Rationalism vs Empiricism (SEP)

Stanford Encyclopedia of PhilosophyRationalismvs.EmpiricismFirstpublishedThuAug19,2004substantiverevisionThuMar21,2013

Thedisputebetweenrationalismandempiricismconcernstheextenttowhichwearedependentuponsenseexperienceinourefforttogainknowledge.Rationalistsclaimthattherearesignificantwaysinwhichourconceptsandknowledgearegainedindependentlyofsenseexperience.Empiricistsclaimthatsenseexperienceistheultimatesourceofallourconceptsandknowledge.

Rationalistsgenerallydeveloptheirviewintwoways.First,theyarguethattherearecaseswherethecontentofourconceptsorknowledgeoutstripstheinformationthatsenseexperiencecanprovide.Second,theyconstructaccountsofhowreasoninsomeformorotherprovidesthatadditionalinformationabouttheworld.Empiricistspresentcomplementarylinesofthought.First,theydevelopaccountsofhowexperienceprovidestheinformationthatrationalistscite,insofaraswehaveitinthefirstplace.(Empiricistswillattimesoptforskepticismasanalternativetorationalism:ifexperiencecannotprovidetheconceptsorknowledgetherationalistscite,thenwedon'thavethem.)Second,empiricistsattacktherationalists'accountsofhowreasonisasourceofconceptsorknowledge.

1.Introduction1.1Rationalism1.2Empiricism

2.TheIntuition/DeductionThesis3.TheInnateKnowledgeThesis4.TheInnateConceptThesisBibliographyAcademicToolsOtherInternetResourcesRelatedEntries

1.Introduction

Thedisputebetweenrationalismandempiricismtakesplacewithinepistemology,thebranchofphilosophydevotedtostudyingthenature,sourcesandlimitsofknowledge.Thedefiningquestionsofepistemologyincludethefollowing.

1. Whatisthenatureofpropositionalknowledge,knowledgethataparticularpropositionabouttheworldistrue?

Toknowaproposition,wemustbelieveitanditmustbetrue,butsomethingmoreisrequired,somethingthatdistinguishesknowledgefromaluckyguess.Let'scallthisadditionalelementwarrant.Agooddealofphilosophicalworkhasbeeninvestedintryingtodeterminethenatureofwarrant.

2. Howcanwegainknowledge?

Wecanformtruebeliefsjustbymakingluckyguesses.Howtogainwarrantedbeliefsislessclear.Moreover,toknowtheworld,wemustthinkaboutit,anditisunclearhowwegaintheconceptsweuseinthoughtorwhatassurance,ifany,wehavethatthewaysinwhichwedivide

uptheworldusingourconceptscorrespondtodivisionsthatactuallyexist.

3. Whatarethelimitsofourknowledge?

Someaspectsoftheworldmaybewithinthelimitsofourthoughtbutbeyondthelimitsofourknowledgefacedwithcompetingdescriptionsofthem,wecannotknowwhichdescriptionistrue.Someaspectsoftheworldmayevenbebeyondthelimitsofourthought,sothatwecannotformintelligibledescriptionsofthem,letaloneknowthataparticulardescriptionistrue.

Thedisagreementbetweenrationalistsandempiricistsprimarilyconcernsthesecondquestion,regardingthesourcesofourconceptsandknowledge.Insomeinstances,theirdisagreementonthistopicleadsthemtogiveconflictingresponsestotheotherquestionsaswell.Theymaydisagreeoverthenatureofwarrantoraboutthelimitsofourthoughtandknowledge.Ourfocusherewillbeonthecompetingrationalistandempiricistresponsestothesecondquestion.

1.1Rationalism

Tobearationalististoadoptatleastoneofthreeclaims.TheIntuition/Deductionthesisconcernshowwebecomewarrantedinbelievingpropositionsinaparticularsubjectarea.

TheIntuition/DeductionThesis:Somepropositionsinaparticularsubjectarea,S,areknowablebyusbyintuitionalonestillothersareknowablebybeingdeducedfromintuitedpropositions.

Intuitionisaformofrationalinsight.Intellectuallygraspingaproposition,wejustseeittobetrueinsuchawayastoformatrue,warrantedbeliefinit.(AsdiscussedinSection2below,thenatureofthisintellectualseeingneedsexplanation.)Deductionisaprocessinwhichwederiveconclusionsfromintuitedpremisesthroughvalidarguments,onesinwhichtheconclusionmustbetrueifthepremisesaretrue.Weintuit,forexample,thatthenumberthreeisprimeandthatitisgreaterthantwo.Wethendeducefromthisknowledgethatthereisaprimenumbergreaterthantwo.Intuitionanddeductionthusprovideuswithknowledgeapriori,whichistosayknowledgegainedindependentlyofsenseexperience.

WecangeneratedifferentversionsoftheIntuition/DeductionthesisbysubstitutingdifferentsubjectareasforthevariableS.Somerationaliststakemathematicstobeknowablebyintuitionanddeduction.Someplaceethicaltruthsinthiscategory.Someincludemetaphysicalclaims,suchasthatGodexists,wehavefreewill,andourmindandbodyaredistinctsubstances.Themorepropositionsrationalistsincludewithintherangeofintuitionanddeduction,andthemorecontroversialthetruthofthosepropositionsortheclaimstoknowthem,themoreradicaltheirrationalism.

Rationalistsalsovarythestrengthoftheirviewbyadjustingtheirunderstandingofwarrant.Sometakewarrantedbeliefstobebeyondeventheslightestdoubtandclaimthatintuitionanddeductionprovidebeliefsofthishighepistemicstatus.Othersinterpretwarrantmoreconservatively,sayasbeliefbeyondareasonabledoubt,andclaimthatintuitionanddeductionprovidebeliefsofthatcaliber.Stillanotherdimensionofrationalismdependsonhowitsproponentsunderstandtheconnectionbetweenintuition,ontheonehand,andtruth,ontheother.Sometakeintuitiontobeinfallible,claimingthatwhateverweintuitmustbetrue.Othersallowforthepossibilityoffalseintuitedpropositions.

ThesecondthesisassociatedwithrationalismistheInnateKnowledgethesis.

TheInnateKnowledgeThesis:Wehaveknowledgeofsometruthsinaparticularsubjectarea,S,aspartofourrationalnature.

LiketheIntuition/Deductionthesis,theInnateKnowledgethesisassertstheexistenceofknowledge

gainedapriori,independentlyofexperience.Thedifferencebetweenthemrestsintheaccompanyingunderstandingofhowthisaprioriknowledgeisgained.TheIntuition/Deductionthesiscitesintuitionandsubsequentdeductivereasoning.TheInnateKnowledgethesisoffersourrationalnature.Ourinnateknowledgeisnotlearnedthrougheithersenseexperienceorintuitionanddeduction.Itisjustpartofournature.Experiencesmaytriggeraprocessbywhichwebringthisknowledgetoconsciousness,buttheexperiencesdonotprovideuswiththeknowledgeitself.Ithasinsomewaybeenwithusallalong.Accordingtosomerationalists,wegainedtheknowledgeinanearlierexistence.Accordingtoothers,Godprovideduswithitatcreation.Stillotherssayitispartofournaturethroughnaturalselection.

WegetdifferentversionsoftheInnateKnowledgethesisbysubstitutingdifferentsubjectareasforthevariableS'.Onceagain,themoresubjectsincludedwithintherangeofthethesisorthemorecontroversialtheclaimtohaveknowledgeinthem,themoreradicaltheformofrationalism.Strongerandweakerunderstandingsofwarrantyieldstrongerandweakerversionsofthethesisaswell.

ThethirdimportantthesisofrationalismistheInnateConceptthesis.

TheInnateConceptThesis:Wehavesomeoftheconceptsweemployinaparticularsubjectarea,S,aspartofourrationalnature.

AccordingtotheInnateConceptthesis,someofourconceptsarenotgainedfromexperience.Theyarepartofourrationalnatureinsuchawaythat,whilesenseexperiencesmaytriggeraprocessbywhichtheyarebroughttoconsciousness,experiencedoesnotprovidetheconceptsordeterminetheinformationtheycontain.SomeclaimthattheInnateConceptthesisisentailedbytheInnateKnowledgeThesisaparticularinstanceofknowledgecanonlybeinnateiftheconceptsthatarecontainedintheknownpropositionarealsoinnate.ThisisLocke'sposition(1690,BookI,ChapterIV,Section1,p.91).Others,suchasCarruthers,argueagainstthisconnection(1992,pp.5354).ThecontentandstrengthoftheInnateConceptthesisvarieswiththeconceptsclaimedtobeinnate.Themoreaconceptseemsremovedfromexperienceandthementaloperationswecanperformonexperiencethemoreplausiblyitmaybeclaimedtobeinnate.Sincewedonotexperienceperfecttrianglesbutdoexperiencepains,ourconceptoftheformerisamorepromisingcandidateforbeinginnatethanourconceptofthelatter.

TheIntuition/Deductionthesis,theInnateKnowledgethesis,andtheInnateConceptthesisareessentialtorationalism:tobearationalististoadoptatleastoneofthem.Twoothercloselyrelatedthesesaregenerallyadoptedbyrationalists,althoughonecancertainlybearationalistwithoutadoptingeitherofthem.Thefirstisthatexperiencecannotprovidewhatwegainfromreason.

TheIndispensabilityofReasonThesis:Theknowledgewegaininsubjectarea,S,byintuitionanddeduction,aswellastheideasandinstancesofknowledgeinSthatareinnatetous,couldnothavebeengainedbyusthroughsenseexperience.

Thesecondisthatreasonissuperiortoexperienceasasourceofknowledge.

TheSuperiorityofReasonThesis:TheknowledgewegaininsubjectareaSbyintuitionanddeductionorhaveinnatelyissuperiortoanyknowledgegainedbysenseexperience.

Howreasonissuperiorneedsexplanation,andrationalistshaveoffereddifferentaccounts.Oneview,generallyassociatedwithDescartes(1628,RulesIIandIII,pp.14),isthatwhatweknowaprioriiscertain,beyondeventheslightestdoubt,whilewhatwebelieve,orevenknow,onthebasisofsenseexperienceisatleastsomewhatuncertain.Anotherview,generallyassociatedwithPlato(Republic479e484c),locatesthesuperiorityofaprioriknowledgeintheobjectsknown.Whatweknowbyreasonalone,aPlatonicform,say,issuperiorinanimportantmetaphysicalway,e.g.unchanging,eternal,perfect,ahigherdegreeofbeing,towhatweareawareofthroughsenseexperience.

Mostformsofrationalisminvolvenotablecommitmentstootherphilosophicalpositions.Oneisacommitmenttothedenialofscepticismforatleastsomeareaofknowledge.Ifweclaimtoknowsometruthsbyintuitionordeductionortohavesomeinnateknowledge,weobviouslyrejectscepticismwithregardtothosetruths.RationalismintheformoftheIntuition/Deductionthesisisalsocommittedtoepistemicfoundationalism,theviewthatweknowsometruthswithoutbasingourbeliefinthemonanyothersandthatwethenusethisfoundationalknowledgetoknowmoretruths.

1.2Empiricism

Empiricistsendorsethefollowingclaimforsomesubjectarea.

TheEmpiricismThesis:WehavenosourceofknowledgeinSorfortheconceptsweuseinSotherthansenseexperience.

EmpiricismaboutaparticularsubjectrejectsthecorrespondingversionoftheIntuition/DeductionthesisandInnateKnowledgethesis.Insofaraswehaveknowledgeinthesubject,ourknowledgeisaposteriori,dependentuponsenseexperience.EmpiricistsalsodenytheimplicationofthecorrespondingInnateConceptthesisthatwehaveinnateideasinthesubjectarea.Senseexperienceisouronlysourceofideas.TheyrejectthecorrespondingversionoftheSuperiorityofReasonthesis.Sincereasonalonedoesnotgiveusanyknowledge,itcertainlydoesnotgiveussuperiorknowledge.EmpiricistsgenerallyrejecttheIndispensabilityofReasonthesis,thoughtheyneednot.TheEmpiricismthesisdoesnotentailthatwehaveempiricalknowledge.Itentailsthatknowledgecanonlybegained,ifatall,byexperience.Empiricistsmayassert,assomedoforsomesubjects,thattherationalistsarecorrecttoclaimthatexperiencecannotgiveusknowledge.Theconclusiontheydrawfromthisrationalistlessonisthatwedonotknowatall.

Ihavestatedthebasicclaimsofrationalismandempiricismsothateachisrelativetoaparticularsubjectarea.Rationalismandempiricism,sorelativized,neednotconflict.Wecanberationalistsinmathematicsoraparticularareaofmathematicsandempiricistsinallorsomeofthephysicalsciences.Rationalismandempiricismonlyconflictwhenformulatedtocoverthesamesubject.Thenthedebate,Rationalismvs.Empiricism,isjoined.Thefactthatphilosopherscanbebothrationalistsandempiricistshasimplicationsfortheclassificationschemesoftenemployedinthehistoryofphilosophy,especiallytheonetraditionallyusedtodescribetheEarlyModernPeriodoftheseventeenthandeighteenthcenturiesleadinguptoKant.Itisstandardpracticetogroupthemajorphilosophersofthisperiodaseitherrationalistsorempiricistsandtosuggestthatthoseunderoneheadingshareacommonagendainoppositiontothoseundertheother.Thus,Descartes,SpinozaandLeibnizaretheContinentalRationalistsinoppositiontoLocke,BerkeleyandHume,theBritishEmpiricists.Weshouldadoptsuchgeneralclassificationschemeswithcaution.Theviewsoftheindividualphilosophersaremoresubtleandcomplexthanthesimplemindedclassificationsuggests.(SeeLoeb(1981)andKenny(1986)forimportantdiscussionsofthispoint.)LockerejectsrationalismintheformofanyversionoftheInnateKnowledgeorInnateConcepttheses,buthenonethelessadoptstheIntuition/DeductionthesiswithregardtoourknowledgeofGod'sexistence.DescartesandLockehaveremarkablysimilarviewsonthenatureofourideas,eventhoughDescartestakesmanytobeinnate,whileLocketiesthemalltoexperience.Therationalist/empiricistclassificationalsoencouragesustoexpectthephilosophersoneachsideofthedividetohavecommonresearchprogramsinareasbeyondepistemology.Thus,Descartes,SpinozaandLeibnizaremistakenlyseenasapplyingareasoncenteredepistemologytoacommonmetaphysicalagenda,witheachtryingtoimproveontheeffortsoftheonebefore,whileLocke,BerkeleyandHumearemistakenlyseenasgraduallyrejectingthosemetaphysicalclaims,witheachconsciouslytryingtoimproveontheeffortsofhispredecessors.Itisalsoimportanttonotethattherationalist/empiricistdistinctionisnotexhaustiveofthepossiblesourcesofknowledge.Onemightclaim,forexample,thatwecangainknowledgeinaparticularareabyaformofDivinerevelationorinsightthatisaproductofneitherreasonnorsenseexperience.Inshort,whenusedcarelessly,thelabelsrationalistandempiricist,aswellasthesloganthatisthetitleofthisessay,Rationalismvs.Empiricism,canretardratherthan

advanceourunderstanding.

Nonetheless,animportantdebateproperlydescribedasRationalismvs.Empiricismisjoinedwhenevertheclaimsforeachviewareformulatedtocoverthesamesubject.Whatisperhapsthemostinterestingformofthedebateoccurswhenwetaketherelevantsubjecttobetruthsabouttheexternalworld,theworldbeyondourownminds.Afullfledgedrationalistwithregardtoourknowledgeoftheexternalworldholdsthatsomeexternalworldtruthscanandmustbeknownapriori,thatsomeoftheideasrequiredforthatknowledgeareandmustbeinnate,andthatthisknowledgeissuperiortoanythatexperiencecouldeverprovide.Thefullfledgedempiricistaboutourknowledgeoftheexternalworldrepliesthat,whenitcomestothenatureoftheworldbeyondourownminds,experienceisoursolesourceofinformation.Reasonmightinformusoftherelationsamongourideas,butthoseideasthemselvescanonlybegained,andanytruthsabouttheexternalrealitytheyrepresentcanonlybeknown,onthebasisofsenseexperience.Thisdebateconcerningourknowledgeoftheexternalworldwillgenerallybeourmainfocusinwhatfollows.

Historically,therationalist/empiricistdisputeinepistemologyhasextendedintotheareaofmetaphysics,wherephilosophersareconcernedwiththebasicnatureofreality,includingtheexistenceofGodandsuchaspectsofournatureasfreewillandtherelationbetweenthemindandbody.Majorrationalists(e.g.,Descartes1641)havepresentedmetaphysicaltheories,whichtheyhaveclaimedtoknowbyreasonalone.Majorempiricists(e.g.,Hume173940)haverejectedthetheoriesaseitherspeculation,beyondwhatwecanlearnfromexperience,ornonsensicalattemptstodescribeaspectsoftheworldbeyondtheconceptsexperiencecanprovide.Thedebateraisestheissueofmetaphysicsasanareaofknowledge.Kantputsthedrivingassumptionclearly:

Theveryconceptofmetaphysicsensuresthatthesourcesofmetaphysicscan'tbeempirical.Ifsomethingcouldbeknownthroughthesenses,thatwouldautomaticallyshowthatitdoesn'tbelongtometaphysicsthat'sanupshotofthemeaningoftheword'metaphysics.'Itsbasicprinciplescanneverbetakenfromexperience,norcanitsbasicconceptsforitisnottobephysicalbutmetaphysicalknowledge,soitmustbebeyondexperience.(1783,Preamble,I,p.7)

Thepossibilitythenofmetaphysicssounderstood,asanareaofhumanknowledge,hingesonhowweresolvetherationalist/empiricistdebate.Thedebatealsoextendsintoethics.Somemoralobjectivists(e.g.,Ross1930)takeustoknowsomefundamentalobjectivemoraltruthsbyintuition,whilesomemoralskeptics,whorejectsuchknowledge,(e.g.,Mackie1977)findtheappealtoafacultyofmoralintuitionutterlyimplausible.Morerecently,therationalist/empiricistdebatehasextendedtodiscussions(e.g.,Bealer1999andAlexander&Weinberg2007)oftheverynatureofphilosophicalinquiry:towhatextentarephilosophicalquestionstobeansweredbyappealstoreasonorexperience?

2.TheIntuition/DeductionThesis

TheIntuition/Deductionthesisclaimsthatwecanknowsomepropositionsbyintuitionandstillmorebydeduction.Manyempiricists(e.g.,Hume1748)havebeenwillingtoacceptthethesissolongasitisrestrictedtopropositionssolelyabouttherelationsamongourownconcepts.Wecan,theyagree,knowbyintuitionthatourconceptofGodincludesourconceptofomniscience.Justbyexaminingtheconcepts,wecanintellectuallygraspthattheoneincludestheother.Thedebatebetweenrationalistsandempiricistsisjoinedwhentheformerassert,andthelatterdeny,theIntuition/Deductionthesiswithregardtopropositionsthatcontainsubstantiveinformationabouttheexternalworld.Rationalists,suchasDescartes,haveclaimedthatwecanknowbyintuitionanddeductionthatGodexistsandcreatedtheworld,thatourmindandbodyaredistinctsubstances,andthattheanglesofatriangleequaltworightangles,wherealloftheseclaimsaretruthsaboutanexternalrealityindependentofourthought.SuchsubstantiveversionsoftheIntuition/Deductionthesisareourconcerninthissection.

OnedefenseoftheIntuition/Deductionthesisassumesthatweknowsomesubstantiveexternalworld

truths,addsananalysisofwhatknowledgerequires,andconcludesthatourknowledgemustresultfromintuitionanddeduction.Descartesclaimsthatknowledgerequirescertaintyandthatcertaintyabouttheexternalworldisbeyondwhatempiricalevidencecanprovide.Wecanneverbesureoursensoryimpressionsarenotpartofadreamoramassive,demonorchestrated,deception.Onlyintuitionanddeductioncanprovidethecertaintyneededforknowledge,and,giventhatwehavesomesubstantiveknowledgeoftheexternalworld,theIntuition/Deductionthesisistrue.AsDescartestellsus,allknowledgeiscertainandevidentcognition(1628,RuleII,p.1)andwhenwereviewalltheactionsoftheintellectbymeansofwhichweareabletoarriveataknowledgeofthingswithnofearofbeingmistaken,werecognizeonlytwo:intuitionanddeduction(1628,RuleIII,p.3).

Thislineofargumentisoneoftheleastcompellingintherationalistarsenal.First,theassumptionthatknowledgerequirescertaintycomesataheavycost,asitrulesoutsomuchofwhatwecommonlytakeourselvestoknow.Second,asmanycontemporaryrationalistsaccept,intuitionisnotalwaysasourceofcertainknowledge.Thepossibilityofadeceivergivesusareasontodoubtourintuitionsaswellasourempiricalbeliefs.Forallweknow,adeceivermightcauseustointuitfalsepropositions,justasonemightcauseustohaveperceptionsofnonexistentobjects.Descartes'sclassicwayofmeetingthischallengeintheMeditationsistoarguethatwecanknowwithcertaintythatnosuchdeceiverinterfereswithourintuitionsanddeductions.Theyareinfallible,asGodguaranteestheirtruth.Theproblem,knownastheCartesianCircle,isthatDescartes'saccountofhowwegainthisknowledgebegsthequestion,byattemptingtodeducetheconclusionthatallourintuitionsaretruefromintuitedpremises.Moreover,hisaccountdoesnottoucharemainingproblemthathehimselfnotes(1628,RuleVII,p.7):Deductionsofanyappreciablelengthrelyonourfalliblememory.

AmoreplausibleargumentfortheIntuition/Deductionthesisagainassumesthatweknowsomeparticular,externalworldtruths,andthenappealstothenatureofwhatweknow,ratherthantothenatureofknowledgeitself,toarguethatourknowledgemustresultfromintuitionanddeduction.Leibniz(1704)tellsusthefollowing.

Thesenses,althoughtheyarenecessaryforallouractualknowledge,arenotsufficienttogiveusthewholeofit,sincethesensesnevergiveanythingbutinstances,thatistosayparticularorindividualtruths.Nowalltheinstanceswhichconfirmageneraltruth,howevernumeroustheymaybe,arenotsufficienttoestablishtheuniversalnecessityofthissametruth,foritdoesnotfollowthatwhathappenedbeforewillhappeninthesamewayagain.Fromwhichitappearsthatnecessarytruths,suchaswefindinpuremathematics,andparticularlyinarithmeticandgeometry,musthaveprincipleswhoseproofdoesnotdependoninstances,norconsequentlyonthetestimonyofthesenses,althoughwithoutthesensesitwouldneverhaveoccurredtoustothinkofthem(1704,Preface,pp.150151)

Leibnizgoesontodescribeourmathematicalknowledgeasinnate,andhisargumentmaybedirectedtosupporttheInnateKnowledgethesisratherthantheIntuition/Deductionthesis.Forourpurposeshere,wecanrelateittothelatter,however:Wehavesubstantiveknowledgeabouttheexternalworldinmathematics,andwhatweknowinthatarea,weknowtobenecessarilytrue.Experiencecannotwarrantbeliefsaboutwhatisnecessarilythecase.Hence,experiencecannotbethesourceofourknowledge.Thebestexplanationofourknowledgeisthatwegainitbyintuitionanddeduction.Leibnizmentionslogic,metaphysicsandmoralsasotherareasinwhichourknowledgesimilarlyoutstripswhatexperiencecanprovide.Judgmentsinlogicandmetaphysicsinvolveformsofnecessitybeyondwhatexperiencecansupport.Judgmentsinmoralsinvolveaformofobligationorvaluethatliesbeyondexperience,whichonlyinformsusaboutwhatisthecaseratherthanaboutwhatoughttobe.

Thestrengthofthisargumentvarieswithitsexamplesofpurportedknowledge.Insofaraswefocusoncontroversialclaimsinmetaphysics,e.g.,thatGodexists,thatourmindisadistinctsubstancefromourbody,theinitialpremisethatweknowtheclaimsislessthancompelling.Takenwithregardtootherareas,however,theargumentclearlyhaslegs.Weknowagreatdealofmathematics,andwhat

weknow,weknowtobenecessarilytrue.Noneofourexperienceswarrantsabeliefinsuchnecessity,andwedonotseemtobaseourknowledgeonanyexperiences.Thewarrantthatprovidesuswithknowledgearisesfromanintellectualgraspofthepropositionswhichisclearlypartofourlearning.Similarly,weseemtohavesuchmoralknowledgeasthat,allotherthingsbeingequal,itiswrongtobreakapromiseandthatpleasureisintrinsicallygood.Noempiricallessonabouthowthingsarecanwarrantsuchknowledgeofhowtheyoughttobe.

ThisargumentfortheIntuition/Deductionthesisraisesadditionalquestionswhichrationalistsmustanswer.Insofarastheymaintainthatourknowledgeofnecessarytruthsinmathematicsorelsewherebyintuitionanddeductionissubstantiveknowledgeoftheexternalworld,theyoweusanaccountofthisformofnecessity.Manyempiricistsstandreadytoarguethatnecessityresidesinthewaywetalkaboutthings,notinthethingswetalkabout(Quine1966,p.174).Similarly,ifrationalistsclaimthatourknowledgeinmoralsisknowledgeofanobjectiveformofobligation,theyoweusanaccountofhowobjectivevaluesarepartofaworldofapparentlyvaluelessfacts.

Perhapsmostofall,rationalistdefendersoftheIntuition/Deductionthesisoweusanaccountofwhatintuitionisandhowitprovideswarrantedtruebeliefsabouttheexternalworld.Whatisittointuitapropositionandhowdoesthatactofintuitionsupportawarrantedbelief?Theirargumentpresentsintuitionanddeductionasanexplanationofassumedknowledgethatcan'ttheysaybeexplainedbyexperience,butsuchanexplanationbyintuitionanddeductionrequiresthatwehaveaclearunderstandingofintuitionandhowitsupportswarrantedbeliefs.Metaphoricalcharacterizationsofintuitionasintellectualgraspingorseeingarenotenough,andifintuitionissomeformofintellectualgrasping,itappearsthatallthatisgraspedisrelationsamongourconcepts,ratherthanfactsabouttheexternalworld.Moreover,anyintellectualfaculty,whetheritbesenseperceptionorintuition,providesuswithwarrantedbeliefsonlyifitisgenerallyreliable.Thereliabilityofsenseperceptionstemsfromthecausalconnectionbetweenhowexternalobjectsareandhowweexperiencethem.Whataccountsforthereliabilityofourintuitionsregardingtheexternalworld?Isourintuitionofaparticulartruepropositiontheoutcomeofsomecausalinteractionbetweenourselvesandsomeaspectoftheworld?Whataspect?Whatisthenatureofthiscausalinteraction?Thatthenumberthreeisprimedoesnotappeartocauseanything,letaloneourintuitionthatitisprime.

Theseissuesaremadeallthemorepressingbytheclassicempiricistresponsetotheargument.ThereplyisgenerallycreditedtoHumeandbeginswithadivisionofalltruepropositionsintotwocategories.

Alltheobjectsofhumanreasonorinquirymaynaturallybedividedintotwokinds,towit,RelationsofIdeas,andMattersofFact.OfthefirstarethesciencesofGeometry,Algebra,andArithmetic,and,inshort,everyaffirmationwhichiseitherintuitivelyordemonstrativelycertain.Thatthesquareofthehypotenuseisequaltothesquareofthetwosidesisapropositionwhichexpressesarelationbetweenthesefigures.Thatthreetimesfiveisequaltohalfofthirtyexpressesarelationbetweenthesenumbers.Propositionsofthiskindarediscoverablebythemereoperationofthought,withoutdependenceonwhatisanywhereexistentintheuniverse.Thoughthereneverwereacircleortriangleinnature,thetruthsdemonstratedbyEuclidwouldforeverretaintheircertaintyandevidence.Mattersoffact,whicharethesecondobjectsofhumanreason,arenotascertainedinthesamemanner,norisourevidenceoftheirtruth,howevergreat,ofalikenaturewiththeforegoing.Thecontraryofeverymatteroffactisstillpossible,becauseitcanneverimplyacontradictionandisconceivedbythemindwiththesamefacilityanddistinctnessasifeversoconformabletoreality.(Hume1748,SectionIV,Part1,p.40)

Intuitionanddeductioncanprovideuswithknowledgeofnecessarytruthssuchasthosefoundinmathematicsandlogic,butsuchknowledgeisnotsubstantiveknowledgeoftheexternalworld.Itisonlyknowledgeoftherelationsofourownideas.Iftherationalistshiftstheargumentsoitappealstoknowledgeinmorals,Hume'sreplyistoofferananalysisofourmoralconceptsbywhichsuch

knowledgeisempiricallygainedknowledgeofmattersoffact.

Moralsandcriticismarenotsoproperlyobjectsoftheunderstandingasoftasteandsentiment.Beauty,whethermoralornatural,isfeltmoreproperlythanperceived.Orifwereasonconcerningitandendeavortofixthestandard,weregardanewfact,towit,thegeneraltasteofmankind,orsomeotherfactwhichmaybetheobjectofreasoningandinquiry.(Hume1748,SectionXII,Part3,p.173)

Iftherationalistappealstoourknowledgeinmetaphysicstosupporttheargument,Humedeniesthatwehavesuchknowledge.

Ifwetakeinourhandanyvolumeofdivinityorschoolmetaphysics,forinstanceletusask,Doesitcontainanyabstractreasoningconcerningquantityornumber?No.Doesitcontainanyexperimentalreasoningconcerningmatteroffactandexistence?No.Commititthentotheflames,foritcancontainnothingbutsophistryandillusion.(Hume1748,SectionXII,Part3,p.173)

Anupdatedversionofthisgeneralempiricistreply,withanincreasedemphasisonlanguageandthenatureofmeaning,isgiveninthetwentiethcenturybyA.J.Ayer'sversionoflogicalpositivism.Adoptingpositivism'sverificationtheoryofmeaning,Ayerassignseverycognitivelymeaningfulsentencetooneoftwocategories:eitheritisatautology,andsotruesolelybyvirtueofthemeaningofitstermsandprovidesnosubstantiveinformationabouttheworld,oritisopentoempiricalverification.Thereis,then,noroomforknowledgeabouttheexternalworldbyintuitionordeduction.

Therecanbenoaprioriknowledgeofreality.Forthetruthsofpurereason,thepropositionswhichweknowtobevalidindependentlyofallexperience,aresoonlyinvirtueoftheirlackoffactualcontent[Bycontrast]empiricalpropositionsareoneandallhypotheseswhichmaybeconfirmedordiscreditedinactualsenseexperience.[Ayer1952,pp.869394]

Therationalists'argumentfortheIntuition/Deductionthesisgoeswrongatthestart,accordingtoempiricists,byassumingthatwecanhavesubstantiveknowledgeoftheexternalworldthatoutstripswhatexperiencecanwarrant.Wecannot.

Thisempiricistreplyfaceschallengesofitsown.Ourknowledgeofmathematicsseemstobeaboutsomethingmorethanourownconcepts.Ourknowledgeofmoraljudgmentsseemstoconcernnotjusthowwefeeloractbuthowweoughttobehave.Thegeneralprinciplesthatprovideabasisfortheempiricistview,e.g.Hume'soverallaccountofourideas,theVerificationPrincipleofMeaning,areproblematicintheirownright.Invariousformulations,theVerificationPrinciplefailsitsowntestforhavingcognitivemeaning.AcarefulanalysisofHume'sInquiry,relativetoitsownprinciples,mayrequireustoconsignlargesectionsofittotheflames.

Inall,rationalistshaveastrongargumentfortheIntuition/Deductionthesisrelativetooursubstantiveknowledgeoftheexternalworld,butitssuccessrestsonhowwelltheycananswerquestionsaboutthenatureandepistemicforceofintuitionmadeallthemorepressingbytheclassicempiricistreply.

3.TheInnateKnowledgeThesis

TheInnateKnowledgethesisjoinstheIntuition/Deductionthesisinassertingthatwehaveaprioriknowledge,butitdoesnotofferintuitionanddeductionasthesourceofthatknowledge.Ittakesouraprioriknowledgetobepartofourrationalnature.Experiencemaytriggerourawarenessofthisknowledge,butitdoesnotprovideuswithit.Theknowledgeisalreadythere.

PlatopresentsanearlyversionoftheInnateKnowledgethesisintheMenoasthedoctrineof

knowledgebyrecollection.Thedoctrineismotivatedinpartbyaparadoxthatariseswhenweattempttoexplainthenatureofinquiry.Howdowegainknowledgeofatheoremingeometry?Weinquireintothematter.Yet,knowledgebyinquiryseemsimpossible(Meno,80de).Weeitheralreadyknowthetheorematthestartofourinvestigationorwedonot.Ifwealreadyhavetheknowledge,thereisnoplaceforinquiry.Ifwelacktheknowledge,wedon'tknowwhatweareseekingandcannotrecognizeitwhenwefindit.Eitherwaywecannotgainknowledgeofthetheorembyinquiry.Yet,wedoknowsometheorems.

Thedoctrineofknowledgebyrecollectionoffersasolution.Whenweinquireintothetruthofatheorem,webothdoanddonotalreadyknowit.Wehaveknowledgeintheformofamemorygainedfromoursoul'sknowledgeofthetheorempriortoitsunionwithourbody.Welackknowledgeinthat,inoursoul'sunificationwiththebody,ithasforgottentheknowledgeandnowneedstorecollectit.Inlearningthetheorem,weare,ineffect,recallingwhatwealreadyknow.

PlatofamouslyillustratesthedoctrinewithanexchangebetweenSocratesandayoungslave,inwhichSocratesguidestheslavefromignorancetomathematicalknowledge.Theslave'sexperiences,intheformofSocrates'questionsandillustrations,aretheoccasionforhisrecollectionofwhathelearnedpreviously.Plato'smetaphysicsprovidesadditionalsupportfortheInnateKnowledgeThesis.Sinceourknowledgeisofabstract,eternalFormswhichclearlyliebeyondoursensoryexperience,itisapriori.

ContemporarysupportersofPlato'spositionarescarce.Theinitialparadox,whichPlatodescribesasatrickargument(Meno,80e),ringssophistical.Themetaphysicalassumptionsinthesolutionneedjustification.Thesolutiondoesnotanswerthebasicquestion:Justhowdidtheslave'ssoullearnthetheorem?TheIntuition/Deductionthesisoffersanequally,ifnotmore,plausibleaccountofhowtheslavegainsknowledgeapriori.Nonetheless,Plato'spositionillustratesthekindofreasoningthathascausedmanyphilosopherstoadoptsomeformoftheInnateKnowledgethesis.Weareconfidentthatweknowcertainpropositionsabouttheexternalworld,butthereseemstobenoadequateexplanationofhowwegainedthisknowledgeshortofsayingthatitisinnate.Itscontentisbeyondwhatwedirectlygaininexperience,aswellaswhatwecangainbyperformingmentaloperationsonwhatexperienceprovides.Itdoesnotseemtobebasedonanintuitionordeduction.Thatitisinnateinusappearstobethebestexplanation.

NoamChomskyarguesalongsimilarlinesinpresentingwhathedescribesasarationalistconceptionofthenatureoflanguage(1975,p.129).Chomskyarguesthattheexperiencesavailabletolanguagelearnersarefartoosparsetoaccountfortheirknowledgeoftheirlanguage.Toexplainlanguageacquisition,wemustassumethatlearnershaveaninnateknowledgeofauniversalgrammarcapturingthecommondeepstructureofnaturallanguages.ItisimportanttonotethatChomsky'slanguagelearnersdonotknowparticularpropositionsdescribingauniversalgrammar.Theyhaveasetofinnatecapacitiesordispositionswhichenableanddeterminetheirlanguagedevelopment.Chomskygivesusatheoryofinnatelearningcapacitiesorstructuresratherthanatheoryofinnateknowledge.HisviewdoesnotsupporttheInnateKnowledgethesisasrationalistshavetraditionallyunderstoodit.Asonecommentatorputsit,Chomsky'sprinciplesareinnateneitherinthesensethatweareexplicitlyawareofthem,norinthesensethatwehaveadispositiontorecognizetheirtruthasobviousunderappropriatecircumstances.AndhenceitisbynomeansclearthatChomskyiscorrectinseeinghistheoryasfollowingthetraditionalrationalistaccountoftheacquisitionofknowledge(Cottingham1984,p.124).

PeterCarruthers(1992)arguesthatwehaveinnateknowledgeoftheprinciplesoffolkpsychology.Folkpsychologyisanetworkofcommonsensegeneralizationsthatholdindependentlyofcontextorcultureandconcerntherelationshipsofmentalstatestooneanother,totheenvironmentandstatesofthebodyandtobehavior(1992,p.115).Itincludessuchbeliefsasthatpainstendtobecausedbyinjury,thatpainstendtopreventusfromconcentratingontasks,andthatperceptionsaregenerallycausedbytheappropriatestateoftheenvironment.Carruthersnotesthecomplexityoffolkpsychology,alongwithitssuccessinexplainingourbehaviorandthefactthatitsexplanationsappeal

tosuchunobservablesasbeliefs,desires,feelingsandthoughts.Hearguesthatthecomplexity,universalityanddepthoffolkpsychologicalprinciplesoutstripswhatexperiencecanprovide,especiallytoyoungchildrenwhobytheirfifthyearalreadyknowagreatmanyofthem.Thisknowledgeisalsonottheresultofintuitionordeductionfolkpsychologicalgeneralizationsarenotseentobetrueinanactofintellectualinsight.Carruthersconcludes,[Theproblem]concerningthechild'sacquisitionofpsychologicalgeneralizationscannotbesolved,unlesswesupposethatmuchoffolkpsychologyisalreadyinnate,triggeredlocallybythechild'sexperienceofitselfandothers,ratherthanlearned(1992,p.121).

Empiricists,andsomerationalists,attacktheInnateKnowledgethesisintwomainways.First,theyofferaccountsofhowsenseexperienceorintuitionanddeductionprovidetheknowledgethatisclaimedtobeinnate.Second,theydirectlycriticizetheInnateKnowledgethesisitself.TheclassicstatementofthissecondlineofattackispresentedinLocke1690.Lockeraisestheissueofjustwhatinnateknowledgeis.Particularinstancesofknowledgearesupposedtobeinourmindsaspartofourrationalmakeup,buthowaretheyinourminds?Iftheimplicationisthatweallconsciouslyhavethisknowledge,itisplainlyfalse.Propositionsoftengivenasexamplesofinnateknowledge,evensuchplausiblecandidatesastheprinciplethatthesamethingcannotbothbeandnotbe,arenotconsciouslyacceptedbychildrenandthosewithseverecognitivelimitations.Ifthepointofcallingsuchprinciplesinnateisnottoimplythattheyareorhavebeenconsciouslyacceptedbyallrationalbeings,thenitishardtoseewhatthepointis.Nopropositioncanbesaidtobeinthemind,whichitneveryetknew,whichitneveryetwasconsciousof(1690,BookI,ChapterII,Section5,p.61).Proponentsofinnateknowledgemightrespondthatsomeknowledgeisinnateinthatwehavethecapacitytohaveit.Thatclaim,whiletrue,isoflittleinterest,however.Ifthecapacityofknowing,bethenaturalimpressioncontendedfor,allthetruthsamanevercomestoknow,will,bythisaccount,beeveryoneofthem,innateandthisgreatpointwillamounttonomore,butonlyanimproperwayofspeakingwhichwhilstitpretendstoassertthecontrary,saysnothingdifferentfromthose,whodenyinnateprinciples.Fornobody,Ithink,everdenied,thatthemindwascapableofknowingseveraltruths(1690,BookI,ChapterII,Section5,p.61).LockethuschallengesdefendersoftheInnateKnowledgethesistopresentanaccountofinnateknowledgethatallowstheirpositiontobebothtrueandinteresting.Anarrowinterpretationofinnatenessfacescounterexamplesofrationalindividualswhodonotmeetitsconditions.Agenerousinterpretationimpliesthatallourknowledge,eventhatclearlyprovidedbyexperience,isinnate.

DefendersofinnateknowledgetakeupLocke'schallenge.Leibnizresponds(1704)byappealingtoanaccountofinnatenessintermsofnaturalpotentialtoavoidLocke'sdilemma.ConsiderPeterCarruthers'similarreply.

Wehavenotedthatwhileoneformofnativismclaims(somewhatimplausibly)thatknowledgeisinnateinthesenseofbeingpresentassuch(oratleastinpropositionalform)frombirth,itmightalsobemaintainedthatknowledgeisinnateinthesenseofbeinginnatelydeterminedtomakeitsappearanceatsomestageinchildhood.Thislatterthesisissurelythemostplausibleversionofnativism.(1992,p.51)

Carruthersclaimsthatourinnateknowledgeisdeterminedthroughevolutionaryselection(p.111).Evolutionhasresultedinourbeingdeterminedtoknowcertainthings(e.g.principlesoffolkpsychology)atparticularstagesofourlife,aspartofournaturaldevelopment.Experiencesprovidetheoccasionforourconsciouslybelievingtheknownpropositionsbutnotthebasisforourknowledgeofthem(p.52).CarruthersthushasareadyreplytoLocke'scounterexamplesofchildrenandcognitivelylimitedpersonswhodonotbelievepropositionsclaimedtobeinstancesofinnateknowledge.Theformerhavenotyetreachedtheproperstageofdevelopmentthelatterarepersonsinwhomnaturaldevelopmenthasbrokendown(pp.4950).

AseriousproblemfortheInnateKnowledgethesisremains,however.Weknowapropositiononlyifitistrue,webelieveitandourbeliefiswarranted.Rationalistswhoasserttheexistenceofinnateknowledgearenotjustclaimingthat,asamatterofhumanevolution,God'sdesignorsomeother

factor,ataparticularpointinourdevelopment,certainsortsofexperiencestriggerourbeliefinparticularpropositionsinawaythatdoesnotinvolveourlearningthemfromtheexperiences.Theirclaimisevenbolder:Inatleastsomeofthesecases,ourempiricallytriggered,butnotempiricallywarranted,beliefisnonethelesswarrantedandsoknown.Howcanthesebeliefsbewarrantediftheydonotgaintheirwarrantfromtheexperiencesthatcauseustohavethemorfromintuitionanddeduction?

Somerationaliststhinkthatareliabilistaccountofwarrantprovidestheanswer.AccordingtoReliabilism,beliefsarewarrantediftheyareformedbyaprocessthatgenerallyproducestruebeliefsratherthanfalseones.Thetruebeliefsthatconstituteourinnateknowledgearewarranted,then,becausetheyareformedastheresultofareliablebeliefformingprocess.CarruthersmaintainsthatInnatebeliefswillcountasknownprovidedthattheprocessthroughwhichtheycometobeinnateisareliableone(provided,thatis,thattheprocesstendstogeneratebeliefsthataretrue)(1992,p.77).Hearguesthatnaturalselectionresultsintheformationofsomebeliefsandisatruthreliableprocess.

AnappealtoReliabilism,orasimilarcausaltheoryofwarrant,maywellbethebestwayforrationaliststodeveloptheInnateKnowledgethesis.Theyhaveadifficultrowtohoe,however.First,suchaccountsofwarrantarethemselvesquitecontroversial.Second,rationalistsmustgiveanaccountofinnateknowledgethatmaintainsandexplainsthedistinctionbetweeninnateknowledgeandaposterioriknowledge,anditisnotclearthattheywillbeabletodosowithinsuchanaccountofwarrant.Supposeforthesakeofargumentthatwehaveinnateknowledgeofsomeproposition,P.WhatmakesourknowledgethatPinnate?Tosharpenthequestion,whatdifferencebetweenourknowledgethatPandaclearcaseofaposterioriknowledge,sayourknowledgethatsomethingisredbasedonourcurrentvisualexperienceofaredtable,makestheformerinnateandthelatternotinnate?Ineachcase,wehaveatrue,warrantedbelief.Ineachcase,presumably,ourbeliefgainsitswarrantfromthefactthatitmeetsaparticularcausalcondition,e.g.,itisproducedbyareliableprocess.Ineachcase,thecausalprocessisoneinwhichanexperiencecausesustobelievethepropositionathand(thatPthatsomethingisred),for,asdefendersofinnateknowledgeadmit,ourbeliefthatPistriggeredbyanexperience,asisourbeliefthatsomethingisred.TheinsightbehindtheInnateKnowledgethesisseemstobethatthedifferencebetweenourinnateandaposterioriknowledgeliesintherelationbetweenourexperienceandourbeliefineachcase.TheexperiencethatcausesourbeliefthatPdoesnotcontaintheinformationthatP,whileourvisualexperienceofaredtabledoescontaintheinformationthatsomethingisred.Yet,exactlywhatisthenatureofthiscontainmentrelationbetweenourexperiences,ontheonehand,andwhatwebelieve,ontheother,thatismissingintheonecasebutpresentintheother?Thenatureoftheexperiencebeliefrelationseemsquitesimilarineach.ThecausalrelationbetweentheexperiencethattriggersourbeliefthatPandourbeliefthatPiscontingent,asisthefactthatthebeliefformingprocessisreliable.Thesameistrueofourexperienceofaredtableandourbeliefthatsomethingisred.Thecausalrelationbetweentheexperienceandourbeliefisagaincontingent.Wemighthavebeensoconstructedthattheexperiencewedescribeasbeingappearedtoredlycausedustobelieve,notthatsomethingisred,butthatsomethingishot.Theprocessthattakesusfromtheexperincetoourbeliefisalsoonlycontingentlyreliable.Moreover,ifourexperienceofaredtablecontainstheinformationthatsomethingisred,thenthatfact,nottheexistenceofareliablebeliefformingprocessbetweenthetwo,shouldbethereasonwhytheexperiencewarrantsourbelief.ByappealingtoReliablism,orsomeothercausaltheoryofwarrant,rationalistsmayobtainawaytoexplainhowinnateknowledgecanbewarranted.Theystillneedtoshowhowtheirexplanationsupportsanaccountofthedifferencebetweeninnateknowledgeandaposterioriknowledge.

4.TheInnateConceptThesis

AccordingtotheInnateConceptthesis,someofourconceptshavenotbeengainedfromexperience.Theyareinsteadpartofourrationalmakeup,andexperiencesimplytriggersaprocessbywhichweconsciouslygraspthem.Themainconcernmotivatingtherationalistshouldbefamiliarbynow:the

contentofsomeconceptsseemstooutstripanythingwecouldhavegainedfromexperience.AnexampleofthisreasoningispresentedbyDescartesintheMeditations.Descartesclassifiesourideasasadventitious,inventedbyus,andinnate.Adventitiousideas,suchasasensationofheat,aregaineddirectlythroughsenseexperience.Ideasinventedbyus,suchasourideaofahippogriff,arecreatedbyusfromotherideaswepossess.Innateideas,suchasourideasofGod,ofextendedmatter,ofsubstanceandofaperfecttriangle,areplacedinourmindsbyGodatcreation.ConsiderDescartes'sargumentthatourconceptofGod,asaninfinitelyperfectbeing,isinnate.OurconceptofGodisnotdirectlygainedinexperience,asparticulartastes,sensationsandmentalimagesmightbe.Itscontentisbeyondwhatwecouldeverconstructbyapplyingavailablementaloperationstowhatexperiencedirectlyprovides.Fromexperience,wecangaintheconceptofabeingwithfiniteamountsofvariousperfections,one,forexample,thatisfinitelyknowledgeable,powerfulandgood.Wecannothowevermovefromtheseempiricalconceptstotheconceptofabeingofinfiniteperfection.(Imustnotthinkthat,justasmyconceptionsofrestanddarknessarearrivedatbynegatingmovementandlight,somyperceptionoftheinfiniteisarrivedatnotbymeansofatrueideabutbymerelynegatingthefinite,ThirdMeditation,p.94.)Descartessupplementsthisargumentbyanother.NotonlyisthecontentofourconceptofGodbeyondwhatexperiencecanprovide,theconceptisaprerequisiteforouremploymentoftheconceptoffiniteperfectiongainedfromexperience.(Myperceptionoftheinfinite,thatisGod,isinsomewaypriortomyperceptionofthefinite,thatismyself.ForhowcouldIunderstandthatIdoubtedordesiredthatislackedsomethingandthatIwasnotwhollyperfect,unlesstherewereinmesomeideaofamoreperfectbeingwhichenabledmetorecognizemyowndefectsbycomparison,ThirdMeditation,p.94).

AnempiricistresponsetothisgenerallineofargumentisgivenbyLocke(1690,BookI,ChapterIV,Sections125,pp.91107).First,thereistheproblemofexplainingwhatitisforsomeonetohaveaninnateconcept.Ifhavinganinnateconceptentailsconsciouslyentertainingitatpresentorinthepast,thenDescartes'spositionisopentoobviouscounterexamples.YoungchildrenandpeoplefromotherculturesdonotconsciouslyentertaintheconceptofGodandhavenotdoneso.Second,thereistheobjectionthatwehavenoneedtoappealtoinnateconceptsinthefirstplace.ContrarytoDescartes'argument,wecanexplainhowexperienceprovidesallourideas,includingthosetherationaliststaketobeinnate,andwithjustthecontentthattherationalistsattributetothem.

Leibniz(1704)offersarationalistreplytothefirstconcern.WhereLockeputsforththeimageofthemindasablanktabletonwhichexperiencewrites,Leibnizoffersustheimageofablockofmarble,theveinsofwhichdeterminewhatsculptedfiguresitwillaccept.

ThisiswhyIhavetakenasanillustrationablockofveinedmarble,ratherthanawhollyuniformblockorblanktablets,thatistosaywhatiscalledtabularasainthelanguageofthephilosophers.Forifthesoulwerelikethoseblanktablets,truthswouldbeinusinthesamewayasthefigureofHerculesisinablockofmarble,whenthemarbleiscompletelyindifferentwhetheritreceivesthisorsomeotherfigure.ButiftherewereveinsinthestonewhichmarkedoutthefigureofHerculesratherthanotherfigures,thisstonewouldbemoredeterminedthereto,andHerculeswouldbeasitwereinsomemannerinnateinit,althoughlabourwouldbeneededtouncovertheveins,andtoclearthembypolishing,andbycuttingawaywhatpreventsthemfromappearing.Itisinthiswaythatideasandtruthsareinnateinus,likenaturalinclinationsanddispositions,naturalhabitsorpotentialities,andnotlikeactivities,althoughthesepotentialitiesarealwaysaccompaniedbysomeactivitieswhichcorrespondtothem,thoughtheyareoftenimperceptible.(1704,Preface,p.153)

Leibniz'smetaphorcontainsaninsightthatLockemisses.Themindplaysaroleindeterminingthenatureofitscontents.Thispointdoesnot,however,requiretheadoptionoftheInnateConceptthesis.

RationalistshaverespondedtothesecondpartoftheempiricistattackontheInnateConceptthesistheempricists'claimthatthethesisiswithoutbasis,asallourideascanbeexplainedasderivedfromexperiencebyfocusingondifficultiesintheempiricists'attemptstogivesuchanexplanation.The

difficultiesareillustratedbyLocke'saccount.AccordingtoLocke,experienceconsistsinexternalsensationandinnerreflection.Allourideasareeithersimpleorcomplex,withtheformerbeingreceivedbyuspassivelyinsensationorreflectionandthelatterbeingbuiltbythemindfromsimplematerialsthroughvariousmentaloperations.Rightatthestart,theaccountofhowsimpleideasaregainedisopentoanobviouscounterexampleacknowledged,butthensetaside,byHumeinpresentinghisownempiricisttheory.Considerthementalimageofaparticularshadeofblue.IfLockeisright,theideaisasimpleoneandshouldbepassivelyreceivedbythemindthroughexperience.Humepointsoutotherwise.

Supposethereforeapersontohaveenjoyedhissightforthirtyyearsandtohavebecomeperfectlyacquaintedwithcolorsofallkinds,exceptoneparticularshadeofblue,forinstance,whichitneverhasbeenhisfortunetomeetwithletallthedifferentshadesofthatcolor,exceptthatsingleone,beplacedbeforehim,descendinggraduallyfromthedeepesttothelightest,itisplainthathewillperceiveablankwherethatshadeiswantingandwillbesensiblethatthereisagreaterdistanceinthatplacebetweenthecontiguouscolorsthaninanyother.NowIaskwhetheritbepossibleforhim,fromhisownimagination,tosupplythisdeficiencyandraiseuptohimselftheideaofthatparticularshade,thoughithadneverbeenconveyedtohimbyhissenses?Ibelievetherearebutfewwillbeoftheopinionthathecan(1748,SectionII,pp.2930)

Evenwhenitcomestosuchsimpleideasastheimageofaparticularshadeofblue,themindismorethanablankslateonwhichexperiencewrites.

Considertooourconceptofaparticularcolor,sayred.CriticsofLocke'saccounthavepointedouttheweaknessesinhisexplanationofhowwegainsuchaconceptbythementaloperationofabstractiononindividualcases.Foronething,itmakestheincorrectassumptionthatvariousinstancesofaparticularconceptshareacommonfeature.Carruthersputstheobjectionasfollows.

Infactproblemsariseforempiricistseveninconnectionwiththeverysimplestconcepts,suchasthoseofcolour.Foritisfalsethatallinstancesofagivencoloursharesomecommonfeature.Inwhichcasewecannotacquiretheconceptofthatcolourbyabstractingthecommonfeatureofourexperience.Thusconsidertheconceptred.Doallshadesofredhavesomethingincommon?Ifso,what?Itissurelyfalsethatindividualshadesofredconsist,asitwere,oftwodistinguishableelementsageneralrednesstogetherwithaparticularshade.Rather,rednessconsistsinacontinuousrangeofshades,eachofwhichisonlyjustdistinguishablefromitsneighbors.Acquiringtheconceptredisamatteroflearningtheextentoftherange.(1992,p.59)

Foranotherthing,Locke'saccountofconceptacquisitionfromparticularexperiencesseemscircular.

Asitstands,however,Locke'saccountofconceptacquisitionappearsviciouslycircular.Fornoticingorattendingtoacommonfeatureofvariousthingspresupposesthatyoualreadypossesstheconceptofthefeatureinquestion.(Carruthers1992,p.55)

ConsiderinthisregardLocke'saccountofhowwegainourconceptofcausation.

Inthenoticethatoursensestakeoftheconstantvicissitudeofthings,wecannotbutobserve,thatseveralparticulars,bothqualitiesandsubstancesbegintoexistandthattheyreceivethistheirexistencefromthedueapplicationandoperationofsomeotherbeing.Fromthisobservation,wegetourideasofcauseandeffect.(1690,BookII,Chapter26,Section1,pp.292293)

Wegetourconceptofcausationfromourobservationthatsomethingsreceivetheirexistencefromtheapplicationandoperationofsomeotherthings.Yet,wecannotmakethisobservationunlesswealreadyhavetheconceptofcausation.Locke'saccountofhowwegainourideaofpowerdisplaysa

similarcircularity.

Themindbeingeverydayinformed,bythesenses,ofthealterationofthosesimpleideas,itobservesinthingswithoutandtakingnoticehowonecomestoanend,andceasestobe,andanotherbeginstoexistwhichwasnotbeforereflectingalsoonwhatpasseswithinitself,andobservingaconstantchangeofitsideas,sometimesbytheimpressionofoutwardobjectsonthesenses,andsometimesbythedeterminationofitsownchoiceandconcludingfromwhatithassoconstantlyobservedtohavebeen,thatthelikechangeswillforthefuturebemadeinthesamethings,bylikeagents,andbythelikeways,considersinonethingthepossibilityofhavinganyofitssimpleideaschanged,andinanotherthepossibilityofmakingthatchangeandsocomesbythatideawhichwecallpower.(1690,ChapterXXI,Section1,pp.219220)

Wecomebytheideaofpowerthoughconsideringthepossibilityofchangesinourideasmadebyexperiencesandourownchoices.Yet,toconsiderthispossibilityofsomethingsmakingachangeinotherswemustalreadyhaveaconceptofpower.

Onewaytomeetatleastsomeofthesechallengestoanempiricistaccountoftheoriginofourconceptsistoreviseourunderstandingofthecontentofourconceptssoastobringthemmoreinlinewithwhatexperiencewillclearlyprovide.Humefamouslytakesthisapproach.BeginninginawayreminiscentofLocke,hedistinguishesbetweentwoformsofmentalcontentsorperceptions,ashecallsthem:impressionsandideas.Impressionsarethecontentsofourcurrentexperiences:oursensations,feelings,emotions,desires,andsoon.Ideasarementalcontentsderivedfromimpressions.Simpleideasarecopiesofimpressionscomplexideasarederivedfromimpressionsbycompounding,transposing,augmentingordiminishingthem.Giventhatallourideasarethusgainedfromexperience,Humeoffersusthefollowingmethodfordeterminingthecontentofanyideaandtherebythemeaningofanytermtakentoexpressit.

Whenweentertain,therefore,anysuspicionthataphilosophicaltermisemployedwithoutanymeaningoridea(asisbuttoofrequent),weneedbutinquirefromwhatimpressionisthatsupposedideaderived?Andifitbeimpossibletoassignany,thiswillconfirmoursuspicion.(1690,SectionII,p.30)

Usingthistest,Humedrawsoutoneofthemostimportantimplicationsoftheempiricists'denialoftheInnateConceptthesis.Ifexperienceisindeedthesourceofallideas,thenourexperiencesalsodeterminethecontentofourideas.Ourideasofcausation,ofsubstance,ofrightandwronghavetheircontentdeterminedbytheexperiencesthatprovidethem.Thoseexperiences,Humeargues,areunabletosupportthecontentthatmanyrationalistsandsomeempiricists,suchasLocke,attributetothecorrespondingideas.Ourinabilitytoexplainhowsomeconcepts,withthecontentstherationalistsattributetothem,aregainedfromexperienceshouldnotleadustoadopttheInnateConceptthesis.Itshouldleadustoacceptamorelimitedviewofthecontentsforthoseconcepts,andtherebyamorelimitedviewofourabilitytodescribeandunderstandtheworld.

Consider,forexample,ourideaofcausation.Descartestakesittobeinnate.Lockeoffersanapparentlycircularaccountofhowitisgainedfromexperience.Hume'sempiricistaccountseverelylimitsitscontent.Ourideaofcausationisderivedfromafeelingofexpectationrootedinourexperiencesoftheconstantconjunctionofsimilarcausesandeffects.

Itappears,then,thatthisideaofanecessaryconnectionamongeventsarisesfromanumberofsimilarinstanceswhichoccur,oftheconstantconjunctionoftheseeventsnorcanthatideaeverbesuggestedbyanyoneoftheseinstancessurveyedinallpossiblelightsandpositions.Butthereisnothinginanumberofinstances,differentfromeverysingleinstance,whichissupposedtobeexactlysimilar,exceptonlythatafterarepetitionofsimilarinstancesthemindiscarriedbyhabit,upontheappearanceofoneevent,toexpectitsusualattendantandtobelievethatitwillexist.Thisconnection,therefore,

whichwefeelinthemind,thiscustomarytransitionoftheimaginationfromoneobjecttoitsusualattendant,isthesentimentorimpressionfromwhichweformtheideaofpowerornecessaryconnection.(1748,SectionVII,Part2,p.86)

Thesourceofourideainexperiencedeterminesitscontent.

Suitablytothisexperience,therefore,wemaydefineacausetobeanobjectfollowedbyanother,andwherealltheobjects,similartothefirstarefollowedbyobjectssimilartothesecondWemay,therefore,suitablytothisexperience,formanotherdefinitionofcauseandcallitanobjectfollowedbyanother,andwhoseappearancealwaysconveysthethoughtoftheother.(1748,SectionVII,Part2,p.87)

Ourclaims,andanyknowledgewemayhave,aboutcausalconnectionsintheworldturnout,giventhelimitedcontentofourempiricallybasedconceptofcausation,tobeclaimsandknowledgeabouttheconstantconjunctionofeventsandourownfeelingsofexpectation.Thus,theinitialdisagreementbetweenrationalistsandempiricistsaboutthesourceofourideasleadstooneabouttheircontentandtherebythecontentofourdescriptionsandknowledgeoftheworld.

Likephilosophicaldebatesgenerally,therationalist/empiricistdebateultimatelyconcernsourpositionintheworld,inthiscaseourpositionasrationalinquirers.Towhatextentdoourfacultiesofreasonandexperiencesupportourattemptstoknowandunderstandoursituation?

Bibliography

WorksCited

Alexander,J.andWeinberg,J.,2007,AnalyticEpistemologyandExperimentalPhilosophy,PhilosophyCompass,2(1):5680.

Ayer,A.J.,1952,Language,TruthandLogic,NewYork:DoverPublications.Bealer,G.,1999,ATheoryoftheApriori,Nos,33:2955.Carruthers,P.,1992,HumanKnowledgeandHumanNature,Oxford:OxfordUniversityPress.Descartes,R.,1628,RulesfortheDirectionofourNativeIntelligence,inDescartes:Selected

PhilosophicalWritings,transl.JohnCottingham,RobertStoothoffandDugaldMurdoch,Cambridge:CambridgeUniversityPress,1988.

Descartes,R.,1641,Meditations,inDescartes:SelectedPhilosophicalWritings,transl.JohnCottingham,RobertStoothoffandDugaldMurdoch,Cambridge:CambridgeUniversityPress,1988.

Descartes,R.1644,PrinciplesofPhilosophy,inDescartes:SelectedPhilosophicalWritings,transl.JohnCottingham,RobertStoothoffandDugaldMurdoch,Cambridge:CambridgeUniversityPress,1988.

Hume,D.,173940,ATreatiseofHumanNature,Oxford:OxfordUniversityPress,1941.Hume,D.,1748,AnInquiryConcerningHumanUnderstanding,Indianapolis,IN:BobbsMerrill,

1955.Kant,I.,1783,ProlegomenatoAnyFutureMetaphysic,transl.JonathanBennett,EarlyModernTexts,

atwww.earlymoderntexts.com.Kenny,A.,1986,Rationalism,EmpiricismandIdealism,Oxford:OxfordUniversityPress.Leibniz,G.,c1704,NewEssaysonHumanUnderstanding,inLeinbiz:PhilosophicalWritings,ed.

G.H.R.Parkinson,transl.MaryMorrisandG.H.R.Parkinson,London:J.M.Dent&Sons,1973.Locke,J.,1690,AnEssayonHumanUnderstanding,ed.Woolhouse,Roger,London:PeguinBooks,

1997.Loeb,L.,1981,FromDescartestoHume:ContinentalMetaphysicsandtheDevelopmentofModern

Philosophy,Ithaca,NY:CornellUniversityPress.Mackie,J.L.,1977,Ethics:InventingRightandWrong,London:PenguinBooks.

Plato,Meno,transl.W.K.C.Guthrie,Plato:CollectedDialogues,editedbyEdithHamiltonandHuntingtonCairns,Princeton:PrincetonUniversityPress,1973.

Ross,W.D.,1930,TheRightandtheGood,Indianapolis,IN:HackettPublishing,1988.

RelatedWorks

Adams,R.,1975,WhereDoOurIdeasComeFrom?DescartesvsLocke,reprintedinStitchS.(ed.)InnateIdeas,Berkeley,CA:CaliforniaUniversityPress.

Aune,B.,1970,Rationalism,EmpiricismandPragmatism:AnIntroduction,NewYork:RandomHouse.

Bealer,G.andStrawson,P.F.,1992,TheIncoherenceofEmpiricism,ProceedingsoftheAristotelianSociety(SupplementaryVolume),66:99143.

Boyle,D.,2009,DescartesonInnateIdeas,London:Continum.Block,N.,1981,EssaysinPhilosophyofPsychologyII,London:Methuen,PartFour.Bonjour,L.,1998,InDefenseofPureReason,Cambridge:CambridgeUniversityPress.Casullo,A.,2003,AprioriKnowledgeandJustification,NewYork:OxfordUniversityPress.Casullo,A.,(ed.),2012,EssaysonAprioriKnowledgeandJustification,NewYork:Oxford

UniversityPress.Cottingham,J.,1984,Rationalism,London:PaladinBooks.Chomsky,N.,1975,RecentContributionstotheTheoryofInnateIdeas,reprintedinS.Stitch(ed.),

InnateIdeas,Berkeley,CA:CaliforniaUniversityPress.Chomsky,N.,1988,LanguageandProblemsofKnowledge,Cambridge,MA:MITPress.DePaul,M.andRamsey,W.(eds),1998,RethinkingIntuition:ThePsychologyofIntuitionandIts

RoleinPhilosophicalInquiry,Lanham,MD:RowmanandLittlefield.Fodor,J.,1975,TheLanguageofThought,Cambridge,MA:HarvardUniversityPress.Fodor,J.,1981,Representations,Brighton:Harvester.Kripke,S.,1980,NamingandNecessity,Oxford:Blackwell.Huemer,M.,2005,EthicalIntuitionism,NewYork,NY:PalgraveMacMillian.Quine,W.V.O.,1966,WaysofParadoxandOtherEssays,NewYork:RandomHouse.Quine,W.V.O.,1951,TwoDogmasofEmpiricism,inW.V.O.Quine,FromaLogicalPointof

View,Cambridge,MA:HarvardUniversityPress,1951.Stitch,S.,1975,InnateIdeas,Berkeley,CA:CaliforniaUniversityPress.

AcademicTools

Howtocitethisentry.PreviewthePDFversionofthisentryattheFriendsoftheSEPSociety.LookupthisentrytopicattheIndianaPhilosophyOntologyProject(InPhO).EnhancedbibliographyforthisentryatPhilPapers,withlinkstoitsdatabase.

OtherInternetResources

[Pleasecontacttheauthorwithsuggestions.]

RelatedEntries

analyticsyntheticdistinction|apriorijustificationandknowledge|Ayer,AlfredJules|Berkeley,George|concepts|Descartes,Ren|Descartes'theoryofideas|epistemology|historicalcontroversiessurroundinginnateness|Hume,David|innate/acquireddistinction|innatenessandlanguage|justification,epistemic:foundationalisttheoriesof|Kant,Immanuel|knowledge:analysis

of|Leibniz,GottfriedWilhelm|Locke,John|Plato|Quine,WillardvanOrman|reliabilism|skepticism|Spinoza,Baruch

Copyright2013byPeterMarkie

OpenaccesstotheSEPismadepossiblebyaworldwidefundinginitiative.PleaseReadHowYouCanHelpKeeptheEncyclopediaFree

TheSEPwouldliketocongratulatetheNationalEndowmentfortheHumanitiesonits50thanniversaryandexpressourindebtednessforthefivegenerousgrantsitawardedourprojectfrom1997to2007.ReaderswhohavebenefitedfromtheSEPareencouragedtoexaminetheNEHs

anniversarypageand,ifinspiredtodoso,[email protected].

TheStanfordEncyclopediaofPhilosophyiscopyright2014byTheMetaphysicsResearchLab,CenterfortheStudyofLanguageandInformation(CSLI),StanfordUniversity

LibraryofCongressCatalogData:ISSN10955054