Hence,therulesofjusticearethemostusefulmeansofextendingcooperationformutualadvantage.Civillawsareformedinresponsetotheneedsandconvenienceofeachcommunity,andthecomplexityofparticularlawsshowstheinfluenceofreasoning,custom,andeducation.Section4extendstheexplicationofjustice,showingthatutilityisalsothesourceofthemeritofthosevirtues,suchasloyaltyandhonor,associatedwithgoodgovernmentandciti-zenship.Indeed,utilityexplainsthemeritofotherconvention-basedvirtues,includingthosespecifictowomen(chastityandmodesty),andgoodmanners.Section5formstheheartofthesecondEnquiry,andexaminesmorefullythereasonwhyutility,ormoreprecisely,whyqualitiesusefultotheagentortoothersandsaidtobevirtuesareasourceofmeritorvalue.Humesexplanationcentersontheimportanceofourcapacitytosympathizewiththeinterests,pleasures,andpainsofothers.Sympathy,notconsiderationsofself-interest,explainswhywefindusefulcharactertraitssovaluable.InthesecondpartofSec-tion5,Humedetailsthepervasiveinfluenceofsympathyinhumanlife.Withoutsympathywewouldbeindifferentnotonlytohuman-itybuttomorality.Therestofthissectiondiscussestheconnectionbetweensympathyandthemoralsentiments.Section6showshowthesameprincipleofsympathyproducesapprovalofself-regardingqualities,suchasindustryorprudence,usefulprimarilyfortheper-sonwhopossessesthem.Sections7and8discussanotherkindofqualitythatproducesattributionsofmerit.Weapproveofsometraitsofcharactersimplybecausetheyareagreeabletotheper-sonpossessingthemortoothers,eventhoughtheymaynothaveCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy315anytendencytothegoodoftheindividualorsociety.Cheerfulness,forexample,pleasesusimmediately.Sotraitsofcharacterthatareagreeablearevirtuesandthosethatwefinddisagreeablearevices.Hume,havinginSections2through8explainedthemeritofvar-iousmentalqualities,givesusinSection9thedefinitionsofvirtueandvice.Virtuesarethosementalqualitiesthatareusefuloragree-abletooneselfortoothers,whilevicesarethementalqualitiesthatareharmfulordisagreeabletooneselforothers.Humemakesitclearthatsomecharactertraitsmeetmorethanoneofthecriteriaforvirtue.Benevolence,forexample,meetsallfourcriteria:itisuse-fulandagreeable,forthepossessoraswellasforothers.Humealsoarguesthatsomevirtues,suchaspride,benevolence,andjustice,areessentialforsuccessfulcooperativeliving,whilethemoreimmedi-atelyagreeablequalities,suchaswitorgoodmanners,enhanceourqualityoflife.LinkinginthiswaytheaimofvirtuewithlivingwellprovidesHumewithgoodgroundsforconcludingthattheobligationtocultivatevirtuecoincideswithourtrueinterest.InthesecondpartofSection9,Humearguesthatfromthepointofviewofinter-est,wherethisincludesaconcernforourreputation,thecultivationofvirtuerecommendsitselftous.Asensibleknavemightthinkthatself-interestandmoralitycomeapartinthecaseofjustice,butthenhelacksasenseofintegrityandthepleasingconsciousnessthat,fortherestofus,reflecttheforceofmanysympathiesandensurethatwekeepacharacterwithourselvesandnotsimplyforappearances(EPM9.16,22,11).HumesstrategyforestablishingthefoundationofmoralityintheEnquiryissignificantlydifferentfromtheonehepursuedintheTreatise.Intheearlierwork,Humeinvokedtheprinciplesofassociationandsympathytogiveacausalstoryofhowthemoralsentiments,aparticularkindofimpression,originateinthemind.IntheEnquiry,Humesetsthehypothesisregardingassociationtoonesideandconcentratesonouractualexperienceofmoraleval-uationasasocialprocessgroundedinlanguageanddiscourse.TheEnquiryalsooffersnewargumentsthatmarksubstantiveadditionsorchangestothemoralphilosophyoftheTreatise.First,Humeexplicitlyarguesthatneitheroursocialmotivesnorthemoralsen-timentsreducetoself-interest.Second,moralevaluationexplicitlyrequiresgoodreasoning,reflectiononexperience,andconversationwithothers.Finally,HumeshowsanewawarenessofhistoricalandCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n316jacquelinetaylorculturalchange,anawarenessthatshapeshisdiscussionofthescopeofmeritandofpossiblevariationsinjustwhichqualitieswerecog-nizeasvirtuesorvices.TheimportanceoftheseadditionstoHumesphilosophyisunderscoredinsofaraseachistreatedinthemainbodyofthetextandinaseparateappendix.ii.self-interestandbenevolenceHumeintheTreatisedidnotdiscusstheimportanceofthevirtuesthatmakeupthebenevolentcharacteruntilthefinalpartofthebook.Priortothat,heemphasizedonlythelimitedscopeofourbenevolenceinordertomakethecasethatwehavenonaturalmotivetoactjustly,andsojusticemustbeestablishedthroughconventions.TheTreatisestrategywasnotparticularlysuccessful,though,sincehiscriticsseizedonhisclaimsaboutjusticeasanarti-ficialvirtueandtriedtolinkhimwiththeselfishtheoristThomasHobbes.4IntheEnquiry,Humetakesadifferenttack.Benevolenceisthefirstvirtuediscussed.Asnotedearlier,Humepresentsjusticeandbenevolenceasthefundamentalsocialvirtues,andhereservesdiscussionofthedifferencesbetweenbenevolenceandjusticeuntilAppendix3.InSection2andAppendix2,thefocusisonthegen-uinenessofourbenevolentdispositionsandaffections.Infact,theissueofhowselfishwearebynatureisimmaterialtomoralpracticesinceincommonlifewerecognizerealfriendshipandgratitude,anddistinguishbetweenthepersonwhoobviouslydoesusagoodturnoutofself-interestandonewhogenuinelydesiresourwell-being.Nevertheless,thequestionofuniversalselfishnessiscertainlyofconsequenceinthespeculativescienceofhumannature.Humehastwotargetshere:thosewhothinkthatallbenevolenceismerehypocrisy,afairdisguisebymeansofwhichwecanmanipulateotherstoserveourowninterests,aswellasthosewhothinkthatall4Inthe1741reviewinBibliothequeraisonn`eedesouvragesdessavansdel’Europe´(referredtoinnote2above),thereviewerwrotethatHumesaccountofjusticewasHobbesssystemclothedinanewfashion(seeHumeStudies32[2006]:36).InhisLetterfromaGentleman,HumecitesWilliamWishartschargethattheTreatiseofHumanNatureaimedatsappingtheFoundationsofMorality,bydenyingthenaturalandessentialDifferencebetwixtRightandWrong,GoodandEvil,JusticeandInjustice;makingtheDifferenceonlyartificial,andtoarisefromhumanConventionsandCompacts(L19).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy317generousactionsaredoneoutofself-love(EPMApp.2.15).BernarddeMandevillesworkwasregardedasaninstanceofthefirstkindofselfishtheory,andHumeidentifiesEpicurus,Hobbes,andLockeasamongthesecond.Humearguesthattheselfishschoolcanmaintainitstenetsonlybydistortingtheevidence.Languageprovideskeyevidenceinfavorofbenevolenceasaprincipleofhumannature.Whenweexaminecommonlifewefindnumeroustermsfortheamiablequalitiesthatshowevidenceofouraltruisticnature.Suchtermsassociable,good-natured,humane,merciful,grateful,friendly,generous,benef-icentnotonlydescribeother-regardingmotivesbutalsoexpressthehighestmeritwecanaccordtooneanother(EPM2.1).InbothSection2andAppendix2Humeadvancesfurtherargumentstosup-portthecontentionthatbenevolenceisreal.Forexample,animalsshowkindnesstooneanother,anditseemsspecioustoattributehypocrisyornarrowself-interesttothem.Ifweallowdisinterestedbenevolenceintheloweranimals,thenbywhatruleofanalogycanwerefuseitinmankind(EPMApp.2.8)?HumecitesJuvenal,whocreditsthemoreextensivebenevolenceofthehumanspeciestooursuperioritytotheotheranimals,andnotesthatthissuperior-itygivesuslargeropportunitiesofspreadingourkindlyinfluence(EPM2.4).HumecontinuestomaintainhisTreatisestancethatbenevolenceislimited.Ourbenevolentattitudesandactionsareforthemostpartinstancesofparticularbenevolence(EPMApp.2,n.60).Wedirectourloveorgratitude,ourkindnessorgenerosity,nottowardmankindingeneral,butrathertowardparticularothers,especiallythosewhomweesteemasvirtuous,orwhohavedoneusaservice,orwhostandmoreintimatelyconnectedtousthroughfamilyorfriendshipthanthegeneralrunofmankind.Thelongdependencyofourchildren,forexample,isevidenceofstrongparentalaffec-tions.AsHumenotes,Itiswiselyordainedbynature,thatprivateconnexionsshouldcommonlyprevailoveruniversalviewsandcon-siderations;otherwiseouraffectionsandactionswouldbedissipatedandlost,forwantofaproperlimitedobject(EPM5.42,n.25).Ouradmirationofbenevolenceinturnreflectsouracceptanceofthisnaturalpartiality.Humepresentsanotherargumentagainstreducingallmotivestoself-interest,versionsofwhichwerealsoadvancedbyJosephCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n318jacquelinetaylorButlerandFrancisHutcheson,andwhichhedescribesasbeingconformabletotheanalogyofnature(EPMApp.2.12).Naturalappetitessuchashungerorthirstalertustothebodysneedforcer-tainthings,viz.,foodorliquid,whicharetheendsoftheappetitesandprecedeanyenjoymentwemaygetfromthem.Ofcourse,weoftendoenjoyourmeals,andpleasurecanbecomeasecondaryobject,sothatwhenhungerstrikesuswenowdesiredeliciousfood.Thesamecanbetrueofourbenevolentdesires.Whenweactfrombenevolentmotivesweaimatthegoodofanotherperson.Itoftenturnsoutthatweenjoyhelpingothers,sothat,asinthecaseofbodilyappetites,thepleasurewereceivefromhelpingotherscanbecomeasecondaryobjectofdesire.iii.sympathyandthemoralsentimentAfterappealingtoutilityasasourceofmoralmerit,Humeiskeentoshowthatwedonotapproveofitonlywhenitbenefitsourselves.Weapproveofbenevolenceandarangeofotherusefulvirtuesfromamorepublicaffection,thatis,onemoresociallyorientedthanself-interest(EPM5.17).Humesetsouthisaccountofourmoralsentiments,thatis,ofwhathegenerallytermsourmoralappro-bationanddisapprobation,instages.Hefirstshowstheimpor-tanceofsympathyforourpreferenceforutility,andthenshowshowsympathyservesasthefoundationofthemoralsentiments.ThediscussioninthebeginningofSection5suggeststhatHumethinksmoralphilosopherstendtoassociateutilitywithmoralskep-ticism.AskepticlikeMandeville,forinstance,thinksthatmoral-ityisaninventionofpoliticians.Cleverpoliticianspersuadeustoattachpraisetothoseactionsthatpromoteandmaintainsocialorder.Humenotes,however,thatincommonlifemostofusappealtoutil-ityandthinkofmoralityassomethingreal.Heasksustoconsiderananalogybetweenthevirtuouscharacterandthosefeaturesoftheenvironmentorhumaninventionsthatmakelifemorecomfortableforus.Wepraisethefertilefield,thewell-builthouse,andthevirtu-ouscharacterforverysimilarreasons:eachpossessesqualitiesthatmakeitusefulforanindividualorsocietyingeneral.Likethefer-tilefield,thevirtuouspersonhasanaturalbeautythatwefindpleasingevenifitprovidesnodirectbenefitforus.Thatis,wetakeaCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy319disinterestedpleasureinthenaturalbeautyexhibitedintheactionsofthevirtuouscharacter.Itisourcapacityforsympathythatexplainsourdisinterestedadmirationandblame.Humeinsiststhatusefulnessisatendencytoanend,andwhattendstotheendcannotaffectoursentimentsunlesswealsocareabouttheend.Inthecaseofcharacter,theusefulqualityisindeedusefulforsomebody:fortheagentherselfinsomecases,orforparticularothers,orfortheinterestsofsociety.Theusefulvirtuestendtopromotewell-being,preventharm,oralleviatedistress.Sympathymakesuscareabouttheinterestsofothersandofsocietygenerally.Nowweshouldnotethatinadditiontothetermsympathy,Humealsousesthetermshumanityandgeneralbenevolencetorefertothecapacitythatexplainswhywetendtobepleasedbyothershappinessorpainedbytheirmisery.5Linkingtogethergen-eralbenevolenceandsympathyisadeparturefromtheassociation-istaccountoftheprincipleofsympathythatHumegaveintheTreatise(EPMApp.2.5,n.60).Somescholarshavethoughtthat,intheEnquiry,HumesreferencestogeneralbenevolencesignalthatheisbeingmoreconciliatorytowardHutchesonsmoralsensethe-ory,whichputsbenevolencesquarelyatthecenterofmorality.6Thereare,however,substantivedifferencesbetweenHumesviewintheEnquiryandHutchesonsposition.First,HumeindicatesintheEnquirythatsympathyfunctions,asitdidintheTreatise,asaprincipleofcommunication.Thatis,sympathyisthemeansbywhichwecommunicateourpassionsandsentimentstooneanother.5Humealsousesthetermsgeneralconcernandnaturalphilanthropy(seeEPM5.16,40).6SeeL.A.Selby-Bigge,whowritesthatintheEnquirysympathyisanothernameforsocialfeeling...ratherthanthenameofaprocess[asitwasintheTreatise]bywhichthesocialfeelinghasbeenconstructedoutofnon-socialorindividualfeeling,Introduction,inHume’sEnquiries(Oxford:ClarendonPress,1975;firstpublished1888),xxvi.Selby-BiggeconcludesthattheethicsoftheEnquiryisvir-tuallyindistinguishablefromHutchesonsmoralsensetheory.SeealsoNicholasCapaldi,Hume’sPlaceinMoralPhilosophy(NewYork:PeterLang,1989).JohnStewartarguesthatHumehasnotrevisedhisviewsofeithersympathyorbenevo-lenceintheEnquiry.Rather,hesayslessaboutsympathybecausehenolongeraimstoprovideanaccountoftheoriginofthemoralsentiments.SeeJohnB.Stewart,TheMoralandPoliticalPhilosophyofDavidHume(NewYork:ColumbiaUniversityPress,1963),especiallypp.32839.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n320jacquelinetaylorSecond,HumesnotionofgeneralbenevolenceisclosertowhatHutchesonmeansbyapublicsensethantoanyoftheformsofbenevolencethelatterdescribes.Iwillexpandonthesetwopointsinturn.Humegivesseveralexamplesofthecommunicationofourpas-sionsorsentimentsthroughsympathy(seeEPM5.26,29,38).Some-timesthishappensasakindofcontagion;anexampleisourtendencytofeelhappysimplybybeingaroundacheerfulperson(EPM7.2).IntheTreatise,sympathywasalsoreferredtoasaprincipleoftheimaginationthatworksanalogouslytocustom:sympathyconveyswhatHumecallsthelivelinessoftheimpressionofselftoanideawehaveofanothersemotionorsentiment,therebytransformingthatideaintothesameemotionthattheotherpersonfeels.ThisearlierexplanationofhowsympathyworksdependedonHumesassociationisttheorytowhichhemakesonlyabriefappealinthesecondEnquiry.Nevertheless,HumesuggestsintheEnquirythathestillregardssympathyasaprincipleoftheimagination,andthatwetypicallyfirstformanideaofanothersemotion(seeEPM6.3,n.26;EPM8.15).Oursympatheticresponsetoothersismorelivelyorintensewhenthesituationisphysicallyclosetous,orisbroughtcloserthrough,forexample,skillfulstorytelling.AsintheTreatise,sympathyisnotitselfanyparticularfeeling,butexplainswhywerespondsympatheticallytoabroadrangeofemotions:nopassion,whenwellrepresented,canbeentirelyindifferenttous;becausethereisnone,ofwhicheverymanhasnot,withinhim,atleasttheseedsandfirstprinciples.Wecanbeswayedbyothersopinionsorsentiments,bothgoodandbad.Humespecificallytakesnoteofthelesslaudableeffectsofthissocialsympathyinproducingpopularsedition,partyzeal,andfactions(EPM5.30,35).Asnotedabove,thedetailedaccountofsympathyintheTreatisereliedontheprinci-plesofassociation.First,Iinferthatsomeoneisexperiencingsomeemotionorotherbyherbehaviororfromhersituation,andbecausesheresemblesme,someofthelivelinessoftheconceptionIhaveofmyselfistransferredtotheideaofthepersonsemotionsothatIcometofeelthatveryemotion.Second,inmorallyevaluatingapersonscharacter,wesympathizewiththeeffectsofitontheagentorherassociates,thatis,withtheirresponses,whichwillincludearangeoffeelingsandattitudessuchaslove,gratitude,hatred,orcontempt,anditisfrommysympathywiththeirpleasuresorpainsCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy321thatIinturnexperiencemoralapprobationorblametowardthepersonscharacter.SinceHumegivesadifferentaccountofmoralevaluationandthestandardofvirtueinthesecondEnquiry(seethenextsection),wehavereasontothinkhesawasproblematictheassociationistaccountofsympathyandmoralevaluation.IntheEnquiry,henolongerappealstoassociationtoexplainsympathy,andatoneplacewritesItisneedlesstopushourresearchessofarastoask,whywehavehumanityorafellow-feelingwithothers.Itissufficient,thatthisisexperiencedtobeaprincipleinhumannature.Wemuststopsomewhereinourexaminationofcauses;andthereare,ineveryscience,somegeneralprinciples,beyondwhichwecannothopetofindanyprinciplemoregeneral....Itisnotprob-able,thattheseprinciplescanberesolvedintoprinciplesmoresimpleanduniversal,whateverattemptsmayhavebeenmadetothatpurpose.(EPM5.17,n.19)Itisabrutefactofournaturethatwetakeaninterestinothersandareresponsivetothem,andthatwecanbeinfluencedbytheiropinionsandfeelings.Humearguesthatitisobviousthatsympathyexertsasstrongaforceinourlivesasdoesself-interest.Sympathyisnotmerelyarealorimaginedself-interest,butmakesitimpossibleforustobecompletelyindifferenttotheconcernsofothers.Weneedonlytothinkofourreactionstodramaorpoetrywhereweenterintothepassionsoffictionalcharacters.Thenewsfromdistantstates,especiallywhenitcentersontragedyorispresentedtousvividly,engagesourcompassionforcompletestrangers.Thesameprincipleexplainsthegenuinefeelingswehaveforhistoricalfigures.Mostofusnaturallyseeourselvesassociablecreatures,takingpleasurefromfriendshipsandcompany,andlosingoutonenjoymentifwearecondemnedtosolitude.Perhapsmosttellingarethoseoccasionswhenself-interestandmoralitydiverge,andyetwestillfeelapprovalfor,orrecognizeaspraiseworthy,thevirtuesofthosewhoopposeus(EPM5.17).Itthusseemsaweakandimplausiblesubterfugetoclaimthatourapprobationofthedistantvirtuouscharacterdependsonourimaginingourselvesasthebeneficiariesofhervirtue.Itisfarmoreplausibleandconsistentwithourexperiencetoconcedethatwejustdoadmireusefuloragreeabletraitsofcharacter,andblameharmfulanddisagreeabletraits.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n322jacquelinetaylorSowhatarewetomakeoftheEnquiry’slinkingofsympathywiththesentimentsofhumanityandbenevolence?IthinkwecangetcleareronthisbycomparingHumesviewwiththatofHutcheson,sincesomescholarsclaimthatHumemovesclosertothelattersview.Hutchesonrecognizesfourdistinctinternalsenses:thatofimagination(ortheaestheticsense);thepublicsense,whichismostlikeHumesprincipleofsympathy;themoralsense;andthesenseofhonor.Thepublicsenseexplainsourtendencytobepleasedbythehappinessofothersandpainedbytheirmisery.Thereisnothinginherentlymoralaboutthistendency,andHutchesonnotesthatitoftenoperatesinpeoplewithouttheirgivinganythoughttovirtueorvice.Themoralsenseisanindependentinternalsensethatrespondstokindorunkindaffection.Hutchesonalsodistinguishesbetweenparticularbenevolenceandwhathecallsgeneralcalmbenevolence.Weactfromparticularbenevolenceinresponsetoasituationcallingforcompassion,friendship,gratitude,oroneoftheotherbenevolentmotives.Generalbenevolence,incontrast,isageneralcalmdesireforthegoodofparticularpersonsorsocieties.Hutchesonrecognizesyetathirdkindofbenevolence,universalcalmbenevolence,whichisdirectedtowardthespeciesingeneral.Particularbenevolence(includingcalmparticularbenevolence)neithernecessarilyarisesfromnorpresupposesuniversalbenevolence;infact,weoftenfindthatpeoplehaveparticularbenevolentaffectionsbut(throughwantofreflection)possessneitheruniversalnorgeneralbenevolence.Allthreeformsofbenevolence,particular,general,anduniversal,countasvirtuesforHutcheson,withuniversalbenevolencebeingthemostesteemed.7IntheEnquiry,Humedistinguishesbetweentwokindsofbenevo-lence,generalandparticular.Weactfromparticularbenevolence,aswesawearlier,whenwearekind,grateful,compassionate,charita-ble,andthelike,towardparticularotherpersons.Particularbenev-olencemotivatesustoactonbehalfofanother,ortakeanattitudeofconcerntowardsomeone,andisalwaysvirtuous.Incontrast,whenHumetalksintheEnquiryaboutgeneralbenevolenceor7SeeFrancisHutcheson,AnEssayontheNatureandConductofthePassionsandAffections,withIllustrationsontheMoralSense,ed.AaronGarrett(Indianapolis:LibertyFund,2002),Sect.II.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy323sympathy,heclearlymeanssomethinglikeHutchesonspublicsense.Thatis,sympathymakesustendtofeelgladaboutothershappinessorpainedbytheirmisery,anditisnotanyactivedesire,evenacalmorgeneralone,topromoteotherswell-beingorexpressanattitudesuchasgratitudethatwouldstrengthenourrelationshiptothem.Generalbenevolencecommonlyoperatesinthosecaseswherewehavenofriendshiporconnexionoresteemfortheper-son,butfeelonlyageneralsympathywithhimoracompassionforhispains,andacongratulationwithhispleasures(EPMApp.2,n.60).Generalbenevolenceorsympathyisoursensitivityandresponsivenesstotheemotionsorinterestsofothers,ratherthanamotivetoaction.AsHumeputsit,oursympatheticresponsivenessneedbenothingmorethanacoolpreferenceofwhatisusefulandserviceabletomankind,abovewhatisperniciousanddangerous(EPM9.4).Theremaybenothingparticularlymoralaboutsympa-thy,aswhenemotionsarecommunicatedasifbycontagion.Yet,asweshallseebelow,wecancultivateamoredelicatesympathyorsenseofhumanity,whichisitselfavirtue.Moreover,andincontrasttoHutchesonsview,Humethinksthatwithoutsympathyorgeneralbenevolencetorenderussensitivetothehappinessormiseryofoth-ers,wewouldneitherperceivemoraldistinctionssuchasvirtueandvice,norrespondtothemwithmoralapprobationordisapprobation(EPM5.39).Nevertheless,onewishesthatHumehadbeenclearerabouttheexactnatureoftherelationshipbetweensympathyandhumanityontheonehand,andthemoralsentimentsontheother.Inatleastoneplacehesuggeststhatoursympatheticconcernforthehappinessandmiseryofsociety,astheseareaffectedbyvirtuousorviciouscharactertraits,isthesameasthemoralsentiment:oursympa-theticpreferencefortheusefulabovetheharmfulisadistinctionthesameinallitsparts,withthemoraldistinction,whosefoun-dationhasbeensooften,andsomuchinvain,enquiredafter.Thesuggestionisthatsinceutilityisasourceofmerit,thenoursympa-theticpreferenceforausefulcharactertraitisthesameasourmoralapprobationofit.Humeconcludesfromthislineofreasoningthatthesesentimentsareoriginallythesame(EPM6.5).Yetthiscon-clusionintroducessomeambiguity.Thetextsuggeststwopossiblereadings,althoughthesearenotincompatiblewithoneanother.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n324jacquelinetaylorOntheonehand,Humecouldmeanthattheoriginofournatu-ralpreferencefortheusefulcharactertraitandourmoralapproba-tionofthetraitisthesame.Bothhavetheirorigininourcapacityforsympathy.ThisinterpretationisconsistentwithHumesaimofshowingthatthemoralsentimentsarenotthedeliverancesofaninnatemoralsense(astheyareforHutcheson).Rather,themoralsentimentsarereflection-informedresponsesthatcorrecttheinflu-enceofinterestandimaginativepropensitiessuchasfavoringwhatiscloseoverwhatisremote.Ontheotherhand,Humemightmeanthatoriginally,insomeearlierstageofthemoraldevelopmentofhumansociety,anaturalsympatheticpreferenceforwhatoneper-ceivedasusefulorharmfulconstitutedmoralapprobationorblame.HumesuggestssomethinglikethisinSection9whenhewritesarude,untaughtsavageregulateschieflyhisloveandhatredbytheideasofprivateutilityandinjury,andhasbutfaintconceptionsofageneralruleorsystemofbehaviour....Butwe,accustomedtosoci-ety,andtomoreenlargedreflections,consider,thatthisman[ourenemy]isservinghisowncountryandcommunity....Andbythesesuppositionsandviews,wecorrect,insomemeasure,ourruderandnarrowerpassions(EPM9,n.57).Thatis,ashumansocietydevelopsandexpands,welearntosympathizewithmoregeneralpreferences,ratherthanwithprivateorpartialones.Noticethatoneitherreading,sympathyisthesourceofthemoralsentiments,butthosesentimentsmustbecorrectedorcultivated.Andineithercase,wemaysafelysaythatsympathyisthesourceorfoundationofthemoralsentiments.AtonepointHumeclaimsthatsympathyalwayshassomeauthorityoverourmoralsentiments.Withoutsympathy,wewouldbeindifferenttomoraldistinctionssuchasvirtueandvice.Forexample,someoneofcoldinsensi-bility...unaffectedwiththeimagesofhumanhappinessormis-ery...mustbeequallyindifferenttotheimagesofviceandvirtue.Similarly,anabsolutelymaliciouscreaturemustbeworsethanindifferenttotheimagesofviceandvirtue,andhissentimentsmustbedirectlyoppositetothose,whichprevailinthehumanspecies.Hence,oursympatheticconcernforthehappinessandmis-eryofsociety,astheseareaffectedbyusefulorharmfulcharactertraits,isaprinciple,whichaccounts,ingreatpart,fortheoriginofmorality(EPM5.3940,17).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy325iv.sympathy,sentiment,andjudgmentInboththeTreatiseandtheEnquiry,Humedrawsattentiontothenaturalpartialityofsympathyandtothemeansbywhichwecorrectit.Justasparticularbenevolencemakesuscaremoreaboutparticularotherpersons,sympathytendsnaturallytofollowourparticularaffectionsandsoitselfexhibitspartiality.Wenaturallysympathizeeasilywiththepeopletowhomwehavespecialtiessuchasthoseoffamily,friendship,ornationality.Oursympathytendstobelessengagedwhenweareconsideringcasesdistantfromusintimeorplace.Soaquestionarisesabouthowthemoralsentimentscanbegroundedinsympathyifsympathyvaries.Incaseswherepartialityornearnessinfluencesympathy,wejudgethattwocharactersarethesame,evenifwefeelmorestronglyabouttheoneclosesttous.Thatjudgmentcorrectstheinequalitiesofourinternalemotions(EPM5.41).Wemakeasimilarcorrectionwhenoursentimentsarearousedbytheactualconsequencesofsomeonesactioninsteadofthecharactertraitthatmotivatedit.Forexample,supposesomeonewhoiswell-offperformsgenerousactionssimplytohavepeoplethinkwellofher,ratherthanoutofaconcernforotherswell-being.Humenotesthatwemustseparatethecharacterfromthefortune,byaneasyandnecessaryeffortofthought,andassessthecharacterratherthantheresults(EPM5.41,n.24).Thepersoninourexampleisunlikelytoactgenerouslywhensucharesponseisneeded,ortobeappropriatelysensitivetowhatothersmostneed.Erratic,ill-timed,orinappropriateinstancesofgivingarenottheactionsofagenerouscharacter(althoughtheymaysuggestthepersonistryingtocultivategenerosity).InbothtextsHumesexplanationfocusesonhowweestablishastandardofvirtue.Thestandardofvirtueisthestandardtowhichoursentimentsshouldconformiftheyaretobeacceptedasappropriatemoralresponsesofpraiseorblame.Buttheaccountofhowweestablishthestandardofvirtuediffersinthetwotexts.IntheTreatiseweextendoursympathytotakeupthepointofviewofthosewhoareaffectedbyanagentscharacter.Humeassertsthattheresponsesofthoseintheagentsspherearemoreconstantthanourownvariableresponses,andsotheycomprisethestandardofvirtuebywhichwecalibrateourapprobationanddisapprobationCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n326jacquelinetaylor(T3.3.1.18,30).Oneproblemwiththisstrategyisthoseintheagentsspheremightthemselvesrespondtohercharacterwithanaturalpartialityoraperniciousbias,andHumeoffersnoexplanationofhowwemightdistinguishbetweenappropriateandinappropriateresponses.8IntheEnquiry,Humedoesnotappealtosympathywithsome-onescircleofassociatesasthemeansforestablishingthestandardofvirtue.WhiletheTreatisementionstheimportanceofconversa-tion,theEnquiryplacestheemphasismoresquarelyonlanguageandsharedconversationaboutwhatisusefulorharmful.Itisthroughconversationwithoneanotherthatweformageneralstandardofviceandvirtue,foundedchieflyongeneralusefulness(EPM5.42,n.25).Humeheredistinguishesbetweenthelanguageofself-loveandthelanguageofmorality.Forexample,whenamandenominatesanotherhisenemy,hisrival,heusesthelanguageofself-love,andexpresseshowthingslooktohimfromtheperspectiveofhispri-vateinterest.Inconversingwithothers,webecomefamiliarizedtothemoregeneralpreferencesandinterestsofthecommunity.Weinventapeculiarsetofterms,inordertoexpressthoseuniversalsentimentsofcensureorapprobation,whicharisefromhumanity.Withmoralterminologyinplace,virtueandvicebecomethenknown:Moralsarerecognized:Certaingeneralideasareframedofhumanconductandbehaviour(EPM9.6,8).Welearntoaffixpraiseorblametoactionsandcharactersinconformitytothesesharedmoralsentiments.Andweusethelanguageofvirtueandvice,ofapprobationandblame,withtheexpectationthatotherswillagreewithourassessments.Sharedreflectionongeneralpreferencesleadstoasharedmoraldiscourse,groundedinourcommonhumanity,andrendersusintelligibletoeachother(EPM5.42).Wecannowconsidertherelationbetweenmoralsentiment,ortaste,andreason.Humestressesthepositivecontributionofeachfaculty,andhesuggeststhatthepersoninwhomthetwofacul-tiesareworkingtogetherproperlywillpossessthevirtuesofgoodevaluation.Tobesure,itisthisinternalfeeling,whichnaturehasmadeuniversalinthewholespecies,ratherthantruth-discovering8IdevelopmorefullytheargumentthatHumesetsoutadifferentandbettersetofcriteriaforthestandardofvirtue,thatis,forgoodmoraljudgmentorevaluation,inHumeontheStandardofVirtue,JournalofEthics6(2002):4362.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy327reason,thatrendersmoralityanactiveprinciple(EPM1.9).Butaswehavejustseen,ourmoralsentimentsdonotreflectmerelyidiosyncraticorindividualpreferences.Ourconversationsaboutgen-eralpreferenceswillinvokereflectiononpastexperiencetoestab-lishwhichstabletraitsofcharactertendovertimetopromotethehappinessoftheagentorothers.Similarly,theduemediumofanycharactertraitisdeterminedbyconsiderationsofutilityandbyreflectingontheconsequencesofhavingtoomuchortoolit-tleofthequality(EPM6.2).Theboundsofdutywithrespecttotheactionsandpracticesthatexhibitparticulartraitsarealsodeter-minedwithreferencetowhatismostusefulfortheinterestsofsociety.Forexample,weneedtoascertainwhichinstancesofgivingreallydosomegood,andwhichextinguishtheindependenceoftherecipient,andthusaredonefromweaknessratherthanfromvirtue.Theboundariesofmoralgoodandevilarethusadjustableaswelearnmoreabouthowtopromotetheinterestsofsociety,orassocialcircumstancesshift(EPM2.1718).InSection1Humespecificallyappealstotheelementsofgoodreasoning:inordertopavethewayforsuchasentiment,andgiveaproperdiscernmentofitsobject,itisoftennecessary,wefind,thatmuchreasoningshouldprecede,thatnicedistinctionsbemade,justconclusionsdrawn,distantcomparisonsformed,complicatedrela-tionsexamined,andgeneralfactsfixedandascertained.Whilesomeformsoftasteremainunsusceptibletotheevidenceofreason,ourmoraltastedemandstheassistanceofourintellectualfacultiesifitistohaveaproperinfluenceonthemind(EPM1.9).HumerepeatsthepointinthefirstAppendix,insistingthatanaccuratereasonorjudgmentmustinformoursentiment-basedresponses(EPM,App.1.2).Whileitisclearthatsomepeoplewillbebettermoralevaluatorsthanothers,Humesuggeststhatthecultivationofgoodreasoningisoftenacollectiveendeavorthattakesplacethroughactivedebateanddiscourse.Sometimesweneedgoodreasoningtomakesenseofcomplexmoralsituations:togatherandsortthroughtherele-vantfacts,tomaketherightsortofdistinctions,tocomparethecurrentcasetopastsituations,andsoon.Inothercasesitisdiffi-culttofigureoutjustwhichmotives,actions,orsocialpoliciesandpracticesreallywilltendtothebestinterestsofsociety.Itisespe-ciallydifficulttomakeassessmentsconcerningutilityinthecaseofCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n328jacquelinetaylorjustice.Humeappealstotheimportanceofthedebatesofcivilians;thereflectionsofpoliticians;theprecedentsofhistoryandpublicrecordstoguideourassessmentsofthetendenciesofparticularlawsorpolicies(EPMApp.1.2).9Yetsincethereisnoantecedentfactofthematteraboutsuchthingsaspropertyrightsorcontracts,lawmakersandjudgesmustoftenexercisetasteandimaginationwherenoprecedentorpositivelawexiststoguidethem(EPMApp.3.10).Thuswhileappropriatemoralsentimentsmustbeinformedbyreasonandreflection,soalsoaretasteandimaginationsometimesneededtosupplementreason.Humethusmakesthecaseforthevirtuesofgoodevaluation:awarmconcernfortheinterestofourspecies;adelicatefeelingofallmoraldistinctions;anaccuratereasonorjudgment;andtheenlargedreflectionsacquiredthroughmoralconversation.Ingeneral,debateandconversationincreaseourawarenessofandsen-sitivitytothecomplexitiesofmoralsituations:themorewehabit-uateourselvestoanaccuratescrutinyofmorals,themoredelicatefeelingdoweacquireofthemostminutedistinctionsbetweenviceandvirtue(EPM5.39,App.1.2,5.14).Withasufficientnumberofgoodjudges,oursenseofmoralitydisplaystheforceofmanysympathies.Bysurveyingourselvesandkeepinginreviewourowncharacterandconduct,aconcernforourownreputationaswellasforthewelfareofothersbecomesthesurestguardianofeveryvirtue(EPM9.1011).v.thescopeofmeritThesecondpartofSection6andthefourthAppendixoftheEnquirycontainargumentsthatHumehadpresentedintwoofthefinalsec-tionsoftheTreatise.Theseargumentsaremeanttoestablish,first,thatsympathyisthesourceofouradmirationofwhateveritisaboutapersonthatcontributestoherowninterestsorthoseofothers.Soouradmirationofnaturaltalents,wealth,bodilyendowments,andbeauty(andourdisdainfortheoppositequalities)hasthesame9Civilianswerethosewhocommentedonthelawsofsocieties.TwocivilianswithwhoseworkHumewasfamiliarareHugoGrotius,authorofOntheLawofWarandPeace(1625),andSamuelPufendorf,authorofOntheLawofNatureandofNations(1672).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy329sourceinsympathyasdoesourmoraladmirationofvirtue.Second,Humearguesthatallofthementalqualities,includingtalentsandself-regardingqualities,andnotmerelythesocialqualitiesusuallystyledvirtues,areonthesamefootinginsofarastheycontributetoassessmentsofanindividualspersonalmerit.IntheEnquiry,Humepointedlydirectstheseargumentsagainstdogmatictheolog-icalmoralists.ButweshouldalsonoticethatHumesargumentdif-fersmarkedlyfromthatofmoralsensetheoristssuchasHutchesonandShaftesbury.Sympathymakesitpossibletofindpleasing,andhencetoascribemeritto,abroadrangeofqualitiesinadditiontobenevolence.Ourmoralsentiments,Humesuggests,donotderivefromaninnatemoralsense,butareformsofthenaturalsympa-theticresponsesthatwecultivateandcorrectinconversationwithoneanother.ThroughouttheEnquiry,Humehasfocusedonpersonalmeritratherthanmorenarrowlyonvirtue.Hehasconsideredallthequalitiesofmindinwhichsomeonemighttakeprideonreview-inghercharacterandconduct.Inadditiontothesocialvirtuesorqualitiesoftheheart,suchasbenevolence,Humeincludesasvirtuesself-regardingqualities,suchasindustriousness,andquali-tiesoftheheadortalents,suchaswitorperseverance.Wealsopraiseinvoluntaryqualities,thosequalitiesthatpeoplemaynotbeabletoachievethroughtheirownefforts.Someinvoluntaryqual-itieswemighttypicallythinkofasgiftsoffortune,including,forexample,witoreloquence.ButHumealsothinksthatsomequali-tiesweregardastypicalvirtuesmaybeinvoluntary.Forexample,wemightdisapproveofthefearfulorimpatientpersondespitehiseffortstocultivatecourageorpatience.Somevirtues,suchasmag-nanimity,requirethatoneoccupyacertainsocialstation,somethingthatoftenliesbeyondthecontrolofmostpeople.Thequalitiesofcourage,patience,andmagnanimityareusefuloragreeable,soweadmirethemanddeploretheiropposites.Humeobservesthattheancientmoralistspaidlittleregardtothedistinctionbetweenvolun-taryandinvoluntary,andincludedallthelaudablequalitiesofmindamongthevirtues.Ouruseoflanguageinmoderntimesshowsthatwedothesame(EPM,App.4.2,1120).Remarkingonthebroadrangeofqualitiesthatcancontributetoapersonsmerit,HumewritesthathehasrepresentedvirtueinallhergenuineandmostengagingcharmssothatthedismaldressCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n330jacquelinetaylorfallsoff,withwhichmanydivines,andsomephilosophershavecov-eredher.First,herecognizesself-regardingqualitiesasimportantcomponentsofmeritanddemerit.Incontrasttothehabitsofself-denial(themonkishvirtues)advocatedbythesedivines,Humehaschampionedtheself-regardingqualitiesthathelpindividualsadvanceintheworldandlivewell(EPM9.15,3).Indeed,hehasallalongemphasizedthatcultivatingboththeself-regardingandthesocialvirtuesenhancesapersonsreputation.Benevolenceismoreessentiallyrequisite,forexample,inthosewithordinarytalentsandwholackwealthorinfluence(EPM2.3).Andaswesawear-lier,HumearguesinSection9thatlookingatthefourdifferentcategoriesofvirtuefromtheperspectiveofourlong-terminterestshowsthattheagreeablesentiment,apleasingconsciousness,agoodreputationthatcomewiththecultivationofvirtueestablishourinterestedobligationtoit(EPM9.21,14).Inaddition,Humenotesthatourlanguagesmarknopreciseboundarybetweentalentsandvirtues.Thisisbecauseourinter-nalsentimentsofpraiseorblame,prideorshame,makelittleornodistinctionbetweenthedifferentkindsofqualitiesthatcontributetoordetractfrompersonalmerit.AttentiontolanguagealsohelpsHumetomakehiscaseagainstatheologicalmorality.Divinesintheguiseofphilosopherspaylittleregardtothephenomenaofnature,ortotheunbiassedsentimentsofthemind,hencereasoning,andevenlanguage,havebeenwarpedfromtheirnaturalcourse.Thetheologicalmoralistswanttodesignateasvirtuesonlythosequali-tiesthatpeoplecanacquirevoluntarily,buttheymaketheerroroftreatingallmorals,asonalikefootingwithcivillaws,guardedbythesanctionsofrewardandpunishment.Humearguesthatspec-ulativephilosophers(incontrasttomoralistswhoseetheirtaskasthatofexhortingustovirtue)shouldseetheancientphilosophersashavinganadvantageoveratleastthosemodernswhowanttonarrowthescopeofmoralityandtherebyexcludethesentimentsofpraiseandblamethatweexperienceindailylife(EPM,App.4.21).ButdoesHumesbroadeningofthescopeofmeritturnouttobeaproblemforhismoraltheory?Howcantheaccountofasympathy-basedmoralsentimentprovidethemeansofclearlydistinguishingourestimationofmentalendowmentsfromourestimationofphysicalendowmentsandexternaladvantagessuchaswealth?Sym-pathyisthecommonsourceofourvariousformsofdisapproval,andCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy331Humewritesthatablemish,afault,avice,acrime;theseexpres-sionsseemtodenotedifferentdegreesofcensureanddisapprobation;whichare,however,allofthem,atthebottom,prettynearlyofthesamekindorspecies(EPM,App.4.22).Sympathyalsogroundsouradmirationofwhateverwefindpleasingaboutaperson,whetheramentaloraphysicalquality,andwhetherthequalityisvoluntarilyacquiredoramatteroffortune.Weesteemthephysicallybeautiful,wealthy,andpowerful,aswellasthevirtuous.YetHumeclaimsthatwecanseparatesuchexpressionsofesteemwhendirectedtowardsomeonebecauseofhisorherwealthorsocialstanding,fromthesen-timentsofesteemwedirecttowardsomeonescharacter.Forexam-ple,hewritesthatsomeonewhovaluesvirtueoverwealthmay,indeed,externallypayasuperiordeferencetothegreatlordabovethevassal;becauserichesarethemostconvenient,beingthemostfixedanddeterminate,sourceofdistinction:Buthisinternalsenti-mentsaremoreregulatedbythepersonalcharactersofmen,thanbytheaccidentalandcapriciousfavoursoffortune(EPM6.34).Butareourestimationsofcharactersoeasilyseparatedfromouradmirationoftheothervaluablequalitiesoradvantagespeoplepossess?Humesinclusionasmeritorioustheimmediatelyagreeablequalitiesofgen-teelness,decency,cleanliness,andgoodmannersqualitiesmostlikelytobefoundinthoseamongthehighersocialrankssuggeststhatsomeonessocialstandingdoesinformourconceptionofthepersonscharacteraswellasourethicalattitudestowardhim.IntheTreatiseandinsomeofhisessays,Humespendsmoretimedis-cussinghowsomeonessocialstandingaffectsourperceptionsofthepersonscharacter.Indeed,intheTreatisethemoredetaileddiscus-sionofsympathyisputtousetoexplainhowweacquireskillsofsocialdiscriminationandhowwetendtoascribequalitiestooneanotheronthebasisofsocialstanding.Thatis,thegeneralizationsweformonthebasisofsocialcategoriesinformourrecognitionandascriptionofcharactertraits.Itmaythusnotbepossibletolocatepurelyethicalformsofpraiseandblamethatareindependentfromoureverydaydiscourseofrespectorcontempt.vi.thevariabilityofmeritIndiscussingthescopeofmerit,HumenotesseveraltimesthatthehesitationaboutincludingtalentsasmeritoriousqualitiesCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n332jacquelinetaylorindicatesatmostaverbaldispute,andheadvisesthatitisofgreaterconsequencetoattendtothingsthantoverbalappellations(EPM,App.4.22).Thissuggeststhatifwerelyonourcorrectedresponsesofpraiseoradmirationtousefuloragreeablequalitiesasrecom-mendationsofmerit,thenwewillreachbroadagreementonwhichagreeablequalitiesbelonginthecatalogueofvirtuesandwhichdisagreeableonesbelonginthecatalogueofvices(EPM1.10).Inonesensethisistrue.Humeidentifiesauniversalfeatureofmoralitythattranscendsculturesorhistoricalages,namely,thesentimentofpraisedirectedtowardusefulandagreeablequalities,andofblametowardqualitieswiththeoppositetendencies.ButHumesansweraboutwhichqualitieswefindusefuloragreeable,andtheiroppo-sites,ismorecomplicated.OneofthemostinterestingaspectsoftheEnquiry,notpresentintheTreatise,istheattentionHumegivestoculturalandhistoricalvariabilityinthequalities,actions,andpracticesthatwefindpraiseworthyorblameworthy.ThisissueofhistoricalandculturalvariabilityisthefocalpointofADialogue,theshortworkappendedtotheEnquiry.ItisalsoathemeinseveralofHumesessays.WeshouldfirstnoticethatHumeisattentive,ashewasintheTreatise,tovariationwithinaparticularsociety.Thiskindofvari-ationtypicallycorrespondstothedifferentsocialrolespeoplehave:forexample,toapersonsstation(whatwethinkofassocioeco-nomicclass),sex(akintoournotionofgender),age,orprofession.Thevariationinquestionmayrelatetothedegreeofavirtueapersonisexpectedtohave,ortoadifferencewithrespecttowhichvirtuessomeoneshouldhave.Modesty,forexample,isanimportantvirtueforeveryone,butatthetimeHumewaswritingitwasoneespe-ciallyimportantforwomenandyoungmen.Theexpectationwouldbethatwomenandyoungmenshowagreaterdegreeofmodestythanotherpeople.Therearealsoqualitiesorvirtuesthatweexpecttofindinsomeindividualsratherthanothers.Forexample,Humeidentifieschastityasimportantforwomen,courageforsoldiers,anddiscretionforpoliticians.Weesteemthepersonwhohasthetalentsandqualitieswhichsuithisstationandprofessionmorethanhewhomfortunehasmisplacedinthepartwhichshehasassignedhim.AlthoughasHumenotes,heismoreexcellent,whocansuithistempertoanycircumstances(EPM6.20,9).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy333Wemaysometimesregardhistoricalchangesincustomsandlawsasanotherkindofintraculturalvariation.Tyrannicide,forexample,washighlyextolledinancienttimes,becauseitwasalmosttheonlymeansavailablefordealingwithoppressiveregimes.We,how-ever,havelearnedfromhistoryandexperiencethatthismethodproducestyrantswhoaremorecruel(EPM2.19).InSection6,Humearguesthattheancientsvaluedmemory,physicalstrength,andcouragemorethandothemodernsbecauseofthecircumstancesoftheirwayoflife:afocusonoratory,ratherthanwriting,madememoryimportant,whileinamartialorheroicsocietystrengthandcouragewereofgreatervaluethanmanyotherqualities(EPM6.19,26).Incontrast,theconventionsofjusticeandtheresultingstabilityofmoderntimesenableindustryandtheartstoflourish.Suchstabilityalsoallowsustoformenlargedreflectionsaboutthegeneralpreferencesofsociety,andthesestandincontrasttotheprivateornarrowerideasofutilityheldbythoseinmartialorlesscomplexsocieties(EPM9.1,n.57).Bylearninghowtomakeourmoralsentimentsmoregeneralwecanapplypraiseandblamemoreuniformlyacrosstheboundariesoffamily,tribe,ornation.Humesuggeststhatwecanhereregardourselvesastheinheri-torsofacommonpastfromwhichwehavelearnedandcanlearnmore.Thedifferencesinvaluesbetweencultures,andtheimplicationsofinterculturaldifferencesforthenatureofmorality(isitrela-tiveoruniversal),formthemaintopicofADialogue.Thedia-loguetakesplacebetweenPalamedesandanunnamednarrator,andbeginswithPalamedesdescribingacountrycalledFourliinwhichthevaluesoftheinhabitantsarediametricallyoppositetothoseoftheeighteenth-centuryBritish.Thiscountryattachespositivevaluetopracticessuchashomosexuality,incest,suicide,andpar-ricide,andtoactionsthatwewoulddescribeastreachery,perjury,andinfidelity.Thenarrator,hardlybelievingPalamedes,assertsthatsuchbarbarousandsavagemannersareincompatiblewithacivi-lized,intelligentpeople,andperhapsevenwithhumannature(D2,12).PalamedesthenrevealsthathehasbeentalkingaboutancientGreece,apparentlythefavoritecultureofthenarrator,andoneofwhichPalamedeshimselfishighlycritical.Thenarratorinturndescribesacultureinwhichadultery,betrayal,dueling,andtortureCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n334jacquelinetaylorarepraisedratherthanblamed.Heisspeaking,asPalamedesguesses,ofeighteenth-centuryFrance.Thisdialogueraisesseveralissuesaboutmoralrelativism.WithrespecttoancientGreece,theissueofrelativismisinpartaboutarelativismofdistance10about,thatis,anolderwayoflifewithwhichwecanhavenoactualencounter,andthatisnot(atleastnotinitsentirety)aviableoptionforus.Wecanbecriticalofpastmoralitiesandwaysofliving,butourcriticismcannotreformtheirvalues.Accordingtothenarrator,weneedtotrytounderstandthecustomsandmannersofdifferentages,andtojudgepastagesbytheirownstandardsratherthanbyourown.Contemporarycompar-isonsraisedifferentandmoregeneralquestions.Howdeepdothedifferencesbetweensocietiesrun?Whatexplainsthesedifferences?Whatexplainsanycommonalitythatmightexistbetweenthediffer-entsocietiesoftheworld,pastorpresent?Thetwocharactersinthedialogueconsideranswerstothesequestions.Palamedesspeculatesthattheremaybenocommonality,andurgesthatfashion,vogue,custom,andlawarethechieffoundationofallmoraldetermina-tions.Thenarratortakesacontraryview,arguingatlength,thatwemusttracemattershigherinordertoexaminethefirstprin-ciples,whicheachnationestablishes,ofblameorcensure.Evenifcustomsandmannersdiffer,suchqualitiesasknowledge,friend-ship,courage,andhonorarealwaysvalued.And,consistentlywiththeargumentsintheEnquiry,thesequalitiesaresaidtobevaluedpreciselybecausetheyareusefuloragreeable,eithertothepersonpossessingthemortoothers.Soalldifferencesinmoralsmaybereducedtothisonegeneralfoundation,andmaybeaccountedforbythedifferentviews,whichpeopletakeofthesecircumstances(D256,37).Accordingtothenarrator,thesamemoralsentimentsofpraiseforwhatisusefuloragreeable,andblameforthecontrarycompriseuni-versallythefoundationofmorals.Thisargumentadmits,however,oftwoimportantqualifications.Thefirstconcernsthesignificanceofthefactthatthedifferentcircumstances(customsandlaws,forexample)ofdifferentsocietiesmaymeanthattheconceptionofcharactertraits,andanunderstandingofhowtheyaremanifestedin10ItakethetermrelativismofdistancefromBernardWilliams,EthicsandtheLimitsofPhilosophy(Cambridge,MA:HarvardUniversityPress,1984),162.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy335virtuousbehavior,willvary.Thesecondconcernsthedifficultyforbothsocietiesandindividualstoreconcileorachieveallthegoodendsoflife.Inthelanguageofrecentsocialtheory,socialinstitutions,includ-ingfamilial,economic,political,andreligiousinstitutions,arewaysoforganizinghumanactivitieswhichgeneratepracticesthatsustaintheseactivitiesandtheirrelatedwaysoflife.Humeappealstocus-tomtoexplainthesamesortofphenomena.Theancientcustomofinfanticidemaybeexplainedbyscarcityandanunwillingnesstoraiseachildinpoverty,whilesomeoftheancientmarriagesthatwewoulddeemincestuoussimplyreflecttheboundariessetonintimaterelationsbytheGreeks,incontrasttothoseweourselvesset.Thus,thecaringparentofancientGreeceactsquitedifferentlyfromthecaringparentofmoderntimes,andthenotionofafaithfulspousediffersbetweenancientandmoderntimes.Nonetheless,parentalcareandspousalfidelityarevaluedinbothcultures(D2930).11Sincecustomsandmannersvarybetweensocieties,theyalsopro-ducedifferentviewsaboutwhichqualitiesareusefuloragreeable,andconsequentlyhavemoralmerit(EPM6.20).Inseveralofhisessays,aswellasinADialogue,Humearguesthattheformofasocietysgovernmentstructurestoagreatextenttheopportuni-tiesavailabletoitsmembers,andexertsaninfluenceonthequal-itiesthatarevalued.12Inamonarchy,politenessandtheartsareatthefore,whereasarepublictendstofosterindustryandscien-tificadvance.Politenesscorrespondstothestrictersocialhierarchytypicalofmonarchies,whereasrepublicspromoterelativelygener-ouscivilliberties.Throughthenarrator,Humetakesnoteofsev-eralpeculiarcircumstances(isanationatwarorenjoyingpeace?isitpoliticallyunitedordivided?hasitanabundanceofmaterialresourcesordoesitfacescarcity?areitspeopleeducatedorigno-rant?),eachofwhichcanvarythesenseandutilityofsomeactionsandqualities.Healsoobservesthathowwomenaretreated,whethertheyhaveanactiveandessentialroleinbusinessandmutualdis-course,orareexcludedfromallsocialcommerce,isthemost11HumegivesamorenegativeassessmentoftheancientpracticeofinfanticideinhisessayOfthePopulousnessofAncientNations;seeE-PA398400.12HumearguesthecaseatsomelengthinanearlyessayOftheRiseandProgressoftheArtsandSciences,anditisalsoathemeinOfRefinementintheArts.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n336jacquelinetaylormaterialdifferenceoccurringinthedomesticsphereandproducesthegreatestvariationinourmoralsentiments(D38,43).13Thisbringsustothesecondconsiderationabouttherelativityofvirtueandmorals.Aswehavejustnoted,oneofthenaturaleffectsofdifferentcustomsisapreferenceforcertainvirtuesoverothers.Althoughwemayidentifyanumberofgoodendsandadvantagesinlifethatcultivatingavirtuouscharacterhelpsonetoachieve,Humearguesthatcircumstancesmayrenderonequalityor,moretypi-cally,onesetofqualitiesmorevaluablethananother.Eighteenth-centuryFrenchsociety,withitsstrongnormsofpoliteness,stressestheagreeablequalitiesovertheuseful,whereastheBritishesteemmoreamanwhoisusefultohimselfandhissociety.Chancealsoplaysaroleinwhatopportunitiesareavailabletopeople,sincewehavenochoiceaboutthesocietyintowhichweareborn.Eveninawell-orderedandpluralisticsociety,individualscannotachieveeveryadvantageinlife.Inchoosingsomeends,weinvariablysacri-ficeothers.Despitethesevariationswithrespecttovirtueandvicebothwithinandbetweencultures,thenarratorofADialoguemaintainsthatthecustomsandmannersthatproducethemvarynottheoriginalideasofmerit...inanyveryessentialpoint(D51).Itisthemoralsentimentsthatfavortheusefulandagreeablethatarethechieffoundationofallmoraldeterminations,andnotthevariablecustomsandmannersthatembodymoreparticularviewsaboutwhichqualities,actions,andpracticesareusefuloragreeable.vii.languageandmoralsentimentWehaveseenthatinthesecondEnquiryHumemakesseveralimpor-tantclaimsabouttherelationbetweenlanguageandmoralsenti-ment.Morallanguage,particularlythetermsweusetopraiseorblamecharactertraits,ismouldeduponoursentimentsofpraiseofusefulandagreeablequalities,andblameofharmfulanddisagree-ableones(EPM9.8).Thelanguageofmoralsisthusdistinctfromthatofprivateinterest.Inturn,weemploythemoralterminology13HumepursuesthethemeofthetreatmentofwomenintheessaysaswellasintheTreatise.SeetheimportantdiscussionsinTreatise3.1.12,Ofchastity,andinhisessaysOftheRiseandProgressoftheArtsandSciencesandOfPolygamyandDivorces.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy337ofcommonlife,andespeciallythetermsweusetonamecharactertraitsormentalqualities,toguidephilosophicalinvestigationintothegeneralprinciplesandfoundationofmorality.Anexaminationofonesetofmoralterms,thoseconnectedwithbenevolence,estab-lishesthefactthathumannatureincludesother-regardingmotives,aswellasself-interestedones.Finally,morallanguageissomethinginwhichwemustacquireacompetence.Wefindthatourconver-sationswithoneanotherhelpustoimproveinourdiscernmentofvirtueandvice.Nevertheless,whatHumehastosayaboutmorallanguageincon-nectionwiththeissuesofculturalandhistoricalvariabilitythreat-enshisclaimthatacommonmoraldiscourserendersusmutu-allyintelligibletooneanother.AtthebeginningofADialogue,PalamedessaysthatduringhisvisittoFourlihehadtosubmittodoublepains;firsttolearnthemeaningofthetermsintheirlan-guage,andthentoknowtheimportofthoseterms,andthepraiseorblameattachedtothem(D2).Hisclaimhighlightshowboththedescriptionofacharactertraitandthevalueattachingtoitcanvarythroughouthistoryandacrosscultures.Humeextendshisdiscussionofthispointinhis1757essay,OftheStandardofTaste.InthisessayHumeclaims,ashedoesintheEnquiry,thatallculturessharethefundamentalsofmoraldiscourse,includingtermsofpraiseandblame,aswellasthemoreparticulartermsthatrefertothecharactertraitsandactionstowhichpraiseandblamearedirected.So,forexample,allcultureshaveterms,suchasshameful,honorable,admirable,orodious,expressingpraiseorblameatagenerallevel.Wealsofinduniversallysuchtermsasvirtueandvice,termsthatrefertopraiseworthyorblameworthycharactertraitsoractions.Andallculturesappeartorecognizeandevaluatesimilarcharactertraits,andusethesamenames(prudence,honesty,orcowardice,forexample)forthem.Moreover,atermsuchasprudencebothreferstoatraitofcharacterthatwecandescribefurtherandindicatessomeevaluativestanceonourpart(forexam-ple,admirationinthecaseofprudence).ButHumesattitudeinthisessayisreallymoreskepticalthanthisaccountsuggests.Hearguesthattheverynatureoflanguage,thatis,thefactthatourmoraltermsaretranslatablefromonetonguetoanother,accountsforwhatisonlyaseemingharmony.Whenweexaminewhatvariouscul-turesmeanbyaparticulartraitterm,wefindthatthedescriptionofCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n338jacquelinetaylorthetraitdiffersbetweencultures.Eventhegeneraltermsvirtueandviceadmitofdifferencesaboutwhichtraitscompriseadmirableorblameablecharacters.HomersdescriptionofUlyssessprudence,forexample,includesmorecunningandfraudthandoesthedescrip-tionofthatsamecharactersprudencebytheseventeenth-centurywriterFenelon.WhatHomerandhisfellowancientsadmired,we´mayfindreprehensible,althoughbothancientsandmodernsemploythesameterms.Ifthemeaningsofourtermsvary,thenouruseofacommonmoraldiscoursemaynotbeevidencethatwehaveashareddiscourse.Wemaybeunintelligibletooneanotherunlesswesub-mittothepainsofunderstandingtheculturalpointofviewofothers(E-ST3,228).Humeoffersabitmorehelpwiththeissueofmutualintelligibil-ityinOftheStandardofTaste.Althoughwhathesaysconcernsaesthetictaste(orwhathetermscriticism),wemayapplyasim-ilarpointtomorals.Withrespecttothetruestandardoftasteandbeauty,thedisagreementatissueisnotaboutwhatisbeautiful,butratheraboutwhoisacompetentcriticofbeauty.Atruejudgeisonewhopossessesthevirtuesrelevantforjudgingwell.Humeconcedesthatwemaydisagreeaboutwhohasthesevirtues,yetsayshereallyneedstoproveonlythatalltasteisnotuponanequalfoot-ing,andthatsomewillbeacknowledgedbyuniversalsentimenttohaveapreferenceaboveothers(E-ST23,25;241,242).Thisisaninterestingmove,since,ifweapplyittothecaseofmorality,wecandismisstheneedforacriterion(otherthanutilityoragreeableness)forassessingtheactions,attitudes,andpracticesofdifferentcul-tures.Wecaninsteadmaintainthatitistheacquisitionandexerciseofthevirtuesofgoodjudgmentthatyieldaptevaluationsofdifferentmoralities.Ofparticularimportanceforevaluatingthemoralitiesofdiffer-entculturesistheuseofgoodsensetochecktheinfluenceofprejudiceonourmoralsentiments(E-ST22,240).Wemustplaceourselvesinthatpointofviewfromwhichwecanappreciateastheirsthepeculiarviewsandprejudicesofhistoricallyorcultur-allydifferentcommunities(E-ST21,239).Whileweshouldnotletthemannersofourownculturepervertoursentiments,wemaynev-erthelessbeguidedbywhatwetaketobeamodernimprovementinmoralvalues,namely,thecultivationofthesenseofhumanityandCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesLaterMoralPhilosophy339decency.14Wemaybejustlyjealousofamoralstandardthatval-ueshumanetreatmentandmoralinclusion,andsoincertaincases,whateverallowancewemakefordifferentwaysoflife,wewillnotletoursentimentsbepervertedbyinhumanity(E-ST31,33;246,247).Ofcourse,disagreementaboutthedescriptivecontentorvalua-tionoftraittermscanalsooccurwithinasociety.AsHumewaswellaware,theremaybewithinasocietyextremefactionsthatrefusetosubscribetothegoalofmutualintelligibility.Membersofasocietymaydifferaboutthemeaningofpatriotismorcompas-sion,forexample.Forthemostpart,though,theseareissuestobeworkedoutalongthelinesHumehassuggested,throughaprocessofcivilconversationanddebate.Inthisrespect,Humeslatermoralphilosophy,intheEnquiryandEssays,showsamaturitylackingintheearliermoralphilosophyoftheTreatise.Theemphasisontheneedtoconverseandevendebatewithothersaboutwhatwefindusefulandagreeable,andonthevirtuesofgoodjudgment,includ-inggoodsenseandhumanity,suggeststhatHumehaddevelopedadeepappreciationofthecomplexitiesofourmoralsentimentsandtheinfluencesonthem.Humerecognizedthatthedevelopmentofauniversallysharedmoralitywillrequiregreat,andcontinuous,effort,butheisoptimisticaboutthepossibilityofmakingthiskindofmoralprogress.15suggestionsforfurtherreadingInadditiontotheworkscitedinthenotestothisessay,forfurtherreadingthefollowingarerecommended.GeneralBackgroundStewart,M.A.TwoSpeciesofPhilosophy:TheHistoricalSignificanceoftheFirstEnquiry.InReadingHumeonHumanUnderstanding,editedbyP.Millican.Oxford:ClarendonPress,2002.6795.14Humepursuesthethemeofhumanityasamodernconceptinseveralessays;see,e.g.,OfthePopulousnessofAncientNations,OfRefinementintheArts,andOfPolygamyandDivorces.15IpresentedversionsofthisessayattheSouthernCaliforniaPhilosophyConfer-enceatUCIrvinein2004,whereMarkColliercommented,andattheThirdInternationalReidSymposiumattheUniversityofAberdeenin2004.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n340jacquelinetaylorOnVirtue,MoralEvaluation,andDeliberationBaier,AnnetteC.EnquiryconcerningthePrinciplesofMorals:Incompara-blytheBest?InACompaniontoHume,editedbyE.S.Radcliffe.Oxford:BlackwellPublishing,2008.Falk,W.D.HumeonPracticalReason.PhilosophicalStudies27(1975):118.OnSympathyAbramson,Kate.SympathyandtheProjectofHumesSecondInquiry.ArchivfurGeschichtederPhilosophie¨83(2001):4580.Debes,Remy.Humanity,SympathyandthePuzzleofHumesSecondEnquiry.BritishJournalfortheHistoryofPhilosophy15(2007):2757.Vitz,Rico.SympathyandBenevolenceinHumesMoralPsychology.JournaloftheHistoryofPhilosophy42(2004):26175.OntheSensibleKnaveBaier,AnnetteC.ArtificialVirtuesandtheEquallySensibleNon-Knaves:AResponsetoGauthier.HumeStudies18(Nov.1992):42940.Gauthier,David.ArtificialVirtuesandtheSensibleKnave.HumeStudies18(Nov.1992):40128.On“ADialogue”Mazza,Emilio.CannibalsinAdialogue(inSearchofaStandardforMorals).InInstructionandAmusement:Leragionidell’Illuminismobritannico,editedbyE.MazzaandE.Ronchetti.Milan:IlPoligrafo(2005).4566.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nknudhaakonssen11TheStructureofHumesPoliticalTheoryDavidHumebelievedthatmostoftheviewsaboutsocietyandpoliticsprevalentinhisdayhadrootsinoneoranotheroftwospeciesoffalsereligion,superstitionandenthusiasm.Bothweredevelopmentsofconflictingtheologicaldoctrinesthatappealedtotwodifferenttypesofpersonalities.Bothhadcometobeassociatedwithopposingpoliticalinterests.Bothsprangfromignorance.And,whilethetwospecieshadbeenuniversallypresentinsocietyandinindividualsinvaryingdegreesthroughouthistory,thepeculiar-ityofmodernpost-ReformationEuropewastheviolentoscillationbetweenthem,asevidencedbythemanywarsofreligion.Theirmoreextremeadherentswerealso,notleast,responsiblefortheplightofmodernBritain,bothnorthandsouth.Oneofthetasksofthephilosophicalhistorian,Humebelieved,wastoexplaintheprepon-deranceatparticulartimesofoneortheotherofthesepersuasions.Thetaskhesetforhispoliticaltheorywastoexplainwhybothwerephilosophicallymisconceived,empiricallyuntenable,and,intheirextremeforms,politicallydangerous.i.thepoliticsofreligionOnepartofhumanity,Humenotes,hasatendencytoweakness,fear,[and]melancholy,togetherwithignorance.Inthisstatetheimaginationconjuresupforcesoperatingunderthesurface,andthemindispronetograspmethodsofinfluencingtheseforcesbycere-monies,observances,mortifications,sacrifices,presents,or[by]anypractice,howeverabsurdorfrivolouswhicheitherfollyorknaveryrecommendstoablindandterrifiedcredulity.ThisconditionandthesepracticesHumecallssuperstition.Inreligion,priests,church341CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n342knudhaakonssenestablishments,andritualsareusedtomediatebetweentheindi-vidualandtheseforces.Insocietyandinpolitics,thesuperstitiouspersonisdisposedtoacceptestablishedformsandpowersasinherentinthenatureofthingsandtoseesocietyasahierarchicalstructurewithamonarchastheunitarysourceofauthorityandsovereigntyasadivineright(E-SE2,74).Incontrast,anotherpartofhumanityhasatendencytohope,pride,presumption,[and]awarmimagination,togetherwithigno-rance.Inthisstate,whichHumecallsenthusiasm,individualstakeflightsoffancyfromtherealworld,presumedirectrapportwithhigherpowers,andinclinetowardungovernableself-assertion.Inreligion,priests,churchestablishments,andritualsarerejected.Insocietyandpolitics,enthusiastsasserttherightsoftheindividual.Theyofteninclinetoforcefulremodelingofauthorityandgenerallyseeself-governmentastheonlypropergovernment,atleastinprin-ciple.Enthusiastsfavorcontractualistaccountsofsuchauthorityastheywillacceptandinsistontheprotectionofindividualcivilliberties(E-SE3,74).1Humespoliticaltheoryismorethananoutrightrejectionofsuchreceivedideasasthoseassociatedwithsuperstitionandenthusi-asm.Hemeanthispoliticalwritingstobealsopoliticalacts,shap-ingtheopinionsorbeliefsthatinturnshapedpoliticsandsociety.Toachievethisend,hesoughttoprovideatheoryofthenatureofsocialandpoliticalphenomenadifferentfromthosethatservedtoreinforcesuperstitionandenthusiasm.Ontheonehand,hepro-ceedsbyanalyzingthosebeliefsthatinrecenthistoryhadtendedtomodifytheidealtypesofsuperstitionandenthusiasm;ontheotherhand,hearguesthatsuchanalysisinitselfformsasetofopinionsorbeliefswithdirectandbeneficialpoliticalconsequences.Speakingintheidiomofthetime,heshowedhowhisprinciplesledhimtotakeoneortheothersideincurrentdebates.Often,ofcourse,histopicalconclusionsobscuredthetheoreticalpremises,notonlyforhiscontemporariesbutforsubsequentgenerationsofinterpretersaswell.ThemainprobleminexplainingHumespoliticalthoughthasalwaysbeenhowtoprovideaclearunderstandingoftheclose1ForexamplesofHumesanalysisoftheoriginsofprominentmodernformsofenthusiasm,namely,CongregationalismandQuakerism,seeHE57.2731,62.719;5:4413,6:1426.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory343coherencebetweenthegeneralandtheparticularandthetheoreti-calandthehistorical.HistheoryofthenatureofsocialandpoliticalphenomenaismainlytobefoundinthethirdbookoftheTreatiseandinthesecondEnquiry,whiletheparticularsofthehistoricallycontingentsituationofmodernBritainandEuropeareanalyzedinmanyofhisEssaysandintheHistoryofEngland.Inordertounder-standeither,wehavetograspthesenseinwhichbasicsocialandpoliticalinstitutionsare,accordingtoHume,artificial.Thiscanbestbeachievedbylookingatthephilosophicalideasunderpinningsuperstitionandenthusiasm.Elsewherethismighthavebeencalledhismetaphysicsofpolitics;Humesaimwastounmaskthepoliticsofreligiousmetaphysics.ii.morals–foundorconstructedHumewaskeenlyawareofthecontinuinginfluenceofideasderivedfromAristotleandmediatedbyscholastictradition.Fromthisper-spective,socialforms(suchaspropertyandcontract)andpoliti-calroles(suchasmagistracy)havetheirfoundationinessences,ininherentstructuresfoundinnatureitself.Onsuchatheory,spe-cificactionsareonlyproperty-holding,contracting,orgoverninginsofarastheyareanattempttoactualizetheinherentmeaningortheessencesignifiedbythesewords.Moral,social,andpoliti-calrelationsbetweenpeoplearenotconstructedbytheindividu-alsinvolved;suchrelationsareestablishedwithreferencetosome-thingoverandabovethepersonsconcerned,namely,anobjec-tivestructureofrealityandmeaningonwhichindividualstrytodraw.Humesawtheseideasasthephilosophicalequivalentofthereligioushocus-pocusofsuperstition(transubstantiation,forexam-ple).Likemostsuchideas,thisphilosophysupportedtheneedforauthoritativeinterpretersofthemeaningssupposedlyinherentin,oressentialto,lifeinsociety.Itwas,inotherwords,thephiloso-phybehindCatholicism,HighChurchAnglicanism,old-fashionedToryism,absolutism,anddivine-rightmonarchism.Thereactionsinpost-ReformationEuropetothesedirectionsinreligionandpoliticswere,asHumerealizedmoreclearlythanmost,immenselycomplexandoftencontradictory.Itwaspossible,how-ever,todiscernsomeofthephilosophicalideasthatwerebasictomuchReformationthought,andthatwereeventuallyspelledoutCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n344knudhaakonssenwithgreatclaritybynaturallawphilosopherssuchasHugoGrotius,ThomasHobbes,SamuelPufendorf,andJohnLocke.ThisProtestantnaturaljurisprudencewasanintellectualphenomenonsopervasivethatthinkerswhowereveryfarfrombeingnaturallawtheoristsalsoadoptedpartsofitsconceptualapparatus.ItwasalsoclearlypartofthecontextinwhichHumedevelopedhisviews.WiththepartialexceptionofGrotius,thesewritersheldthattherewerenomoralorpoliticalmeaningsinherentinthestructureofthings.Allmeaning,orvalue,iswilledorconstructedandimposedonanat-uralworldthatinitselfisamoralandapolitical.ThebasicactofwillisthatofGodwho,inchoosingtheparticularhumannatureHedid,delegatedtohumansthetaskofcreatingmoralandpoliticalformsthatwouldmakepossiblethecultureofhumanity.Accord-ingtomostProtestantnaturallawthinkers,humanreasoncould,unaidedbyrevelation,derivefromthecharacterofhumannatureandthehumanpositionintheworldacertainguidanceinmoralsandpolitics,andthisiswhattheycalledthelawofnature.Gener-allyspeaking,thebasiclawofnatureheldthat,sincepeopleweresociableand,indeed,hadtobesociableinordertoexistatall,var-iousmeasureshadtobetaken.Thesemeasureswerecontainedinderivativelawsofnaturethatspecifiedthecreationofmoralandpoliticalinstitutionsrangingfrommarriageandpropertytocivilgovernmentandthelawofnations.Afewthinkers,notablyGrotiusandHobbes,triedtoformulateatheorythatweakenedtheroleofnaturallawasaguideforthehumanwill.Onthisview,socialandpoliticalformsaresettlementsnegotiatedbetweenindividualswithoftenconflictingclaimsandintentions,orrights.Naturallawinthisschemeissimplythelessonslearnedfromsuchsettlements,nottheprescriptionforhowtomaketheminthefirstplace.2ThedivisionbetweenanaturallawdirectionandanaturalrightsdirectioninProtestantnaturallawtheorywasoffundamental2Inaddition,thelawofnaturewas,ofcourse,consideredapositivelawofGodasrevealedinHisWord,butinthisguiseitcouldonlybeconsideredalawforthosewhoreceivedtheWord,namely,Christianbelievers.ConcerningtherelationshipbetweennaturallawandnaturalrightsinGrotiusandHobbes,seeKnudHaakon-ssen,NaturalLawandMoralPhilosophy:FromGrotiustotheScottishEnlight-enment(Cambridge:CambridgeUniversityPress,1996),Chap.1;andRichardTuck,NaturalRightsTheories(Cambridge:CambridgeUniversityPress,1979),esp.Chaps.3and6;andHobbes(NewYork:OxfordUniversityPress,1989).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory345importanceforthefurtherdevelopmentofpoliticalthought,asweshallsee.Forthemoment,however,thesignificantpointisthatbothformsofnaturallawtheoryapparentlysubscribedtotheviewthattheinstitutionsofmoralandpoliticallifearecontractuallycon-structedbyindividuals.Theseideasofpersonalautonomy,ofindividualrights,oftheabsenceofmediatingfactorsbetweenGodandman,andofthecon-sequentconstructionofmoralsandpoliticsaccordingtoourownlightsideasidentifiedsincewithconstructiviststhesepartsofthephilosophicalargumentcouldbetakentotheextremesofenthu-siasminreligionandfanaticalfactionalisminpolitics.Thishadhap-penedrepeatedlyinmanypartsofEurope,inHumesopinion,butneverwithmoredevastatingeffectsthaninseventeenth-centuryBritain,markedasitwasbyreligiousstrifeandcivilwarfare.Eveninhisowntimethepoliticaleffectsoftheenthusiasticcastofmindremainedadangertobeguardedagainst;asHumegrewolder,hesometimesfearedthatthebattleagainstitmightyetbelost.3iii.humeonjusticeThetheoryofsocialartificepresentedinthethirdbookoftheTrea-tiseisanattempttoformulateapositionmediatingthetwophilo-sophicaltraditionsbrieflyoutlinedintheprecedingsection.Hume,ofcourse,hasnotimeforscholasticessences,andhisnaturalismprecludesanyroleforthedivinevoluntarismofmuchProtestantnaturallaw.Humesindividualscanexpectneitherinherentstruc-turesnortranscendentguidance.OnlyHobbeshadisolatedhuman-itymetaphysicallyandreligiouslyascompletelyasHume,yetthetwothinkersreachverydifferentconclusionsaboutthehumancon-dition.ItisnotonlythatHumegivesagooddealmorecredittothegeneroussideofhumannature.Healsogivesanaccountofthesocialrelationsbetweenindividualsthat,whilesharingtheindividualisticnaturalismofHobbes,isprofoundlyun-Hobbesian.3Inanumberoflettersinthelate1760sandearly1770s,HumeexpressedhisfearandloathingfortheLondonmobsriotinginsupportofthereelectiontoParliamentoftheoutlawedJohnWilkes.HumesawitasadegenerationofthedemandforlibertytoasenselessfanaticismthatEnglishfreedomallowedtofeedonitself,thuscreatingfactionalismandbarbarismofasortthatcouldendangerthisveryfreedom.SeeHL2:1801,1912,20911,21213,216,261;NHL196,199.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n346knudhaakonssenTheactionsthatspringfromthenaturalvirtuesandvices(benef-icence,clemency,moderation,andtheiropposites,forexample)are,accordingtoHume,entirelynatural,andhavenodependanceontheartificeandcontrivanceofmen(T3.3.1.1).4Eachoftheseactionsisasimpleorself-containedactthatestablishesrelationsbetweenpar-ticularagentsandparticularpatients.Anactofbenevolence,oritsopposite,iscompletedasoneactoroneoccasionwhenthatvirtue,orvice,isbeingexpressed,forexample,bythegivingofagiftorthedenialofaservice.Suchanactivitymaystretchovertime,butitisneverthelessinasignificantsenseoneact.Actsofbenevolencemay,ofcourse,promptreactions,suchasgratitude,butthesereac-tionsareclearlyotheracts.Actsthatresultfromnaturalvirtueandvicearecoherentandself-containedbecausetheyhaveapointorameaningwhentakeninisolationevenwhenseenasnothingmorethanrelationsbetweenspecificindividuals,apointHumeempha-sizesbynotingthatwevalueeachperformanceofanaturalvirtue(T3.3.1.1112).Thenaturalvirtues,commonlydenominatedthesocialvirtues,tomarktheirtendencytothegoodofsociety,providethebasisforfamilylifeandintimatecirclesoffriendship,butsociallifeatlargerequiressomethingelseentirely,namely,asetofartificialvirtues(T3.3.1.11;cf.3.2.2.46).WhenIasagentabstainfromtakingthefruitofmyneighborspeartree,paymylandlordhisrent,oranswerthegovernmentsmilitarycall-up,myactionscannotbeunderstoodinisolationasmereexpressionsofinherentfeaturesofmynature.Theseactionshavereferencetosomethingelse,tosomethingbeyondtheotherpersonorpersons,thepatient(s)affectedbythem.Thispatientmaybeunknowntotheagentormayhavebeenundeservingoftheagentsbehavior:theneighbormayneverharvesthisfruit,thelandlordmaybeexcessivelyrichandgrasping,thegovernmentmaybeconductinganunjustwar.Insuchcasesanagentsbehaviorcanonlyhavemeaningandonlybeevaluatedthroughitsrelationtosomeadditionalfactorbeyondboththeagent(s)andpatient(s)involved.Ithasmeaningonlywithinaframeworkthatisinanimportantsenseobjectiveanddistinctfromindividualsandtheir4Foranotherdiscussionofthedistinctionbetweenthenaturalandtheartificialvirtues,seeinthisvolumetheessayFoundationsofMoralityinHumesTreatise,PartIIIA.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory347qualities.Therelationsbetweenpeoplewhoholdpropertytotheexclusionofothers,whocontractforexchangeofgoodsorservices,andwhooweallegianceorsupportofsomesorttheserelationscanbeestablishedonlybecausethepeopleinvolvedhavesomethingotherthaneachothersintentionstoreferto,somethingthatcanshapetheirintentions.Mygivingmoneytoanotherpersondoesnotconstitutepayingrentmerelybecausewehave,respectively,intentionsofgivingandreceiving.Thetransactionisgivenitspar-ticularmeaningbecauseitinvolvesasocialpracticeorinstitution,inthiscaseaspecialformofcontract.Inotherwords,individualactionsofthissortarenotself-containedandcomplete.Wecannotseetheirpointandevaluatethemwithoutinvokingthesocialprac-ticetowhichtheyrelateoronwhichtheyrely.Individualactionscanbeapprovedofasinstancesofsuchinstitutionalizedpracticesasholdingprivateproperty,keepingpromisesandcontracts,payingallegiance,andthelike,becausesuchpracticesalreadyexistandareapprovedof.Thispeculiarcircumstanceis,asHumeexplains,wellillustratedbyactionsthatseemabsurdwhentakeninisolation,butthatacquiremeaningandcanbeevaluatedonceweassumetheirreferencetoasocialpracticeofthesortmentioned.Consequently,whenweseeapoorpersonpayingmoneytoarichone,weassumethataloanisbeingsettledorgoodspaidfor(T3.2.1.9).Humesanalysisofthenatureofsocialactionsisathoroughrejec-tionofwilltheories,suchascontracttheories.Thatis,herejectstheoriesaccordingtowhichsuchsocialactionsasrespectingprop-ertyclaimsacquiremeaningbecausetheyderivefromactsofwilloftheparticipatingindividuals.LikethinkersintheAristotelianandThomistictradition,Humeholdsthatactsofwillcanonlyestab-lishsocialrelationsoutsideintimategroupsiftheseactsaregivenmeaningbysomethingoverandabovethemselves.5Incontrastto5ThescholastictheoryofcontractderivesfromAquinasstheoryofpromises,SummaTheologica2.2.88.Thelatescholastics,especiallyintheSpanishschools,madeasophisticatedcombinationofthisdoctrineandtheRomanlawoncontracts.ThiscombinationhadanenormousinfluencethroughtheseventeenthandeighteenthcenturiesoneventhenaturallawyerswhohelpedunderminethephilosophicalbasisforthedoctrineandwefinditincivilianlawyerslikeJeanDomatandRobertJosephPothier,whoinfluencedtheFrenchCodeCivil(1804).SeeDomat,Lesloixcivilesdansleurordrenaturel(Paris,1689),BookI;Pothier,Traitedesobligations´(17614).ThemodernalternativetotheAristotelian-ThomisticideaofcontractsastheactualizationoftheinherentessenceofcontractingwascommonlyseentoCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n348knudhaakonssenthattradition,however,Humerejectedtheviewthattherearefixedandessentialmeaningsforsuchsocialinstitutionsaspropertyandcontract.Suchinstitutionsarenomorethanpractices,afacthesig-nalsbycallingthemandtheirassociatedvirtuesandvicesjusticeandinjustice,forexampleartificial(T3.2.1.1719).Theyarearti-ficialbecausetheyarehumancreations.Atthesametime,Humehasdeprivedhimselfofthesimplecontractualistaccountoftheseinstitutionsasexpressionsofwill.Onhisaccount,propertyandcon-tractmustexistassocialpracticespriortoanyactsofwillrelatingtothem.Humehasthussaddledhimselfwithageneticproblem,namely,howtoaccountfortheoriginsofthesocialpracticesthatconstitutebasicsocialinstitutions.Thesolution,HumesuggestsinOftheoriginofjusticeandprop-erty(T3.2.2),involvesluck,moderateforesight,andimitativebehav-ior.Weinevitablyliveinfamilyunits,andalthoughthisislargelyaresponsetonaturalpassionsandnaturalvirtues,aswellastothenumberlesswantsandnecessitieswithwhichnaturehaslumberedhumanbeings,itprovidessomeexperiencewithrelativedivisionsintomineandthineandwithtrust(T3.2.2.2).Itrequiresonlymodestluckandprudencetoattempttoimitatethisinrelationswithpeo-pleoutsidethefamilygroup.Thescarcityofgoodsandabilities,inrelationtoneedsanddesires,putsapremiumonmakingasuccessofsuchattempts.Itisthereforeeasytoseehowitmaybecomecommonpracticetorespectpeoplepossessing,transferring,andexchangingthingsthatinonewayoranotherareassociatedwiththem,andthencomingtotrusteachotherswordaboutfutureactions.Thegeneralpatternofsuchpracticesmaybeexplainedbythewaytheimag-inationworksalongempiricallyestablishedassociativelines(seeT3.2.35).bethecombinationofnominalisticdefinitionsandwilltheoriesinthinkerslikeHobbesandLocke.SeeHobbes,Leviathan,ed.RichardTuck(Cambridge:Cam-bridgeUniversityPress,1991),945;Locke,AnEssayconcerningHumanUnder-standing,ed.P.H.Nidditch(Oxford:ClarendonPress,1975),1.3,2.28,4.4;seealsoLocke,TwoTreatisesofGovernment,ed.P.Laslett(Cambridge:CambridgeUni-versityPress,1960),2.81.Itmustbestressed,however,thatthescholasticformofteleologywaswidelyreplacedbytheteleologicalschemeofnaturalreligion,andthelatterwasnotmuchmoresuitedtosupportpurewilltheoriesofpromisingandcontractingthanitspredecessor.Eighteenth-centurytheoriesofpromiseandcontractlegalaswellaspoliticalarethereforemostlycomplicatedandconfused,acircumstancethatmakesHumestheoreticalclarificationthemoreremarkable.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory349Itisaquestionofhowsuchpracticesgainsufficientstrengthtowithstandthepressureofconflict,forinstance,insituationsofsocialexpansionandscarcity.Thetwobasicrequirementsarethateachpracticeshouldcometobevaluedindependentlyofitsindividualinstantiationsandshouldbeseenasbindingorobligatoryontheindividual.Ratherthanbeingjustthesumtotalofwhatpeopledo,socialpracticeshavetobecomeindependentrulesspecifyingwhatisgoodandtobedone.Humehereoffersaradicalsolutiontowhathadprovedtobeoneofthemostintractableproblemsinmoralphilosophy,therelationshipbetweengoodnessandobligation.Atoneextremewerethosewhothoughtthathumannaturehadbeensoimpairedbyoriginalsinthathumankindhadnoinsightintomoralgoodnessandcouldbedirectedandgovernedonlybybeingobligedtocertainformsofbehavior.Theobligingwillsmightbethoseofahierarchyofauthorities,terminat-inginGod,asinmuchLutheranthought;ortheymightbethoseofeachindividual,reflectingdirectlythewillofGod,asinmuchCalvinistthought.Eitherway,wehaveawilltheoryofmoralsandpoliticsofthesortHumethoughtimpossible,andwedonothaveanaccountthatmakesanynecessarylinkbetweenmoralgoodnessandbeingobliged.Incontrasttothislineofthoughtwereawidevarietyoftheoriesthatallallowedthat,eveninitsfallenstate,humanitywasleftwithsomenaturalcapacityformoralinsight.InHumesrecentpast,theyrangedfromCambridgePlatonismandtheratio-nalismofSamuelClarketothemoralsensetheoriesofShaftesburyandHutcheson.Theproponentsofsuchtheoriesallhadthetaskofexplainingwhetherandhowinsightintomoralgoodnesshadimpli-cationsformoralobligation.Theyallthoughtitdid,andtheyallhadextremedifficultyinaccountingforit.6Theproblemwasaserioustheologicalone.Ifeachpersonhadanaturalmoralfacultythatcouldbringmoralunderstanding,andifsuchunderstandingimposedamoralobligation,thenGodsmoralroleinhumanlifewasseverelycurtailed.ThemorallygoodpersonwouldnotneedGod,whosemoralfunctionwouldbereducedtothatofpolicingthemorallywayward.Thiswasclearlyunacceptable,foritwouldmakemoralsideallyindependentofGod.Accordingly,inall6SeeStephenDarwall,TheBritishMoralistsandtheInternal‘Ought,’1640–1740(Cambridge:CambridgeUniversityPress,1995).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n350knudhaakonssenthesetheorieswefindsomeresidualelementofdivinevoluntarism.Generallyspeaking,awayoutwassoughtinsomevariationonthefollowingtheme.SincetherelationshipinGodsnaturebetweenmoralinsightandmoralwillisunbrokenwhateveraccountoftherelationshiptheologiansmaygiveandsincehumanstosomeextentshareinthemoralinsight,apalereflectionofthisrelationshipmaybeestablishedinhumannatureifmenandwomencanpartakeinGodswillandinsomemeasuremakeittheirown.Thechiefwayofaccountingforthiswithoutresorttorevelationwasteleological.Theparticularandconfinedmoralgoodthateachpersonandcom-munityisabletoeffectmaybeunderstoodasacontributiontotheoverallgoodofthemoraluniverseofallmoralagentspast,present,andfuture.Thisuniversalgoodisunderstoodtobetheintentionofthedivinityasshowninthepurposefulnessofcreation.Conse-quentlyourparticularwilltodotheparticulargoodinourpowerispartofGodsgeneralwillforthemoralcreationasawhole.Ifonoccasionwelackthatparticularwill,orifitisweakandunde-cided,thethoughtoftheteleologicalarrangement,thatis,ofGodswill,isabletosupplythewant.Wearethenactingoutofasenseofobligation.ThistypeofteleologyandtheassociatedChristianutilitarian-ism,asitisnowoftencalled,wasprobablythemostpervasivestyleofmoralandpoliticalthoughtintheeighteenthcentury.7AnearlyformulationbyRichardCumberlandhadsomeinfluence,butthemostimportantversionwasundoubtedlyFrancisHutchesons.8ThislineofargumentprovidedthebasisfortheempiricalscienceofmoralsthatcharacterizedagreatdealofEnlightenmentsocialthought.Sincesomuchdependedontheteleologicalarrangementoftheuniverse,animportanttaskofthescienceofmoralswastoprovideamapofthemoralworldshowinghowitsvariouscom-ponentsideallyfittedtogether.Thepopularscienceofmoralswasthusadescriptionoftheproperworkingofthemoralinstitutions7SeeJ.E.Crimmins,JohnBrownandtheTheologicalTraditionofUtilitarianEthics,HistoryofPoliticalThought4(1983):52350.8SeeHaakonssen,TheCharacterandObligationofNaturalLawaccordingtoRichardCumberland,inEnglishPhilosophyintheAgeofLocke,ed.M.A.Stewart(Oxford:ClarendonPress,1993),2947,andNaturalLawandMoralPhilosophy,6399.ForfurtherdiscussionofHutchesonstheory,seeinthisvolume,FoundationsofMoralityinHumesTreatise,PartI.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory351currentlymakingupsocietyproperbeingdefinedintermsofmak-ingsociallifepossibleasacontributiontothegeneralhappinessofhumankind.Humematchedthisagendapointforpoint.Oncesuchformsofbehaviorasrespectforthepossessionsofothersandkeepingofpromiseshavebecomefairlycommon,itwillbeevidenttoallthattheyaresociallyusefulbyallowingthingstobedonecollaborativelythatotherwisecouldnotbedone.Thissocialutility,orpublicgood,ismerelytheoutcomeofindividualactions,butitappearsasthoughitweretheresultofashareddesign.Consequentlyindividualsareinclinedtoapproveofthebehaviorthatbringsaboutthepublicgood,foritappearsasthoughsuchbehaviorwereaimedatthisout-come,and,contrariwise,todisapproveofbehaviorhavingcontraryeffects.Inthiswaythebasicrulesofjusticepertainingtopropertyandcontractcometobeacceptedasmoralrules.Inshort,whilethepurposefulnessofcertaingeneralpatternsofbehaviorisonlyappar-ent,theperceptionofthisapparentpurposefulness,orteleology,initselfbecomesanindependentcauseofsuchbehaviorinthefuture.Theproblemisthatnoteveryapplicationoftherulesofjus-ticeproducesgoodresultsforalltheindividualsconcerned,or,inextremecases,foranyone.Nonetheless,becauseoftheirgeneralpublicutility,westillthinkthattherulesshouldbekept,orthattheyareobligatory.WhilethepopularmoralphilosophysketchedpreviouslyinvokedwhatwemaycalltheinternalizationofGodswillinordertoaccountforobligation,Humesuggestsinsteadthatweinternalizeasocialwill.Inasocialgroupwherejustbehaviorisgenerallyapprovedasgoodbecauseitproducessocialutility,peoplewhoinaparticularcaselackanymotiveforjusticeperhapsbecauseneithertheynoranyotherassignablepersonstandtogainanythingfromtheactioninquestionwilltendtohaveamotivesupplied.Becauseeveryonegenerallyapprovesofjustbehaviorasifitsprangfromaseparatelaudablemotive,peoplelackingsuchamotivewillfeelmorallydeficientascomparedwiththeirsurroundingsandwillcometodisapproveoforhatethemselvesonthataccount.Inthistheywillbereinforcedbythedisapprovaloftheirfellows.Thisself-loathingbecomesthemotiveorthewillbywhichpeopleactjustlyasamatterofobligation.Wemayalsosaythatjustbehaviorhasbecomeanartificialaccre-tiononthenaturalperson.WedisapproveofdeficienciesalackofCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n352knudhaakonssenacertaindegreeofbenevolence,forexampleinthenaturalmoralqualitiesandseeitasanobligationtoreinstatebenevolencetoitsnaturalplaceamongourmotives.Inthesamemannerwehavelearnedtoseethefailuretohaveamotiveforjusticeasadeficiency.Sincethereisnomotivetobereinstatedinthiscase,wehavetoinventone,namely,thewilltobefullmoralcharacterslikeotherpeopleinoursociety.AcrucialconceptinHumesanalysisofobli-gationisthusthatofcharacter.Partofourmoralcharacterisnatu-ral,partofitderivesfromsocialliving.Deficienciesintheformerevokeanatural,inthelatterasociallyinduced,desiretorepairourcharacter.Thesedesiresare,respectively,ourobligationtothenat-uralandtotheartificialvirtues.9Thiscausalaccountofthemoralobligationtopursuetheartificialvirtues,typifiedbyjustice,isthecrowningeffortinHumessubversionofthereigningparadigmsofmoralandpoliticalphilosophy.Ithadanumberofnolesssubversiverepercussions.iv.thebasisforauthorityTherulesofjusticeformthebasisofsociallifeinextendedgroups.Yettheobligationtoobeytherulesofjusticedependsonnothingmorethaneachpersonsperceptionofthegeneralsocialopinionoftheserules.Althoughtheformationofsuchopinionisastronganduniversaltendencyinhumanlife,itisclearlysubjecttoseveredis-ruptionandfluctuation.Peopleaccordinglyseektoprotecttherulesofjusticebytheinstitutionofgovernment(T3.2.8.5).Butthenonemustask,whatisthebasisfortheauthorityofgovernmenttoadmin-isterjusticeortodoanythingelse?OrinHumeslanguage,whatisthesourceofallegiancetogovernment?Inansweringthisquestion,hefollowsapatternsimilartothatemployedinhisanalysisoftherulesofjustice.9SeeT3.2.1.48,3.2.2.238,3.2.5.47,1013,3.2.8.7,3.3.3.2,andthewholeofSect.3.3.1(Oftheoriginofthenaturalvirtuesandvices).ForfurtherdiscussionsofHumestheoryofobligation,seeHaakonssen,TheScienceofaLegislator:TheNaturalJurisprudenceofDavidHumeandAdamSmith(Cambridge:CambridgeUniversityPress,1981),305.SeealsointhisvolumeHumesMoralPsychology(PartV)andFoundationsofMoralityinHumesTreatise(PartIII).BothoftheseessaysconsiderHumesaccountofobligationandtheroleofsympathyinhismoraltheory.ThelatteralsoconsidersHumesaccountofjustice.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory353ThetraditionalTorynotionthatauthorityisinherentinthesocialworldintheformofadivinerighthastoberejectedbecauseitinvokesforcesaboutwhichhumanscanknownothing.Thetradi-tionalWhignotionthatauthorityderivesfromcontractualarrange-mentsis,Humeargues,empiricallyfalseandconceptuallycon-fused.10Theessentialfeatureofacontractualarrangementisthatitinvolveschoosingwhetherornottoenterintothearrangement:butachoicethatisunknowntoachooserisnotachoice.Itseemsimpossibletoidentifyanycontractbyvirtueofwhichanygrouplivingunderaparticulargovernmentowesallegiancetothatgov-ernment.Thegeneralityofhumankindknowsnothingofsuchacontract,andeveniftherehadbeensomecontractinthepast,itwouldnotcarryauthoritybeyondtheoriginalcontractors.Humethusfindsincoherentthecommonsuggestionthatthereisatacitcontract,acontractaboutwhichapeopledoesnotknoworthink.Furthermore,individualsonthewholehavenochoice.Wearegen-erallybornintosocietiesthatarealreadysubjecttogovernmentandfindourselvesobligedtoobeythelawsofthatgovernment.Peopleofaparticularlyenthusiasticcastmay,ofcourse,saythattheyalwayshavethechoiceofdyingratherthanlivingwithwhattheyconsideratyrannicalgovernment.TheseareexactlythepeopleHumefearsmostofallbecauseintheirfanaticismtheycoulddestroyexistinggovernment,andtheirwildnessoftempercouldneversustainalast-inggovernment(seeT3.2.79).10HumeexplainsthelabelsWhigandToryinHE68.12,6:381,andaccountsfortheemergenceoftheWhigandTorypartiesattheRevolutionof16889inHE71.5672,6:52334.ThebasicpartyprinciplesandtheirconnectionwiththelaterdivisionbetweencourtandcountryinterestsarelaidoutinE-PGB612,6972,andE-CP.Pre-RevolutionToryideasofdivinerighttoruleowedagreatdealtoRobertFilmersPatriarchaof1680;seePatriarchaandOtherWritings,ed.J.P.Sommerville(Cambridge:CambridgeUniversityPress,1991);amongmanypost-Revolutionrestatements,CharlesLesliesvoluminousoutputisrepresentative,forexample,TheConstitution,Laws,andGovernmentofEnglandVindicated(London,1709)andTheFinishingStroke(London,1711).TheroleofJohnLockesSecondTreatiseofGovernment(1689;inTwoTreatisesofGovernment)fortheformationofWhigprinciplescontinuestobeamatterofdispute,asdoesthesignificanceofradicalcontractarianismingeneral.RepresentativeexamplesofthesortofWhiggismHumehasinmindaretheanonymousVoxPopuli,VoxDei:BeingTrueMaximsofGovernment(London,1709);DanielDefoe,TheOriginalPoweroftheCollectiveBodyofthePeopleofEngland,ExaminedandAsserted(London,1702);andBenjaminHoadly,TheOriginalandInstitutionofCivilGovernmentDiscuss’d(London,1710),esp.Part2.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n354knudhaakonssenThecontracttheoryofallegiancetogovernmentisinanycasemuddledinexactlythesamewayasthecontractualaccountofprop-erty.Ittriestoreduceallegiancetoactsofwillbyindividuals,butindoingsoitpresupposesthatthereisagovernment,thatis,anauthoritywithsomeclaimtoallegiancetowhichindividualspledgethatallegiance.Individualactsofobedience,intheformofpromises,forinstance,canonlyberecognizedasexpressionsofallegianceiftheobjectofsuchbehavioristhesortofpersonorgroupofpersonstowhomallegianceisdue.Governmentalauthoritymustthereforerestonsomethingexistingpriortoanysuchpromiseofallegiance.Inthetermsusedintheaccountofproperty,wecanseethatthesubjectsofgovernmentmusthaveaninterestingovernmentdis-tinctfromtheirinterestinkeepingtheirpledgeofallegiance.Theinterestinquestionis,ingeneralterms,aninterestinexternalandinternalprotectionand,especially,intheadministrationofjustice.Totheextentthatsuchinterestestablishesobedienceasageneralpatternofbehavior,allegiancebecomes,likejustice,anotherartifi-cialaccretiononthenaturalpersonalityofthoseinvolved.Oncethishashappened,theabsenceofsentimentsofallegianceisperceivedasapersonaldeficiency.Inthisway,allegiance,likejustice,becomesamatterofnotonlythenaturalobligationsofinterest...butalsothemoralobligationsofhonourandconscience(T3.2.8.7).Humesideaoftheobligationofallegiancehasacertainsimilaritywithaformofcontracttheorythathadsomecurrencyinhistime,butthathenevermentionsatall,namely,impliedcontract.Infact,inhisrejectionoftacitcontract,heseemstosuggestthathedidnotseeanydifferencebetweenthesetwotheories.Thosewhodiddistinguishbetweentacitandimpliedcontractssawtheformerasavoluntarycommitmentsignaledinanonverbalway,butstillasanidentifiablebehaviorialevent.Ontheotherhand,animpliedcontractdoesnotarisefromanyparticularevent;thereisnoactofwill.Thecommitmentofanimpliedcontractfollowsfrom,isimpliedby,whatapersonisorwhatpositionoroffice(spouse,child,doctor,neighbor,citizen,magistrate)heorsheholds.ThiswasawayofthinkingaboutsocialrelationsthathadAristotelian,Stoic,andChristianorigins,andthathadbeentranslatedintothecommonteleologicallybasedsystemsofmoralsoutlinedhere.Hume,too,thoughtthatdutiesarisefromwhatapersonis,butthiscouldnotbeaccountedforteleologicallyintermsoftheoverallaimofthesystemCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory355ofmoralbeings,nor,becausethereisnoactofwillinvolved,wasthereanyreasontoinvokecontracttoaccountfortheseduties.Humestheoryofallegiancealsosaddleshimwithageneticprob-lem:howtoaccountforthefirstoriginsofgovernment.Inhisear-lierworksheiscontenttogiveabriefandblandexplanationtotheeffectthat,sincegovernmentissuperimposedonsocialgroupsthatalreadyrecognizetherulesofjustice,includingtheobligationtokeeppromises,itispossibletoseethefirstinstitutionofgovern-mentasamatterofmutualpromises.Itisclear,however,thathisconcernistodiscredittheideathatthishasanyimplicationsforacontinuingallegiancetogovernment(T3.2.8.3;E-OC8,4701).Afteralifetimeofreflectingontheproblemand,presumably,afterdiscussingitwithhisfriendAdamSmith,Humealteredhisargu-mentinthelastessayhewrote,OftheOriginofGovernment.Inthisessayhesuggeststhatgovernmenthasitsoriginsinpeo-pleshabitofsubmissiontomilitaryleadersintimeofwar.Suchleadershipwouldnaturallyattractnonmartialfunctions,forexam-ple,theadministrationofjusticeandthecollectionofrevenue,andgraduallybecomecommonplacebetweenboutsofwarfare(E-OG6,3940).11v.opinionandthescienceofpoliticsIrrespectiveofthehistoricalaccountoftheoriginsofgovernment,Humealwaysmaintainedhispositionthatcontractandconsentarenot,andcannotbe,thebasisforcontinuedallegiancetogovernmen-talauthority.ThebasisforgovernmentisacombinationofthetwofactorsdiscussedintheTreatiseandnotedintheprecedingsection:apeoplesperceptionofthepublicinterestinprotection,especiallythroughenforcementoftherulesofjustice,andtheirperceptionoftheirobligationtoallegiance.IntheEssaysheprovocativelyformu-latesthisviewbysayingthatitisonopiniononlythatgovern-mentisfounded.Thisopinionisoftwokinds,towit,opinionof11SeeAdamSmith,LecturesonJurisprudence,ed.R.L.Meek,D.D.Raphael,andP.G.Stein(Oxford:ClarendonPress,1978),Reportof17623,5:11419,1278,1348;Reportof1766:1518,936.OnSmith,seeHaakonssen,TheScienceofaLegislator,12931,andDuncanForbes,Hume’sPhilosophicalPolitics(Cambridge:CambridgeUniversityPress,1975),76.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n356knudhaakonsseninterest,andopinionofright(E-FP12,323).Peoplearegener-allyborninto,andcontinuetolivein,societiesthatareundersomeformofgovernment.Theopinionsofthesesubjectsthattheirgov-ernmentcancareforthepublicinterest,andhastherighttoexerciseauthority,arethefoundationofthisgovernment.Consequentlythecentraltaskofthescienceofpoliticsistoaccountfortheformationandtransformationofthesefundamentalopinions.Someofthecausesofopinionaresouniversalthattheycanbeexplainedincompletelygeneralterms;theyareoperativeinpracticallyallcircumstancesofhumanlife.Thisappliestobeliefsconcerningtheinterestandobligationonwhichpregovernmentalinstitutionsresttherulesofjusticepertainingtopropertyandcontractandtothebeliefsunderlyinggovernmentitself.Afewadditionalfeaturesofpoliticsmaybeexplainedinsimilarlygeneralterms,butitsoonbecomesnecessarytodrawonmoreparticularfactors,factorsthataremorehistoricallyspecific.Althoughitispossibletodiscussingeneraltermstherelationshipbetweenlib-ertyandslaveryingovernment,onecannotintroduceconceptslikepartiesintoonesaccountwithoutdrawingontheexperienceofparticularformsofgovernment.Thisrequiresknowledgeofspe-cificeventsinindividualcountries.ConsequentlyHumesscienceofpoliticsrangesfromaconsiderationofwhatsomeofhiscon-temporarieswouldhavecalledthenaturalhistoryofthehumanspecies,thatis,fromhisexaminationofhumanunderstandingandtheprinciplesofmorals,throughhistoricallybasedgeneralmax-ims,tothecivichistoryofparticularculturesandstates.Thisentirerangeofmaterialisnecessarilypartofhisscienceofpoliticsbecause,evenintheexplanationofthemostspecificevent,therewillberef-erencestotheuniversalprinciplesofhumannatureunderlyingallmoralthought,andtotheinstitutionstowhichthoseprincipleshaveled.Onlyrarelywillourexplanationsdependontheidiosyncraticwhimsofindividuals.Andeveninthoserarecases,as,forexample,theextremesofenthusiasticmadness,deviationsfromprinciplecanonlybeunderstoodassuchbecauseweknowtheregularitythatisbeingbroken.Atthesametime,thefullrangeofexplanations,fromthemostgeneraltothemostspecific,ispartofascienceofpoliticsbecauseallexplanationsareconcernedwiththeformationofthoseopinionsthatsupporttheinstitutionsofsociety.ThemoreCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory357generalpartofpoliticsexplainsthatsuchinstitutionsarethekindofthingsthatmusthaveahistory,whilethemorespecificpartsrevealthehistorytheyhaveactuallyhad.Thegeneralprinciplesofpoli-ticsteachusthatpoliticalactionmuststartfromanunderstandingoftheparticularpoliticalconditionstowhichhistoryhasbroughtus.Humespoliticaltheoryis,inotherwords,anexplanationofwhypoliticaltheorizinginabstractionfromhistoricalconditionsisfutileandoftendangerous.Humewasacutelyconsciousofthefactthatthiswasinitselfapoliticalopinioncalculatedtoinformpoliticalconductataparticulartimeandplace.Indeed,thiswasundoubtedlypartofthereasonwhyhewenttosuchlengthstopopularizehisthe-orybymeansofhisEssaysandtheHistory.Theformationofsoundpoliticalopinionsisthemostbasicpoliticalactivity,andHumespoliticaltheorizingwassuchanactivity.ThereisoftenasenseofurgencyinHumespoliticalwritings,forhewasalwayskeenlyawarethatpeoplesopinionsarefickle.Undertheinfluenceofpassionsofavarice,offactionalordynasticorconfessionalallegiances,ofutopiandreamsofperfectionourunderstandingofoursituationandthatofoursocietytoooftenbecomesclouded,particularlyinsituationsofuncertaintyandinstability.Whenthereisuncertaintyaboutwhohasauthorityoraboutwhatthoseinauthoritymaydo,ourhabitualwaysofthinkingandbehavingarebroken.Undersuchcircumstancesopinionsandactionsaremuchmorelikelytobeinflu-encedbyimaginedsituationsthanbyactualconditions,andpas-sionateflightsoffancytendtotakeover.Sinceopinionsareformedbyexperience,wecanonlyhaveempiricallywell-foundedopinionsaboutwhoisdoingwhatinsocietyifthereisacertainregularityofbehavior.ThemessageofHumestheoryconcerningthebasicfeaturesofsocietyisthatsuchregularitycannotcomefromindi-vidualmindsandwillsalone;itdependsonsomethingoutsidetheindividual,namely,regularorrule-boundinstitutionsthatcanguideourbehaviorandconsequentlyourexpectationsofeachother.Ifsuchinstitutions,onceacquired,arelightlygivenup,welosehabitandregularity;welose,thatis,themostimportantmeansofori-entingourselvestoothers.Consequentlywecannotknowwhatweourselvesmaydowithsuccess,andwewillhavelostourmostele-mentaryfreedom.ThisistherationalefortheenormousemphasisHumeplacedoninstitutionalstability.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n358knudhaakonssenvi.thedistributionofjusticeStabilitycanbeseenfromtwoperspectives:thestabilityofwhatthoseinauthoritydoandthestabilityofwhomtheyare.ThesetwotopicsarefundamentaltoHumespoliticalthought.Theconductofgovernmentisonlystableandpredictableifitfollowspubliclyknowngeneralrulesonlyifitisgovernmentinaccordancewithlaw.Governmentmustthereforebeconcernedwithissuesthataresuitablesubjectsoflaw.Theseareprimarilyformsofbehaviorthatareinthepublicinterest,butnotnecessarilyintheinterestofeachindividualconcernedintheparticularinstance.Weare,therefore,tolookuponallthevastapparatusofourgovernment,ashavingultimatelynootherobjectorpurposebutthedistributionofjustice,or,inotherwords,thesupportofthetwelvejudges.Kingsandparliaments,fleetsandarmies,officersofthecourtandrevenue,ambassadors,ministers,andprivy-counsellors,areallsubordinateintheirendtothispartofadmin-istration.Eventheclergy,astheirdutyleadsthemtoinculcatemorality,mayjustlybethought,sofarasregardsthisworld,tohavenootherusefulobjectoftheirinstitution.(E-OG1,378,emphasisadded)AlthoughatfirstsightanexampleofthehyperboletowhichHumeoccasionallyresortsintheEssays,thispassagemakesclearwhatcarriesmostweight.Humehasnodoubtaboutthenecessityofagovernmentalagendaindefenseandforeignaffairsaswellasineconomicsandculture,buthegivesprioritytomaintainingthosetwobasicinstitutionsofjusticepropertyandcontractthatmakesociallifepossible.Insofarasthepopulacehasaclearopinionthatthisbalanceofprioritiesconstitutesthepublicinterestandthatthegovernmentprotectsthisinterestaswellasanypossiblegovernmentcould,tothatextentthegovernmenthasasecuresourceofallegiance(E-FP2,33).ItfollowsfromthisthatHumemustrejectpoliciesthatsignif-icantlybreaktherulesofjustice.Hedisparages,forexample,thesuggestionthatgovernmentsshouldtreatindividualcitizensaccord-ingtotheirnaturalmerit.Suchapolicywouldcreatethegreatestuncertainty.Meritissodependentoneachparticularsituationthatitisimpossibletoformulategeneralrulesforit;consequently,noorderlyallocationofgoodscouldbebasedonit.Thesamecriticismappliestoallotherschemesforthedistributionofgoodsorstatusonthebasisofallegedpersonalmeritsorvirtues.HumecriticizesinCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory359particulartheclaimsofthosereligiousfanaticswhosaythatdominionisfoundedongrace,andthatsaintsaloneinherittheearth,andpointsoutthatEnglandhadexperiencedsuchenthusi-asmfromthePuritansandfromoneoftheirpoliticalsubsects,theLevellers,whoclaimedthatthereoughttobeanequaldistributionofproperty(EPM3.235).Regardingthedistributionofproperty,Humeaddssomefurtherconsiderationsofimportance.Evenifweassumethatequalityofpropertycouldsomehowbeachieved,itsmaintenancewouldbeextremelypernicioustohumansociety.Renderpossessionseversoequal,mensdifferentdegreesofart,care,andindustrywillimme-diatelybreakthatequality.Inordertokeeppeopleequalintheirpossessions,thesevirtueswouldhavetobecontrolled.Todosowouldrequireamostrigorousinquisition,wouldimpoverishsoci-ety,andwouldbreakdownsocialsubordinationandorder(EMP3.246).TheseremarksmakeitclearthatHumesnotionofjusticeisnotpurelyformalandprocedural.TherulethateveryoneshouldhavethesamequantityofexternalpossessionsisasuniversalinformasHumesrulesconcerningtheallocationofproperty.Butherejectssucharulebecauseitwouldrequiretyrannicalinterferencewithindividualsnaturalqualitieswiththeirvirtuesandwiththeirpersonalfreedom.Theobjectofjustlawsisthusindividualliberty,and,sincethemostobviousandmostendangeredexpressionofsuchlibertyistheacquisitionanduseofproperty,justiceiscentrallycon-cernedwithproperty,and,itfollows,withcontracts.Thisorderofjustificationisnoteworthy,forintheTreatiseHumehassometimesseemedtolimittheobjectoftherulesofjusticetosecuringpropertyperse.Hetheresaysthatwehavethreespeciesofgoods,theinternalsatisfactionofourmind,theexternaladvan-tagesofourbody,andtheenjoymentofsuchpossessionsaswehaveacquirdbyourindustryandgoodfortune.Ofthese,thefirstcannotbetakenfromus,andthesecond,whiletransferable,canbeofnousetoothers.Thelastonlyarebothexposdtotheviolenceofothers,andmaybetransferrdwithoutsufferinganylossoralteration;whileatthesametime,thereisnotasufficientquantityofthemtosupplyeveryonesdesiresandnecessities(T3.2.2.7).Externalgoodsare,accordingly,thedirectobjectsofjustice.WhatthepassagesfromthesecondEnquirymakeclearisthatthroughtheprotectionofpropertythetwootherspeciesofgoodsarebeingindirectlyprotectedaswell(EPM3Part2).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n360knudhaakonssenvii.theroleofrightsHumescarcelyusedthetraditionalnotionsofrightsinhismoralandpoliticalphilosophy.WritersonthesesubjectscommonlyusedaschemebasedonmaterialsfromRomanlawanddevelopedbynat-urallawyersfromHugoGrotiusonwards.12Onthisscheme,certainfeatureswereinherentineachpersonquahumanbeing,whileotherswereacquiredandaddedtothepersonthroughhisorheractivityinlife.TheformerwerenaturalorinnaterightsandcorrespondroughlytoHumesnaturalvirtues;thelatterwereadventitiousoracquiredrightsandcorrespondroughlytoHumesartificialvirtues.Someofthenaturalrightswereimperfect,otherswereperfect,aswereallacquiredrights.Kindness,benevolence,gratitude,andthelikecouldbeclaimedasrightsonlyimperfectlybecausethequalitiesoftheclaimantthatwouldjustifytheclaimweretoouncertainandvariabletobethesubjectoflaw,andthemoralurgencyofclaimsforthemwastoolimitedtowarranttheuseoflegalforcetosecurethem.Buttheperfectnaturalrights,life,liberty,personaljudgmentorbodily,behaviorial,andmentalintegrityandtheiradventitiousorartificialextensionofthepersontopropertyandcontractualrela-tionsweresufficientlyascertainabletoberegulatedbylaw,andtheirprotectionbytheforceoflawwasdeemedsoimportantthatitpro-videdthemainjustificationfortheinstitutionofgovernment.ThedistinctionbetweenperfectandimperfectrightsinrespectoftheircertaintyandenforceabilitysoundsverymuchlikeHumesdistinc-tionbetweenartificialvirtuessuchasjusticeandthenaturalvirtuessuchasbeneficence.Yet,aswesawattheendoftheprevioussection,herecognizedthatinadditiontoproperty,certainnaturalqualitiesthegoodsofmindandbodyrequiretheprotectionoflaw,andthattheyreceivesuchprotectionwhenpropertyislegallysafeguarded.Thesenaturalqualitiesorgoodsaretheareasoflifethat,intheoriesofnaturallaw,areprotectedasperfectnaturalrights.Inotherwords,insubstance,Humewasinagreementwiththepopularnaturallaw12OnrightsinProtestantnaturaljurisprudence,seeHaakonssen,NaturalLawandMoralPhilosophy,1562,31041,andTheMoralConservatismofNatu-ralRights,NaturalLawandCivilSovereignty:MoralRightandStateAuthorityinEarlyModernPoliticalThought,ed.I.HunterandD.Saunders(Basingstoke:PalgraveMacmillan,2002),2742.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory361systemsofmorals,buthecouldnotusetheconceptofrightstoformulatehisargument.Whenhedoestalkofrights,itiscasuallyandinconnectionwithpropertyandcontract,oritisinthecontextofauthoritytherighttogovern.Humecouldnotusetheconceptofrightsbecausebothoftherightstraditionswereunpalatabletohim.Ononeview,rightswerequalitiesofthepersonasamoralagent;theyweretheprimaryfea-tureofallmorals,andallmoralinstitutions,suchasrulesofprop-ertyorstructuresofauthority,arosewhenindividualsadaptedtheirrespectiverightstoeachother.ThisviewhadbeenapproximatedbyGrotiusandhadreceivedadaringphilosophicalformulationbyHobbes,forwhomthequalities,orrights,inquestionwerenothingmoremysteriousthanthevariousclaimsofindividualsontheirsur-roundingworldandoneachother.InmanyrespectsthisviewwasclosetoHumeswayofthinking,butthereweretwogoodreasonswhyhecouldnotacceptit.First,thisformofrightsargumentleddirectlytothecontractarianwilltheoriesofsocialinstitutionsthatwehaveseenhimreject.Onlyifhehadfoundawayofseeingtheascriptionofrightstoindividualsaspartoftheprocessofsocial-izationcouldhehavereconciledrightswithhismoraltheory.13Asecondreasonforrejectingthisformofrightsargumentwasprobablythatitwastooreadilyassociatedwithpolitico-religiousenthusiasmandwaspoliticallydangerous.Religiouslybasedclaimstoafreedomofthespirittogovernoneselfwereonlytooeasilycouchedintermsofrights.Onasecondview,rights,farfrombeingtheprimarymoralfea-tureoftheperson,werederivativefromanaturallawthatascribeddutiesandrightstoindividuals.Thiswasbyfarthemostperva-siveviewofthephilosophicalstatusofrightsandwassharedbythinkerswhootherwisewereextremelydifferent,rangingfromtheneoscholasticGottfriedWilhelmLeibnizandChristianWolff,theCambridgePlatonistsandtheirassociateNathanielCulverwell,13ThiswasoneofthemostsignificantphilosophicalachievementsofAdamSmith,whoexplainedjusticeintermsofrights,rightsintermsofinjury,andinjuryintermsofthereactionsofspectatorswhowereconditionedbytheircircumstances.SeeSmith,LecturesonJurisprudence,Reportof17623,1.1,925;Reportof1766:511;Haakonssen,TheScienceofaLegislator,99104.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n362knudhaakonssenthequasi-CartesianRichardCumberland,totheeclecticvolun-taristsSamuelPufendorfandChristianThomasius.14Oneleadingcharacteristicofthesetheorieswasthatrights,andespeciallyper-fectrights,weredependentonduties;whenonepersonhasarighttosomething,othershaveadutytoabstainfromit.ThisissimilartoHumesreasoningaboutthemoralqualityoftherulesofjustice.ButifinthesecircumstancesHumehadinvokedaconceptofrights,hewouldhavebeeningreatdangerofbeingmisunderstood.Hehadtoavoidthetraditionalargumententirelybecause,aswehaveseen,thenaturallawinvolvedwascommonlyconsideredpartofateleologicalandprovidentialistschemeofjustification.Insum,thereareverygoodreasonsembeddedinHumestheoryofmoralsandpoliticsforrejectingallthecommontheoriesofrights.Butthisdidnotleadhimtorejecttheentirejurisprudentialapproachtopolitics.Hisbasicidealofstablegovernmentalactioniscouchedinthejuridicaltermsoftherulesofjustice,andtheserulescoverthecentralareasofprivatejurisprudenceinthesystemsofnaturallaw.Theycover,thatis,theprotectionofnaturalandadventitiousrights,especiallyrealandpersonalrightssuchasproperty,succes-sion,contract,anddelict.Thisidealofgovernment,ortheruleoflaw,was,intheBritishpoliticaldebate,associatedwithfreegovernments,whetherpurelyrepublicanlikethoseoftheItaliancitystatesandtheUnitedProvinces(Netherlands),ormixedliketheBritishgovernment.OneofHumesmostprovocativecontributionstothisdebatewashispartialdivorceofthequestionofthenatureandstabilityofgovernmentfromthatofthenatureandstabilityofgovernmen-talaction.Heshowed,first,thatabsolutemonarchieslikeFrancewereundercertaincircumstancesperfectlyabletoadopttheruleoflawandservethepublicinterest;and,second,thatfreegov-ernmentsliketheBritishoneharboredforcesthattendedtowardanarchy,andtherebytyrannyandtheunderminingofthepublicinterest.14Foranoverviewoftheseventeenth-centurythinkers,seeHaakonssen,NaturalLawandMoralPhilosophy,1562,andofthoseintheeighteenthcentury,GermanNaturalLaw,CambridgeHistoryofEighteenth-CenturyPoliticalThought,ed.M.GoldieandR.Wokler(Cambridge:CambridgeUniversityPress,2006),25190.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory363viii.therighttogovernHavingseenwhatHumemeantbystabilityofgovernmentalaction,weareleftwithasecondquestionaboutstability,thequestionofwhogoverns.Allgovernments,Humesays,arefoundedontwoopin-ions,opinionofrightandopinionofinterest.Wehavediscussedopin-ionofinterest,orwhatHumedescribesasthesenseofthegeneraladvantagewhichisreapedfromgovernment,intermsoftheregu-laradministrationofjusticeastheidealofwhatgoodgovernmentshoulddoandwhatcitizensshouldseekfromtheirgovernment.Opinionofrightisconcernedwithwhothepeoplethinkshouldrule,anditisdividedintotwokinds:righttopowerandrighttoproperty(E-FP3,33).Agovernmentgenerallyheldbythepeopletohavearighttopowerandtoservethepublicinterestwillbesta-ble,unlessitsconstitutionallowsforsomepopularinfluence,asinarepublicanormixedconstitution.Inthesecasespeoplesopinionoftherighttopropertynormallyincludestheideathatthereshouldbesomeproportionalitybetweenpropertyandpoliticalinfluence.Humeremarksthatanotedauthor[JamesHarrington]hasmadepropertythefoundationofallgovernment;andmostofourpolit-icalwritersseeminclinedtofollowhiminthatparticular(E-FP4,334).15ButHumerejectsHarringtonsclaimthatthebalanceofpoliticalpowerisdirectlydependentonthebalanceofproperty.Thereisacertaintendencyforpowertogravitatetowardtheproper-tied,butthisprocessisnormallyinfluencedbyseveralotherfactors,suchasreverenceforsettledconstitutionalformsthatis,itisinfluencedbytheopinionofrighttopower.OtherwisetheBritishgovernmentwouldhavebecomerepublican,giventheweightofthepropertiedgentryrepresentedintheHouseofCommons.Inconsti-tutionswherepropertycanhaveinfluence,thereisalwaysadangerthatthismayconflictwithbeliefsabouttherighttopower,andconsequentlythereisadangerofinstability.ThisistheframeworkforHumesanalysisoffactionalisminfreegovernmentingeneral15HumeisreferringtooneofthecentralthesesofJamesHarringtoninsuchworksasOceanaandThePrerogativeofPopularGovernment.SeeHarrington,ThePoliticalWorks,ed.J.G.A.Pocock(Cambridge:CambridgeUniversityPress,1977),1635,1812,231ff.,404ff.,458ff.HumediscussesHarringtoninE-BG1,478,andE-IPC45,50,51416,5223.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n364knudhaakonssenandinthatofBritaininparticular.Thedangerofinstabilityisnotgreatingovernmentsthatrestprimarilyontheopinionofrighttopower,suchasabsolutemonarchies,butmonarchiesmaybefraughtwithotherdangers.Humesanalysisoftheopinionofagovernmentsrighttopowerisinaccordancewithhisgeneralviewsregardingtheconnectionsbetweenhabitualbehavior,thecreationofexpectations,andthemakingofmoraljudgments.16Hesuggeststhatthefactorsthatformsuchopinionmaybedividedintofivecategories,longpos-session,presentpossession,conquest,succession,andpositivelaws(T3.2.10.415).Longpossessionofpoweristhestrongestandmostcommonsourceofauthority,aswasdramaticallydemonstratedinBritainbythecontinuinginfluenceoftheStuartslongaftertheyhadexhaustedmostothersourcesofauthority,includingthatofpresentpossession.Presentpossessionofpowerwillalwaysinfluencepeoplesopinionabouttowhomtheyoweallegiance,asisshownbytherepeatedchangesofsovereigntyinBritainduringtheseventeenthandearlyeighteenthcenturies.Conquestisaparticularlyforcefuldemonstra-tionofpresentpossessionandhasbeenusedefficientlythroughoutrecordedhistory.Intheeyesofsome,theaccessionofWilliamofOrangetotheBritishthronewasanexampleofconquest.Bysuc-cession,Humemeansasituationinwhichthesonsucceedstothefathersauthorityasifthisauthoritywerepropertyeventhoughsuchsuccessionhadnotbeenlongestablished.Finally,positivelawsthatregulatewhoshouldholdpowerwillalwayshavesomeimpactonapeoplesopinionofrightfulauthority,andthiswouldundoubtedlybethecaseinBritainfollowingtheActofSettlement(1701),whichsecuredtheHanoveriansuccessiontothethroneofEngland(after1707,alsothatofScotland).Alltheseprinciplesinfluencepeoplesopinionofrightfulauthority,andiftheyallconcur,thegovernmenthasthestrongesttitletosovereignty,andisjustlyregardedassacredandinviolable.Often,ofcourse,theprinciplesdonotpointinthesamedirection,andthereisnogeneralprinciplethatwilleffec-tivelysortthemout.Intheend,allpoliticsisentirelysubordinatetotheinterestsofpeaceandliberty(T3.2.10.15).Whateverthe16ThisprocessisoutlinedinconnectionwiththeobligationtojusticeinPartIIIofthisessay.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory365principlesonwhichagovernmentmaytrytorestitsauthority,ifittoogrosslyinvadestheseinterests,therationaleforgovernmenthasbeenremoved.Inthatsensethepeoplealwayshavearighttoresistance.Irrespectiveoftheprincipleorprinciplesonwhichagovernmentbasesitsclaimtosovereignty,theinvocationofhistorywillsoonplayarole.Inmonarchiestheimportanceofhistoryisreflectedintheweightlaidonthehereditaryprinciple.Electivemonarchiestendtobeunstablebutoftenmakeupforitthroughtheprincipleofsuc-cession.Inrepublicsandmixedgovernments,historicaljustificationissoughtintheancientoriginsoftheconstitutionalformsfollowed.Theseinvocationsofthepastforthepurposesoflegitimationareoftenmeremyths,ofcourse,andHumecertainlyrejectedaspurefictionthevariousWhigideasofanancientEnglishconstitution.17Heclearlytookitasoneofthehallmarksofmoderncivilizationthatsuchmythscouldbesubjecttocriticismwithoutendangeringthestabilityofgovernment.MuchofhishistoricalandpoliticalwritingwasmeanttoeducatemodernBritonsinthisregard.Bygivingacandidviewofthepast,Humehopedtoprovidearealisticunder-standingofhowthepassageoftimeinfluencesthepresent.Timeandcustomgiveauthoritytoallformsofgovernment,andallsuc-cessionsofprinces;andthatpower,whichatfirstwasfoundedonlyoninjusticeandviolence,becomesintimelegalandobligatory(T3.2.10.19).ThiswascruciallyimportantinBritain.EveniftheaccessionofWilliamofOrangecouldbeseenasusurpationin1688,thecourseofhistoryhadlentlegitimacytothewholeofthesucces-sionsetintrainthen.ItwasthelatterthatwasimportantfortheallegianceofBritishsubjectsinthemiddleoftheeighteenthcentury.Thetaskofnonpartisan,philosophicalhistoryintheserviceofthescienceofpoliticswastodisregardfactionsandfactionalmythsandexplaintheprocessbywhichthenationhadarrivedatitsparticularpresent:bythisprocesstheinterestsofpeaceandlibertyhadbeenshaped.Itwasnecessaryforthepoliticallyrelevantpartofthepop-ulationtoholdenlightenedopinionsaboutthegovernmentsrightsonthebasisofitspresentperformancewithregardtotheseinter-ests.OneofthemostremarkablefeaturesofmodernEuropewas,17Forfurtherdiscussionofthistopic,seeinthisvolumetheessayDavidHume:TheHistorian,PartIV.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n366knudhaakonssenHumesuggested,thatthisenlightenmentwastakingplacenotonlywithinthemixedconstitutionofBritain,butalsointhecontinentalmonarchies,oratleastinFrance,themostmodernofthese.Traditionallymonarchieshadwhenevernecessarycreatedsuit-ableopinionsofgovernmentalauthoritythroughthetyrannicalandarbitraryexerciseofforce.AmongBritonsthiswasstilltheentrenchedcaricatureofFrenchslavery,acaricaturethatHumethoughtitwasimportanttodislodge.Francewasinthevanguardofanentirelynewspeciesofmonarchy,thecivilizedmonarchy.18Thisadmittedlydidnothavethedynamismtogeneratethecentralele-mentsofcivilizationinthefirstplace;itimitatedfreesocietieslikeBritain.Onceadopted,however,civilizedmodesoflifewerefairlysecureinamonarchy,insomerespectsperhapsevenmoresecurethanunderamixedconstitution.Humesanalysisoftheprocessofcivilizationissubtleandrichandbeyondeasysummary.Thethreemainfociaretheexpressionofthehumanspiritinartsandsciences,theprotectionofthepersonbymeansoflaw,andtheacquisitionandexchangeofthegoodsoftheexternalworld(seeT3.2.2.7).Indealingwiththesethreefac-tors,Humeisshowingtherelationshipbetweenmerelylivingandlivingwell,tousetheAristoteliandistinction.Forasocietytoliveatall,itneeds,inadditiontoagovernmentstrongenoughtopro-tectitexternally,aminimalsystemofjusticeandthewherewithaltofeeditself.Inasocietywherethegovernment,forwhateverrea-son,isrestrainedfromdoingmuchmorethansecuringthesethings,aspiritofenterpriseandindividualismwilltendtopredominate.Therewillbeagrowthofknowledgearisingfromexperimentsinlivingandproducing,anditisonthisbasisthatacommercialsoci-etyliketheBritishemerges.Bylivingatall,afreesocietycomestolivewell.Intriguingly,monarchiescanalsobecomecivilizedbywantingtoliveequallywell:inrealizingthiswish,theymayadoptsomeofthebasicfeaturesofafreesociety.Monarchiesarecharac-terizedbyacrustofnobility,whosestatusisdependentonthegood18HumesanalysisofmodernmonarchyingeneralandofthatofFranceinparticularisscatteredthroughtheEssays.ThemostimportantpassagesonwhichthepresentdiscussionisbasedareE-LP2,1011;E-PR1011,214;E-CL;E-RP515and2540,11219and12433;E-OC45,4856.TheissueisdiscussedinForbes,Hume’sPhilosophicalPolitics,15260;andNicholasPhillipson,Hume(London:Weidenfeld&Nicolson,1989),6170.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory367willofthemonarchratherthantheirownenterprise,andwhoselifeisguidedbycodesofhonorandritualizedshow.Suchaclasswillfeedofftheartsandcraftsdevelopedinafreesocietyandthenwilloftenoutstripthatsocietyinthefinerarts,asexemplifiedbythesuperi-orityofFrenchliterature.Cocoonedastheyarewithinsuchaclassofculture,monarchsarelittleinclinedtotakemuchinterestinthelifeofsocietyatlarge,andnosocialgroupissufficientlypropertiedtomakeitnecessaryforthemtodoso.Aslongasthecivilorderismaintainedbytheenforcementoflaw,societycanbeleftalone,andthisfreedom,combinedwiththeneedforforeigngoods,eventuallyleadstothegrowthofcommerce.ThiswasthemodelofthemoderncivilizedmonarchyemerginginFrance,whichHumeadmiredandaboutwhichhetriedtoenlightenhiscountrymen.ItwaswithintheframeworkofthispoliticalanalysisofcivilizationthatHumedevelopedhispoliticaleconomy.However,thisperspectivemaybeturnedaround,andithasinfactbeenarguedthatHumesfunda-mentalconcernwaswithcommerceasanewformofsociability,andthathispoliticaltheorywasderivedfromthisconcern.19DespitehisadmirationforFrance,Humeneverforgotthatsuchasocietyenjoyedaregularadministrationofjusticeonlybydefault.Therewerenoconstitutionalguaranteesbecausetherewasnocon-stitutionalcounterweighttothecrown.Forallitsfreedomandciv-ilization,modernmonarchyhadnopoliticalliberty.Humethoughtsociallifewithpoliticallibertyhighlyprecarious,andinhismorepessimisticperiods,whenfacedwithlibertarianexcessessuchastheWilkesriotsinLondoninthe1760s,hethoughtacivilizedmonarchythesafestlong-termsolution.Whathefearedinafreeconstitutionwasitstendencytobreedfactionsandthetendencyoffactionalismtodegenerateintofanaticism,disorder,andanarchy,outofwhichwouldgrowtyranny.Inotherwords,theveryengineofcivilized19SeeIstvanHontsimposingJealousyofTrade:InternationalCompetitionandtheNationStateinHistoricalPerspective(Cambridge,MA:HarvardUniversityPress,2005),whichprovidesarichEuropean,notleastFrench,contextforHume;seeesp.537andChaps.3and4.AlsoJohnRobertsoninTheCasefortheEnlightenment:ScotlandandNaples1680–1760(Cambridge:CambridgeUniversityPress,2005)marginalizesHumespoliticaltheoryinfavorofpoliticaleconomyderivedfromwhatisseenasneo-EpicureanelementsinHumesmoralthought.Incomparison,seetheaccountofHumeseconomicviewsintheessayHumesPrinciplesofPoliticalEconomyinthisvolume.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n368knudhaakonssenliving,namely,freedomunderlaw,founditsmostrefinedprotectioninasystemofpoliticallibertythatinevitablyharboredforcesthatcouldbecomedestructiveofthatengine.Thiscouldhappenwhenanignorantpeoplesufferedfromhistoricaldelusions,stokedbycorruptpoliticians,suchastheneedtobejealousofFranceandfearfulofitsambitionsforuniversalmonarchy.Suchmadnesswasleadingthecountrytomeetanimagineddangermilitarilyatthecostofamountingpublicdebt,apolicythatinHumeseyeswasbeingtrans-formedintocommercialempirebuildingthrougharms,somethinghefearedwouldleadtodisaster.ThiswasthesituationinwhichcontemporaryBritainfounditself,andtheanatomyoffactionalismwasconsequentlyacentralconcerninHumesliteraryinterventioninpubliclife:hisEssaysandmuchoftheHistory.ThenewanddifficultpointHumehadtoimpressonhisread-erswasthatinafreeconstitutionpoliticaldifferencescouldnotbeabouttheconstitution;theyhadtobewithintheconstitution.Fac-tionalismasheknewitwasinconsistentwiththis:theinfluenceoffactionisdirectlycontrarytothatoflaws(E-PG2,55).Thegeneraldangerinfactionalismwasthatitwouldleadtofragmentationbypittinggroupinterestsagainsteachotherattheexpenseofthepublicinterest.Evenworse,ittendedtotransformtherecurringquestionofwhoshoulddischargetheofficesofgovernmentintoaquestionofthebalancebetweenthepowersoftheconstitutionitself.ThiswasparticularlydangerousinamixedconstitutionsuchastheBritish,wherethemainfactionsnaturallywouldformaroundtwodifferentprinciplesofgovernment,themonarchicalandtherepublican.TheextraordinarythingwasthatBritain,asHumesawit,wasintheprocessofbreakingawayfromthisdivision.Buthiscontemporariesdidnotappreciatethisand,bycontinuingtheoldfactionalrant,theyendangeredtheprecariousconstitutionalandpoliticalbalancethatwasemerging.Areadableanalysisoffactionalismwasneeded.ix.abovepartiesFactions,orpartiesHumeoftenusesthetwowordsinterchange-ablyfallintotwobroadcategories,personalandreal.Personalfactionsareheldtogetherbypersonalrelations,normallyextensivefamilyties.Althoughsuchrelationscanplayaroleinanyparty,theymosteasilydominatepoliticsasawholeinsmallrepublics,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory369suchasthoseofItaly.Realfactionsaretheonesthatcanhelpusunderstandlargerstates,andespeciallyBritain.Realfactionsmaybedividedintothosefrominterest,fromprinciple,andfromaffec-tion.Factionsbasedoninteresttypicallyarisewhentwodifferentsocialgroups,likethenobilityandthecommonpeople,have,orthinktheyhave,opposinginterests.Sinceinterestisinevitablyadrivingforceinallhumanendeavor,suchfactionsarethemostreasonable,andthemostexcusable.InEnglandithadoftenbeenthoughtthattherewasafundamentaloppositionbetweentheinter-estsofthelandedandthoseofthetradingpartofthenation,andthisbeliefwasanimportantaspectofthedivisionofthepoliti-calnationintoCourtandCountryfactions.Butthebeliefwassimplynotjustified.Ifpeoplearetoavoidsuchfalseoppositions,theymustbeenlightenedsothatthepursuitofinterest,whichisconstitutiveofhumanbehavior,isguidedbythebeliefthatthepublicinterestisalsothemostimportantprivateone(E-PG910,5960).Incontrast,politicalfactionsinspiredbyprinciple,especiallyabstractspeculativeprinciple,areknownonlytomoderntimes,andare,perhaps,themostextraordinaryandunaccountablephænomenon,thathasyetappearedinhumanaffairs(E-PG11,60).Thekeywordhereisspeculative.Ifdivisionbetweenfactionsisconcernedonlywithdifferencesofaspeculativeortheoreticalsort,thenthereisnoobjectivenecessityforanydivisioninpoliticalbehavior.Thatistosay,thereisnothingoutsidethemindsofthoseinvolvedoverwhichtodivide.Ifthefactionalprinciplesconcernedthingslikepowerorgoodsthatonlyoneorotherpartycouldhave,thentherewouldbeaprimafaciecasefordivision.Inmattersspec-ulative,however,eachmindcouldholditsown,wereitnotforanaturaltendencytoconvinceothermindstoconformtoonesownandthustoonesgroup.Thefactorthatgavethisnaturaltendencysuchswayinthemodernworldwas,inHumesopinion,theChris-tianreligion.Initsorigins,Christianity,incontrasttomostotherreligions,wasnotanestablishmentfaith.Itcouldsurviveonlybydevelopingastrongpriesthoodtoprotectthesectagainstsecularpower.Thepriesthoodthereforehadavestedinterestincontin-uingtogoverntheirflockinseparationfromthestateandfromothersects.Inordertodoso,theyhadtoinventspeculativeprin-ciplesaroundwhichtorallytheirfollowers,andinthisthepriestsCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n370knudhaakonssensoughtreinforcementfromspeculativephilosophy.Whentheuni-versalchurchbrokeup,theopposingforcesburstonmodernEuropeinthedisastrousreligiouswars.Suchdivisions,therefore,onthepartofthepeople,mayjustlybeesteemedfactionsofprinciple;but,onthepartofthepriests,whoaretheprimemovers,theyarereallyfactionsofinterest(E-PG13,62).Thedangerfromthepeopleisfactionsbasedonenthusiasm;fromthepriests,factionsprimedbysuperstition.Humefearedfactionalismbasedontheopposingprinciplesofsuperstitionandenthusiasmmostofallbecauseofitsrabble-rousingpotential.Couchingtheirrhetoricinwhateverwasthepoliticaljar-gonoftheday,leaderscouldtaketothestreetsandthemeeting-housesandappealtothelargesectionofthepopulationthatwasoutsidethereachofpropereducation.Theonlywaytodealwithsuchfactionalismwastoenlightenthepotentialleadersofthefac-tions.TothispurposeHumesupportedeverymovethatcouldsecuretheinclusionoftheclergyintheworldofletters.Clergymenoftasteandlearningwouldtendtoseeissuesofdoctrinaltheologyasmat-tersfordiscussionamongtheeducatedratherthanasreasonsforsocialdivisions,andtheywouldseetheirrolevis-a-vistheircongre-`gationsasamoralizingandcivilizingoneratherthanasasectarianone.TheBritishpoliticalsystem,however,alsobredleadersofsecu-larfactionswhobasedtheircausesonprinciple.InthewakeoftheconstitutionalsettlementaftertheRevolutionof1688,membersoftheoldToryandWhigfactionshadbeenweavingacomplicated(andshifting)patchwork,themaincomponentsofwhichwereagov-ernmentorcourtfactionconsistinglargelyofmodernWhigsandacountryoppositionconsistingofgroupsofToriesandold-fashionedWhigswhowereonlyrarelyabletoactcoherentlyasagroup.Humethoughtthatthisfactionalismshouldbedealtwithintwoways.Atoneleveltheprinciplesinvokedbythefactionsshouldbecrit-icized.Atanotherlevelthiscriticismshouldnottaketheusualformofpoliticalpolemicsbutratherthedetachedformofpoliteliterarydebate.Politicshadtobemadepoliteandsubjecttocivi-lizedmannersjustlikeartandliterature;ithadtobewrittenaboutaccordingtoliterarystandardsasinHumesEssaysandHistoryandnotintheformofpolemicordiatribe.ThesubstantialcriticismoffactionalprinciplesHumeapproachedinavarietyofways.IntheCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory371Treatise,thesecondEnquiry,andsomeoftheEssays,hetriedtoshowtheuntenabilityofthebasicphilosophicalprinciplesbehindthefactionsthatweconsideredatthebeginningofthisessay,thatis,theideasofnaturalhierarchyandauthority,ononeside,andwilltheoriesofcontract,ontheother.IntheHistoryandsomeoftheEssaysherejectedasspuriousthehistoricalauthorityinvokedforpartyprinciples.20AndinseveraloftheEssayshearguedthatthefactionswerepoliticallyblindtotherealitiesofthecontemporarysituationandthereforepotentiallydisastrous.Thiswasnotleastthecasewithfactionsarisingfromaffection,asdistinctfromthosefrominterestandfromprinciple.Factionsfromaffectionarefoundedonthedifferentattachmentsofmentowardsparticularfamiliesandpersons,whomtheydesiretoruleoverthem(E-PG15,63).Suchfactionswerepowerfulforcesinhistory,andHumeanalyzedatsomelengththeattachmentofEnglishmenandScotsmentotheStuarts,andofthenewbreedofBritonstotheHanoverians,anissuethatremainedattheforefrontofBritishpoliticsuntilthedefeatofthelastJacobiterisingin1745(E-PS;E-PGB1012;HE71.702,6:5304).Politicaloppositionbasedonsuchprincipleswasirrationalsinceitcouldseldomservetherealinterestsofthoseinvolved.Thevoiceofreasoncouldonlytrytoper-suadepeopleofthisand,moregenerally,trytoshowthatitmatteredlesswhogovernedthanhowtheygoverned.Thebestconstitutionwasoneofsuchstableproceduresthatevenpoorrulersmightgov-erninthepublicinterest.Atleastinhismoreoptimisticmomentsbeforethepessimismofthe1760sand1770ssetin,HumethoughtthattheBritishconstitutionwasapproaching,orcouldapproach,suchastableandpositiveform.TheproblemwasthatfactionalcantwasblindingBritonstothispossibility.x.thestabilityofgreatbritainTheloudestchargeagainsttheBritishpoliticalsystemasitoper-atedaftertheRevolutionof1688,andespeciallyagainstthelongregimeofSirRobertWalpole,wasthatofcorruption,bywhichwasmeantthemanipulationoftheHouseofCommonsbytheCrownanditsministers.Rejectingtheuseofsuchchargedlanguage,Hume20AsDavidWoottonshowsinhisessayDavidHume:TheHistorian,PartII.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n372knudhaakonssenpointedoutthatitwasinfactasystemofmutualdependenceandtheverythingthat,howeverprecariously,enabledpoliticallibertytobecombinedwithstabilityinamixedconstitution.TheCrownwaseconomicallyinfinitelyweakerthanthepropertyrepresentedintheCommonstakenasawhole,astateofaffairsconstitutionallyratifiedintheCrownsdependenceonParliamentforsupply.Inbal-ance,theCrownhadacquiredameasureofcontroloverpartsofthelowerhousetakenindividuallyandinthatwaysecuredthestabilityofthepolicypursuedatanygiventime.ThiswaspossiblebecauseoftherespectgiventotraditionalconstitutionalformsandbecausetheCrownwasagreatdealricherthananyindividualsubject.Throughdistributionofofficesingovernmentandchurch,pensionsandhon-ors,procurementofsecureparliamentaryseats,andthelike,thekingandhisministersenrolledmembersofParliamentinsupportofcourtpositionsondecisiveissues.Themotivesandindeedthechar-actersofthoseinvolvedmightnotstandtheclosestmoralscrutiny,butthatwasnottobeexpectedofpeopleinpowerinanysystem.Thepointwasthatthissystemconvertedprivateandnotsoprivatevicesintopublicinterest.Thesamecouldnot,inHumeseyes,besaidabouttheotherpartofthechargeofcorruption,theunderminingoftheeconomythroughpublicdebt.Thegovernmentincreasinglyfinanceditsbusi-ness,includingsuccessivelarge-scalewars,bymeansofpublicloansfromthecommunity,guaranteedbythepublictreasury.Tradinginthesebondsbecameamajorpartofthefinancemarket.Thestabil-ityofthiswholesystemwasassumedtodependontheabilityofthegovernmenttohonortheloans,andasthepublicdebtgrewandgrew,itseemedthattheonlybarriertonationalbankruptcywastrustinthefuture,whichmeanttrustinthestabilityofthegov-ernmenttosecureafuture.Likemanymoretraditionalthinkers,Humebelievedthatlandedpropertywasastabilizinginfluenceongovernment.Sincerealestatecouldnotberemovedfromthecoun-try,thelandedinterestwastheinterestofthecountry.But,inacommercialsocietywherelanditselfwasincreasinglyacommodityandsubjecttothefluctuationsoffinancialexchange,evenlanddidnotprovideaterrafirmaforagovernmentengagedinloanfinancingonalargescale.Thewholefinancialsystemappearedincreasinglytobeamentalconstructoftheplayersinvolved,akindofeconomicsuperstition,withdecreasingreferencetoanythingobjectiveandCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory373extra-mental.ItwaslefttoHumesonlypeerinsuchmatters,AdamSmith,toshowthatinthisregardfinancialsystemsoperatedonrulesnotmuchdifferentfromtherulesofjustice,asHumehadexpoundedthem.Thefactthatsuchsystemswereusefulandtheresultofchoicedistinguishedthemfromthosebasedonsuperstition.21xi.openingtheeyesofthepublic:twocenturiesofreadinghumeInthePrefacetothefirsteditionofhisEssays(1741),Humesetshisnewliteraryeffortintothecontextofthepoliteessaysofthetime,referringtoTheSpectatorofAddisonandSteeleandtoTheCrafts-maninwhichBolingbrokepublishedsomeofhismostimportantpoliticalwritings(W3:412).Itisunclear,becauseunexamined,howmuchimpactHumesinterventionhadonthepublicdebateingen-eralorhowfaritcompensatedfortheineffectivenessoftheTreatise.WhileHumehadclearlydamnedhimselfasaninfidelwhocouldnotbeappointedtoauniversitychairinphilosophy,hewasprobablyofnogreatsignificanceinthepublicdiscussionofmoralsandpoliticsuntilhepublishedhisPoliticalDiscourses(1752)and,especially,theStuartvolumeofhisHistoryofEngland(1754).Inanagewhenthepoliticalbattlesofthepresentwerehabituallyfoughtthroughthepast,HumejudgedcorrectlythathistoryEnglishhistorywasthemosteffectivemediumforhispoliticalviews.Yeteffec-tivenesswasboughtataprice.ThephilosophicallybasedattemptatimpartialityandbalancebetweenToryandWhigreadingsofthepastservedonlytoconcentratethepublicattentionontheparty-politicalissue,tothedetrimentofdiscussionoftheunderlyingphilosophicalideas.DespiteprivateprotestationsandpublicpursuitofevenfinerbalanceinsubsequentvolumesoftheHistory,theworkeffectivelymarkedhimoutasaToryapologistinhisprinciples.Theposthu-mouspublicationin17024ofClarendonsHistoryoftheRebellionandastringofsubsequenthistorieshadsoinuredtheBritishpublictoseeingtheirpastandthereforetheirpresentinTory-WhigtermsthatHumesprinciplesofimpartialityhadlittlechanceofbeingper-ceivedand,typically,itwasprimarilyonthebattlegroundofhistory21SmithalsorejectedHumestheoryoftrade-ledeconomicdevelopment.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n374knudhaakonssenthathehadtobecountered,forexample,inthemonumentalWhighistoryofCatharineMacaulay.22HumesHistorybecamethestandardworkofitskindforsixtyorseventyyears,untileventuallyovertakenbyT.B.Macaulaysgreatwork,whichwasexplicitlydesignedasaWhigreplacement.Thus,eveninitsdemise,HumesworkservedtomaintainthenarrowagendaofToryversusWhig.ThroughoutthenineteenthandthefirsthalfofthetwentiethcenturyHumewasrarelythoughtofasapoliticaltheoristatall,exceptoccasionallyontheissueofthesocialcontract.23Notuntilthe1960sand,especially,the1970sdidHumefigureasmuchmorethanaToryhistorianwhorejectedtheoriginalcontract.ItwasDuncanForbessdetailedscholarshipandJ.G.A.PococksgrandvisionofAnglo-AmericanpoliticalculturethateffectivelydrewattentiontoHumeasacomplexsocialandpoliticaltheorist.2422EdwardHyde,FirstEarlofClarendon,TheTrueHistoricalNarrativeoftheRebel-lionandCivilWarsinEngland,3vols.(London,17024);CatharineMacaulay,HistoryofEnglandfromtheAccessionofJamesItothatoftheBrunswickLine,8vols.(London,176383).ThehistoriespostClarendon,whichhelpedcreatetheclimateinwhichHumeseffortsatpolitical-philosophicalhistoryhavetobeunder-stood,includeLaurenceEchard,HistoryofEnglandfromJuliusCaesarto1689,3vols.(London,170718);WhiteKennett,CompleteHistoryofEngland(London,1706)andACompassionateEnquiryintotheCausesoftheCivilWar(London,1708);ViscountBolingbroke(HenrySt.John),RemarksontheHistoryofEng-land(London,[1743]);PauldeRapin-Thoyras,Histoired’Angleterre,10vols.(TheHague,17237;Englishtrans.,London172631);JohnOldmixon,CriticalHistoryofEngland,2vols.(London,17246)andHistoryofEnglandduringtheReignsoftheRoyalHouseofStuart(London,1730);DanielNeal,TheHistoryofthePuritans,4vols.(London,17328);JamesRalph,TheHistoryofEnglandduringtheReignsofKingWilliam,QueenAnne,andKingGeorgeI;WithanIntroductoryReviewoftheReignsoftheRoyalBrothersCharlesandJames(London,1744);WilliamGuthrie,HistoryofEnglandfromtheInvasionofJuliusCaesarto1688,4vols.(London,174451);ThomasCarte,AGeneralHistoryofEngland,4vols.(London,174755).23See,e.g.,ErnestBarkersSocialContract.EssaysbyLocke,Hume,andRousseau(Oxford:ClarendonPress,1947).24ThemostcentralworksareDuncanForbes,Hume’sPhilosophicalPolitics;Scep-ticalWhiggism,Commerce,andLiberty,inEssaysonAdamSmith,ed.A.S.SkinnerandT.Wilson(Oxford:ClarendonPress,1975),179201;andHumesScienceofPolitics,inDavidHume:BicentenaryPapers,ed.G.P.Morice(Edin-burgh:EdinburghUniversityPress,1977),3950.J.G.A.Pocock,TheMachiavel-lianMoment.FlorentinePoliticalThoughtandtheAtlanticRepublicanTradition(Princeton:PrincetonUniversityPress,1975)andHumeandtheAmericanRev-olution:TheDyingThoughtsofaNorthBriton,inMcGillHumeStudies,ed.D.F.Norton,N.Capaldi,andW.Robison(SanDiego:AustinHillPress,1979),CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory375ForbessetintrainthesearchforconnectionsbetweenHumespurelyphilosophicalprinciples,especiallyintheTreatise,andhisapplica-tionofthesetoformulateascepticalorscientificWhiggismthatnotonlycutacrosstheoldTory-Whigdividebutalso,andmuchmoreimportantlyforourunderstandingofHumesimme-diatesituation,sortedoutthedivisionbetweencourtandcountryfactionsthathadcometodominateBritishpoliticssince1688.ThequestionofthecoherencebetweenHumesphilosophicalendeavorintheTreatiseandAnEnquiryconcerningthePrinciplesofMoralsandhispoliticalprincipleshassubsequentlydominatedstudiesofhispoliticalthought.InaparticularlyusefulintroductiontoHumespoliticalphilosophy,DavidMillerarguedthatthereisagapbetweenHumesgeneralskepticismandhispoliticalconservatismandthatthisgapisfilledbywhathecallsHumesideology.ThisthesiswasrejectedbyJohnB.Stewart,whoarguedthatHumesviewofpracticalknowledgeledhimtotheprinciplesofliberalism.InabookthatpresentsthemostdetailedaccountofHumesideaoftheartificialityofsociopoliticalinstitutions,FrederickWhelanarguedforthecoherenceofHumespoliticalthought,whileDonaldLiv-ingstonsuggestedthatsuchcoherencearosefromHumesnarrativeapproach,notonlyinhishistoriesandessaysbutalsointhemorenarrowlyphilosophicalwritings.25ThepicturepaintedbyDuncanForbeswassimultaneouslyandindependentlygivendepthandnuancebyPococksevocativeimageofanAtlanticpoliticaldebatecentrallyconcernedwiththecivic-humanistvaluesofclassicalrepublicanismthathadbeenrevivedin32543,nowrepublishedinPocock,Virtue,Commerce,andHistory(Cambridge:CambridgeUniversityPress,1985),12541,aworkthatfurtherdeepensPococksinterpretation.SeealsotheIntroductiontoWealthandVirtue:TheShapingofPoliticalEconomyintheScottishEnlightenment,ed.I.HontandM.Ignatieff(Cambridge:CambridgeUniversityPress,1983).ThepointmadeinthetextisnotmeanttoimplythattherewerenovaluablecontributionstoscholarshiponHumespoliticspriortotheseworks,butonlythatnonewereparticularlyeffectiveincreating,inthescholarlydebate,abroaderviewofHumespoli-tics.25DavidMiller,PhilosophyandIdeologyinHume’sPoliticalPhilosophy(Oxford:ClarendonPress,1981);JohnB.Stewart,OpinionandReforminHume’sPoliticalPhilosophy(Princeton:PrincetonUniversityPress,1992);FrederickWhelan,OrderandArtificeinHume’sPoliticalPhilosophy(Princeton:PrincetonUniversityPress,1984);DonaldW.Livingston,Hume’sPhilosophyofCommonLife(Chicago:Uni-versityofChicagoPress,1984).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n376knudhaakonssentheItaliancity-statesoftheRenaissance.Theidealofapolitycon-sistingofcitizensofpropertysufficienttokeepthemindependentandarmedtoprotecttheirfreedomwas,itwassuggested,theback-groundagainstwhichwehadtounderstandthepost-1688debatesaboutthecorruptinginfluenceoftransientcommercialwealth,asopposedtopermanentrealestate;aboutcivicorpublicvirtue;aboutthevirtuesofacitizenmilitiaandthedangersofastandingarmy;aboutthebalanceoftheconstitutionbetweenexecutiveandciti-zenryand,consequently,aboutthedurationandindependenceofParliament.26Whensetagainstthisgeneralframework,Hume,itbecameclear,hadapoliticalagendaofhithertounsuspectedrich-ness.Partofthisagendawasthewritingofphilosophicalhistory,andthishasbeenextensivelyexploredinavastEuropeancontextinPococksmostrecentmajorwork.27ThenotionofanAtlanticMachiavellianmomentofneoclassicalrepublicanismstrengthenedtheattentiontoanotherpublicwhoseeyesHumemighthaveopened.Inthe1960sAmericanscholars,intheirattemptstounderstandthemeaningofAmerica,rediscov-eredrepublicanismanditsassociatedcivicethicsasanalterna-tivetoLockeanliberalism.28ThiswasreinforcedbythecontinuingEuropean,andespeciallyBritish,discussionofrepublicanism.Thegeneralpropositionwasthis.29Intheirsearchforprinciplesinthelightofwhichtheycouldunderstandtheirproblemsandjustifytheirsolutions,theNorthAmericancolonistswereparticularlyreceptivetotheneorepublicanandanticourtideasofthecountryoppositioninthemothercountry.Butamongtheproblemstheyfacedafterindependencewastheclassicaldogmathatarepublicanformof26ConcerningHumeandthemilitiaissue,seeJohnRobertson,TheScottishEnlight-enmentandtheMilitiaIssue(Edinburgh:JohnDonald,1985),esp.Chap.3.27Pocock,BarbarismandReligion,Vol.2:NarrativesofCivilGovernment(Cam-bridge:CambridgeUniversityPress,1999),Chaps.1115.28Theliteratureonthistopicisextensiveandbeyondbriefsummary.ThepioneeringstudieswereBernardBailyn,TheIdeologicalOriginsoftheAmericanRevolution(Cambridge,MA:BelknapPressofHarvardUniversityPress,1967);andGordonS.Wood,TheCreationoftheAmericanRepublic1776–1787(ChapelHill:UniversityofNorthCarolinaPress,1969).29Again,Pococksworkwasthemostimportantinclearingtheground;seeesp.TheMachiavellianMoment,Chap.15,andVirtue,Commerce,andHistory,1608,1838,26376.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory377governmentcouldexistonlyinasmallcountry.Thesolutiontothisproblem,asoutlinedbyJamesMadisoninFederalistNo.10,was,itwassuggested,directlyinspiredbyHumesspeculativeIdeaofaPerfectCommonwealth.30WhiletheexplosionofscholarlyinterestintheScottishEnlight-enmenthasbenefitedHumescholarshipgenerally,hispolitical(andmoral)thoughthasbeenparticularlywellservedbymuchnewworkonAdamSmith.ThebetterweunderstandSmithswork,thebetterwemayappreciatethesharpestreadingHumespoliticshasreceived.ItwasthepublicationofanewsetofstudentsnotesfromSmithslecturesonjurisprudence,aspartofthenewcollectededitionofSmithsworks,thatbegantoopenupthequestionofhowHumestheoryofjusticerelatedtotraditionaltheoriesofnaturallaw.31Smithstheorywas,itwassuggested,ineffectofferingaHumeanbasisforajurisprudentialsystemthathithertohadrestedonentirelydifferentphilosophicalfoundations.32Therelationshipbetweennat-urallawandHumespoliticshadalreadybeenputunderdebatebyDuncanForbesspioneeringsketches.33ThisandthenewworkonSmithinspiredfurtherworkonnaturallaw,anditsoonbecameclear30Thisproblemwasstudiedearlyinanarticlethathasbecomeaclassic,DouglasAdair,ThatPoliticsMayBeReducedtoaScience:DavidHume,JamesMadisonandtheTenthFederalist,HuntingtonLibraryQuarterly20(1957):34360;repub-lishedinAdair,FameandtheFoundingFathers:EssaysbyDouglasAdair,ed.T.Colbourn(Indianapolis:LibertyFund,1998),13251.Forawide-rangingdiscus-sionofHumeinthecontextofeighteenth-centuryideasofconstitutionalism,seeDavidWootton,Liberty,MetaphorandMechanism:ChecksandBalancesandtheOriginsofModernConstitutionalism,inLibertyandAmericanExperienceintheEighteenthCentury,ed.D.Womersley(Indianapolis:LibertyFund,2006),20974.31AdamSmith,LecturesonJurisprudence,ed.R.L.Meek,D.D.Raphael,andP.G.Stein(Oxford:ClarendonPress,1978),partofTheGlasgowEditionoftheWorksofAdamSmith,6vols.(Oxford:ClarendonPress,197683).32SeeHaakonssen,TheScienceofaLegislator;andDonaldWinch,AdamSmith’sPolitics:AnEssayinHistoriographicalRevision(Cambridge:CambridgeUniver-sityPress,1978),whichincludeimportantdiscussionsofHume.33Forbes,Hume’sPhilosophicalPolitics,Chaps.12,andNaturalLawandtheScot-tishEnlightenment,inTheOriginsandNatureoftheScottishEnlightenment,ed.R.H.CampbellandA.S.Skinner(Edinburgh:EdinburghUniversityPress,1982),186204.SeealsoStephenBuckle,NaturalLawandtheTheoryofProperty:GrotiustoHume(Oxford:ClarendonPress,1991);RussellHardin,DavidHume,MoralandPoliticalTheorist(Oxford:OxfordUniversityPress,2007).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n378knudhaakonssenthatcivichumanistrepublicanismlivedsidebysidewithasimilarlyancientandrevivednaturaljurisprudence.34IfHumesinterventionsinthepublicpoliticaldebatesaroundhimhadtobeunderstoodagainstthebackgroundofcivichumanism,hisunderlyingpoliticalphilosophyhad,amongotherthings,tobeappreciatedinitscriticalrelationshiptoProtestantnaturallaw.Foramajorbranchofthiscurrentofthought,derivinginparticularfromSamuelPufendorf,preparedthewayforHumesandSmithsdecisiverejectionofmeta-physicalformsofnaturallaw.35Atthesametime,naturaljurispru-denceasageneralgenredominatedtheScottishmoralphilosophycourses,atleastinGlasgowandEdinburgh,aswellastheteachingoflaw.36ItwasinmanywaysthesystematicframeworkforacommonsocialethicswhichallofHumesreaderswouldhaveacceptedasamatterofcourse.ThisdebateaboutthemajorconceptualworldstowhichHumespoliticshadtoberelatedwasbeingfurthercomplicatedbytheworkofIstvanHont,whoarguedthatHumefollowedbySmithwerethethinkerswhofullyunderstoodthemostfundamentalchangetooccurintheearlymodernworld.Thiswastheemergenceofanewformofsocialintercourse,namely,nationalandinternationalcom-merceasthebasisforpoliticalsociety.ThisunsocialsociabilitywhichHontalsosawbeingpreparedbyPufendorfmarginalizedalltraditionalpoliticaldiscourse,andHumesessaysinpoliticalecon-omywerethepioneeringeffortinthisregard.37MorerecentlythisprimacyofpoliticaleconomyinHumesthoughthasbeenforcefullyarguedbyJohnRobertsononadifferent,thoughlargelycompatiblebasis.RobertsonseesHumespoliticaleconomy(andthatofSmith)34TherelationshipbetweenrepublicanismandnaturallawisdiscussedbyJ.G.A.Pocock,CambridgeParadigmsandScotchPhilosophers:AStudyoftheRela-tionsbetweentheCivicHumanistandtheCivilJurisprudentialInterpretationofEighteenth-CenturySocialThought,WealthandVirtue,23552;andinHaakon-ssen,NaturalLawandMoralPhilosophy,esp.Chap.2.35SeeHaakonssen,ProtestantNaturalLawTheory:AGeneralInterpretation,inNewEssaysontheHistoryofAutonomy:ACollectionHonoringJ.B.Schneewind,ed.N.BrenderandL.Krasnoff(Cambridge:CambridgeUniversityPress,2004),92109.36SeetheIntroductiontoThomasReid,onPracticalEthics.LecturesandPapersonNaturalTheology,Self-Government,NaturalJurisprudenceandtheLawofNations,ed.K.Haakonssen(Edinburgh:EdinburghUniversityPress,2007).37Hont,JealousyofTrade,whichincludeshismajorpapersstretchingbacktothe1980s.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPoliticalTheory379astheimplicationandpracticalapplicationofamoralphilosophyderivedfromneo-Epicureanstock,especiallyfromPierreBayleandBernardMandeville.ThisispartofaclaimforaunitaryEnlighten-mentcharacterizedinthiswayandreachingitsepitomeinHume.38InterestintheconnectionsbetweenHumesscienceofpoliticsandAdamSmithsincompletescienceofalegislatormeantthatHumebecamepartoftheexplorationsofthefateofthelatter.Asaresult,wecannowseethatinthegenerationfollowinghisdeathHumewasnotreadexclusivelyasaToryhistorianbutalsoasathinkerofmixedpoliticalprinciplesandasapioneerempiricalhis-torianofpolitics.39suggestionsforfurtherreadingForfurtherreadingonthistopic,thefollowingareespeciallyrecommended.Forbes,Duncan.Hume’sPhilosophicalPolitics.Cambridge:CambridgeUni-versityPress,1975.Hardin,Russell.DavidHume,MoralandPoliticalTheorist.Oxford:OxfordUniversityPress,2007.Hont,Istvan.JealousyofTrade:InternationalCompetitionandtheNationStateinHistoricalPerspective.Cambridge,MA:HarvardUniversityPress,2005.Livingston,DonaldW.Hume’sPhilosophyofCommonLife.Chicago:Uni-versityofChicagoPress,1984.Miller,David.PhilosophyandIdeologyinHume’sPoliticalPhilosophy.Oxford:ClarendonPress,1981.Phillipson,Nicholas.Hume.London:Weidenfeld&Nicolson,1989.Pocock,J.G.A.BarbarismandReligion.Vol.2:NarrativesofCivilGovern-ment.Cambridge:CambridgeUniversityPresss,1999.Chaps.1115.HumeandtheAmericanRevolution:TheDyingThoughtsofaNorthBriton.InPocock,Virtue,Commerce,andHistory.Cambridge:Cam-bridgeUniversityPress,1985.12542.38Robertson,TheCasefortheEnlightenment,289324,36076.39Seeesp.StefanCollini,DonaldWinch,andJohnBurrow,ThatNobleScienceofPolitics:AStudyinNineteenth-CenturyIntellectualHistory(Cambridge:Cam-bridgeUniversityPress,1983),Chaps.12;BiancamariaFontana,RethinkingthePoliticsofCommercialSociety:The“EdinburghReview”1802–1832(Cambridge:CambridgeUniversityPress,1985),esp.Chap.3;andMarkSalberPhillips,Soci-etyandSentiment:GenresofHistoricalWritinginBritain,1740–1820(Princeton:PrincetonUniversityPress,2000).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n380knudhaakonssenRobertson,John.TheCasefortheEnlightenment:ScotlandandNaples1680–1760.Cambridge:CambridgeUniversityPress,2005.Stewart,JohnB.OpinionandReforminHume’sPoliticalPhilosophy.Princeton:PrincetonUniversityPress,1992.Whelan,Frederick.OrderandArtificeinHume’sPoliticalPhilosophy.Princeton:PrincetonUniversityPress,1984.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nandrews.skinner12HumesPrinciplesofPoliticalEconomyDavidHumesessayswerethecradleofeconomics,suggestedJohnHillBurton,inhisimportantbiographyofHume.1Althoughthismaybeabiographersexaggeration,therecanbenodoubtthatHumesworkprovidedanimportantcontributiontopoliticalecon-omyasadiscipline,togetherwithasignificantcritiqueofthemer-cantilesystemthatwaslaterattackedbyhisfriendAdamSmith.i.economics:thebackgroundMercantilismisdifficulttodefine.AsthehistorianP.J.Thomasputit:Mercantilismhasoftenbeendescribedasadefiniteanduni-fiedpolicyordoctrine,butthatithasneverbeen.Inrealityitwasashiftingcombinationoftendencieswhich,althoughdirectedtoacommonaimtheincreaseofnationalpowerseldompossessedaunifiedsystemofpolicy,orevenaharmonioussetofdoctrines.Itwasaverycomplicatedwebofwhichthethreadsmingledinex-tricably.2Intheseventeenthandeighteenthcenturiestheobjectofpolicywastheenhancementofthepowerofthenationstate,astrategythatwastobeattainedinanumberofways,atleastoneof1LifeandCorrespondenceofDavidHume,2vols.(Edinburgh,1846),1:354.2P.J.Thomas,MercantilismandtheEastIndiaTrade(London:P.S.King&Son,1926),3.TheclassicauthoritiesonthemercantilesystemincludeGustavvonSchmoller,TheMercantileSystemandItsHistoricalSignificance(London,1896);EdgarS.Furniss,ThePositionoftheLabourerinaSystemofNationalism(Boston:HoughtonMifflin,1920);JacobViner,EnglishTheoriesofForeignTradebeforeAdamSmith,JournalofPoliticalEconomy38(1930):249301,40457;P.W.Buck,ThePoliticsofMercantilism(NewYork:HenryHolt,1942);JosephA.Schumpeter,HistoryofEconomicAnalysis(London:Routledge,1954),33576;EliHeckscher,Mercantilism(London:GeorgeAllen&Unwin,1955).381CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n382andrews.skinnerwhichwaseconomic.Thepowerofthisstatewastobeenhancedbytheaccumulationoftreasurethroughtrade,themaximizationofemployment,andtheencouragementofpopulationgrowth.Hume,unlikeAdamSmith,madenoattempttotreatthemer-cantilesystemasasystem.Butdespitethefactthatheapproachedmercantilismasapolicy,hedidalsotreatitatthetheoreticallevel.Aswillbeseenbelow,Humeinhiseconomicessaysidentifies,crit-icizes,andsometimesusesanumberofthethreadsassociatedwithmercantilism.Mercantilistssupposed,forexample,thatforeigntradeismoreimportantthandomestic,apointofviewadmirablysummarizedbythetitleofThomasMunnsEngland’sTreasurebyForraignTrade.Or,TheBallanceofourForraignTradeistheRuleofourTreasure(1630).Hume,too,thoughtforeigntradeofgreatimportance,buthisreasoningissubstantiallydifferentfromthatofthemercantilists.Therateofinterest,mercantilistssupposed,isdependentonthesupplyofmoney.Humefocusedonthisissue,butpresentedabroaderaccountofthedeterminantsofinterestrates.Mercantilistssupposedthatstateregulationoftradewasbothneces-saryandbeneficial.ThispositionisneatlycaughtinWilliamPettysBritanniaLanguens(1680):NothingcansoeffectuallyandcertainlysecurethepeaceoftheNation,astheRegulatingofourTrade,sinceitwillsetallMensheadsandhandsatworkinallmannerofInnocentandProfitableImployments,andintroduceageneralsatisfactionandHarmony.3Althoughfundamentallycommittedtofreetrade,Humeacceptedamodifiedformofthismercantilistviewinsomecircumstances,hegranted,thestatewouldhavetoproducelegisla-tiontopromoteandregulateindustry.Mercantilistsheldthatlowwagesareanincentivetoindustrytheutilityofpovertyargument.4Humesawthatlowwagescouldinsomecircumstancesprovideanincentivetoanestablishedindustry,butalsosawthatrelativelyhighwageswouldproduceadditionalmarketsforgoods.3BritanniaLanguens,oraDiscourseofTrade:ShewingtheGroundsandReasonsoftheIncreaseandDecayofLand-Rents,NationalWealthandStrength,citedfromASelectCollectionofEarlyEnglishTractsonCommerce,ed.J.R.McCulloch(London,1856),501.McCullochsworkincludeseighttextsfirstpublishedfrom1621to1701.4SeeFurniss,PositionoftheLabourer,esp.Chap.4,andE.A.G.Johnson,Unem-ploymentandConsumption:TheMercantilistView,QuarterlyJournalofEco-nomics46(1932):698719.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy383Itwouldbewrong,however,tosuggestthatthemercantilistsamongHumespredecessorsprovideanadequateexpressionofthenatureofeconomictheoryatthetimeHumewascomposinghisessays.Bytheseventeenthcenturyspeculationoneconomicques-tionswasbeginningtofollowanewanddifferentmethodology,evenasmercantilistscontinuedtoarguethemeritsofcloselyregulatedtrade.5Astrikingexampleofthismethodologicalrevolutionispro-videdbyWilliamPetty:ThemethodItake...isnotveryusual;forinsteadofusingonlycompar-ativeandsuperlativewords,andintellectualarguments,Ihavetakenthecourse(asaspecimenofthepoliticalarithmeticIhavelongaimedat)toexpressmyselfintermsofnumber,weight,ormeasure;touseonlyargu-mentsofsense,andtoconsideronlysuchcauses,ashavevisiblefoundationsinnature;leavingthosethatdependuponthemutableminds,opinions,appetites,andpassionsofparticularmen,totheconsiderationofothers.6AnotherexampleofthenewmethodisprovidedbyGregoryKingsposthumouslypublishedPoliticalArithmetic.ThemanuscriptofthisworkmayhavehadaprofoundinfluenceonCharlesDavenantsclassicworksAnEssayontheEastIndiaTrade(1697)andDiscoursesuponthePublicRevenueandTrade(1698).Pioneeringworkonpop-ulationstatisticswasdonebyPettysfriendJohnGraunt,whosePoliticalObservationsontheBillsofMortalityappearedin1662.EdmundHalleyfurtheradvancedpopulationstudiesinhisDegreesofMortalityofMankind(1693).Anadditionalremarkableexample,butthistimeinaCartesianmode,isprovidedbySirDudleyNorthsDiscoursesuponTrade(1691).North,likeDavenant,wasanadvocateoffreetrade.InthePrefacetothisworkRogerNorthacknowledgedthedebtofSir5WriterswhoreviewthisdebateincludeSchumpeter,HistoryofEconomicAnal-ysis,andE.A.G.Johnson,PredecessorsofAdamSmith:TheGrowthofBritishEconomicThought(NewYork:PrenticeHall,1937);WilliamLetwin,TheOriginsofScientificEconomics:EnglishEconomicThought,1660–1776(London:Methuen,1963);andTerenceW.Hutchison,BeforeAdamSmith:TheEmergenceofPoliticalEconomy,1622–1776(Cambridge:BasilBlackwell,1988).Mydebtstotheseworks,andespeciallytothatofHutchison,willbeevidentinwhatfollowsinthissection.DouglasVickers,StudiesintheTheoryofMoney,1690–1776(London:PeterOwen,1959),remainstheclassicaccountofmonetarytheoryintheperiodandishighlyrecommended.6PoliticalArithmetic(London,1690),quotedfromHutchison,BeforeAdamSmith,37.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n384andrews.skinnerDudley,hisbrother,toDescartes,whoseexcellentDiscourseonMethodissomuchapprovedandacceptedinourAge.Ofhisbrotherswork,Northremarks:IfindTradehereTreatedatanotherrate,thanusuallyhathbeen;ImeanPhilosophically:fortheordinaryandvulgarconceits,beingmeerHuskandRubbish,arewaived;andhebeginsatthequick,fromPrinciplesindisputablytrue;andsopro-ceedingwithlikecare,comestoaJudgmentofthenicestDisputesandQuestionsconcerningTrade.Tothisheadded:Andhenceitis,thatKnowledgeingreatmeasureisbecomeMechanical;whichwordIneednotinterpretfarther,thanbynoting,itheremeans,builtuponclearandevidentTruths.7Itistemptingtoseesubsequentdevelopmentsineconomicsasinvolvingacombinationoftwodifferentmethodologicalapproaches.InhisPrinciplesofPoliticalOeconomy(1767),SirJamesSteuartexplicitlyadoptedthetechniquesofbothinductionanddeductionwhilemakingaformalapproachtotheconstructionoftheory.Buttherewereother,andearlier,examplesnotablyHenryMartinsConsiderationsupontheEast-IndiaTrade(1701).InhisprefatoryremarkstothisworkMartinnotesthathehasendeavourdafterthemannerofthePoliticalArithmetick,toexpresshimselfinTermsofNumber,Weight,andMeasure;andhehopes,heshallnotbethoughttospeakwithconfidence,ofanythingthatisnotascertainastheveryPrinciplesofGeometry.8TherewerestillfurtherdevelopmentsthatcouldhaveattractedHumesattention.Especiallyimportantweretheworksofthosewhoplacedemphasisontheeconomicconsequencesofthefundamentalprinciplesofhumannature,and,moreparticularly,ontheimpor-tantroleplayedbyself-interest.ThefirstandmostobviousofthesewriterswasBernardMandeville,authorofTheFableoftheBees(170523),whoseinfluenceHumeacknowledgesinhisATreatiseofHumanNature(TIntro.7).LessobviousisPierredeBoisguil-bert,whoseDetaildelaFrance(1695)placesself-interestatcenterstageandgivesprominencetoissuesthatHumewaslatertoempha-size,namely,theinterdependenceofeconomicphenomenaandthe7CitedfromMcCulloch,EarlyEnglishTracts,[51415].8OnSteuart,seeSirJamesSteuart,PrinciplesofPoliticalOeconomy,ed.A.Skinner(Edinburgh:EdinburghUniversityPress,1966),lxlxii.MartinsConsiderationsisreprintedinMcCulloch,EarlyEnglishTracts,541629.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy385operationofthecircularflowofmoney.IsaacGervaise,lessinterven-tionistthanBoisguilbert,attemptedtoputthecaseforthefreedomoftradeonscientificgrounds,andsuccessfullyarticulatedtheconceptofthebalanceoftrade,aconceptlateremphasisedbyHume.9JacobVanderlintscontribution,MoneyAnswersAllThings,oranEssaytomakeMoneyPlentiful...andincreaseourForeignandDomestickTrade(1734),advancedthesessimilartothosewewillfindHumetohaveadumbratedandfurtheranticipatedtheroleofself-regulatingmechanismsininternationaltrade.DugaldStewartwastocompareVanderlintwithHumeinpointofgoodsenseandliberality,whileMarxwastochargethatHumefollowedVanderlintsworkstepbystep.10WhileStewartidentifiedaparallelbetweenHumeandVander-lint,debtsaremoredifficulttoestablish.ButBurtonhasdrawnattentiontoHumesknowledgeofSebastianVauban,MarshallofFrance,whoseDixmeRoyale(1701)isremarkableforitsempiricalcontentandfortheanalysisoftaxationanotherimportantthemeinHumesessays.BurtonhasalsoremindedusofHumesdebttoMontesquieu.11GiventhatHumereadsomeofJosiahTuckersworkinmanuscriptandcorrespondedwithhim,12itislikelythathewasfamiliarwiththelattersEssayonTrade(1749),aworkthatalsoemphasizedtheneedtoplaceeconomicstudiesonascientificbasis.HumealsocitesJean-FrancoisMelonsEssaiPolitiquesurleCom-merce(1734),DutotsReflectionsPolitiquessurleCommerceetlesFinances(1738),andJosephParis-Duverneyscommentary(1740)onthelatterswork(E-Mo,8n,2878).HumesknowledgeoftheworkofFrancisHutchesonopensupintriguingpossibilitiesevenifitislikelythattheshapethatHutch-esongavetothestudyofeconomicshadagreaterinfluenceonhisstudentAdamSmiththanitdidonHume.13LikeHume(andlaterSmith),Hutchesontreatedquestionsofpoliticaleconomyasintegral9Hutchison,BeforeAdamSmith,1269.10TheNewPalgrave:ADictionaryofEconomics,s.v.Vanderlint,Jacob.11LifeandCorrespondence,1:92,3045.12SeeHL1:27072;2:180,182,205.13W.R.Scott,FrancisHutcheson(Cambridge,1900),23043;seealsoW.L.Taylor,FrancisHutchesonandDavidHumeasPredecessorsofAdamSmith(Durham,NC:DukeUniversityPress,1965),1228;andTheNewPalgrave:ADictionaryofEconomics,s.v.Hutcheson,Francis.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n386andrews.skinnerwithissuesinethicsandjurisprudence.14Wearethusremindedofanother,distinctive,approachtothestudyofpoliticaleconomy,onethattookitsorigins,inpart,fromtheworkofGrotiusandPufendorf,andthatwasfurtherstimulatedbyphilosophicalconsiderations.15ButthepathHumewastofollowwaslargelyhisown.ItdiffersfromthatofHutchesonandfromthatofJohnLocke,anothermajorphilosopherofanearlieragewhomadeasignificantcontributiontothedevelopmentofscientificeconomics.16ii.fundamentalprinciplesofhume’seconomicsPsychologyInhisvaluableintroductiontoDavidHume:WritingsonEco-nomics,17EugeneRotweinremindsusthatHumesdiscussionofeconomicissuesreliesheavilyonprincipleselaboratedinhisTrea-tiseofHumanNature.RotweinnotesespeciallyHumesconvic-tionthatallthescienceshavearelation,greaterorless,tohumannature;andthathoweverwideanyofthemmayseemtorunfromit,theystillreturnbackbyonepassageoranother.EvenMathemat-ics,NaturalPhilosophy,andNaturalReligionareinsomemeasuredependentonthescienceofman;sincetheylieunderthecognizanceofmen,andarejudgdofbytheirpowersandfaculties(TIntro.4).Humewasalsoconvincedthatthescienceofmanitselfmustbefoundedonexperienceandobservation.But,becausethissciencecannotmakeexperimentspurposely,heconcludedthatwemustthereforegleanupourexperimentsinthissciencefromacautiousobservationofhumanlife,andtakethemastheyappearinthecommoncourseoftheworld,bymensbehaviourincompany,inaffairs,andintheir14RichardF.TeichgraeberIII,FreeTradeandMoralPhilosophy:RethinkingtheSourcesofAdamSmith’sWealthofNations(Durham,NC:DukeUniversityPress,1986).15Schumpeter,HistoryofEconomicAnalysis,Chap.2.16See,e.g.,LockesConsequenceoftheLoweringofInterest(London,1691),andFurtherConsiderationsconcerningtheRaisingoftheValueofMoney(1695).17EugeneRotwein,DavidHume:WritingsonEconomics(Edinburgh:Nelson,1955).ProfessorRotweinhasprovidedthemostimportantcommentaryonHumeseconomicwritingsthusfarpublished.HiseditionincorporatescorrespondencebetweenHume;CharlesdeSecondat,BaronMontesquieu;JamesOswaldofDun-niker;HenryHome,LordKames;Anne-Robert-JacquesTurgot;andtheAbbe´Morellet,174076.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy387pleasures.Whereexperimentsofthiskindarejudiciouslycollectedandcompard,wemayhopetoestablishonthemascience,whichwillnotbeinferiorincertainty,andwillbemuchsuperiorinutilitytoanyotherofhumancomprehension.(TIntro.7,10)ThisapproachtohissubjectgivesHumegroundsformaintainingapointofviewthatwastoproveprofoundlyinfluentialintheeigh-teenthcentury,namely,thatbothhumannature,and,toalesserextent,humanbehavior,areuniformthat,ashewastoputitinAnEnquiryconcerningHumanUnderstanding,thereisagreatuni-formityamongtheactionsofmen,inallnationsandages,andthathumannatureremainsstillthesame,initsprinciplesandopera-tions(EHU8.7).AmongtheconstantprinciplesHumeidentifiedasessentialtohumanhappinessareaction,pleasure,andindolence.Inpractice,heplacedthemostemphasisonthefirstofthese.Thereis,hesaid,nocravingordemandofthehumanmindmoreconstantandinsa-tiablethanthatforexerciseandemployment;andthisdesireseemsthefoundationofmostofourpassionsandpursuits.Hemakesadirectapplicationofthisneedforactiontothesphereofeconomics:Intimeswhenindustryandtheartsflourish,menarekeptinper-petualoccupation,andenjoy,astheirreward,theoccupationitself,aswellasthosepleasureswhicharethefruitoftheirlabour(E-RA3,26970;E-In11,300;seealsoT2.3.10.6).Self-InterestandthePursuitofGainInadditiontoisolatingtheimportanceoftheloveofaction,Humealsocallsattentiontothedesireforgain:Avarice,orthedesireofgain,isanuniversalpassion,whichoperatesatalltimes,inallplaces,anduponallpersons(E-RP6,113).Thisdesireislinkedtovanityandpride:Wefoundavanityuponhouses,gardens,equipages,aswellasuponpersonalmeritandaccomplishments,butthemostcommonofthesesourcesofvanityisproperty(T2.1.9.1,2.1.10.1).Indeed,richesaretobeconsiderdasthepowerofacquiringthepropertyofwhatpleases,andtheveryessenceofrichesconsistsinthepowerofprocuringthepleasuresandconvenienciesoflife.Theveryessenceofthispowerconsistsintheprobabilityofitsexercise,andinitscausingustoanticipate,byatrueorfalsereasoning,therealexistenceofthepleasure(T2.1.10.3,10).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n388andrews.skinnerHumeusedthisargumenttothrowimportantlightonwhatAdamSmithwaslatertodescribeasmansdrivetobetterhiscondition,andinsodoinganticipatedSmithsclaimthatthisdrivehasgenerallyasocialreferenceinasmuchasitisrootedinthedesireforapprobation.AsHumeputit:Therearefewpersons,thataresatisfydwiththeirowncharacter,orgenius,orfortune,whoarenotdesirousofshowingthemselvestotheworld,andofacquiringtheloveandapprobationofmankind(T2.2.1.9).Thispositionheelaboratedbyarguingthatthesatisfactionwetakeintherichesofothers,andtheesteemwehaveforthepossessorsmaybeascribdtothreedifferentcauses.First,Totheobjectstheypossess;suchashouses,gardens,equipages;which,beingagreeableinthemselves,necessarilyproduceasentimentofpleasureineveryone,thateitherconsidersorsurveysthem.Secondly,Totheexpectationofadvantagefromtherichandpowerfulbyoursharingtheirpossessions.Thirdly,Tosympathy,whichmakesuspartakeofthesatisfactionofeveryone,thatapproachesus.(T2.2.5.2)Inhisexplanatorypractice,Humeplacedthemostemphasisonthethirdofthesecauses,sympathy,sayingthatthepleasureofastranger,forwhomwehavenofriendship,pleasesusonlybysympa-thy.Tothisprinciple,therefore,isowingthebeauty,whichwefindineverythingthatisuseful....Whereveranobjecthasatendencytoproducepleasureinthepossessor,orinotherwords,isthepropercauseofpleasure,itissuretopleasethespectator,byadelicatesympathywiththepossessor(T3.3.1.8).18Self-InterestandConstraintHumedrewattentiontoaproblemcreatedbytheactivepursuitofgain,namely,theproblemofmaintainingsocialorder.Ouravid-itytoacquiregoodsandpossessionsforourselvesandournearestfriends,hewrites,isinsatiable,perpetual,universal,anddirectly18AdamSmithwascriticalofHumescontentionthatthesameprincipleproduces,inmanyinstances,oursentimentsofmorals,aswellasthoseofbeauty(T3.3.1.9).SmithscriticismisfoundinTheTheoryofMoralSentiments,ed.D.D.RaphaelandA.L.Macfie(Oxford:OxfordUniversityPress,1976),Part4,17993;thisworkwasfirstpublishedin1759.ButSmithaccepted,andindeedelaboratedon,Humesuseofsympathytoexplaineconomicphenomena.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy389destructiveofsociety,andournaturalbenevolencetostrangersistooweakapassiontocounter-ballancetheloveofgain.Infact,wefindthatournaturalself-interestcanbeconstrainedonlybyitself,andredirectedtoconstructiveends:Thereisnopassion,therefore,capableofcontroulingtheinterestedaffection,buttheveryaffectionitself,byanalterationofitsdirection.Thisredirectionoralter-ationasHumecallsit,mustnecessarilytakeplaceupontheleastreflection;sincetisevident,thatthepassionismuchbettersatisfydbyitsrestraint,thanbyitsliberty,andthatbypreservingsociety,wemakemuchgreateradvancesintheacquiringpossessions,thanbyrunningintothesolitaryandforlorncondition,whichmustfollowuponviolenceandanuniversallicence(T3.2.2.1213).Wesee,then,theimportanceofsocietyandoftheconventionsofjusticeonwhichsocietyisfoundedasbasicpreconditionsofthesocialorderwefindbeneficialandprize.Justiceitselforiginates,Humewrites,fromtheselfishnessandconfin’dgenerosityofman,alongwiththescantyprovisionnaturehasmadeforhiswants(T3.2.2.18).Havingobservedthattisimpossibletoliveinsocietywithoutrestrainingthemselvesbycertainrules,humanshave(per-hapswithoutconsciousforethought)devisedtheneededrules.But,oncetherulesareinplace,ithappensthatwereceiveapleasurefromtheviewofsuchactionsastendtothepeaceofsociety,andanuneasinessfromsuchasarecontrarytoit.Thisfactandthepublicinstructionsofpoliticians,andtheprivateeducationofpar-ents,contributetothegivingusasenseofhonouranddutyinthestrictregulationofouractionswithregardtothepropertiesofoth-ers(T3.2.6.11).19Asthereferencetopoliticiansperhapsindicates,thefinalconditionforsocialorder,andonethatisessentialtotheconductofeconomicaffairs,issomesystemofgovernment.Giventhatweasindividualsare,inagreatmeasure,governdbyinterest,itisobviousthatweneedtoinstitutionalizesomeformofcontrolovertheseindividualinterests.Weneedtochangeourcircum-stancesandsituation,andrendertheobservanceofthelawsofjusticeournearestinterest,andtheirviolationourmostremote.Thiswehavedonebyestablishingarelativelyfewpersonsasmagistratesor19Forfurtherdiscussionofthesepoints,seeinthisvolumetheessaysTheFoun-dationsofMoralityinHumesTreatiseandTheStructureofHumesPoliticalTheory.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n390andrews.skinnergovernorsandrulers.Theseareindividualswhoaremeanttohavenointerest,orbutaremoteone,inanyactofinjustice;andbeingsatisfiedwiththeirpresentcondition,andwiththeirpartinsociety,haveanimmediateinterestineveryexecutionofjustice,whichissonecessarytotheupholdingofsociety....Thesepersons,then,arenotonlyinducdtoobservethoserulesintheirowncon-duct,butalsotoconstrainotherstoalikeregularity,andenforcethedictatesofequitythrothewholesociety(T3.2.7.1,6).TheUseofHistoryThestudentofHumeswritingsoneconomicsshouldbeawarethatthegreatbulkofhispublishedworkwashistorical,andofhiscon-victionthathistoryisnotonlyavaluablepartofknowledge,butopensthedoortomanyotherparts,andaffordsmaterialstomostofthesciences.Itisaninventionthatextendsourexperiencetoallpastages,andtothemostdistantnations(E-SH6,566).20Lookedatinthisway,historicalstudiesaffordinvaluableinforma-tionwithregardtotheprinciplesofhumannatureandtothefactthattheexpressionoftheseprincipleswouldbeprofoundlyaffectedbythesocioeconomicenvironmentthatmayhappentoexistandbychangesinhabits,customs,andmanners.TotheeconomistthemostinterestingpartsofHumesHistoryofEnglandmayinitiallybetheappendicesandreportsofmiscellaneoustransactionsintroducedthroughoutthework.TheseansweredtoHumesdesiretotakeageneralsurveyoftheage,sofarasregardsmanners,finances,arms,commerce,artsandsciences.Thechiefuseofhistoryis,thatitaffordsmaterialsfordisquisitionsofthisnature;anditseemsthedutyofanhistoriantopointouttheproperinferencesandconclusions(HE62.65,6:140).HumeappliesthisproceduretohisaccountoftheentirerangeofEnglishhistory,fromthetimeofJuliusCaesartotheRevolutionof1688.21Quiteapartfromtheintrinsicvalueofthematerial,Humes20HumewithdrewtheessayOftheStudyofHistoryin1760,butitdoesnotfollowthathethoughttheessaymistakeninallrespects.21ForthetitlesoftheseveralvolumesofTheHistoryofEngland,andinformationabouttheorderinwhichthevolumesappeared,seePartIoftheBibliographytothisvolume.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy391accountsareinformedbyanattempttounderstandspecificpoli-ciesintheirinstitutional,economic,andpoliticalsettings,withtheimportanceofeconomiclibertyemphasizedthroughout.Thus,forexample,inhisdiscussionoftheecclesiasticalaffairsunderHenryVIII,HumeremarksthatMostoftheartsandprofessionsinastateareofsuchanature,that,whiletheypromotetheinterestofthesociety,theyarealsousefuloragreeabletosomeindividuals;andinthatcase,theconstantruleofthemagistrate,except,perhaps,onthefirstintroductionofanyart,is,toleavetheprofessiontoitself,andtrustitsencouragementtothosewhoreapthebenefitofit.22Unlesstheprocessisdisturbedbyanyinjudicioustampering,thecommodityisalwayssuretobeatalltimesnearlyproportionedtothedemand(HE29.2,3:135).Thesameemphasisisapparentinthetreatmentoftheregulationofinterest,wages,theprohibitionsontheexportofspecie,and,aboveall,inthecontextofinternationaltrade:Itisevident,thatthesemattersoughtalwaystobeleftfree,andbeentrustedtothecommoncourseofbusinessandcommerce(HE26.40,3:78).AtthesametimeithastoberecognizedthattheHistorywaspri-marilyconcernedwiththebroaderthemeofthestudyofcivilizationandwiththeinterconnectionsbetweenthegrowthofcommerce,thechangingformsofgovernment,andliberty.Inshort,Humescon-cernwaswiththeoriginsandnatureofthepresentestablishmentsinEurope,wheretheeconomicdimensionwasonlyonepartofawiderwhole.HumesperceptionoftheinterplaybetweeneconomicgrowthandlibertymovedAdamSmithtoremark:Commerceandmanufacturesgraduallyintroducedorderandgoodgovern-ment,andwiththem,thelibertyandsecurityofindividuals,amongtheinhabitantsofthecountry,whohadbeforelivedalmostinacontinualstateofwarwiththeirneighbours,andofserviledependencyupontheirsuperi-ors.This,thoughithasbeentheleastobserved,isbyfarthemostimportantofalltheireffects.Mr.Humeistheonlywriterwho,sofarasIknow,hashithertotakennoticeofit.2322ThisisapreambletoHumesargumentinfavorofanestablishedchurch.Suchaninstitution,heargues,curbstheperniciousandinteresteddiligenceoftheclergy.23AnInquiryintotheCausesoftheWealthofNations,ed.R.H.Campbell,A.S.Skinner,andW.B.Todd(Oxford:ClarendonPress,1976),3.4.4(1:412).Thisworkwasfirstpublishedin1776.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n392andrews.skinnerTherelevanceofthesepositionsforthecontemporaryunderstandingofHumestreatmentofeconomictheoryandpolicywillbereadilyapparent.24iii.theeconomicessaysItisusualtoidentifynineofHumesessays,eightofwhichwerefirstpublishedin1752,theninthin1758,astheeconomicessays.25Theseareessays,ratherthanatreatiseoraworkthataddressessep-aratesubjectsfromonepointofview.YetHumebelievedeconomicquestionstobeamenabletoscientifictreatmentlargelyasaresultofhisbeliefintheconstantprinciplesofhumannatureandtheempha-sisthathegavetoself-interest.Inafamouspassageheassertedthatitiscertain,thatgeneralprinciples,ifjustandsound,mustalwaysprevailinthegeneralcourseofthings,thoughtheymayfailinpar-ticularcases;anditisthechiefbusinessofphilosopherstoregardthegeneralcourseofthings(E-Co2,254).Humealsonotedthatthereareareasofexperienceaboutwhichgeneralizationisdifficult:Whatdependsuponafewpersonsis,inagreatmeasure,tobeascribedtochance,orsecretandunknowncauses:Whatarisesfromagreatnumber,mayoftenbeaccountedforbydeterminateandknowncauses.Fromthisprinciplehecon-cludesthatthedomesticandthegradualrevolutionsofastatemustbeamorepropersubjectofreasoningandobservationthanthefor-eignandviolent.Healsoconcludesthatitiseasiertoaccountfortheriseandprogressofcommerceinanykingdom,thanforthatoflearning;andastate,whichshouldapplyitselftotheencouragementoftheone,wouldbemoreassuredofsuccess,thanonewhichshouldcultivatetheother(E-RP2,5,6,11213).Itshouldalsobeobservedthattheseparateessaysshowaunityofpurpose.26Allofthemillustratethefundamentalpropositionsin24DuncanForbessuggeststhattheessayOfRefinementintheArtsisanabridgedversionofthoseaspectsoftheHistorybrieflytouchedonhere.Hume,TheHistoryofGreatBritain,ed.DuncanForbes(Harmondsworth:Penguin,1970),297.25ThenineessaysareOfCommerce,OfRefinementintheArts,OfMoney,OfInterest,OftheBalanceofTrade,OftheJealousyofTrade,OfTaxes,OfPublicCredit,andOfthePopulousnessofAncientNations.Thosepub-lishedin1752wereapartoftheimmediatelysuccessfulPoliticalDiscourses.OftheJealousyofTradewaspublishedin1758.26ThispointhasbeenmadebyRotwein,Hume:WritingsonEconomics,cv.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy393SectionII.Itisthisunityofpurposeandmethodthatenablesustoidentifythreemajorthemes:historicaldynamics,ortheprocessofhistoricalchange;theuseofthehistoricalmethod;andHumesuseofbothoftheseperspectivesinthetreatmentofinternationaltrade.HistoricalDynamicsandtheExchangeEconomyAssoonasmenquittheirsavagestate,wheretheylivechieflybyhuntingandfishing,HumesuggestsinOfCommerce,theybecomefarmersormanufacturers,thoughtheartsofagricultureemployatfirstthemostnumerouspartofthesociety.Inanearlyanticipationofthetheorythathumanityhaspassed,bystages,fromhuntingandgatheringtothecommercialsocietyofeighteenth-centuryEurope,Humenotedthatwherethereislittlestimulustochange,peoplemustapplythemselvestoagriculture.Becausetheycannotexchangeanysurplusforothercommodities,humansinthissituationhavenotemptationtoencreasetheirskillandindustry.Asaresult,thegreaterpartofthelandliesuncultivated.Whatiscultivated,yieldsnotitsutmostforwantofskillandassiduityinthefarmers(E-Co5,10,256,2601).Incontrast,Humecontinued,Whenanationaboundsinman-ufacturesandmechanicarts,theproprietorsofland,aswellasthefarmers,studyagricultureasascience,andredoubletheirindustryandattention.Thesuperfluity,whicharisesfromtheirlabour,isnotlost;butisexchangedwithmanufacturesforthosecommodi-ties,whichmensluxurynowmakesthemcovet(E-Co11,261).Inshort,Humesuggeststhatthereislikelytobeagradualprogres-siontowardtheinterdependenceofthetwomainsectorsofactivity.Playinganimportantsupportingrolearethemerchants,oneofthemostusefulracesofmen,whoserveasagentsbetweenthosepartsofthestate,thatarewhollyunacquainted,andareignorantofeachothersnecessities(E-In10,300).Humesargumenthasitrootsinhisdeploymentofafavoritethesisoftheeighteenthcentury,namely,thatmenhavenaturalwantsthatgraduallyextendinaself-sustainingspiral:Everythingintheworldispurchasedbylabour;andourpassionsaretheonlycausesoflabour(E-Co11,261).ItwasthisthesisthatMandevilleaddressedwithsuchamusingconsequencesinTheFableoftheBees,andthatdrewfromHumethecommentthattoimagine,thattheCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n394andrews.skinnergratifyingofanysense,ortheindulgingofanydelicacyinmeat,drink,orapparel,isofitselfavice,canneverenterintoahead,thatisnotdisorderedbythefrenziesofenthusiasm(E-RA1,268).Butthereismoretothethesisthanaconcentrationonagrad-ualinstitutionalchange;itisalsoapartofHumesargumentthattheemergenceofwhatcametobeknownasthestageofcommercewouldinduceanacceleratingrateofchangeduetochangesinhabitsandmannersnotablybyencouragingthedesireforgainandbygiv-ingprogressivelyincreasingscopetomansactivedisposition.27Thehistoricalprocessofeconomicdevelopmenthadbeenstimulatedbythediscoveryofgold,forexample,andthishadbeenafactor,Humethought,intherapidrateofeconomicgrowthduringthereignofCharlesIandduringtheperiodfromtheRestorationtotheRevolu-tionof1688(E-Mo7,286;HE62.83,6:148;71.81,6:537).Humewentontoobservethatindustryandartsandtradeencreasethepowerofthesovereignaswellasthehappinessofthesubjectsinsofarastheystoreupsomuchlabour,andthatofakindtowhichthepublicmaylayclaimthroughtaxation.Hereagainthemoderncommercialstatehasanadvantage,forwhentherichesaredispersedamongmultitudes,theburthenfeelslightoneveryshoul-der(E-Co9,11,17;260,262,265).Hewarnedagainstsucharbitraryimpositionsasapolltax,expressingastrongpreferencefortaxesonconsumptioninordertominimizedisincentives:Thebesttaxesaresuchasarelevieduponconsumptions,especiallythoseofluxury;becausesuchtaxesareleastfeltbythepeople.Theyseem,insomemeasure,voluntary;sinceamanmaychusehowfarhewillusethecommoditywhichistaxed:Theyarepaidgraduallyandinsensibly:Theynaturallyproducesobrietyandfrugality,ifjudiciouslyimposed:Andbeingconfoundedwiththenaturalpriceofthecommodity,theyarescarcelyper-ceivedbytheconsumers.Theironlydisadvantageis,thattheyareexpensiveinthelevying.(E-Ta5,345)Indeed,Humeevencontendedthattaxationcouldproveanencour-agementtoindustry:Wheretaxesaremoderate,arelaidongradu-ally,andaffectnotthenecessariesoflife,hewrites,theyoftenservetoexcitetheindustryofapeople,andrenderthemmore27Rotwein,Hume:WritingsonEconomics,19,suggeststhatthehistoricalvariabilityofmoralstandardsisathemeofHumesADialogue,aworkroutinelypublishedattheendofAnEnquiryconcerningthePrinciplesofMorals.ForSmithsviews,seeTheTheoryofMoralSentiments,5.2(194211).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy395opulentandlaborious,thanothers,whoenjoythegreatestadvan-tages(E-Ta2,343).28Themodernstatehasafurtheradvantageinthatitcanborrowpresentresourcesthroughthesaleofsecuritiesthatarewithusbecomeakindofmoney,adevelopmentthatencouragesasetofmen,whoarehalfmerchants,halfstock-holdersandwhoarewillingtotradeinsecuritiesforsmallprofits.Consequently,moreindividuals,thosewithlargestocksandincomes,maynaturallybesupposedtocontinueintrade,wheretherearepublicdebts;andthis,itmustbeowned,isofsomeadvantagetocommerce,bydiminishingitsprofits,promotingcirculation,andencouragingindustry(E-PC89,3534).29TheessaysarealsoremarkablefortheemphasisthatHumegavetotheother,noneconomicadvantagesthataccruefromtheprocessofhistoricaldevelopment:Themindsofmen,beingoncerousedfromtheirlethargy,andputintoafermentation,turnthemselvesonallsides,andcarryimprovementsintoeveryartandscience.Profoundignoranceistotallybanished,andmenenjoytheprivilegeofrationalcreatures,tothinkaswellastoact,tocultivatethepleasuresofthemindaswellasthoseofthebody.Henotedthatthemoretheserefinedartsadvance,themoresociablemenbecome;theyflockintocities;lovetoreceiveandcommunicateknowledge(E-RA45,271).HealsoemphasizedsociologicalandpoliticaldevelopmentsinanotablepassageintheHistory.DuringthereignofHenryVII,hewrites,Thecommonpeople,nolongermaintainedinviciousidlenessbytheirsuperiors,wereobligedtolearnsomecallingorindustry,andbecameusefulbothtothemselvesandtoothers.Anditmustbeacknowledged,inspiteofthosewhodeclaimsoviolentlyagainstrefinementinthearts,orwhat28SeeM.Arkin,TheEconomicWritingsofDavidHumeAReassessment,SouthAfricanJournalofEconomics24(1956):20420.HumecorrespondedwithTurgotonthesubjectoftaxation;seeespeciallyTurgottoHume,7September1766and25March1767(Rotwein,2067;21013),andHumetoTurgot,September1766(HL2:934).29Humewasforsomeyearsconvincedthatgovernmentborrowingservedtoreduceinterestrates.From1752to1764OfPublicCreditincludedanotereading:Onthishead,Ishallobserve,withoutinterruptingthethreadoftheargument,thatthemultiplicityofourpublicdebtsservesrathertosinktheinterest,andthatthemorethegovernmentborrows,thecheapermaytheyexpecttoborrow;contrarytofirstappearance,andcontrarytocommonopinion.Theprofitsoftradehaveaninfluenceoninterest(E-PCVariants,637).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n396andrews.skinnertheyarepleasedtocallluxury,that...anindustrioustradesmanisbothabettermanandabettercitizenthanoneofthoseidleretainers,whoformerlydependedonthegreatfamilies;somuchisthelifeofamodernnoblemanmorelaudablethanthatofanancientbaron.(HE26.37,3:767)ThisthemeiselaboratedinOfRefinementintheArts.Whereluxurynourishescommerceandindustry,Humewrites,thepeas-ants,byapropercultivationoftheland,becomerichandindepen-dent;whilethetradesmenandmerchantsacquireashareoftheprop-erty,anddrawauthorityandconsiderationtothatmiddlingrankofmen,whoarethebestandfirmestbasisofpublicliberty.Thisdevelopmentbroughtaboutmajorconstitutionalchanges,atleastinEngland,wherethelowerhouseisthesupportofourpopulargovernment;andalltheworldacknowledges,thatitoweditschiefinfluenceandconsiderationtotheencreaseofcommerce,whichthrewsuchabalanceofpropertyintothehandsofthecommons.Howinconsistentthenisittoblamesoviolentlyarefinementinthearts,andtorepresentitasthebaneoflibertyandpublicspirit!(E-RA1617,2778).Thisdynamicenvironment,buttressedbyequallaws,furtherenhancesthepossibilitiesforeconomicgrowth.30ButitisimportanttonoteherethatHumealsoofferedasharpcritiqueofegalitarianism:howeverattractiveandcompellingtheideasofperfectequalitymayseem,theyarereally,atbottom,impracticable;andweretheynotso,wouldbeextremelypernicioustohumansociety.Renderpossessionseversoequal,mensdifferentdegreesofart,care,andindustrywillimmediatelybreakthatequality(EPM3.26).TheHistoricalorInstitutionalMethodHumesinterestinthehistoricalprocessledhimquitenaturallytodevelopadistinctivetechniquefordealingwithpurelyeconomicquestions;inturn,thistechniqueledhimtogiveprominencetoinstitutionalbackgroundand,inparticular,totheroleofcustoms30InOfCommerce,Humesaysthat,Everyperson,ifpossible,oughttoenjoythefruitsofhislabour,infullpossessionofallthenecessaries,andmanyoftheconveniencesoflife.HewentontonotethatinthiscircumstanceconsiststhegreatadvantageofEnglandaboveanynationatpresentintheworldorthatappearsintherecordsofanystory(E-Co17,265).ThelinkbetweencommerceandlibertyisoneofthethemesdevelopedinOfCivilLiberty.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy397andmanners.Whilethistechniqueshapesalltheessays,threeinparticularstandoutinthisregard:theessaysonpopulation,money,andinterest.InthelongessayOfthePopulousnessofAncientNations,aworkthathasscarcelyreceivedtheattentionitdeserves,HumeaddressedapropositionadvancedbyRobertWallaceinhisDisser-tationontheNumbersofMankindinAncientandModernTimes(1753).Wallacemaintainedthatpopulationlevelshadbeenhigherinancientthaninmoderndays.31Inresponse,Humearguedthatthereisinallmen,bothmaleandfemale,adesireandpowerofgeneration,moreactivethaniseveruniversallyexerted.Conse-quently,inaddressingthequestionatissueitisnecessarytoknowthesituationofsocietyandtocompareboththedomesticandpoliticalsituationofthesetwoperiods,inordertojudgeofthefactsbytheirmoralcauses(E-PA4,3,5;381,383).Inarguingthatmodernsocietywasthemorepopulousone,Humepointedoutthattheuseofslaveryinancienttimeshadbeeningeneraldisadvantageousbothtothehappinessandpopulousnessofmankind.Healsopointedoutthatancienttimeshadbeencharac-terizedbyarelativelyhighincidenceofmilitaryconflictandbypolit-icalinstability.ButperhapsthemoststrikingaspectofhisargumentishisclaimthatTrade,manufactures,industry,werenowhere,informerages,soflourishingastheyareatpresentinEurope(E-PA34,83;396,416).Inshort,Humesawthatpopulationisultimatelylimitednotjustbypoliticalfactors,butalsobythefoodsupply,andthisinturnisaffectedbythetypeofeconomicorganizationprevailing.Humegrantedthatagricultureisthespeciesofindustrychieflyrequisitetothesubsistenceofmultitudes;anditispossible,thatthisindustrymayflourish,evenwheremanufacturesandotherartsareunknownandneglected(E-PA88,419).But,headded,Themostnaturalway,surely,ofencouraginghusbandry,is,first,toexciteotherkindsofindustry,andtherebyaffordthelabourerareadymarketforhiscommodities,andareturnofsuchgoodsasmaycontributetohispleasure31WallacesworkwasnotpublisheduntilafterHumes,butHumehadseenthemanuscript,andapparentlyencouragedWallacetopublishit.SeeRotwein,Hume:WritingsonEconomics,184.HumealsocitesMontesquieu,withwhomhecorre-spondedonthesubject(E-PA3,37980).Forasummaryofthetheoryofpopulationinthisperiod,seeSchumpeter,HistoryofEconomicAnalysis,2508.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n398andrews.skinnerandenjoyment.Thismethodisinfallibleanduniversal;and,asitprevailsmoreinmoderngovernmentthanintheancient,itaffordsapresumptionofthesuperiorpopulousnessoftheformer.(E-PA89,41920)Itisclear,then,thatHumesawnosimplerelationshipbetweenpop-ulationandthefoodsupply.Muchdependsontheformofeconomicorganization,onthedegreetowhichsectorsofactivityareinterde-pendent,andonthedegreetowhichmenaremotivatedbythedesireforgain.ThesamebasicviewinformsOfMoney.ThereHumerejectstheconventionalwisdomthatmoneycanberegardedaswealth,andgoesontostatethefamousrelationshipbetweenchangesinthemoneysupplyandthegeneralpricelevel:Ifweconsideranyonekingdombyitself,itisevident,thatthegreaterorlessplentyofmoneyisofnoconsequence;sincethepricesofcommoditiesarealwaysproportionedtotheplentyofmoney.32LessfamiliaristhefactthatHumeconsistentlycontrastedthesituationofaprimitiveeconomywithamoresophisticatedversion.Inaprimitiveeconomy,wemustconsider,that,inthefirstandmoreuncultivatedagesofanystate,erefancyhasconfoundedherwantswiththoseofnature,men,contentwiththeproduceoftheirownfields,orwiththoserudeimprovementswhichtheythemselvescanworkuponthem,havelit-tleoccasionforexchange,atleastformoney,which,byagreement,isthecommonmeasureofexchange.Inamoreadvancedstateofsoci-ety,Greatundertakers,andmanufacturers,andmerchants,ariseineverycommodity;andthesecanconvenientlydealinnothingbutinspecie.Andconsequently,inthissituationofsociety,thecoinentersintomanymorecontracts,andbythatmeansismuchmoreemployedthanintheformer(E-Mo1,1415;281,2901).Thechangedformofeconomicorganizationheraldsachangeinmannersbygivinggreaterscopetoindividualeffortandmustthere-foremassivelyincreasethesupplyofcommoditiesthataresubjecttoexchange.HumethusconcludedthatalthoughpricesinEuropehadrisensincethediscoveriesintheWestIndiesandelsewhere,thesepriceswereinfactmuchlowerthantheincreaseinthemoneysupplyitselfmightsuggest:32Humesinitialstatementofthisrelationshipoccursinaletterof10April1749,toMontesquieu(HL1:1368).Forasummaryoftreatmentsofthequantitytheoryofmoney,seeSchumpeter,HistoryofEconomicAnalysis,31117.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy399Andnoothersatisfactoryreasoncanbegiven,whyallpriceshavenotrisentoamuchmoreexorbitantheight,exceptthatwhichisderivedfromachangeofcustomsandmanners.Besidesthatmorecommoditiesarepro-ducedbyadditionalindustry,thesamecommoditiescomemoretomarket,aftermendepartfromtheirancientsimplicityofmanners.Andthoughthisencreasehasnotbeenequaltothatofmoney,ithas,however,beencon-siderable,andhaspreservedtheproportionbetweencoinandcommoditiesnearertheancientstandard.(E-Mo17,2923)TheessayOfInterestdiscussesaninstanceofthefallacyoftakingacollateraleffect...foracause.Aloweredrateofinterestisascribedtotheplentyofmoney;thoughitbereallyowingtoachangeinthemannersandcustomsofthepeople(E-Mo22,294).Thecontentionisthathighinterestratesarisefromthreecircum-stances:Agreatdemandforborrowing;littlerichestosupplythatdemand,andgreatprofitsarisingfromcommerce,whilealowrateofinterestwillreflectthecontrarycircumstances(E-In6,297).Inaprimitiveeconomy,theessaygoeson,therewillbelittleevidenceoffrugality,butoftenaconsiderabledemandforborrowingforthepurposeofconsumption.Thisstateofhabitsormannersisconsistentwithhighratesofinterest.Inthemoderneconomytherewillbehighlevelsofdemandforfundstobeusedforproductivepurposes,butalsoanenhancedsupplyofsuchfundsbecauseCommerceencreasesindustry,byconveyingitreadilyfromonememberofthestatetoanother,andallowingnoneofittoperishorbecomeuseless.Itencreasesfrugality,bygivingoccupationtomen,andemployingthemintheartsofgain,whichsoonengagetheiraffection,andremoveallrelishforpleasureandexpence.Itisaninfallibleconsequenceofallindustriousprofessions,tobegetfrugality,andmaketheloveofgainprevailovertheloveofpleasure.(E-In12,301)Inshort,theincreaseofcommercebyanecessaryconsequence,raisesagreatnumberoflenders,andbythatmeansproduceslow-nessofinterest.Thisresultisaccompaniedbyafurthertendencytoreducetherateofprofit:whencommercehasbecomeextensive,andemployslargestocks,theremustariserivalshipsamongthemer-chants,whichdiminishtheprofitsoftrade.Humethusconcludedthatthemostimportantsinglefactorwasnotsimplythesupplyofmoney,butachangeinmannersandintheformofeconomicorga-nization.Interest,hewrote,isthebarometerofthestate,anditsCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n400andrews.skinnerlownessisasignalmostinfallibleoftheflourishingconditionofapeople(E-In1314;3023).33Thetechniquejustconsideredcounselscautioninofferinggen-eralizationsineconomics.Thewayinwhicheconomicrelation-shipsdevelopwillnecessarilybeaffectedbymannersandbytheprevailinginstitutionalstructures.ItisthereforeimportanttonoteHumesawarenessofafurtherfact,namely,thateconomicrelation-shipswillbeaffectedbytheconditionofaneconomyevenwheretheinstitutionalstructureisstable.Hemakesthispointregularly,butaptlyillustrateditinOfMoney:Itseemsamaximalmostself-evident,thatthepricesofeverythingdependontheproportionbetweencommoditiesandmoney,andthatanyconsiderablealter-ationoneitherhasthesameeffect,eitherofheighteningorloweringtheprice.Encreasethecommodities,theybecomecheaper;encreasethemoney,theyriseintheirvalue(E-Mo12,290).Thisstatementseemsclearlytomeanthatanincreaseinthemoneysupplywillgenerateachangeinthepricelevelincaseswhereresourcesarefullyemployed,whileasimilarchangeinthesupplyofmoneycouldbeexpectedtoresultinanincreaseinthesupplyofcommoditiesifthereareunemployedresources.Humesanalysisoftheprocessbyvirtueofwhichchangesinthemoneysupplyaffecttheeconomyembracesbothresultsandatthesametimetakestheargumentastepfurther:Hereareasetofmanufacturersormerchants,weshallsuppose,whohavereceivedreturnsofgoldandsilverforgoodswhichtheysenttoCadiz.Theyaretherebyenabledtoemploymoreworkmenthanformerly,whoneverdreamofdemandinghigherwages,butaregladofemploymentfromsuchgoodpaymasters.Ifworkmenbecomescarce,themanufacturergiveshigherwages,butatfirstrequiresanencreaseoflabour;andthisiswillinglysubmittedtobytheartisan,whocannoweatanddrinkbetter,tocompensatehisadditionaltoilandfatigue.Hecarrieshismoneytomarket,wherehefindseverythingatthesamepriceasformerly,butreturnswithgreaterquantityandofbetterkinds,fortheuseofhisfamily.Thefarmerandgardener,finding,thatalltheircommoditiesaretakenoff,applythemselveswithalacritytotheraisingmore;andatthesametimecanaffordtotake33Rotwein,Hume:WritingsonEconomics,xivxvi;Schumpeter,HistoryofEco-nomicAnalysis,32734;andHutchison,BeforeAdamSmith,passim,provideadditionalinformationabouteighteenth-centuryviewsofinterestrates.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy401betterandmoreclothsfromtheirtradesmen,whosepriceisthesameasformerly,andtheirindustryonlywhettedbysomuchnewgain.Itiseasytotracethemoneyinitsprogressthroughthewholecommonwealth;whereweshallfind,thatitmustfirstquickenthediligenceofeveryindividual,beforeitencreasethepriceoflabour.(E-Mo7,2867)Thisreasoning,Humeadded,leadsustotheconclusionthatthedomestichappinessofacountryisentirelyindependentofthesizeofthesupplyofmoney.Allthatmatterswhatconstitutesgoodpolicyofthemagistrateisthatthesupplyofmoneycontinuallyincrease.Ifthemagistratecanachievethatgoal,hekeepsaliveaspiritofindustryinthenation,andencreasesthestockoflabour,inwhichconsistsallrealpowerandriches(E-Mo9,288).Butasweshallsee,aratherdifferentappreciationofthismatterwastoemergeinthecourseofHumesdiscussionofinternationaltrade.InternationalTradeThefinalmajoraspectofHumesdiscussionoftheseissuesconcernstheproblemofinternationaltrade;hisdiscussionagainproceedsatanumberoflevels.Hefirstdrawsattentiontothegeneralbene-fitsofforeigntrade.InOfCommerce,forexample,hepointsoutthatifweconsulthistory,weshallfind,that,inmostnations,for-eigntradehasprecededanyrefinementinhomemanufactures,andgivenbirthtodomesticluxury(E-Co15,263).Inthesameessayhemakesthefurtherpointthat,asSmithwaslatertoputit,imi-tationleadsdomesticmanufacturestoemulatetheimprovementsofforeignones.34HumerepeatsthisclaiminOftheJealousyofTrade,askingustheretocomparethesituationofGreatBritainatpresent,withwhatitwastwocenturiesago.Alltheartsbothofagricultureandmanufactureswerethenextremelyrudeandimper-fect.Everyimprovement,whichwehavesincemade,hasarisenfromourimitationofforeigners;andweoughtsofartoesteemithappy,thattheyhadpreviouslymadeadvancesinartsandingenuity(E-JT2,328).Thissentimentsetsthetoneofthisparticularessay,whichexplic-itlycriticizeswhatHumetooktobeacharacteristicfeatureof34WealthofNations,3.4.4(I:414).SeealsoHE33.58,3:328.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n402andrews.skinnermercantilistpolicy,namely,anunfoundedjealousyorsuspicionofthecommercialsuccessofothernations.Nothingismoreusual,amongstateswhichhavemadesomeadvancesincommerce,hewrote,thantolookontheprogressoftheirneighbourswithasuspiciouseye,toconsideralltradingstatesastheirrivals,andtosupposethatitisimpossibleforanyofthemtoflourish,butattheirexpence.Inoppositiontothisnarrowandmalignantopinion,Iwillventuretoassert,thattheencreaseofrichesandcommerceinanyonenation,insteadofhurting,commonlypromotestherichesandcommerceofallitsneighbours;andthatastatecanscarcelycarryitstradeandindustryveryfar,whereallthesurroundingstatesareburiedinignorance,sloth,andbarbarism.(E-JT1,3278)InapassagethatmaywellhavestruckachordwiththeFrencheconomistJ.B.Say,whofirstformulatedhisfamouslawwhiledis-cussingexactlythistopic,Humecontinued:35Theencreaseofdomesticindustrylaysthefoundationofforeigncommerce.Whereagreatnumberofcommoditiesareraisedandperfectedforthehome-market,therewillalwaysbefoundsomewhichcanbeexportedwithadvan-tage.Butifourneighbourshavenoartorcultivation,theycannottakethem;becausetheywillhavenothingtogiveinexchange.Inthisrespect,statesareinthesameconditionasindividuals.Asinglemancanscarcelybeindustri-ous,whereallhisfellow-citizensareidle....Norneedsanystateentertainapprehensions,thattheirneighbourswillimprovetosuchadegreeineveryartandmanufacture,astohavenodemandfromthem.Nature,bygivingadiversityofgeniuses,climates,andsoils,todifferentnations,hassecuredtheirmutualintercourseandcommerce,aslongastheyallremainindus-triousandcivilized.(E-JT34,329)HeclosedtheessaywithapassagethatmusthaveattractedtheattentionofAdamSmith:Ishallthereforeventuretoacknowledge,that,notonlyasaman,butasaBritishsubject,IprayfortheflourishingcommerceofGermany,Spain,Italy,andevenFranceitself.Iamatleastcertain,thatGreatBritain,andallthosenations,wouldflourishmore,didtheirsovereignsandministersadoptsuchenlargedandbenevolentsentimentstowardseachother.(E-JT7,331)35SaysLawisthatsupplycreatesitsowndemand.Ontheoriginofthislaw,seemySaysLaw:OriginsandContent,Economica34(1967):15366.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy403ThesecondaspectofHumesanalysissupportsthisattitudetowardforeigncommerceongroundsthatareessentiallytechni-cal.BuildingontheanalysisfoundinOfMoney,Humeexaminedthecaseoftwoormoreeconomieswithnounemployedresourceswithaviewtodemonstratingthefutilityofthemercantilistpreoc-cupationwithapositivebalanceoftrade.Againstthismercantilistconcern,Humecontendedthatanetinflowofgoldwouldinevitablyraisepricesinthedomesticeconomy,whilethelossofgoldfromtheforeigneconomieswouldreducethegeneralpricelevelinthem.Thenetresultwouldbeanincreaseinthecompetitivenessoftheforeigneconomyandadecreaseinthecompetitivenessofthedomes-ticeconomy.InOftheBalanceofTradeHumehadconcludedthatmoney,inspiteoftheabsurdjealousyofprincesandstates,hasbroughtitselfnearlytoalevel,justasallwater,whereveritcom-municates,remainsalwaysatalevel(E-BT15,11;314,312).Inthesameessayhegoesontoconclude:Fromtheseprincipleswemaylearnwhatjudgmentweoughttoformofthosenumberlessbars,obstructions,andimposts,whichallnationsofEurope,andnonemorethanEngland,haveputupontrade;fromanexor-bitantdesireofamassingmoney,whichneverwillheapupbeyonditslevel,whileitcirculates;orfromanill-groundedapprehensionoflosingtheirspecie,whichneverwillsinkbelowit.Couldanythingscatterourriches,itwouldbesuchimpoliticcontrivances.(E-BT35,324)ThethirddimensiontoHumestreatmentofforeigntradeismuchmorecomplex.Itisbasedonthepremisethatcountrieshavediffer-entcharacteristicsanddifferentratesofgrowth,thusopeningupadifferentanddistinctivepolicypositionascomparedtothosesofarconsidered.ThepresenceofanargumentreflectingajudgmentthateconomicconditionsarelikelytobediverseisnotperhapssurprisinginawritersuchasHume.AsRichardTeichgraeberhaspointedout,HumesperspectivewasEuro-Centric,ratherthanAnglo-Centric.36WhilecriticalofMontesquieusclaimthatmannersandcustomsdependonphysicalfactors,Humewasnonethelessconsciousofthefactthatdifferentcountriescouldhavedifferentphysicalendowments,36Teichgraeber,“FreeTrade”andMoralPhilosophy,106.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n404andrews.skinnerandwasclearlyawarethatclimatecouldhavesomeinfluenceoneconomicactivity(E-Co21,267;seealsoE-NC).Itisworthrecallingherethattheuseofthehistoricalmethodinvolvedthecomparisonofdifferenteconomictypes,whileempha-sisonthedynamicelementdrawsattentiontotheimportanceofindividualeffortandtoanacceleratingrateofchangeasinstitu-tionsandmannersthemselveschange.Ontheonehandthereaderisremindedoftheimportanceofadiversityofgeniuses,climatesandsoils,whileontheotherhandemphasisisplacedonthefactthattheextenttowhichmenmakeuseofart,care,andindustrymayvaryinonesocietyovertime,andbetweendifferentsocietiesatanygiventime.Otherfactorsthatwillaffecttherateofgrowthandcausevariationsinratesofgrowthindifferentcommunitiesincludetheformofgovernmentandthedegreetowhichpublicpoli-ciessuchastraderegulations,taxes,anddebtaredeployedwithintelligence.Humeillustratedthisnewaspectoftheproblembyreferringtotheissueofregionalimbalance(aconcernhesharedwithJosiahTucker),citingthecaseofLondonandYorkshire(E-PC11,3545).InhisearlyessayThatPoliticsMayBeReducedtoaScience,hemadetheinterestingfurtherclaimthat,thoughfreegovernmentshavebeencommonlythemosthappyforthosewhopartakeoftheirfreedom;yetaretheythemostruinousandoppressivetotheirprovinces(E-PR9,1819).Thisregionaldimensionisasrelevanttotherichcountrypoorcountrydebate(canapoorcountryhopetocatchandovertakearichcountry?)asistheinternationaldimension,althoughitwasonthelatterthatHumechosetoplacemostemphasis.Humestreatmentoftheperformanceofthemoderneconomy,especiallyinOfMoneyandOfInterest,impliesthatanincreaseinproductivitymaygivethedevelopedeconomyanadvantageintermsofthepriceofmanufactures.Healsorecognizedthataninflowofgoldintoagrowingeconomyneednotgenerateadversepriceeffects.AsheobservedinalettertoJamesOswald:Inevermeanttosaythatmoney,inallcountrieswhichcommunicate,mustnecessarilybeonalevel,butonlyonalevelproportionedtotheirpeople,industry,andcommodities.Tothisheadded,Iagreewithyou,thattheincreaseofmoney,ifnottoosudden,nat-urallyincreasespeopleandindustry(HL1:1423).Lookedatfromthispointofview,HumemighthaveagreedwithTuckersbeliefthatCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy405thepoorcountry,accordingtomyapprehension,canneverovertaketherich,unlessitbethroughthefaultandmismanagementofthelatter37InOfMoneyHumehadalreadynotedthatwhereonenationhasgottenthestartofanotherintrade,itisverydifficultforthelat-tertoregainthegroundithaslost;becauseofthesuperiorindustryandskilloftheformer,andthegreaterstocks,ofwhichitsmer-chantsarepossessed,andwhichenablethemtotradeonsomuchsmallerprofits(E-Mo3,283).Butheobservedthatthehistoricalincreaseinthequantityofmoneythatquickeneddiligencecouldalsoresultinageneralincreaseinthepricelevel,anincreasethatwouldbedisadvantageousinthecontextofinternationaltrade.Theadvantagesenjoyedbyarelativelyadvancedeconomy,hecontinued,arecompensated,insomemeasure,bythelowpriceoflabourineverynationwhichhasnotanextensivecommerce,anddoesnotmuchaboundingoldandsilver.Manufactures,thereforegraduallyshifttheirplaces,leavingthosecountriesandprovinceswhichtheyhavealreadyenriched,andflyingtoothers,whithertheyarealluredbythecheapnessofprovisionsandlabour;tilltheyhaveenrichedthesealso,andareagainbanishedbythesamecauses.And,ingeneral,wemayobserve,thatthedearnessofeverything,fromplentyofmoney,isadisadvantage,whichattendsanestablishedcommerce,andsetsboundstoitineverycountry,byenablingthepoorerstatestoundersel[l]thericherinallforeignmarkets.(E-Mo3,2834)HumeclearlyfeltthatthesetrendswerebeginningtomanifestthemselvesinEngland.There,somedisadvantagesinforeigntradebythehighpriceoflabour,whichisinparttheeffectoftherichesoftheirartisans,aswellasoftheplentyofmoney,werealreadyfelt(E-Co19,265).ThepositionwhichhewasstrivingtoformulatewaswellputinalettertoLordKamesinthecourseofadiscussionofadvantagesenjoyedbyrichcountries:Thequestionis,whethertheseadvantagescangoon,increasingtradeininfinitum,orwhethertheydonotatlastcometoaneplusultra,andcheckthemselves,bybegettingdisadvantages,whichatfirstretard,andatlastfinallystoptheirprogress....ItwasneversurelytheintentionofProvidence,thatanyonenationshouldbeamonopoliserofwealth:andthegrowthofallbodies,artificialaswellasnatural,isstoppedbyinternalcauses,derived37QuotedfromRotwein,Hume:WritingsonEconomics,205.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n406andrews.skinnerfromtheirenormoussizeandgreatness.Greatempires,greatcities,greatcommerce,allofthemreceiveacheck,notfromaccidentalevents,butnecessaryprinciples.(HL1:2712)ThesesentimentsexpandonapointthathadalreadybeenmadeinOfMoney,whereHumehadsaidthatthereseemstobeahappyconcurrenceofcausesinhumanaffairs,whichchecksthegrowthoftradeandriches,andhindersthemfrombeingconfinedentirelytoonepeople(E-Mo3,283).38ThepossibilitiesthatHumeoutlinedarenotwithouttheirimpli-cationsforeconomicpolicy.Arelativelybackwardeconomymight,forexample,finditinitsinteresttoadoptapolicyofprotectionforinfantindustries.Moreadvancedeconomiesconfrontingagenerallossofmarketsmighthavetoadoptapolicyofprotectiontosustainthelevelofemployment.Humeregardedthepossibilityofsucharesponsewithsomeequanimity,notingthatasforeigntradeisnotthemostmaterialcircumstance,itisnottobeputincompetitionwiththehappinessofsomanymillionswhomightotherwisefindthemselvesunemployed(E-Co19,265).Thus,althoughthereisinHumeswritingsamarkedpresumptioninfavoroffreetrade,healsorecognizedthatgovernmentinterven-tionmaybebeneficial.Butanypoliciessoinstitutedmustalwaysbeconsistentwiththeprevailingcircumstances.ThisperspectiveisitselfentirelyconsistentwiththatwhichHumeadoptedwhendealingwithquestionsofamorepurelytheoreticalnature.Humesconcernwithpolicyservestoremindusofotheraspectsofhiscontributiontoeconomictheory.Aswesawwhendiscussinghistoricaldynamics,Humestoneisthoroughlyoptimisticinthesensethathesaweconomicchangeasresultingfromaseriesof38Therichcountrypoorcountryissuewasthesubjectofconsiderabledebate.See,forexample,JamesOswaldtoHume,10October1749(Rotwein,1906)whereitisargued,amongotherthings,thatrichcountriesarelikelytoenjoycontinu-ingadvantages.HumetookissuewiththispositioninalettertoOswalddated1November1750(HL1:1424).ThetopicisalsoaddressedincorrespondencebetweenHumeandKames(HL1:2701)andJosiahTuckerandKames(Rotwein,2024).OnthisdebateseealsoIstvanHont,TherichcountrypoorcountrydebateinScottishclassicalpoliticaleconomy,inWealthandVirtue:TheShapingofPoliticalEconomyintheScottishEnlightenment,ed.IstvanHontandMichaelIgnatieff(Cambridge:CambridgeUniversityPress,1983),271315.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy407institutionalchangeswhosenetresultistogiveincreasingscopetohumanitysactivedispositionandinparticulartothepursuitofriches.Thisvisionofthefutureis,however,qualifiedbytheintroductionoftheclassicalthesisofgrowthanddecay,athe-sisthatmanifestsitselfinHumesbeliefthatmatureeconomieswilleventuallyandnecessarilyconfrontconstraintstotheirfurtherdevelopment.AfurtherqualificationofHumesoptimismemergesfromhisdis-cussionofwhathebelievedtobeacharacteristicfeatureofthemod-ernstate,namely,publiccredit.InthismoderninstitutionHumesawseveraldangers.First,nationaldebtscauseamightyconfluenceofpeopleandrichestothecapital,bythegreatsums,leviedintheprovincestopaytheinterest.Second,publicstocksbeingakindofpaper-credit,haveallthedisadvantagesattendingthatspeciesofmoney.39Third,holdersofthiskindofstockhavenoconnexionswiththestateandcanenjoytheirrevenueinanypartoftheglobe;theyareagroupliabletosinkintothelethargyofastupidandpam-peredluxury,withoutspirit,ambition,orenjoyment.Fourth,thisformofwealthconveysnohereditaryauthorityorcredittothepossessor;andbythismeans,theseveralranksofmen,whichformakindofindependentmagistracyinastate,institutedbythehandofnature,areentirelylost;andeverymaninauthorityderiveshisinfluencefromthecommissionaloneofthesovereign(E-PC1112,23;3545,3578).Humeconcludedthatthemodernstaterely-ingonpubliccreditcouldbeaffectedbythosecircumstancesthatwouldoffsetthepoliticalandconstitutionaladvantagesthathadbeenemphasizedinOfRefinementintheArts:Noexpedientremainsforpreventingorsuppressinginsurrections,butmercenaryarmies:Noexpedientatallremainsforresistingtyranny:Electionsareswayedbybriberyandcorruptionalone:Andthemiddlepowerbetweenkingandpeoplebeingtotallyremoved,agrievousdespo-tismmustinfalliblyprevail(E-PC23,358).39SeealsoOfMoney(E-Mo4,284)andOftheBalanceofTrade(E-BT212,317).TheseconcernsreflectthedoubtsrasedbyJohnLaw,MoneyandTradeConsidered(Edinburgh,1705),whosepositionwasechoedinpartbyGeorgeBerkeleyinhisQuerist(173537).Itisinterestingtonotethatboththesewriterswereconcernedwiththeproblemsofunderdevelopedeconomies;seeHutchison,184.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n408andrews.skinneriv.conclusionThemajorScottishfigureswhocontributedtothedevelopmentofpoliticaleconomyinthetwodecadesfollowingthepublicationofthePoliticalDiscourseswereSirJamesSteuartandAdamSmith.Thetwomencouldhardlyhavebeenmoredifferent.SteuarthadbeenacommittedJacobitewhereasSmithwasaWhig.Aseconomists,theformerwas,seemingly,anadvocateofinterventionism,whereasthelatterisnowregardedasaleadingadvocateofeconomicliberalism.YetSteuartandSmithhadtwothingsincommon:bothwerepro-foundlyinfluencedbyHumeseconomicessays,andbothenjoyedhisclosefriendship.Fromabiographicalpointofview,SteuartsknownlinkswithHumearefewindeed.WedoknowthatHumevisitedSteuartinhisancestralhomeofColtness,inLanarkshire,onanumberofocca-sions,duringatleastoneofwhichthetwomendiscussedHumesHistory.ThePrinciples(1767)alsofiguresintheonelongletterfromSteuarttoHumethathassurvived,aletterthatisremarkableforitsgoodhumorandfamiliarityandthatatteststhemanyproofsyouhavegivenmeofyourfriendship.HumehadprobablygivensomeassistanceinthevexedquestionofSteuartspardon,buthadalsoreadthePrinciplesindraft.Inaletterdated11March1766,Profes-sorRouetwrotetoBaronMurethatGeorgeScottandDavidHumehavelookedintoourfriendsmanuscriptandareexceedinglypleasedwithitalthoughHumewaslatersaidtohavebeencriticalofitsformandstyle.40ItisnotdifficulttoseewhyHumemighthaveapprovedatleastofthestructureofthePrinciples.ThebookparallelsHumespre-occupationwiththesocialandpoliticalimplicationsofeconomicgrowthandplacesasimilaremphasisontheroleofnaturalwants.Inittootherearediscussions,similartothosefoundinHume,ofthetheoryofpopulationandofthenatureoftheexchangeeconomy.Indeed,itcouldbeclaimedthatSteuartcarriedHumesargumentfurtherinthesensethatheaddressedtheproblemsthatcouldbefacedbyaneconomyintheprocessoftransitionfromanadvancedagrarianstagetoaprimitivestageoftheexchangeeconomy.SteuartalsowentfurtherthanHumeinaddressingtheissuespresentedby40Skinner,SirJamesSteuart,xlvxlvi.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy409variationsinratesofgrowthbothregionallyandinternationally;ananalysisthatresultedinageneralizedstatementofthethreestagesoftrade:infant,foreign,andinland.41IfSirJamesSteuartofferedalegitimatedevelopmentofHumestreatmentofpoliticaleconomy,itisequallytruethatAdamSmithmorefullycomprehendedthelattersviewsastotheappropriateshapeandscopeofthisdiscipline.ItisnowacommonplacethatSmithendeavoredtolinkphilosophy,history,andeconomicsaspartofagrandplanthatwasannouncedintheclosingpagesofthefirsteditionofTheTheoryofMoralSentimentsandrepeatedintheadver-tisementtothesixthandlasteditionofthatwork.ButwhenwetaketheTreatiseofHumanNatureinconjunctionwiththeEssaysMoralPoliticalandLiterary,itbecomesapparentthattheoutlinesofthemodelhadalreadybeenestablishedbyHume.HeitwaswhosawthecloserelationshipbetweentheUnderstandingandPassions,thesubjectsofthefirsttwobooksoftheTreatise,andMorals,Pol-itics,andCriticism,theremainingsubjectsheprojectedaspartofafive-volumeTreatiseofHumanNature(TAdv.).ItisimportanttonotethatAdamSmithhadacloseknowledgeofHumesphilosophy,soclosethatDugaldStewartwouldconcludethatthePoliticalDiscoursesofMrHumewereevidentlyofgreaterusetoMrSmiththananyotherbookthathadappearedpriortohislectures.42ItisequallynoteworthythatSmithshouldhaveacknowledgedHumeshistoricalanalysisofthelinksbetweencom-merceandliberty.SmithwouldhaveagreedwiththeviewthatHumedeservestoberemembered...forhismorefundamentalattempttoincorporateeconomicsintoabroaderscienceofhumanexperience.43ButSmithsformaleconomicanalysisdiffersfromthatofHume(andofSteuart)partlybecauseitwasfinallydevelopedaftertheappearanceofsomeofthegreatsystematicperformancesoftheperiod.NotableamongtheseisRichardCantillonsEssaisurlaNatureduCommerceenGeneral,writteninthe1730sbutnot41SeemyDavidHume:PrecursorofSirJamesSteuart,inDiscussionPapersinEco-nomics(UniversityofGlasgow),No.9003;andTheShapingofPoliticalEconomyintheEnlightenment,ScottishJournalofPoliticalEconomy37(1990):14565.42DugaldStewart,AccountoftheLifeandWritingsofAdamSmith,inEssaysonPhilosophicalSubjects(Oxford:ClarendonPress,1980),3201.43Rotwein,Hume:WritingsonEconomics,cxi.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n410andrews.skinnerpublisheduntil1755.Cantillonsteachingwasdisseminated,inpart,bytheMarquisdeMirabeauintheAmidesHommes(1756)andprob-ablyhadaprofoundinfluenceonthePhysiocrats.44ThemostnotableofthisgroupincludeFrancoisQuesnay,whoseTableauEconomique(1757)providedacoherentaccountofamacroeconomicmodel,andTurgot,whoseReflectionsontheFormationandDistributionofRichesdatesfrom1766.45AdamSmithwastoobjectthatthemembersofthePhysiocraticSchoolallfollowimplicitly,andwithoutanysensiblevariation,thedoctrineofMr.Quesnai.46Perhapswiththisinmind,Humein1769wrotetotheAbbeMorellet:´Ihopethatinyourworkyouwillthunderthem[thePhysiocrats],andcrushthem,andpoundthem,andreducethemtodustandashes!Theyare,indeed,thesetofmenthemostchimericalandmostarrogantthatnowexist,sincetheannihilationoftheSorbonne.Iaskyourpardonforsayingso,asIknowyoubelongtothatvenerablebody.Iwonderwhatcouldengageourfriend,M.Turgot,toherdamongthem;Imean,amongtheeconomists.(HL2:205)ButthetruthisthatwriterssuchasQuesnayandTurgotproducedamodelofacapital-usingsysteminwhichanumberofsectorsofactivitywerefeatured.Inaddition,socioeconomicgroupswerepresentedasbeingfullyinterdependent.AdamSmithknewofthiswork,anditseemstohaveinfluencedthemacroanalysisofBook2oftheWealthofNations.47Humesessays,mostlywrittenby1752,areinnocentofamodelofthiskindandsotoowasSteuartsPrinciples,thefirsttwobooksofwhichwerecompletedintheisolationofTubingenearlyin1759.Humeseconomicessaysdonot¨comparewiththegreatsystematictreatisesofhisfriends,different44RichardCantillon,Essaisurlanatureducommerceengeneral,ed.HenryHiggs(London:Macmillan,1931);HenryHiggs,ThePhysiocrats;SixLecturesontheFrenchEconomistesofthe18thCentury(London:Macmillan,1897);RonaldL.Meek,TheEconomicsofPhysiocracy(London:AllenandUnwin,1962).ForabriefdiscussionoftheviewsofthePhysiocrats,seeTheNewPalgrave:ADictionaryofEconomics,s.v.Physiocracy.45ThelatterworkisincludedinTurgotonProgress,SociologyandEconomics,ed.andtrans.RonaldL.Meek(Cambridge:CambridgeUniversityPress,1973).46WealthofNations,4.9.38(2:6789).47SeemyASystemofSocialScience(Oxford:OxfordUniversityPress,1979),Chap.5.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy411astheywereincharacter,ortotheanalyticalcontributionsofthePhysiocrats.Thatbeingsaid,itmustbenotedthatHumemadesignificantcontributionsinthefieldsofpopulationandofmoney,especiallytoquantitytheoryandtotheanalysisofspecieflow.Therehisworkremainedsubstantiallyunchallengeduntilthe1920s.48Humealsosucceededinestablishingthatthereisarelationshipbetweentheproductionofcommoditiesandthelevelofaggregatedemand,arelationshipmorecommonlyassociatedwiththeworkofJ.B.Say.CertainlyHumesanalysisofthesectoraldivisionoflabor,histreat-mentofthetheoryofpopulation,andhisconsiderationofinterna-tionaltradeseparatelyandseverallypromptaconclusionthat,inthewordsofSay,mayatfirstsightappearparadoxical;viz.thatitisproductionwhichopensademandforproducts.49IfHumesessaysdonotconstituteasinglecoherenttreatise,theydo,asthisessayhasendeavoredtoshow,discloseevidenceofsys-tematictreatment.Perhapsthemostimportantsinglefeatureofthistreatmentistobefoundintheuseofhistoryandofthehistoricalmethod:Humeconsistentlysoughttolinkeconomicrelationshipswiththeenvironmentandthestateofmanners.ThispositionwastofindlaterexpressionintheworkoftheGermanHistoricalSchoolandoftheAmericanInstitutionalists.50ButitisimportanttonotethatHumeshistoricaltechniqueisdifferentfromthatlateradoptedbyAdamSmith.InSmithshands,thehistoryofcivilsocietyisessentialforourunderstandingoftheexchangeeconomyandofthesocialandpoliticalenvironmentwhichitmayproduce.ButinSmithhistoryistheprefacetopoliticaleconomyratherthanintegraltothetreatment.Infact,ithasbeensaidthatSmithdidnotusethehistor-icalmethodindealingwitheconomicquestions:Onemaysaythat,despiteitspronouncedemphasisoneconomicdevelop-ment,SmithsapproachtoitsmoregeneralaspectsislessbasicallygeneticorevolutionarythanHumes....Withregardparticularlytohistreatment48Schumpeter,HistoryofEconomicAnalysis,367;seealsoVickers,TheoryofMoney,21739.49QuotedfrommySaysLaw:OriginsandContent,Economica,NewSeries,34,no.134(May1967):159.50LeoRogin,TheMeaningandValidityofEconomicTheory:AHistoricalApproach(NewYork:Harper&Brothers,1956),113.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n412andrews.skinnerofthetheoreticalissuesofpoliticaleconomy,SmithclearlyexhibitsthetendencytoabstractfromhistoricalinfluenceswhichwassocharacteristicofRicardoandthelaterclassicaleconomists.51AfurtherpointofinteresttothemoderneconomistisHumessystematiccomparisonofdifferenteconomicstagesandhiscon-cernwiththeprocessoftransitionbetweenthem.Thisproce-durethrowsimportantlightontheproblemsofeconomicandsocialdevelopment.Sotoodoeshisconcernwithinternationaltradebetweeneconomieswithdifferentcharacteristicsanddifferentratesofgrowth.Humesargumenteffectivelyintroducedtherichcountrypoorcountrydebatethatwasalsoaddressedby,amongothers,Tucker,Wallace,andSteuart.Suchaperspectivemeansthatpolicyrecommendationsmustalwaysberelatedtothecircum-stancesthatprevail.JosephSchumpetersdescriptionoftheworkdonebyanothereighteenth-centuryeconomist,theItalianFerdi-nandoGaliani,thusapplies,despiteHumesbeliefintheuniformityofhumannature,equallytoHume:Onepointabouthisthoughtmustbeemphasised...hewastheoneeighteenth-centuryeconomistwhoalwaysinsistedonthevariabilityofmanandoftherelativitytotimeandplace,ofallpolicies;theonewhowascom-pletelyfreefromtheparalysingbelief,thatcreptovertheintellectuallifeofEurope,inpracticalprinciplesthatclaimuniversalvalidity;whosawthatapolicythatwasrationalinFranceatagiventimemightbequiteirrational,atthesametimeinNaples.52TheanalyticalsuccessoftheWealthofNationsinthefirsttwodecadesofthenineteenthcenturyhadsomeunfortunateresults.53Thedominantclassicalorthodoxymadeitpossibletothinkofeco-nomicsasquiteseparatefromethicsandhistory,thusobscuringthetruepurposesofSmithandHume.Inreferringtotheseproblems,Hutchisonwasmovedtoremark,inatellingpassage,thatSmithwasunwittinglyledbyaninvisiblehandtopromoteanendnotpart51Rotwein,Hume:WritingsonEconomics,cixcx;seealsoHutchison,BeforeAdamSmith,21314.52Schumpeter,HistoryofEconomicAnalysis,2934.53RichardF.TeichgraeberIII,LessabusedthanIhadreasontoexpect:TheRecep-tionofTheWealthofNationsinBritain,177690,HistoricalJournal30(1987):33766.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumesPrinciplesofPoliticalEconomy413ofhisintention,thatofestablishingpoliticaleconomyasaseparateautonomousdiscipline.54ThedominanceofaversionofSmithseconomicsysteminthenineteenthcenturyledtothebeliefthatthehistoryofthesubjectdatedfrom1776,thusobscuring,temporarilyatleast,thecontribu-tionsofSmithspredecessors,English,French,Italian,andScottish.TheacceptanceofSmithsaccountofthemercantilesystemalsocausedadvocatesofintervention,suchasSteuart,toberegardedasmercantilistsonthisgroundalone,andsometimestocausecom-mentatorstoviewwithmildembarrassmenttheoccasionaldepar-turesoftheenlightenedHume,notfromtheprinciple,butfromtheapplication,ofapolicyoffreetrade.suggestionsforfurtherreadingInadditiontotheworkscitedinthenotestothisessay,forfurtherreadingthefollowingarerecommended.Hont,Istvan.JealousyofTrade:InternationalCompetitionandtheNationStateinHistoricalPerspective.Cambridge,MA:HarvardUniversityPress,2005.Rotwein,Eugene.DavidHume.InTheNewPalgrave:ADictionaryofEconomics,editedbyJ.Eatwell,M.Milgate,andP.Newman.London:Macmillan,1987.Schabas,Margaret.DavidHumeonExperimentalNaturalPhilosophy,Money,andFluids.HistoryofPoliticalEconomy33(2001):41135.Schabas,Margaret,andCarlWennerlind,eds.DavidHume’sPoliticalEcon-omy.London:Routledge,2007.54Hutchison,BeforeAdamSmith,355.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\npeterjones13HumeontheArtsandTheStandardofTasteTextsandContextsHumesobservationsontheartsaresetintheframeworkofsociallife.Thatiswhyheconsidersboththemakingof,andresponseto,worksofartashumanactionssubjecttotheanalysishehasofferedofotherhumanactions.Heneverpublishedhisintendedtreatiseoncriticism(TAdv.),andnodevelopedtheoriesofbeauty,art,orcriticismaretobefoundinhisworks.Butbybringingtogetherhisscatteredremarksonthesesubjects,andbylookingathisgeneralaimsandthecontextinwhichhewrote,wecanidentifyhisprincipalviewsonthesetopics.i.culturalcontextItisalwaysimportanttoestablishthecontextanddateofawritersviews;itisespeciallyimportantwhensettingoutHumesideasonwhattodaywecallaesthetics.Hisinterestsandreferences,inalmosteveryrespectexceptthecrucialoneofclassicalliterature,werenar-rowerthanthoseofaninformedmodernreader.Weneedtoknowwhathemight,andwhathecouldnot,haveexperienced,andtorec-ognizethathisculturalenvironmentdifferedessentiallyfromours:conceptsof,andattitudesto,thevariousmediumsofartinthe1740swereevolvingrapidly,aswereartisticpracticesandexpectations.Exceptingonlytheveryrich,andtheirhouseholds,mostScotsandthemajorityoftheEnglishhadverylimitedaccesstowhatcountastheartstoday.MostofHumesobservationsweremadewithinathirty-yearperiodbeginninginthelate1720s,attheoutsetofsocialandintel-lectualrevolutionsthatweretogainrapidmomentuminthesecondhalfoftheeighteenthcenturyandtotransformEurope,although414CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste415thereweresignificantgeographicalvariations.Aesthetics,under-stoodprimarilyasreflectiononthenatureoftheartsandofdis-cussionaboutthem,achieveditsmodernformsonlyafterHumesdeath.Itsdevelopmentisinseparablefrommanyintellectual,social,political,andeconomicfactors:thespreadofwealthandincreaseofleisureamongthemiddleclassesinmanyEuropeancountries;greatereaseoftravel,andthebeginningofpublicconcertsandpublicmuseums(inwhichworkswouldberemovedfromtheiroriginalcontexts);adeclineinindividualpatronage,andaresult-ingfreedomforartiststosatisfyagrowingluxurymarket,ortosurvivebyfollowingtheirownpreferences;thebeginningoftheformalstudyofthearts,especiallyliterature,bynonpractitionersandnonownersincollegesanduniversities,thusaugmentinganinformedaudience;agreateravailabilityofbooksandillustrationsassecondarysourcesofinformationaboutthearts;theincreasinginfluenceofcritics,throughjournals;andtheinstitutionalizationoftheartsandsciencesandthusofprofessionaldistinctionsbetweenthem.Humesreferencestoartsotherthanliteratureareinfrequentandfleeting.Healmostneverreferstomusic,ortosculpture,hisasidesonpaintingareinconsequential,andarchitecturegainsmorethanapassingmentiononlyinhislettersfromthemainlandofEuropein1748;whatlittletheoreticalorphilosophicalwritingwasavailabletohimontheseartsreceivesalmostnomention.1Hiscriticalviews1Themusicalsceneineighteenth-centuryScotlandisexploredinDavidJohnson,MusicandSocietyinLowlandScotlandintheEighteenthCentury(Oxford:OxfordUniversityPress,1972).SeealsoNealZaslaw,ed.,TheClassicalEra:Fromthe1740stotheEndofthe18thCentury(London:Macmillan,1989);EnricoFubini,MusicandCultureinEighteenth-CenturyEurope(Chicago:UniversityofChicagoPress,1994);RuthSmith,HandelsOratoriosandEighteenth-CenturyThought(Cambridge:CambridgeUniversityPress,1995);WilliamWeber,TheRiseofMusi-calClassicsinEighteenth-CenturyEngland(Oxford:ClarendonPress,1992);Cyn-thiaVerba,MusicandtheFrenchEnlightenment(Oxford:ClarendonPress,1993);ThomasChristensen,RameauandMusicalThoughtintheEnlightenment(Cam-bridge:CambridgeUniversityPress,1993);DonaldBurrowsandRosemaryDunhill,MusicandTheatreinHandelsWorld(Oxford:OxfordUniversityPress,2002).ThecontextsofpaintingandarchitectureinBritainareanalyzedinJohnSum-merson,GeorgianLondon(London:PleiadesBooks,1945);IainPears,TheDiscoveryofPainting:TheGrowthofInterestintheArtsinEngland,16801768(NewHaven,CT:YaleUniversityPress,1988);BasilSkinner,ScotsinItalyinthe18thCentury(Edinburgh:NationalGalleriesofScotland,1966);JamesHolloway,PatronsandPainters:ArtinScotland16501760(Edinburgh:ScottishNationalPortraitGallery,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n416peterjonesseemtohavebeenformedwithmainlypoetryanddramainmind,althoughitwascommonplaceinhisdaytocompareandeveniden-tifypoetrywithpainting,asDrydenhaddoneinhisprefacetothetranslationofC.A.duFresnoyspoemDeArteGraphica(TheArtofPainting),entitledAParallelbetwixtPaintingandPoetry(1695).2ThisworkwaswidelyquotedbycontemporariessuchasGeorgeTurnbullandAlexanderGerard,3andHumemayevenalludetoit(T2.2.5.19).Humeacceptedastandardviewofhistimethatpaint-ingscouldconveyanarrativeoractasanhistoricalrecordorsymbol,buthemoreoftenregardedthemasessentiallypiecesofusefullydecorativefurniturelikeJamesNoriesfashionableovermantlepaintingsinEdinburghandtheLowlands.ThepainterAllanRamsaywasaclosefriend,and,ofcourse,HumevisitedScottishBorderand1989);GeorgeFairfull-Smith,TheFoulisPressandtheFoulisAcademy(Glasgow:GlasgowArtIndex,2001).Forgardens,seeA.A.Tait,TheLandscapeGardeninScotland17351835(Edin-burgh:EdinburghUniversityPress,1980).TheFrenchcontextforarchitecture,painting,andcriticismisexaminedinJosephRykwert,TheFirstModerns(Cam-bridge,MA:MITPress,1980);AntoinePicon,FrenchArchitectsandEngineersintheAgeofEnlightenment(Cambridge:CambridgeUniversityPress,1992);ThomasE.Crow,PaintersandPublicLifein18th-CenturyParis(NewHaven,CT:YaleUniversityPress,1985).SeealsoAndrewWiltonandIlariaBignamini,eds.,GrandTour:TheLureofItalyintheEighteenthCentury(London:TateGalleryPublish-ers,1996);JohnIngamells,ed.,ADictionaryofBritishandIrishTravellersinItaly17011800(NewHaven,CT:YaleUniversityPress,1997);J.PaulHunter,BeforeNovels:TheCulturalContextsofEighteenth-CenturyEnglishFiction(NewYork:Norton,1990);M.H.Abrams,Art-as-Such:TheSociologyofModernAesthetics,AmericanAcademyofArtsandSciences38(1985);RemyG.Saisselin,TheTrans-´formationofArtintoCulture:FromPascaltoDiderot,andStudiesonVoltaireandtheEighteenthCentury70(1970);idem,Painting,WritingandPrimitivePurity,StudiesonVoltaireandtheEighteenthCentury217(1983):25796;BernardDenvir,TheEighteenthCentury:Art,DesignandSociety16891789(London:Longman,1983).2JohnDryden,AParallelbetwixtPaintingandPoetry,inC.A.duFresnoy,DeArteGraphica:TheArtofPainting(London,1695).OnthelistofbooksinthelibraryorganizedbyProfessorRobertSteuartin1724,calledthePhysiologicalLibrary,andtowhichHumehadaccessasastudent,seeMichaelBarfoot,HumeandtheCultureofScienceintheEarlyEighteenthCentury,inStudiesinthePhilosophyoftheScottishEnlightenment,ed.M.A.Stewart(Oxford:ClarendonPress,1990),15190.Forastudyofcomparisonsbetweenpoetryandpainting,seeR.W.Lee,UtPicturaPoesis:TheHumanisticTheoryofPainting(NewYork:W.W.Norton,1967).3GeorgeTurnbull,ATreatiseonAncientPainting(London,1740);AlexanderGerard,AnEssayonTaste(London,1759);HenryHome,LordKames,ElementsofCriticism(London,1762).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste417Lowlandhouses,andpossiblyalsothebesthousesinRheimsandLaFleche,whenstudyingthereinthe1730s(HL1:5,12).Nevertheless`hehadprobablyseenratherfewpaintingsotherthanportraitsuntilheaccompaniedGeneralSt.ClairtoViennaandnorthernItalyin1748(HL1:64).WeshouldrememberthatstrictCalvinistchurchesdisplayednopaintings,andthatthelabelportraitcoveredfancifulhistoricallikenessesaswellasfacesofactualsitters.UntilafterthemidcenturytherewerefewcollectorsinScotland:theClerksofPenicuikandthesecondMarquisofAnnandalecertainlyhadnotablecollections,althoughtheDukeofArgylldidnot,andseveralpromi-nentScotsknowntoHumemadetheGrandTour,aspatronsortutors.ButeveninEnglandthemarketandaudienceforpainting,togetherwithcriticaldebate,hadnotdevelopedtotheextentHumewitnessedinParisinthe1760s.Therewere,ofcourse,illustratedbooksandengravings,butsomeengravingswerebothreversedandconsistedonlyofoutlinesofthemainformsinapainting.Inanycasetheirsmallscaleandcontentencouragedaliteraryapproach.UnlikeGeorgeTurnbullinhisTreatiseonAncientPainting(1740)forwhomRamsaycommissionedengravingswhiletheywerebothinItalyAlexanderGerardinhisEssayonTaste(1759),orevenLordKamesinhislaterElementsofCriticism(1762),Humemakesnoref-erencetotheinfluentialideasofRogerdePiles,CharlesAlphonseduFresnoy,orAndreFelibien;andofJonathanRichardson,whose´challengingwritingsfirstappearin1715,thereisnotaword.4Thereisampletextualevidence,nevertheless,thatHumederivedmanyofhisideasaboutcriticismfromFrenchwriters,andwenowknowthatheowned,atsomestage,manyimportantFrenchtexts.54RogerdePiles,Conversationssurlaconnoissancedelapeintureetsurlejuge-mentquondoitfairedestableaux(n.p.,1677);CharlesAlphonseduFresnoy,DeArteGraphica(Paris,1668);AndreF´elibien,´LIdeedupeintreparfait´(Paris,1707);JonathanRichardson,TwoDiscourses.I.AnEssayonthewholeArtofCriticismasitrelatestoPainting.II.AnArgumentinbehalfoftheScienceofaConnoisseur(London,1719).DePiles,intheopeningbookandinaverybriefessay,OfTaste,withwhichheendsTheArtofPainting(2nded.,London,1744),listsanumberofideastakenupbyDubos,andlaterbyHume.InScotland,Kames,inparticular,discussedarchitectureatsomelength;Smithwroteaboutmusiclateinhislife,butinthecontextofHumesessay,themostinterestingdiscussionofmusicisbyJohnGregory,AComparativeViewoftheStateandFacultiesofMan(London,1765).5ForadetailedanalysisofHumesviewsontheartsandcriticismseemyHumesSentiments:TheirCiceronianandFrenchContext(Edinburgh:EdinburghUniver-sityPress,1982).IdiscussthepossibleinfluenceonHumeofEphraimChambers,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n418peterjonesHumedisplaysnointerestinmusic,buthewasnotwithoutopportunitytolearnaboutit.Althoughformallyconstitutedonlyin1728,theEdinburghMusicalSocietyhadflourishedsincethe1690s,providingencouragementforseveralgoodamateurperform-ersandcomposers;butunlikeKames,WilliamCullen,orJosephBlackheseemsnottohavebeenamember.Therewasastrongtra-ditionofdanceandfolksongthroughoutmuchofScotland,however,withwhichHumewouldhavebeenfamiliar.EnglishoperaarrivedinEdinburghin1751,butwehaveproofonlythatHumeattendedtheoperainTurinin1748andinParisin1764,andreportedthathehadheardthecastratoAngeloMariaMonticelliinVienna,alsoin1748,butpossiblynotinopera.Therearenoreferencestothecon-certsoftheeight-year-oldMozartinParisin1764.Inthefirsthalfofthecentury,therewaslittletheoreticaldiscussionofmusicoutsideFrance,andHumereferstonooneotherthanAbbeJean-Baptiste´Dubos,althoughhemusthaveheardaboutthedisputebetweenJeanJacquesRousseauandJeanPhilippeRameauinthe1750s.HealsocametoknowDr.CharlesBurney,thegreathistorianofmusic,CountAlgarotti,whowroteonopera,andJean-FrancoisMarmontel,whowroteseveraloperalibretti.Bywayofcontrast,AlexanderGer-ard,JohnGregory,andJamesBeattie,inAberdeen,alongwithKamesandAdamSmith,allwrotesomethingaboutmusic.AlthoughHumelaterseemstohaveownedotherbooksthatdiscussedmusic,byauthorssuchasDAlembert,WilliamDuff,DanielWebb,theMar-quisdeChastellux,aswellaslibrettibyAbatoPietroMetastasio,theirideasappearnowhereinhisownwritings.Cyclopaedia:oranUniversalDictionaryofArtsandSciences(London:EphraimChambers,1728)inHumeandtheBeginningsofModernAesthetics,inTheScienceofManintheScottishEnlightenment,ed.P.Jones(Edinburgh:EdinburghUniversityPress,1989).InformationabouttheHumeLibraryistakenfromDavidFateNortonandMaryJ.Norton,TheDavidHumeLibrary(EdinburghBibliograph-icalSociety,1996).EarlierstudiesonHumeandhiscontextincludeT.Brunius,DavidHumeonCriticism(Stockholm:UniversityofUppsala,1951);W.J.Hipple,Jr.,TheBeautiful,TheSublime,andthePicturesqueinEighteenth-CenturyBritishAestheticTheory(Carbondale:SouthernIllinoisUniversityPress,1957).SeealsoRemyG.Saisselin,´TheEnlightenmentagainsttheBaroque:EconomicsandAes-theticsintheEighteenthCentury(Berkeley:UniversityofCaliforniaPress,1992).SeealsomyItalianOperasandTheirAudiences,inTheEnlightenmentWorld,ed.MartinFitzpatricketal.(London:Routledge,2004).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste419Itwouldbenaturalforarchitecture,ratherthanpaintingormusic,tocapturetheattentionofsomeoneengrossed,asHumewas,inthedebatebetweentheAncientsandModerns;Addisonhadhim-selfdemonstratedthatfact,andcountlessauthorsbeforeandafterHumelikedtorefertotheDivineArchitectwhencomparingtheworksofGodandman.6Humeownedatleastthreeofthemajorvolumesthatessentiallychampionedthemoderns.7HefurthermorecommentsonthedebateinletterswrittenduringhisViennesemis-sion,andin1767expressesprideinthearchitecturalachievementofRobertAdam(HL1:11827;2:173).Herefersto,andthuspre-sumablyhadseen,thebeautifullyillustratedtranslationsofAndreaPalladiothatwereavailablefromthe1720s;andhequotesasignifi-cantpassagefromClaudePerraultsinfluentialcommentaryontheRomanwriterMarcusVitruviusPollio(EPMApp.1.15;T2.1.8.2).8LikePascal,bothCharlesandClaudePerraultemphasizedfashionasinfluencingarchitecturalstylesanddetails,andthispointwas6SeemyHumeonContext,SentimentandTestimonyinReligion,inReligionandHumesLegacy,ed.D.Z.PhillipsandTimothyTessin(NewYork:St.MartinsPress,1999),25177.7NortonandNortonindicatethatthefollowingworkswereintheHumeLibrary,andtheymaywellhavebeentherebeforeHumewrotehisessayof1757:CharlesPerrault,Parallelesdesanciensetdesmodernes`(Amsterdam,1693;lsted.,1688);WilliamWotton,ReflectionsuponAncientandModernLearning(London,1694);Jean-BaptisteDubos,ReflexionsCritiquessurlapo´esieetsurlapeinture´(Utrecht,1732).HumeownedaneditionofBernardleBovierdeFontenelle,Oeuvres(Paris,1742)andwascertainlyfamiliarwithitscontents,suchasDigressionsurlesanciensetlesmodernes(Paris,1688),andDialoguesdesMorts(Paris,1683):andhealsoownedDominickBouhours,LesEntretiensDAristeetDEugene(Paris,1671);HoudardelaMotte,ReflexionssurlaCritique´(TheHague,1715);JamesHarris,ThreeTreatises,ConcerningArt,Music,Painting,Poetry,andHappiness(London,1744);andRogerdePiles,ArtofPainting(London,1744).ThelastworkoriginatesindePilessoriginaltranslationofandcommentaryonDuFresnoy,DeArteGraphica(Paris,1668).SeeThomasPuttfarken,RogerdePilesTheoryofArt(NewHaven,CT:YaleUniversityPress,1985);EdgarWind,HumeandtheHeroicPortrait(Oxford:ClarendonPress,1986);WilliamHogarth,TheAnalysisofBeauty,ed.R.Paulson(NewHaven,CT:YaleUniversityPress,1997;firstpublished1753).8OnRobertAdam,seeJohnFleming,RobertAdamandHisCircle(Cambridge,MA:HarvardUniversityPress,1962);GeoffreyBeard,TheWorkofRobertAdam(Edin-burgh:JohnBartholomew&Son,1978);JosephandAnneRykwert,TheBrothersAdam(London:Collins,1985).ClaudePerrault,LesDixLivresdarchitecturedeVitruve(Paris,1673).ProfessorSteuartsclasslibrarylistof1724citesacopyofPerraultsAbreg´e´ofVitruvius,alongwithworksontheconstructionandgeometryoffortifications.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n420peterjonesadaptedandbroadenedbybothAllanRamsayandAdamSmithintheirdiscussionswithHumeabouttaste.ButwhatelsehadHumeread?TherewaslittleonarchitectureofaphilosophicalnatureineitherFranceorEnglandbeforethe1760s.Therewerepracticalhand-booksforpatronsandbuilders,ofcourse,andwhatwereessentiallypatternbooks,butthereisnoreasontothinkthattheywereofinteresttoHume.9Vitruvius,LeonBattistaAlberti,Palladio,Sebas-tianoSerlio,VincenzoScamozzi,andGiacomeBarozzidaVignolawereallavailableinEnglish,andHumecouldreadilyhavecon-sultedtheworkofFreartdeChambrayaswellasthatofPerrault.´10Butif1757,andtheappearanceofhisessayontaste,istakenasthelastdateforinfluencingHumespublishedphilosophicalthoughtsonsuchmatters,Marc-AntoineLaugiers(asthenanonymous)AnEssayonArchitecture(1753,Eng.trans.,1755)wasavailable,butSirWilliamChamberssATreatiseonCivilArchitecture(1759)isruledout.TherewerealsothearticlesonarchitecturalmattersintheearlyvolumesofthefamedEncyclopedieouDictionnaireRaisonn´e´desSciences,desArtsetdesMetiers´(Paris,175180),andseveralworksbyRobertMorris,butHumereferstothesenomorethantotheothers.11Andyet,evenifonlyindirectly,Humeabsorbedideasfromsomeofthesewriters.Forexample,Vitruviushadinsisted,andalmostallhisfollowersemphasizedthepoint,thatsomeformof9SomekeyreferencesareJohnArcher,TheLiteratureofBritishDomesticArchitec-ture17151842(Cambridge,MA:MITPress,1985);EileenHarris,BritishArchi-tecturalBooksandWriters,15561785(Cambridge:CambridgeUniversityPress,1990);HowardColvin,ABiographicalDictionaryofBritishArchitects16001840,3rded.(NewHaven,CT:YaleUniversityPress,1995).10MarcusVitruviusPollio,TheTenBooksofArchitecture(lstEnglishtrans.ofPerraultsAbridgement,1692);LeonBattistaAlberti,DeReAedificatoria(Flo-rence,1485;lstcompleteEnglishtrans.,1726);AndreaPalladio,IQuattrolibridellarchitettura(Venice,1570;lstcompleteEnglishtrans.,1715);SebastianoSer-lio,Architettura,(VeniceandParis,153747;lstEnglishtrans.,1611);VincenzoScamozzi,LIdeadellaarchitetturauniversale(Venice,1615;lstEnglishtrans.,1676);GiacomeBarozzidaVignola,Regoledellecinqueordinidarchitettura(Rome,1563;lstEnglishtrans.,1655);RolandFreart,SieurdeChambray,´Parallele`delarchitectureantiqueetdelamoderne(Paris,1650;lstEnglishtrans.,1664).11Marc-AntoineLaugier,AnEssayonArchitecture(London,1755);WilliamCham-bers,ATreatiseonCivilArchitecture(London,1759);RobertMorris,LecturesonArchitecture(London,1734,1736).OntheEncyclopedie,´seeKevinHarrington,ChangingIdeasonArchitectureintheEncyclopedie´,17501776(AnnArbor,MI:UMIResearchPress,1985).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste421expertjudgmentisneededinbothmakingandrespondingtoworks,albeitjudgmentnothide-boundbyrules.Londonwitnessedahugebuildingprogramthroughouttheeigh-teenthcentury,andtheincreasedwealthofEnglishpatronsprovidedopportunitiesfordesignerssuchasThomasChippendale,whowerequicktopublicizetheirwork.ButHumesawlittleofthisuntilafterhehadwrittentheTreatiseandearlyEssays,andhediedbeforemorethanahandfulofeleganthouseshadappearedinEdinburghsNewTown.12Fortyyearsearlier,intheTreatise,hehadstatedthatbuildings,furniture,andutensilsaremadetofulfillspecificfunc-tions,andthattheirbeautyderiveslargelyfromtheirsuccessinthisregard:Mostoftheworksofartareesteemdbeautiful,inpropor-tiontotheirfitnessfortheuseofman,andthebeautyoftables,chairs,scritoires,chimneys,coaches,sadles,ploughs,andindeedeveryworkofartischieflyderivdfromtheirutility(T3.3.1.8,2.2.5.17).13Suchviewshad,infact,beencanvasedbyClaudePer-rault,explicitlyreferringtotheroleoftheassociationofideas.More-over,Perraultalsodeclaredthattheorywasoftenneitheruppermost,norpresent,inthemindsofcraftsmen,andshouldnotthereforebetheobjectofaspectatorsorcriticssearch:assumptionsaboutthe-oryhavebeengeneratedbymenofletters,althoughthereisnogreatobstacletobelievingthatthethingsforwhichtheycanfindnoreasonare,infact,devoidofanyreasonmaterialtothebeautyofthething.14ParalleltosuchviewswerepassagesinCicero,quotedbynumerouswritersthroughouttheeighteenthcentury,totheeffectthatalmosteverybodyisabletojudgeartwithouttheneedofanytheory(DeOratore,3.l.195).Severalissuesoverlappedinsuchdis-cussions.Throughouttheevolvingdebatebetweentheancientsand12HumesownmoremodestNewTownhouse,situatedonwhatisnowcalledSt.DavidStreet,atthesouthwestcornerofSt.AndrewSquare,wasbuiltforhimin17701.Forfurtherdetails,andspeculationaboutthenamingofSt.DavidStreet,seeErnestCampbellMossner,TheLifeofDavidHume(Edinburgh:ThomasNelson&Sons,1954),5626.13Artmeantskill,andthephraseworkofartwasusedwellintothe1750stomeananyworksinvolvingskill;thetermwasnotsynonymouswiththefinearts,whichdenotedpaintingandthedecorativearts.14Ordonnancedescinqespecesdecolonnesselonlam`ethodedesAnciens´(Paris,1683);citedfromthetranslationbyIndraKagisMcEwen(SantaMonica,CA:GettyCenterfortheHistoryofArtandHumanities,1993),58.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n422peterjonesmoderns,muchinkwasspiltonthewideninggapbetweenpracti-tionersandcommentators,betweenprofessionalsandamateurs.Pro-fessionalscausedthingstohappen,andinitiatedchange;amateurs,bycontrast,experiencedandsavoredtheeffectsofsuchchanges,andbecameactiveonlyintheirsubsequentparasiticreflections.Inaddition,however,atransitionwastakingplacebetweenaviewofthefineartsaspleasurableluxuries,usefultoforestallboredom,andaviewofthemastheproductofcitizenswhoseprimarymoraltaskwastobepubliclyuseful.J.-F.Blondel,apracticingarchitectwithanunusuallyextensiveinterestintheory,rejectedafavoritedistinctionoftheamateursbetweenuseandpleasureasameansofidentifyingthearts,andhiscolleagueDenisDiderotclaimedthatspeculationandpratiquewerebothessentialinthesuccessfulexecutionofanyart.15Sinceantiquity,ithadbeenrecognizedthatarchitec-turedifferedfromtheotherarts,andfittedcomfortablyintononeoftheexistingcategoriesofendeavor.Itisaphysicalandsocialphe-nomenon,occupying,ordering,andenclosingspaces:experienceofitinvolvesallthesenses.Albertihadseductivelyarguedthatthemeritofaworkofarchitectureisinseparablefromitsintendeduseorfunc-tion,andmightnotbeexplicableintermsofrules.J.-G.Sulzer,whotranslatedHumeandwasafriendofKant,arguedforarigorouscon-textualisminjudgingarchitecture,andagreedwithAristotlethatinalltheartsparticularrulesarediscoveredandunderstoodbeforeanyformulationofgeneralprincipleswhererulesareunderstoodtobepracticalmeanstodesiredends.Healsoheldthataresolutionbetweenthedemandsofreasonandtastewaspartpsychologicalandpartpolitical.16Humealwaysproclaimedthatliteraturewashisprincipalpassion,andhecertainlyknew,met,orcorrespondedwithalmosteveryoneofnoteinhisdayassociatedwiththearts,inBritainandFrance.Hewaswidelyreadinclassical,English,French,andItalianauthorsandfrequentlyalludestothem.Butitmustbeemphasizedagainthatforhimliteraturewasageneralcategorythatincludedhistoryandphilosophy.Thatexplainswhyheassessesliterature,ofalmostany15DiderotsremarksoccurinhisentryforArtinLEncyclopedieouDictionnaire´RaisonnedesSciences,desArtsetdesM´etiers´(Paris,1751),1.714.16SulzersremarksoccurintheentriesforArchitecture,Art,andEsquisseinSupplement´aLEncyclop`edie´(Amsterdam,17767),1.540,588,595;2.872.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste423kind,asthecoherentexpressionofthought.Moreover,thenotionofjudgment,whichbecameassociatedwiththatofcritic,involvedthreedecisionsabouttheappropriatenessofexpression:namely,tothespeakersstateofmind,totheintendedlistenersowncapaci-ties,andtotheparticularcontext.Eighteenth-centuryBritishandFrenchwritersalikeclaimedJosephAddisonselevenpapers,OnthePleasuresoftheImagination,andaprecedingpaperontaste,fortheSpectatorof1712,assignificantsourcesoftheirtheoreticalideas.HumeadmiredAddisonsskillandsuccessasapopularessay-istand,ofcourse,referstothesepapers,buthedidnotthinktheexplicitLockeanaccountofimaginationusedinthemcouldexplainthephilosophicalissuesraised.OtherinsightsHumeadoptedfromJeandelaBruyere,NicolasBoileau-Despreaux,orBernardleBovier`deFontenelleneededamoresubstantialgroundingthancouldbeprovidedbythepassingreflectionsofAnthonyAshleyCooper,thirdEarlofShaftesbury,orhisavowedchampionFrancisHutcheson.17HumefoundmanyoftheconclusionshewantedtoadoptinaFrenchworkthatforatleastfiftyyearswasthemostinfluentialworkofitskindthroughoutEurope:butwiththeAbbeJean-BaptisteDubos,´Humestaskwastoprovideastrongerandnon-Cartesianphilosoph-icalunderpinningforthoseconclusions,byreferencetohisowntheoryofhumannature.TheAbbeDuboswasSecretaryoftheFrenchAcademyfrom1723´untilhisdeathin1742;hehadbeenafriendoftheHuguenotscholarPierreBayle,whosephilosophicalskepticismhefoundincreasinglycongenialandwhogreatlyinspiredHume.AtthebeginningofthecenturyDuboshadalsohelpedtopublicizeAnEssayconcerningHumanUnderstandingbyhisfriendJohnLocke,inPierreCostesFrenchtranslation.LikeFontenelle,whoseworkheadmired,Duboswasalearnedandculturedman,andhisvolumesaboundinref-erencestoancientandmodernworks,andinallusionstorecent17JeandeLaBruyere,`LesCaracteres`(Paris,1688);NicolasBoileau-Despreaux,Oeu-vresdiverses(Paris,1701);BernardleBovierdeFontenelle,Oeuvresdiverses(LaHaye,1728);AnthonyAshleyCooper,thirdEarlofShaftesbury,CharacteristicksofMen,Manners,Opinions,Times(London,1711);FrancisHutcheson,AnInquiryintotheOriginalofourIdeasofBeautyandVirtue,TreatiseI,ConcerningBeauty,Order,Harmony,Design(London,1725).ForClaudePerrault,seehisOrdonnancedescinqespecesdecolonnesselonlam`ethodedesAnciens´(Paris,1683);forJ.-G.Sulzer,seeSupplement´alEncyclop`edie´.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n424peterjonesscientificdiscoveries.NowhereisthismoreapparentthaninReflexionscritiquessurlaPo´esieetsurlaPeinture,´whichappearedfirstin1719,wentthroughseveraleditions,andwastranslatedintoEnglishin1748.HumereferredtoitinhisEarlyMemoranda(pos-siblyoftheearly1740s),anditsimpactisdiscerniblebothintheEssaysMoralandPoliticalof17412,especiallywhenthetopicisart,andmostdramaticallyinhisOftheStandardofTasteof1757.18(AdamSmith,inthe1780s,alsoadoptedfromDubosasig-nificantportionofhisideasaboutthearts.)Likemanyinfluentialwriters,Duboswasnothimselfveryoriginal:hisskilllayinsyn-thesizingmanyofthecriticalideasintheaireveniftheyweremutuallycontradictoryandintherangeofissuesheidentifiedascallingforanalysisandreflection.TheReflexions,´amongotherthings,isacontributiontothedebatebetweentheAncientsandModerns,Humesinterestinwhichiseverywhereapparentinhis1741Essays.19WritersinEnglandhadimportedthisdebateontherelativemeritsofancientandmod-ernlearningandculturefromFrance,andbytheearly1700sseveralissueswerebeingdiscussed.Theseincludedthenatureofjudgmentsoftasteandtheinfluenceofhistoryandsocietyonsuchjudgments;thelimitsofcriticismandtheroleofrulesinit;thenatureofbeautyandtherespectiverolesofexpertsandthepublicinitsdetermina-tion;overallcomparisonbetweentheAncientsandtheModerns,withparticulardisputeoverthestatureofHomer;andthenatureofprogress,particularlyasrevealedinmoderninstitutionsandprac-tices,andinrelationtoChristianity.Humeexpressedviewsonallthesematters,andalthoughhisherowasinmostthingsCicero,18SeeErnestCampbellMossner,HumesEarlyMemoranda,17291740,JournaloftheHistoryofIdeas9(1948):492518,esp.500;andespeciallyM.A.Stewart,TheDatingofHumesManuscripts,inTheScottishEnlightenment:EssaysinReinterpretation,ed.P.Wood(Rochester,NY:UniversityofRochesterPress,2000),267314.ForafulldiscussionofHumesdebtstoDubos,andthedebatebetweentheAncientsandModerns,seemyHumesSentiments.19SeeP.O.Kristeller,TheModernSystemoftheArts,inJournaloftheHistoryofIdeas,vols.12and13(1951,1952);R.F.Jones,AncientsandModerns:TheBattleoftheBooks(St.Louis,1936);H.Rigault,Histoiredelaquerelledesanciensetdesmodernes(Paris:L.Hachette,1856);CharlesKerby-Miller,ed.,TheMemoirsoftheExtraordinaryLife,Works,andDiscoveriesofMartinusScriblerus(NewHaven,CT:YaleUniversityPress,1950);JosephM.Levine,BetweentheAncientsandtheModerns(NewHaven,CT:YaleUniversityPress,1999).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste425hisadmirationforrecentpoliticalprogressandmaterialadvance,togetherwithhisrefusaltoappealtoauthorityortoChristianity,alignedhimwiththemoderns.Moreover,HumesdebttoDubosgoessomewaytoexplainwhyalmostallofhisownremarksontheartsaresetintheframeworkofoursociallife.Humesrecordedartisticpreferences,togetherwithhiscriticalobservationsonparticularworks,areentirelyorthodoxfortheage,andarelikethoseofhisfriendAdamSmithratheruninterest-ing.Thesereferencesaresecurelyanchoredinclassicalandmodernneoclassicalliterature.In1741heassertedthatVirgilandRacinerep-resentthepeaksofancientandmodernliteraryachievement,andthatFrancepossiblysurpassesancientGreeceinartisticmerit.Atthesamedateheobjectstoanythingfallingundertheeighteenth-centurynotionofGothic,andtoanyexcessornament(E-SR4,193;E-CL6,91).In1757heobjectedtoHomerbecauseofthemoralattitudesrepresented(E-ST31,246),andataboutthesametimehedeclaresthatDonneisguiltyofthemostuncouthexpressionandthatShakespeare,althoughagenius,istoooftentasteless(HEApp.4,5:1512).OnhisjourneytoParisin1763hecarriedwithhimtheworksofVirgil,Horace,Tasso,andTacitushisHomerwastoolarge(HLI:401).Suchflatverdicts,whichappearinhisHistoryofEnglandofthelate1750s,shouldbejuxtaposed,however,withhispatrioticandromanticenthusiasmforJohnHomesDouglas(1755)and,ini-tially,forJamesMacphersonsputativetranslationsofGaelicpoetryinthe1760s.ButHumesbroadnotionofliteraturemustbeunder-lined,sinceheneverdisplayedasmuchinterestinpoetryanddramaasinhistoryandphilosophy.Indeed,noneofhisallusionstoworksofartrevealsspecialknowledge,interest,orinsight,andmanyarederivedsecond-hand,fromtextsaboutthem.In1739,atleast,hisproposedtreatiseoncriticismwouldhavebeenpartofhisoverallaccountofthescienceofman,andwouldnothavetakenitsdepar-turefromparticularconcernwithoneormoreofthemediumsofart,orofindividualworks:wehavenoevidencethatheconceivedofaprojectofthiskindatanytimeinhislife.ii.beautyandjudgmentsofbeautyInATreatiseofHumanNatureHumehardlymentionsthearts,butheattemptstomarkoutthedomainofreasonandsentimentinCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n426peterjonesmattersofbeauty.Althoughhisremarksonbeautyarestrictlysub-ordinatetoother,usuallymoral,concerns,theyarerelevanttohislaterreflectionsontheartsbecauseofhisobservationsonjudgmentingeneral,anddisinterestedevaluationinparticular.Inoutlinehisviewisthatbeautyisanindefinablepowerinobjectsthatcausesapleasurablesentiment(T2.1.8.2);beautyisnotitselfasentiment,norevenapropertydiscerniblebythefivesenses,butratherapropertywhosepresenceisfelt(byasixthorevenseventhsense,asDubosandHutcheson,respectively,said)onlywhenobjectswithcertaindetectablepropertiescausallyinteract,underspecifiableconditions,withmindshavingcertainproperties(E-Sc1318,1646;EPMApp.1.15).20Discussioncanfocusontheobjectinwhichapersontakesdelight,andbyalteringhisperceptionsofit,cansetoffanewcausalchainthatresultsinanewsentiment.Humedistinguishesbeauty,perceptionofbeauty,andjudgmentsofbeauty,butheconcentratesonthelast,furtherdistinguishing,inlinewithShaftesbury,betweenbeautyofform,ofinterest,andofspecies.21Humegivestwocloselyrelatedexamplesofintrinsicbeauty,perceptionofwhichisbarely,ifatall,mediatedbyconcep-tualjudgment.Hesayswemightattendtothebeautyoftheformofsomesenselessinanimatepieceofmatter;orwemightfindthatsomespeciesofbeauty,especiallythenaturalkinds,ontheirfirstappearance,commandouraffectionandapprobation(T2.2.5.16;EPM1.9).22Twoimportantprinciplesoperateinjudgmentsofbeauty:com-parisonandsympathy.Thefirstfunctionsinourclassificationofobjects:Wejudgemoreofobjectsbycomparison,thanbytheir20Dubosprobablyderivedhisnotionofasixthsense(Dubos2.22)fromFontenelle,Entretienssurlapluralitedesmondes´3(1688),butitwascommonplacetociteCicero(DeOratore3.1.195)astheclassicalsourceoftheidea.OnHutcheson,seePeterKivy,TheSeventhSense(NewYork:B.Franklin,1976).21Shaftesbury,Characteristics,ed.J.M.Robertson,2.vols.(Indianapolis:LibraryofLiberalArts,1964),2:1368.22ThereferencesheretonaturalkindsareechoedlaterinImmanuelKant,TheCritiqueofJudgement(1790),trans.J.C.Meredith(Oxford:ClarendonPress,1952),1545.Kantmaintainedthattheempiricalinterestinthebeautifulexistsonlyinsociety,andthatinterestcancombinewithajudgmentoftasteonlyafterithasoncebeenpositedasapureaestheticjudgment.Humewouldprobablyreplythathisownagreementwiththeformerpointwaspartofhisattempttoavoidanythinglikethelatter.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste427intrinsicworthandvalue.Sympathyoperateswheneverwethinkofobjectsinassociationwithpeople;thusoursympathywiththeownerofahouseenablesustoderivepleasurefromtheconve-nienceofhishouse(T3.3.2.4,2.2.8.2,2.2.5.16).Thejustificationofjudgmentsofbeautydependsonthespeciesandnatureoftheobjecttowhichitisattributed.Beautyofutilityisrelativetospecies,whethertheutilitybenefitstheanimalitself,ortheownerofanobject;italsovariesbetweencultures(T2.1.8.2,3.2.1.18,3.3.5.3).Althoughitisdetectedbyasentiment,beautyisasrealascolorandotherallegedlysecondaryqualities;more-over,discussionofitcanbeobjective,howeverdifficulttoachievethismaybe.Threefactorsarenecessarytotheobjectivityofsuchjudgments:theconventionsoflanguage,theuniversalpsychologicalmake-upofhumanbeings,andthepossibilityofpubliclyshareableviewpoints.Everyoneacknowledgedthatreferencetoutilityrequiredtheexer-ciseofjudgment,andHumeemphasizestheimportanceofboththekindofbeautyinquestionandthekindofthingthatissaidtobebeautiful.Everycommunity,Humethinks,agreesonhowtodescribewhatmostconcernsit.Nospecialmysterysurroundstheconventionsgoverningsuchdiscourse,althoughtheirhistoricalandpsychologicaloriginsmaybeobscure;butwithinanysocialgroupitisonlydeviationfromtheconventionsthatcallsforexplanation.Thus,whatcountsasabeautifulplaindependsonaparticularcom-munitysnotionofaplain,andinHumescontextpooragriculturallandalmosteverywhereaplaincannotbebothovergrownandbeautiful.Ifwelimitedourattentiontosomeaspectofitsform,say,itscolor,orthoughtofitassomethingotherthanaplain,wemightthenbeabletothinkofthenewitasbeautiful,butnottofeelittobeso,becausesuchbeautywouldbemerelyofimagination(T2.2.5.18,3.3.1.203,2.1.20.4).Fortherequiredcausalinteractiontooccurbetweenobjectandobserver,forourjudgmentstobeobjective,andforsocialcommu-nicationtotakeplaceatall,itisnecessarytoestablishandagreeonappropriateviewpoints.Themetaphoricalnotionofviewpointherecoversthedescriptionsunderwhichanobjectisconsidered,aswellastheobserversbeliefs,attitudes,andinterests.ItiscentraltoHumespositionthattisimpossiblemencoudeveragreeintheirsentimentsandjudgments,unlesstheychoseCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n428peterjonessomecommonpointofview,fromwhichtheymightsurveytheirobject,andwhichmightcauseittoappearthesametoallofthem.Thisgeneralviewpointisthegroundofthegeneralinalterablestan-dard,bywhichwemayapproveordisapproveofcharactersandman-ners.Andthotheheartdoesnotalwaystakepartwiththosegeneralnotions,orregulateitsloveandhatredbythem,yetaretheysuffi-cientfordiscourse,andserveallourpurposesincompany,inthepulpit,onthetheatre,andintheschools(T3.3.1.30,3.3.3.2).Suchstandardsarerevisable,becausetheyservetheneedsofthecommu-nity,andthoseneedsmaychange.Moreover,generalrulesareoftenextendedbeyondtheprinciple,whencetheyfirstarise;andthisinallmattersoftasteandsentiment(EPM5.42,4.7;T2.2.5.12);onlycloseattentiontothecontextwillenableustodistinguishbetweentheoriginsofaprincipleanditspresentfoundationsinotherwords,ithasahistory,andmayhavechanged(E-OC4,469).Itiscontingent,ofcourse,whichstandardsareacceptedwithinaparticularcontext,sincethejudgmentismadeongroundsofutility;butitisnecessarythattherearesomestandards.Wecanadopttherequiredgeneralpointsofviewonlyinourthoughts,buttheyarenecessarytoallsociallife:twereimpossi-blewecoudevermakeuseoflanguage,orcommunicateoursenti-mentstooneanother,didwenotcorrectthemomentaryappearancesofthings,andoverlookourpresentsituation.Strictlyspeaking,theadoptionofageneralviewpointenablesustocorrectourlanguageratherthanoursentiments;first,becauseourpassionsdonotread-ilyfollowthedeterminationofourjudgment,andchangemoreslowlythantheoperationsoftheimagination;second,becauseoursentimentsarenotinfluencedimmediately,butonlymediatelybyjudgments(T3.3.1.1518).SuchpassagessupporttheviewthatHumeisoneofthefirstBritishwriterstoemphasizethecentralimportanceofcontexttoourcriticaljudgments.Thepassion,inpronouncingitsverdict,considersnottheobjectsimply,asitisinitself,butsurveysitwithallthecircumstances,whichattendit(E-Sc35,172),whileinmanyordersofbeauty,particularlythoseofthefinerarts,itisrequisitetoemploymuchreasoning,inordertofeelthepropersentiment(EPM1.9).Wemayhavetolearnwhatcomplexitiesneedtobeconsidered,butdiscussioncanchangehowwethinkofsomething,andtherebysetoffanewcausalsequenceendinginnewsentimentsandverdicts.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste429iii.earlyessaysIntheEssaysof17412thereareseveraldiscussionsoftheoriginsandsocialdevelopmentofthearts,andHumefrequentlyechoesDubossconclusions,withoutnecessarilyagreeingwiththeimpliedroutetothem.HumeagreeswithDubosthatthefineartscandeveloponlywhengroupsorsocietiesexistbeyondtheconditionsofbaresubsistence,andindeedonlywhenproductionoftheneces-sitiesoflifeexceedsdemand,viewsthatAdamFergusonexplicitlyrejects.(J.-G.SulzerisuniqueamongHumescontemporariesforrec-ognizingthatthechoicesofcolor,size,shape,decoration,ortextureofcontainersordwellings,amongso-calledearlypeoples,mightberegardedasartorincipientart.)23LikeDubos,Humeclaimsthattheartsandsciencesariseonlyamongpeopleswhohavewhathecallsafreegovernment,andsomemeasureofsecurity;moreover,strongrivalstatesstimulateinventionwhilealsocurbingterritorialexpansion.Thoseartsthatrequirepatronagearelikelytoflourishbestinacivilizedmonarchy,whereassciencestypicallyflourishinarepublic,butinbothdomainstherearecyclesinwhichperfec-tionisnecessarilyfollowedbydecline.Thenotionofcyclesreferredlesstoskills,assuch,thantothefactthatbothstylesandcontentcanfalloutoffavorthroughoverfamiliarityorrepetition,orwhenparticularpracticesseemtobeexhausted.Humeholdsthatnothingcanbedonetoalteranyonespersonal,innerconstitution,althoughitexertsacrucialinfluenceoverindividualtaste.Andreasoninghasimportantrolestoplaybecauseitcanmodifythewaysweperceiveanddescribethings.Duboshadechoedancientwritersinemphasiz-inghowphysicalandphysiologicalfactors,suchasaging,affectourcriticaljudgments;bycontrast,Humeemphasizessocialormoralfactors.Hearguesthathappiness,consistinginabalancebetweenaction,pleasure,andindolence,isnecessarytothephysicalwell-beingofindividuals,aswellastothepoliticalhealthofsociety.Dubosarguedthatworksofartraiseartificial,notnatural,pas-sions,andthateveryoneexceptfellowartistsandscholarsreadsworksofartforpleasure.Thecontrastbetweenartistsandschol-ars,ontheonehand,andspectators,ontheother,isimportant.Fellow-artistsareinterestedintechniquesandknow-how,butas23J.G.Sulzer,inhisarticleonArtinSupplement´alEncyclop`edie,´1.588.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n430peterjonesrivalcraftsmenandpotentialcompetitorsforattention,theycannot,inthatframeofmind,adoptaproperlydisinterestedattitude.(Dubosmayhavebeenthefirsttousethisnotioninanexclusivelyaestheticcontext.)Dubosinsiststhatwecanderivesustainedpleasurefromaworkonlyifweunderstanditinsomeway,andtheminimalrequirementisforordrewhichmightbetranslatedasdiscerniblestructure.Thepublic,andnottheself-proclaimedprofessionalcritics,aretheproperjudgesofartbecause,havingnoself-interestinthetransaction,theycanmoreeasilyanswertheprimaryquestionofwhethertheyhavebeenmovedoraffectedbyaworkandthatquestionisnotthetaskofreasonbutofaninternalsensecalledsentiment.24Thetaskofreasonistoidentifythefeaturesofaworkthatcauseuspleasure,andtherebyjustifytheverdictofsentiment.Thesetasksofidentificationandjustificationtypicallybelongtothecritic.Thereareimportantdistinctions,therefore,betweenanartistwhomakes,aspectatorwhoresponds,andacriticwhoexplains.Thepublic,itshouldbeadded,turnsouttobeaprivilegedgroupthathaslearnedthroughexperiencetoexercisecomparativetaste;theyarethetrueamateurs,becausethelearnedconnoisseursareindangeroflosingtouchwiththeverypointofthearts,whichistoplease.DubosinfluencedHumenotonlyinhisreflectionsonthephysi-cal,social,andpoliticalconditionsofthearts,butalsowhenhecametoconsidertheconditionsfortheproperresponsestothem.Thiswillbecomeapparentinthenextsection.BeforeexaminingHumesmostimportantessayonmattersofcriticism,however,briefmen-tionshouldbemadeofashortessayhepublishedatthesametime,underthetitleOfTragedy.There,Humetakesupthefashionabletopicofwhyspectatorsshouldderivepleasurefromrepresentationsoftragiceventsthatinreallifetheywouldabhor.Mostwell-known24ItshouldbeemphasizedthattheambiguityintheFrenchtermsentiment,deploredbyFrenchandEnglishwritersalike,ispreciselymirroredinHumestermsenti-ment,bywhichhesometimesmeansemotion,passion,andsometimesjudg-ment,opinion;indeed,forhim,thetermtreacherouslycoversbothfeelingandthought.AdamSmith,in1762,usedthetermtomeanmoralobservations;seehisLecturesonRhetoricandBellesLettres,ed.J.C.Bryce(Oxford:ClarendonPress,1983),Lecture11,58.Inthesameyear,Kamesdefinedit:Everythoughtsuggestedbyapassionoremotion,istermedasentiment;ElementsofCriticism,2:149.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste431writersinFranceandBritainhadsomethingtosayonthesubject,includingAddison,Dubos,andLordKames.HumecombinestheviewsofDubosandFontenelle,andaugmentsthem.25Heagreeswiththemthatweneverfailtoknowthatweareinatheater,thatalmostanypassionisbetterthannone,andthatalmostanyformofimitationorrepresentationarrestsourattentionandpleasesus.Theclue,hethinks,liesinthemasteryoflanguageandpresentation,whichallowsoneoftheconflictingpassionstobecomedominantovertheother.Theintensehorrorweexperiencefromthestoryisitselfconvertedtosomethingpleasingbytheoverwhelmingplea-surefromthebeautywithwhichitispresented.Theconversioncanoccuronlywhenthestoryisfictional,oratleastisnarratedaboutthepast.Sinceantiquity,ofcourse,theskillwithwhichasculp-tororpainterrepresentedhorror,hadbeenadmiredasinthecaseofLaocoon.Inthefieldofmusic,bythemid-eighteenthcentury,astaplecriterionofmeritconsistedintheresolutionachievedbyaskillfulcomposerinfirstsettinghimself,andthenovercoming,dif-ficultiesachievementsdetectable,ofcourse,onlybyaninformedlistener.By1759itwasbecomingcommonplacetocomplainthat,inVoltaireswordsfromCandide,musictodayisnothingbutthetech-nicalskillofperformingscoresthatarehardtoplay(Chapter25).Inconnectionwithtragedies,however,Humedoesnotmentionthenotionofsympathy,perhapsbecauseitwoulddiminishpleasuretotheextentthatitinducedidentificationwiththesufferers(E-Tr).iv.“ofthestandardoftaste”In1755theEdinburghSocietyforEncouragingArts,Sciences,Man-ufactures,andAgricultureinScotlandhadproposed,butfailed,toawardamedalforthebestessayontaste.AdamSmith,LordKames,WilliamRobertson,AdamFerguson,andHumeweremem-bersofthesociety,aswasAllanRamsay,whohadjustpublishedhisownADialogueonTaste.TheSocietyreneweditsproposalin1756andawardedagoldmedaltoAlexanderGerard:heexploredhisideasfurtherintheAberdeenPhilosophicalSocietyin1758,25Addison,TheSpectator,no.418;Dubos,Reflexionscritiques,´1.1;1.44;HenryHome,LordKames,EssaysonthePrinciplesofMoralityandNaturalReligion(Edinburgh,1751),1.1;Fontenelle,Reflexionssurlapoetique,´36.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n432peterjonesinthepresenceofThomasReid,JohnGregory,andGeorgeCamp-bell,allofwhomlaterwroteonaspectsoftheproblem.Gerardsexpandedsubmissionwaspublished,atHumesurging,in1759asAnEssayonTaste,togetherwiththreedissertationsonthesamesubject,byVoltaire,dAlembert,andMontesquieu;theselastwere,infact,unacknowledgedtranslationsfromentriesunderGoutinvolume7oftheEncyclopedie.´26InthesameyearSmithpublishedTheoryofMoralSentiments,whichincludeshisdiscussionoftaste,whileLordKamespublishedhisdiscussioninElementsofCriticismin1762.Meanwhile,in1757Humehadpublished(inFourDissertations)anessayofhisown,OftheStandardofTaste,togetherwithOfTragedyandtwootheressays;hetoldacorrespondentthathisessayontastewasasubsti-tuteforoneongeometrythatLordStanhopepersuadedhimtowith-draw.Ramsay,likeHume,citesShaftesburyasapointofdeparturewhendiscussingwhethertherecouldbeastandardoftaste,andoffersasociologicalexplanationofchangesinfashion,placinggreatweightonhabitandsocialstatus.HumesownessayiscondensedandadoptsmanyofAbbeDubossconclusions,albeitnowimplicitlysupported´byhisownphilosophicalsystem.Indeed,Humeseekstofindinhumannature,aswellasinsocialpractices,aresolutionfortheprob-lemRamsaylocatedonlyinsocialpractices.GerardsmuchlongerbooktakesitsdeparturefromaHumeanpositionanddiscussesmanyofthesameissues:theneedforattentionandcomparisoninordertoestablishtheendsandmeritofawork;theneedforgoodsense,rea-soning,andmodels;theparallelsbetweentasteandvirtue;theneedtogroundourconclusionsinexperience.Likewise,KamessevenlongerbookclearlyaddressesHumesviews,andincludesunusuallydetaileddiscussionsofarchitecture,gardens,andmusic.Reason,Sentiment,andJudgmentInOftheStandardofTasteHume,ineffect,extendshisreflec-tionsfromAnEnquiryconcerningthePrinciplesofMoralsonthe26AllanRamsay,DialogueonTaste,TheInvestigator322(1755);H.LewisUlman,ed.,TheMinutesoftheAberdeenPhilosophicalSociety17581773(Aberdeen:AberdeenUniversityPress,1990).JamesAdamwrotetoKamesfromRomeaboutthematterin1761anddraftedhisownessayonarchitecturaltheorythefollow-ingyear:theessayisreprintedinFleming(seenote8above).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste433respectiverolesofreasonandsentimentintherealmofvalues.Someso-calledjudgmentsoftasteare,hebelieves,palpablyfoolishandindefensible:thetasteofallindividualsisnotuponanequalfooting,andweshouldnotgiveunrestrictedlicensetotheclaimthatitisfruitlesstodisputeconcerningtastes.Herecognizesthatthosewhointroducesentimentintotheanalysismustneverthelessavoidclaimingthateveryoneisequallyrightinmattersofsenti-ment.Indeed,ifrationaldiscourseiseventobepossible,theremustbesomestandard,rule,orcriterionbywhichdisputescanberesolved.Consequently,Humehopestoshowthatcriticismisafac-tuallybased,rational,socialactivity,capableofbeingintegratedintotherestofintelligiblehumandiscourse,andheattemptstoestab-lishthatsentimentcanbeacriterion.Ofcourse,iftherearerules,whetherofcompositionorcriticism,theymustnotbethoughtofasfixedbyreasoningsapriori,or...beesteemedabstractconclusionsoftheunderstanding(E-ST25,7,6,9;242,230,229,231).Humeholdsthattodiscernthesubtleorthedefiningpropertiesofsomething,apurelypassiveattitudeisnotenough.Theobservermustself-consciouslyattendtotheobjectinquestionand,moreover,beinaproperstateofmindwhendoingso;amerelycausalreactionwillbereplacedbyanappropriatecausalinteraction,towhichtheobserversignificantlycontributes.FollowingDubos,Humestatesthatthreetraitsareneeded:Aperfectserenityofmind,arecollec-tionofthought,adueattentiontotheobject(E-ST10,232).Inarttheproblemiscomplex:Inordertojudgearightofacompositionofgenius,therearesomanyviewstobetakenin,somanycircum-stancestobecompared,andsuchaknowledgeofhumannaturerequisite,thatnoman,whoisnotpossessedofthesoundestjudg-ment,willevermakeatolerablecriticinsuchperformances(E-DT4,6).Therearethreemaincausesoffailuretorespondproperlytoworksofart,andeachoneisderivedverbatimfromDubos:apersonmaylackdelicacy,maylackgoodsense,ormaysufferfromprejudice.Thesearealltransliteratedseventeenth-centuryFrenchtechnicalterms,essentiallyCartesianandcognitiveinforce,famil-iartoallofHumeslearnedcontemporaries(E-ST14,201,234,23940).Humeholdsthatdelicacyoffeelingenablesonetobesensiblytouchedwitheverypartofawork(E-DT2,4):suchdiscriminat-ingperceptionclearlyrequiresjudgmentofsomekind.Cartesianshaddefinedgoodsenseastruejudgmentofsensiblethings,itsroleCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n434peterjonesbeingtoguardagainstfalsejudgment,orprejudice.Humeacceptsthepoint:reason,ifnotanessentialpartoftaste,[not,thatis,thedefiningelement]isatleastrequisitetotheoperationsofit.Toformatruejudgmentacriticmustplacehimselfinthesamesit-uationastheaudienceforwhomtheworkwasconceived,andtowhomitwasoriginallyaddressed.Inotherwords,anapproximationtotheoriginalcontextisafirstconditionofdiscerningwhatwasachieved.Goodsenseattendstofourfeaturesofthecontext:theendsforwhichaworkhasbeencalculated,theeffectivenessofthemeanstothoseends,themutualrelationsofthepartsandofthepartstothewhole,andtheintelligibilityofthewhole.Forexam-ple,theobjectofeloquenceistopersuade,ofhistorytoinstruct,ofpoetrytopleasebymeansofthepassionsandtheimagination.Theseendswemustcarryconstantlyinourview,whenweperuseanyperformance.Inaddition,everykindofcomposition,eventhemostpoetical,isnothingbutachainofpropositionsandreason-ings,sothatintelligibilityiscentral(E-ST21,22,239,240).HumesfriendAdamFerguson,inAnEssayontheHistoryofCivilSociety(1767),agreedthataknowledgeofcontextisindeedessentialforunderstandinghumanachievements,butinsistedthatwetoooftenascribedesign,intention,andtheorywherenoneexisted(inPart3,SectionII,forexample).Theantitheoryviewofmanypractic-ingartistswasacceptedbyseveralphilosophersinbothFranceandBritainthroughouttheeighteenthcentury,andseveralcriticsofHumesHistorywhodisapprovedofhispoliticalinterpretationsalsoobjectedtohisimplausibleattributionofrationalthoughtandplanningtohischosensubjects.Practice,Comparison,InterpretationToovercomefailurescausedbyprejudiceorthelackofdelicacyandgoodsense,twostepsarenecessary(againfromDubos):prac-ticeandcomparison.Practiceisnecessarytoovercomesuperficialfirstimpressions,sinceanyveryindividualperformanceshouldbemorethanonceperusedbyus,andbesurveyedindifferentlightswithattentionanddeliberation.Onlyinthiswaycanwedeterminetherelationofthepartsandtheirrespectivemerits.Likewise,comparisonisessentialbetweentheseveralspeciesandCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste435degreesofexcellence,becausesomeonewhohashadnooppor-tunityofcomparingthedifferentkindsofbeauty,isindeedtotallyunqualifiedtopronounceanopinionwithregardtoanyobjectpre-sentedtohim.Bycomparisonalonewefixtheepithetsofpraiseorblame,andlearnhowtoassigntheduedegreeofeach.Aprej-udicedcriticfailstoplacehimselfinthatpointofview,whichtheperformancesupposes,andobstinatelymaintainshisnaturalposition.Humeinsiststhateveryworkofart,inordertoproduceitsdueeffectonthemind,mustbesurveyedinacertainpointofview,andcannotbefullyrelishedbypersons,whosesituation,realorimaginary,isnotconformabletothatwhichisrequiredbytheperformance(E-ST1821,2379).WeseehereHumesviewthataworkofartisanintentionalact,callingforself-consciousmentalactiononthepartofthespectator.Althoughherarelyusestheterm,Humeisclearlyconcernedwiththenatureandconditionsofinterpretationwhichleadstoanunderstandingofawork(seeEPM9.4);practiceisneededinordertoachievediscrimination,andcomparisoninordertoplaceaworkinitspropercategories.ApassageinSection3ofPhilosophicalEssaysconcerningHumanUnderstandingof1748(laterretitledasAnEnquiryconcerningHumanUnderstanding)isheresignificant.ThereHumearguesthattheprinciplesofhumanagencyarethem-selvesrepresentedin,oratleastleavetracesin,whatwedo;andtheyunderliethespectatorsactsofinterpretation.Thatiswhyhesays,in1757,thesameaddressanddexterity,whichpracticegivestotheexecutionofanywork,isalsoacquiredbythesamemeans,inthejudgingofit(E-ST18,237).ForHume,ourcapacitytounder-standtheworlddependsonourcapacitytounderstandthecausesinoperation.InAnEnquiryconcerningHumanUnderstandingheaskscausalquestionsabouteachoftheissuesheraiseslaterinOftheStandardofTaste:theartistandtheconditionsofcreation,theartproductwhichresults,andtheaudienceandtheconditionsofresponse.Moreover,likeDubos,Humesinterestcentersonbroadlyrepresentationalworksofart,andheaskscausalquestionsaboutthethingsoreventsrepresented,inordertodeterminetheconsis-tencyandintelligibilityoftheworkscontent.Hisinternalquestionsabouttheconsistencyoftheworkitselfpresupposeanswerstotheexternalmatteroftheproperviewpoint,andtherelationsbetweenCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n436peterjonestheworkandotherthingssuchassocietyatlarge,ormorality.HedeclaresthatAsmanisareasonablebeing...heseldomactsorspeaksorthinkswithoutapurposeandintention....Inallcompositionsofgenius,therefore,itisrequisite,thatthewriterhavesomeplanorobject...theremustappearsomeaimorintention,inhisfirstsettingout,ifnotinthecompositionofthewholework.Aproductionwithoutadesignwouldresemblemoretheravingsofamadman,thanthesobereffortsofgeniusandlearning.(EHU3.45)Innarrativecompositions,itisarulethatadmitsofnoexcep-tion,thatthenarratedeventsmustbeconnectedtogetherbysomebondortye,mustformakindofUnity,whichmaybringthemunderoneplanorview,andwhichmaybetheobjectorendofthewriterinhisfirstundertaking.Aboveall,itisnecessarythatsuchworkshaveasufficientunitytomakethembecomprehended(EHU3.617).27TheMindsContributionIntheTreatiseHumearguedthatconsistencyofideasensurestheireasytransition,togetherwiththeemotionsorimpres-sions,attendingtheideas;thenaturalrequirementforsucheasytransitionsliesbehinddemandsforconsistencyoftreatmentandtoneinliterature,andforbalancedfiguresinpaintingandstatuary(T2.2.8.1820,2.2.5.19;EPM6.6).Thedesigns,andprojects,andviewsofmenareprinciplesasnecessaryintheiroperationasheatandcold,moistanddry(T3.1.2.9);theyare,inbrief,theconditionsofhumanagency,andaknowledgeofthemisapreconditionforunderstandingwhatamandoes.Werequirearttobeintelligiblepreciselybecauseitisahumanactivity;durablepleasure,indeed,isdependentonunderstandingintherelevantways(T2.2.4.5,2.3.10.6).Ouraffectionsarearoused,27Thefirstfiveeditions(174860)ofEHUincludedanadditionalparagraphinwhichHumesaidthatitisincumbentoneveryWriter,toformsomePlanorDesign,beforeheenteronanyDiscourseorNarration,andtocomprehendhisSubjectinsomegeneralAspectorunitedView,whichmaybetheconstantObjectofhisAttention.SeetheClarendoneditionofEHU,237.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste437oratleastsustained,onlyiftheactionsofourfellowsdisplaywhatmightbecalledacertaintransparency.Evenif,ashemaintainedinthesecondEnquiry,thefinalsen-tencedependsonsomeinternalsenseorfeeling,whichnaturehasmadeuniversalinthewholespecies,ajudgmentonanyworkofartinvolvesmorethanamerereportofsuchafeeling(EPM1.9;E-ST23,241).Judgmentinvolvesidentifyingthecausesofthepleasurablesentimentand,althoughtheyaredetectableonlyfromcertainviewpoints,thesecausesaretobefoundamongtheproper-tiesoftheworkitself.LikeSanchoskinsmen,inCervantessDonQuixote(2.13),whowerevindicatedbythediscoveryoftheleatherthong(Humegetsthestoryslightlywrong),acriticwhoidentifiesthecausesofhissentimentwillhavejustifiedtheverdict(E-ST16,235);hewillhaveestablisheditsappropriatenessbyestablish-ingitsrepeatablecausalconditions.Duboshadclaimed,in1719,thattheroleofreasonwastojustifythejudgmentthatsentimenthadalreadymade(justifierlejugementquelesentimentaporte´),bydeterminingthecausesofourpleasure;critics,indeed,cantellusthecauseofaneffectonehasalreadyfelt(lacauseduneffetquonsentaitdej´a`)onlyif,onHumesview,theantecedentisiden-tifiable,andtherelationrepeatable.28Reason,therefore,thatis,inductive,experimentalreasoning,isrequisitetotheoperationsoftaste(E-ST22,240)becausethepropersentimentdependsontheproperdiscernment,whichinturninvolvesthinkingoftheworkinparticular,determinate,ways.Criticscanreasonanddisputemoreplausiblythancooks(E-Sc11,163)becausecooksareconcernedsolelywiththephysicalcausesofsensationsandrequireminimalmentalcontributionfromthecustomer,andtheirdishesarenotbearersofmeaning;conceptsstilloperate,ofcourse,andchildrenarenotaloneinrejectingunfamiliardishestheyfailinitiallytoclas-sifyassoup.Bycontrast,criticsrequiresoundjudgmentinordertodiscerntheconsistencyandmeaningofanintentionalhumanper-formance,andinordertounderstandit.Ofcourse,manyancientandmodernwritersinsisted,inShaftesburyswords,thatwemayjudgewellofpoetrywithoutbeingpoetsorpossessingtheleastofapoeticvein.29Itmustbestressed,nevertheless,thatagenuineman28Dubos,Reflexionscritiques,´2.22,32.29Shaftesbury,ALetterConcerningEnthusiasm(1708),inCharacteristics,1.30.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n438peterjonesoftastemustexperienceapleasurablesentimentwhenheattendstoaworkinspecifiedways,becausethatisthesentimenthissignalwhosecauseandjustificationhewishestolocateintheworkitself.Inbrief,thepropersentimentisacomplexresponsetoawork,involvingcausalinteractionbetweenitandaspectatorwhoattendstoitinspecifiableways;acriticstaskincludesthedeterminationofthespectatorsviewpoint,andhejustifieshisverdictsbybring-ingotherstoperceiveandthinkoftheworkinthewayhehas.Itcouldevenbesuggestedthatbyreplacingapurelycausalaccountofresponsebyoneessentiallyinvolvinginteraction,HumerevealsanewcontextinwhichNewtonianthinking,bythemidcentury,haddisplacedtheearlierCartesianmodel.PretendersandBadCriticsWhendiscussingthecausesoffailuretofeelthepropersentiment,Humeraisestwoimportantquestions,althoughhedirectlyanswersonlyoneofthem.First,heaskshowweshouldsilencethebadcriticwhoinsistsonhissentimentinthefaceofcounterevidencetohiscausalclaims(E-ST16,236).Humesansweristhatwemustappealtoparallelcaseswhoserelevancethebadcriticacknowledges.Humesviewisthatintheabsenceofestablishedmodelsandgen-eralprinciples,timewillultimatelyvindicateanycriticaljudgment;someonewithunusualpowersofdiscriminationmayfailtocon-vincehispeers,butweshouldallhopethattruthtriumphsintheend.Hewouldadmit,however,thatevenifanestablishedcriticcanconvincehisaudienceaboutanewcase,weareleftinigno-ranceasindividualsorasunestablishedcritics,becauseself-doubtwilltypicallyaccompanyourfailuretoconvinceothers.Thispointleadstothesecondissue:inthecontextofartcriticism,howcanpretendersbedetected?Incontrasttothebadcritic,whoinsistsonhisownsentiment,thepretendersayswhatothersdosayorarewillingtosay,buteitherexperiencesnopleasure,orderivesitfrompropertiesotherthanthoseheclaimstobethecauseoftheplea-sure.Humesquestionaboutpretendersunjustifiedclaimantsisalegacyofthesearchforacriterionofknowledge,andofdebatesaboutgenuinefaithinreligion.Thequestionforcesustoclarifytheroleofthecalmpassionsinhistheory.If,asispossible,apretendersjudgmentseventuallygainacceptance,theydosoinspiteoftheCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste439factthathedoesnothavethesamesentimentsashislisteners.Butwhatmatterstothem,astoeachoneofus,isnotwhatthepre-tenderpersonallyfeels,althoughtheyhappentobedeceivedaboutthat,butfindingtheviewpointsfromwhichpropertiesoftheworkcanbediscerned.Supportingthisideaisanancientinsight,muchtrumpetedintheeighteenthcentury,thatweallbeginbylearningwhattosayfromothers,andnoonecanbeginbybeingapretender:pretenceisparasiticonknowledge,andknowledgeinprincipleispublic.Humehasalreadyemphasizedthatitisbycomparisonthatwelearnhowtoassigntheepithetsofpraiseorblame(E-ST20,238).Evenapretendermustfirstlearnthesameconventionsashispeers,inordertobeabletocommunicateatall.Inbrief,nohumanbeingcanproclaimspecialauthorityforaself-absorbedreportofhisorherthoughtsabouttheworld,withoutfirsthavinglearnedfromothershowtoformulateandexpressthoughtsofsuchakind:inthisrespect,judgmentsaboutartdiffernotatallfromanythingelse.IntheTreatiseHumeobservedthatinchangingthepointofview,thotheobjectmayremainthesame,itsproportiontoourselvesentirelyalters(T2.2.10.3);andlaterhenotedthataverysmallvariationoftheobject,evenwherethesamequalitiesarepreserved,willdestroyasentiment(EPM5n.17).Sodisagreementsinjudg-mentsmayhavetheirsourceinvariationsbetweentheobservers,orintheobjectsobserved,orintheviewpointsadopted.ButHumeneverdoubtsthepossibilityofreachingagreementondescriptionsofstatesofaffairs,althoughitisanempiricalfactwhetheraparticularcommunitypossessesadequateconventionstoachievethem.Whatsortofpersonisapretender?Ifinnersentimentsplaynoroleindiscussionsofpublicobjects,apretendermustbesomeonewhowantstobeadmiredforhisjudgmentsonthosethings,eventhoughheknowshelacksthenormalpleasurablesentimentsderivedfromattendingtothem.Therearemanypossiblemotivesforwantingtodeceiveothersaboutourinternalstates.Humebelievesthatmenwanttoagreeprimarilybecausetheyaresocialbeings;thebondofsocialsympathyissostrongthatwecanrarelyholdoutagainstthegeneralopinionofothers(T3.3.2.2,2.1.11.2;DP2.10,W4:152).Moreover,asmembersofacommunitywebeginbybeingtoldwhattosay,andonlysubsequentlydiscoveranyinternalaccompanimentstoourutterances.Noonethrustintoanewcultureandsociety,asShaftesburyandothersrecognized,wouldknowwhatjudgmentsitCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n440peterjoneswasappropriatetomake,andatfirstwouldnotevenknowhowtodescribethemattersinhand.Ofcourse,asHumeoftenstates,educa-tioncanleadtopurelymechanicalreactionsthat,inthiscase,mightmeanjudgmentsmadeintheabsenceoftheinnersentimentsthataretheirground(EPM5.3,3.23).Butamancanbecomeapretenderonlywhenhediscoversthathisinternallifediffersfromthoseofhispeers,andbythenhemaywellhaveestablishedalternativeassoci-ationsthatactascues.Humespretenderincriticism,indeed,hasmuchincommonwithacolor-blindman.Both,forthemostpart,cangetbywiththeirlearnedresponses;both,inhisview,sufferfromadefectintheirmentalconstitution,asaresultofwhichtheyarelackingincertainbasicallynaturalreactions(E-Ep6,140);neither,onceweknowtheirhandicap,istobereliedonasourguide.IntheconcludingpagesoftheessayHumecontinuestoechoDubos.Psychologicalfactsaboutindividuals,andsocialfactsaboutcommunities,explainresidualvariationswithinotherwiseagreedjudgments.Thus,Ovid,Horace,andTacitusareallworthyofesteem,althoughatdifferenttimesofhislifeamanmaypreferoneaboveanother.Suchpreferencesareinnocent,becauseitisalmostimpossiblenottofeelapredilectionforthatwhichsuitsourpartic-ularturnanddisposition.Itoftenrequiressomeefforttojudgetheworksofotheragesandcultures,butadverseverdictsshouldberestrictedtothosewhichconfoundtheboundariesofviceandvirtue(E-ST30,31,33;244,2447;seealsoEPM5.15).Humeimpliedamoraldistinctionbetweenabadcriticandapretender,becauseintheworstcases,apretender,settingouttodeceive,maysecureapprovalofwhatisreallyreprehensible.v.conclusionForHume,thenotionsofmistake,error,falsityareassociatedwithacriticsfailuretoconvinceasuitablepeergroup,overasuitableperiodoftime,abouttheallegedfacts;thepossibilityofmistakesentailsthepossibilityofcorrection,andabadcriticisonewhopersistsinhismistakes.Itisaquestionoffactwhothearbitersoftasteareatanyparticulartime,andhowtheygainrecognition,althoughHumebelievesthatmostofusareapttoreceiveamanforwhateverhehasamindtoputhimselfofffor(E-IM2,553).Onthisviewhecouldclaimthatmostcriticsareself-proclaiming.FewCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste441peopleareproperlyqualifiedtobecritics,however,becauseitisraretofindthecombinationofstrongsenseanddelicatesentiment,improvedbypractice,perfectedbycomparison,andclearedofallprejudice(E-ST23,241).SomemoderncommentatorshaveallegedthatHumesaccountisentirelycircular:thestandardoftasteisestablishedby,orrecognizedin,thetruecritics;andthetruecriticsarerecognizedbyadherencetothestandardoftaste.30ButthisisnotHumesposition.Heholdsthatwhenlearningsocialpractices,andtheconventionsthatgov-ernthem,welearnatthesametimewhocurrentlycountsastheexperts,andwhatareacceptedasthebestexamples.Butifthisistruewhenapracticeisalreadyestablished,itdoesnotexplainhowthefirstexpertsinagivenfieldgainrecognition,orhowestablishedjudgmentsaremodifiedor,exceptionally,overturned.Humecannotappealsolelytothepassingoftimesince,byitself,thatdoesnoth-ingtoestablisheithercorrectnessorappropriateness;anditisuse-lessasacriterioninthepresent.Headmitsthateighteenth-centuryresponsestoHomerandtheGreektragediansdifferfromthoseoftheirowntime,buthealsoholdsthat,onceprejudicesareremoved,audiencescanperceivethelastingandtruebeautiesoftheworks.Yetoncethepropertiesofaworkaretrulydiscerned,theendorsementoflatertimesdoeslittlemorethanclarifywhichworksshouldfunc-tionasmodelsatdifferentstagesinatradition.Humedoesnotraisethequestionofwhetherlongadmiredworkseitherareormustbeunderstooddifferentlyatdifferenttimes;norwhethersomeworksadmitofsignificantlyvaryinginterpretations;although,inhisHis-tory,hedoesobserveofpoliticalmattersthatitseemsunreasonabletojudgeofthemeasures,embracedduringoneperiod,bythemax-ims,whichprevailinanother(HE52.66,5:240).Since,atleastin1741,hecanvassedacyclicalviewoftheprogressofthearts,itissurprisingthathedidnotdirectlyanalyzethefactorsinchangingtastes.Ifeveryindividualisconstrainedinhisjudgmentsbythetra-ditionshehasinherited,heisunlikelytowitnessthevindicationofanylarge-scalereassessmentshemayattempt.AndsinceHumealsoholdsthatwhateverpleasescannotbeafault(E-ST9,232),itmaybeaskedwhethersomeonemightnotderivegreaterpleasure30PeterKivy,HumesStandardofTaste:BreakingtheCircle,BritishJournalofAesthetics7(1967):5766.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n442peterjonesfromaworkthat,intheviewofexperts,hehasmisunderstoodormisperceived.AttheendofOftheStandardofTaste,havingconcededthattheexpressionofmerelypersonallikesanddislikesadmitsofnoworthwhilediscussion,Humeassertsthattheoverallmeritofaworkstandsorfallsonanestimateofitsmoralstance,ifthereisone.Threekindsofobservationcanbemadebyacritic,therefore:Ilikethework,Theworkisgoodofitskind,Theworkismorallypraiseworthy.Referencetothestandardoftastecoverstheconditionsforestablishingwhatsomethingis,themodelsagainstwhichitistobemeasured,andthetruemoralstandards.Humeisquiteclearthat,evenifpleasureistheoccasionofsustainedatten-tiontoawork,thecriticstaskistoconcentrateonthework,notonhimself.IntheTreatisehestatedthatourapprobationisimplydintheimmediatepleasure(T3.1.2.3),andinthissenseamanhim-selfneedsnojustificationforthepleasuresheexperiences.Butasasocialbeinghewishestocommunicatethatpleasure,andtoseekreassurancethathedoesnotdeviatemarkedlyfromhispeers.Thishecandoonlybygettingagreementfromothersoverthecausesofshareablepleasures.Humehimself,forexample,admitsdistressatbeingunabletochangehissentimentstoaccordwiththoseofmenherespects;hehopesthatthelong-termverdictofposteritywillvindicatehim(HL2:133).Inanotherletterheremarkedthatweoftenconcealourdislikeofsomethingbecauseofourinabilitytogivereasonsforourverdict(HL1:30).Toexpressourdislikeisoftentopublicizeourdeviationfromtheacceptedevaluation,and,assuch,ourjudgmentcallsforexplanation.Wecanretreat,ofcourse,inthedirectionofourpsychologicalidiosyncrasies,buttodosoistowithdrawouroriginalverdictfrompublicdiscussion.Butifinsteadweadvancetowardaviewpointavailabletoothers,wethenadvancetowardobjectiveverdicts.Humenowherediscussesquestionsaboutthemeaningofworksofart,eventhoughhewasentirelyfamiliarwithlong-runningbat-tlesovertheinterpretationofbiblicalandotherhistoricaltexts.Thisispartlybecausehebelievesthat,inthegeneralcontextofcommu-nication,anydistortionofthenaturalmeaningoftermsissociallyirresponsible.Itissurprising,however,thathedidnotdiscusstheimplicationsofthefactthatnotraditionscanbereliedonuncrit-ically,sincetheycanchangewithinalifetime,andwefrequentlyCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste443projectourpresentconcernsontothepastasAnthonyCollinsandotherfreethinkersforcefullyinsisted.31Itshouldalsobenotedthat,comparedwiththeabsolutenecessitiesoflife,artisoneoflifessuperfluities;thisaside,thehabitofconversingtogether,andofcontributingtoeachotherspleasures,increasesthelevelofbothknowledgeandhumanity(E-RA7,272).Theoverallviewcanbeoutlinedasfollows.Someonebecomesconsciousofpleasureinacertainobject.Heneedsonlyavagueawarenessofitspropertiestobeabletoconcentratemorefully,andtorenderhisfirstobscureandconfusedsentimentclearanddis-tinct(E-ST18,237;T2.3.9.12).Butitisnot,strictly,theoriginalsentimentthatbecomesclear.Attentionrequiresthefullestpercep-tualandintellectualeffort,andthespectatornaturallyseeksastateofequilibriumandconsistency.DAlembert,forexample,under-linestheimportanceofsustainedinterest,andthisnotionclearlyfitsHumesscheme.32For,ifaspectatorcanmakesomesenseofwhatheperceives,hewillexperiencenewsentiments;thesewilldisplacehisearlierones,andmaylooselybedescribedasenhanc-ingthem.Todeterminethefocusofhispleasureshemustappealtopubliclydiscerniblethings,usingtheconventionsofhiscommu-nity,andtheseconventionssecurealltheobjectivityweneedforourjudgmentsandverdicts.Moreover,asasocialbeing,aspectatorwishestoreassurehimselfthathisresponsesresemblethoseofhisneighbors,andtosharewiththemwhatheenjoys.vi.finaleThisessaybeganwiththeclaimthatHumesobservationsonartaresetintheframeworkofsociallife,andthatthisiswhyheconsidersboththemakingof,andtheresponseto,worksofartashumanactionssubjecttotheanalysishehasofferedofotherhumanactions.Inthebroadestsense,worksofartarepleasurablemeansofcommunicationbetweenhumanbeings,andsothepreconditions31AnthonyCollins,AnEssayConcerningtheUseofReasoninPropositions(Lon-don,1707);ADiscourseofFree-Thinking(London,1713);APhilosophicalInquiry(London,1715);ACollectionofTractswrittenbyMr...(London,1717).32JeanleR.dAlembert,Delalibertedelamusique,in´Melangesdeliterature,´dhistoire,etdephilosophie(Amsterdam,1759),4.404.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n444peterjonesofeffectivecommunicationapplytoartasmuchastoothermeans.Certainworkspleaseusbecauseoftheparticularpropertiestheypossess;oneofourtasksistoidentifythesecausesinordertoenableotherstoshareinourenjoyment.Wecannotcomprehendaworkofartmerelybybeinginitspresence,however.Twopropertiesthatbelongtohumanactions,andthataregoalsofourcomprehension,aremeaningandvalue;neitherisdiscerniblebythefiveordinarysensesalone.Themindmustbeactivelyinvolved.Justasinferencebeyondthepresentdataisnecessaryforallfactualreasoning,sointerpretationisnecessarytoestablishthemeaningofwhatanotherpersonhasdone.Humescommitmenttoclassicallearning,andinparticulartotheviewsofCicero,explainshisinterestintheprac-ticeandtheoryofrhetoric.Fromthatdomainhederivedtheviewthatonecan,andshould,askasetofquestionsofanytext,inordertograspitsmeaning:whowroteit?forwhom?aboutwhat?how?when?where?why?Hedidnotsufficientlyconsider,however,howcomplexordifficultsomeofthesequestionscanbe.Nooneinthefirstpartoftheeighteenthcenturyenvisagedthemultipleinterpretationsofworksofartthat,today,arecommon-place.Humesemphasisisprimarilyonthecontextinwhichworksareencountered.Thegreatdanger,asDubossaw,isthatthecriticbecomesaparasite:adisengagednonpractitionerwhopassesjudg-mentandexercisesauthorityoverthelivesofbothmakersandspec-tators,andwhotransformsthecriticalrolefromthatofadispensableintermediarytothatofanindispensableoracle.Atthatstagecriti-cismhasassumedautonomyasapracticeinitsownright,butonethatfundamentallydependsontheagencyofothers.Indeed,fromtheseventeenthcenturyonwardsandprobablyearlierartistsinmanyfieldsdeploredagrowingtendencyintheiraudiencestosub-stitutetalkforclosescrutinyand,intheabsenceofanypracticalknowledgethemselves,toimputetheoriestoartistswhennonehadbeenentertained.33Humesviewsontastewerediscussed,sometimesatconsider-ablelength,byhisimmediatecontemporaries,suchasLordKames,33AlmostacenturyafterPerrault,HumesacquaintanceDr.CharlesBurneyforce-fullymadethesamepoint;seeJones,ItalianOperasandTheirAudiences,326.CharlesBurney,AGeneralHistoryofMusicfromtheEarliestAgestothePresentPeriod(London,17769).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume,theArts,andTheStandardofTaste445AdamSmith,EdmundBurke,AlexanderGerard,ThomasReid,GeorgeCampbell,andHughBlair.HumesclosefriendAdamSmithacceptedmostofhisviewsonthearts,andthoseofDubos,butunderlinedtheimplicationofHumesemphasisoncontextbystress-ingthatouroverridingconcerniswiththemeaningofworks.34ImmanuelKantundoubtedlyconceivedthefirstpartofhisCritiqueofJudgmentasaresponsetoHume,whoseessayontasteSulzertranslatedforhim.OnthecontinentofEuropeafewwritersdur-ingthenineteenthcentury,inRussia,Hungary,Italy,andGreece,forexample,becameinterestedinHumesviews,oftenthroughthemediationnotonlyofKant,butalsotheworksofHughBlair,Kames,andarticlesinencyclopedias.35Afteragapofalmost200years,sev-eralWesternphilosophershaveturnedtoHumesessay;theirworkisbestdescribedasinspiredbyHume,ratherthanasthoughtplau-siblyascribabletoHumehimself.36Humesownexperienceoftheartsincreaseddramaticallyfromthe1760s,butwehavenoevidencethathelaterrevisedanyofhisearlierthoughtsabouttheartsortaste.Althoughheregardedphilosophyasanchoredinandreferrablebacktodailylife,weregrettablydonotknowhowheexpectedhisviewsontaste,inparticular,tobeabsorbedintoeithercreativeorcriticalpractice.TheprecisenatureofthecontextinwhichHumewaswriting,anditsprofounddifferencesfromourown,areusuallyignoredbymoderncommentatorsonHume,whoseenoneedtoanchortheir34ForananalysisofSmithsviews,seemyTheAestheticsofAdamSmith,inAdamSmithReviewed,ed.P.JonesandA.S.Skinner(Edinburgh:EdinburghUniversityPress,1992),5678.35SeeTheReceptionofDavidHumeinEurope,ed.P.Jones(London:ThoemmesContinuum,2005);PeterJones,WhyBotherwithHumeandKamesonCriticismandtheArts?inInstructionandAmusement,ed.E.MazzaandE.Ronchetti(Padova:IlPoligrafo,2005),2744.36Someanachronismmaybeinevitableintheinterpretationofpasttextsandcon-texts;seePeterJones,PhilosophyandtheNovel(Oxford:ClarendonPress,1975).Mosttwentieth-andtwenty-first-centuryphilosophersreflecttheirviewofphilos-ophyitselfbyusingtextsfromthepastfortheirownpurposesintheirowncon-texts.Forvariationsinthisapproach:MaryMothersill,BeautyRestored(Oxford:ClarendonPress,1984);GeorgeDickie,TheCenturyofTaste(Oxford:OxfordUni-versityPress,1996);PaulGuyer,TheStandardofTasteandtheMostArdentDesireofSociety,inPursuitsofReason;EssaysinHonorofStanleyCavell,ed.T.Cohen,P.Guyer,andH.Putnam(Lubbock:TexasTechUniversityPress,1993),3766.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n446peterjonesownreflectionsinpracticalknowledgeofoneormoreofthearts.Moreover,itisoftenclaimedthatphilosophymustnotbecon-flatedwiththehistoryofideas,andthattheessentiallogicofanargument,whateverthecontingenciesofitsexpression,iscontext-independent.Howevermuchhisownpracticevaried,Humehimselfisunlikelytohavesympathizedwithsuchaview.ThosewhofindKantianaestheticslargelyconvincingaskwhetheranempiricistaes-thetics,alongHumeanlines,isevenpossible,withinsomespecifiedcontext.Therolesofimaginationandtheemotions,inbothcreationandresponse,whichHumedidnotexploreindetail,stillrequiresat-isfactoryanalysis;theissuesofinterpretationandmeaningoccupymanymoderntheoristsoflanguageandcriticism,asdothecom-plexsocialcontextsinwhichworksofartaremade,recognized,andassessed,andthatacceleratechangesinthecentralconceptsthem-selves.Everyreader,however,shouldaskwhat,intheiroptimisticmoments,Humeandhiscolleagueshopedmightorcouldhappenasaresultofpublishingtheirviewsinthewaytheydid.suggestionsforfurtherreadingOftheworkscitedinthenotestothisessay,thefollowingareespeciallyrecommended.Dickie,George.TheCenturyofTaste:ThePhilosophicalOdysseyofTasteintheEighteenthCentury.Oxford:OxfordUniversityPress,1996.Guyer,Paul.TheStandardofTasteandtheMostArdentDesireofSoci-ety.InPursuitsofReason:EssaysinHonorofStanleyCavell,editedbyT.Cohen,P.Guyer,andH.Putnam.Lubbox:TexasTechUniversityPress,1993.3766.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\ndavidwootton14DavidHume:TheHistorianThefirstvolumeofHumesHistoryofEngland,dealingwiththeearlyStuarts,appearedin1754.Thefinalvolumes,coveringtheperiodfromtheinvasionofJuliusCaesarto1485,appearedin1762,althoughHumewasoccupiedwithrevisionsofthewholeworkuntilhisdeath.1Inwritinghistory,Humewaspartlycreating,partlyrespondingto,anewmarket.In1757hethoughthistorythemostpopularkindofwritingofany(HL1:244).In1770hewrote,IbelievethisisthehistoricalAgeandthis[Scotland]thehistoricalNation(HL2:230).Heknewofnofewerthaneighthistoriesthatwerecurrentlybeingwritten.Theyearbefore,inEngland,hehaddeclared,History,Ithink,istheFavouriteReading(HL2:196).HumeandhisfriendWilliamRobertsonwereinlargepartresponsi-bleforthisnewpopularityofhistory,muchofitwrittenbyScots-menforEnglishreaders.2HumehimselfhadreceivedunprecedentedpaymentsforhisHistory(forwhichhesoldthecopyrightoneachvolume,ratherthancollectingroyalties):hemadeatleast£3,200onthewhole,atatimewhenafriendofhiscouldconsiderhimselfwell1MytitlecomesfromtheentryforourDavidHumeintheBritishLibrarycatalog,which,tothepuzzlementofgenerationsofphilosophers,distinguisheshimfromothersofthesamenamebyidentifyinghimasthehistorian.2RobertsonsHistoryofScotlandappearedin1759.ForarecentdiscussionofRobert-son,seeM.Fearnley-Sander,PhilosophicalHistoryandtheScottishReformation,HistoricalJournal33(1990):32338.HumeandRobertsonhadalowopinionoftheirchiefEnglishcompetitor,TobiasSmollett,whoseHistoryofEnglandappearedin1757(HL1:302).WhenGibbonpublishedtheDeclineandFall,HumeexpressedsurprisethatanEnglishmancouldhavebeencapableofsuchawork.HeexpectedGibbontounderstandthatitwasthehighestofcomplimentstosaythatthebookhadbeenwellreceivedbyalltheMenofLettersinthisPlace[Edinburgh](HL2:30912).447CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n448davidwoottontodoon£80perannum(HL1:193,255,266,314).Althoughinprac-ticehisHistoryseemstohavesoldlesswellduringhislifetimethanthevariousvolumesofhisEssays,Humewasconsistentlyoftheopinionthatthiswashisbooksellersfault.Themarketforhistorybookswaspotentiallyfarlarger(HL2:106,229,233,242).i.thehistoricalageHumewasthebeneficiaryofanimmenseexpansioninthereadingpublicthattookplaceinthemid-eighteenthcentury.3Muchofthatpublicwasmiddleclassandfemale,andperhapsitsfavoriteread-ingwasthenewgenrethatseemstohavebeeninventedespeciallyforit,thenovel:RichardsonsPamelaappearedin1740,Clarissain1748.4Asearlyas1741,however,Humehadfeltthatthehistorybookcouldsupplantthenovel.ThereisnothingwhichIwouldrecommendmoreearnestlytomyfemalereadersthanthestudyofhistory,hewrote,asanoccupation,ofallothers,thebestsuitedbothtotheirsexandeducation,muchmoreinstructivethantheirordinarybooksofamusement,andmoreentertainingthanthoseseriouscompositions,whichareusuallytobefoundintheirclos-ets.Onlyawomanwhowasacquaintedwiththehistoryofherowncountry,andwiththoseofGreeceandRome,couldengageincon-versationthatcanaffordanyentertainmenttomenofsenseandreflection.Moreover,historyprovidedthebestwayofbecomingacquaintedwithhumanaffairs,withoutdiminishingintheleastfromthemostdelicatesentimentsofvirtue.Thesamecouldnotbesaidfortheactualpracticeoftheworld,orforpoetry,sincepoetsoftenbecomeadvocatesforvice.Philosophy,foritspart,couldrarelymovethepassions;itishistorianswhoarethetruefriendsofvirtue(E-SH1,5,7,563,5667).Itwasasafriendofvirtue,onewritingincompetitionwithnov-elists,thatHumedeliberatelysoughttomovehisaudiencetotearsbyhisaccountoftheexecutionofCharlesI(HL1:210,222,also3JeromeChristensen,PracticingEnlightenment:HumeandtheFormationofaLit-eraryCareer(Madison:UniversityofWisconsinPress,1987).4WedonotknowwhatHumethoughtofRichardson.In1773hethoughtLaurenceSternesTristramShandythebestBookthathasbeenwritbyanyEnglishmanthesethirtyyears(HL2:269).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian449344;MOL11).5Hewascertainlysuccessful:indeed,wehavelet-tersfromfemaleadmirerstestifyingtohowhishistoryhadmovedtheirpassions.Oneofthemtoldhimthatshehadneverhadsuchagoodopinionofherselfaswhenreadinghishistory:evidentlyHumehadinspiredvirtuoussentimentsinher,andtherebymadeherfeelvirtuousandadmirable(HL2:347,3667).6Humethusearlysawacentralroleforhistoryincontemporaryculture.Hehadhimselflongthoughtofwritinghistory,andhewasevidentlyspurredonbytheconvictionthatthereisnopostofhonourintheEnglishParnassusmorevacantthanthatofHistory.PrevioushistoriansevenPauldeRapin-Thoyras,themostwidelyadmiredofthemhadlackedstyle,judgement,impartiality,care(HL1:170).7Onlyacontemporaryhistoriancouldhopetomeetthefirsttest.In1741Humewasabletowrite:Thefirstpoliteprosewehave,waswritbyaman[Swift]whoisstillalive(E-CL8,91).SincetherewerenocontemporaryBritishhistoriansofnote,Humehadtolookabroadforhismodels.HoraceWalpolethoughtHumesstylewasinfluencedbyVoltaire,butVoltairesSiecledeLouisXIV`didnotappearuntil1751.8ThemodelsHumehimselfreferstoweremoredistantintime.HetellsushewaswritingafterthemanneroftheAncients,atleastinthathewaswritingaconcisenarrative(HL1:170).TheAnnalsofImperialRome(c.a.d.117)byGaiusCorneliusTacitusweresoontobehismodelforworkingbackwardsratherthanforwardsintime(HL1:378),anditisprobablyfromTacitusthathetookhishabitofexplainingwhatdifferenttypesofpeoplethoughtabouteventsatparticularmomentsofcrisis:Hume5J.C.Hilson,Hume:theHistorianasManofFeeling,inAugustanWorlds,ed.J.C.Hilson,M.M.B.Jones,andJ.R.Watson(Leicester:LeicesterUniversityPress,1978),20522.6HumesHistoryofEnglandwastranslatedintoFrenchbyOctavieGuichard,MadameBelot,andpublishedinAmsterdamin1763.TheleadingWhigreplytohisHistorywasCatherineMacaulayseight-volumeHistoryofEnglandfromtheAccessionofJamesItothatoftheBrunswickLine(176383);seeNHL,802.HumeandMacaulayarecomparedinNatalieZemonDavis,HistorysTwoBod-ies,AmericanHistoricalReview93(1988):130.7ThefirstvolumeofRapinsHistoiredAngleterreappearedin1724,followedbyanEnglishtranslationin1725.8InthisCountrey,theycallmehisPupil,andthinkthatmyHistoryisanImitationofhisSiecledeLouisXIV.ThisOpinionflattersverymuchmyVanity;buttheTruthis,thatmyHistorywaspland,&inagreatmeasurecomposd,beforetheAppearanceofthatagreeableWork(HL1:226).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n450davidwoottonthustookoveraversionofthefictitiousspeechesbelovedofancienthistorians.In1758HumeapologizedtoWalpolefornothavingprovidedfoot-notereferencestohissourcesinthevolumesontheStuarts:itwasadefecthewastoremedywhenherevisedthem(HL1:316,379).Hisapologyprovidesuswiththebestguidetohismorerecentmodels:Iwasseducdbytheexampleofallthebesthistoriansevenamongthemoderns,suchasMatchiavel,Frapaolo,Davila,Bentivoglio;with-outconsideringthatthatpractice[ofprovidingreferences]wasmoremodernthantheirtime,andhavingbeenonceintroducd,oughttobefollowdbyeverywriter(HL1:284).9(Machiavelli,theold-est,wasbornin1469;Bentivoglio,theyoungest,diedin1644.)FraPaoloSarpi(15521623)inparticularwassingledoutforpraiseintheHistory:hisaccountoftheCouncilofTrentwasanadmirableexampleofhistoricalwriting,soeffectivethattheRomanCatholicchurchwould,Humethought,neverdarecallanothergeneralcoun-cilwhiletherewasaprospectofsuchahistorianbeingalivetowriteaboutit(HE39n.F,4:3889).Sarpihadthussingle-handedlychangedthecourseofhistory:Humecouldhopenomoreofhisownundertaking.10NiccoloMachiavelli,Sarpi,EnricoDavila(authorofthe`HistoriadelleguerrecivilidiFrancia[1630]),andGuidoBentivoglio(authorofDellaguerradiFiandra[16329])havemoreincommonthanthattheyareallsixteenth-centuryItalians,andallinfluencedbyTacitus.Theyhadalltakenanactivepartinpoliticallife,andhadwrittenabouteventswithinthememoryoftheircontemporaries.Forthem,historywasapracticalaccountofhowpowerpoliticsfunctioned,and9ThestrikingpeculiarityamongstthislistofthemodernsistheomissionofFrancescoGuicciardini,Machiavelliscontemporary,andhissuperiorasahisto-rian.ButHumewasfamiliarwithGuicciardiniswork,andmentionshimonthreeseparateoccasionsinAnEnquiryconcerningthePrinciplesofMorals,andonceintheTreatise.10OnSarpi,seemyPaoloSarpi:BetweenRenaissanceandEnlightenment(Cam-bridge:CambridgeUniversityPress,1983).Those,suchasRichardH.Popkin,whothinkthatoneofHumesmajorachievementswastobreakwithprovidential-istandprophetichistory,attributetoHumeatransformationthatHumehimselfwouldprobablyhaveattributedtolateRenaissanceItaly.SeeHume:Philosoph-icalversusPropheticHistorian,inDavidHume,Many-SidedGenius,ed.K.R.MerrillandR.Shahan(Norman:UniversityofOklahomaPress,1976),8395.ThestandardworkonthesubjectisE.Cochrane,HistoriansandHistoriographyintheItalianRenaissance(Chicago:UniversityofChicagoPress,1981).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian451anindispensableeducationforthosewhoplannedtoparticipateinpoliticallife.Theirnaturalaudiencewascomposedofthemembersofthepoliticaleliteall,ofcourse,men.Inordertomakehimselflikesuchmen,Humehadsetoutin1748toacquireaknowledgeofcourtsandcampsbyenteringtheserviceofGeneralSt.Clair(HL1:109;seealsoNHL23).ButHumesHistory,asitwasfinallywrit-ten,placeslittleemphasisoncourtintriguesandmilitaryprowess.Itseesevents,aboveall,fromthepointofviewoftheintelligentspectator,nottheparticipant,justasitsreadersareexpectedtobespectators:AddisonsSpectatorhadprovidedthemodelforsomeofHumesmoralandpoliticalessays,andhaddonemuchtoeducatethenewpolitepublicthatHumesoughttoaddress.IfHumehadamodelamonghistorians,itislikely,onceagain,tohavebeenSarpi:foriftheaudiencetowhichSarpidirectedhisHistoryoftheInqui-sitionwasoneofVenetianpoliticians,theintendedaudienceoftheHistoryoftheCouncilofTrent(1619)wasmuchwider.HemadethenotionthatageneralcouncilexpressedthewillofGodincredibletoaneducatedpublicwhodidnotparticipateinchurchpoliticsbutwishedmerelytodecidewhattobelieveaboutthechurch.HumesHistorysoughttoaddressanewaudience:anaudienceneitherofpoliticiansnorofantiquarians,butofthosewhoaspiredtoparticipateinpoliteconversation.Thisgavehistoryanewrole:thatofretellingastoryalreadytold.Wetakeitforgrantedthattherewillbemorethanonehistorydealingwithaparticularsubject,buttothoselivingintheeighteenthcenturytherewassomethingnovelaboutthisidea.AsHumesettoworkontheHistory,CharlesRollinhadrecentlypublishedhisHistoireRomaine(173841):thiswasthefirstattempttoretellthehistoryofancientRomeforamod-ernaudience.11UntilthenithadalwaysbeenassumedthatthosewhowantedtolearnaboutRomanhistorywoulddosothroughadirectreadingofthegreathistoriansofancientRome,GaiusSal-lustiusCrispus(Sallust:c.8635b.c.),TitusLivius(Livy:c.59b.c.a.d.17),andTacitus(c.a.d.56120).ThisassumptionwasalsoheldaboutEnglishhistory.InthesamelettertoWalpole,HumedefendedhimselfagainstaviewthathefeltWalpoleratherinsinuatedthanadvancd;asifitweresuperfluoustorewritetheEnglishhistory,orpublishonthatsubjectanythingwhichhaseverbeforeinanyshape11AnEnglishtranslationbegantoappearin1739.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n452davidwoottonappeardinprint.Humeinsisted,first,thattherewastoomuchmaterialforanyonebutaprofessionaltocometogripswithitall:Theoriginalbooks,whichinstructusinthereignofQ.Elizabethalone,wouldrequiresixmonthsreadingattherateoftenhoursaday(HL1:285).Thehistorianthus,inpart,providedaprecis,and´itwaspartlyasaprecisofEnglishhistorythatHumesvolumeshad´suchastoundingsuccess:sevencompleteeditionsduringhislife-time,and175inthecenturyafterhisdeath.12Why,though,beyondthefactthatHumewasmoreconcise,readHumeontheCivilWarratherthanthegreathistorianandstatesmanEdwardHyde,EarlofClarendon(160974)?ItwasClarendon,notHume,whoembodiedthetraditionalidealofthehistorian:anintelligentparticipantintheeventshehimselfdescribed.Humesresponsewasthatmostpeoplewhoreadtheoriginalsourceswoudattainbutaveryconfusdideaofthetransactionsofthatperiod.Theproblemwasnotmerelytheabundanceofmaterial,butthedifficultyofmakingsenseofit:Toallegethereforethenumberofhistoricalmonumentsagainstcom-posingahistoryseemsnotmuchbetterfounded,thanifoneshoudgiveitasareasonfornotbuildingahouse,thathelaynearaquarry(HL1:285).Thustheaccountsofcontemporarieswerenolongertoberegardedasself-sufficientnarratives,buttobetreatedasmeresources,aquarryforthemodernhistorian.Itwaswrongtothinkthathistory,oncewritten,needneverberewritten.AttheheartofHumesundertakingwas,therefore,anovel,andlargelyunstated,conceptionofprogressinhistoricalknowledge.Humeturnedtothewritingofhistorypartlyindeliberatepursuitofliteraryfame,though,despitewhathesaysinhisautobiography,thiscanscarcelyhavebeenhisoverridingmotive,forhisEssayshadalreadywonhimrenown.Justasimportantwashisdesiretoexplore,throughthewritingofahistoricalnarrative,philosophical,political,andmoralquestionsthatlayattheheartofhispreviousenquiries.1312DavidBerman,DavidHumeonthe1641RebellioninIreland,Studies:AnIrishQuarterlyReview65(1976):10112,110;DavidHume:PhilosophicalHistorian,ed.D.F.NortonandR.H.Popkin(Indianapolis:Bobbs-Merrill,1965),109,41317.13ThepointiswellmadeinDavidFateNorton,HistoryandPhilosophyinHumesThought,inDavidHume,PhilosophicalHistorian,xxxii1,whichisparticularlyhelpfulonthelinksbetweenHumesskepticismandhistoricalstudy.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian453ii.historicalcriticismWhatdidHumehavetotellhisreadersthatwasnew?BeforewelookatthesubstantivecontentoftheHistory,weneedtounderstandinwhatsenseHumewasaphilosophicalhistorian.14OneHumescholarhasrashlyclaimedthattherewasnophilosophyofhistoryintheeighteenthcentury.15Infact,therewasawell-establishedliteratureonhistoricalmethodology,onethatdealtwithphilosoph-icalproblemscentraltoHumesowninterests.ThefoundingtextforthisliteraturewasLalogique,oulartdepenser,commonlyknownasThePort-RoyalLogic,writtenbytwoJansenisttheolo-gians,AntoineArnauldandPierreNicole,andfirstpublishedin1662.16InthatworkArnauld(forheseemstohavebeenprimarilyresponsibleforthechaptersofinteresttous)hadinventedmodernprobabilitytheorybyarguingthatanumberofdifferentactivitiesgamesofchance,theauthenticationoflegaldocuments,beliefin14Philosophicalhistoryissometimesidentifiedwithconjecturalhistory,aterminventedbyDugaldStewart(17531828),andonethathetooktobeidenticalwithHumestermnaturalhistoryasusedinTheNaturalHistoryofReligion.Sincethisworkliesoutsidethescopeofthepresentchapter,Ihavenotdiscussedcon-jecturalhistoryhere.See,forexample,H.H.Hopfl,FromSavagetoScotsman:ConjecturalHistoryintheScottishEnlightenment,JournalofBritishStudies17(1978):1940.IalsodonotdiscussheretheviewthatHumesinsistenceintheTrea-tiseontheuniformityofhumannaturemeantthathisoutlookwasfundamentallyahistorical:see,e.g.,L.Pompa,HumanNatureandHistoricalKnowledge:Hume,Hegel,andVico(Cambridge:CambridgeUniversityPress,1990).Pompaisremark-ableforfailingtograspthatHumesconceptofhistoricalknowledgeisintendedtobearesponsetoscepticalarguments(289,33),andforfailingtoconsiderHumesactualpracticeasahistorian.15DonaldW.Livingston,HumesPhilosophyofCommonLife(Chicago:UniversityofChicagoPress,1984),211.ItmaybeadmittedthattherewaslittlephilosophyofhistoryinEnglish:characteristicallyindifferenttophilosophicalquestionsis[P.Whalley],AnEssayontheMannerofWritingHistory(London,1746).AgoodguidetothewayinwhichhistoricalwritingwasconceivedwhenHumewaswritingtheHistoryarelectures1720ofAdamSmithsLecturesonRhetoricandBellesLettres(17623),ed.J.C.Bryce(Oxford:ClarendonPress,1983).Smith,whowasaclosefriendofHumes,wasawarethatmodernhistorywasdifferentfromancienthistorybecauseitwaspreoccupiedwithprovingmattersoffact,butdisapprovedofthisbecauseitinterruptedthenarrative(102).Healsocontinuedtothinkthatthebesthistoriansweregenerallythosewhowroteabouteventsofwhichtheyhadpersonalexperience(934).16OnthesignificanceofthisworkforHumesphilosophyofhistory,seemyHumesOfMiracles:ProbabilityandIrreligion,inStudiesinthePhilosophyoftheScot-tishEnlightenment,ed.M.A.Stewart(Oxford:ClarendonPress,1990),191229.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n454davidwoottonmiracleswerephilosophicallyrelatedbecausetheyallinvolvedjudgmentsofprobability.Courtshadestablishedrulestodeterminewhichdocumentscouldbeadmittedinatrialbecausetheywantedtominimizethechanceoffraud,misrepresentation,anderror.His-torians,whentheyreportedthatAugustinehadwitnessedamiracle,wereagreeingtogivecredencetoAugustinestestimonydespitetheinherentimprobabilityoftheeventsdescribed:Augustinesgoodcharacterandintelligenceeliminatedthepossibilityoffraud,mis-representation,orerror.Arnauldthussoughttoformulatetherulesthatshouldgovernthecriticismofsourcesbyapplyingprobabilitytheory.Crucially,heseparatedjudgmentoftheactuallikelihoodofaneventinitself(heagreedmiracleswereimprobableinthemselves)fromjudgmentofthequalityofthetestimony(goodtestimonycouldmakeanimprob-ableeventlikely).Fromthisanalysistherederivedaseriesofworksthatsoughttoformulatetherulesthatshouldbeemployedinassess-inghistoricaltestimony,andthatinsistedthatseveralindependentwitnessesshouldbegivencredenceiftheycontradictedoneisolatedindividual,andthatdirectparticipantsandeyewitnessesweretobebelievedwhentheycontradictedthosewhoreliedonsecond-handinformation.17Animmediateconsequenceofthiswasadowngrad-ingofthereliabilityoforaltradition:apointmadewithparticu-larforcebyProtestantssuchasJohnLocke,sinceitunderminedtheCatholicclaimthattheologicaltruthsarebasedintraditionalknowledge.18Thenew(largelyFrench)literatureonhistoricalsourcecriticismthrewupanumberofkeyproblems,allofwhichHumesoughttoaddress.Firstofall,howcouldweclaimtohavefirst-handknowledgeoftheexistenceofJuliusCaesarwhenthesourceswereliedonwereinfactcopiesofcopies?DebelloGallicowasonlyinappearanceaprimarysource:suchaworkwasinfactasecondaryauthority,sincethereaderhadtotrustinthefidelityofgenerationsofprinters,and,17C.Borghero,Lacertezzaelastoria(Milan:FrancoAngeli,1983),givesanexcellentaccountoflate-seventeenth-andearly-eighteenth-centuryphilosophiesofhistory.18DanielWoolf,TheCommonVoice:History,FolkloreandOralTraditioninEarlyModernEngland,PastandPresent120(1988):2652;andmyHumesOfMiracles,198,2223.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian455beforethem,ofscribes.Forancienthistorytherewere,onthisview,virtuallynoprimarysourcesatall.IntheTreatiseHumeinsistedthatthisskepticalargumentwasspecious:scribesandprintersweregenerallytobetrusted(T1.3.13.6).Themostimportanttheoristinthisfield(andonewhowasalmostcertainlyknowntoHume)wasNicolasFreret,whose´Reflexionssurl´etudedesancienneshistoires´etsurledegredecertitudedeleurspreuves´(1729)wasaresponsetotheskepticalargumentsofJ.LevesquedePouillys´ADissertationupontheUncertaintyoftheHistoryofthefirstfouragesofRome.19FreretarguedthatwecouldtrusthistorianslikeLivywhenthey´reportedeventsthatwecouldnolongerconfirmbyappealingtoindependentsources,fortheyhadaccesstosourcesnowlosttous.20Therewasnothinginherentlyimpossibleaboutknowledgeofthepast.Tomakethisclaim,Frerethadtojustifyacriticalreadingof´Livy.Livyfreelyreportednaturalprodigiesandevenmiracles.Freret´arguedthatwecouldbeskepticalofthefirstandincredulouswithregardtothesecondbutstilltrustLivysaccountsofmilitaryandpoliticalconflicts.InOfMiracles,Humesetouttojustifyasystematicskepticismwithregardtoallreportsofmiraculouseventsthroughanovelapplicationofprobabilitytheory.OriginallyintendedforpublicationintheTreatise,thisessaywouldhavebal-ancedhisdefenseofthereliabilityofoursourcesfortheknowl-edgeofancienthistory,parallelingFreretstreatmentofthesubject.´Hisargument(whichwaseventuallypublishedintheEnquirycon-cerningHumanUnderstanding)proceededbyimaginingamiracleforwhichtherewasapparentlyunimpeachabletestimony.Yetonewouldhavetobalancethistestimonyagainsttheinherentunlike-lihoodofaneventcontrarytotheconstantcourseofnature.Thusexternalandinternalevidencewouldseemalmostperfectlybal-anced.Takeintoaccount,though,thefactthatthosewhoadvocatedabeliefinmiracleswereself-interestedwitnesseswhoweretryingtowinadherentstotheirowncause,andatonceitbecomesclear19IanRosstellsmethatitwasprobablyAdamSmithwhoarrangedfordePouillysessay,whichfirstappearedinFrenchin1729,tobepublishedinEnglish,inThePhilologicalMiscellany,vol.1(1761).20SeemyHumesOfMiracles,2003.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n456davidwoottonthattheirtestimonymustnotbetakenatfacevalue.Theinternalevidencemustalwaysoutweightheexternal,andonecouldonlyproperlybelieveinamiracleifthetestimonyinfavorofitweresostrongthatitwouldtakeamiracleforittobefalsethatis,onlyifthealternativetobeliefinamiraclewasbeliefinagreatermiracle.InhislongandbrilliantessayOfthePopulousnessofAncientNations(firstpublishedin1752),Humepresentedatourdeforceinthecriticismofsources.Regularlyonefoundapparentlyreli-ableancientsourcesreportingcitieswithenormouspopulationsandbattlesinvolvingarmiesofextraordinarysize:itwaseasytoconcludethattheancientworldwasfarmorepopulousthanthemodern.Humesetouttoshowthattheseclaimswereinherentlyimplausible:slaveshadfewopportunitiestobreedandraisechildren;massacresanddestructivewarswerecommon;thesmallphysicalspacethatcitiesoccupiedwasirreconcilablewiththeclaimsmadefortheirpopulations;andhighinterestrateswereinfallibleevi-denceofprimitiveeconomies,economiesconsequentlyincapableofsupportingdensepopulations.Thusbydrawingtogetherdiffer-enttypesofevidencefromanimmenserangeofsources,Humeshowedthatancientclaimswithregardtopopulationsizeswerecompletelyunreliable;roundnumbers,likereportsofmiracles,rep-resentedaspecialcategoryofevidencetobetreatedwithextremeskepticism.Humesdetailedandtechnicalargumentwasnotofmerelyantiquarianimportance:itprovideddecisiveevidencefortheclaimthatmoderncivilizationwassuperiortoancient.HumeemergedvictoriousfromadebatewithRobertWallace,oneofthelastmajorbattlesinthelong-runningwarbetweenthosewhosawhistoryasarecordofdeclinefromancientglory,andthosewhosawitintermsof(atleasttemporary)progresstowardmodernsophistication.21InTheHistoryofEngland,sourcecriticismalsohadacentralroletoplay,forfactualclaimsaboutthepastwerecrucialtocontem-porarydebatesbetweenWhigsandTories,andbetweenProtestantsandCatholics.OneofHumescentralpurposeswastoexposeparty21SeeinthisvolumetheessayHumeontheArtsandTheStandardofTaste,PartI,and,fortheseventeenth-centuryoriginsofthecontroversy,R.F.Jones,AncientsandModerns,2nded.(Berkeley:UniversityofCaliforniaPress,1965).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian457myths:parties,likereligions,dependedonhistoricalclaimsthatcouldbesubjectedtoimpartialcriticism.Detailedhistoricalenquiryintosuchclaimscouldleadonetoconclusionsthatnoreasonablepersonwouldquestionandcouldintheprocessprovideevidencethatdecisivelyunderminedtherhetoricofpoliticalextremists.Humeidentifiedthreeepisodesthatplayedacrucialroleincontemporaryhistoricalmythologies,anditwasnolittlepartofhispurposeintheHistorytoshowhowhistorianscouldseparatefactfrommythwithregardtoepisodessuchasthese:Thereareindeedthreeeventsinourhistory,whichmayberegardedastouchstonesofpartymen.AnEnglishWhig,whoassertstherealityofthepopishplot,anIrishCatholic,whodeniesthemassacrein1641,andaScotchJacobite,whomaintainstheinnocenceofqueenMary,must[inthelightofHumesHistory]beconsideredasmenbeyondthereachofargumentorreason,andmustbelefttotheirprejudices.(HE39n.M;4:395)22Humespositiononthesequestionsinevitablyinvolvedhimincontroversy,evenwiththose,suchasRobertson,whosejudgmentheusuallyrespected.Inhisview,onesidewasobviouslyright,theotherwrong.Thiswasnotthecasewithanequallyvexedquestion,thatoftheauthorshipoftheEikonBasilike.Thiswork,suppos-edlybyCharlesIandpublishedimmediatelyafterhisexecution,wasundoubtedlythemostsuccessfulpieceofpoliticalpropagandainEnglishhistory.Itportrayedthekingaspious,honest,andcon-cernedwiththewelfareofhissubjects.ItspopularsuccessplayedamajorpartinpreparingthewayfortheRestoration.Butdiditreallyprovideaninsightintothekingssecretthoughts,orwasit,asmanyclaimed,thehypocriticalconcoctionofaroyalistchaplain,22HumeschoiceofthreeeventsinparticularispresumablygovernedbyhisdesiretohaveoneeachfromEnglish,Irish,andScottishhistory.TherealityofthePopishPlotasupposedplottoassassinatethekingandimposeCatholicismonEnglandbyforcewasfirmlyassertedbyShaftesburyin1678,anditwasthecampaigntohuntdowntheplottersthatbroughttheWhigpartyintoexistence.NumerousProtestantsweremurderedinaCatholicIrishuprisingin1641:thisprovidedthejustificationforCromwellssavagecampaignagainsttheIrish.QueenMaryofScotland,themotherofJamesIofEngland,wascomplicitinthemurderofherhusband,Darnley:thiswasheldtojustifyherexpulsionfromthethrone,aprecedentfortheexpulsionofJamesIIin1688,thelegitimacyofwhichwasdeniedbyJacobites.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n458davidwoottonDr.Gauden?Hume,whowantedtodefendCharlessgoodcharac-ter,hadaninterestinfindingthebooktobebyCharleshimself,buthedidnotinanywayunderstatethedifficultyofreachingadecision:Theproofsbroughttoevince,thatthisworkisorisnotthekings,aresoconvincing,that,ifanimpartialreaderperuseanyonesideapart,hewillthinkitimpossible,thatargumentscouldbeproduced,sufficienttocounter-balancesostronganevidence:Andwhenhecomparesbothsides,hewillbesometimeatalosstofixanydetermination.Shouldanabsolutesuspenceofjudgmentbefounddifficultordisagreeableinsointerestingaquestion,Imustconfess,thatImuchinclinetogivethepreferencetotheargumentsoftheroyalists.Thetestimonies,whichprovethatperformancetobethekings,aremorenumerous,certain,anddirect,thanthoseontheotherside[thoughHumeaddedafootnoteadmittingthismightbeanoverstatement].Thisisthecase,evenifweconsidertheexternalevidence:Butwhenweweightheinternal,derivedfromthestyleandcomposition,thereisnomannerofcomparison.Thesemeditations...aresounlikethebombast,perplexed,rhetorical,andcorruptstyleofDr.Gauden,towhomtheyareascribed,thatnohumantestimonyseemssufficienttoconvinceus,thathewastheauthor.(HE59.144,5:5478)HereHumehadfoundarealcasethatappearedatfirstsighttocorrespondtothehypotheticalcasehehaddiscussedinOfMira-cles,onewheretwobodiesofevidenceareinperfectbalance.Asinthatcase,Humegoesontoinsistonthesuperiorityofinternaloverexternalevidence:intheonecase,theinherentimprobabilityofaneventcontrarytothelawsofnature;intheother,theinherentimprobabilityofDr.Gaudensprosebeingtransformed.Suchevi-dencecouldbesostrongthatnotestimonycouldservetocontradictit:thecaseagainstDr.Gaudensauthorshipwaspreciselyasstrongasthecaseagainstmiracles.Humehadarguedthatunskilledindividuals,eveniftheycouldsparethetimeandenergytostudythemonumentsofthepast,wouldnotnecessarilyknowwhattomakeofthem.Conflictingclaimsregardingthecasketletters,supposedlywrittenbyMaryQueenofScots(inHumesviewherauthorshipcouldbedemonstrated),ortheauthorshipofEikonBasilike,couldonlybreedconfusionunlessonehadanadequatetraininginhistoricalmethodandunderstoodthephilosophicalprinciplesinvolvedintheassessmentoftestimony.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian459iii.historicalarchitectureWhenwespeaknowadaysofthephilosophyofhistorywesometimesmeantorefertothedetailedphilosophicalanalysisofthenatureofhistoricalevidence,andsometimestolarge-scaletheoriesthatclaimtoidentifyapatternorameaninginthecourseofhistory.Humewasfamiliarwithbothsortsofphilosophyofhistory.Hewouldnothavethoughtitunreasonabletoregardhishistoryasatestofthelarge-scaletheoriesofJamesHarrington(161177),Charles-LouisdeSecondat,barondeMontesquieu(16891755),andAnne-Robert-JacquesTurgot,barondelAulne(172781),whowouldhavebeeninhiseyesthemostimportantexponentsofthesecondsortofphilosophyofhistory.HarringtonwasanimportantfigurefortheanalysisofEnglishpoliticsforfourreasons:first,hehadinOceana(1656)givenanaccountofanidealcommonwealth,layingdownprinciplesaccord-ingtowhichafreestateoughttobeestablished,andprovidingastan-dardbywhichtheeighteenth-centuryEnglishconstitutioncouldbemeasured.Second,Harringtonbelievedthat,bygroundinghisidealstateintheprincipleofself-interest,hehadconstructedanewsci-enceofpolitics,althoughheacknowledgedthatinthisrespecthewasindebtedtoMachiavelliandtoThomasHobbes.Third,hehadprovidedanaccountofEnglishhistoryaccordingtowhichthemid-seventeenth-centurypoliticalcrisiswastheinevitableoutcomeofalongperiodofsocialchange.Fourth,hisfollowersaftertheRestora-tionhadmountedasustainedattackontwocontemporarydevelop-mentsthattheybelievedhadshiftedthebalanceofpowerinfavorofthemonarchy:thedevelopmentofaprofessionalorstandingarmy,andthegrowthofroyalrevenueandroyalbureaucracytothepointthatthekingcouldsignificantlyinfluencethepoliticalprocessbytheofferofplacesandpensions,bybriberyandcorruption.HumeengagedwithallfouraspectsoftheHarringtonianandneo-Harringtonianposition.23IntheIdeaofaPerfectCommon-wealth,helargelyapprovedofHarringtonsidealmodelofarepub-lic,whileofferingsomeimprovements.Second,inThatPoli-ticsmaybereducedtoaScienceandinTheIndependencyof23JamesMoore,HumesPoliticalScienceandtheClassicalRepublicanTradition,CanadianJournalofPoliticalScience10(1977):80939.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n460davidwoottonParliament,heacceptedtheHarringtonianpositionthatpoliticalinstitutionscreatedaframeworkwithinwhichpeoplepursuedtheirinterests;andthatitwasthereforepossibletopredicthowtheywouldbehave.Suchpredictionsmustbebasedontheassumptionthatpeoplewouldactselfishly:althoughinfactnoteveryonewasselfish,enoughpeopleweretoundermineanyinstitutionthatreliedonmenactingselflessly.Thekeytopoliticswasthusthestudyofinstitutions,notmen;goodinstitutionswouldcausementoactasiftheythemselvesweregood.Third,inOftheProtestantSucces-sion,hebyandlargeacceptedtheHarringtonianaccountofthesocialchangesthathadmadeanincreaseinthepoweroftheCom-monsinevitableinthemid-seventeenthcentury.InTheIndepen-dencyofParliament,however,hesoughttoshowthatHarringtonwaswrongtoconcludefromthisthatthemonarchywasboundtobeabolished.Finally,inWhethertheBritishGovernmentinclinesmoretoAbsoluteMonarchy,ortoaRepublic,heacceptedtheneo-HarringtonianaccountoftherisingpoweroftheCrown.HumeisneveruncriticalinhisattitudetoHarrington,butheisalwaysafriendly,neverahostile,critic(see,forexample,OftheFirstPrinciplesofGovernment).Harringtonhadconstructedthefirstdeterministaccountofhistory:hehadarguedthatthedistri-butionoflandedpropertydeterminedthedistributionofmilitarystrength,andthatthisinturnmustpredeterminetheoutcomeofpoliticalconflict.Allonehadtodowastoextendhisdefinitionofpropertytoincludecommercialwealthtohaveaspeciesofeconomicdeterminism.Ifonethenarguedthatcommercialactivityhadciv-ilizingconsequences,onewouldhaveaneweconomicexplanationfortheprogressofliberty.AdamSmithbelievedonecouldfindthisargumentinHumesHistory.HewroteinTheWealthofNations:commerceandmanufacturesgraduallyintroducedorderandgoodgovern-ment,andwiththem,thelibertyandsecurityofindividuals,amongtheinhabitantsofthecountry,whohadbeforelivedalmostinacontinualstateofwarwiththeirneighbours,andofserviledependencyupontheirsuperi-ors.This,thoughithasbeentheleastobserved,isbyfarthemostimportantofalltheireffects.Mr.Humeistheonlywriterwho,sofarasIknow,hashithertotakennoticeofit.2424AnInquiryintotheCausesoftheWealthofNations,ed.R.H.Campbell,A.S.Skinner,andW.B.Todd(Oxford:ClarendonPress,1976),3.4.4(1:412);C.N.Stock-ton,EconomicsandtheMechanismofHistoricalProgressinHumesHistory,inCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian461HumewouldcertainlyhaveagreedwithSmiththatthesubjecthadhithertobeenignored.In1741hehadwritten:Tradewasneveresteemedanaffairofstatetillthelastcentury;andtherescarcelyisanyancientwriteronpolitics,whohasmadementionofit.EventheItalianshavekeptaprofoundsilencewithregardtoit,thoughithasnowengagedthechiefattention,aswellofministersofstate,asofspeculativereasoners.Thegreatopulence,grandeur,andmilitaryatchievementsofthetwomaritimepowers[EnglandandHolland]seemfirsttohaveinstructedmankindintheimportanceofanextensivecommerce.(E-CL2,889)Thus,writingacenturyafterHarrington,HumewasboundtogivemuchmoreimportancetocommercethanHarringtonhaddone.AmajorpurposeofahistoryofEnglandmustbetotracetheroleofcommercialexpansioninthecountrysrisetoopulenceandgrandeur.WhatwasnottobeexpectedwasthatHume,insteadoftreatingpoliticallibertyasacauseofcommercialexpansion,asheoftenhaddoneintheEssays(seeE-CL10,923;E-RA6,113;E-Co1719,2656),wouldreversetheorderofcausationandinsistthatcommerce,inEnglandatleast,hadcausedtheappearanceandexpansionofliberty.Itwasinthetownsthatpoliticallibertyhadfirstexisted;itwasthedevelopmentofthemoneyeconomythatmadeserfdomananachronismthatthelordssawnoneedtopreserve;andtheriseofpersonalfreedompavedthewayfortheincreaseofpolit-icalorcivilliberty(HE23.42,2:524).Humedidnotofferafourstagestheoryofhistorythatexplainedhistoricalchangeintermsofthedevelopmentofthemeansofproduction:thiswastocomelater,inSmithandhisassociates.25Buthecertainlyhadprovidedanewtypeofeconomicexplanationofhistory,whichwemaytermacommercializedHarringtonianism.Hume,likeHarrington,sawintheriseofthegentrythepro-foundcauseoftheshiftinthebalanceofpowerthatunderminedthemonarchyintheearlyseventeenthcentury.IftheHarringtoniancharacterofhisanalysisisnotalwaysobvious,itisbecauseHumesawintheriseofthegentrytheexplanation,notoftheCivilWar,butoftheconstitutionalrevolutionthatprecededtheCivilWaritself.Socialchange(combinedwiththefactthatParliamentprovidedanHume:ARe-evaluation,ed.D.W.LivingstonandJ.T.King(NewYork:FordhamUniversityPress,1976),296320.25SeeAndrewSkinnersessayinthisvolume,andR.L.Meek,Smith,TurgotandtheFourStagesTheory,HistoryofPoliticalEconomy34(19712):927.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n462davidwoottoninstitutionalcontextwithinwhichthatchangecouldfindpoliticalexpression)mightexplaintheeventsof16401;onlyreligiousenthu-siasmcouldexplainParliamentsinabilitytocometotermswiththeking.Harringtonsawtheexecutionofthekingasthenaturalout-comeofthepoliticalprocesshehadtraced;Humesawthemixedconstitutionoftheearlyeighteenthcenturyasitsnaturaloutcome.FortheaberrationoftheCromwellianregime,otherexplanationswerenecessary.Harringtonhadfailedtotakeaccountofthefactthatthoughmenbemuchgovernedbyinterest;yeteveninterestitself,andallhumanaffairs,areentirelygovernedbyopinion(E-BG5,51).ItismuchhardertobesureaboutMontesquieusinfluenceonHumethanitisaboutHarringtons,forthesimplereasonthatHumesthinkingseemstoparallelMontesquieusbeforeaswellasafterthepublicationofTheSpiritoftheLaws(1748).OnonekeyissuehecertainlydisagreedwithMontesquieu:OfNationalCharacters(1748)readsasarefutationofMontesquieusaccountoftheroleofclimateinshapingpoliticalandculturallife.Itisnotimpossiblethatthisisexactlywhatitis,forHumemayhaveknownsomethingoftheargumentsofTheSpiritoftheLawspriortotheirpublication.26InotherrespectsHumesthinkingissoclosetoMontesquieusthatreadingMontesquieucanhelpintheinterpretationofHumesHistory.27MontesquieuandHumebothworkwithathreefoldtypol-ogyofregimes:despotisms,civilizedmonarchies,andconstitutionsbasedonliberty,suchastheEnglish.Englandisforbothofthemthemostsingularexampleoflibertytheworldhasseen.Bothusethewordlibertyinseveraldistinctsenses,butbothareprimarilyconcernedwithonethatcommentatorssometimescallprivateorcivilliberty,andforwhichrepresentativegovernmentservedasmerelyameans.28Montesquieudefinedpoliticallibertyasthat26P.E.Chamley,TheConflictbetweenMontesquieuandHume,inEssaysonAdamSmith,ed.A.S.SkinnerandT.Wilson(Oxford:ClarendonPress,1975),274305.27Montesquieu,Charles-LouisdeSecondat,baronde,TheSpiritoftheLaws,trans.anded.A.M.Cohlerandothers(Cambridge:CambridgeUniversityPress,1989),esp.18,22,15666,197,32533,388,456,608.28DavidMiller,PhilosophyandIdeologyinHumesPoliticalThought(Oxford:ClarendonPress,1981),1489;DuncanForbes,HumesPhilosophicalPolitics(Cambridge:CambridgeUniversityPress,1975),12592.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian463tranquillityofspiritwhichcomesfromtheopinionthateachonehasofhissecurity,andinorderforhimtohavethislibertythegovern-mentmustbesuchthatonecitizencannotfearanothercitizen.29Hume,inaposthumouslypublishedessay,stressed(surelyfollowingMontesquieuonthedivisionofpowers)theneedforcitizenstohavenocausetofear,notonlyeachother,butalsotheirgovernment:Thegovernment,which,incommonappellation,receivestheappellationoffree,isthatwhichadmitsofapartitionofpoweramongseveralmembers,whoseunitedauthorityisnoless,oriscommonlygreaterthanthatofanymonarch;butwho,intheusualcourseofadministration,mustactbygeneralandequallaws,thatarepreviouslyknowntoallthemembersandtoalltheirsubjects.Inthissense,itmustbeowned,thatlibertyistheperfectionofcivilsociety.(E-OG7,40l)ThusbothseethekeytoEnglishlibertyaslyingintheestab-lishmentofadivisionofpowersbetweenlegislativeandexecutive,butbothregardotherconstitutionaldevelopmentsasindispensable:thejurysystem,habeascorpus,anindependentjudiciary,andthefreedomofthepressarepillarsofliberty.Bothbelievethatthecom-petinginterestsoflegislativeandexecutivemustnecessarilycreateconflictingpartiesinsupportofthesedifferentinterests,andthatthesepartieswillinevitablyrewritehistorytojustifytheirpolicies.30Finally,bothbelievethattheessentialpatternoftheevolutionofEnglishhistorysincetheNormanConquesthasbeenoneoffeudalanarchy,whichgavewaytodespotism,whichitselfgavewaytothelibertyofamixedconstitution.BothagreethatthereneverhasbeenacivilizedmonarchyinEngland,partlybecausesuchaninstitutiondependsonastrongnobility,whiletheEnglishnobilitywasfatallyweakenedbeforetheemergenceofastrongmonarchy.Finally,MontesquieusTheSpiritoftheLawsforcefullyremindsusofacentralproblemthatfacedHumeashewrotetheHistory.The29Montesquieu,TheSpiritoftheLaws,157.Montesquieuwrites,Ihavehadnewideas;newwordshavehadtobefoundornewmeaningsgiventooldones.Hume,Iwish,thatPeopledonottakeaDisgustatLiberty;aword,thathasbeensomuchprofandbythesepollutedMouths[thesupportersofWilkes],thatmenofSensearesickattheverymentionofit.Ihopeanewtermwillbeinventedtoexpresssovaluableandgoodathing(NHL,196).30Humethought(lettertoMontesquieu)thatMontesquieusomewhatunderesti-matedthedangersofconstitutionalconflict(HL1:138).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n464davidwoottonclassicalmodelsforhistoricalwritingwereallalmostentirelynarra-tiveinform.Montesquieu,however,showedthatonecouldanalyzeaconstitutionandacultureatamomentintime,presentingitasanidealtype.Indeed,oneneededtodothisifonewastograspthelogicofevents.Humethereforecommittedhimselftoextendeddigres-sionsexplainingthecharacteroftheconstitutionatdifferenttimes,andsurveyingevents(forexample,developmentsintheartsandsci-ences)thatdidnotfitcomfortablywithinanarrativeframework.Nevertheless,thesedigressionsseemedtohimatoddswiththenar-rativeformpropertohistoricalwriting,andinlatereditionsoftheHistoryhemovedpartsofthemintonotes(HL1:294).31Moreover,althoughHumeclearlyfeltthatconstitutionalinstitutionsandprac-ticeswerecrucialtotheexplanationofpoliticalevents,heresistedfollowingMontesquieusofarastryingtoestablishacloserelation-shipbetweendevelopmentsinanationscultureandthoseinitspoliticallife:Humeneverclaimedtoidentifyasinglespiritrunningthroughallaspectsofthelifeofanepoch.Turgot,unlikeHarringtonandMontesquieu,hadlittleinfluenceonHume,buthisviewsprovideahelpfulcontrasttoHumesown.ItisnotpossibletodeterminewhenHumefirstreadTurgotsOnUni-versalHistory(1751),whicharguesthatallhistoryisthehistoryofprogress.In1768,fullofpessimismoverthedemandsforlibertymadebyWilkes,HumepointedtoeventsinEnglandasarefutationofTurgotsviews:Iknowyouareoneofthose,whoentertaintheagreeableandlaudable,ifnottoosanguinehope,thathumanSoci-etyiscapableofperpetualProgresstowardsPerfection....Pray,donotthelateEventsinthisCountryappearalittlecontrarytoyourSystem?Humewasconvincedthatdangerouspoliciesaboveallthegrowthofthenationaldebtcouldendangercivilization.HewasnotevenpersuadedbyTurgotsviewthatsincetheDiscoveryofPrintingweneednolongerDreadtheusualReturnsofBarbarismandIgnorance(HL2:180).Onthecontrary,readingthefirstvolumeofEdwardGibbonsDeclineandFalloftheRomanEmpire(1776)31Hewasobviouslyhappierwithanothertypeofdigression:thediscussionofanindividualscharacter.Thiswasnecessary,bothtoevokethereaderssympathiesandpassions,andtomakeitpossibleforthemtomakemoraljudgments.InHumesviewitischaracterratherthanindividualactionsthatoughttobejudgedwhenassessingthemoralvalueofaperson.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian465madehimremarkonthemarksofDeclineinEngland,wherethePrevalenceofSuperstition...prognosticatestheFallofPhilosophyandDecayofTaste(HL2:310).HeforesawanewandasuddenInroadofIgnorance,SuperstitionandBarbarism(NHL199).Tur-gotsconfidencethatprogresswasguaranteedbyprovidencemustindeedhaveseemedtoHumetobeitselfasophisticatedsupersti-tion.HewasmuchhappierwithVoltairesCandide,ouloptimisme(1759):ItisfullofSprightliness&Impiety,&isindeedaSatyreuponProvidence,underPretextofcriticizingtheLeibnitianSys-tem(NHL53).Historyitself,inHumesview,providednogroundsforanyfaithinprovidence,eveninthesecularizedformofabeliefintheinevitabilityofprogress.iv.whigsandtoriesWecomenowtothecentral,andmostdifficult,questionintheassessmentofHumeasahistorian:thatofthepoliticsoftheHis-tory.32Scholarshaveadoptedanumberofviewsonthisquestion,andHumesownexplicitdescriptionsofthepoliticalsignificanceoftheHistoryinhislettersareopentomorethanoneinterpreta-tion.Forsome,theHistorypresentsaconservativeviewofpolitics:itreflectsthegrowingconservatismofHumesviewsaftertheJacobiteuprisingof1745andwasrevisedfurtherinaconservativedirectionafterHumerespondedwithhostilitytotheWilkesandLibertycampaigns.ItisthusessentiallyaToryhistory.33Accordingtoanotherview,thewholepointoftheHistoryisitsimpartiality:HumepresentstheviewsofthedifferentpartiesinvolvedinEnglishpoliticsandadmitsthatallhaveacertainjustification.34However,onthisviewHumeheldthatpastpracticewasalmosttheonlybasisforestablishingwhetheragovernmentwaslegitimate.HumesdeclaredsupportfortheRevolutionof1688intheTreatise,andthe32ForHumesownsummaryaccountofpartypoliticsinEngland,andhisexplanationoftermssuchasCourtandCountry,WhigandTory,seeE-PGB.33GiuseppeGiarrizzo,DavidHumePoliticoeStorico(Turin:Einaudi,1962),andAncorasuHumestorico,Revistastoricaitaliana83(1971):42949.34DuncanForbes,PoliticsandHistoryinDavidHume,HistoricalJournal6(1963):28095;HumesPhilosophicalPolitics;andhisintroductiontoHumesHistory,areprintofthefirsteditionoftheearlyStuartvolume(Harmondsworth:Penguin,1970).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n466davidwoottonHistoryinvolvedhim,itisargued,innolittleinconsistency.AthirdviewisthatHumewrotetodefendtheexistingconstitution,andthatheapprovedthosemenandmeasuresthatbroughtitabout,eventhoughtheycouldonlybeseentohavebeenadmirablewiththeadvantageofhindsightandhadnoproperideaofwhattheyweredoingatthetime.35HumecertainlydescribesintheTreatisethepsy-chologicalprocessthatcausesonetomakeretrospectivejudgmentsinthelightoftheoutcomeofevents(T3.2.10.19),butahistorywrit-tenfromsuchapointofviewwouldfailinwhatwasforHumeitsessentialpurpose,thatofdistinguishingvirtuefromvice:todothisthehistorianwouldhavetogiveuptheadvantageofhindsight.AfourthviewholdsthatHumethoughtthatthecontemporaryabso-lutistmonarchiesofEuropeweresuperiortothemixedmonarchyofEngland,thathenotonlyhopedtoseeabsolutismestablishedinEnglandinhisownday,butregrettedthefailureoftheStuartstoestablishitintheseventeenthcentury.36BeyondHumesinsistencethatcivilizedabsolutistgovernmentsarelegitimatewheretheyarealreadyestablished,thisviewseemstometohavelittleevidencetosupportit.Imustadmittoacertainsympathywithafifthview,namely,thattheHistoryisinformedthroughoutbyHumesloveofliberty,althoughitwouldbequitewrongtoequateliberty,asHumeunderstoodtheterm,withdemocracy.37ThereisnodoubtthatthefirstvolumeoftheHistorywaswellreceivedinFrancepartlybecauseitgavetheStuartmonarchyasym-pathetictreatment.DuringthereactiontotheFrenchRevolution,HumeshostilitytotheexecutionofCharlesImadehimseemthefirstcounter-revolutionaryhistorian,aprecursorofBurke.38LouisXVIsownresponsetothenewsofhisdeathsentencewastosetabout35Livingston,HumesPhilosophyofCommonLife,2578.36NicholasPhillipson,Hume(NewYork:St.MartinsPress,1989).37J.V.Price,HumesConceptofLibertyandTheHistoryofEngland,StudiesinRomanticism5(1966):13957.LibertyinHumesHistoryofEngland,ed.N.CapaldiandD.W.Livingston(Dordrecht:Kluwer,1990),collectsanumberofessaysthatdefendthisview.DavidMiller,PhilosophyandIdeology,providesabalancedassessment.AnimportantcontributiontothisdebateisJohnB.Stew-art,OpinionandReforminHumesPoliticalPhilosophy(Princeton:PrincetonUniversityPress,1992).38LaurenceL.Bongie,DavidHume:ProphetoftheCounter-Revolution(Oxford:OxfordUniversityPress,1965).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian467rereadingHumesaccountofCharlessexecution.39EvenbeforetheFrenchRevolution,theanti-WilkesreactionhadincreasedHumespopularityinEngland.Hewrotein1768:Licentiousness,orratherthefrenzyofliberty,hastakenpossessionofus,andisthrowingeverythingintoconfusion.HowhappydoIesteemit,thatinallmywritingsIhavealwayskeptataproperdistancefromthattemptingextreme,andhavemaintainedadueregardtomagistracyandestablishedgovernment,suitablytothecharacterofanhistorianandaphilosopher!Ifindonthataccountmyauthoritygrowingdaily;andindeedhavenownoreasontocomplainofthepublic.(HL2:1912)Suchreadings,however,ignoreimportantaspectsofHumesargu-ment,aspectsthatsurvivedhismanyrevisionsofhiswork,allofwhich,hesaid,favoredtheTories.Hedescribedthe1770revisionsasfollows:IamrunningoveragainthelastEditionofmyHistoryinordertocorrectitstillfarther.IeithersoftenorexpungemanyvillainousseditiousWhigStrokes,whichhadcreptintoit.IwishthatmyIndignationatthepresentMadness,encouragdbyLyes,Calumnies,Imposture,andeveryinfamousArtusualamongpopularLeaders,maynotthrowmeintotheoppositeExtreme.Iam,however,sensiblethatthefirstEditionsweretoofullofthosefoolishEnglishPrejudices,whichallNationsandallAgesdisavow.(HL2:216)Itwouldbewrong,however,tooverestimateHumesconservatism,eventowardtheendofhislife.Notonlywashe,asheputit,anAmericaninmyPrinciples(HL2:303),unconditionallysupportingAmericanindependencefromanearlydate,buthewasalsohappytosaybluntlythattheFrenchmonarchywouldhavetobeoverthrownsothattheFrenchpeoplecouldescapetheburdenofthegovernmentdebt(HL2:242).Indeed,itgaveHumegreatpleasuretothinkpeoplewouldhavetroublecategorizinghim:WhetheramIWhigorTory?ProtestantorPapist?ScotchorEnglish?Ihopeyoudonotallagreeonthishead;&thatthere[are]disputesamongyouaboutmyprinciples(HL1:196).Clearlyhewoulddelightinthecontinuingdisputesoverhowtocategorizehim.39Livingston,HumesPhilosophyofCommonLife,317.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n468davidwoottonTomakesenseofHumesHistory,wemusttakemoreseriouslythancommentatorsusuallydothefactthathisisanarrativehis-torydescribingchangingsocialandpoliticalcircumstances.Humesemphasisonchangewas,indeed,oneofthemostoriginalaspectsofhishistory.Toryhistoriansbefore1688,suchasRobertBrady,thefirstvolumeofwhoseCompleteHistoryofEnglandappearedin1685,andJacobitehistorians,suchasThomasCarte,authorofAGeneralHistoryofEngland(174755),hadinsistedthattheancientconstitutionofEnglandwasthatofdivine-rightmonarchy:therevo-lutionsof1642and1688werescarcelyforgivablefromthispointofview,andthemid-eighteenth-centuryconstitutionanoutrage.Whighistoriansbefore1714hadarguedtheoppositecase:theancientcon-stitutionofEnglandwasonethatguaranteedthelibertiesofthesubject.Onthisview,whichwassupportedbyJamesTyrrellinhisGeneralHistoryofEngland(16961704)andlaterbyRapin,Parlia-ment,includingtheHouseofCommons,wasaninstitutionofSaxonorigin,andonewithanunbrokenhistory.TheStuartshadsoughttounderminethisancientconstitution;theRevolutionof1688hadrestoredit.InHumesday,elementsofthisviewwereadoptedbyToryopponentsofthecourt,suchasHenrySaint-John,viscountBolingbroke.40Thiscountryoppositionattackedthecourtforcor-ruption,andattackedtheWhiggovernment,particularlyWalpoleintheperiodimmediatelybeforehisfall,forextendingexecutivepower.ThereplyofthecourtWhigstothisassaultwasanovelone.LordHervey,forexample,arguedthattherehadbeennoancientcon-stitution.41TheconstitutionofEnglandhadbeenconstantlyinflux.Therehadbeennosecurelibertybefore1688.Thepresentadmin-istrationshouldbejudgednotbyitsfidelitytoamythicalsetofconstitutionalprinciples,butbyitsabilitytopreserveeffectiveandbeneficentgovernment.Placedinthiscontext,thereisnodoubtthatHume,forallhisclaimstoimpartiality,effectivelyendedupsupportingtheargument40BolingbrokewastheauthorofADissertationuponParties(17334)andofLet-tersontheStudyandUseofHistory,2vols.(London,1752).ForaselectionofBolingbrokesworks,seeHistoricalWritings,ed.I.Krammick(Chicago:Univer-sityofChicagoPress,1972).41John,BaronHervey,AncientandModernLibertyStatedandCompared(London,1734).HumedidnotsharethecourtWhigsfavorableassessmentofWalpole(E-CR5746).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian469ofthecourtWhigs.ThefirstvolumehewrotewasthevolumeontheearlyStuarts.AlthoughinwritingthatvolumehecameincreasinglytosuspectthattraditionalWhighistorywasdeficient,andalthoughhedefendedatlengththecharacterandmotivesofCharlesI,hestilltendedtotheWhigviewthattheStuartswereinnovatingtothedetrimentofParliament.OncehehadworkedontheTudorperiod,hebecameconvincedthatTudorrulehadbeendespoticandrevisedhisaccountofStuartaspirationsaccordingly.42HumesrevisionsofhisHistorymayhavebeenhostiletoWhigprejudices,buttheybroughthimclosertothecourtWhigposition.InHumesview,thebarbarismoftheMiddleAgesandthedespo-tismoftheTudorswerenotsurprising:theywerepreciselywhatonewouldexpecttofindinaprimitivesociety.Avastgulfseparatedusfromtheworldofourancestors.InplaceoftraditionalToryandWhigarguments,whichstressedconstitutionalcontinuityandpaidlittleattentiontosocialchange,Humestressedthedistanceatwhichwestoodfromboththesocialandthepoliticallifeofourancestors:heattacked,alongsidetheWhigmythoftheancientconstitution,themythoftheRoastBeefofOldEngland.Apoordietandadespoticgovernmentwereintimatelyrelated.Incorrespondenceheexplainedhisviewbluntly:MyNotionis,thattheuncultivatedNationsarenotonlyinferiortocivilizdinGovernment,civil,military,andeclesiastical;butalsoinMorals;andthattheirwholemannerofLifeisdisagreeableanduneligibletothelastDegree.IhopeitwillgivenoOffence(andwhetheritdoornot,Imustsayit)ifIdeclaremyOpinion,thattheEnglish,tillnearthebeginningofthelastCentury,areverymuchtoberegardedasanuncultivatedNation;andthatevenWhengoodQueenElizabethsatontheThrone,therewasverylittlegoodRoastBeefinit,andnoLibertyatall.TheCastleoftheEarlofNorthumberland,andnodoubtthatoftheEarlofWarwick,theKingMakerandothers,wasnobetterthanaDungeon:NoChimneytoletouttheSmoak;noGlassWindowstokeepouttheAir;aglimmeringCandlehereandthere,whichcoudscarcekeeptheirRagamuffinsofServantsandRetainersfrombreakingtheirShinsorrunningfoulofeachother:NoDietbutsaltBeefandMutton42ThefullextentofRapinsdeficiencywasobviouslynotapparentin1753:HL1:170;contrastHL1:258.ForthecomparisonbetweentheTudorsandtheStuarts,HL1:264,461.Forlaterrevisions:HL1:379;HL2:216,2601;NHL6971.ForHumesviewsin1758,E-CP493501.Forananalysis,seeE.C.Mossner,WasHumeaToryHistorian?JournaloftheHistoryofIdeas2(1941):22536.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n470davidwoottonfornineMonthsoftheYear,withoutVegetablesofanykind:FewFiresandtheseverypoorones....WhenQueenCatherineofArragonhadaFancytoeataSallad,shecoudnotfindoneinallEngland,shewasobligdtosendaMessengertotheLowCountriesforthatPurpose:AndIsupposewhenherTyrantofaHusband[HenryVIII]thoughtshewaswithChild,andthattheLifeofhisroyalIssuedependeduponit,hewoudindulgeherinthatCaprice.(NHL198)Humehimselfcamefromwhathadlongbeen,byhisdescription,therudest,perhaps,ofallEuropeanNations;themostnecessitous,themostturbulent,andthemostunsettled(HL2:310).Theextraor-dinaryrapidityofsocialchangeintheScotlandofhisdaymadehim,alongwiththerestoftheenlightenedScots,acutelysensitivetothesocialevolutionthathadtakenplaceinthepast;thecontinu-ingbackwardnessoftheHighlandsmusthaveprovidedthemwithavividpictureofmedievallifeinFranceorEngland.Nevertheless,HumewaspreparedtofindevenintheMiddleAgesthestirringsofmodernliberty.AlthoughheinsistedonplacingtheMagnaCarta,thegreatcharteroflibertiesof1215,inafeudalandbaronialcontext,hewasalsopreparedtoseeinitstatementsofprinciplethatwouldhavemetwiththeapprovalofaLockeoraSidney:Itmustbeconfessed,thattheformerarticlesoftheGreatChartercontainsuchmitigationsandexplanationsofthefeudallawasarereasonableandequitable;andthatthelatterinvolveallthechiefoutlinesofalegalgovern-ment,andprovidefortheequaldistributionofjustice,andfreeenjoymentofproperty;thegreatobjectsforwhichpoliticalsocietywasatfirstfoundedbymen,whichthepeoplehaveaperpetualandunalienablerighttorecal,andwhichnotime,norprecedent,norstatute,norpositiveinstitution,oughttodeterthemfromkeepingeveruppermostintheirthoughtsandattention.(HE11.56,1:445)Thisishardlythelanguageofthethirteenthcentury,anditcomesperilouslyclosetoapprovingviewsthatHumehadattackedinhisessayOftheOriginalContract.43Whenhedescribestheexecu-tionofAlgernonSidney(d.1683),whohadbeenconvictedbecausehehaddefended,inaprivatemanuscript,viewssuchasthese,Hume,43Onwhich,seeinthisvolumetheessayTheStructureofHumesPoliticalThe-ory.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian471farfromattackingcontractarianism,cautiouslydefendsit:Sidneyhadmaintainedprinciples,favourableindeedtoliberty,butsuchasthebestandmostdutifulsubjectsinallageshavebeenknowntoembrace;theoriginalcontract,thesourceofpowerfromacon-sentofthepeople,thelawfulnessofresistingtyrants,thepreferenceoflibertytothegovernmentofasingleperson(HE69.48,6:436).ThoughhehadindeedbeeninvolvedinaconspiracyagainstCharlesII,hisconvictionwascontrarytolaw,hissentenceiniquitous.Suchpassages,takeninisolation,wouldappeartojustifyacourtWhigreadingofHume:althoughlibertyhadnotbeenlongestab-lished,theprinciplesoflibertyhadalwaysbeenadmirable.Humehimselfinsisted:MyviewsofthingsaremoreconformabletoWhigprinciples;myrepresentationsofpersonstoToryprejudices.Noth-ingcansomuchprovethatmencommonlyregardmorepersonsthanthings,astofindthatIamcommonlynumberedamongtheTories(HL1:237).DuncanForbes,whosestudyofHumesPhilo-sophicalPoliticsisthemostinfluentialaccountofHumeshistoricalthought,preferstoemphasizeHumesownclaimstoimpartiality,anddismissesthisparticularremarkassuperficial:HumehadnocommitmenttoWhigprinciples.44ButHumesclaimstoimpar-tialityarealsomisleadingunlessonenotesthatHumeclaimstosupportWhigsandToriesalternately:inotherwords,heisalwayspartial,evenifheisnotalwaysonthesameside(HL1:179,369).ThekeytothevolumeontheearlyStuartslies,infact,inthefollowingdescription:IamnotsurelyunfavorabletotheParliament.TilltheypushdtheirAdvantagessofarastoexciteacivilWar,sodangerous&unnecessary,IesteemtheirConductlaudable;&tothisExtrem-itynothingcarrydthembuttheirfuriousZealforPresbytery:AlowBigotry,withwhichtheysullydanobleCause(HL1:222).Indeed,HumesHistoryisParliamentary,notRoyalist,initsaccountoftheperiodfrom1604to1641,andParliamentaryagaininitsaccountoftheperiodfrom1681to1688.Thereisnothingarbitraryaboutthischangingofsides:inHumesview,itwasacharacteristicofEnglishhistorythatthedisinter-ested(thatistosay,theimpartial)fluctuatedbetweenthefac-tions;andgavethesuperioritysometimestothecourt,sometimestotheopposition(HE66.43,6:3078).Thehistorywaswrittenfrom44Forbes,HumesPhilosophicalPolitics,292.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n472davidwoottonthepointofviewofsuchdisinterestedindividuals,whoseconcernwaswiththepublicgood.Suchpeople,Humebelieved,supportedParliamentuntiltheendof1641,untiltheGrandRemonstrancemadewarinevitable.TheyoughttohaveadmiredJohnPymandJohnHampden,Parliamentsleadingspokesmen,andapprovedthePetitionofRightof1628andtheconstitutionalrevolutionof1640and1641,whichhadestablishedtheprinciplesofthemixedconsti-tution.TheyshouldhavecontinuedtosupportParliamentdespitetheexecutionin1641ofThomasWentworth,EarlofStrafford,thekingschiefminister,aneventthatforeshadowedthetrialofCharleshimself.ButthereafteritwasParliament,nottheking,thatendan-geredstability(despite,forexample,thekingsattempttoarresttheleadersoftheCommons,includingPymandHampden,in1642).Thekinghadacceptedthathiswasalimitedmonarchy,andtherewasnothinginhischaractertosuggestthathecouldnotbetrusted.OnlytheirlowBigotrydrovemenontoresistance,whilethosewhowerenotbigotsralliedtotheroyalistside.ThetrueheroesofHumesaccountwerethusthemenwhohadknownwhentochangesides,ClarendonandLuciusCarey,viscountFalkland,whohadjoinedthekingin1642.45HewillinglyechoesClarendonspraiseofFalkland,attributingtohimtheauthorshipofthekingsreplytotheParliament,HisMajestiesAnswertotheNineteenPropositions(June1642),inwhichforthefirsttimetheprinciplesofthemodernEnglishconstitutionweredescribedanddefended.Falkland,whodiedinbattlein1643,wastheepitomeofHumesconceptionofvirtue:devotedtothepursuitsoflearning,andtothesocietyofallthepoliteandelegant,[he]hadenjoyedhimselfineverypleasure,whichafinegenius,agenerousdisposition,andanopulentfortunecouldafford.Calledintopub-liclife,hestoodforemostinallattacksonthehighprerogativesofthecrown;anddisplayedthatmasculineeloquence,andundauntedloveoflib-erty,which,fromhisintimateacquaintancewiththesublimespiritsofantiquity,hehadgreedilyimbibed.Whencivilconvulsionsproceededtoextremities,anditbecamerequisiteforhimtochusehisside;hetempered45Itisworthnoting,however,thatHumeregardedBulstrodeWhitelockeasamoreimpartialhistorianthanClarendon(HE64,6:215).SinceWhitelockesupportedParliamentwhileClarendonsupportedtheKing,thisjudgmentscarcelyaccordswithaviewofHumeasaTory.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian473theardourofhiszeal,andembracedthedefenceofthoselimitedpowers,whichremainedtomonarchy,andwhichhedeemednecessaryforthesup-portoftheEnglishconstitution.Stillanxious,however,forhiscountry,heseemstohavedreadedthetooprosperoussuccessofhisownpartyasmuchasoftheenemy;and,amonghisintimatefriends,oftenafteradeepsilence,andfrequentsighs,hewould,withasadaccent,re-iteratetheword,Peace.(HE56.59,5:41617)OnceweseeHumechangingsides,wecanrecognizethat,whetherhewritesinfavorofkingorParliament,healwaysgiveshissupporttothemixedconstitution.Forthosewhofavoredsuchaconstitu-tionthekeyquestionattheendof1641was,hebelieved,Canthekingbetrusted?HencetheimportanceofHumesaccountofCharlesscharacter,andthestrategicsignificanceofhisclaimthattheargumentsoftheroyalistsin1642werestrongestwhendiscussingnotprinciple,butthepastbehavioroftheking.Simi-larly,betweentheExclusionCrisisand1688,thekeyquestionisthebehaviorofJamesII.HumerefusestotakesidesbetweentheWhigswhotrumpedupthechargesagainstToriesduringthePopishPlot(1678),andtheTorieswholedthewitchhuntagainstWhigsaftertheRyeHouseplot(1683).46Butfrom1682JamessbehaviortowardScotlandwasunambiguous.Hesoughttoexecutemenguiltyonlyofhavingconversedwithotherswhoweresuspectedofhavingbeenrebels.Twowomenweretiedtostakesandslowlydrownedbyarisingtidebecausetheyrefusedtotakeanoathofloyalty.Jameshim-selfwasbelievedtohaveparticipatedinthetorturingofsuspects.IfClarendonrepresentsthereasonablemenandwomenwhocon-vertedtoRoyalismbeforetheoutbreakoftheCivilWar,ArchibaldCampbell,ninthearlofArgyll,representstheloyalsubjectswhowereforcedtoturnagainstJames.TriedforhislifebecauseofhisloyaltytoProtestantismandtheconstitution,hewasdrivenintoexilelatein1682:fromthismoment,conspiracyagainstJameswasjustified.Therewas,ofcourse,animportantdifferencebetween1642and1682:aftertheRestoration,CharlesIIandJamesIIhadnoexcuseforfailingtorecognizethatEnglandwasnowalimitedmonarchy.46TheRyeHousePlotwasaplotbyleadingWhigs,includingAlgernonSidney,toassassinateCharlesIIandhisbrotherJames.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n474davidwoottonAsHumehadarguedintheTreatise,inanylimitedmonarchytheremustbearightofrebellionindefenseoftheconstitution,forother-wisetherewillbenoeffectivelimitsonroyalpower.IfHumehadgoodreasontodefendrebellionin1688,evenonhisownconser-vativeprinciples,oneisboundtoask,OnwhatgroundscouldhejustifyParliamentaryinnovationin1641?Perhaps,inthewordsoftheTreatise,nothingismoreessentialtopublicinterest,thanthepreservationofpublicliberty(T3.2.10.16),butwhatrightcanthepublicclaimtoconstructpubliclibertywheretherehasbeennonebefore?Eveniftheyknowthatlibertywillbenefitsociety,howcantheybeconfidentthattheattempttoestablishitwillnotfatallyundermineauthority,andthatinnovationwillnotcausemoreharmthangood?Inattemptingtoanswerthisquestion,thefirstthingtonoteisthatEnglandundertheTudorsdidnothavewhatHumetermedacivilizedmonarchy(E-RP28,125).Itsufferedunderaspeciesofdespotism.Thus,underHenryVIIItheEnglishweresothor-oughlysubduedthat,likeeasternslaves,theywereinclinedtoadmirethoseactsofviolenceandtyranny,whichwereexercisedoverthemselves,andattheirownexpence.UnderElizabeth,thegovernmentofEnglandboresomeresemblancetothatofTurkeyatpresent:Thesovereignpossessedeverypower,exceptthatofimpos-ingtaxes(HE33.47,App.3.10;3:323,4:360).Thislimitation,asinTurkey,couldonlyhaveperniciousconsequenceswhilethemonarchsoughttoexerciseabsolutepower(E-Ta11,3478).InEnglanditmeantthattheCrownhadrecoursetoforcedloans,orratherarbi-traryconfiscations,andtothesaleofmonopoliesandprivileges.Overtime,thefirstpracticewouldhavedestroyedallsecurityofproperty,whilethesecondwouldhavedestroyedalltrade,reducingEnglishsocietytothepovertyofTurkish.Anabsolutegovernmentinwhichthemonarchcouldlevytaxeswouldhavebeenpreferable,forthenthemonarchywouldhavehadaninterestinthewealthofitssubjects.Tudordespotismwasthusdangeroustocivillibertyandeconomicprosperity:itwasinthepublicinterestthatsuchdespo-tismshouldgiveway,eithertocivilizedabsolutismortoamixedconstitution.InhisAppendixtothereignofElizabeth,Humesaysthatoneshouldgenerallylimitonesquestionsaboutconstitutionstoquestionsaboutthefacts:tofindingwhathasinpracticebeenestablished.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian475Ifanyotherrulethanestablishedpracticebefollowed,factionsanddis-sentionsmustmultiplywithoutend:Andthoughmanyconstitutions,andnonemorethantheBritish,havebeenimprovedevenbyviolentinnova-tions,thepraise,bestowedonthosepatriots,towhomthenationhasbeenindebtedforitsprivileges,oughttobegivenwithsomereserve,andsurelywithouttheleastrancouragainstthosewhoadheredtotheancient[thatis,established]constitution.(HEApp3.1,4:355;seealso23.245,2:51315)WhenHumefirstwrotetheHistory,hethoughtthatintheearlysev-enteenthcenturybothRoyalistsandParliamentarianshadsoughttounderminetheexistingconstitution.IntimehecamemoreandmoretobelievethatmanyoftherightsclaimedbytheStuartswereestab-lishedinpreviouspractice;nevertheless,theycouldnotbeclearedofthechargeoftryingtoinnovatebymakingthekingmasteroftaxation.SinceParliamentrefusedtofinancethegovernmentade-quately,suchinnovationmightbejustifiable,but,alas,theresultwouldhavebeennotacivilizedabsolutism,butanalmostunparal-leledconcentrationofpower.Humethoughtthatmostsupposedlyabsolutegovernmentswereinfactlimited.AsheputitinOftheOriginofGovernment,Thesultanismasterofthelifeandfortuneofanyindividual;butwillnotbepermittedtoimposenewtaxesonhissubjects:aFrenchmonarchcanimposetaxesatpleasure;butwouldfinditdangeroustoattemptthelivesandfortunesofindividuals(E-OG7,40).CharlesI,how-ever,wasablebothtotaxandtoimprisonatwill.Inmanycoun-tries,Humesays,religionactsasarestraintonpower,andindeedinhisaccountoftheelevenyearstyranny(162940),HumesaysthattheChurchofEnglandwastheonlyeffectiverestraintonroyalauthority(HE53.1,5:250).Thecommentwasdeliberatelyironical,forthiswasthechurchofLaud,achurchthatpreachedunrestricteddivine-rightmonarchy.InFrance,asMontesquieuhadstressed,thepowerofthenobilityprovidedapracticalrestraintonthekingswill;inEngland,bycon-trast,thenobilityhaddeclinedinthefifteenthcentury,creatingapowervacuumthatthemonarchyhadsoughttofill.InFrance,theParlementsprovidedaguaranteeofdueprocessinlaw;inEngland,theStarChambercouldcircumventthenormalprocessesoflaw,evencommonlawjudgesservedattheroyalpleasure,andjurieswereeasytointimidate(HEApp.3.38,4:35660).InFrance,thechurchwasinlargemeasureindependentoftheCrown;inEngland,itwasCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n476davidwoottonentirelyunderroyalcontrol.Suchcomparisons,implicitratherthanexplicitinHumesHistory,couldalonejustifytheinnovationstowhichtheLongParliamentresortedinitsfirstsessions.HumewasthusabletoconcludehisaccountoftheTudors:Onthewhole,theEnglishhavenoreason,fromtheexampleoftheirances-tors,tobeinlovewiththepictureofabsolutemonarchy;ortoprefertheunlimitedauthorityoftheprinceandhisunboundedprerogatives,tothatnobleliberty,thatsweetequality,andthathappysecurity,bywhichtheyareatpresentdistinguishedaboveallnationsintheuniverse.(HEApp.3.32,4:370)ThekeytoEnglishliberty,equality,andsecuritydidnotlie,how-ever,aswemightthink,inrepresentativegovernment.Represen-tativegovernmentwasnotonitsownaguaranteeoffreedom,forelectionsprovidedscopefortheCrowntobringitsinfluencetobear,andthesmallnumberofmembersofParliamentmadeitpossibleforittomakefulluseofitspowersofpatronage.Humewould,inprinciple,havepreferredmoredecisionstobetaken,eitherbytheelectorateasawhole,orbymuchlargerrepresentativebodies.Lib-ertyderived,infact,fromtheconstantefforttorestrictroyalauthor-itywithinthemixedconstitution.Itwasthisthathadgivenbirthtoinstitutionsthatwere,inHumeseyes,unprecedentedadvancestowardcivilliberty.Firstamongthesewashabeascorpus,uniquetoEngland,ofwhichhewrote:asithasnotplaceinanyotherformofgovernment,thisconsiderationalonemayinduceustopreferourpresentconstitutiontoallothers(HE67.59,6:367).Secondwasthelibertyofthepress,alsouniquetoEngland,ofwhichhewaspre-paredtoclaim(untilthedisadvantagesoftoomuchlibertybecameapparenttohimduringtheWilkesandLibertycrisis):thislibertyisattendedwithsofewinconveniencies,thatitmaybeclaimedasthecommonrightofmankind,andoughttobeindulgedthemalmostineverygovernment(E-LP6.13n,604[published174168]).47HumessupportforParliamentundertheearlyStuartsthusrests,notonanyprincipledcommitmenttorepresentativegovern-ment,butontheconvictionthatonlythegentry,representedinParliamentandgivenapoliticalvoicebyit,wereinapositiontocontrolwhatshowedeverysignofbecomingatyrannicalpower.47TheonlyexceptionHumerecognizedwasecclesiasticalgovernment,whichhetooktobeincompatiblewithfreedomofdiscussion.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian477Fortunatelythespreadoflearninghadensuredthatmanyofthegentryentertainednobleambitions:Afamiliaracquaintancewiththepreciousremainsofantiquityexcitedineverygenerousbreastapassionforalimitedconstitution,andbegatanemulationofthosemanlyvirtues,whichtheGreekandRomanauthors,bysuchani-matingexamples,aswellaspatheticexpressions,recommendtous(HE45.35,5:1819).TheHouseofCommonswastheonlypoten-tialrestraintonroyalpower,andHampdenandhisassociateswererighttorallytoitsdefense.ThesuccessoftheParliamentarycause,however,owedmoretobigotrythantogoodjudgmentandnoblesentiment:Andthoughitwasjustlyapprehended,thatsuchprece-dents[astheimprisonmentofHampden],ifpatientlysubmittedto,wouldendinatotaldisuseofparliaments,andintheestablish-mentofarbitraryauthority;Charlesdreadednooppositionfromthepeople,whoarenotcommonlymuchaffectedwithconsequences,andrequiresomestrikingmotive,toengagetheminaresistanceofestablishedgovernment.Itwasanironythatwithoutreligiousbigotry(whichwastoprovidethestrikingmotive),libertymightneverhavebeenestablishedinEnglandin1640,andthat,withouttheinvasionofWilliamofOrangein1688,itmightneverhavebeensecured.Recognitionofsuchironieswas,Humebelieved,oneofthepleasuresofhistoricalstudy(HE53.1,5:24950;23.31,2:51819).HumethuswritesinfavorofthemixedconstitutiontowhichbothmoderateParliamentariansandRoyalistsaspiredin1641,andthatwasfinallyestablishedin1688.TowardtheendoftheHistory,hetellsusthattheEnglishhaveestablishedthemostentiresys-temofliberty,thateverwasknownamongstmankind(HE71.70,6:531),and,inoneofthevolumeswrittenlast,heclaimsthatthemainutilityofhistoryliesininstructingustocherishthepresentconstitution(HE23.45,2:525).Earlierhehadwrittenofthatsin-gularandhappyGovernmentwhichweenjoyatpresent.In1772,inafitofSpleenandIndignation,hestruckoutthewordsandhappy,buthisconsideredopinionwasthattheyshouldberestored,fortheEnglishGovernmentiscertainlyhappy,thoughprobablynotcalculatedforDuration,byreasonofitsexcessiveLiberty(HE49.34,5:114;HL2:2601).Thus,theEnglishhadestablishedmorelibertythanhadeverexistedatanyothertimeorinanyotherplace;toomuchliberty,perhaps,forthestabilityoftheirpoliticalsys-tem,butnottoomuchforthephilosopherorthehistorianashesoughttogoabouthisbusiness.Hume,inimpartiallyweighingtheCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n478davidwoottonmeritsofWhigandToryprinciples,wastakingfulladvantageofthatlibertyofthepressthatwasuniquetoEngland.TheconclusiontotheHistorymightwellhavebeenthesameastheepigraphtotheTreatise,afamoussentencefromTacitus:Raratemporumfelicitas,ubisentire,quævelis;&quæsentias,dicerelicet(Theraregoodfortuneofatimewhenyoumaythinkwhatyoulikeandsaywhatyouthink;Histories1.1).Humemayhavebeenincreasinglyunsurehowmuchlibertywascompatiblewithstability;heneverdoubtedthatitwastheperfectionofcivilsociety.suggestionsforfurtherreadingInadditiontotheworkscitedinthenotestothisessay,theeditorsrecom-mendthefollowing.48Hume,David.DavidHume:AHistoryofEngland.Avariorumedition,editedbyF.vanHolthoon.Charlottesville,VA:InteLexCorporation,2000.AdigitaltextofHumesHistoryofEngland.ArticlesCohen,Alix.InDefenceofHumesHistoricalMethod.BritishJournalfortheHistoryofPhilosophy13(2005):489502.Evnine,Simon.Hume,ConjecturalHistory,andtheUniformityofHumanNature.JournaloftheHistoryofPhilosophy31(1993):589606.Perinetti,Dario.PhilosophicalReflectiononHistory.InTheCambridgeHistoryofEighteenth-CenturyPhilosophy,editedbyK.Haakonssen.2vols.NewYork:CambridgeUniversityPress,2006.2:110740.Phillips,MarkSalber,andDaleR.Smith.CanonizationandCritique:HumesReputationasaHistorian.InTheReceptionofDavidHumeinEurope,editedbyP.Jones.London:ThoemmesContinuum,2005.299313.Slater,Graeme.HumesRevisionsoftheHistoryofEngland.StudiesinBibliography45(1992):13057.Spencer,MarkG.HumeandMadisononFaction.WilliamandMaryQuarterly59(2002):86996.Stewart,M.A.HumesHistoricalViewofMiracles.InHumeandHumesConnexions,editedbyM.A.Stewart.Edinburgh:EdinburghUniversityPress,1994.171200.48Fortheirassistanceincompilingthislist,theEditorsgratefullyacknowledgeDarioPerinettiandMarkSpencer.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHume:TheHistorian479VanHolthoon,F.L.Humeandthe1763EditionofHisHistoryofEngland:HisFrameofMindasaRevisionist.HumeStudies23(1997):13352.Wood,Paul.TheNaturalHistoryofManintheScottishEnlightenment.HistoryofScience28(1990):89123.BooksFieser,James,ed.EarlyResponsestoHumesHistoryofEngland.InEarlyResponsestoHume,2nded.10vols.Bristol:ThoemmesContinuum,2005.Vols.78.Foster,StephenPaul.MelancholyDuty:TheHume-GibbonAttackonChristianity.Dordrecht:KluwerAcademic,1997.Herdt,JenniferA.ReligionandFactioninHumesMoralPhilosophy.Cam-bridge:CambridgeUniversityPress,1997.Phillips,MarkSalber.SocietyandSentiment:GenresofHistoricalWritinginBritain17401820.Princeton,NJ:PrincetonUniversityPress,2000.Spencer,MarkG.,ed.HumesReceptioninEarlyAmerica.2vols.Bristol:Thoemmes,2002.2:1182.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nj.c.a.gaskin15HumeonReligionHumescritiqueofreligionandreligiousbeliefis,asawhole,sub-tle,profound,anddamagingtoreligioninwaysthathavenophilo-sophicalantecedentsandfewsuccessors.Someofthedamageandalittleofthesubtletywill,Itrust,becomeevidentinPartIIofthisessay,whereHumesseminaldiscussionsofthedesignargumentfortheexistenceofGod,miracles,morality,andnaturalbeliefareexamined.Beforethis,however,certainpreliminariesneedatten-tion.First,thereisthedifficultycausedbytheold-fashionedorunfamiliarterminologyusedbyHumeandhiscommentatorsindescribingandassessingwhathehastosay.Second,althoughthescaleofHumeswritingonreligionisreasonablyobvious(itexceedshisoutputconcerninganyothersubjectexcepthistory),thefactthatitisdispersedoveranumberofpublicationsandpartlyembedded(sometimesnonetooclearly)inseveralmore,aswellashavingtobedrawnfromessays,letters,andminorwritings,needstobeunder-stoodbeforeanyinformeddiscussionispossible.Third,thereistheproblemofseeingwhathewrotenotasadhoccriticismsturnedoutpiecemeal,butasacomprehensivecriticalstrategy.Finally,aprob-lemofinterpretationresultsfromHumesabundantprudenceincoveringhisrealopinionswithambiguousironyandeven,onocca-sions,withdenialsofhisownapparentconclusions.IshallattemptsomeclarificationofthesefourpreliminaryissuesinPartIofthisessay,beginningwiththeterminology,amatterthatinfectsandinformsallelsethatcanbesaid.480CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion481i.textandcontextTerminologyAbasicclassificationofreligiousinformationaccordingtoitssourceinreasonorinhistoricallyparticulardisclosureshaslongbeenestab-lishedinthecontrastbetweennaturaltheologyornaturalreligion,ontheonehand,andrevelationorrevealedreligion,ontheother.Naturalreligion(thephraseusuallydoesdutyintheeighteenthcen-turyforthenowmorecommontermnaturaltheology)isthesystemofconclusionsaboutGods(orthegods)existenceandnaturesup-posedlyattainablefromevidenceandbyreasoningaccessibletoanyintelligentpersonirrespectiveofanyspecialinformationconveyedintheBible,Koran,orotherrevelatorysource.Forexample,thecon-clusionthatadesigningagent,notchance,isneededtoexplaintheorderofthecosmosispartofnaturalnotrevealedreligion.Revela-tionorrevealedreligion,ontheotherhand,isthebodyofallegedtruthsaboutthedivinethatcanbeobtainedonlyfromparticularhis-toricalandsupposedlyinspiredsourcessuchastheBibleorKoran.Forexample,theclaimsthatanindividualhumanpersoncanexpectresurrectionafterdeath,orthatGodoncesenthissonintotheworld,arepartsofarevelation.Adistinctionissometimesmadebetweenparticularrevelationandgeneralrevelation.Particularrevelationisrevelationasjustdescribed.GeneralrevelationisthesupposedgeneralexperienceofthepresenceofGodinthereligiouslifeofeachbeliever.Withinnaturalreligion,twotypesofargumentinvariousver-sionsare,andalwayshavebeen,conspicuous.ThesearecalledbyHume(andsomeotherswhousethepre-Kantianterminology)theargumentaposterioriandtheargumentapriori.TheargumentaposterioriisthephrasebymeansofwhichHumeusuallyreferstoversionsofwhatwewouldnormallycallthedesignargument,thatis,theargumentthatGodexistsbecausehiscreativeintelligencecanbeobservedintheorderorpurposivenesstobefoundinthenaturalworld(DNR2.5,9.1;143,188,forexample).TheargumentaprioriinHumesusagereferstohisparaphraseoftheparticularcosmologicalargumenttobefoundinSamuelClarkesBoyleLecturesfor1704,laterpublishedasADiscourseconcerningtheBeingandAttributesofGod.TheargumentinHumeswordsbeginsWhateverexistsCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n482j.c.a.gaskinmusthaveacauseorreasonofitsexistence,andconcludeswiththeclaimWemust,therefore,haverecoursetoanecessarilyexis-tentBeing,whocarriesthereasonofhisexistenceinhimself;andwhocannotbesupposednottoexistwithoutanexpresscontradic-tion(DNR9.3,1889).Boththepositiverejectionofrevelationasasourceofreligiousknowledgeandoneofthepossibleconclusionsattainablefromtheargumentsandevidenceofnaturalreligioncanbereferredtoasdeism.Thetermdeismwaswidelyusedintheeighteenthcentury,butwithvaguemeaning,toindicateaviewofreligionthatheldthatourreliableknowledgeofGodisbasedonreasonalone(thatistosay,onnaturalreligionandnotonrevelation).ThetermisnotmuchusedbyHumeexcepttorejectitsapplicationtohimself.1Itisalsoatermusedtoindicatebelief(arrivedatfromreasoningalone)inagodwhosettheuniverseinmotionorcausedtheuniversetoexistandthenleftitalone.Anotherwayofexpressingthislimitedviewistosaythatdeismistheclaimtorationallysubstantiatedbeliefinagodlackingprovidence.Providence,whilesometimesusedasasynonymforGod,ismoreparticularlyusedtorefertothataspectofGods(orthegods)naturethatconsistsinexertingcontrol,guidance,orforethoughtinthemoralaffairsofhumanbeingsorthephysicalprocessesoftheworld.HumeusesprovidenceinjustthissenseinSection11oftheEnquiryconcerningHumanUnderstanding,wherehearguesthatthereisnoevidenceforGodsprovidence.WhentheGod(eitherontheevidenceofrevelation,ornaturalreligion,orinsomeotherway)isheldtobeasingleandeternalGodwhocreatedallthings(possiblyexnihilo)andcontinuestosustainandworkwithinhiscreation(thatis,toexerciseprovidence),thebeliefisusuallycalledtheism.ThusthecommonrootoftheJudaic,Christian,andIslamicreligionsistheism.TwocorruptionsofreligionwereofgreatconcerntoHumeandothereighteenth-centurywriters.Thesecorruptionsweresupersti-tion,usuallyassociatedwithidolatryandwiththeChurchofRome,andenthusiasm,usuallyassociatedwiththenewlyconvertedandwithextremeProtestantsects.Superstitionisthestateinwhichunknownevilsaredreadedfromunknownagents.Itssourceis1SeemyHumesAttenuatedDeism,ArchivfurGeschichtederPhilosophie¨65(1983):16073.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion483weakness,fear,melancholy,togetherwithignorance,anditman-ifestsitselfinceremonies,observances,mortifications,sacrifices,presentsdirectedtowardtheunknownagent.Enthusiasmisreli-gioncorruptedbyemotionalfanaticismorreligiousmania:rap-tures,transports,andsurprisingflightsoffancythatareattributedtotheimmediateinspirationofthatDivineBeing,whoistheobjectofdevotion(E-SE23,734).2Anattitudetoreligionoftenassociatedinthelateseventeenthandeighteenthcenturieswithenthusiasm,butalsohavingapedi-greethatincludesTertullian,Pascal,andpossiblySt.Paul,wassometimesreferredtobyHumesnearcontemporariesasimplicitbelieforblindbelieforthesubmissionofreasontofaith.Inthenineteenthcenturythisattitudewasdevelopedintothepositionknownsinceabout1870asfideism.Thisistheview,arguedbysomeChristianapologiststobereinforcedbyHumesskepticism,thatreligiousbeliefisjustifiedbyfaithalone,quiteapartfromreasonsorevidence,becauseallknowledgerestsonpremisesacceptedbyfaith.Finallytherearetwoconfusingtermsthatbothcontainthewordnatural,butthatareusedindifferentsenses:theeighteenth-centurytermnaturalhistoryandthetwentieth-centurytermnaturalbelief.Naturalhistory(asinHumesNaturalHistoryofReligion)indicatesanaccountofsomethingasanaturalphenomenon.InthetitleofHumeswork,theaccountisofthecausesandconditionsthatnat-urallyproducereligion(as,forexample,thepresenceofairandwaternaturallyproducesrustoniron)withoutreferencetoanyreasonsthatcanbeproducedinfavoroforagainstthereligioninquestion.NaturalbeliefontheotherhandisnotaphrasetobefoundinHumesownwritings.ItisaphraseintroducedbyNormanKempSmith,3andmuchusedsince,toindicatebasicorindispens-ablebeliefs.2FormoreonHumesdistinctionbetweensuperstitionandenthusiasmandtruereligion,seehisunpublishedPrefacetowhathasbecomeVol.6ofhisHistoryofEngland,quotedinfullinErnestCampbellMossner,LifeofDavidHume(Edin-burgh:ThomasNelson,1954),3067.Foradiscussionofthepoliticaldimensionsofthedistinction,seeinthisvolumetheessayTheStructureofHumesPoliticalTheory.3NormanKempSmith,ThePhilosophyofDavidHume(London:Macmillan,1941),particularly10437,44263.Seealsonote25below.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n484j.c.a.gaskinTheWorksHumestwomainworksdirectlyonreligionaretheNaturalHis-toryofReligion(1757)andtheDialoguesconcerningNaturalReli-gion(1779,butfirstwritteninthe1750s).Theformerdealswithreligionsnaturaloriginsinhumannatureandsocietyitscauses.ThelatterexaminesthesupposedrationalgroundsforbeliefinGodorgodsitsreasons.PartsofthelatterexaminationhadalreadybeengivenapreliminaryruninSection11oftheEnquiryconcern-ingHumanUnderstanding(1748)inwhichtherealsoappearedthechapterOfMiracles,Humescelebratedonslaughtonthecreden-tialsoftheChristianrevelation.ButtheEnquiryasawholealsodevelopsanepistemologicalattackonmetaphysicsandphilosoph-icalreligionwhosefinaloutcomeisnotunliketheconclusionsoftwentieth-centurylogicalpositivism(EHU12.2434).Lessobvi-ously,thesecondEnquiry,theEnquiryconcerningthePrinciplesofMorals(1751),isalsoconcernedwithreligion.InitHumegivesanaccountofamoralityinwhichwhatisaddedbyreligiontothesecularcorealltoooftenamountstospuriousvirtuesandimaginarycrimesthatresultincruel,bigoted,andanti-utilitarianinterferencesinhumanaffairs.SomeoftheseinterferencesarechronicledinhisHistoryofEngland(publishedbetween1754and1762).AmongHumesfiftyorsoindividualessays,therearenumerousreflectionsonreligion.Theserangefromthelengthyfootnoteonthehypocrisyoftheclergy,whichisattachedtoOfNationalCharac-ters,tothedamagingaccountofthetwoconceptsdevelopedinOfSuperstitionandEnthusiasm.Thisaccountwouldseemtoleavelittleoftruereligiononcethetwocorruptionsofreligionhavebeenunderstood.Butbyfarthemostimportantessaysarethetwothatoughttohaveappearedin1757alongwiththeNaturalHistoryofReligion.TheseareOfSuicide(whicharguesthatsuicideisnei-therimmoralnorirreligious)andOftheImmortalityoftheSoul(whicharguesthatthereisgoodevidenceformansmortality).BothessayswerewithdrawnbyHumebeforepublicationafterthreatsagainsthimorhispublisher.Anauthorizedandsomewhatinaccu-rateprintingappearedin1777,theyearafterHumesdeath.HumesauthorizedtextispreservedintheNationalLibraryofScotland.Finally,lettersandshortdocumentsapart,thereistheTreatiseofHumanNature(173940).TheTreatise,Humesfirstwork,is,toCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion485oureyes,notovertlyconcernedwithreligion.PartofthereasonforthisisthatHumeexcisedsomeofitsnoblerpartsbeforepubli-cation,includingsomeversionofOfMiracles(whichcouldhavebeenlocatedinBook1,Part3,betweensections13and14),4andpossiblysomeversionofOftheImmortalityoftheSoul(whichcouldhaveformedtheconcludingpagestothesectionofBook1entitledOftheimmaterialityofthesoul).ButamoreimportantreasonthattheTreatiseaspublisheddoesnotseemtousmuchcon-cernedwithreligionisthatoursensitivitiesregardingwhatwouldconstituteanattackonreligionaremuchweakerthanthoseofHumescontemporaries.Thenatureoftheirsensitivitiesisillumi-natedbythepamphletALetterfromaGentleman.ThetextisdrawnfromaletterbyHumeandwasrushedintoprintonhisbehalfin1745whenhewasacandidatefortheChairofMoralPhilosophyatEdinburghUniversity.InthisbriefworkHumeisdefendedagainstsixcharges,fourofwhicharethattheTreatisesubvertsreligion(L78,1518,2135).Tousthemostobviouschargeisthatsomeofhisargumentsaboutcausation(particularlythesectioninBook1entitledWhyacauseisalwaysnecessary)constituteasignificantcriticismoftheaprioriargumentforGodsexistence.TheStructureofHumesCritiqueSupposeweputthefundamentalquestionthus:WhydoesanyonebelieveinGodorgods,orcleavetotheteachingsofsuchtheisticreligionsasChristianityorIslam?Theanswermaybegiven(nonex-clusively)intermsofeitherreasonsorcauses,anditisunderthisdivisionthatHumesexaminationofreligionbeginstolooklikeacomprehensivecritiqueratherthanacollectionofchallengingbutdiscretesections.Intraditional(andparticularlyeighteenth-century)religiousapologetic,thereasonsforbeliefinGodusuallytooktheformofappealstoargumentsandrevelation.Theappealtorevelationwasneithertothegeneralrevelationassociatedwithdedicatedreligiouspracticenortoindividualclaimstohavedirectinformationabout4ThesuggestionisendorsedinDavidWoottonsarticleHumesOfMiracles,inStudiesinthePhilosophyoftheScottishEnlightenment,ed.M.A.Stewart(Oxford,1990),199.WoottonismainlyconcernedwiththebackgroundinfluencesonHume.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n486j.c.a.gaskintheDivine,butspecificallytotheparticularrevelationofChristian-ityassetoutintheNewTestament.This,itwassupposed,carriedwithitcertainguaranteesofitsownauthenticity.Theseguaranteeswerethattherevelationfulfilledprophecyandwasattendedwithmiracles.MiraclescouldbebroughtaboutonlybyGod(andnotanygod,butonlybytheonetrueGod).ThereforearationalmanhadgroundsforacceptingtheChristianrevelationasgenuine.Itis,ofcourse,preciselythesegroundsthatHumesetouttoundermineinSection10ofthefirstEnquiry,whereincidentally,hetreatsfulfilledprophecyasaspeciesofmiracle(EHU10.41).TheappealtoargumentstosupportbeliefinGodwasmostcom-monlyanappealtothosetypesofargumentthatHumecallstheargumentapriori(cosmologicalarguments)andtheargumentapos-teriori(designarguments).Thesewerethetraditionalcoreofnaturalreligion.TheformerisdismantledbyHumeintheTreatise,thefirstEnquiry,andagaininPart9oftheDialogues.ThelatterissubjectedtoauniquelythoroughandhostileexaminationinSection11ofthefirstEnquiryandagainthroughouttheDialogues.Butif,asHumecontends,theargumentsofnaturalreligiondonotestablishtheexistenceofanydeitythatcouldbeanobjectofreligiousbelief,andifrevelationisnotauthenticatedinanywaythatcouldconvincearationalman,thenitmightseemthattheonlyanswerthatcanbegiventothequestionWhydoesanyonebelieveinGodorgods?isthatthebeliefhasnaturalcauses.AninvestigationoftheseisHumessubjectintheNaturalHistoryofReligion.Attheheartofhisanalysisisthecontentionthattheoriginofbeliefingodsistobefoundinfearoftheunknowncausesofthesometimesmalevolent,sometimesbenevolent,andfrequentlycapriciouseventsthatgovernhumanlife(NHR2.45,3.16,8.12,13.16,14.78).That,Ithink,isthemainstructureofHumescritiqueofreligion,butitsdetailsextendvastlyfurther:toamitigatedscepticism(carefullydevelopedinthefirstEnquiry)thatwouldputreligiousmetaphysicsbeyondourunderstanding;toasharpenedtheologi-caldilemma(EHU8)betweenGodsomniscienceandmansmoralanswerability;toaseparationofmoralityandreligion(impliedintheTreatiseandemphasizedbythesecondEnquiry)withcommentsonparticularissuessuchassuicide;toaphilosophicalaccountofper-sonalidentityandofthesoul(T1.4.56),whichinvitestherejectionCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion487ofimmortalitycontainedinOftheImmortalityoftheSoul;toanexposeinthe´HistoryofEnglandofthemiseryproducedbyreli-giousfanaticismandsuperstition;andontolettersthatcontainallmannerofdetailedcommentsandcriticisms(note,forexample,hisremarksonthepsychologyofworshipandtheinappropriatenessofprayerinNHL13).HumesStanceandtheProblemofInterpretationTheproblemwiththeinterpretationofHumeisthat,althoughhisargumentsandthefactsheadducesareregularlyhighlycriticalofreligionanddamagingtoanybeliefinthedivine,hisaffirmations(andsometimestheconclusionsthatheseemstodraw)donotalwayslookliketherealoutcomeofhiscriticisms.Thus,forexample,theNaturalHistoryofReligionreadslikeareductionofreligiontoitscausesinhumannature,butinhisbriefIntroductiontotheworkHumeremarks:Thewholeframeofnaturebespeaksanintelligentauthor;andnorationalenquirercan,afterseriousreflection,suspendhisbeliefamomentwithregardtotheprimaryprinciplesofgenuineTheismandReligion(NHRIntro.1).Similaraffirmationsappearatleastfivetimesinthemaintext.Seventeenyearsearlier,inafootnotetotheAppendixtotheTrea-tise,HumehadunequivocallycountermandedwhateverdamagetobeliefinGodtheTreatisemighthavebeensupposedguiltyof:Theorderoftheuniverseprovesanomnipotentmind;thatis,amindwhosewillisconstantlyattendedwiththeobedienceofeverycrea-tureandbeing.Nothingmoreisrequisitetogiveafoundationtoallthearticlesofreligion(T1.3.14.12,n.30).5AndyettheargumenttoGodsexistencefromtheorderoftheuniverse,describedinthefirstEnquiryasthechieforsoleargumentforadivineexistence(EHU11.11),isthere,andagainandmostcelebratedlyintheDialogues,subjectedtodevastatingcriticism.Mostparadoxicallyofall,thiscriticismisitselffollowedbyanaffirmationfromtheskepticPhilothatApurpose,anintention,ordesignstrikeseverywherethemostcareless,themoststupidthinker(DNR12.2,214).AnotherinstanceofHumesargumentsapparentlybeingatoddswithhisconclusion5ThisnotewasaddedbymeansoftheAppendixpublishedin1740,attheendofthethirdandfinalvolumeoftheTreatise.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n488j.c.a.gaskinisinhisonslaughtonmiracles.Therehisattackonthecredentialsofrevelationconcludeswitharecommendationtofaith:OurmostholyreligionisfoundedonFaith,notonreason(EHU10.40).Some-whatsimilarly,hisaphoristicdemolitionofthegroundsforbelievinginimmortalityinOftheImmortalityoftheSoulbeginsandendswitharecommendationtothegospelalone,thathasbroughtlifeandimmortalitytolight(E-IS1,43,590,598).Itisnotpossiblewithinpresentconstraintsofspacetodiscusstheseissuesinfull.Theyarecomplexandhave,moreover,alreadybeenexaminedatdepthinrecentHumeanexegeses.6Butanoutlineinterpretationwillbeuseful.InthefirstplaceneitherHumenoranyotherwriterineighteenth-centuryBritain(orelsewhereinEuropeforthatmatter)wasfreetoexpressatheisticalorantireligiousviewswithoutthethreatoforactualprosecutionorsocialpenaltiesofaverynastysort.HencewewouldexpectHumetocoverhisappar-entlyskepticalviewswithprotestationsoforthodoxywithwhichhecoulddefendhimselfwhenneedarose.Inthisheisincompanywithmostothereighteenth-centuryexpressionsofreligiousskepti-cismoratheism.7Hisisolateddirectiontofaithasthefoundationofourmostholyreligionisthusalmostcertainlyadefensiveironyfollowingonhisattackonmiracles,oraruefulacknowledgmentoftheultimateirrationalityofreligiousbelief,notasincerefideisticdefenseofwhatreligiousbeliefreallyis.Itwouldalsobepossi-bletoconstruesomeofhisblanderaffirmationsofbeliefinGodasthedesignerinthisway,particularlythefulsomeandthenhighlyqualifiedconcessionbytheskepticPhiloinDialogues,Part12.Buthavingacknowledgedtheprudentialirony,thereremainsanimpression,fromthecarefulcomplexityofhisarguments,fromhisskepticismaboutmetaphysicalarguments,andfromlettersandanecdotalevidence,thatHumereallywasunwillingtodenytheexistenceofGodandalllessersupernaturalagentsintheunequivo-calsensenowconveyedbythenotionofatheism.Itisasifhewastooconsistentaskeptictopronouncepositivelyonanyremoteand6Forafulldevelopmentoftheinterpretationthatfollows,seemyHumesPhilosophyofReligion,2nded.(London:Macmillan,1988).ForfurtherworksinthesameareaandotherrecentexegesisoftheDialogues,seenote8below.7ThematterisdocumentedbyDavidBermaninhisAHistoryofAtheisminBritain(London:CroomHelm,1988).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion489abstrusesubjects(EHU1.12),atheismincluded;and,moreover,itisasifthecloserhelookedatthedefectsofthedesignargument,themoresomethingofitremainedunrefuted,sothat,attheendoftheDialogues,inaparagraphveryprobablyaddedtotheinitialversionofthemanuscriptin1757,hecanwrite,surelywithouthintofirony:Ifthewholeofnaturaltheology,assomepeopleseemtomaintain,resolvesitselfintoonesimple,thoughsomewhatambiguous,atleastundefinedproposition,thatthecauseorcausesoforderintheuniverseprobablybearsomeremoteanalogytohumanintelligence:Ifthispropositionbenotcapableofextension,variation,ormoreparticularexplication:Ifitaffordnoinferencethataffectshumanlife,orcanbethesourceofanyactionorforbearance:Andiftheanalogy,imperfectasitis,canbecarriednofartherthantothehumanintelligence;andcannotbetransferred,withanyappear-anceofprobability,totheotherqualitiesofthemind:Ifthisreallybethecase,whatcanthemostinquisitive,contemplative,andreligiousmandomorethangiveaplain,philosophicalassenttotheproposition,asoftenasitoccurs;andbelievethatthearguments,onwhichitisestablished,exceedtheobjectionswhichlieagainstit?(DNR12.33,227)8SoIwouldsuggestforworkingpurposesthatoneshouldtakeasprudentialironyHumesaffirmationswheretheyareblandlyatvariancewithanystraightforwardreadingofwhatprecedesorfol-lowsthem.Butoneshouldalsoacknowledgethathisregardforthelimitationsofhumanunderstanding,andhiscautionconcern-ingsoextraordinaryandmagnificentaquestionasthebeingandnatureofGod(DNR12.33,227),makehimgenuinelyunabletoadvocatestraightforwardatheismofthesortlaterassociatedwithDHolbachorRussell.Thushisskepticismaboutalltheologicalandotherclaimsbasedonabstrusemetaphysicsdoesnotattheendpermithimtorejectintotoobviousclaimsbasedontheorderapparentintheuniverse.Buttheseobviousclaimsamounttoverylittleasfarasanyrealreligionisconcerned.Theyimplynodutiesandnoactionorforbearancefromaction.Theyinvolveno8Smithsupposesthisparagraphwasaddedin1776,butM.A.Stewartshowsthatthepaper,watermarks,andspellingonthepageonwhichtheparagraphwaswrittenallsupporta1757date.SeeM.A.Stewart,TheDatingofHumesManuscripts,inTheScottishEnlightenment:EssaysinReinterpretation,ed.P.B.Wood(Rochester,NY:UniversityofRochesterPress,2000),267314,esp.288305,3078,andFigure14.IamindebtedtoProf.Stewartforcorrespondencebearingonthisquestion.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n490j.c.a.gaskindevotion.Ihaveelsewheresuggestedthatsuchanemasculatedcon-cessiontothepropositionthereisagodshouldbecalledattenu-ateddeism.9Thisisdeisminwhichsuchevidenceandreasonsasremainuncontrovertedadduptonomorethanadimpossibilitythatsomenonprovidentialgodexists,apossibilitytooill-understoodtobeaffirmedordeniedbyawiseman.Butwhetherthedesignationattenuateddeismisappropriateornot,itisHumesactualargumentsthatcontributetothephi-losophyofreligion,togetherwiththeexcitementofthechallengesthathebringstobearonquestionsconcerningreligionandtheexis-tenceofGodorgods.TheseargumentsandchallengesforthemostpartstandorfallontheirownphilosophicalmeritswithoutneedtorefertoHumesownhard-to-identifystance.Inwhatfollows,andforpresentpurposes,IshallthereforetakeHumetobeidentifiedwithanyinterestingpositionsetoutinhisownworks.10ii.argumentandoutcomeTheCoreofNaturalReligionInthefirstEnquiryHumereferstothedesignargumentasthechieforsoleargumentforadivineexistence(EHU11.11).Heisherenotmakingajudgmentbutreportingafact.Therearestrong9SeemyHumesPhilosophyofReligion,21923.10TheonlycasewherethismayneedspecialjustificationistheDialogues,wherethethreespeakersareinevidentconflictanditisnotalwaysclearwhohasthebetteroftheargument.GiventhatHumesmodelisCicero,andthebalancedpresentationexemplifiedbyDeNaturaDeorum,andnotPlato,with(generally)hispro-Socraticyes-men,itisstillpossibletosaythat(a)DemeaspeaksverylittleforHumeandingoodpartforboththehighrationalismofSamuelClarkeand,somewhatper-versely,forsuchblindbeliefasHumeadmitsintothediscussion;(b)CleanthesspeakssomewhatmoreforHume,especiallywhenheopposesDemea,butmostlyforthemoderaterationalistsandusersoftheaposterioriargumentswhosebestknownrepresentativeisJosephButler;(c)PhiloisclosesttoHumesmouthpiecebutinclinedtooverstatehispositionsothatretreatisoccasionallypossible.ThereisaconsiderableliteratureontheinterpretationoftheDialogues.SeeinparticularKempSmithsIntroduction,Dialogues,ed.N.K.Smith(London:ThomasNel-son,1935);JamesNoxon,HumesAgnosticism,PhilosophicalReview73(1964):24861;myHumesPhilosophyofReligion,Chap.12;NelsonPike,HumeontheArgumentfromDesign,DialoguesconcerningNaturalReligion,ed.N.Pike(Indianapolis:Bobbs-Merrill,1970).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion491hintsoftheargumentintheBible.11ItplayedasignificantpartinGreekphilosophicalmonotheism.12InitsteleologicalversionitappearsastheFifthWayofAquinas.Ineighteenth-centuryliteratureitssoundnessisvirtuallytakenforgranted,andthesameappliesformuchnineteenth-centuryliterature.Ithasevenenjoyedsomerehabilitationinthetwentiethcentury.13Apartfromnumerousshortreferences,Humeattemptsthreestatementsoftheargument:141.Thereligiousphilosophers...paint,inthemostmagnificentcolours,theorder,beauty,andwisearrangementoftheuniverse;andthenask,ifsuchagloriousdisplayofintelligencecouldproceedfromthefortuitousconcourseofatoms,orifchancecouldproducewhatthegreatestgeniuscanneversufficientlyadmire.(EHU11.10)2.Thecuriousadaptingofmeanstoends,throughoutallnature,resemblesexactly,thoughitmuchexceeds,theproductionsofhumancontrivance;ofhumandesign,thought,wisdom,andintelligence.Sincethereforetheeffectsresembleeachother,weareledtoinfer,byalltherulesofanalogy,thatthecausesalsoresemble,andthattheAuthorofnatureissomewhatsimilartothemindofman;thoughpossessedofmuchlargerfaculties,proportionedtothegrandeurofthework,whichhehasexecuted.(DNR2.5,143)3.Consider,anatomizetheeye:Surveyitsstructureandcontrivance;andtellme,fromyourownfeeling,iftheideaofacontriverdoesnotimmediatelyflowinuponyouwithaforcelikethatofsensation.(DNR3.7,154)Ineffect(althoughIdonotthinkitcanbeshownthatHumeintendedanythingsosystematic)thesethreestatementsshowthedesignargumentindistinctversions.In(1),theargumentispre-sentedaswhatIhaveelsewheresuggestedshouldbecalledthenomologicalargument,15namely,astheappealtotheinexplicability11Note,e.g.,Psalm19;2Maccabees7:28;Romans1:20.12See,e.g.,Xenophon:Memorabilia,1.4,67;Plato:Timaeus,47;Cicero:DeNaturaDeorum,2.345.13NoteparticularlytheremarkablymodernsoundingTheFitnessoftheEnvironmentbyL.J.Henderson(NewYork:Macmillan,1913),and,amongmorerecentbooks,AlvinPlantinga,GodandOtherMinds(Ithaca,NY:CornellUniversityPress,1967),andR.G.Swinburne,TheExistenceofGod(Oxford:ClarendonPress,1979),Chap.8.14ItisdifficultevennowtoidentifyanydefinitivestatementofthedesignargumentandweknowHumeencounteredthesameproblem.SeeHL1:155.15SeeAnEncyclopaediaofPhilosophy,ed.G.H.R.Parkinson(London:Routledge,1988),33942.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n492j.c.a.gaskinofnaturalorderifthisisnotaccountedforastheoutcomeofintelli-gentdesign.In(2),Humemakesaverycarefulattempttorepresenttheformofwhatisusuallycalledtheteleologicalargument:theappealtothesignificanceofthepurposessupposedlyevidentinnat-uralphenomena.In(3),Cleanthes,theadvocateoftheargumentintheDialogues,isnotsomuchpresentinganewversionoftheargumentassuggestingthatitsconclusionissomethingvergingontheperceptuallyobvious.Wecannotseethestructuresofnature,orbecomeawareoftheallpervadingregularitiesweexpressaslawsofnature,withoutfeelingtheirsourceasintelligent.Thequestionthenbecomeswhetherthisfeelingisjustifiablyrelatedtowhatelicitsit(likeourfeelingoffearaboutatomicradiation)orunjus-tifiablyrelated(likesomepeoplesfeelingoffearaboutdarknessperse).IntheDialoguesandthefirstEnquiry,Section11,Humesubjectstheseargumentstoanintricateandcumulativelydevastatingseriesofobjections,themajorityofwhichapplytoboththenomologicalandtheteleologicalarguments.Hismainobjectionsareasfollows:a.IfwesupposeGod(orgods)tobethecauseoforderintheworld,thensinceallthatwecaninferaboutGod(orgods)isinferredfromtheworld,wecanonlyattributetoGod(orthegods)whateverdegreeofpower,intelligence,foresight,andsoforthissufficienttoproducewhatweactuallyfindintheworld(EHU11.1223).Inpar-ticular,Humeargues,whenappliedtodivineprovidence,thatitisimpossibletoinferfromtheworldinfiniteorevenverygreatbenev-olenceinitsdesigner(DNR1011).AsPhilointheDialoguesputsit,Thewholepresentsnothingbuttheideaofablindnature,impreg-natedbyagreatvivifyingprinciple,andpouringforthfromherlap,withoutdiscernmentorparentalcare,hermaimedandabortivechil-dren(DNR11.13,211).HumeisalsoatpainstopointoutintheDialoguesandfirstEnquirythatwemaytortureourbrains(toadaptHumesphrasefromEHU11.17)intoreconcilingthesufferingofliv-ingthingswiththepresuppositionthatGodisperfectlybenevolent,butwhatwecannotdoisjustifythatpresuppositionbyinferencefromthegivensuffering.b.Ifvalid,theinferencefromdesigncouldequallywellestablishanumberofconclusionsincompatiblewithmonotheism:forexam-ple,thattheuniverse,likemosthumancontrivance,istheproductCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion493ofcooperatingdesigners;thatitisadiscardedexperimentinuniversemakingortheproductofasecond-rategod;thatitisthecreationofadeisticgod,thatis,onewhohassetitallgoingandthenletitrunonatitsowndevices;andsoon(DNR5,1669).c.Humearguesextensivelyinhisgeneralphilosophythattheconceptofcauseappliesonlytokindsofobject.ThatistosaywehavewarrantforcallingAthecauseofBonlyifobjectsofkindAhaveconstantlyoccurredpriorto,andcontiguouslywith,objectsofkindB.Whenthishashappened,webothhaveevidencethat,andbelievethat,AscauseBs.Giventhisanalysisofcausation,itmakesnosensetotalkaboutauniqueobjectsuchastheuniverseasawholebeingcausallyproducedbyauniqueandotherwiseunknownentityoutside(inthesenseofnotbeingoneamong)therepeatingcausalsequencesoftheuniverseitself.Evidentlywecannothaveexperienceofaneventoreventspriorto,contiguouswith,andcon-stantlyconjoinedwiththebeginningoftheuniverse(EHU11.30;DNR2.248,149ff.).d.Theanalogy,Humecontends,betweenartifactsobjectsknowntoproceedfromdesignandnaturalobjectsistooweakandremotetosuggestsimilarcauses.(ThisobjectionisdevelopedalmostthroughouttheDialogues.)e.Therelationbetweenorderanddesignisexperiencebased:order,arrangement,ortheadjustmentoffinalcausesisnotaprioriproofofdesign,itisindicativeofdesignonlysofarasithasbeenexperiencedtoproceedfromthatprinciple(DNR2.14,146).f.Thedesignargumenttakesitasobviousthattheorderinnature(ormoregenerallythematerialworldperse)requiresanexplana-tion:thatitcannotitselfbeacceptedastheultimatebrutefact.But,Humeargues,ifthisisso,andtheexplanationofnaturalorderhastobesoughtinsomethingotherthanitselfnamely,intheactivityofanintelligentagenttakentobewhattheisticreligionscallGodthenwhatisitaboutthisexplanationwhichlegiti-matelyblocksthequestforfurtherexplanation?Butifwestop[attheagentexplanation]...whygosofar?Whynotstopatthemate-rialworld?(DNR4,161).16Humesquestionisabsolutelycrucial.16ForacriticaldiscussionofthisandsomeofHumesotherobjectionstothedesignargument,seeR.G.Swinburne,TheArgumentfromDesign,Philosophy43(1968):199212.SeealsoL.Dupre,TheArgumentfromDesignToday,JournalCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n494j.c.a.gaskinItchallengesthemostfundamentalofallJewish,Christian,andIslamicassumptions:thatthereissomethingaboutGodthatmakesitmorereasonabletostopallwhy?andwherefrom?questionswiththatentitythanwiththemostfundamentalrealitiesofthematerialworlditself.Eachoftheaboverequires,andisgivenbyHumeandthesec-ondaryliteraturewhichfollowshisagenda,detailedphilosophicaldiscussionthatcannotbeattemptedhere.Butatleastoneotherimportantandhighlyoriginalcountertothedesignargumentrelatedto(f)aboveissuggestedbyHume.Wecanallagreethattheinfer-encetoadesignerdependsontheassumptionsthat(1)theorderinnatureneedsexplanationandthat(2)noexplanationispossibleotherthanbyreferencetosomedesigningintelligence.InPart8oftheDialoguesboththeseassumptionsarequestioned.ThefirsthadbeenclassicallychallengedbyatenetoftheEpi-curean(orGreekAtomist)systemthatattractedmuchancientridiculeandcriticism.ThiswasEpicurusscontentionthattheworldjusthappenedbytheunguidedcollisionandgroupingofnumberlessprimaryparticlestakingplaceoveraninfinitetimeininfinitespace.ThusBalbus,oneofCiceroscharactersinhisdialoguesinDeNaturaDeorum(2.37),deridestheEpicureans:Shoulditnotastoundmethatanyone...canpersuadehimself...Thataworldoftheutmostsplendourandbeautyiscreatedbyanaccidentalcom-binationofthose[primaryparticles]?Idonotseehowthepersonwhosup-posesthatthiscanhappencannotalsobelievethatifcountlessinstancesofthetwenty-oneletterswerethrownintoacontainer,thenshakenoutontotheground,itwerepossibletheymightformareadableversionoftheAnnalsofEnnuis.Imnotsurethatluckcouldmanagethistotheextentofasingleline!ThefundamentalclaimagainsttheEpicureansisthatorder,beauty,andthearrangementoftheuniverseneedexplanation,andrandomcollisionsofinfinitenumbersofprimaryparticlesdonotprovideaprobableone.Thesameclaimiselicitedbytheseventeenth-centuryrevivalofEpicureanatomism.ofReligion54(1974):112;GaryDoore,TheArgumentfromDesign:SomeBetterReasonsforAgreeingwithHume,ReligiousStudies16(1980):14561;andmyHumesPhilosophyofReligion,Chaps.23.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion495In1682,atOxford,atranslationintoEnglishverseofLucretiusssixbooksofEpicureanphilosophyappeared,tobefollowedbyparaphrasesbyDrydenandothers.Anangryreactiontothepopular-ityofsuchanirreligiousworkfollowed,andin1712thereappearedanansweronanepicscale:TheCreationbySirRichardBlackmore.AtseveralpointsBlackmoreconfrontstheEpicureanaccountoftheoriginoftheordereduniverseinpreciselythemannerinwhichithadbeenconfrontedbyBalbusinCicerosdialoguesixteenhundredyearsearlier:CouldAtomes,whichwithundirectedflightRoamdthrotheVoid,andrangdtheRealmsofNight;OfReasondestitute,withoutIntent,DeprivdofChoice,andmindlessofEvent,InOrdermarch,andtotheirPostsadvanceLedbynoGuide,butundesigningChance?Thechallengeisagainclear:theordermanifestedbytheuniverseneedsexplanation.Butdoesit?Humeisinclinedtoanswersee(f)abovethatitdoesnot;orrather,ifwethinkitdoes,thenhavingtraceditsorigintoadivineorderer,theorderinthatoughtjustasmuchtorequireexplanationastheorderinmatter:Tosay,thatthedifferentideas,whichcomposethereasonofthesupremeBeing,fallintoorder,ofthemselves,andbytheirownnature,isreallytotalkwithoutanyprecisemeaning.Ifithasameaning,Iwouldfainknow,whyitisnotasgoodsensetosay,thatthepartsofthematerialworldfallintoorder,ofthemselves,andbytheirownnature?Cantheoneopinionbeintelligible,whiletheotherisnotso?(DNR4,162)Humeaddsthatwehaveindeedexperienceofideas,whichfallintoorder,ofthemselves,andwithoutanyknowncause(presumablyourownideas),butwehaveamuchlargerexperienceofmatter,whichdoesthesame.AreplytoHumeisthatthereductionoftwosortsofautonomousorder,materialandmental,toone,mentalorder,effectsadesirableeliminationofasuperfluousexplanatoryentity.17Theproblemforthetheist,however,istoshow,against17ThisandrelatedmatterswereinvestigatedintheSwinburne-Oldingexchangeintheearly1970s.SeeA.Olding,TheArgumentfromDesignAReplytoR.G.Swinburne,ReligiousStudies7(1971):36173;R.G.Swinburne,TheArgu-mentfromDesignADefence,ReligiousStudies8(1972):193205;A.Olding,DesignAFurtherReplytoR.G.Swinburne,ReligiousStudies9(1973):22932.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n496j.c.a.gaskintheeverrisingtideofscientificevidence,thatmentalorder,andnotmaterialorder,hasexplanatoryprimacy:thatmaterialorderisexplainableintermsofmentalorderandnotviceversa,andthatmentalorderandmaterialorderaregenuinelydifferentcategories.Butthereisafurtherreasontothinkthattheordermanifestedintheuniverseisnotinneedofspecialexplanation.ThepointisthatCicerosBalbus,andBlackstone,andotherswhohavewalkedintheirfootstepshaveassumedthatinsomesenseachaoticuni-verseismoreprobable,atleastwouldbelessinneedofexplanation,thantheorderlycosmoswefind.Butthisassumptionisinneedofjustification.Why?Becausetheassumptionimpliesthatwecancomparetheorderedcosmosthatactuallyexistswithachaosthatdoesnotexist,andfindtheexistentcosmoslessprobablethanthenonexistentchaos.Butthecrucialpointisthatwecannotmakesuchacomparison.Wehaveabsolutelynogroundsforsupposingthatwhatactuallyexistshasanyprobabilityatallbycomparisonwithanythingelse,sinceinthisspecialinstancethereisnothingelse.Similarlywehaveabsolutelynogroundsforholdingthatorderinnatureismore(orless)inneedofexplanationthanchaoswouldhavebeen.Orderiswhatwehave,andthereisnothingelsebycon-trastwithwhichthatorderisinanysenseprobableorimprobable.EvenifatbestHumehimselfcandonomorethansheddoubtontheneedtoexplainnaturalorder,orontheusefulnessofdoingso,thesecondassumptionrequiredforinferencetoadesignerremains,namely,thatthereisnoexplanationofnaturalorderifwedonotattributeittoadesigningintelligence.However,we(butnotHume)mightbeabletoargue,inthelightofthebigbangtheoryfavoredbymoderncosmology,thattheinitialeventoutofwhichallsub-sequentsequencesofeventsemergedcould(atleastwehavenoreasonstothinkthatitcouldnot)havesetabsolutelyanysortofuniversedeveloping.Buthavingsetgoingthisuniverse,thosefirstdevelopmentswerecontinuouswithwhatwesubsequentlyreadasthelawsofnature.Theinitialeventhavingsetthingsgoinginoneway(thatis,thewayitactuallydid),thatonewayiswhatweseeasnaturalorder,andindeednoexistentthingscandevelopinanyotherwaygiventheinitialevent.ThereisevenahintofthistypeofthinkingintheDialogues(althoughitisarrivedatinasomewhatdifferentway):Insteadofadmiringtheorderofnaturalbeings,weshouldclearlysee,thatitwasabsolutelyimpossibleforthem,inCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion497thesmallestarticle,evertoadmitofanyotherdisposition(DNR6,175).ButinthepagesoftheDialoguesthatfollowthisremark,HumedevelopswithoutaidfromourbigbangtheoryanextensivereplytothetraditionalStoicandChristianassumptionthatordercouldnothaveemergedfromchaoswithoutintelligentdesign.Humesnewhypothesisofcosmogony(DNR8,183)isaformoftheEpicureantheoryrevisedbytheassumptionthatthenum-berofprimaryparticlesofmatterisverylargebutnot,asEpicurussupposed,infinite.Suppose,saysHume,inapassageofremarkableinsight,matterwerethrownintoanyposition,byablind,unguidedforceandthatthisforcewasnotexhaustedatthemomentofthefirstthrow,butremainedactiveineverypartofmattersothatmove-mentcontinued(DNR8,184).Iswhatweactuallyfindnamely,stablestructurescomposedofdisorderlyprimaryparticlesnotapossibleoutcomeofsuchafiniteamountofmatterundergoingtranspositionsoveraverylongperiodoftime?Inparticular,certainstructuresandsequences,oncestruckon,willbeofacharacterthatenablesthemtoendure.IneffectHumeissuggestingthatgivenaninitialblindforce(abigbang?)subsequentrandommovementsofalargebutfiniteamountofmattercouldproducethestableentitiesandsequenceswenowobserveintheuniverse.Lawsofnatureandinorganicstructures,justasmuchasnaturalspecies,couldbearrivedatbyaprocessakintothatofnaturalselection:Itisinvain,therefore,toinsistupontheusesofthepartsinanimalsorvegetables,andtheircuriousadjustmenttoeachother.Iwouldfainknowhowananimalcouldsubsist,unlessitspartsweresoadjusted?(DNR8,185).Evenifthebestreadingoftheavailableevidencewouldnowseemtoshowthatthemostfundamentallawsofnaturehavenotevolved,buthaveoperateduniformlyfromtheremotestaccessiblepast,Humesnewhypothesis(DNR8,183)remainsastonishinglyimpressiveasanattempttoprovideanalternativetothereligioushypothesis(EHU11.18).Itis,moreover,anattemptthat,whenfleshedoutbyDarwinsobservations,vastlydevaluestheteleologi-calargumentevenifthenomologicalargumentpartiallyescapes.WhatthenisHumesachievementinthisarea?Attheveryleastheputamassiveandpermanentquestionmarkagainstacrucialpieceofreligiousapologeticpreviouslytakenasunquestionable.IntheprocesshebrilliantlyanticipatedlaterideasandestablishedCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n498j.c.a.gaskinthegroundsonwhichallsubsequentphilosophicaldiscussionshavetakenplace.TheCredentialsofRevelationTwo-and-one-halfcenturiesafteritspublication,OfMiracles,Sec-tion10inthefirstEnquiry,isstillspawningbook-lengthresponsestogetherwithanunabatedstreamofdiscussionarticles.18Indeed,OfMiraclesismanifestlyoneofthoserarephilosophicalpieceswhoseveryinconsistenciesandambiguitiesaremorefruitfulthanthecautiousbalanceofathousandlesserworks.Itsmainstructureissimple.InPart2anumberofcasehistoriesandwhathavebeencalledaposterioriargumentsarereproduced(forthemostparttheyarenotoriginaltoHume)toshowthatthereneverwasamiraculouseventestablishedonsofullanevidencethatthefalsehoodoftheevidencewouldbearealprodigy(EHU10.14).IneffectPart2isconcernedwiththecriteriaforgoodevidence,withthesignificanceofincompatiblereligiousclaimsbasedonrivalmiracles,19andwiththegeneralconclusionHumedrawsfromhisargumentsthatamiraclecanneverbeproved,soastobethefoundationofasystemofreligion(EHU10.36).Giventhecontemporarybackgroundofcontroversyconcerningmiracles,andtheuseofmiraclestovalidateaparticularrevelationofChristianity,IhavesuggestedthatthisguardedconclusionshouldbeunpackedasTheResurrectioncanneverbeprovedinsuchawaythatitcanfunctionasagoodreasontoaccepttheChristianrevelation.2018Thefirstbook-lengthreplywasWilliamAdams,AnEssayonMr.HumesEssayonMiracles(London,1752).GeorgeCampbellscritique,ADissertationonMiracles(Edinburgh,1762),elicitedfromHumeoneofhisrarephilosophicalreplies:seeHL1:34851.AninterestingrecentdiscussionofthegeneralphilosophicalproblemsconcerningreportedprodigiescanbefoundinC.A.J.Coady,Testimony:APhilo-sophicalStudy(Oxford:ClarendonPress,1992).SeealsotheworksbyEarman,Fogelin,andJohnsoncitedbelowinnote21.19SeemyContraryMiraclesConcluded,HumeStudies10(Suppl.1985):114.20ThecontroversyismostreadablydocumentedbySirLeslieStepheninhisHistoryofEnglishThoughtintheEighteenthCentury,Chap.4,Sect.4(originallypub-lishedin1876).TheareahasbeenmorethoroughlyandmorerecentlyreviewedbyR.M.BurnsinTheGreatDebateonMiracles(Lewisburg,PA:BucknellUniversityPress,1981).FortheinterpretationofOfMiraclesasacommentontheevi-dentialsignificanceoftheResurrection,seemyDavidHumeandtheEighteenthCenturyInterestinMiracles,Hermathena99(1964):8092.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion499InPart1anaprioriargument(socalledbycommentatorsonHume)isproducedtoactasacheckonsuperstition.Theargumentpurportstoshowthatnowiseman(thatisonewhosebeliefisproportionedtotheevidence)couldbelievereportsofmiracles.AparaphraseofHumesargumentisasfollows:1.Aweakerevidencecanneverdestroyastronger.2.Awisemanproportionshisbelieftotheevidence.3.Somethingshappeninvariablyinourexperience,forexam-ple,thatmendie.Inmattersoffacttheseinvariableexperi-encesconstitutecertaintiesandarecalled,orformthebasisof,lawsofnatureafirmandunalterable[unalterablebecausepast]experiencehasestablishedtheselaws.(EHU10.12)4.Otherthingshappenlessthaninvariablyinourexperience,forexample,thatonewillsurviveaheartattack.Inmat-tersoffactthesevariableexperiencesconstituteprobabilitieswhichadmitofdegreesrangingfromstrong(almostalwayshappens)toweak(veryseldomhappens).5.Theveracityofhumantestimonyis,fromexperience,nor-mallyastrongprobabilityandassuchamountstoaproofthatwhatisreportedtookplace.Butsometimestheveracityofhumantestimonyisaweakprobability(asisalwaysthecase,accordingtoHumesargumentsinPart2,withreportsofmiracles).Therefore,from3and4,whentestimonyisgiventhatiscontrarytoourinvariableexperience,aproba-bility,whetherweakorstrong,isopposingacertainty,and(from1and2)thewisemanwillbelievethecertainty.6.Butamiracleisatransgressionofalawofnature[see3]byaparticularvolitionoftheDeity.Therefore,Theremust...beauniformexperienceagainsteverymiraculousevent,otherwisetheeventwouldnotmeritthatappellation.Andasauniformexperienceamountstoaproof,thereishereadirectandfullproof,fromthenatureofthefact,againsttheexistenceofanymiracle.(EHU10.12andn.23)Theaboveargumenthasprovokedsomefundamentalquestions,includingthefollowing,whichcanonlybestatedhere:(1)Whatismeantbyalawofnature,andhowcanonedistinguishbetweenaneventthatfalsifiesalaw(showsthatitisaninaccuratedescriptionCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n500j.c.a.gaskinofthewaythingsareinthenaturalworld)andaneventthatresultsfromasuspensionofthelaworanintrusionintothenaturalworldbyasupernaturalagentsuchasagodorotherinvisiblespirit?(2)CanHume,onthebasisofwhathesayselsewhereintheTreatiseandfirstEnquiry,formulateanyconceptofnaturalcausationstrongenoughtogivecontenttothenotionofitsviolation?(3)IsHumesdefinitionofamiracle(whichisentirelyreportive)inneedofsupplementation,particularlybythequalificationofreligioussignificance,sothatmereinexplicablefreaksofnaturedonotgetcountedasmiracles?(4)IsHumecorrectinimplying(EHU10.1113)thatinorderforsomethingtobecalledamiracleitmustnothappenmorethanonce?Andif,asbiblicalreportswouldseemtosuggest,heisnotcorrect,atwhatstagewillrepeatedmiraclesbecomeclustersofpara-normalphenomenainneedofexplanationwithinthenaturalworld?(5)CanHume,oranyonearguingonhisbehalf,oronbehalfofthosewhoneedsuchaconceptintheirdefinitionofwhatamiracleis,giveadequatecontenttothenotionofaphysicallyimpossibleevent?Butthereisafurtherandcrucialquestion.Humesaprioriargu-mentisaddressedtoreportsofevents,nottoourowneye-witnessingofthem,andhisargumentdependsonourreluctancewhetherrationalorirrationalremainstobeestablishedtoacceptasveridi-calreportsthatarecontrarytoouruniformexperience.So,(6)withwhatjustificationcanweusetheexceptionalnatureofaneventasgroundsforrejectingtestimonythattheallegedeventtookplace?Itisthisfinalquestionthatiscrucialinassessingandunderstand-ingHumesaprioriargumentsince,itmustagainbeemphasized,theargumentisaddressedtoreportsofevents,nottoourowneye-witnessingofthem.ThepositionIwoulddefendwithregardtoquestion(6)isthis:Humesargumentisanaccurateformalrepresentationofthenormofrationalityweallinfactapply,ortrytoapply,inoursearchforhistoricaltruth.Furthermore,whenappliedtothereportstowhichHumehastoapplyitinordertodamagethecredentialsoftheChristianrevelationnamely,tothebiblicalreportsofmiraclesingeneralandtotheResurrectioninparticularthenormissuccessfulinshowingthatthesereportswouldberejectedforthereasonshegives,iftheyoccurredinhistoricalcontextsinwhichreligiousfaithwasnotinvolved.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion501Consideranonbiblicalexample.TowardtheendofhisdialogueAgricola,theaugustRomanhistorianTacitusdescribesadecisivebattlewiththeaboriginaltribesnorthofPerthinScotlandatMountGrampius.Thelocationofthebattlehasneverbeenidentified,butsupportingarchaeologicaltracesofAgricolascampaignhavebeendiscovered,andthereisnothingimprobableaboutabattle,inthecircumstancesTacitusdescribes,thatwouldinvitetheapplicationofHumesargument.Henceweacceptthetestimony.Now,attheendofthesamehistoriansaccountofGermany,whenhesurveysthelandtotheeast,heconcludes:Whatcomesafterthemisthestuffoffables:HellusiiandOxioneswiththefacesandfeaturesofmen,butthebodiesandlimbsofanimals.ConcerningsuchunverifiablesIwillexpressnoopinion.SincethejudiciousTacitusmerelyitemizesfablesandthensuspendsjudgment,wearenotfacedwithareporttoassess,howeverfar-fetched.Butsupposehehadwritten:IntheborderlandsoftheworldtotheeastoftheDniepertherearehuman-likecreatureswho(A)haveasingleeyeinthemiddleoftheirskulls,and(B)donotmoveasothercreaturesdo,butwhentheydesiretotraverseadistancetheymerelywishitso,whereupontheydisappearintheplacetheywereinandreappearintheplacewheretheywishtobe.Thesecreaturesarecalledcyclopoids.WhatwouldbetheresultofapplyingHumesnormofhistoricalrationalitytothissupposedreport?Item(A)hassometraceofcor-roborationintheOdysseybutlacksanyzoologicalorarchaeologicalsupport,andneverhasoccurredinourexperience.Hence,despiteTacitusreputation,weareunlikelytoacceptastruethereportIamsupposinghimtohavegiven.Itistooimprobable.Item(B)isofadifferentorder.Likethereconstitutionofadeadbodyintoalivingman,suchwish-locomotionwouldbeviolationofawholeclusterofwhatwearejustifiedintakingaslawsofnature,andassuchthereisadirectandfullproof,fromthenatureofthe[alleged]factagainstitsexistence.Cyclopoids(B)justdonotexist.Thereportisatvari-ancewiththenormofhistoricalrationalityformalizedinHumesargumentbecausethereportconcernstheimpossibleasthatconceptwouldnormallybeunderstoodandiscommonlyapplied.Now,itislargelyagreedthatdespitehisobviousinclinationtoregardmiraclesasimpossible,HumedidnotputforwardtheofficialversionofhisaprioriargumentinordertoprovethatmiraclesareCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n502j.c.a.gaskinimpossible.Whathesetouttoshowwasthatitwouldneverberea-sonabletobelieveonthebasisofreportedevidencethatamiraclehadtakenplace.Butonceitisgrantedthathe,andyouandI,neverhaveourselvesexperiencedamiracleinthesenseofsomethingthatisclearlyatvariancewithwhatwecalllawsofnature,theeffec-tivepracticaldifferencebetweenneverreasonabletobelieveandimpossiblebecomesnegligible.Intermsofwhatwehaverationalwarranttobelieve,thereisnodifferencebetweenrejectingancienttestimonytocyclopoid(B)ortheResurrectiononthegroundsthatitconflictswithallourexperienceascodifiedinthelawsofnature,andsayingthatcyclopoid(B)ortheResurrectionisimpossibleasthatwordiscommonlyemployed.Itisthis,Isuggest,thatgivesHumesaprioriargument,hischecktoallkindsofsuperstitiousdelusion(EHU10.2),itspeculiarlysharpambiguityinwhichonefeels,andis,takentoamoreradicalconclusionthanonebelievestobewarranted.21ThePreposterousDistribution...ofPraiseandBlameTheattentionjustlygiventotheTreatiseas,amongotherachieve-ments,HumesmaincontributiontoanalyticmoralphilosophyhastendedtoeclipsehisotheraccountofsocialandpersonalmoralityinthesecondEnquiry.22Thisaccount,supplementedbythefinalpart21Asalreadyindicated,theliteratureisveryextensive.AmongmoderndiscussionsofthelogicandtheinterpretationofHumesargumentthefollowingprovidesomeofthebasicdiscussion:AntonyFlew,HumesPhilosophyofBelief,2nded.(London,1961),Chap.8;andDavidHume,PhilosopherofMoralScience(Oxford:OxfordUniversityPress,1986),Chap.5;myHumesPhilosophyofReligion,Chap.8.TheaboveparaphraseofHumesaprioriargumentisdrawnfromthislatterwork,152f.Ausefulcollectionofhighlyrelevantarticles(includingHumesessay)isprovidedinMiracles,ed.R.G.Swinburne(NewYork:Macmillan,1989).UsefuladditionstotheliteratureincludeDorothyColeman,Hume,Miracles,andLotter-ies,HumeStudies14(1988):32846;J.Houston,ReportedMiracles:ACritiqueofHume(Cambridge:CambridgeUniversityPress,1994);JohnEarman,HumesAbjectFailure:TheArgumentagainstMiracles(NewYork:OxfordUniversityPress,2000);RobertJ.Fogelin,ADefenseofHumeonMiracles(Princeton,NJ:PrincetonUniversityPress,2003).Seealsonotes1820above,and33below.22Forexample,inHumesMoralTheory(London:Routledge&KeganPaul,1980),J.L.MackiementionsthesecondEnquiryinhisfirstsentenceandthencontinuesanotherwiseadmirablebookasifonlytheTreatiseexisted.SimilartreatmentofthetwoworksisevidentinJonathanHarrisonsHumesTheoryofJustice(Oxford:CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion503oftheDialoguesandtheessayOfSuicide,hastwothingstosayaboutreligionthattomanypeopleareasunacceptableatthebegin-ningofthetwenty-firstcenturyastheywerewhenHumefirstpub-lishedhisideasinthemiddleoftheeighteenth.Thefirstisthatthepreceptsofmoralityandourpracticalobligationstoobservethemareindependentofreligiousbeliefsandreligioussanctions.Thesec-ondisthatwhenreligiondoesintrudeintomorality,itservesonlytodistortnaturalmoralitybytheintroductionoffrivolousspeciesofmeritandthecreationofartificialcrimes.Thisdistortionresultsinapreposterousdistribution...ofpraiseandblameandingratu-itoushumansuffering(DNR12.16,222).1.Theindependenceofmorality.Itisamatteroffacteverywhereobservable,HumecontendsinthesecondEnquiry,thatnormalhu-manbeingsarenotabsolutelyindifferenttothewealandwoeofothersphysicallyorimaginativelyneartothem.Thisresponsive-nesstootherpeopleis,accordingtoHume,ultimatelytraceabletotheoperationofsympathy,thenaturaltraitbymeansofwhichweactuallyshareinoraredirectlymovedby,thefeelingsofoth-ers.(IntheTreatise,HumehadattemptedtotracetheoriginsofsympathybacktomorefundamentalfeaturesofhumannaturethanheisconcernedwithinthesecondEnquiry.Inthelaterworkheexplicitlytakestheoperationofsympathyasanexperiencedfactabouthumanbeingsthat,inanaccountofsocialmorality,neednotbeprovidedwithanydeeperexplanation.SeeEPM5.17,n.19,App.2.5,n.60.)Now,continuesHume,sincehumanbeingshavetoacer-tainextentacommonnature,whatismiserytoone,ismiserytomost;andwhatproduceshappinessinone,produceshappinessinmost.Thusitisthatcertaindevicesanddoingsattractourgeneralcondemnationbecausetheycommonlyproducemisery,whileoth-ersattractourgeneralapprovalbecausetheycommonlypromotehappiness.Thisgeneralityofapprovalforwhateverpromoteshappi-nessinhumansocietyis,accordingtoHume,theultimatesourceClarendonPress,1981).BothMackieandHarrisontypifythestandardanddis-proportionateemphasisontheTreatiseastheonlyworthwhilesourceforHumescontributiontoethics.AstudythatgivessomeattentiontoAnEnquiryconcerningthePrinciplesofMoralsisNicholasCapaldi,HumesPlaceinMoralPhilosophy(NewYork:Lang,1989).ForfurtherdiscussionsofHumesethics,seetheessaysinthisvolumebyTerencePenelhum,DavidFateNorton,andJacquelineTaylor.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n504j.c.a.gaskinofmoraldiscriminations.Onthisshowingmoralrules(andtheparticularlawsofastate)will,intheabsenceofdistortingpreju-dicesormisinformation,expressthegeneralpoliciesthathavebeenfoundtopromotetheobjectivesofminimizingmiseryandmaxi-mizinghappiness.Thesourcesofmoralrulesarethuslocatedinthegoodofsocietyanditsmembers,andnotinmansrelationtoGodortosomeothernonworldlyorspiritualentity.ThepointwaswellmadebytheEmperorJulianin361a.d.whenherejectedtheJudeo-ChristianclaimtohavehadaspecialmoralrevelationintheTenCommandments:ExceptforthecommandmentThoushaltnotworshipothergodsandRememberthesabbathday,whatnationisthere...whichdoesnotthinkitoughttokeeptheothercommandments?23Humewouldhaveagreed.Theothercommand-mentscommendthemselvestousquiteapartfromreligionbecausetheyareperceivedtocodifysomeoftheconductgenerallyneededtoensurethehappinessofanysociety,andthisperceptionistrue,asamatterofcommonhumanexperience,notasaresultofsurprisinginformationconveyedbyaGodonMountSinai.Butevenifitisconcededthatmoralruleshave,orneedhave,nosourcebeyondouropen-mindedandnatural(Ireturntothiswordbelow)approvalofwhatisgenerallyusefulinpromotinghappiness,surelyourcommitmenttoobservingthemdependsonreligion?Dowenottothisday,andnotinfrequently,comeacrossutterancesbypoliticians,religiousbelievers,andlaymenblamingtheincreaseofcrimeandthedropinstandardsofbehavioronlackofreligiousbeliefandteaching?Andifreligiousteaching(asHumeandtheEmperorJulianwouldhaveit)isnotanecessarypreconditionfordiscover-ingthat,forexample,stealingandmurderhavetobeprohibited,thenitmustatleastbethecasethatreligionisanecessarycondi-tionforourenforcementofthesecommandmentsonourselvesasindividualswhenwearedisinclinedtoobeythem.Inshort,religionisthesourceofmoralobligation.Humewoulddisagree:themoralobligationholdsproportionwiththeusefulness(EPM4.3).Yes,butthatistoassertaproportionbetweenobligationandusefulness,nottogiveanaccountofthesourceoftheobligation.Wemayagreethatthemoresomething23TheEmperorJulian,AgainsttheGalilaeans,152D(Spanheim-Neumannpagina-tion).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion505contributeshappinesstoindividualsortosociety,themoreweoughttodoit.Butthenatureofoughtisnottherebyexplained.Humesexplanation,hishighlydistinctivesecularanalysisofobli-gation,isforthemostpartlocatedintheconclusiontothesecondEnquiry(EPM9.1415).Whathethereproducesisanaccountofwhathecallsourinterestedobligationtovirtue.Itisinterestedbecauseitisacombinationofallthefactorsthatpressonus,asmen-tallynormalpeopleinournormalsocialrelationswithourselvesandothers.Thesefactorsincludeourself-interestindoingtootherswhatwewouldwishotherstodotous,ournaturalinterchangeofsym-pathy,ourdesiretobewellthoughtofbyourneighbors,ourwishtoliveateasewithourselveswheninwardpeaceofmind,conscious-nessofintegrity,asatisfactoryreviewofourownconductispartofwhatisrequiredtobeahappyperson(EPM9.23).Butifthesearesomeofthefactorsthatinterestusinwhatiscalledmorality,howdotheyadduptoanobligation?Because,apartfrombeingunderstand-ableandcapableofanalysisintoseparateinfluences,theyconstitutesomethingnaturallyfelt,andfeelings,unlikethoughtsorfactsinHumesestimation,constitutedirectsourcesofaction.Feelings,or,inHumespreferredterm,passions,arethemainspringsofaction.Now,clearlyalotmoredeservestobesaidandwillnodoubtbesaidaboutHumesaccountofsocialmorality,butforpresentpurposesthepointisthathoweverdebatabletheoutcome,whatHumeoffersisaseriousaccountofmoralitythatmakesnoreferencewhatsoevertoGod,ortoreligiousbelieforteaching.ButHumegoesfurtherthanaseparationofreligionandmorality.Healsoholdsthattheinputofreligionintomoralityispositivelymischievousinthesensethatreligioninventscrimes(suchassuicideortheuseofcontraceptives)thatarenotnaturalcrimes,thatis,arenotactivitiesthatnormallyproducemisery;anditinventsvirtues(suchasself-mortificationordoctrinalorthodoxy)thatarenotnaturalvirtues,thatis,arenotactivitiesthatnormallypromotehappinessinoneselforothers.2.Thereligiousdistortionofmorality.ThekeytothepointHumeismakingistobefoundinmyinsistentuseaboveofsuchphrasesasintheabsenceofdistortingprejudices,ournaturalapproval,asnormalpeople.ThepointisthatHumeisattemptingtocharacterizemoralityasitisorwouldbewhenitoperatesbetweenCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n506j.c.a.gaskinnormalpeopleinnaturalconditions:normalinthesense(a)thatthepersonorpersonsconcernedarenotpathologicallydefective(fromwhatevercause)intheiremotionalresponses,feelings,orlev-elsofintelligence,andnaturalinthesense(b)thattheconditionsdonotincludespecialinfluencesthatovercomenormalfeelings.Item(a)willmakeaspecialcaseof,forexample,thecriminallyinsane,orthosewhoseconductisexplainableintermsoftheirreallackofthefeelingsthatcommonlyoperatebetweenpersons(forexample,thepersonwhosehurtingofchildrenreallydoesnotfeeltohimorherasaharmbecausethatwasthewayheorshewastreated).InsuchcasesthosewhofollowthedirectionofHumesthoughtwouldconcludethatspecialtreatment,notmoraldisap-proval,iscalledfor:moraldisapprovalbeingreservedforthevol-untaryactionsofpeoplewhoarenormalinthesensejustgiven.Item(b),vastlymoreseriousbecausecapableofvastlymoregeneraloperationthan(a),attemptstosingleoutasunnaturalconductthatoverridesthenaturalsystemofmorality(basedonhappiness)intheinterestsofnonmoralsuperstitions.ThesuperstitionsHumewasthinkingaboutasoverridingnaturalmoralitywerereligious,thoseinwhichthecommitmenttothereligionovercameallsenseofnaturalgood,forexample,intheburningofwitchesandhereticsandtherighteousinflictionofpainonothersfortheir(nonnatural)good,orforthegoodofthereligionperse.ButthetwentiethcenturycouldaddpoliticalsuperstitionsNationalSocialistandMarxistinwhichallfeelingsofnaturalgoodhavegivenway(andhavereallygivenwayinthefeelingsofthemanyconcerned)tothenonnaturalgoodthatconsistsofloyaltytothepartyorstateirrespectiveofthehappinessresulting,orthemiserycausedtoactualmenandwomen.Humessubstantialaccountofsecular,this-worldly,utilitarianmoralityinthesecondEnquiryiscertainlypolishedliterature,butitisalso,asIhopetohaveshown,revolutionarythoughtofeverwiden-ingapplication.Therevolutionisstillgoingon,andthethoughtisstillcontentious.24Ifitwerenot,itisdifficulttoseewhysooftenreligionandmoralityarestillpopularlylinked,orhow,forexample,amajorreligioncanstillstigmatizeassinfulthenatural(Humessense)goodinherentineffectivefamilyplanning.24Forfurtherdiscussion,seeDavidFateNorton,Hume,Atheism,andtheAuton-omyofMorals,inHumesPhilosophyofReligion,ed.M.Hester(Winston-Salem,NC:WakeForestUniversityPress,1986),particularly12033.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion507NaturalBeliefIf,asHumemaintains,theevidenceofnaturalreligionisatbesthighlyproblematicandambiguous,iftheevidenceofrevelationissuchaswouldnotbeacceptedifitcamefromanonreligioussource,ifwecanbothunderstandthenaturalcausesofreligionanddeploreitsunnaturaleffectsonconduct,andif,asseemstobeHumesarguedpositionatthebeginningandendofboththefirstEnquiryandtheDialogues,allspeculationsaboutthepowersandoperationsofoneuniversalspiritarebeyondourunderstanding(DNR1,135),whyisitthatreligiousbeliefpersists,evenamongwell-informedpeople?OnepossibleansweristhatwhichseemstobeimpliedbyafullreadingofHumeonreligion:thatbeliefinthedivineretainsjustenoughwispsofrationalsupportforourpropensitytoseetheworldasintelligible,inconjunctionwiththestill-operatingcausesofreli-gion,tosustainreligiondespitephilosophicalcriticism.Anotheranswer,notstrictlyanansweratall,ischaracterizedbythegestureofastonishmentwithwhichHumeendshisessayonmiraclesinthefirstEnquiry:thegesturethathasledsomeapologistsintothefalseviewthatHumeisadvocatingfideismasadefensibleaccountofhowwedoandwhyweshouldretainreligiousbelief.Butathird,andpotentiallyveryfruitfulanswer,issometimesgivenonHumesbehalf:thatbeliefinthedivineisanaturalbelief.Theconceptofanaturalbeliefwasassembledbyanumberoftwentieth-centurycommentatorsonHumefromthecharacteristicsofthosefewandverygeneralbeliefsthatHumeidentifiesasulti-matelyresistanttoallskepticalargumentbeliefinthecontinuousexistenceofanexternalworldindependentofourperceptionofthatworld,beliefthattheregularitiesofthepastwillcontinueintothefuture,thatoursensesarenormallyreliable,areexamples.25Thecharacteristicsofthesenaturalbeliefsarethefollowing:a.Thattheyarearrivedatpriortoanyprocessofreasoning,andcannotforlongbedislodgedbyanyprocessofskepticalreasoningbecause:b.Theyareindispensableaspresuppositionsofknowledgeandconductforanysentientbeingwholivesinacoherent25TheconceptwasfirstdevelopedbyN.K.SmithinhisarticleTheNaturalismofHume,Mind14(1905):14973,33547.ThismaterialreappearsasChaps.46ofhisPhilosophyofDavidHume.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n508j.c.a.gaskinrelationtothegivenappearancesofthings.Inpracticaltermsnoonecanactintheworldunlessheorshehasthesebeliefs.Hence:c.Thesebeliefsareuniversalnotmerelythecherishedordominantorunquestionedassumptionsofaparticularcul-tureorofalearnedorunlearnedpopulation,butsuchasallhumanbeingsalwaysandeverywherehave.Setoutthus,itisallbutobviousthatbeliefinthedivinedoesnothavethecharacteristicsofanaturalbelief.26Evenifitcouldbeshownthatformost,oratleastformanypeople,religiousbeliefisattainedandretainedaccordingto(a),itisanincontrovertiblematteroffactthatreligiousbeliefisnotuniversalinthemannerof(c).Itisalsoevidentthatindividualscananddoactperfectlyadequatelyintheworldwithoutreligiousbelief,andthatreligiousbeliefisnotanepistemicrequirementforanycoherentrelationtothegivenappearancesofthings,thatis,(b)doesnotholdeither.Thereis,moreover,noclearevidencethatHumeeverseriouslyentertainedthethoughtthatbeliefinthedivinemightbeaninstinctofnatureimpervioustoskepticisminthewaythatourbeliefinanexternalworldis.ThenearestwegettosuchathoughtisCleanthesrestatementofthedesignargumentinwhichthereisanappeal:tellme,fromyourownfeeling,iftheideaofacontriverdoesnotimmediatelyflowinuponyouwithaforcelikethatofsensationfollowedbyareferencetotheuniversalandirresistibleinfluenceoftheargument(thedesignargument)fortheism(seeDNR3,154,26ThethesisthatbeliefinthedivineisanaturalbeliefwasfirstarguedatlengthbyR.J.ButlerinNaturalBeliefandtheEnigmaofHume,ArchivfurGeschichteder¨Philosophie42(1960):73100.Itisdiscussed,andrejected,inmyHumesPhilos-ophyofReligion,Chaps.67,andissubjecttoadditionalrefinementsinTerencePenelhumsimportantThemesinHume:TheSelf,theWill,Religion(Oxford:ClarendonPress,2000),Chap.10.Withtheadvantageofacenturyofdiscussionbehindusitmaybefeltthatthetermnaturalbelief,whichisnotfoundinHumesownwritings,wasunfortunate.Thereisalackofclarityabouthowitrelatestotermswhichhediduse,suchasoriginalprinciples,originalinstinct,primaryimpressionofnature,ornaturalpropensity.SomeaccountofthesetermsinrelationtonaturalbeliefisgivenbyMiriamMcCormickinHumeonNaturalBeliefandOriginalPrinciples,HumeStudies19(1993):10316.Butthephilosophicallyimportantpointremains:thatwhateverwedecidetocallthesebeliefsHumesdeviceforrejectingthesustainedapplicationofskepticismtobeliefsintheabsenceofwhichhumanbeingscouldnot,forexample,learnfromexperienceoranticipatethefuturebeliefinGodisnotoneofthem.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion509quotedas(3)inSectionIIabove).TheforceofCleanthespointseemstobethatournaturalpropensitytoseeandexpectorderinnatureissoclosetoseeinganordererthatournaturalbeliefinthefor-merbringswithitthelatter.ButevenifCleanthes,contrarytothemajorityviewamongcommentators,canbetakentobespeakingforHume,hisviewisdefectiveinthismatter.Inthefirstplace,asPhilopointsoutneartheendofPart4oftheDialoguesandagaininPart7,theactivityofanorderingagentisnottheonlypossibleexpla-nationoforder;andsecond,evenifthefeelingthatacontriverisresponsiblefortheordereduniverseisdifficulttokeepatbaywithskepticalargument,itisnotirresistiblebecauseitisresisted,anditisnotuniversalbecauseatleastsomepeopledonotsuccumbtotheinfluenceoftheargumentfortheism.Thatsomethingiswidelyfelt,influential,anddifficulttodislodgebyargument,isnotofitselfsufficienttogiveittheexceptionallyprivilegedstatusofanaturalbelief.ButthisstillleavesHumewiththedifficultywhichhepartlyfacesintheNaturalHistoryofReligionofexplainingthepersistenceofreligiousbeliefoncetheargumentsandevidenceforitareshowntobeallbutnegligible.Humedidnotandperhapscouldnothaveanticipatedthenineteenth-centuryexplanationforthispersistencedevelopedbySchopenhauer,Feuerbach,andaboveallbyFreud.Namely,thatwearesoconstitutedthatemotionallyandpsychologically(butnotrationallyandepistemically)weneedsomesortofreligiousbelief.NorcouldHumehaveexpected(1)thathisskepticalphilosophyofreligionwouldleadtoaredeploymentoffideismor(2)thathisnaturalbeliefcountertoextremeskepticismwouldsuggestthedevelopmentofotherandnewdefensesofChristianity.Howdidthiscomeabout?Inthefirstplace,HumesunderminingofthetraditionalrationalgroundsforbeliefinGodwassothoroughthatoncehispositionhadbeenabsorbedintothemainstreamsofEuropeanthought(via,amongothers,DHolbach,Kant,andShelley)afundamentalreappraisalofthenatureofreligioncommenced.ThusfirstSchleiermacher(17681834)andlaterKierkegaard(181355)soughttomakereligionrelylessonevidenceandreason,andmoreonfeeling,subjectiveexpe-rience,andfaith.SuchafideisticreliancelargelyevadesHumesrationalisticcritique,butitdoessoattheriskofmakingreligiousbeliefarbitrary,whileatthesametimebothinvitingHume-typeCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n510j.c.a.gaskinaccountsofitsnaturalhistoryandleavingintacthiscriticismofitsmoralandsocialeffects.TheologicalfideismhasaphilosophicalcounterpartinwhatTer-encePenelhumhascalledTheParityArgument.27Theargumentcanbeusedbysomeonewhoagrees,asHumedoes,withtheskep-ticaltraditionthatatleastsomeofthefundamentalphilosophicalcommitmentsofsecularcommonsensearewithoutrationalfounda-tionbutwhoneverthelessyieldstoournaturaltendencytobelievethem:TheParityArgumentsuggeststosuchapersonthatheisinconsistentifherefusestoyieldalsotothedemandsofreligiousbeliefmerelybecauseheconsidersthatit,too,doesnothavearatio-nalfoundation.28Thecoreobjectiontothisargumentisthattheinconsistencyclaimedisnotaninconsistencyunlessitcanbeshownthatthepressuretoyieldtoreligiousbeliefisequalinallrespectstothepressuretoyieldtonaturalbeliefs.Butsetagainstthecriteria(a),(b),and(c)above,wehavealreadyseenthatthemetarationaldemandstobelieveinthedivineareinmanyrespectsnotequaltothedemandstobelievein,forexample,anexternalworld.AnadditionalobjectiontotheParityArgumentisthatifitjustifiesbeliefinthedivine,italsojustifiesanycherishedpersonalorgroupbeliefforwhichthereisnorationalfoundation,forexample,thattherearewitcheswithdiabolicalandsupernaturalpowers.ItwillbenotedthatHumesaccountofnaturalbeliefscannotbeusedtojustifysuchcherishedirrationalitiesbecausethecriteriaforanaturalbeliefareenormouslytougherthantheirrationalitycriterionappealedtointheParityArgument.Despitethefailuretoidentifybeliefinthedivineasagenuinenat-uralbelief,modernphilosophicaltheologyismarkedwithattemptstoemploysomenotionofnaturalbeliefforapologeticpurposes.Thus,forexample,JohnHickassertsananalogybetweenthereli-giouspersonsclaimtobeconsciousofGodandanymansclaimtobeconsciousofthephysicalworldasanenvironment,exist-ingindependentlyofhimself.29Thesamethoughtturnsupinthe27SeeTerencePenelhumsimportantworkGodandSkepticism(Dordrecht:Reidel,1983),particularlyChaps.2,5,and6.28Penelhum,GodandSkepticism,139.29JohnHick,ArgumentsfortheExistenceofGod(London,1970),110.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion511writingsofJohnMacquarrie:ItisnotinappropriatetocomparetheconvictionoftheindependentrealityofGodtotheconvictionoftheindependentrealityoftheworldorofotherselves,30andagain,morerecently,inthewritingsofHansKung:Thehistoryofmod-ernepistemologyfromDescartes,Hume,andKanttoPopperandLorenzitseemstomemadeclearthatthefactofanyrealityatallindependentofourconsciousnesscanbeacceptedonlyasanactoftrust,hencealikeactoftrustisappropriatetobeliefinGod.31Asimilarmove,butdifferentlypresented,isevidentintheAmericanschoolofBasicBeliefApologists,associatedwithAlvinPlantinga.32ThesemovesderivefromHumesnaturalbeliefcountertoexcessiveskepticism,butthederivationislessacceptablethanHumesoriginalcounterforthetworeasonsalreadyidentifiedinconnectionwiththeParityArgument,namely,thederivationadmitsanybeliefthatonemaychoosetoassertbaselessly,anditfailstodifferentiatebetweenanoptionalbelieflikebeliefinGod(optionalsinceplainlysomeofusdonothaveit)andanonoptionalbelieflikebeliefinanexternalworld:Towhateverlengthanyonemaypushhisspeculativeprinciplesofscep-ticism,hemustact,Iown,andlive,andconverselikeothermen;andforthisconductheisnotobligedtogiveanyotherreasonthantheabsolutenecessityheliesunderofsodoing(DNR1,134).Nosuchabsolutenecessityattachestoanyparticularbeliefinthedivine.IsaidabovethattherearethreepossiblewaysinwhichHumecouldhaverespondedtothepuzzleabouttheresistanceofreligiousbelieftoskepticalreasoning.Hedoesnottakethewayofnaturalbelief.Heworksatthewayofcausalexplanationsforreligioncou-pledwithavestigialrationality.Thethirdway,characterizedby30JohnMacquarrie,God-Talk(London,1967),244.31HansKung,EternalLife?,trans.E.Quinn(London:Collins,1984),275.SeealsoKungsDoesGodExist?AnAnswerforToday,trans.E.Quinn(GardenCity,NY:Doubleday,1980),56883.IamindebtedtoPhilipBarnesforthesereferences.32SeeparticularlyAlvinPlantinga,RationalityandReligiousBelief,inContempo-raryPhilosophyofReligion,ed.S.KahnandD.Shatz(NewYork:OxfordUniversityPress,1982);andFaithandRationality:ReasonandBeliefinGod,ed.A.PlantingaandN.Wolterstorff(NotreDame,IN:NotreDameUniversityPress,1983).CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n512j.c.a.gaskinthegestureofastonishmentwithwhichHumeendshisessayOfMiracles,isperhapsaveryrealisticperceptionofthefundamentalirrationalityofhumanbeingsconcerningthosespeciallycherishedbeliefscalledreligious:Sothat,uponthewhole,wemayconclude,thattheChristianreligionnotonlywasatfirstattendedwithmir-acles,butevenatthisdaycannotbebelievedbyanyreasonablepersonwithoutone(EHU10.41).Thisisnot,assomewouldhaveit,toclearthewayforfideisticChristianityaconceptionalienbothtoHumesmitigatedskepticismandtohisworldlymoral-ity.Itissimplytonotethecontinuedmiraclebywhichreli-giousfaithsurvivesinthesecularworldagainstalltheintellectualodds.suggestionsforfuturereadingInadditiontotheworkscitedinthenotestothisessay,forfurtherreadingthefollowingarerecommended.Bernard,Christopher.HumeandtheMadnessofReligion.InHumeandHumesConnexions,editedbyM.A.StewartandJ.P.Wright.Edinburgh:EdinburghUniversityPress,1994.22438.Fogelin,RobertJ.WhatHumeActuallySaidaboutMiracles.HumeStud-ies16(1990):816.Garrett,Don.HumeonTestimonyconcerningMiracles.InReadingHumeonHumanUnderstanding,editedbyP.Millican.Oxford:Claren-donPress,2002.30334.Gaskin,J.C.A.Religion:TheUselessHypothesis.InReadingHumeonHumanUnderstanding,editedbyP.Millican.Oxford:ClarendonPress,2002.34969.Herdt,Jennifer.ReligionandFactioninHumesMoralPhilosophy.Cam-bridge:CambridgeUniversityPress,1997.Johnson,David.Hume,Holism,andMiracles.Ithaca,NY:CornellUniver-sityPress,1999.OConnor,David.RoutledgePhilosophyGuidebooktoHumeonReligion.London:Routledge,2001.Owen,David.HumeversusPriceonMiraclesandPriorProbabilities:Testi-monyandtheBayesianCalculation.InReadingHumeonHumanUnder-standing,editedbyP.Millican.Oxford:ClarendonPress,2002.33548.Phillips,D.Z.,andT.Tessip,eds.ReligionandHumesLegacy.NewYork:St.MartinsPress,1999.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nHumeonReligion513Reich,Lou.HumesReligiousNaturalism.Lanham,MD:UniversityPressofAmerica,1998.Stewart,M.A.HumesHistoricalViewofMiracles.InHumeandHumesConnexions,editedbyM.A.StewartandJ.P.Wright.Edinburgh:Edin-burghUniversityPress,1994.171200.Yandell,Keith.HumesInexplicableMystery:HisViewsonReligion.Philadelphia:TempleUniversityPress,1990.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nappendix:hume’sautobiographiesi.akindofhistoryofmylifeInthespringof1734,HumeacceptedapositionwithaBris-tolmerchant.Hisphilosophicalendeavorswerenotgoingwell,andsohedeterminedtoputtheseasideforsometime,inorderthemoreeffectuallytoresumethem.AshetraveledtoBristol,hewrotetoanunnamedphysician,probablyeitherJohnArbuthnotorGeorgeCheyne,1toaskadviceabouthowtogetonwithhisphilosophicalwork.WhetherHumeactuallysentsuchaletterisnotknown,butthesurvivingmanuscriptfurnishesuswithavaluableaccountofthefirstyearsofhisadultlife.ThetextprintedhereisbasedontheoriginalmanuscriptdepositedintheNationalLibraryofScotland,andispublishedwiththeper-missionoftheRoyalSocietyofEdinburgh.Thetitleistakenfromthefirstparagraphoftheletter.Sir[1]Notbeingacquaintedwiththishand-writing,youwillprobablylooktothebottomtofindtheSubscription,¬findingany,willcertainlywonderatthisstrangemethodofaddressingtoyou.Imusthereinthebeginningbegyoutoexcuseit,&toperswadeyoutoreadwhatfollowswithsomeAttention,[and]musttellyou,thatthisgivesyouanOpportunitytodoaverygood-natur’dAction,whichIbelieveisthemostpowerful1Forahelpfuldiscussionofthisletter,seeJohnP.Wright,“Dr.GeorgeCheyne,ChevalierRamsay,andHume’sLettertoaPhysician,”HumeStudies29(2003),125–41.515CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n516appendix:hume’sautobiographiesArgumentIcanuse.Ineednottellyou,thatIamyourCountryman,aScotchman;forwithoutanysuchtye,IdarerelyuponyourHumanity,eventoaperfectStranger,suchasIam.TheFavourIbegofyouisyourAdvice,&thereasonwhyIaddressmyselfinparticulartoyouneednotbetold.AsonemustbeaskilfulPhysician,amanofLetters,ofWit,ofGoodSense,&ofgreatHumanity,togivemeasatisfyingAnswer,IwishFamehadpointedouttomemorePersons,inwhomtheseQualitiesareunited,inordertohavekeptmesometimeinSuspense.ThisIsayintheSincerityofmyHeart,&withoutanyIntentionofmakingaComplement:Fortho’itmayseemnecessary,thatinthebeginningofsounusualaLetter,Ishou’dsaysomefinethings,tobespeakyourgoodOpinion,&removeanyprejudicesyoumayconceiveatit,yetsuchanEndeavortobewitty,wou’dbutillsuitwiththepresentConditionofmyMind;which,Imustconfess,isnotwithoutAnxietyconcerningtheJudgementyouwillformofme.TrustinghowevertoyourCandor&Generosity,Ishall,withoutfurtherPreface,proceedtoopenuptoyouthepresentConditionofmyHealth,&todothatthemoreeffectuallyshallgiveyouakindofHistoryofmyLife,afterwhichyouwilleasilylearn,whyIkeepmyNameaSecret.[2]YoumustknowthenthatfrommyearliestInfancy,IfoundalwiseastrongInclinationtoBooks&Letters.AsourCollegeEducationinScotland,extendinglittlefurtherthantheLanguages,endscommonlywhenweareabout14or15YearsofAge,IwasafterthatlefttomyownChoiceinmyReading,&founditenclinemealmostequallytoBooksofReasoning&Philosophy,&toPoetry&thepoliteAuthors.Everyone,whoisacquaintedeitherwiththePhilosophersorCritics,knowsthatthereisnothingyetestablishtineitherofthesetwoSciences,&thattheycontainlittlemorethanendlessDisputes,eveninthemostfundamentalArticles.UponExaminationofthese,IfoundacertainBoldnessofTemper,growinginme,whichwasnotenclin’dtosubmittoanyAuthorityintheseSubjects,butledmetoseekoutsomenewMedium,bywhichTruthmightbeestablisht.AftermuchStudy,&Reflectiononthis,atlast,whenIwasabout18YearsofAge,thereseem’dtobeopen’duptomeanewSceneofThought,whichtransportedmebeyondMeasure,&mademe,withanArdorCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nI.AKindofHistoryofMyLife517naturaltoyoungmen,throwupeveryotherPleasureorBusinesstoapplyentirelytoit.TheLawwhichwastheBusinessIdesign’dtofollow,appear’dnauseoustome,&Icou’dthinkofnootherwayofpushingmyFortuneintheWorld,butthatofaScholar&Philosopher.IwasinfinitelyhappyinthisCourseofLifeforsomeMonths;tillatlast,aboutthebeginningofSeptr1729,allmyArdorseem’dinamomenttobeextinguisht,&Icou’dnolongerraisemyMindtothatpitch,whichformerlygavemesuchexcessivePleasure.IfeltnoUneasynessorWantofSpirits,whenIlaidasidemyBook;&thereforeneverimagindtherewasanybodilyDistemperintheCase,butthatmyColdnessproceededfromaLazinessofTemper,whichmustbeovercomebyredoublingmyApplication.InthisConditionIremain’dfornineMonths,veryuneasytomyself,asyoumaywellimagine,butwithoutgrowinganyworse,whichwasaMiracle.[3]Therewasanotherparticular,whichcontributedmorethananything,towastemySpirits&bringonmethisDistemper,whichwas,thathavingreadmanyBooksofMorality,suchasCicero,Seneca&Plutarch,&beingsmitwiththeirbeautifulRepresentationsofVirtue&Philosophy,IundertooktheImprovementofmyTemper&Will,alongwithmyReason&Understanding.IwascontinuallyfortifyingmyselfwithReflectionsagainstDeath,&Poverty,&Shame,&Pain,&alltheotherCalamitiesofLife.Thesenodoubtareexceedinguseful,whenjoin’dwithanactiveLife;becausetheOccasionbeingpresentedalongwiththeReflection,worksitintotheSoul,&makesittakeadeepImpression,butinSolitudetheyservetolittleotherPurpose,thantowastetheSpirits,theForceoftheMindmeetingwithnoResistance,butwastingitselfintheAir,likeourArmwhenitmissesitsAim.ThishoweverIdidnotlearnbutbyExperience,&tillIhadalreadyruin’dmyHealth,tho’Iwasnotsensibleofit.[4]SomeScurvySpotsbrokeoutonmyFingers,thefirstWinterIfellill,aboutwhichIconsultedaveryknowingPhysician,whogavemesomeMedicines,thatremov’dtheseSymptoms,&atthesametimegavemeaWarningagainsttheVapors,which,thoIwaslaboringunderatthattime,Ifancy’dmyselfsofarremov’dfrom,&indeedfromanyotherDisease,exceptaslightScurvy,thatIdespis’dhisWarning.AtlastaboutAprile1730,whenIwasCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n518appendix:hume’sautobiographies19YearsofAge,aSymptom,whichIhadnotic’dalittlefromthebeginning,encreas’dconsiderably,sothattho’itwasnoUneasyness,theNoveltyofitmademeaskAdvice.ItwaswhattheycallaPtyalismorWatrynessinthemouth.UponmymentioningittomyPhysician,helaughtatme,&toldmeIwasnowaBrother,forthatIhadfairlygottheDiseaseoftheLearned.OfthishefoundgreatDifficultytoperswademe,findinginmyselfnothingofthatlownessofSpirit,whichthose,wholaborunderthatDistempersomuchcomplainof.HoweveruponhisAdvice,IwentunderaCourseofBitters,&Anti-hystericPills.DrunkanEnglishPintofClaretWineeveryDay,&rode8or10ScotchMiles.ThisIcontinu’dforabout7Monthsafter.[5]ThoIwassorrytofindmyselfengag’dwithsotediousaDistemperyettheKnowledgeofit,setmeverymuchatease,bysatisfyingmethatmyformerColdness,proceedednotfromanyDefectofTemperorGenius,butfromaDisease,towhichanyonemaybesubject.InowbegantotakesomeIndulgencetomyself;studiedmoderately,&onlywhenIfoundmySpiritsattheirhighestPitch,leavingoffbeforeIwasweary,&triflingawaytherestofmyTimeinthebestmannerIcould.Inthisway,Iliv’dwithSatisfactionenough;andonmyreturntoTownnextWinterfoundmySpiritsverymuchrecruited,sothat,thotheysunkundermeinthehigherFlightsofGenius,yetIwasabletomakeconsiderableProgressinmyformerDesigns.IwasveryregularinmyDiet&wayofLifefromthebeginning,&allthatWinter,madeitaconstantRuletoridetwiceorthriceaweek,&walkeveryday.FortheseReasons,IexpectedwhenIreturn’dtotheCountrey,&cou’drenewmyExercisewithlessInterruption,thatIwou’dperfectlyrecover.ButinthisIwasmuchmistaken.FornextSummer,aboutMay1731theregrewupon[me]averyravenousAppetite,&asquickaDigestion,whichIatfirsttookforagoodSymptom,&wasverymuchsurpriz’dtofinditbringbackaPalpitationofHeart,whichIhadfeltverylittleofbefore.ThisAppetite,however,hadanEffectveryunusual,whichwastonourishmeextremely;sothatin6weekstimeIpastfromtheoneextremetotheother,&beingbeforetall,lean,&rawbon’dbecameonasudden,themoststurdy,robust,healthful-likeFellowyouhaveseen,witharuddyComplexion&achearfulCountenance.InexcuseformyRiding,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nI.AKindofHistoryofMyLife519&careofmyHealth,Ialwisesaid,thatIwasafraidofaConsumption;whichwasreadilybeliev’dfrommyLooks;butnoweveryBodycongratulatemeuponmythorowRecovery.ThisunnaturalAppetiteworeoffbydegrees,butleftmeasaLegacy,thesamePalpitationoftheheartinasmalldegree,&agooddealofWindinmyStomach,whichcomesawayeasily,&withoutanybadGout,asisordinary.However,theseSymptomsarelittleornoUneasynesstome.Ieatwell;Isleepwell.HavenolownessofSpirits;atleastnevermorethanwhatoneofthebestHealthmayfeel,fromtoofullameal,fromsittingtoonearaFire,&eventhatdegreeIfeelveryseldom,&neveralmostintheMorningorForenoon.ThosewholiveinthesameFamilywithme,&seemeatalltimes,cannotobservetheleastAlterationinmyHumor,&ratherthinkmeabetterCompanionthanIwasbefore,aschoosingtopassmoreofmytimewiththem.ThisgavemesuchHopes,thatIscarceevermistadaysriding,exceptintheWinter-time;&lastSummerundertookaverylaborioustask,whichwastotravel8MileseveryMorning&asmanyintheForenoon,to&fromamineralWellofsomeReputation.Irenew’dtheBitters&AntihystericPillstwice,alongwithAnti-scorbuticJuiceslastSpring,butwithoutanyconsiderableEffect,exceptabatingtheSymptomsforalittletime.[6]ThusIhavegivenyouafullaccountoftheConditionofmyBody,&withoutstayingtoaskPardon,asIoughttodo,forsotediousaStory,shallexplaintoyouhowmyMindstoodallthistime,whichoneveryOccasion,especiallyinthisDistemper,haveaverynearConnexiontogether.HavingnowTime&Leizuretocoolmyinflam’dImaginations,Ibegantoconsiderseriously,howIshou’dproceedinmyPhilosophicalEnquiries.IfoundthatthemoralPhilosophytransmittedtousbyAntiquity,labor’dunderthesameInconveniencethathasbeenfoundintheirnaturalPhilosophy,ofbeingentirelyHypothetical,&dependingmoreuponInventionthanExperience.EveryoneconsultedhisFancyinerectingSchemesofVirtue&ofHappiness,withoutregardinghumanNature,uponwhicheverymoralConclusionmustdepend.ThisthereforeIresolvedtomakemyprincipalStudy,&theSourcefromwhichIwou’dderiveeveryTruthinCriticismaswellasMorality.Ibelieve’tisacertainFactthatmostofthePhilosophersCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n520appendix:hume’sautobiographieswhohavegonebeforeus,havebeenoverthrownbytheGreatnessoftheirGenius,&thatlittlemoreisrequir’dtomakeamansucceedinthisStudythantothrowoffallPrejudiceseitherforhisownOpinionsorforth[ose]ofothers.AtleastthisisallIhavetodependonfortheTruthofmyReasonings,whichIhavemultiply’dtosuchadegree,thatwithinthesethreeYears,IfindIhavescribledmanyaQuireofPaper,inwhichthereisnothingcontain’dbutmyownInventions.ThiswiththeReadingmostofthecelebratedBooksinLatin,French&English,&acquiringtheItalian,youmaythinkasufficientBusinessforoneinperfectHealth;&soitwou’d,haditbeendonetoanyPurpose:ButmyDiseasewasacruelIncumbranceonme.IfoundthatIwasnotabletofollowoutanyTrainofThought,byonecontinuedStretchofView,butbyrepeatedInterruptions,&byrefreshingmyEyefromTimetoTimeuponotherObjects.YetwiththisInconvenienceIhavecollectedtherudeMaterialsformanyVolume;butinreducingthesetoWords,whenonemustbringtheIdeahecomprehendedingross,nearertohimsoastocontemplateitsminutestParts,&keepitsteddilyinhisEye,soastocopythesePartsinOrder,thisIfoundimpracticableforme,norweremySpiritsequaltososevereanEmployment.HerelaymygreatestCalamity.IhadnoHopesofdeliveringmyOpinionswithsuchElegance&Neatness,astodrawtometheAttentionoftheWorld,&Iwou’dratherlive&dyeinObscuritythanproducethemmaim’d&imperfect.[7]SuchamiserableDisappointmentIscarceeverremembertohaveheardof.ThesmallDistancebetwixtme&perfectHealthmakesmethemoreuneasyinmypresentSituation.TisaWeaknessratherthanaLownessofSpiritswhichtroublesme,&thereseemstobeasgreataDifferencebetwixtmyDistemper&commonVapors,asbetwixtVapors&Madness.[8]Ihavenotic’dintheWritingsoftheFrenchMysticks,&inthoseofourFanaticshere,that,whentheygiveaHistoryoftheSituationoftheirSouls,theymentionaColdness&DesertionoftheSpirit,whichfrequentlyreturns,&someofthem,atthebeginning,havebeentormentedwithitmanyYears.AsthiskindofDevotiondependsentirelyontheForceofPassion,&consequentlyoftheAnimalSpirits,IhaveoftenthoughtthattheirCase&minewereprettyparralel,&thattheirrapturousAdmirationsmightCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nI.AKindofHistoryofMyLife521discomposetheFabricoftheNerves&Brain,asmuchasprofoundReflections,&thatwarmthorEnthusiasmwhichisinseperablefromthem.[9]Howeverthismaybe,IhavenotcomeoutoftheCloudsowellastheycommonlytellustheyhavedone,orratherbegantodespairofeverrecovering.TokeepmyselffrombeingMelancholyonsodismalaProspect,myonlySecuritywasinpeevishReflectionsontheVanityoftheWorld&ofallhumaneGlory;which,howeverjustSentimentstheymaybeesteem’d,Ihavefoundcanneverbesincere,exceptinthosewhoarepossestofthem.BeingsensiblethatallmyPhilosophywou’dnevermakemecontentedinmypresentSituation,Ibegantorouzeupmyself;&beingencourag’dbyInstancesofRecoveryfromworsedegreesofthisDistemper,aswellasbytheAssurancesofmyPhysicians,Ibegantothinkofsomethingmoreeffectual,thanIhadhithertotry’d.Ifound,thatastherearetwothingsverybadforthisDistemper,Study&Idleness,sotherearetwothingsverygood,Business&Diversion;&thatmywholeTimewasspentbetwixtthebad,withlittleornoShareoftheGood.ForthisreasonIresolvedtoseekoutamoreactiveLife,&tho’Icou’dnotquitmyPretensionsinLearning,butwithmylastBreath,tolaythemasideforsometime,inorderthemoreeffectuallytoresumethem.[10]UponExaminationIfoundmyChoiceconfin’dtotwokindsofLife;thatofatravellingGovernor&thatofaMerchant.Thefirst,besidesthatitisinsomerespectsanidleLife,was,Ifound,unfitforme;&thatbecausefromasedentary&retir’dwayofliving,fromabashfulTemper,&fromanarrowFortune,Ihadbeenlittleaccustom’dtogeneralCompanies,&hadnotConfidence&KnowledgeenoughoftheWorldtopushmyFortuneorbeserviceableinthatway.IthereforefixtmyChoiceuponaMerchant;&havinggotRecommendationtoaconsiderableTraderinBristol,Iamjustnowhasteningthither,withaResolutiontoforgetmyself,&everythingthatispast,toengagemyself,asfarasispossible,inthatCourseofLife,&totossabouttheWorld,fromtheonePoletotheother,tillIleavethisDistemperbehindme.[11]AsIamcometoLondoninmywaytoBristol,Ihaveresolved,ifpossible,togetyourAdvice,tho’Ishou’dtakethisabsurdMethodofprocuringit.AllthePhysicians,Ihaveconsulted,CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n522appendix:hume’sautobiographiestho’veryable,cou’dneverenterintomyDistemper;becausenotbeingPersonsofgreatLearningbeyondtheirownProfession,theywereunacquaintedwiththeseMotionsoftheMind.YourFamepointedyououtastheproperestPersontoresolvemyDoubts,&Iwasdetermin’dtohavesomebodiesOpinion,whichIcou’drestuponinalltheVarietiesofFears&Hopes,incidenttosolingeringaDistemper.IhopeIhavebeenparticularenoughindescribingtheSymptomstoallowyoutoformaJudgement;orratherperhapshavebeentooparticular.Butyouknow’tisaSymptomofthisDistempertodelightincomplaining&talkingofitself.[12]TheQuestionsIwou’dhumblyproposetoyouare:WhetheramongalltheseScholars,youhavebeenacquaintedwith,youhaveeverknownanyaffectedinthismanner?WhetherIcaneverhopeforaRecovery?WhetherImustlongwaitforit?WhethermyRecoverywilleverbeperfect,&mySpiritsregaintheirformerSpring&Vigor,soastoenduretheFatigueofdeep&abstrusethinking?WhetherIhavetakenarightwaytorecover?IbelieveallproperMedicineshavebeenus’d,&thereforeIneedmentionnothingofthem.ii.myownlifeByApril1776,Humewasconvincedthattheboweldisorderthathadafflictedhimforsomemonthswouldsoonleadtohisdeath.JustpriortosettingoutforBath,theretoseekacurefromthewaters,hepreparedhiswillandthebriefauto-biographythatfollows.Thetitle,MyOwnLife,isHumesown;thetextprintedhereisthatofthefirsteditionofthiswork,TheLifeofDavidHume,Esq.WrittenbyHimself,London,1777.[1]Itisdifficultforamantospeaklongofhimselfwithoutvanity;therefore,Ishallbeshort.ItmaybethoughtaninstanceofvanitythatIpretendatalltowritemylife;butthisNarrativeshallcontainlittlemorethantheHistoryofmyWritings;as,indeed,almostallmylifehasbeenspentinliterarypursuitsandoccupations.Thefirstsuccessofmostofmywritingswasnotsuchastobeanobjectofvanity.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nII.MyOwnLife523[2]Iwasbornthe26thofApril1711,oldstyle,atEdinburgh.Iwasofagoodfamily,bothbyfatherandmother:myfather’sfamilyisabranchoftheEarlofHome’s,orHume’s;andmyancestorshadbeenproprietorsoftheestate,whichmybrotherpossesses,forseveralgenerations.MymotherwasdaughterofSirDavidFalconer,PresidentoftheCollegeofJustice:thetitleofLordHalkertoncamebysuccessiontoherbrother.[3]Myfamily,however,wasnotrich,andbeingmyselfayoungerbrother,mypatrimony,accordingtothemodeofmycountry,wasofcourseveryslender.Myfather,whopassedforamanofparts,diedwhenIwasaninfant,leavingme,withanelderbrotherandasister,underthecareofourmother,awomanofsingularmerit,who,thoughyoungandhandsome,devotedherselfentirelytotherearingandeducatingofherchildren.Ipassedthroughtheordinarycourseofeducationwithsuccess,andwasseizedveryearlywithapassionforliterature,whichhasbeentherulingpassionofmylife,andthegreatsourceofmyenjoyments.Mystudiousdisposition,mysobriety,andmyindustry,gavemyfamilyanotionthatthelawwasaproperprofessionforme;butIfoundanunsurmountableaversiontoeverythingbutthepursuitsofphilosophyandgenerallearning;andwhiletheyfanciedIwasporinguponVoetandVinnius,CiceroandVirgilweretheauthorswhichIwassecretlydevouring.[4]Myveryslenderfortune,however,beingunsuitabletothisplanoflife,andmyhealthbeingalittlebrokenbymyardentapplication,Iwastempted,orratherforced,tomakeaveryfeebletrialforenteringintoamoreactivesceneoflife.In1734,IwenttoBristol,withsomerecommendationstoeminentmerchants,butinafewmonthsfoundthatscenetotallyunsuitabletome.IwentovertoFrance,withaviewofprosecutingmystudiesinacountryretreat;andItherelaidthatplanoflife,whichIhavesteadilyandsuccessfullypursued.Iresolvedtomakeaveryrigidfrugalitysupplymydeficiencyoffortune,tomaintainunimpairedmyindependency,andtoregardeveryobjectascontemptible,excepttheimprovementofmytalentsinliterature.[5]DuringmyretreatinFrance,firstatReims,butchieflyatLaFleche,inAnjou,IcomposedmyTreatiseofHumanNature.Afterpassingthreeyearsveryagreeablyinthatcountry,IcameovertoCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n524appendix:hume’sautobiographiesLondonin1737.Intheendof1738,IpublishedmyTreatise,andimmediatelywentdowntomymotherandmybrother,wholivedathiscountry-house,andwasemployinghimselfveryjudiciouslyandsuccessfullyintheimprovementofhisfortune.[6]NeverliteraryattemptwasmoreunfortunatethanmyTreatiseofHumanNature.Itfelldead-bornfromthepress,withoutreachingsuchdistinction,aseventoexciteamurmuramongthezealots.Butbeingnaturallyofacheerfulandsanguinetemper,Iverysoonrecoveredtheblow,andprosecutedwithgreatardourmystudiesinthecountry.In1742,IprintedatEdinburghthefirstpartofmyEssays:theworkwasfavourablyreceived,andsoonmademeentirelyforgetmyformerdisappointment.Icontinuedwithmymotherandbrotherinthecountry,andinthattimerecoveredtheknowledgeoftheGreeklanguage,whichIhadtoomuchneglectedinmyearlyyouth.[7]In1745,IreceivedaletterfromtheMarquisofAnnandale,invitingmetocomeandlivewithhiminEngland;Ifoundalso,thatthefriendsandfamilyofthatyoungnoblemanweredesirousofputtinghimundermycareanddirection,forthestateofhismindandhealthrequiredit.–Ilivedwithhimatwelvemonth.Myappointmentsduringthattimemadeaconsiderableaccessiontomysmallfortune.IthenreceivedaninvitationfromGeneralSt.Clairtoattendhimasasecretarytohisexpedition,whichwasatfirstmeantagainstCanada,butendedinanincursiononthecoastofFrance.Nextyear,towit,1747,IreceivedaninvitationfromtheGeneraltoattendhiminthesamestationinhismilitaryembassytothecourtsofViennaandTurin.Ithenworetheuniformofanofficer,andwasintroducedatthesecourtsasaid-de-camptothegeneral,alongwithSirHarryErskineandCaptainGrant,nowGeneralGrant.Thesetwoyearswerealmosttheonlyinterruptionswhichmystudieshavereceivedduringthecourseofmylife:Ipassedthemagreeably,andingoodcompany;andmyappointments,withmyfrugality,hadmademereachafortune,whichIcalledindependent,thoughmostofmyfriendswereinclinedtosmilewhenIsaidso;inshort,Iwasnowmasterofnearathousandpounds.[8]Ihadalwaysentertainedanotion,thatmywantofsuccessinpublishingtheTreatiseofHumanNature,hadproceededmoreCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nII.MyOwnLife525fromthemannerthanthematter,andthatIhadbeenguiltyofaveryusualindiscretion,ingoingtothepresstooearly.I,therefore,castthefirstpartofthatworkanewintheEnquiryconcerningHumanUnderstanding,whichwaspublishedwhileIwasatTurin.ButthispiecewasatfirstlittlemoresuccessfulthantheTreatiseofHumanNature.OnmyreturnfromItaly,IhadthemortificationtofindallEnglandinaferment,onaccountofDr.Middleton’sFreeEnquiry,whilemyperformancewasentirelyoverlookedandneglected.Anewedition,whichhadbeenpublishedatLondonofmyEssays,moralandpolitical,metnotwithamuchbetterreception.[9]Suchistheforceofnaturaltemper,thatthesedisappointmentsmadelittleornoimpressiononme.Iwentdownin1749,andlivedtwoyearswithmybrotherathiscountry-house,formymotherwasnowdead.ItherecomposedthesecondpartofmyEssays,whichIcalledPoliticalDiscourses,andalsomyEnquiryconcerningthePrinciplesofMorals,whichisanotherpartofmytreatisethatIcastanew.Meanwhile,mybookseller,A.Millar,informedme,thatmyformerpublications(allbuttheunfortunateTreatise)werebeginningtobethesubjectofconversation;thatthesaleofthemwasgraduallyincreasing,andthatneweditionsweredemanded.AnswersbyReverends,andRightReverends,cameouttwoorthreeinayear;andIfound,byDr.Warburton’srailing,thatthebookswerebeginningtobeesteemedingoodcompany.However,Ihadfixedaresolution,whichIinflexiblymaintained,nevertoreplytoanybody;andnotbeingveryirascibleinmytemper,Ihaveeasilykeptmyselfclearofallliterarysquabbles.Thesesymptomsofarisingreputationgavemeencouragement,asIwasevermoredisposedtoseethefavourablethanunfavourablesideofthings;aturnofmindwhichitismorehappytopossess,thantobeborntoanestateoftenthousandayear.[10]In1751,Iremovedfromthecountrytothetown,thetruesceneforamanofletters.In1752,werepublishedatEdinburgh,whereIthenlived,myPoliticalDiscourses,theonlyworkofminethatwassuccessfulonthefirstpublication.Itwaswellreceivedabroadandathome.InthesameyearwaspublishedatLondon,myEnquiryconcerningthePrinciplesofMorals;which,inmyownopinion(whooughtnottojudgeonthatsubject),isofallmyCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n526appendix:hume’sautobiographieswritings,historical,philosophical,orliterary,incomparablythebest.Itcameunnoticedandunobservedintotheworld.[11]In1752,theFacultyofAdvocateschosemetheirLibrarian,anofficefromwhichIreceivedlittleornoemolument,butwhichgavemethecommandofalargelibrary.IthenformedtheplanofwritingtheHistoryofEngland;butbeingfrightenedwiththenotionofcontinuinganarrativethroughaperiodof1700years,IcommencedwiththeaccessionoftheHouseofStuart,anepochwhen,Ithought,themisrepresentationsoffactionbeganchieflytotakeplace.Iwas,Iown,sanguineinmyexpectationsofthesuccessofthiswork.IthoughtthatIwastheonlyhistorian,thathadatonceneglectedpresentpower,interest,andauthority,andthecryofpopularprejudices;andasthesubjectwassuitedtoeverycapacity,Iexpectedproportionalapplause.Butmiserablewasmydisappointment:Iwasassailedbyonecryofreproach,disapprobation,andevendetestation;English,Scotch,andIrish,WhigandTory,churchmanandsectary,freethinkerandreligionist,patriotandcourtier,unitedintheirrageagainsttheman,whohadpresumedtoshedageneroustearforthefateofCharlesI.andtheEarlofStrafford;andafterthefirstebullitionsoftheirfurywereover,whatwasstillmoremortifying,thebookseemedtosinkintooblivion.Mr.Millartoldme,thatinatwelvemonthhesoldonlyforty-fivecopiesofit.Iscarcely,indeed,heardofonemaninthethreekingdoms,considerableforrankorletters,thatcouldendurethebook.ImustonlyexcepttheprimateofEngland,Dr.Herring,andtheprimateofIreland,Dr.Stone,whichseemtwooddexceptions.Thesedignifiedprelatesseparatelysentmemessagesnottobediscouraged.[12]Iwas,however,Iconfess,discouraged;andhadnotthewarbeenatthattimebreakingoutbetweenFranceandEngland,Ihadcertainlyretiredtosomeprovincialtownoftheformerkingdom,havechangedmyname,andnevermorehavereturnedtomynativecountry.Butasthisschemewasnotnowpracticable,andthesubsequentvolumewasconsiderablyadvanced,Iresolvedtopickupcourageandtopersevere.[13]Inthisinterval,IpublishedatLondonmyNaturalHistoryofReligion,alongwithsomeothersmallpieces:itspublicentrywasratherobscure,exceptonlythatDr.HurdwroteapamphletagainstCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nII.MyOwnLife527it,withalltheilliberalpetulance,arrogance,andscurrility,whichdistinguishtheWarburtonianschool.Thispamphletgavemesomeconsolationfortheotherwiseindifferentreceptionofmyperformance.[14]In1756,twoyearsafterthefallofthefirstvolume,waspublishedthesecondvolumeofmyHistory,containingtheperiodfromthedeathofCharlesI.tilltheRevolution.ThisperformancehappenedtogivelessdispleasuretotheWhigs,andwasbetterreceived.Itnotonlyroseitself,buthelpedtobuoyupitsunfortunatebrother.[15]ButthoughIhadbeentaughtbyexperience,thattheWhigpartywereinpossessionofbestowingallplaces,bothinthestateandinliterature,Iwassolittleinclinedtoyieldtotheirsenselessclamour,thatinaboveahundredalterations,whichfartherstudy,reading,orreflectionengagedmetomakeinthereignsofthetwofirstStuarts,IhavemadealloftheminvariablytotheToryside.ItisridiculoustoconsidertheEnglishconstitutionbeforethatperiodasaregularplanofliberty.[16]In1759,IpublishedmyHistoryoftheHouseofTudor.TheclamouragainstthisperformancewasalmostequaltothatagainsttheHistoryofthetwofirstStuarts.ThereignofElizabethwasparticularlyobnoxious.ButIwasnowcallousagainsttheimpressionsofpublicfolly,andcontinuedverypeaceablyandcontentedlyinmyretreatatEdinburgh,tofinish,intwovolumes,themoreearlypartoftheEnglishHistory,whichIgavetothepublicin1761,withtolerable,andbuttolerablesuccess.[17]But,notwithstandingthisvarietyofwindsandseasons,towhichmywritingshadbeenexposed,theyhadstillbeenmakingsuchadvances,thatthecopy-moneygivenmebythebooksellers,muchexceededanythingformerlyknowninEngland;Iwasbecomenotonlyindependent,butopulent.IretiredtomynativecountryofScotland,determinednevermoretosetmyfootoutofit,andretainingthesatisfactionofneverhavingpreferredarequesttoonegreatman,orevenmakingadvancesoffriendshiptoanyofthem.AsIwasnowturnedoffifty,Ithoughtofpassingalltherestofmylifeinthisphilosophicalmanner,whenIreceived,in1763,aninvitationfromtheEarlofHertford,withwhomIwasnotintheleastacquainted,toattendhimonhisembassytoParis,withanearCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n528appendix:hume’sautobiographiesprospectofbeingappointedsecretarytotheembassy,and,inthemeanwhile,ofperformingthefunctionsofthatoffice.Thisoffer,howeverinviting,Iatfirstdeclined,bothbecauseIwasreluctanttobeginconnexionswiththegreat,andbecauseIwasafraidthatthecivilitiesandgaycompanyofParis,wouldprovedisagreeabletoapersonofmyageandhumour:butonhislordship’srepeatingtheinvitation,Iacceptedofit.Ihaveeveryreason,bothofpleasureandinterest,tothinkmyselfhappyinmyconnexionswiththatnobleman,aswellasafterwardswithhisbrother,GeneralConway.[18]Thosewhohavenotseenthestrangeeffectsofmodes,willneverimaginethereceptionImetwithatParis,frommenandwomenofallranksandstations.ThemoreI[recoiled]fromtheirexcessivecivilities,themoreIwasloadedwiththem.Thereis,however,arealsatisfactioninlivingatParis,fromthegreatnumberofsensible,knowing,andpolitecompanywithwhichthatcityaboundsaboveallplacesintheuniverse.Ithoughtonceofsettlingthereforlife.[19]Iwasappointedsecretarytotheembassy;and,insummer1765,LordHertfordleftme,beingappointedLordLieutenantofIreland.Iwaschargedaffaires´tillthearrivaloftheDukeofRichmond,towardstheendoftheyear.Inthebeginningof1766,IleftParis,andnextsummerwenttoEdinburgh,withthesameviewasformerly,ofburyingmyselfinaphilosophicalretreat.Ireturnedtothatplace,notricher,butwithmuchmoremoney,andamuchlargerincome,bymeansofLordHertford’sfriendship,thanIleftit;andIwasdesirousoftryingwhatsuperfluitycouldproduce,asIhadformerlymadeanexperimentofacompetency.But,in1767,IreceivedfromMr.ConwayaninvitationtobeUnder-secretary;andthisinvitation,boththecharacteroftheperson,andmyconnexionswithLordHertford,preventedmefromdeclining.IreturnedtoEdinburghin1769,veryopulent(forIpossessedarevenueof1000l.ayear),healthy,andthoughsomewhatstrickeninyears,withtheprospectofenjoyinglongmyease,andofseeingtheincreaseofmyreputation.[20]Inspring1775,Iwasstruckwithadisorderinmybowels,whichatfirstgavemenoalarm,buthassince,asIapprehendit,becomemortalandincurable.Inowreckonuponaspeedydissolution.Ihavesufferedverylittlepainfrommydisorder;andCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nII.MyOwnLife529whatismorestrange,have,notwithstandingthegreatdeclineofmyperson,neversufferedamoment’sabatementofmyspirits;insomuch,thatwereItonametheperiodofmylife,whichIshouldmostchoosetopassoveragain,Imightbetemptedtopointtothislaterperiod.Ipossessthesameardouraseverinstudy,andthesamegaietyincompany.Iconsider,besides,thatamanofsixty-five,bydying,cutsoffonlyafewyearsofinfirmities;andthoughIseemanysymptomsofmyliteraryreputation’sbreakingoutatlastwithadditionallustre,IknewthatIcouldhavebutfewyearstoenjoyit.ItisdifficulttobemoredetachedfromlifethanIamatpresent.[21]Toconcludehistoricallywithmyowncharacter.Iam,orratherwas(forthatisthestyleImustnowuseinspeakingofmyself,whichemboldensmethemoretospeakmysentiments);Iwas,Isay,amanofmilddispositions,ofcommandoftemper,ofanopen,social,andcheerfulhumour,capableofattachment,butlittlesusceptibleofenmity,andofgreatmoderationinallmypassions.Evenmyloveofliteraryfame,myrulingpassion,neversouredmytemper,notwithstandingmyfrequentdisappointments.Mycompanywasnotunacceptabletotheyoungandcareless,aswellastothestudiousandliterary;andasItookaparticularpleasureinthecompanyofmodestwomen,IhadnoreasontobedispleasedwiththereceptionImetwithfromthem.Inaword,thoughmostmenanywiseeminent,havefoundreasontocomplainofcalumny,Ineverwastouched,orevenattackedbyherbalefultooth:andthoughIwantonlyexposedmyselftotherageofbothcivilandreligiousfactions,theyseemedtobedisarmedinmybehalfoftheirwontedfury.Myfriendsneverhadoccasiontovindicateanyonecircumstanceofmycharacterandconduct:notbutthatthezealots,wemaywellsuppose,wouldhavebeengladtoinventandpropagateanystorytomydisadvantage,buttheycouldneverfindanywhichtheythoughtwouldwearthefaceofprobability.Icannotsaythereisnovanityinmakingthisfuneralorationofmyself,butIhopeitisnotamisplacedone;andthisisamatteroffactwhichiseasilyclearedandascertained.April18,1776.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nselectedbibliographyPartIofthisBibliographyprovidesreaderswithachronologicallist,bydateoffirstpublication,ofHume’sprincipalphilosophical,historical,andlit-eraryworks,andrecommendscurrentlyavailableeditionsoftheseworks.PartIIrecommendsseveralhelpfulandgenerallyreliable(butincomplete)printedandonlinebibliographiesofworkonHumeandthreebiographies.PartIIIrecommendsseventeenanthologiesofarticlesonawiderangeofHume’sthought.PartIVrecommendsapproximatelyfiftymonographsonHumeorsomeaspectofhisthought.Thosewhoareabletolookatthebibli-ographieslistedinPartIIwillrealizethattherearenowthousandsofarticlesandbooksdevotedtoHume’swork,andthatnocompletebibliographyisavailable.Foralistoftheeditionsandabbreviationsusedinthisvolume,seeabove,“MethodofCitation.”i.hume’swritingsWepresenthere,intheapproximatechronologicalorderoforiginalpublica-tion,theprincipalworksofHume.Wealsoprovidecross-referencestothe“MethodofCitation”atthebeginningofthisvolume,becausethisindicateswhichcurrentlyavailableeditionswehaverecommendedbyusingtheminthisvolume.Attheendofthislistwediscusssomeofthecurrentlyavailableelec-troniceditionsofselectedworks.ATreatiseofHumanNature:BeinganAttempttointroducetheexperimen-talMethodofReasoningintoMoralSubjects.3vols.London,1739–40.ThecriticaleditionofthistextpublishedbyOxfordUniversityPressisusedinthisvolume.Fordetails,seeT,“MethodofCitation.”AnAbstractofaBooklatelyPublished;Entituled,ATreatiseofHumanNature,&c.WhereintheChiefArgumentofthatBookisfartherIllus-tratedandExplained.London,1740.Priortotherediscoveryofthiswork531CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n532SelectedBibliographybyJ.M.KeynesandP.Sraffainthe1930sitwassupposedthatthismiss-ingworkhadbeenwrittenbyAdamSmith,eventhoughthosewhohadseentheworkattributedittoHumeuntilabout1825.ScholarsarenowwidelyagreedthatHumewasitsauthor.ThecriticaleditionofthistextpublishedbyOxfordUniversityPressisusedinthisvolume.Fordetails,seeA,“MethodofCitation.”Essays,MoralandPolitical.2vols.Edinburgh,1741–2.Foradditionaldetails,seebelow,Essays,Moral,Political,andLiteraryinEssaysandTreatisesonSeveralSubjects.ALetterfromaGentlemantoHisFriendinEdinburgh:ContainingSomeObservationsonASpecimenofthePrinciplesconcerningReligionandMorality,saidtobemaintaindinaBooklatelypublishd,intituled,ATreatiseofHumanNature,&c.Edinburgh,1745.ThecriticaleditionofthistextpublishedbyOxfordUniversityPressisusedinthisvolume.Fordetailsregardingfacsimileandcriticaleditionsofthiswork,seeL,“MethodofCitation.”AnEnquiryconcerningHumanUnderstanding,firstpublishedasPhilo-sophicalEssaysconcerningHumanUnderstanding.London,1748.ThecriticaleditionofthistextpublishedbyOxfordUniversityPressisusedinthisvolume.Fordetails,seeEHU,“MethodofCitation.”AnEnquiryconcerningthePrinciplesofMorals.London,1751.ThecriticaleditionofthistextpublishedbyOxfordUniversityPressisusedinthisvolume.Fordetails,seeEPM,“MethodofCitation.”PoliticalDiscourses.Edinburgh,1752.Theessaysinthisworkwereeven-tuallyincludedinEssaysandTreatisesonSeveralSubjects,acollectionfirstpublished1753–6.Fordetails,seeE,“MethodofCitation.”EssaysandTreatisesonSeveralSubjects.Beginningin1753,thefirstEnquiry(EHU),secondEnquiry(EPM),andessays(includingmaterialsincludedinthePoliticalDiscoursesof1752and,beginningin1758,theFourDissertationsof1757)werebroughttogetherandpublishedunderthistitle.Humesawthroughthepresseditionsthatwerewerepublishedin1753–6,1758,1760,1764,1767,1768,1770,1772,andpreparedthetextofthefirstposthumousedition,thatof1777.FourDissertations.London,1757.Thefouritemsincludedinthisvolume(“TheNaturalHistoryofReligion,”“OfthePassions,”“OfTragedy,”and“OftheStandardofTaste”)werefrom1758includedinEssaysandTreatisesonSeveralSubjects.ThecriticaltextsofADissertationonthePassionsandTheNaturalHistoryofReligion,publishedbyOxfordUniversityPress,areusedinthisvolume.Fordetails,seeDPandNHR,“MethodofCitation.”TheHistoryofEngland,fromtheInvasionofJuliusCaesartoTheRevolu-tionin1688.TheworknowpublishedunderthistitlewasfirstpublishedCambridgeCollectionsOnline©CambridgeUniversityPress,2009\nSelectedBibliography533insixvolumesoveraneight-yearperiod,1754–62.Humefirstpublished(1754,1757)twovolumesontheperiod1603–88,thentwomore(1759)abouttheperiod1485–1603,and,finally,twovolumes(1762)abouttheperiodfromc.50b.c.toa.d.1485.Fordetailsabouttheeditionusedinthisvolume,seeHE,“MethodofCitation.”Forinformationaboutsomeofthemorethan175posthumouseditionsofthiswork,seeDavidHume:PhilosophicalHistorian.EditedbyD.F.NortonandR.H.Popkin.Indi-anapolis:Bobbs-Merrill,1965.AppendixA.TheLifeofDavidHume,Esq.WrittenbyHimself.London,1777.PublishedintheAppendixtothisvolume.Fordetails,seeMOL,“MethodofCita-tion.”DialoguesconcerningNaturalReligion.London,1779.Firsteditedandpub-lishedbyHume’snephew,DavidHumetheYounger.Fordetailsoftheeditionusedinthisvolume,seeDNR,“MethodofCitation.”TheLettersofDavidHume.EditedbyJ.Y.T.Greig.2vols.Oxford:Claren-donPress,1932.NewLettersofDavidHume.EditedbyR.KlibanskyandE.C.Mossner.Oxford:ClarendonPress,1954.ManyadditionallettersofHumehavebeenpublishedsince1954,inmanyjournalsandbooks.AfewofthehundredsofletterstoHumearepublishedinLettersofEminentPersonsAddressedtoDavidHume,ed.J.H.Burton(Edinburgh,1849).AcatalogofseveralhundredsuchlettersisincludedinCalendarofHumeMss.inthePossessionoftheRoyalSocietyofEdinburgh,ed.J.Y.T.GreigandH.Beynon(Edinburgh:RoyalSocietyofEdinburgh,1932).Computer-ReadableCollectionsAsthisvolumegoestopressthemostcomprehensivecollectionofHume’stextsavailableinelectronicformisthatinthePastMastersSeriesoftheInteLexCorporation.Thiscollectioncontainstheworks,includingthethreevolumesofHume’sletters,mentionedabove.Unfortunately,withoneexception,thetextsavailablefromthissourceareassembledfromahotchpotchofinauthenticandcasuallypreparededitions,someofwhichmodernizespellingandpunctuation,andonlyoneofwhichderivesfromauthenticeditionspreparedforthepressbyHumehimself.Theoneexcep-tionisTheHistoryofEngland.TheInteLextextofthisworkisbasedontheeditionpublishedin1778,aneditionthatHumerevisedbeforehisdeathin1776,butdidnotseethroughthepress.Fortunately,OxfordUniversityPressplanstomakeavailableonlinethecriticaleditionspublishedaspartoftheClarendonEditionoftheWorksofCambridgeCollectionsOnline©CambridgeUniversityPress,2009\n534SelectedBibliographyDavidHume(Generaleditors:T.L.Beauchamp,D.F.Norton,andM.A.Stewart).Theseare:ATreatiseofHumanNature,AnAbstractoftheTrea-tise,ALetterfromaGentleman,AnEnquiryconcerningHumanUnder-standing,AnEnquiryconcerningthePrinciplesofMorals,ADissertationonthePassions,andTheNaturalHistoryofReligion.Inaddition,thetextofthe1777editionofAnEnquiryconcerningHumanUnderstanding,anothertextthatHumerevisedbeforehisdeathin1776butdidnotseethroughthepress,isavailableatwww.etext.leeds.ac.uk/hume/.partii.selectedbibliographiesandbiographiesA.BibliographiesFieser,James.ForanonlinelistofmanyoftheresponsestoHume’sworks,1739–1900,gotohttp://www.utm.edu/staff/jfieser/humebib/.Hall,Roland.FiftyYearsofHumeScholarship:ABibliographicalGuide.Edinburgh:EdinburghUniversityPress,1978.ThisbibliographyincludestheprincipalwritingonHumefortheyears1900–24andacomprehensivelistofHumeliterature,1925–76.Hall,Roland,WilliamE.Morris,andJamesFieser.“TheHumeLiteraturefor1976.”HumeStudies2(1977):94–102,andsubsequentyears.HumeStudieshaspublishedannualbibliographiesofworkonHumefortheyears1976–2003.ThesebibliographieshavebeenpreparedbyRolandHall(1977–85),WilliamE.Morris(1986–2003),and,from2004,JamesFieser.Jessop,T.E.ABibliographyofDavidHumeandofScottishPhilosophyfromFrancisHutchesontoLordBalfour.London:BrownandSon,1938.Facsimileedition,NewYork:Garland,1983.B.BiographiesBurton,JohnHill.LifeandCorrespondenceofDavidHume.Edinburgh,1846.Greig,J.Y.T.DavidHume.Oxford:OxfordUniversityPress,1932.Mossner,ErnestCampbell.TheLifeofDavidHume,2nded.Oxford:Claren-donPress,1980.partiii.selectedanthologiesonhumeCapaldi,Nicholas,andDonaldW.Livingston,eds.LibertyinHumesHis-toryofEngland.Dordrecht:Kluwer,1990.Chappell,V.C.,ed.Hume:ACollectionofCriticalEssays.NewYork:Dou-bleday,1966.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nSelectedBibliography535Cohon,Rachel,ed.Hume:MoralandPoliticalPhilosophy.Aldershot,Eng-land:Dartmouth,2001.Frasca-Spada,Maria,andP.J.E.Kail,eds.ImpressionsofHume.Oxford:OxfordUniversityPress,2005.Jacobson,AnneJaap,ed.FeministinterpretationsofDavidHume.Univer-sityPark:PennsylvaniaStateUniversityPress,2000.Jones,Peter,ed.TheReceptionofDavidHumeinEurope.Bristol:Thoemmes-Continuum,2005.Livingston,DonaldW.,andJamesT.King,eds.Hume:ARe-evaluation.NewYork:FordhamUniversityPress,1976.Mazza,Emilio,andEmanueleRonchetti.NewEssaysonDavidHume.Milan:FrancoAngeli,2007.Millican,Peter,ed.ReadingHumeonHumanUnderstanding:EssaysontheFirstEnquiry.Oxford:ClarendonPress,2002.Morice,G.P.,ed.DavidHumeBicentenaryPapers.Edinburgh:UniversityofEdinburghPress,1977.Norton,DavidFate,NicholasCapaldi,andWadeL.Robison,eds.McGillHumeStudies.SanDiego:AustinHillPress,1976.Owen,DavidW.D.,ed.Hume:GeneralPhilosophy.Aldershot,England:Dartmouth,2000.Read,Rupert,andKennethA.Richman,eds.TheNewHumeDebate.Lon-don:Routledge,2000.Schabas,Margaret,andCarlWennerlind,eds.DavidHumesPoliticalEcon-omy.London:Routledge,2007.Stewart,M.A.,ed.StudiesinthePhilosophyoftheScottishEnlightenment.OxfordStudiesintheHistoryofPhilosophy,vol.1.Oxford:ClarendonPress,1990.Stewart,M.A.,andJohnP.Wright,eds.HumeandHumesConnexions.Edinburgh:EdinburghUniversityPress,1994.Traiger,Saul,ed.TheBlackwellGuidetoHumesTreatise.Oxford:Black-well,2006.partiv.selectedmonographsA.Eighteenth-andNineteenth-CenturyWorksBeattie,James.AnEssayontheNatureandImmutabilityofTruth;InOppo-sitiontoSophistryandScepticism.Edinburgh,1770.Brown,Thomas.InquiryintotheRelationofCauseandEffect.3rded.Edinburgh,1818.OriginallypublishedasObservationsontheNatureandTendencyoftheDoctrineofMr.HumeconcerningtheRelationofCauseandEffect.Edinburgh,1805.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n536SelectedBibliographyCampbell,George.ADissertationonMiracles:ContaininganExaminationofthePrinciplesAdvancedbyDavidHume,Esq;inanEssayonMiracles.Edinburgh,1762.Fieser,James,ed.EarlyResponsestoHume.2nded.10vols.Bristol:Thoemmes-Continuum,2004.Green,T.H.“GeneralIntroduction”and“IntroductiontotheMoralPartoftheTreatise.”InATreatiseofHumanNature,editedbyT.H.GreenandT.H.Grose.2vols.London,1874.1:1–299,2:1–71.Home,Henry,LordKames.EssaysonthePrinciplesofMoralityandNaturalReligion.Edinburgh,1751.Leland,John.AViewofthePrincipalDeisticalWritersoftheLastandPresentCentury.3rded.3vols.London,1755–7.Paley,William.ViewoftheEvidencesofChristianity.London,1794.Priestley,Joseph.LetterstoaPhilosophicalUnbeliever.Bath,1780.Reid,Thomas.AnInquiryintotheHumanMind,onthePrinciplesofCom-monSense.Edinburgh,1764.Shepherd,LadyMary.EssayupontheRelationofCauseandEffect,Contro-vertingtheDoctrineofMr.Hume.London,1824.Warburton,William,andRichardHurd.RemarksonMr.DavidHumesEssayontheNaturalHistoryofReligion.London,1757.Whately,Richard.HistoricDoubtsRelativetoNapoleonBuonaparte.Lon-don,1819.B.Twentieth-andTwenty-First-CenturyWorksArdal,P´allS.´PassionandValueinHumesTreatise.2nded.Edinburgh:EdinburghUniversityPress,1989.Baier,AnnetteC.AProgressofSentiments:ReflectionsonHumesTreatise.Cambridge,MA:HarvardUniversityPress,1991.Baxter,DonaldL.M.HumesDifficulty:TimeandIdentityintheTreatise.NewYork:Routledge,2008.Bongie,LaurenceL.DavidHume:ProphetoftheCounter-Revolution.Oxford:ClarendonPress,1965.Box,M.A.TheSuasiveArtofDavidHume.Princeton:PrincetonUniversityPress,1990.Bricke,John.MindandMorality.Oxford:ClarendonPress,1996.Buckle,Stephen.HumesEnlightenmentTract:TheUnityandPurposeofAnEnquiryconcerningHumanUnderstanding.Oxford:ClarendonPress,2001.Costelloe,TimothyM.AestheticsandMoralsinthePhilosophyofDavidHume.NewYork:Routledge,2007.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\nSelectedBibliography537Flew,Antony.HumesPhilosophyofBelief:AStudyofHisFirstInquiry.London:Routledge&KeganPaul,1961.Fogelin,RobertJ.ADefenseofHumeonMiracles.Princeton:PrincetonUniversityPress,2003.Forbes,Duncan.HumesPhilosophicalPolitics.Cambridge:CambridgeUni-versityPress,1975.Garrett,Don.CognitionandCommitmentinHumesPhilosophy.NewYork:OxfordUniversityPress,1997.Haakonssen,Knud.NaturalLawandMoralPhilosophy:FromGrotiustotheScottishEnlightenment.Cambridge:CambridgeUniversityPress,1996.Hendel,CharlesW.StudiesinthePhilosophyofDavidHume.Princeton:PrincetonUniversityPress,1925;2nded.,Indianapolis:Bobbs-Merrill,1963.Herdt,JenniferA.ReligionandFactioninHumesMoralPhilosophy.Cam-bridge:CambridgeUniversityPress,1997.Jones,Peter.HumesSentiments:TheirCiceronianandFrenchContext.Edinburgh:EdinburghUniversityPress,1982.Kail,P.J.E.ProjectionandRealisminHumesPhilosophy.NewYork:OxfordUniversityPress,2007.Laird,John.HumesPhilosophyofHumanNature.London:Methuen,1932.Livingston,DonaldW.HumesPhilosophyofCommonLife.Chicago:Uni-versityofChicagoPress,1984.Loeb,LouisE.StabilityandJustificationinHumesTreatise.NewYork:OxfordUniversityPress,2002.Mackie,JohnL.HumesMoralTheory.London:Routledge&KeganPaul,1980.MacNabb,D.G.C.DavidHume:HisTheoryofKnowledgeandMorality.2nded.Hamden,CT:ArchonBooks,1966.Mall,R.A.ExperienceandReason:ThePhenomenologyofHusserlandItsRelationtoHumesPhilosophy.TheHague:MartinusNijhoff,1973.Mercer,P.SympathyandEthics:AStudyoftheRelationshipbetweenSym-pathyandMorality,withSpecialReferencetoHumesTreatise.Oxford:ClarendonPress,1972.Miller,David.PhilosophyandIdeologyinHumesPoliticalThought.Oxford:ClarendonPress,1981.Norton,DavidFate.DavidHume:Common-SenseMoralist,ScepticalMeta-physician.Rev.ed.Princeton:PrincetonUniversityPress,1984.Orr,James.DavidHumeandHisInfluenceonPhilosophyandTheology.NewYork:Scribner’s,1903.Owen,David,HumesReason.Oxford:OxfordUniversityPress,1999.Passmore,John.HumesIntentions.2nded.NewYork:BasicBooks,1968.CambridgeCollectionsOnline©CambridgeUniversityPress,2009\n538SelectedBibliographyPenelhum,Terence.DavidHume:AnIntroductiontoHisPhilosophicalSystem.WestLafayette,IN:PurdueUniversityPress,1992.Price,H.H.HumesTheoryoftheExternalWorld.Oxford:ClarendonPress,1940.Rivers,Isabel,Reason,Grace,andSentiment:AStudyoftheLanguageofReligionandEthicsinEngland,1660–1780.2vols.Cambridge:CambridgeUniversityPress,1991,2000.Russell,Paul.FreedomandMoralSentiment:HumesWayofNaturalizingResponsibility.NewYork:OxfordUniversityPress,1995.TheRiddleofHumesTreatise:Skepticism,Naturalism,andIrreligion.NewYork:OxfordUniversityPress,2008.Siebert,DonaldT.TheMoralAnimusofDavidHume.Newark:UniversityofDelawarePress,1990.Smith,NormanKemp,ThePhilosophyofDavidHume:ACriticalStudyofItsOriginsandCentralDoctrines.London:Macmillan,1941.Spencer,MarkG.DavidHumeandEighteenth-CenturyAmerica.Rochester,NY:UniversityofRochesterPress,2005.Stewart,JohnB.OpinionandReforminHumesPoliticalPhilosophy.Princeton,NJ:PrincetonUniversityPress,1992.Strawson,Galen.TheSecretConnexion:Causation,Realism,andDavidHume.Oxford:ClarendonPress,1989.Stroud,Barry.Hume.London:Routledge&KeganPaul,1977.Waxman,Wayne.HumesTheoryofConsciousness.Cambridge:CambridgeUniversityPress,1994.Whelan,FrederickG.OrderandArtificeinHumesPoliticalPhilosophy.Princeton,NJ:PrincetonUniversityPress,1985.Wright,JohnP.TheScepticalRealismofDavidHume.Manchester:Man-chesterUniversityPress,1983.Zabeeh,Farhang.Hume:PrecursorofModernEmpiricism.2nded.TheHague:MartinusNijhoff,1973.CambridgeCollectionsOnline©CambridgeUniversityPress,2009