WhatMattersin
PersonalIdentityandOtherPossibilities
DOUGLASEHRING
GreatClarendonStreet,Oxford,OX26DP, UnitedKingdom
OxfordUniversityPressisadepartmentoftheUniversityofOxford. ItfurtherstheUniversity’sobjectiveofexcellenceinresearch,scholarship, andeducationbypublishingworldwide.Oxfordisaregisteredtrademarkof OxfordUniversityPressintheUKandincertainothercountries
©DouglasEhring2021
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:2020952856
ISBN978–0–19–289471–7
DOI:10.1093/oso/9780192894717.001.0001
PrintedandboundintheUKby TJBooksLimited
LinkstothirdpartywebsitesareprovidedbyOxfordingoodfaithand forinformationonly.Oxforddisclaimsanyresponsibilityforthematerials containedinanythirdpartywebsitereferencedinthiswork.
Tomyamazingwife,Ann,andmywonderfuldaughters,KatandBella.
Acknowledgments
Myinterestinthetopicofpersonalidentitygoesbacktoametaphysicsclass ItookfromBernardBerofskyingraduateschoolatColumbiaUniversity.It wasinthatclassthatI firstreadDerekParfit.Asaconsequence,Parfit becameoneofmyphilosophicalheroesandhasremainedsoforthesemany years.MydebttoParfitwillbeobvioustothereaderthroughoutthiswork. Ihavealsobenefitedfromtheconversationsandcommentsofmycolleagues inthePhilosophyDepartmentatSouthernMethodistUniversityinDallas, Texas.IamespeciallygratefultoRobertHowellandBradThompson,who readandcommentedontheentiremanuscript.Iamalsogratefulto SouthernMethodistUniversityforanumberofresearchleavesthatmade possiblemuchoftheresearchthatwentintothisbook.Iamalsoquite thankfulfortheextensiveandinsightfulcommentsfromtherefereesat OxfordUniversityPress.IwouldalsoliketothankRebeccaMarinforher verydetailededitorialworkontheentiremanuscript.Mywife,Ann,and children,KatandBella,alsodeservetobethankedfortheirsupportand encouragement.
Thisbookincorporatesmaterialfromseveralpublishedarticles,listed below.Permissionoftheoriginalpublishers(OxfordUniversityPress,John WileyandSons,Taylor&Francis,andSpringer)isgratefullyacknowledged.
“SurvivalandTrivialFacts.” Analysis 47(1987):50–4
“PersonalIdentityandTimeTravel.” PhilosophicalStudies 52(1987): 427–33
“PersonalIdentityandthe R-Relation.” TheAustralasianJournalof Philosophy 73(1995):337–46
“Fission,FusionandtheParfitRevolution.” PhilosophicalStudies 94 (1999):329–32
“WhyParfitDidNotGoFarEnough.” PhilosophicalStudies 165(2013): 133–49
“WhyParfitCannotGeneralizefromFission. ” AnalyticPhilosophy 59 (2018):413–25
“JohanssononFission.” ActaAnalytica 34(2019):155–63
“WhyParfitCanRebutJohnston’ s Reductio. ” Theoria 86(2020):583‒594
“FissionandAnticipatingHavinganExperience.” Synthese (2020)https:// doi.org/10.1007/s11229-020-02860-4
5.TheTrivialityArgument122
6.TheNon-TrivialityPrincipleandObjectionstoIts
Introduction
Whatrelationtoafutureindividual,ifany,givesmeareasontohave prudential(“special”)concernforthatfutureindividualand,intheabsence ofthatrelation,Ihavenoreasontohaveprudentialconcernforthat individual?Callthisthe “FuturePersonQuestion.” Thecommonsense answeristhatIhavesuchareasonjustincaseIamidenticaltothatperson. Thisanswercanbealternatelyphrasedastheclaimthatidentityis “what mattersinsurvival” ifthelatteristakentomean “reasontohaveprudential concernwithrespecttoafutureindividual.” Thereisalsotherelated question,theFutureTimeQuestion,towhichcommonsensegivesaclosely relatedanswer:whatrelationtoafuturetime,ifany,givesmeareasonto haveprudential(“special”)concernforthattimeand,intheabsenceofthat relation,Ihavenoreasontohaveprudentialconcernforthattime?¹The commonsenseansweristhatIhaveareasontodirectmyprudentialconcern tothattimejustincaseIamidenticalwithsomeoneatthattime.
Mighttheseanswersbewrong?Thereare many relationsbetweenme nowandmeinthefuture,andpersonalidentitycannotconsistinallof them,sothereareotherpossibleanswers.²Coulditbethatoneofthese relations notpersonalidentity givesmeareasontodirectmyprudential concerntothefutureme?Or,perhaps,thereisnorelationthatgivesmea reasonforfuture-directedprudentialconcern.Thesequestionscanbeclarifiedbybrieflycharacterizingthenotionsof “prudentialconcern” and “ reason ” atissue.
Thespecialconcernthatonenormallydirectstowardsoneselfdiffersin kindfromtheconcernonenormallydirectstowardsothers.
Thereisadifferenceinhowyouwouldrespondtobeingtoldthatsomeone otherthanyourselfwillbetorturednextweekandhowyouwouldrespond tobeingtoldthat you willbetorturednextweek.Inbothinstances,youwill
¹Forthedistinctionbetweenthesequestions,see(Johansson2010:31). ²Thelateryoucanremembersomeoftheexperiencesoftheearlieryou.Theyhavethesame brain.Theyhavethesamebody.Theyhavesomeofthesamehairs.And,soon.
WhatMattersinSurvival:PersonalIdentityandOtherPossibilities.DouglasEhring,OxfordUniversityPress(2021). ©DouglasEhring.DOI:10.1093/oso/9780192894717.003.0001
2
beconcerned.However,thekindofconcernyouwouldhaveinthesecond instancewouldbedifferentinnaturethaninthe firstinstance.
Whatarethedistinctivefeaturesofprudentialconcern?First,ourspecial concernisinstinctive.Onbeingtoldthatyouwillbeinpain,yourconcernis immediate,entirelynatural,and,mostprobably,selectedforbyevolution (Johnston1992:599).Inaddition,self-concernisnotvoluntary.Weseemto be “boundtobespeciallyconcernedaboutourselves” (Parfit2007:23).
Third,ourself-concernisgenerallystrongerthanother-directedconcern, butnotnecessarilyso. “IfIknowthatmychildwillbeinpain,Imaycare aboutthischild’spainmorethanIwouldaboutmyownfuturepain” (Parfit 2007:80).Fourth,ourspecialconcernforourselvesisnotderivative;itdoes notderivefromsomeotherconcern,desireorinterest(Parfit2007:20–1). Forexample,wedon’tderiveourself-concernfromourconcernforothers orfromourinterestinthecompletionofourvariousprojects.Fifth,this kindofconcerniscloselyassociatedwiththeabilitytoanticipatehavingthe experiencesofthefuturepersontowhomitisdirected. “ThoughImaycare moreaboutmychild’spain,Icannot,itseems,fearfully anticipate thatpain” (Parfit2007:22).³
By “ reason ” Iwillmeannormativereason,notmotivatingreason. Anormativereasonissomethingthatcountsinfavoroftakingacertain actionorhavingacertainattitude.Ontheotherhand,amotivatingreasonis somethingthatexplainswhysomeonetookacertainaction.
Ifyouranawayfromtheangrysnake,yourmotivatingreasonwouldbe providedbyyourfalsebeliefthatthisactwouldsaveyourlife.But...you havenonormativereasontorunaway.(Parfit2011,Vol.1:37)
Giventhatstandingstillisthebestreactiontoanangrysnake,your motivatingreasonisnotanormativereason,andyournormativereason (“stayingstillissafest”)isnotamotivatingreason.Iwillalsoassumean “externalist,” ratherthanan “internalist,” conceptionofnormativereasons. Onthelatterconception,aconsiderationisareasonforanagenttotakea certainactionorhaveacertainattitudeonlyifcertainmotivationalfacts
³Parfitthinksthatprudentialconcernshouldnotbeexplicitlydefinedasarelationbetween apersonatonetimeandthatsamepersonatanothertime.Rather,itshouldbecharacterizedas the kind ofconcernwehavetowardsourselvesinthefuturesoastoleaveopenthepossibility thatwemightdirectthisconcerntowardssomeonewithwhomwearenotidentical(Parfit 2007:20).
holdofthatagent.Forexample,asimpleinternalistviewmightholdthatwe haveareasontodoonlythosethingsthatwouldfulfillsomedesirethatwe nowhave.Moresophisticatedformsofinternalismappealnottoone’ s actualpresentdesires,buttocertainhypothetical,non-instrumentaldesires (aimsorchoices) desiresthatwemaynotcurrentlyhave,but would have hadnowhadcertaincircumstancesobtained,suchashavingnofalsebeliefs, havingfullinformation,orhavingdeliberatedrationally.Fora “ reasons externalist,” somefactmaygiveanagentareason,evenintheabsenceof acorrespondingactualdesireoraimandevenintheabsenceofcorrespondinghypotheticaldesiresoraimsthathewouldhavenowhadcertainconditionsobtained.Forexample,itisarguablethatIhaveareasonnowto avoidfuturepain,evenifIdonotnowcareaboutmyfuturepainandevenif IwouldnotcareaboutmyfuturepainwereInowtohavenofalsebeliefs, havefullinformation,orhavedeliberatedrationallywithfullinformation (Parfit2011,Vol.1:73–82).Assuming “reasonsexternalism,” asIdo,our questionsareaboutwhat,ifany,externalreasonstherearefordirecting one ’sdistinctiveegoisticconcerntowardsafutureindividualora futuretime.
Accordingtocommonsense,identityandonlyidentitygivesoneareason fordirectingone’sprudentialconcerntowardsafutureindividual. ⁴
IdentityAlwaysMatters (1)Foranysituation s,if P₁ at t₁ hasareason r to directhisprudentialconcernto P₂ at t₂,thenthefactthat P₁ isidenticalto P₂ gives P₁ thatreason.(2)Foranysituation s,if P₁ at t₁ isidenticalto P₂ at t₂, thenthefactthat P₁ at t₁ isidenticalto P₂ at t₂ gives P₁ at t₁ areasontodirect hisprudentialconcernto P₂ at t₂. ⁵
Given(1),foranysituation s,if P₁ att₁ isnotidenticalto P₂ att₂,then P₁ at t₁ hasnoreasontodirecthisprudentialconcernto P₂ at t₂. ⁶
⁴ Theremightbereasonsthatarenotdecisive,sufficient,orrequiredinsomedomains. Suppose,forexample,thatthereisasetoffacts say, xyz suchthat xy isadecisivereasontodo w andasis xz andasis yz,butthateachof x, y and z isbyitselfnotadecisiveorsufficientreason todo w.Eachfact x, y,and z ineachpairisrequiredforthatpairtobedecisiveorsufficient, butnosub-fact,say, x,isarequiredreasonfordoing w.If x undertheseconditionsisareasonto do w,areasonfordoing w mayfailtobearequired,sufficient,ordecisivereasonfordoing w.
⁵ Iwillsometimessaythatcertainfactsarereasonsforusbutthisshouldbetakentomean thatcertainfactsgiveusreasons.
⁶ Thatsomethinggivesoneareason,say,todo x doesnotmeanitgivesoneadecisivereason todo x.Forexample,onemighthaveareasontodo x, buttheremaybeotherreasonsagainst doing x suchthatonedoesnothavemostreasontodo x.Inaddition,areason r fordoing x may besufficientbutnotdecisive: r mightbeastrongenoughreasontopermitdoing x,butthereisa
But,accordingtoParfit,wearewrongtofollowthethinkingofcommon senseinthisdomain.Inparticular,commonsenseiswrongatleastwith respecttothescopeoftheclaimthatidentitymattersinsurvival.Thereare atleast some casesofsurvivalinwhichwhatmattersisnotidentity.
Fission ThehemispheresofMr.Fissionyareeachtransplantedintoanew, brainlessbody,adifferentbodyforeachhemisphere.Eachofthepostfissionpeople,LeftyandRighty,ispsychologicallycontinuous/connected tothepre-fissionperson.
ByParfit’saccounting,itisnottruethatthe fissionerisidenticaltoeither fissionee.However,the fissionerdoesgetwhatmattersinsurvivalwith respecttoeach fissionee,sohemustdosobywayofarelationotherthan identity.Mr.FissionyhasareasontodirecthisspecialconcerntoLeftyand Rightythatisnotidentity.⁷
ForParfit,(1) fissionisacaseofsurvival thereisasenseinwhichMr. FissionysurvivesasLeftyandasRighty and(2)Mr.Fissionyhasareason todirecthisprudentialconcerntoLeftyandRighty.⁸ Withrespectto(1), althoughwenormallyassociate “survival” withidentity,since fissionisnot reasontodosomethingotherthan x thatmaybeequallystrong(Parfit2011,Vol.1:32–3).We might addtothis “IdentityAlwaysMatters” formulationthatidentityalwaysgivesonea decisive reasonforprudentialconcern outweighinganyotherreasons,nomatterthesituation,which favorone’snotdirectingprudentialconcerntooneselfinthefuture.Iwillleaveitopenasto whetherornotcommonsensegoesthisfar.Inparticular,Iwillleaveitopenthatcommonsense mightallowforexceptions;forexample,ifone’sfutureissohorrific,commonsensemightallow thatonemighthaveaverystrongreasonnottodirectone’sprudentialconcerntoone’sfuture selfthatisnotoutweighed.Whetherweaddinan “alwaysdecisive” clausewillnotbecrucialto ourdiscussion.
⁷ Moreprecisely,Parfitthinksthathisclaimthatidentitydoesnotmatterin fissioncanbe demonstratedoneithertheassumptionthatitistruethatthe fissionerisnotidenticaltoeither fissionee,orthatitisindeterminatethatthe fissionerisidenticaltoeither fissionee.
(1)Myrelationtoeachoftheresultingpeoplewouldcontainwhatmatters.(2)Itis nottruethatthisrelationwouldbeidentity.Either(A)itisnottruethatIwouldbe eitherofthesepeople,or(B)itistruethatIwouldbeneitherofthem.Therefore(3) Identityisnotwhatmatters.Premise(2)couldbedefendedintwoways.Wemight claimthatthereisnotrueanswertothequestion ‘WouldIbeeitheroftheresulting people?’.Thatwouldsupport(2)(A).Orwemightclaimthatit’sdeterminatelytrue thatIwouldbeneitherofthesepeople.Thatwouldsupport(2)(B)(Parfit1993:26).
⁸ Belzerpointsoutthatthisview fitsbetterwithParfit(1971)thanwithParfit(1984).Inthe former,Parfitallowsthat “Iwillsurvive” doesnotentail “therewillbesomeonealivewhoisthe samepersonasme, ” butinthelatterParfit “ argues not thatsurvivalshouldbepriedapartfrom identity,butthat whatmattersinsurvival shouldbepriedapartfrombothsurvival and identity. Onecanhavewhatmattersinsurvival,andonewouldhaveitina fissioncase,eventhoughone wouldnotsurvive” (Belzer2005:137).Iwillassumetheformerunderstandingofsurvival.
equivalenttodeath,wehaveagoodreasonforwideningthenotionof “survival” toinclude fission.Accordingto(2),Mr.Fissionyhasprudential concernforLeftyandRighty,andhehasagoodreasonforsuchconcern evenifitturnsoutthatitisnottruethatMr.Fissionyisidenticaltoeither. The fissioner’ s “relationtoeachofthesepeoplewouldcontainwhatmatters” (Parfit1993:25).⁹
Thus,in some casesofsurvival,identitydoesnotmatter.Somethingelse does,ifanythingdoes.ButParfitgoesevenfurther.Heclaimsthatidentity nevermatters(forshort,IDM),rejectingeventheintermediateposition accordingtowhichidentitysometimesmattersandsometimesdoesnot.¹⁰
IdentityNeverMatters(IDM
)(1)Foranysituation s,if P₁ at t₁ hasareason r todirecthisprudentialconcernto P₂ at t₂,thenthefactthat P₁ isidentical to P₂ doesnotgive P₁ thatreason.(2)Foranysituation s,if P₁ at t₁ is identicalto P₂ at t₂,thenthefactthat P₁ at t₁ isidenticalto P₂ at t₂ doesnot give P₁ at t₁ areasontodirecthisprudentialconcernto P₂ at t₂.
ItfollowsfromIDMthatthefactthat P₁ at t₁ isidenticalto P₂ at t₂ doesnot give P₁ at t₁ arequired,suf ficient,ordecisivereasontodirecthisprudential concernto P₂ at t₂ sincethatfactnevergivesoneareasonforprudential concern.
Tobeclear,Parfitacceptsthatwenormally think thatwhatmattersis identity,andadmitsthatifourdiscussionof “whatmattersinsurvival” were about “whatwethinkmatters,” then “identity” wouldbethecorrectanswer. Butthatisnotthequestion.Thequestioniswhatweshouldthinkmattersin survival:
By “whatmatters” Ididn’tmean...whatwetaketomatter.Itcannotbe arguedthat,inthissense,identityisnotwhatmatters.Imeantwhatwehave reason tocareaboutorshouldbelievetomatter.(Parfit2007:84,n5)
⁹ Parfitdoesnotruleoutthepossibilitythatweneverhaveareasonforprudential concern includingin fission ifreductionismaboutpersonalidentityistrue(the “Extreme Claim”).However,hetendstosupportthe “ModerateView” thatreductionismiscompatible withtherebeingsomethingthatmattersinsurvival.
¹⁰ Accordingtoanintermediateview,(1)forsomesituation s, inwhich P₁ at t₁ hasareason r todirecthisprudentialconcernto P₂ at t₂,thefactthat P₁ isidenticalto P₂ doesnotgive P₁ that reason,and(2)forsomesituation s,inwhich P₁ at t₁ isidenticalto P₂ at t₂,thefactthat P₁ at t₁ is identicalto P₂ at t₂ doesgive P₁ at t₁ areasontodirecthisprudentialconcernto P₂ at t₂
AlonglistofphilosophershasrespondedtoIDM.Somephilosophersclaim thatIDMgoestoofarandothersthatitgoesjustfarenough.Inthiswork, IclaimthatIDMdoesnotgofarenough.Iclaimthatthereisabigger mistakemadebycommonsense:themistakeinthinkingthatthereisevera relationbetweenaperson P₁ at t₁ and P₂ at t₂ thatgives P₁ at t₁ areasonto haveprudentialconcernfor P₂ at t₂.Sowhereasmanyphilosophersargue thatcommonsenseisrightthatthereis somerelation thatgivesoneareason forprudentialconcern,whichmayormaynotbeidentity,Iclaimthatitis wrongtothinkthereissucharelation.
Notonlydoespersonalidentitynevermatterinsurvival,ordinaryor otherwise,nothingmattersinsurvival(SurvivalNihilism).
Thereisnoalternativetoidentityforwhatmattersinsurvival.Itisfalsethat thereissomerelation X thatprovidesareasonforprudentialconcern.This isamoreradicalnegativethesisthanPar fittendstoputforward.¹¹
InordertoclarifySurvivalNihilism,itwillbeusefultocompareittowhat ParfitcallstheExtremeView.TheExtremeViewcanbereadashavinga positiveandanegativecomponent.¹²Thepositivecomponentisthatthereis somerelation X thatincludestheidentityovertimeof P₁’sCartesianegoor soul(a “separatelyexistingentity”)suchthatif X heldbetween P₁ and P₂, P₁ wouldhaveareasonforprudentialconcernfor P₂.Thenegativecomponent isthatthereisnorelation X thatdoesnotincludetheidentityovertimeof P₁’sCartesianegoorsoulsuchthatif X heldbetween P₁ and P₂, P₁ would haveareasonforprudentialconcernfor P₂.So,forexample,ifpersonal identityisreducibletorelationsofpsychologicalcontinuityorconnectedness,relationsofphysicalcontinuityorconnectedness,orrelationsinvolvingsomecombinationofthetwo,thatdoes not includetheidentityofa Cartesianegoorsoul,personalidentitydoesnotprovideareasonfor prudentialconcern.¹³Furthermore,thereisnorelationthatisanalternative
¹¹Asindicated,Ilargelyassumethat “reasonsexternalism” istrueinmydefenseofSurvival Nihilism.Theassumptionimpliesthatthisdefense,ifsuccessful,showsthatnotevenwitha permissive viewofreasons(onethat,unlike “reasonsinternalism,” doesnotsaythatcertain motivationalstatesarenecessaryforthepresenceofreasons)dowehavereasonsforprudential concern.
¹²SometimestheExtremeViewisdescribedonlyintermsofthenegativecomponent.
¹³Onareductionisttheoryof persons,apersonconsistsinabrainandabodyandthevarious physicalandpsychologicalmomentaryeventsassociatedwiththatbrainandbodyandvarious relationsbetweenthoseevents(Parfit1984:211).Onareductionisttheoryofpersonalidentity, personalidentityiswhollyreducibletorelationsofpsychologicalcontinuityorconnectedness, relationsofphysicalcontinuityorconnectedness,orrelationsinvolvingsomecombinationof
topersonalidentitythatdoesnotincludetheidentityovertimeof P₁’s Cartesianegoorsoul,whichprovidesareasonfor P₁ todirecthisprudential to P₂.Inparticular,therearenorelationsofpsychologicalcontinuityor connectedness,relationsofphysicalcontinuityorconnectedness,orrelationsinvolvingsomecombinationofthetwo,whetherornottheyare realizedinsuchawayastoconstitutepersonalidentity(onareductionist theoryofpersonalidentity),thatprovidebythemselvesareasonforprudentialconcern.Mymainthesisincludesthenegativecomponentofthe ExtremeView,butIrejectthepositivecomponentoftheExtremeView.My viewisthe MoreExtremeView.¹⁴
SomephilosophersdefendthenegativecomponentoftheExtremeView onthebasisoftheclaimthattherelataoftherelevantpsychological continuityandconnectednessrelations(andtherelevantphysicalcontinuity andconnectednessrelations)arethemselvesmomentaryentities(e.g., momentaryepisodesoftheconsciousnessortemporalpartsofpersons). Forexample,ifpersonshavetemporalpartsandtherelevantcontinuityand connectednessrelationsholdbetweentemporalparts,itisarguedthatthere canbenoreasonfortheprudentialconcerndirectedfromonetemporalpart toanyfuturetemporalpartthatiswhollydistinct.Incontrast,Itrytogetto thenegativecomponentoftheExtremeViewbyaroutethatfocusesonthe candidatesforthe “mattering” relation,notonthepossiblymomentary natureoftherelataofthatrelation.
Thenegativecomponentincludes,butisnotexhaustedby,IDMona reductionistaccountofpersonalidentity.Iwillworkmywaytothenegative componentoftheExtremeViewbyfocusingonIDM.Inparticular,Ifocus ontwolinesofargumentforIDM.The firstfailsasanargumentforIDM, whilethesecondsucceedsinnotonlydemonstratingIDMbutalso,more broadly,thenegativecomponentoftheExtremeView.
The firstargumentforIDMbeginswith fissioncases.Itisargued, first, thatidentitydoesnotmatterin fission(IDMF)and,second,fromIDMF (alongwithother “divergent” casesifthereareany)wecangeneralizetoall casesofsurvival.InattemptingtodemonstrateIDMF,therearethree thetwo.Personalidentityis not afurtherfactaboutaCartesianegoorimmortalsouloverand abovefactsaboutsuchphysicalandpsychologicalrelations.Nordoespersonalidentityeven partiallyinvolveafurtherfactaboutCartesianegosorsouls.Accordingtoanon-reductionist account,personalidentitydoesconsistinorincludeafurtherfactofthissort.
¹⁴ AccordingtotheModerateView,thereissomerelationofpsychologicalcontinuityand connectednessthatcangiveoneareasonforprudentialconcernwhetherornotthoserelations arerealizedinsuchawayastoconstituteidentity.
possibleapproaches.Oneapproachistodeterminethecorrectwayof mappingidentityontothe fissioncase,andthenarguethatthismapping showswhatmattersin fissionisnotidentity.Anotherapproachistoargue thatallmappingsofidentityonto fissionleadtoIDMF.Thethirdistoargue thatsomeinterpretationsareincorrectandtheremainingonesleadto IDMF.IwillfollowthesecondapproachtoIDMF.Par fitmainlytakesthe firstapproachbutshowssomeinclinationtothethirdapproachin(1976). InadoptingthissecondapproachtoIDMF,IwillstructureChapters1 through3aroundthefollowingsixwaysofconstruing fissionintermsof identity.(EachwillberecastintermsofPerdurantisminthenextchapter.)
1.Mr.FissionyisidenticaltoLeftyandMr.Fissionyisidenticalto Righty.(Chapter1)
2.Mr.FissionyisidenticaltoneitherLeftynorRighty.(Chapter1)
3.Mr.FissionyisidenticaltothefusionofLeftyandRighty.(Chapter1)
4.Mr.FissionyisidenticaltoonebutnottheotherofLeftyandRighty. (Chapter1)
5.LeftyandRightywhollyorpartiallyoverlappre- fission.(Chapter2)
6.ItisindeterminatethatMr.FissionyisidenticaltoLeftyanditis indeterminatethatMr.FissionyisidenticaltoRighty.(Chapter3)
Iwillarguethat,fromeachoftheseoptions,itfollowsthatidentitydoesnot matterin fission.Alongtheway,Itrytoshowthatsomeoftheseoptionsare moreplausiblethantheyinitiallyappeartobe.However,sinceallthese optionsleadtothesameresult identitydoesnotmatterin fission thereis noneedtodeterminewhichoneiscorrect.
Nonetheless,IargueinChapter4thatwecannotgeneralizetoIDMfrom IDMF (alongwithother “divergent” cases)withoutadoubtfuladditional premise,thepremisethatthereisnosuchthingas “matteroverdetermination,” accordingtowhichbothidentityandsomethingotherthanidentity haveindependentnon-derivativeimportancesufficientforwhatmattersin survival.Soevenifidentitydoesnotmatterin fissionandsomethingelse does,thatleavesopenthepossibilitythatbothidentityandthissomething elsematterinordinarysurvival.
ThesecondargumentforIDMisbasedontheNon-TrivialityPrinciple. Theideaencodedinthisprincipleisthattheimportantcannotdependon thetrivial.Iwillcallthissortofargument, “thetrivialityargument. ” Onecan findinParfitatrivialityargumentforIDMconsistingintheclaim thatanyadequatereductionistaccountofpersonalidentitywillviolatethe
Non-TrivialityPrinciplebecausesuchanaccountwillincludeanonbranchingconditionthatwillleadtosuchviolations.InChapter5, IdevelopatrivialityargumentforIDMthatisquitedifferentinnature bothinstructureandoutcome.Ifocusonthecausalnatureofidentitygiven reductionismandonthecausalnatureofothernon-identitycandidatesfor whatmatters.Myargumentdemonstrates,ifitworks,morethanIDM:for anyX,eitheridentity,givenreductionism,orsomenon-soul-based,nonCartesian-ego-basednon-identitycandidateforwhatmattersinsurvival, X iseitherhighlyimplausibleaswhatmattersordoesnotsatisfytheNonTrivialityPrinciple.
If X iswhatmattersinsurvival,then X cannotfailtoholdbecauseofatrivial differenceinthefacts.Butanyplausiblecandidatefor X notinvolvingsouls orCartesianegosmustincludeacausalcomponentand,becauseofthat component,willnotmeetthiscondition.
Inshort,thenegativecomponentoftheExtremeViewcanbedemonstrated bywayofthecausalcharacterofplausiblenon-Cartesian-egoandnon-soulbasedcandidatesfortherelationthatmattersinsurvival.Furthermore,one shouldnotinferthatmyargumenthastheimplicationthatthereisnocausal componentinwhatmattersinsurvivalifreductionismistrue.Iarguethat anti-causalistaccountsofwhatmattersfail.
AsforthepositivecomponentoftheExtremeView,Iargue against itby suggestingthattheenduranceofaCartesianegoorsoulalonewouldnot provideareasonforprudentialconcern.Norwouldtheabsenceofan enduringCartesianegoorsoulmeanthatonehasnoreasonforprudential concern.CombiningthepersistenceofaCartesianegoorsoulwithpsychologicalorphysicalcontinuityorconnectednessalsodoesnotprovidea reasonforprudentialconcern,forreasonshavingnothingtodowith Cartesianegosorsouls.Hence,onmyview,nothing,includingpersonal identity,mattersinsurvival.Specialconcernisalwayswithoutreason.Ithus rejectwhatmightbecalled MatterRealism ,theviewthatthereissomething thatmattersinordinarysurvival Nothingmattersinsurvivalinthatsense. ThisleavesopenthatImighthaveareasontodirectsomeotherkindof concerntoalaterindividual,perhapsaltruisticconcernorimpartial concern.
TheDivergenceArgument
AccordingtoParfit’ s “DivergenceArgument,” itcandemonstratedthat identitydoesnotmatterin fission(IDMF)andfromIDMF (alongwith other “divergent” casesifthereareany)wecangeneralizetoallcasesof survival(Parfit2007).IwillfocusonIDMF inthischapterthrough Chapter3.IwillconsiderthegeneralizinginferenceinChapter4.
TherearethreewaysonemightestablishIDMF:
(1)Determinethecorrectwayofconstruing fissionintermsofidentity andarguethatthisconstrualshowsthatwhatmattersin fissionisnot identity.
(2)Arguethatallmappingsofidentityonto fissionleadtoIDMF.
(3)Determinethatsomemappingsareincorrectandthattheremaining mappings,whichmayormaynotbeincorrect,leadtoIDMF.
Parfitlargelymakesuseofthe firstapproachtoIDMF butnotconsistently (Parfit1971;1984;1993;2007).¹Iwilladoptthesecondapproach.
1.1Fission
Considerthefollowing fissioncase:
ThehemispheresofMr.Fissionyareeachtransplantedintoanewbrainless body,adifferentbodyforeachhemisphere.Aftertheoperation,eachofthe post-fissionpeople,LeftyandRighty,ispsychologicallycontinuous/connectedtothepre-fissionperson,Mr.Fissiony.
LeftyandRightycaneach “remember” theexperiencesofMr.Fissiony,and theyhavethebeliefs,plans,andcharactertraitsofMr.Fissiony.
¹ThereissomeevidencethatParfitdoesnotalwaysfollowthe firstapproach(Parfit1976).
WhatMattersinSurvival:PersonalIdentityandOtherPossibilities.DouglasEhring,OxfordUniversityPress(2021). ©DouglasEhring.DOI:10.1093/oso/9780192894717.003.0002
Therearesixmainwaysofconstruing fissionintermsofidentity.Iwill considerthe firstfourinthischapter.Iwillmakeuseofthefollowing provisionalassumptions:
FuturePersonQuestion:Inthischapter,IwillrestrictfocustotheFuture PersonQuestion. “Concerninganyfutureindividual,whatmattersinmy relationtohim?” (Johansson2010:31)²
Perdurantism: Personshavepropertemporalparts,andapersonpersists from t toanon-overlappingtime t’ byhavingatemporalpartat t anda distincttemporalpartat t’.³
I-RelationAssumption:TheproperPerdurantistanalogueof “identity” in thecommonsenseproposition “identitymatters” isthe I-relation.Stage s₁ is I-relatedtostage s₂ justincase s₁ and s₂ arestagesofthesameperson.Forit tobetrue/falsethat “identity” mattersin fissionisforittobetrue/falsethat the I-relationmattersin fission.⁴
NoSharedStages: Foreachpersonstage s,thereisonlyoneperson P of which s isastage.⁵
Our firstfourmappingsof “identity” ontothe fissioncasecanbeformulated inthelanguageofPerdurantismasfollows:
(1)Thepre-fissionstageofMr.Fissiony, s₁,theLeftystageimmediately after fission, s₂,andtheRightystageimmediatelyafter fission, s₃,are allstagesofthesameperson,Mr.Fissiony.
(2)Thereisnoperson P ofwhichthepre-fissionstageofMr.Fissiony, s₁, andthepost- fissionstageofLefty, s₂,arebothstages,andthereisno person P ofwhichthepre-fissionstageofMr.Fissiony, s₁,andthe post- fissionstageofRighty, s₃,arebothstages.
(3)Thereisaperson P ofwhichthepre-fissionstageofMr.Fissiony, s₁, andthefusionofthepost-fissionstagesofLeftyandRighty, s₂ + s₃, arebothstages.
²Insubsequentchapters,IwillalsoexpandmyfocustoincludetheFutureTimeQuestion, whichconcernswhatmatterswithrespecttoafuturetime.
³Onemightalsorequirethat P haveatemporalpartateachtimebetween t₁ and t₂ if temporallygappypeople/objectsarenotpossible.
⁴ IwillalsoassumeprovisionallyMatterRealism:thereisasituation s,suchthataperson C₁ at t₁ getswhatmattersinsurvivalwithrespecttoaperson C₂ at t2, and C₁ at t₁ getswhatmatters withrespectto C₂ at t₂ invirtueofsomerelation X between C₁ at t₁ and C₂ at t₂.
⁵ ThisassumptionisnotnaturaltoPerdurantismandwillbedroppedinthenextchapter.
(4)Thereisaperson P ofwhichthepre-fissionstageofMr.Fissiony, s₁, andthepost-fissionstageofLefty, s2, arebothstages,butthereisno personofwhichthepre-fissionstageofMr.Fissiony, s₁,andthepostfissionstageofRighty, s3, arebothstages,orthereisaperson P of whichthepre-fissionstageofMr.Fissiony, s₁,andthepost- fission stageofRighty, s3, arebothstagesbutthereisnopersonofwhichthe pre-fissionstageofMr.Fissiony, s₁,andthepost-fissionstageof Lefty, s2, arebothstages.
1.1.1TheFirstCandidate
The firstcandidate forthemappingofidentityonto fissionisthat Mr.FissionyisidenticaltoLefty,Mr.FissionyisidenticaltoRighty,and LeftyandRightyareidenticaltoeachother.Wehaverestatedthisinterpretationinthelanguageofpersonstages. s₁,thepre-fissionstageofMr. Fissiony,is I-relatedto boths₂,theLeftystageimmediatelyafter fission, and s₃,theRightystageimmediatelyafter fission.⁶ Itfollowsthat s₂ and s₃ are I-relatedonourtemporaryassumptionthattherecanbenosharedstages. s₂ and s₃ arestagesofMr.Fissiony.⁷ Isthereanaccountofpersonalidentity underwhichthisinterpretationiscorrect?⁸
Consider, first,atheoryofpersonalidentitythatmakesthiscandidate interpretationfalseof fission.
SimpleMemoryTheorys₁ andalaterstage, s₂,arestagesofthesame personjustincase s₂ includesorcouldincludeamemoryofanexperience in s₁,or s₂ standsintheancestralofthisrelationto s₁.
⁶ Thismappingwillconflictwiththecommonidea developed,forexample,inSider(2001: 60) ofa “temporalpart” ifapersonstageisatemporalpartsinceonthatcommonideaa personstageincludesallofthepersonduringthetimethatthestageexists.Infact,thisissuehas alreadybeenraisedindiscussionsoftimetravelandFour-Dimensionalism.Inacaseinwhich Jonestravelsbackintimeandmeetshisearlierself,theFour-Dimensionalistwillsaythatthere aretwopersonstagesofJonesthatexistsidebysideatthesametime.ButonSider’sdefinitionof atemporalpartthatwillentailthatolderJonesstageandyoungerJonesstagearenottemporal partssinceneitherofthemincludesallofJonesatthattime.Sider’ssuggestedsolutioninvolves understanding “person-stage” as “person-like” partsoftemporalparts. “Ordinarilymytemporal partatanytimeisapersonstage,butnotincaseoftimetravel” (Sider2001:101).Wemight extendthispointto fissionunderthis firstmapping thetwopost-fissionperson-stagesarethe person-likepartsofasinglepost-fissiontemporalpart.
⁷ Asiswellknown,DavidLewisrejectstheno-shared-stageassumptioninhisattemptto avoidIDMF.Fornow,however,wewilloperateundertheno-shared-stageassumption.
⁸ Parfitsetsthisinterpretationasideastoocounterintuitive.
SimpleMemoryTheoryisincompatiblewithour firstcandidateforinterpreting fissionintermsofidentitysinceitrulesoutthatLeftyandRightyare identicaltoeachother. s₂ doesnot/couldnotincludeamemoryof s₃’s experiences(nordoes s₃ includeamemoryof s₂’sexperiences),nordoes eitherstandintheancestralofthememoryrelationtotheother.Hence, s₂ and s₃ arenotbothstagesofsomecommonpersonunderSimpleMemory Theory.Butconsider,second,asophisticatedversionofMemoryTheory thatisresponsivetothefollowingcase:
SenileGeneral Asenilegeneralcannotremember evenwiththeaidofthe sortsofpromptingthatarecompatiblewithmemory performingabrave deedasayoungofficeroranyexperienceslaterthanthatdeed,butcanor doesremembervariouschildhoodexperiences.Thesenilegeneralstageisa stageofthesamepersonastheyoungofficerstagedespitethefactthatit doesnotandcouldnotcontainamemoryoftheexperiencesofthatstage. Nordoesthesenilegeneralstagestandintheancestralofthememory relationtothatstage.
PerrydiscussesarevisionofMemoryTheorythatinvolvesacertainkindof sequence.
Thereisasequenceofperson-stages(notnecessarilyintheorderthey occurintime,andnotexcludingrepetitions),the firstofwhichis A andthe lastofwhichis B,suchthateachperson-stageinthesequenceeither(i) contains,orcouldcontain,amemoryofanexperiencecontainedinthe next,or(ii)containsanexperienceofwhichthenextperson-stagecontains amemoryorcouldcontainamemory.(Perry1975:19)
Callsuchasequencea “ memorysequence. ” x₁ and x₂ arestagesofthesame personjustincasethereisamemorysequencebetweenthem.Asumof person-stagesisapersonjustincaseallthestagesinthatsumarememorysequencerelatedtoeachother,andtherearenostagesoutsidethatsumthat arememory-sequencerelatedtoanyofthestagesinthatsum.In Senile General,thereisamemorysequencelinkingthesenilegeneraltothebrave youngofficer senilegeneral–child–braveyoungofficer.So,thesenilegeneralisthesamepersonasthebraveyoungofficerunderthisrevision. ⁹
⁹ Althoughtherearelinesofmentalcausation butnotmemorycausation in Senile General runningfromtheyoungofficertothegeneral,wecangenerateamoregeneralversion
Nownoticethatthepost-fissionstages s₂ and s₃ inour fissioncasebearthe “ memorysequence ” relationtoeachother. s₂ canremembersomeofthe experiencesof s₁ and s₁ containssomeexperiencesthatarerememberedby s₃. Hence,onthismoresophisticatedmemorytheory,LeftyisidenticaltoRighty. Thereisoneperson,Mr.Fissiony,andheis Y shaped.¹⁰ Hence,thisfairly reasonableversionofamemorytheorymakessomemetaphysicalsenseoutof this firstinterpretationof fissionintermsofidentity.¹¹Wecouldalsoconsider asophisticatedversionofapsychologicalaccountofidentitythatbringsinto playpsychologicalrelationsotherthanmemory,suchascharactertrait constancy,beliefretention,actionplancontinuity,andthelike.
Nevertheless,thereareobjectionstotheclaimthatthesimultaneous, post-fissionstages s₂ and s₃ arestagesofthesamepersonthatmustbe considered,evenifitfollowsfromthissophisticatedmemorytheory.There arethreecharacteristicseachofwhichappearstoprovideconclusiveevidenceforthinkingthatthesepost- fissionstagesarenotstagesofthesame person.Iwillarguethat,infact,noneofthesecharacteristicsisincompatible withthe I-relation.
(1)Thepost-fissionstageslackthecapacityforsharedconsciousnessatthe sametime.¹²Forexample,Leftyimmediatelyafter fissionmaybeaware ofseeingsomethingblue,butRightyimmediatelyafter fissionfailsat thatsamemomenttoseeanythingblue.SoLefty’sstage, s₂,atthattime isnotastageofthesamepersonasRighty’sstage, s₃,atthattime.
(2)Therearetwosimultaneousbodystagespost-fission,oneassociated withLeftyimmediatelyafter fissionandtheotherassociatedwith Rightyimmediatelyafter fission.Iftwosimultaneousbutwholly distinctbodystagesareassociatedwithpersonstages,theseperson stagesarenotstagesofoneandthesameperson.So s₂ and s₃ arenot stagesofthesameperson.
of SenileGeneral thateliminatesanysuchmentalconnectionsfromtheyoungofficertothe general amoregeneralkindofpsychologicalsenility.
¹⁰ s₂ islinkedbyachainofmemoriesto s₁,and s₁ islinkedby “ reversememory ” sequenceto s₃,sothereisapersonthatincludesallthree. s₃ ismemory-sequencerelatedtoastage, s₂,and thatsequenceincludes s₁.
¹¹Istherealsoapersonthatincludes s₁ and s₂ butnot s₃?No,thereisnot.Since s₂ ismemorysequencerelatedto s₁,thepersonofwhich s₂ ispartwillincludeallthestagestowhich s₁ is memory-sequencerelated,including s₃.Thistheoryalsoexcludessharedstages.
¹²However,theremightbeabackward-lookingnotionofdiachronicunity.Theexperiences of s₂ and s₃ mayinvolvedifferentcontinuationsoftheearlierexperienceof s₁ sothatthereis somesenseofdiachronic-sharedconsciousnessbetween s₂ and s₃ althoughthereisnosynchronicunity.
Asitturnsout,inpointoffact,simultaneousstagesofthesame personcanhavebothcharacteristic s.Considerthefollowingtimetravelcase:
Self-MeetingTimeTraveler In2021,Jonesgetsintohistimemachine,and “travels” backto1956whenhewasfouryearsold.Hemeetshisearlierself andholdsaconversationwiththeyoungerJones.
ThemostnaturaldescriptionofthiscaseisthatJonesistalkingtohimself. Thereisreallyonlyonepersoninvolvedinthe1956conversation.Anearlier stageofJones,say,Jones-4,isinteractingwithalaterstageofJones,say Jones-68.DistinctstagesofJonesexistconcurrently,andthesestageshave characteristicsthatnoonestagecouldpossess.Ifthistimetravelcaseis possibleanditiscorrectlydescribedasinvolvingtwosimultaneousstagesof Jonesinaconversation,thenthe firsttwocharacteristics failureofshared consciousnessandtheexistenceofmorethanonebodystageatthesame time donotguaranteethattherelevantstagesarenotstagesofthesame person.
Thereis,however,athirdcharacteristic,physical-causalindependence, thatwouldseemtoruleouttheidentityofLeftyandRighty.Onecannot causeascarinRightyafter t₂ byinjuringLeftyat t₂.Purelyphysicalevents thatoccurtoone fissionproductcannotcausallyinfluence,bywayofpurely physicaleventsequencesthatrunthroughthestages,anypurelyphysical eventthatoccurstotheother fissionproduct.
Infact,itisatleastarguablethatthiskindofindependencedoesnot guaranteenon-identity.
FissionwithPre-Cognition
ThehemispheresofMr.Fissionyareeach transplantedintoanewbody,adifferentbodyforeachhemisphere.Each ofthepost- fissionpeople,LeftyandRighty,ispsychologicallycontinuous/ connectedtoMr.Fissiony.However,Mr.FissionyhasthepowerofprecognitionwithrespecttoRighty.Throughacombinationofmemoryand Mr.Fissiony’ s “pre-cognition” withrespecttoRighty,Leftycan “remember” experienceshadbyRighty.
Atthelevelofhismentallife,thestagesofLeftybeararelationtothe experiencesofRightythatissimilartotherelationwebeartoourpastselves. Forexample,LeftywillseemtorememberexperiencesfromRighty,and thosememoryexperienceswillbecausallydependentuponexperiencesof
