https://ebookmass.com/product/belief-inference-and-the-self-
Instant digital products (PDF, ePub, MOBI) ready for you
Download now and discover formats that fit your needs...
Conscious Mind, Resonant Brain: How Each Brain Makes a Mind Stephen Grossberg
https://ebookmass.com/product/conscious-mind-resonant-brain-how-eachbrain-makes-a-mind-stephen-grossberg/
ebookmass.com
The Undivided Self: Aristotle and the 'Mind-Body' Problem
David Charles
https://ebookmass.com/product/the-undivided-self-aristotle-and-themind-body-problem-david-charles/
ebookmass.com
The Implicit Mind: Cognitive Architecture, the Self, and Ethics Michael Brownstein
https://ebookmass.com/product/the-implicit-mind-cognitivearchitecture-the-self-and-ethics-michael-brownstein/
ebookmass.com
Critical Concept Mastery Series: Acid-Base Disturbance Cases 1st Edition Zachary Healy https://ebookmass.com/product/critical-concept-mastery-series-acidbase-disturbance-cases-1st-edition-zachary-healy/
ebookmass.com
Respiratory Care Made Incredibly Easy (Incredibly Easy! Seriesu00ae) 2nd Edition, (Ebook PDF)
https://ebookmass.com/product/respiratory-care-made-incredibly-easyincredibly-easy-series-2nd-edition-ebook-pdf/
ebookmass.com
Sensation and Perception 10th Edition, (Ebook PDF)
https://ebookmass.com/product/sensation-and-perception-10th-editionebook-pdf/
ebookmass.com
The Unplanned Life of Josie Hale 1st Edition Stephanie Eding
https://ebookmass.com/product/the-unplanned-life-of-josie-hale-1stedition-stephanie-eding/
ebookmass.com
Site Engineering for Landscape Architects 6th Edition, (Ebook PDF)
https://ebookmass.com/product/site-engineering-for-landscapearchitects-6th-edition-ebook-pdf/
ebookmass.com
The Metaphysics of Light in the Hexaemeral Literature: From Philo of Alexandria to Gregory of Nyssa Isidoros C. Katsos
https://ebookmass.com/product/the-metaphysics-of-light-in-thehexaemeral-literature-from-philo-of-alexandria-to-gregory-of-nyssaisidoros-c-katsos/
ebookmass.com
1st Edition, (Ebook PDF)
https://ebookmass.com/product/essentials-of-marketing-research-ahands-on-orientation-1st-edition-ebook-pdf/
ebookmass.com
Belief,Inference,andthe Self-ConsciousMind Belief,Inference,andthe Self-ConsciousMind ERICMARCUS GreatClarendonStreet,Oxford,OX26DP, UnitedKingdom
OxfordUniversityPressisadepartmentoftheUniversityofOxford. ItfurtherstheUniversity’sobjectiveofexcellenceinresearch,scholarship, andeducationbypublishingworldwide.Oxfordisaregisteredtrademarkof OxfordUniversityPressintheUKandincertainothercountries
©EricMarcus2021
Themoralrightsoftheauthorhavebeenasserted FirstEditionpublishedin2021
Impression:1
Allrightsreserved.Nopartofthispublicationmaybereproduced,storedin aretrievalsystem,ortransmitted,inanyformorbyanymeans,withoutthe priorpermissioninwritingofOxfordUniversityPress,orasexpresslypermitted bylaw,bylicenceorundertermsagreedwiththeappropriatereprographics rightsorganization.Enquiriesconcerningreproductionoutsidethescopeofthe aboveshouldbesenttotheRightsDepartment,OxfordUniversityPress,atthe addressabove
Youmustnotcirculatethisworkinanyotherform andyoumustimposethissameconditiononanyacquirer
PublishedintheUnitedStatesofAmericabyOxfordUniversityPress 198MadisonAvenue,NewYork,NY10016,UnitedStatesofAmerica
BritishLibraryCataloguinginPublicationData Dataavailable
LibraryofCongressControlNumber:2021942711
ISBN978–0–19–284563–4
DOI:10.1093/oso/9780192845634.001.0001
Printedandboundby CPIGroup(UK)Ltd,Croydon,CR04YY
LinkstothirdpartywebsitesareprovidedbyOxfordingoodfaithand forinformationonly.Oxforddisclaimsanyresponsibilityforthematerials containedinanythirdpartywebsitereferencedinthiswork.
Formydaughters,LolaandBea. Acknowledgments ix
1.BeliefandJudgment9
I.BeliefandTruth11
II.JudgmentandtheLimitsofIrrationality22
III.Objections30
2.TheSelf-ConsciousnessofBelief39
I.BeliefandHonestAssertion42
II.HonestAssertionandBeliefAvowal46
III.BeliefAvowalandDoxasticSelf-Knowledge50
IV.ExplainingtheSelf-ConsciousnessofBelief55
3.MakingNonsenseofMoore’sParadox62
I.WhatisMoore’sParadox?63
II.Neo-Expressivism69
III.WhatAvowalsExpress76
4.TheChallengeforanAccountofInference84
I.TheTakingCondition85
II.TheCausalTheory86
III.DispositionstotheRescue?90
IV.Broome’sRule-FollowingAccount92
V.TheConstitutionTheory95
VI.InferenceasanEvaluative,Causation-ConstitutingAct99
5.InferencewithoutRegress104
I.TwoRegresses104
II.WhatICan’tBelieve107
III.WhatIMustBelieve109
IV.TakingasUnderstanding(andNotIntuition)112
V.FurtherIssues116
VI.Summary120
Acknowledgments Thisbookwouldhavebeenfarworseandmightneverhavebeenatallwithout thehelpofmanyfriendsandcolleagues.Forfruitfulconversations,email exchanges,and/orcommentsondrafts,thankstoDoritBar-On,MattBoyle, JasonBridges,JimConant,KerenGorodeisky,MatthiasHaase,Adrian Haddock,ArataHamawaki,JonasHeld,UlfHlobil,DavidHorst,David Hunter,AndreaKern,ChristianKietzmann,IradKimhi,NicholasKoziolek, RamNeta,AlexandraNewton,GillyNir,JohnPhillips,SebastianRödl,Guy Rohrbaugh,JohnSchwenkler,JamesShaw,WillSmall,ChrisBlake-Turner, andMarkosValaris.PeterMomtchiloffandtwoanonymousrefereesat OxfordUniversityPressalsoprovidedvaluablefeedbackandadvice.Iam gratefultoRyanSimonelliforcompilingtheindex.AndthankstoWileyfor permissiontousebitsandpiecesof “ToBelieveistoKnowYouBelieve,” dialectica 70(3):375–405,and “InferenceasConsciousnessofNecessity,” AnalyticPhilosophy 61(4):304-322.
MygreatestdebtistoLydiaMarcus,withoutwhomIwouldneverhave foundthehappinessandpeaceofmindthatenablesmetowritephilosophy. Shehasmydeepestgratitudeandloveforthis,asforeverythingelse.
Introduction Wedophilosophywhenwecannotseehowsomethingsimpleandobvious issomuchaspossible.Acaseinpoint:Wetakeourselvestospeakwith distinctiveauthorityaboutwhatwebelieveandwhywebelieveit.Inour struggletodefendthispresumption eventosaywhatexactlyitamountsto weareledvariouslytoskepticism(thepresumptioniswrong),denialism (thereisinfactnosuchpresumption),deflationism(this ‘speakingauthoritatively’ isreallyjustaho-huminstanceof__),inflationism(wejustneedto postulateanewkindof__),anddefeatism(wewillnotmakeprogresswithout helpfromoutsidephilosophy).Forswearingtheseoptions,thisbookarticulatesanddefendsanunderstandingoftherationalmindthatincorporatesthe presumptionwithoutdistortionorcompromise,puttingitatthecenterofa philosophicaltheoryofthenatureofbeliefandinference.
Beliefandinference,Icontend,areessentially self-conscious.Toa first approximation,thismeansthattoholdabeliefortomakeaninferenceisat thesametimetoknowthatonedoes.Myargumentforthisthesisexploitsthe followingstrikingfact:Itisimpossibletoholdpatentlycontradictorybeliefsin mindtogetheratonce.Itisimpossibletodoso,Iargue,becausewecannot believewhatweknowtobefalse.Thisimpossibilityisaspeciesof rational necessity,ametaphysicalnecessitywhosesourceistherationalityofthe subject.Wealso findrationalnecessityininference.Asubjectwhomakesan inferencerepresentstheconclusionaswhatmustbetrue,giventhetruthofthe premises,andthereinbelievesit.Rationalnecessityisaphenomenonthat uniquelycharacterizestherelationbetweenoneperson’sbeliefs.Andwhat explainsitisthatbeliefsandtheinferencesbywhichweacquirethemare constitutedbyaparticularkindofendorsementofthoseverystatesandacts.
Inthecourseofexploringtheconnectionbetweenself-knowledgeand assessmentsofwhatshouldbebelievedandinferred,wewillarriveatan understandingoftheunityoftherationalmind.Consider:Whatmakesapair ofbeliefspartofasingleviewoftheworld,asopposedtoshardsofseparate views?Whenoneofmycurrentlyheldbeliefsclasheswithoneofyours,we disagree.Whenoneofmycurrentlyheldbeliefscontradictsanotherofmy own,Iaminconsistent.Thedifferenceturnsonwhetherornottheclashing
beliefsbelongtoasinglemind,butwhatmakesfor asinglemind?Icontend thatwhatmakesapairofbeliefs mine ispreciselythatIhaveadistinctive firstpersonalknowledgeoftheirbeingmine.Surprisingly,thisisnotknowledgeof apreexistingunitysuppliedinsomeotherfashion,buttheverysourceofthat unity.Anexaminationofthenatureofbeliefandinference,inlightofthe phenomenonofrationalnecessity,revealshowtheunityoftherationalmind isafunctionofourknowledgeofourselvesasboundtobelievethetrue. Rationalself-consciousnessistheformofmentaltogetherness.
Inowsketchthislineofthought.
Itisoftensaidthatbeliefaimsattruth.Thissayingmeansatleastthismuch: Beliefsareassessableaccordingtowhetherornottheyaretrue.Whataccounts forthisfactaboutbelief?Itisnotenoughtopointtotheinternalstructureof belief thatitconsistsinasubject’srepresentingapropositionastrue.That wouldbetooeasy,forthisispreciselywhatthosewhoaskthequestionare tryingtounderstand.Itisoftenassumedthatasatisfactoryanswerwilltake somethinglikethefollowingform:Thecognitivesystemofabelieverregulates certaininternalstatessoastointroduce,maintainandreviseappropriately thosethatrepresentwhat is thecasewhileweedingoutthosethatrepresent what isn’t thecase;itisinvirtueofbeingsotreatedbythesystemthatthose statesare beliefs.Moregenerally,abeliefisthoughttobewhatitis astate whoseaimistruth invirtueof ‘acting’ likeone.
Acoreideaofthisbookisthatinsofarasbeliefaimsattruth,itisnot becauseoftherelationsintowhichbeliefsenter,butsimplybecauseofhowthe thinkerrepresentsthebelievedproposition.What’srequired,inotherwords, isnotaswitchfromthe first-personpointofview(howdoI,inbeinga believer,representtheproposition?)tothethird-personpointofview(how doesabeliefassuchinteractwithotherstates?)butadeeperunderstandingof theformer:Whatexactlyisittorepresentapropositionastrue,intherelevant sense?Myanswer:Itistorepresentitaswhat shouldberepresentedastrue or (asIputit)as tobebelieved.This,andnottherelationsintowhichastate enters,makesitbelief.Itsaimisinternaltotheattitudeitself.Moreprecisely, whatmakesitthecasethatIbelievepisthatItake myself tobebound(ina specifiedsense)torepresentpastrue.Thecorrectaccountofthenatureof beliefentailsthatbeliefisessentiallyknowntothebeliever.Because,in believing,onerepresentsthebelievedpropositionasto-be-believed,Ialso knowinbelievingthatIamopentocriticismifthebeliefisfalse.Thesource ofthetruth-assessabilityofbelief,Iargue,rendersthisnormativitymetaphysicallyinseparablefromdoxasticself-consciousness.
Insofarasbeliefaimsattruth,itisbadinsomesensetoholdcontradictory beliefs.Afterall,oneofanysuchpairmustbefalse.Butitisnotjust bad.In paradigmaticcases,itismetaphysically impossible toholdcontradictory beliefs.Whenbeliefsareheld(asIwillputit)inmindtogether intheabsence ofdistraction,self-deception,confusion,andthelike itisimpossiblebothto believethatpandalsotobelievethatnot-p.Itisimpossiblebecauseanyone capableofholdingabeliefatallknowsthatapropositionanditscontradictory cannotbothbetrue(evenifveryfewwouldputitthisway).Itisan understandingofthelawofnon-contradictionthatmakesitimpossible (whenitisimpossible)formetoholdcontradictorybeliefs.
Thisimpossibilityisakindof rational necessity.Byrationalnecessity, Idon’tsimplymeanaprincipleitwouldnecessarilybeirrationaltoviolate. Imeanakindmetaphysicalnecessitywhosesourceistherationalityofthe subject.Rationalnecessityis,Iargue,essentialtounderstandingthenatureof themind:tounderstandingbeliefitself,thespecialcharacteroftherational bondsthatlinkourbeliefstooneanother,and,moregenerally,thesourceof themind’sunity.
Consideragainthequestionsraisedabove:whatconfersonaworldviewthe unityofbeing someone ’ s?Howcanweexplainthedifferencebetweeninconsistencyanddisagreement?Whatmakesonemind one?Inwhatdoestheunity ofamindconsist?Whatisthecharacterofthedoxasticbondsthataccountfor thetogethernessofthebeliefsofasinglesubject?
Inordertoanswerthesequestions,aswellastofullyunderstandthenature ofbeliefitself,Ifocusoninferenceinthebook’ssecondhalf.Itselusive characterwasbroughtintoreliefoveracenturyagobyLewisCarroll.Inhis famoustale,thetortoisedrawsAchilles’ attentiontoapairofobviouslytrue premisesfromwhichaconclusionobviouslylogicallyfollows.Thetortoise challengesAchillesto “force” him “logically” toaccepttheconclusion.Achilles findsthathecannotdoit,sincenomatterhowmanybeliefshecoaxesthe tortoiseintoaccepting,nomatterhowobvioustheseextrapremisesrenderthe validityoftherelevantinference,itremainspossiblethatthetortoisefailsto drawtheconclusion.ManythinkthatCarroll’sParadoxcanbeavoidedsimply byrefrainingfromcastingrulesofinferenceintheroleofpremises.Butthatis notso.Theproblemisthatwhatmovessomeonetoactuallydrawthe conclusioncannot,itseems,bethemerepossessionofcertainbeliefs whethertheyarepremises,backgroundbeliefs,orwhathaveyou butrather thefactthatthosebeliefsarefortuitouslyensnaredinacausalprocessthat operatesoutsidethescopeofherbeliefs.Nooneeverreallydrawsaconclusion inlightofthefactthatthepremisessupporttheconclusion;thisisjust
somethingthathappens to thethinkerwhenthepossessionofcertainbeliefs triggersaconvenientlysituatedmechanism.Inference,asweordinarilypresumeittotranspire,canthusseemimpossible.
Thetheoryofbeliefoutlinedinthe firsthalfofthebookhelpsusescape Carroll’sParadox.Justasthe(rationallygroundedmetaphysical)necessityof rejectingabelief(say,~p)canbeduemerelytothefactthatsomeonewillnot surrendertheirbeliefthatp,the(rationallygroundedmetaphysical)necessity ofacceptingabelief(say,q)canbeduemerelytothefactthatsomeonewillnot surrendertheirbeliefsthatpandthatpentailsq.Nothingmoreisrequired, psychologicallyorrationally,toreject~pthanthebeliefthatp.Similarly, nothingmoreisrequired,psychologicallyorrationally,toacceptthebeliefthat qthanthebeliefsthatpandthatpentailsq.IbelieveqbecauseI must where, again,thisisrationallygroundedmetaphysicalnecessity believeit,given whatelseIbelieve.
Inferenceisconsciousnessofnecessity.Itisatonceconsciousnessofthe necessityofthetruthofaconclusion,giventhepremises,andconsciousnessof thatconclusionaswhatImustbelieve,giventhatIbelievethepremises. Whereastobelieveistorepresentapropositionasto-be-believed,toinferis torepresenttheto-be-believed-nessoftheconclusionasto-be-inferredfrom theto-be-believed-nessofthepremises.Because,ininferring,Irepresentthe inferenceitselfasgood,IalsoknowininferringthatIamopentocriticismif theconclusiondoesnotinfactfollowfromthepremises.Thesourceofthe validity-assessabilityofinference,Iargue,rendersthisnormativityinseparable frominferentialself-consciousness.
Thataperson’sbeliefsconstrainoneanother notjustnormativelyand causally,butmetaphysically helpstodrivehomeanimportantlessonabout thenatureofbelief.Beliefisatermweusetodivide,withpotentiallymisleadingartificiality,ouroverallviewoftheworldintoproposition-sizedpieces. Theriskisthatwecometoviewbeliefsaspossessingakindofparticularity a constitutiveindependencefromotherbeliefs thattheylack.Andinsodoing, weareunabletorecaptureinourphilosophyofmindthetruecharacterofthe mind’sunity.
Myanswertothequestionofthesourceofthemind’sunityrunsroughlyas follows: My beliefs,unlikethebeliefsofothers,affectwhatelseIdoordonot believesimplyinmyunderstandingtheirimplicationsforwhatelseIshouldor shouldnotbelieve.Inparadigmaticcases,themerefactthatoneunderstandsa propositiontofollowfrom(ortobeinconsistentwith)othersthatone believes,andwhichonewillnotsurrender,therebymakesthatproposition metaphysicallynecessary(orimpossible)tobelieve.Theunityoftherational
mindiswhatconstitutesthiscausal–normativenexus.Andwhatconstitutesit is,Iargue,self-consciousness.
Theveryideaofaninferentialrequirementcontainswithinittheideaofa unifiedsubjectwhoderivestheconclusionfromthepremises.Itisbecause S believespthat S (andnotnecessarilyanyonewhoisnotS)mustbelieveq.And inthinkingofsomeoneasinferring i.e.,inthinkingofsomeoneasrepresentingtheto-be-believed-nessoftheconclusionasto-be-inferredfromthetobe-believed-nessofthepremises Iemploythatveryconcept.Iunderstand thegoodnessoftheinferenceasjustifyingasinglesubjectwhobelievesthe premisesinacceptingtheconclusion.Andwhen I maketheinference, IunderstandthisStobe me,i.e.,Igrasptheunityfromthe ‘inside’ . Iunderstandmyself,quabelieverofp,asalsoboundtobelieveq.This recognitioniswhatmakesitthecasethatI do believeqbecauseIbelieve p.AnditiswhatexplainsthefactthatIknowthatitisbecausepthatIbelieve q,andknowthisnotonthebasisofobservationorinference,butsimply becauseIbelieveqontheinferentialbasisofp.Thetogethernessofourbeliefs is,inthissense,constitutedbyconsciousnessofthetogetherness;i.e.,consciousnessofmybeingthesubjectofbothoftheseobligationsiswhatmakesit thecasethatIconformtothem.Consciousnessoftheunityofthemindisnot anotherstate;itistheformofmentaltogetherness.Theunityofthemindis constitutedbyself-consciousness.
Theforegoingsummaryconsistschieflyofaseriesofsuggestiveclaimsthat will(Ihope)conveythegeneralshapeofthecentrallineofthoughtthatruns throughthebook.Thebooktofollowarguesindetailfortheseclaims, considersalternativetheories,andaddsmanyelementsthatIhaveomitted inthissketch.
Thebookisorganizedasfollows:
Idevelop,inchapterone,anaccountofbeliefbywayofdiscussingthe following fiveinterrelatedtruths:(i)tobelievethatpistorepresentpastrue; (ii)truthisastandardbywhichbeliefsareassessed;(iii)acertainsenseofthe question ‘Why?’ hasapplicationtobeliefs;(iv)beliefsaregovernedbythelaw ofnon-contradiction;and(v)abeliefisthesortofthingonecanbringto mind.Overthecourseofrefining,qualifying,anddefendingthesetheses, Iarriveattheviewthattobelieveistorepresentapropositionaswhatone shouldrepresentastrue.Beliefsthatare inmind (whichIcall judgments)are suchthattheyaregovernedmetaphysically,andnotjustnormatively,bythe lawofnon-contradiction.Onecannotclearlyanddistinctlyatoncerepresent astruebothapropositionanditscontrary.Irespondtoavarietyofobjections
tothisposition,including atlength theobjectionthattheaccountisunderminedbythebeliefsofnonhumananimalsandhumaninfants.
Thesecondchapterpresentsanargument,initiallywithoutrelyingonthe accountdevelopedinchapterone,forthefollowingthesis:Tobelievethatpis toknowthatonebelievesthatpandtoknowitsimplybybelievingit.Thebulk oftheargumentisadefenseofthreepremisesthatentailtheinseparabilityof belieffromknowledgeofbelief:(i)SbelievesthatponlyifSisabletohonestly assertthatp;(ii)SisabletohonestlyassertthatponlyifSisabletoavowthe beliefthatp;and(iii)SisabletoavowthebeliefthatponlyifSknowsthatshe believesthatp.Idefendeachpremisefromobjections,payingspecialattention tothephenomenonofself-deception.Ishowthatonecanholdboththatbelief isessentiallyself-conscious and thatpeoplearesometimeswrongaboutwhat theybelieve.Ultimately,Iusethetheoryofbeliefdevelopedinchapteroneto accountfordoxasticself-consciousness.
Ishow,inchapterthree,howtheviewonofferresolvesMoore’sParadox, furtherdevelopingthethemesandthesesdiscussedinpreviouschapters. Mooreanabsurditydoesnot,asitisgenerallyheld,involveaninter-level conflictbetween first-andsecond-orderbeliefs(orthecorrespondingutterances);itisjustaconflictbetweendoxasticstancesonthequestionofasingle proposition’struth.Moore-Paradoxicalstatementsare,inthisway,similarto statementsoftheform “pandnot-p”.Whatmakesthetwosortsofstatements seem fundamentallydifferentistheapparentnecessityofconstruing “Idon’t believethatp” and “Ibelievethatnot-p”,astheyoccurinMooreanutterances, asexpressiveofdoxasticattitudestowardone’sowndoxasticstates,rather thantowardtherelevantp.Thatis,whatmakesMooreanabsurditypuzzlingly puzzlingisthatwhereasanutteranceof “ p ” expressesthespeaker’sstanceon thetruthofp,thecorrespondingutteranceof “Ibelievethatp” wouldseemto expressthespeaker’sstanceonthelogicallyunrelatedpropositionthatshe believesthatp.Butthisisamistake.ThecorrectexplanationofMoorean absurdityisthatbothassertionandexplicitbelief-avowalmanifestprecisely thesamething:self-consciousknowledgeof(first-order)belief.Butonewillbe blindtothissolutionsolongasonefailstoappreciatethecharacterofdoxastic self-knowledge.
Thebook’ssecondhalfboresdownonthetopicofinference.Atouchstone ofrecentdiscussionsofinferenceisTheTakingCondition:Ininferringqfrom p,onedrawstheconclusioninlightofone’srepresentingpassupportingq. Thisconditionhighlightstwoelementsofinference.First,ininferringqfrom p,one drawsaconclusion.Itistherebybroughtaboutthatthesubjectbelieves thatq.Thisbringing-aboutsuggests causation inthebroadestsense.Second,
ininferringqfromp,thisbringing-abouthappensinlightofthesubject’ s endorsingthethought:psupportsq.Inferencethusjoinstwoelements:one causal,theotherevaluative.Ishowinthischapterthatextantaccountsfailto properlyjointhesetwoelements.Theyfailbecausethetwoelementsarenot, asisgenerallyassumed,separable.Inferenceistheendorsementofathought thatthegroundsupportsthegroundedand thisendorsementamountstoa causalconnectionbetweenthecorrespondingbeliefs.Thedifficultyisinseeing howathoughtcouldnotmerelyregisterthepresenceof,butactuallyconstitutethecausalrelationbetweenbeliefs.
Onewhoinfersrepresentspremisesassupportingaconclusioninamanner thatconstitutesacausalconnectionbetweenthepremise-beliefsandthe conclusion-belief.Thetrickforanaccountofinferenceistosayhowthis schemacanbesatisfied.Idothatinchapter five,usingCarroll’staleasa lodestar.Inferringisamatterofunderstandingtherelevantpropositionswell enoughtorecognizethatitisimpossibleforpremisesthatoneacceptstobe trueandtheconclusiontobefalse,sothatoneseestheconclusionaswhat mustbetrue,therebybelievingit.(Thiswillraisetheworry,towhichIrespond atlength,thattheaccountcoversonly deductive inference.)Mycentral formulationofthisthesiswillultimatelyextend,bothsubstantivelyand terminologically,theaccountofbeliefexpoundedinearlierchapters.To believeapropositionistorepresentitas to-be-believed;ininferringqfrom p,onerepresentstheto-be-believed-nessofqas to-be-inferred fromthetobe-believed-nessofp.Justasonebelievesthatpinseeingpaswhatoneought tobelieve,someonebelievesthatqfortheinferentialreasonthatpinseeingq aswhatoneoughttobelieveinlightoftheto-be-believed-nessofp.
Inferenceisaself-consciousact.Iestablishthis,inchaptersix,usingan argumentpreciselyanalogoustotheonedefendedinchaptertwoinconjunctionwithbelief.Herearethepremises:(i)Sbelievesthatqforthereasonthatp onlyifSisablehonestlytogivepasherreasonforbelievingq;(ii)Sisable honestlytogivepasareasonforbelievingthatqonlyifSisableto ‘justsay’ thatshebelievesqinlightofp,i.e.,onlyifsheisabletoavowthereason;and (iii)Sisableto ‘justsay’ thatshebelievesqinlightofponlyifSknowsthatpis herreasonforbelievingthatq.Itfollowsthatoneinfersonlyifoneknowsone does.Ithenshowhowtheself-consciousnessofinferenceresolvesaninferentialversionofMoore’sParadox.
IaccountforinferenceintermsofwhatIlabelthe ‘being-in-mind-together’ ofbeliefs.Beliefsthatareinmindtogetheraresubjecttorationallygrounded metaphysicalnecessity.Becauseonebelievesp,onecan’tormustbelieveq.Up untilthispointinthebook,Ihavesimplyreliedonthereadertoacknowledge
therealityofthisphenomenon.Tomakefurtherprogress,wemustgobeyond suggestivelabeling.Ifitisimpossibletoholdapairofmentalstatesinmindat onceandtheimpossibilityhasitssourceinourunderstandingofthenecessary falsehoodofaconjunction,thenthesubjecthasknowledgenotjustofthe individualstatesthey’reinbutalsooftheircombination.What,then,isthe relationbetweentheunityofourbeliefsandconsciousnessofthisunity?My answer:Theunityoftherationalmindconsistsinthesubject’sconsciousness ofthatunity.
Myapproachtoeachoftheseissuesisshowntobeplausibleinitsownright. But,astheforegoingoutlinewillhavemadeplain,thebooktofollowisnota collectionofdiscretesolutionstoputativelyseparateproblems.RatherItreat alloftheproblemsasinterrelated,andthusmysolutionisaunifiedconceptionofthemindinitstheoreticalorientation.Myhopeisthattheunityofthe analysiswilllendadditionalcredencetomytreatmentofwhataregenerally takentobeseparateissues.
BeliefandJudgment Theterm ‘belief ’ livesamessylifeoutsidephilosophy,andtheprecisionofa philosophicalaccountcancreatetheimpressionthatsomethingimportantis beingleftout.Theriskofmisunderstandingherewillbereduced,Ihope,by declaringfromthestartthatmygoalisnottogiveanaccountthatcaptures everyacceptable,literal,andtrueoccurrenceoftheterm.Iaimtounderstand whatwearegenerallysayinginordinarycontextsaboutthecognitivelytypical, adulthumanstowhomweapplyit.
Here,presentedintheordertobediscussed,areseveninterrelatedtruths aboutbelief,sounderstood,thatwillorientmydiscussioninthisandlater chapters:
(i)Tobelievethatpistorepresentpastrue.
(ii)Truthisastandardbywhichbeliefsareassessed.
(iii)Acertainsenseofthequestion ‘why?’ hasapplicationtobeliefs.
(iv)Beliefsaregovernedbythelawofnon-contradiction.
(v)Abeliefisthesortofthingthatonecanbringtomind.
(vi)Oneknowsone’sownbeliefsbetteranddifferentlythananyone elsecan.
(vii)Sometimespeoplearewrongaboutwhattheybelieve.
Whilethesestatementshavethesyntaxofuniversalgeneralizations,theyare notintendedtomakesubstantivegeneralclaimsaboutanindependently specifiablekindofstate.Rather,theyare topic-fixing.Theseprinciplesare meanttofocusthereaderonwhatIwilldiscuss.Tobesure,eachneedstobe clarified,refined,and/orqualified.Inthecourseofdoingso,Idevelopatheory ofbeliefthatbestexplainstheirjointtruth.(Thecomplementarytheoryof inferencedefendedinthesecondhalfofthebookwillextendtheexplanatory gainsofits firsthalfbyarticulatinghowcertainmentalactsmeet ‘thetaking condition,’ anddosowithoutengenderingregress.)
Itwillbeimportantinwhatfollowsnottolosesightofthefactthatmy accountofwhatIcallbelief thatis,thoseentitiesdescribedbythetheory thatexplains(i)–(vii) iscompatiblewiththeexistenceofdistinctyet
unproblematicusesof ‘belief,’ usesonwhichitreferstosomethingelse.For example,insofarascognitivepsychologists,ethologists,andpetownersusethe term ‘belief ’ torefertostatesthatare not beliefsinmysense e.g.,tosubpersonalstatesthatcarryinformationfromoneplaceinourbodiestoanother, toperson-levelstatesthatguidetheintelligentbehaviorofrationalcreatures butdosowithoutthepossibilityoftheirbecomingconscious,totheconscious cognitionsof(non-human)animals thisdoesnotthreatenmyargumentsin anyway.Manydifferentkindsofthingsarecalled “belief,” and(i)–(vii)don’t allholdtrueofallofthem.Butmyinterestisexclusivelyinthekindofstateof which(i)–(vii)arealltrue.Ofcourse,ifwehadreasontobelievethatthereis nokindofstateofwhich(i)–(vii)arealltrue,thenthiswouldunderminemy project.Butthereisnosuchreason.Relatedly:Cognitivesciencetellsusabout manydifferentkindsofmovement-guidingrepresentationsinthebrain,but itdoesn’ttelluswhetherthoserepresentationsexplain(i)–(vii)orevenif (i)–(vii)aretrueofthem.Oncethenatureofbeliefs(inmysense)isbetter understood,wewillthenbeinapositiontoassesstherelationbetweenthem andthesortofstatesthatcognitivescientistsdiscussunderthesamelabel.But Idonotundertakethatinquiryhere.
Beforebeginning,somediscussionofthe kind ofaccounttobegivenisin order.Latetwentieth-centuryaccountsofbelieftendedtobereductively naturalistic.Philosophersofmindattemptedtoexplainbeliefexhaustivelyin behavioral,functional,dispositional,and/orteleo-functionalterms.These approachesliveon,butanotherfamilyofviewshasjoinedthem.Sincethe turnofthecentury,manyhavebeguntofavoraccountsofbeliefintermsofits aim,specificallyitsaimingattruth,orperhapssomethingelse.¹Thisapproach getsatacrucialnormativefeatureofbelief:Afalsebeliefisdeficient.Some (thoughnotall)ofthisworktoohasareductivecharacter:Itisthecognitive system,ratherthantheperson,thatregulatescertaininternalstatessoasto introduce,maintain,andreviseappropriatelythetrue,whileweedingoutthe false;anditisinvirtueofbeingsotreatedbythesystemthatthosestatesare beliefsandsoassessableaccordingtothenormoftruth.Accordingtoother membersofthisfamily,thebelieverherselfhastheaimofbelievingtruly,and itisinvirtueof her regulatingcertainofherstatesaccordingtothisaimthat thosestatescountasbelief,andsoareassessableaccordingtothenormof truth.Bothstrategiessharetheassumptionthatbelief ’sconnectiontotruth isamatterofhowastateistreatedandisinthatsenseexternaltothebeliefs themselves.Thatis,itisastate’sbeingregulatedeitherpersonallyor
¹BernardWilliamsintroduces ‘aim’-talktotheliteratureonbeliefinWilliams(1973).
sub-personallyinacertainmannerthataccountsforitsaimingattruthand, therefore,foritsbeingsubjecttothetruth-norm.Itisbecauseofhowbelief ‘behaves’ (i.e.,itsexternalrelations)andnotbecauseofhowthebeliever representstheproposition(itsinternalstructure),thatbeliefisassessable accordingtothetruth-norm.
ThegistofthetheoryIwilldefendinwhatfollowsisthattobelieveisto representapropositionaswhatoneshouldrepresentastrue.Itispreciselyin recognizingthisdoxasticobligationthatoneconformstoit.Thus,insofaras beliefaimsattruth,itisnotbecauseofhowthestateisregulated,butbecause ofhowthethinkerrepresentsthebelievedproposition.
I.BeliefandTruth NishiShahandDavidVellemanarguethat
(i)tobelievethatpistorepresentpastrue.
doesnotsingleoutbelief.Supposingp,assumingp,andimaginingpalso couldbedescribedasinvolvingrepresentingpastrue.²Thisshowsnotso muchthatthereis more tobelievingpthanrepresentingitastrue,butthatthe relevantsenseof ‘representing-as-true’ needstobefurtherspecified.Tothat end,notethatsupposing,assuming,andimaginingareallaction-types,inthe sensethatitispossibletosuppose,assume,orimagine atwill.Thesensein whichbelievingthatpisrepresentingpastrueisoneinwhichtherelevantsort of ‘representing’ isnotsomethingthatcanbedoneatwill.³
Butthisclarificationwillstillnotenableustoidentifytherelevantattitude, asthereareothernon-voluntaryattitudestowardpthatmightbedescribedas formsofrepresentingpastrue.If,forexample,Ihaveavisceralfearof cockroaches,onemightdescribemeasfeelingandactingasifitweretrue thatcockroachesweredangerous,andindoingsoasrepresentingtheproposition ‘cockroachesaredangerous’ astrue.Nonetheless,itmightbethatIdo notbelievethatcockroachesaredangerous notevenalittlebit.Perhapswe canruleoutthesenon-voluntaryrepresentings-as-truebyfollowingthrough ontheobservationthatbeliefisthesortofattitudethatoneholdsonthebasis ofreasons,i.e.,therearereasons for believingandnotjustreasons why people
²ShahandDavidVelleman(2005).
³Therelatedissueofpragmaticreasonsforbeliefwillbediscussedbelow.
believe.EvenifthereisasenseinwhichIrepresent ‘cockroachesaredangerous ’ astrue,itis(wecanimagine)notanattitudeIhold for anyreason.
Isitenoughtosingleouttherelevantsortofrepresenting-as-truetosaythat itisatoncenon-voluntaryandalsoheldforreasons?No.ConsiderthatifIam sadthatpIamsadthatpistrue.Inbeingsadthatp,Itherebyinsomesense representpastrue.Icannotadoptthisattitudeatwill,butIamsadforreasons, e.g.,IamsadthatIbrokethedishforthisreason:Itwasafamilyheirloom. Moregenerally,theemotionsseemtobemodesofrepresentingpastruethat arebothnon-voluntaryand ‘held’ forreasons.
Truthisnot,however, theobject oftheemotions.Ifoneissadthatp,one mustbelieve(know,even)thatp.Butifpturnsouttobefalse,themistakeis notinthesadnessbutinthebelief.Whetherthesadnessitselfismistakendoes notdependonwhetherit’safactthatp,butonlyonwhetherp(factornot) warrantssadness.Ifafriendissadthatherapplicationtobecomeamemberof theRussianRouletteteamhasbeenturneddown,sheissadaboutsomething thatsheshouldnotbesadabout.Themistakeisinthesadness.If,ontheother hand,sheissadbecausetheworldisgoingtoendtomorrow orsoshe believes themistakeisinthebelief.Thetruthofp,then,issomethinglike abackgroundconditionratherthantheobjectofsadnessaboutp.
Butwhatdoesitmeantosaythattruthistheobjectofbelief?Hereismybasic thought.Tobelievethatpisnotsimplytotreatpastrueinthoughtandaction asperhapsonemightifoneweresupposingitorifonewereinthegripofan irrationalfear buttorepresentsotreatingitascorrect.Ipropose,then,thatthe followingprinciplecanexplaintheothersixtruthsintroducedabove:
(i)*tobelievethatpistorepresentpaswhatoneshouldrepresentastrue.
Toseehowitcandothis,i*itselfmustbeclarified.Respondingtoapairof objectionswillbringitintofocus.
First,itmightseemopentoobviouscounterexamples.Ifsomeonemadean offerofabilliondollarstoanyonewhorepresentedtheproposition ‘thesunis madeofbutter’ astrue,Imightrepresentitaswhatoneshouldrepresentas true,therebysatisfyingtheletterofi*.ButIdonotbelieveit.Onecanholdthat acertainpropositionshouldberepresentedastruewithoutbelievingit.
Second,itlooksasifbelievingthatpisbeingidentifiedwithanapparently differentbelief:thebeliefthatpiswhatoneoughttobelieve.Thiswouldbe problematiconitsface,astheseseemtobedifferentcontentsandalsowould appeartogenerateaviciousregress.Believingthatoneoughttobelievep wouldthenhavetobeidentifiedwithbelievingthatoneoughttobelievethat oneoughttobelievethatp,andsoforth.
Thebutterexampleshowsthati*ishopelessif ‘shouldrepresentastrue’ coverspracticalimperatives.Plainly,onedoesnotbelievepinvirtueof representing holdingptobetrue asbeneficial.Replyingtothisobjection requiresspecifyingadifferentsenseof ‘should.’
Tothatend,considertheoreticaldeliberation,i.e.,deliberationonthe questionofwhetheracertainpropositionistrue.Inanordinarycase,one considerswhatonehasseenwithone’sowneyes,whatonehasbeentold, relevantmemories,anyevidencethatbearsonthequestion,etc.To findaset ofconsiderationssufficientlydemonstrativeofthetruthofpistorecognizep aswhatone should believe,andtotherebybelieveit atleastintheparadigmaticcase(moreonthisbelow).Moreprecisely,partofwhatitisforthe drawingofone’sconclusiontomanifestrationalresponsivenesstosupporting considerationsisforthatresponsivenesstoincorporateanunderstandingof thoseconsiderationsasconducivetothetruthofone’sconclusion.Themotion frompremisetoconclusionintheoreticaldeliberationisthusfueledpartlybya graspofthisparticularsortofnormativeconnectionbetweenpropositions.
ThisideaismoreorlessequivalenttowhatPaulBoghossiancalls “TheTaking Condition” oninference.⁴ Iwilldiscussitatlengthinlaterchaptersandsayin moredetailwhatformtheincorporationofthis “understandingofsupporting considerations” musttake.Butifoneacceptsthatthetheoreticaldeliberatormust possesssomesuchunderstanding,thenonemustthinkofherasviewingherown believingthatpasresponsivetothevariousconsiderationsthatconstituteher reasonsforbelievingthatp(supposingshehasany).Shethinksthat,inlightof herreasons,pis tobebelieved.Tothinkofitas ‘tobebelieved’ intherelevant sensehasnothingtodowiththinkingofthebenefitsofholdingthebeliefas outweighingthecosts.Ithastodoonlywiththetruth-conducivenessofthe relevantconsiderations.Thisindicatesthesenseof ‘should’ that figuresin(1*). Myproposalthenisthattobelievethatp evenwhennosuchdeliberationhas occurred issimplytorepresent believingthatp aswhatoneshoulddo,inthe just-sketchedsense.Toavoidanyfurther confusion,Iwillrelyintheremainder ofthebookonthefollowingformulation:
(i)**Tobelievethatpistorepresentpasto-be-believed.⁵
⁴ Boghossian(2014).Cf.,Lavin(2011).
⁵ IthusgomuchfartherthanShah,who(inShah2003)arguesthatdeliberationaboutwhetherto believepisamatterofemployingourknowledgethatbeliefsaredeficientiftheyarefalse.Forme(and notforShah)beliefis itself anexerciseoftheconceptofbeliefand(consequently)anapplicationofour knowledgeofitsnormativity.ForShah,beliefsarebodilystatesthatarecausallyregulatedbytruthtrackingmechanisms.Thosecreatureswhopossesstheconceptofbelief andnotallbelievers,on Shah’sview,dopossesstheconcept caninfluencetheirbeliefsbytriggeringamechanismthatprotects beliefsfromtheinfluenceofnon-evidentialbelief-formingdispositions.Thusforme(butnotShah),
Wearestillnotdonewiththe firstobjection,however.Onemight,itseems, representapropositionasto-be-believed(evenwiththeforegoingclarification)withoutbelievingit.Forexample,theremightbeacertainproposition thatonesimplycan’ttakeseriously say,thatone’sspouseisanundercover Russianspy.TheCIAhas,suppose,requestedameetingonthesubjectandfor prudentialreasons,onehasdecidedtotakeit.Theevidenceisoverwhelming. Attheendofthesession,onemaygenuinelystillnotbelievethatsheisaspy, butoneisinastateexpressiblebysayingsuchthingsas “Ioughttobelievethat mywifeisaspy,butIstillcan’tquitebringmyselftobelieveit.” Suchacase appearstoshowthatrepresentingapropositionasbeingsufficientlywell supportedtowarrantbeliefisonething,actuallybelievingitisanother. (Notethatthiscounterexampledoesnotrelyonapragmatic ‘should.’)
Onegetsadifferentimpressionfromthefollowingexample.Supposea womanisplayingpoker,tryingto figureoutwhetherheropponentisbluffing. Sheconsidershisbehavioronpriorroundsofbetting,whichshe finds inconclusive.Hopingtogetareadonhim,sheaskshimwhetherhe’llshow herhishandifshefolds.Heenthusiasticallysayshewill,therebyrevealingthat hewantshertofold,asignthathishandisweak.Thisevidencesettlesthe matterforher he’sbluffing.Herethereisnotemptationtobreakdownthe transitionbetweendeliberationandbeliefintotwosteps: first,astepinwhich sherecognizesthatsheoughttobelievehe’sbluffing;andsecond,astepin whichsheactuallycomestobelievethathe’sbluffing.Shesimplyrecognizes whattheevidenceshowsandthereinbelievesit.
Furthermore,thecounterintuitiveideathatthereis necessarily agap betweenanormativejudgmentofbelief-worthinessandabeliefhasan unacceptableconsequence.Onewouldthenhavetosaythatarationalexplanationofwhyapersonbelievespexplainsdirectlyonlywhyshejudgesthat sheoughttobelievep.Anexplanationofwhysheactuallybelievesitwould alsohavetomentionthattheought-judgmentledtothebelief.Youaskwhy shebelieveshe’sbluffingandshesays “becausehesaidhe’dshowmehis hand.” Youreply “that’swhyyouthinkyou ought tobelievehe’sbluffing,but whydoyouactuallybelieveit?” Thepokerprowouldbeutterlyconfused.Her originalanswerseemstoleavenothingofthissorttoexplain.Onthetwo-step view,therewouldbequestionsabouthowpreciselytheought-judgmentgives risetothebelief,whetheritisanautomaticprocessorwhethermoreactivity fromthethinkerisrequired.Ifitisautomatic,onemightaskhowlongittakes
believingthatprequiresconceivingofyourattitudetowardsthepropositionthatpasgovernedbythe normoftruth.ForShah,thisisonlyarequirementfordeliberationaboutp.
andwhethertheprocesscanbespeduporsloweddown,whetheritmightstill beaccomplishedwhilethesubjectwasasleepordrunk.Andwouldreasoning onthebasisofpsubsequenttotherecognitionthattheevidencedecisively favorspbutpriortothebelief ‘takingroot’ somehowhavetobequalified (unbeknownsttothereasoner)asmerelyconditionalreasoning?Ifmore activityfromthethinkerisrequired,thenwemustaskwhatsortofactivity thismightbe ifnotmoredeliberation,thenwhat?Absentfealtytothetwosteppicture,thesequestionswouldbedismissedasabsurd.
Here’sadifferentangleonthesameworry.Belief,onewouldthink,isthe conclusionoftheoreticalreasoning:the finalstepinthedeliberationofa rationalcreature.Thepracticalanalogueofthisthesis thatactionisthe conclusionofpracticalreasoning iscontroversialbecauseanaction,such asmywalkingtothestore,seemslikethewrongsortofthingtobeanelement ofreasoning.⁶ Butaccordingtothetwo-steppicture,rationalactivityendswith thejudgmentthatoneoughttobelievep,onestepshortofactuallybelieving thatp.Buthowdoesonegofrombelievingthatpistobebelievedtobelieving pitself,supposingthatthesearedistinct?Notonthebasisofmoreevidence. Afterall,theevidenceinfavorofbelievingpispreciselytheevidencethatone oughttobelievep.Sothetransitionfromthelattertotheformer(iftheremust besuchatransition)cannotbeunderstoodasthedirectexpressionofthe subject’srationalactivity.Itwouldhavetobeaprocessof ‘sinkingin’ thatone hopestakesplacesubsequenttoone’srecognitionthatpdeservestobe believed.Onecouldlabelthisprocess ‘rational’ ornot,butitwouldamountto viewingthemakingupofone’smindasoutsourced,atthe finalstage,to unconsciousorsubpersonalmechanismsthatonecouldatbestcheeronas theydidordidn’tchurnoutthebeliefatwhichonehopedtoarrive.Betterto rejectthegap:representingpaswhatoneoughttobelievejust is believingthatp.
Notethatthedifficultythatthegapposesforthetwo-steppicturehas nothingtodowiththeimmediacyoftheallegedtransitionbetweenbelieving thatpoughttobebelievedandsimplybelievingthatp.Andsoitisnouse respondingthat,insomecases,thelattermightbetheimmediateeffectofthe former.(Andwhy,onemightwonder,istheformeralessproblematic stoppingpointfortheoreticalreasoningthanthelatter?)Thepointisthat nomatterhowimmediatethetransition,thebeliefthatponthetwo-stepview issomethingexternaltotheagent’sdoxasticreasoningaboutp,forsuch
⁶ Iwouldandhaveargued,however,thatinfactaction is (asAristotleheld)theconclusionof practicalreasoning.Toxjustistorepresentx-ingastobedone.SeeMarcus(2012),ch.2andMarcus (2018).
