Counterparts of Persons and Their Bodies - David Lewis

8/12/2019 Counterparts of Persons and Their Bodies - David Lewis

1/10

Journal of Philosophy, Inc.

Counterparts of Persons and Their BodiesAuthor(s): David LewisReviewed work(s):Source: The Journal of Philosophy, Vol. 68, No. 7 (Apr. 8, 1971), pp. 203-211Published by: Journal of Philosophy, Inc.Stable URL: http://www.jstor.org/stable/2024902 .Accessed: 19/08/2012 18:48

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Journal of Philosophy, Inc. is collaborating with JSTOR to digitize, preserve and extend access to The Journalof Philosophy.

http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=jphilhttp://www.jstor.org/stable/2024902?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/2024902?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=jphil


2/10

COUNTERPARTS OF PERSONS AND THEIR BODIES 203

COUNTERPARTS OF PERSONSAND THEIR BODIES*

M/rATERIALISTS like myself oldthat persons ndtheir

bodies are identical. But there s a simple argument oshowthat this dentity hesis s refuted y the mere possi-

bility hat a person might witch bodies.To defeat the argument tseemsnecessary o revise my counterpart heory by providing ormultiplicity f counterpart elations.This revisionhas an odd result.Modal predictions may be de re, yet not referentially ransparent.

The thesis wish to defend here may be stated more precisely, s

follows:(T) Necessarily, person ccupies body t a time f and only f that

person s identicalwith hat body t that time.Note that the thesis T) is formulated ot n terms f dentity tself,a two-place elation, ut in terms f a derivative hree-place elationof dentity t a time. wish to regard nduring hings uch as personsand bodies as aggregates-sets, mereological ums, or somethingsimilar-of momentary stages. Enduring things X and Y are

identical t a time if and only f they both have stages at t-that is,exist t t-and their tages at t are identical. Therefore X and Yareidentical impliciter f and only f they re identical whenever itherone exists. Note that (T) does not say that persons nd bodiesmustbe identical simpliciter. t does imply that if a certain personoccupies a certain body whenever either the person or the bodyexists, hen heperson nd the body are identical. n such a case, alland only those tages which re stages of the person re stages of thebody he occupies. But (T) also permits ther cases: for nstance,body consisting f the stages of a certain person together with omefinal dead stages that are not stages of any person (and some initialprenatal stages that perhaps are not stages of any person); or aperson consisting f stages of a certain body together with someinitial r finalghostly tages that are not stages of any body; or evena body-switching erson consisting artly of stages of one body andpartly f stages of another body. A person consists of stages relatedpairwise by a certain relation we may call the relation of personal

unity; body consists of stages related pairwise by another relationwe may call the relation of bodilyunity.' Since for the most part*I am ndebtedoM. J. Cresswell,avidKaplan,ndJohn erry orlluminat-

ing iscussionf hesematters.1The o-calledproblemfpersonaldentity sthe roblemf xplicatinghe

relation fpersonal nity etweentages. hisview sexpounded ore ully yJohn Perry in Can the Self Divide? , presented at the APA Pacific Divisionmeetings, Mlarch1970.


3/10

204 THE JOURNAL OF PHILOSOPHY

persons occupy bodies and bodies are occupied by persons, itfollows ccording o (T) that the two relations f unity re relationson almost the same set of stages. The exceptions re dead stages,perhaps prenatal tages, nd perhaps ghostly tages. Moreover, f weleaveout the dead or ghostly r perhaps prenatal tages, then t leastfor the most part the two relations of unity are coextensive. Theexceptionswould be body-switchers nd perhaps split personalities.Nevertheless, he two relations of unity are different elations-in-intension; o they re coextensive nly contingently f at all.

Now I shall present an argument gainst (T). I regard t as a

simplified escendant of an argument ut forth y Jerome haffer,2but I have changed t so much that he might not wish to acknowl-edge it as his own.

Body-switching s logically possible. Because I might haveswitched ut of my present ody yesterday, hough n fact did not,I and my actual present body are such that the former might nothave occupied the atter oday. Whether r not persons re identicalwith bodies they occupy, certainly ersons re never dentical withbodies they do not occupy. So we have:(1) I and my body are such that they might ot have been dentical

today.Suppose that, as is surely t least possible, occupy the same bodyfrom he time when t and I began until the time when t and I willend. Then, by (T), it and I are identical whenever t or exist. Hencemy body and I, enduring hings, re identical simpliciter. y thisidentity nd Leibniz's law, (1) yields (2):

(2) My body and my body are such that they might not have beenidentical oday.Since (2) is self-contradictory, T) has apparently been refuted.

To rescue (T) without denying he possibility f body-switchingand without enying hat might ccupy the same body throughoutthe time that it or I exist, plan to show that the step from 1) to(2) by Leibniz's law is illegitimate-in other words, that (1) is notreferentially ransparent with respect to the term I'.

I have used the familiar are such that construction o indicatethat (1) and (2) are to be taken as de re rather han de dicto modalpredications. hat is, we are to consider what happens n other pos-sible worlds to the things denoted here n our actual world by theterms ITand 'my body', not what happens in other worlds to the

2 Persons nd TheirBodies, Philosophical eview, XXV, 1 (January 966):59-77.


4/10


5/10


only those things which resemble X closely enough in importantrespects, nd more closely than do the other things n their worlds.It is easier for a Republican than for a cockatrice to resemble meenough to be my counterpart.

The counterpart elation serves as a substitute for dentity be-tween things n different orlds. The principal advantage of themethod of counterparts ver the method of interworld dentities sthat if we adopted the latter n its most plausible form, we wouldsay that things were dentical with all and only those things whichwe would otherwise all their counterparts. ut that could not be

correct: first, ecause the counterpart elation s not transitive rsymmetric, s identity s; and second, because the counterpart ela-tion depends on the relative importances we attach to variousdifferent espects f imilarity nd dissimilarity, s identity oes not.

To recapitulate: n each possible world here s a set of momentarystages and a set of enduring hings omposedof stages related pair-wise by various relations f unity. An enduring hing nd its stagesexist only n one world, but may have counterparts n other worlds.We shall be concerned here only with counterparts f enduringtlhings, houghwe can allow that stages also have their ounterparts.

Applying he method of counterparts o the problem t hand, weimmediately ncounter bothersome distraction. The translationof (2), which eemed self-contradictory, s this:

There are a world W, a counterpart in W of my body, nd acounterpart in Wofmy body, uch hatX and Y are not denticaltoday.

Unfortunately, his translation omes out true, but for n irrelevantreason. I, and also my body whether r not I am identical with t,might have been twins. My body therefore oes have two differentcounterparts n certain worlds. Not only is the translation rue; itseems to me to show that (2) itself s true. But the argumentagainst (T) can easily be repaired. Replace (1) and (2) by:(1') I and my body are such that (without ny duplication f either)

they might ot have been dentical oday.(2') My body nd my body re such that (without ny duplication f

either) hey might ot have been dentical oday.The argument works s well with 1') and (2') as it did with 1) and(2). Indeed, (1') and (2') correspond o (1) and (2) as we would haveunderstood hem f we had forgotten hat I might have been twins.4

4 I do not know ow r whether 1') and (2') canbeexpressed nthe anguage fquantified odal ogic.That does not botherme. knowhowto express hem nEnglish nd in counterpart heory.


6/10

COUNTERPARTS OF PERSONS ANDTHEIR BODIES 207

Applying ounterpart heory o the repaired rgument, we obtainthese translations f (1') and (2'):(1*) There re a world W, a unique ounterpart in TVof me, and a

unique ounterpart in W of mybody, uch hat X and Y are notidentical oday.

(2*) There re a worldW, unique ounterpart in W of mybody, nda unique ounterpart in W of my body, uchthat X and Y arenot dentical oday.

The argument gainst (T) seems to go through, sing 1*) and (2*):(1*) seems true because I might have switched bodies yesterday;(2*) is self-contradictory; et (1*) implies (2*) by Leibniz's law,given (T) and the supposition hat I occupy the same body when-ever I or it exist.

In defense f (T), however, claim that (1*) isfalse, despite thefact that I might have switched bodies yesterday. What is truebecause I might have switched bodies is not (1*) but rather 1**):(1**) There re a world W, uniquepersonal ounterpart in W of me,

and a uniquebodily ounterpart in Wof my body, uchthat X

and Yare not dentical oday.I now propose a revision of counterpart heory to the effect hat,at least in the present ontext, 1**) rather han (1*) is the correcttranslation f (1'). What follows rom 1**) by Leibniz's law, given(T) and the supposition hat I occupy the same body wheneverexist, s not the self-contradiction 2*) but rather he truth:

There re a world TV, uniquepersonal ounterpart in W of mybody, nd a uniquebodily ounterpart in Wofmybody, uch hatX and Yare not dentical oday.

Two other truths ollow from 1**) in the same way:There re a worldW, a uniquepersonal ounterpart in WVf me,anda unique odily ounterpart in Wofme, uch hat X and Yarenot dentical oday.There re a world TV, uniquepersonal ounterpart in Wof mybody, nd a uniquebodily ounterpart in W of me, uch hat X andYare not dentical oday.

However, the translation f (2') is none of these. If the translationof (1') is (1**),the translation f (2') should be (2**):(2**) There re a worldW, a uniquebodily ounterpart in Wof my

body, nd a uniquebodily ounterpart in Wof my body, uchthat X and Yare not dentical oday.

Though (2**) is not (2*), it is still a self-contradiction.


7/10


As we already noted, ounterpart elations re a matter f over-allresemblance n a variety of respects. If we vary the relative im-portances f different espects f similarity nd dissimilarity, e willget different ounterpart elations. Two respects f similarity r dis-similarity mong enduring hings re, first, ersonhood nd personaltraits, nd, second, bodyhood and bodily traits. f we assign greatweight to the former, e get the personal ounterpart elation.Onlya person, or something ery ike a person, an resemble person nrespect of personhood nd personal traits enough to be his personalcounterpart. ut if we assign great weight to the latter, we get the

bodily ounterpart elation. Only a body, or something ery like abody, can resemble body in respect of bodyhood nd bodily traitsenough to be its bodily counterpart.

If I am my body, then n many worlds there re things hat areboth personal and bodily counterparts f me and ipso facto of mybody. These things, ike me, are both persons and bodies. But inother worlds (and my body) have neither personal counterpartsnor bodily counterparts; or personal counterparts that are notbodily counterparts; r bodily counterparts hat are not personalcounterparts; or personal and bodily counterparts hat are notidentical. A world n which switched ut of my body-that is, mypersonal counterpart witched out of my bodily counterpart-yesterday s of this ast sort. I and my body have there personalcounterpart hat is a person but not a body and also a bodilycounterpart hat s a body but not a person. These are not identicaltoday, and not identical simpliciter, hough they were identical attimes before yesterday ince they shared their earlier tages. How-

ever, my personal counterpart s identical today with a differentbody. My bodily counterpart s identical today with a differentperson (if the body-switching as a trade) or with none.

We may draw an analogy between the relations of personal andbodily unity among stages of persons and bodies and the personaland bodily ounterpart elations mong nduring ersons nd bodies.If I ask of something hat is both a stage of a body and a stage ofa body-switching erson Was this ever in Borneo? you shouldask whether mean this person or this body. If the former, amasking whether he given stage is linked by personal unity to anearlier tage located n Borneo. If the latter, am asking whether tis linked by bodily unity to an earlier stage located in Borneo.Similarly, f ask of something hat s both an enduring erson ndan enduring ody Might this have been an orangutan? you shouldagain ask whether mean this person or this body. If the former,


8/10

COUNTERPARTS OF PERSONS AND THEIR BODIES 209

am asking whether t has an orangutan for personal counterpart;if the latter, whether t has an orangutan for bodily counterpart.

But the analogy is imperfect. he two relations of unity areequivalence relations, t least for the most part and as a matter ofcontingent act.Therefore t is easy and natural to form he conceptof an enduring erson or body, consisting f stages linked togetherpairwise by a relation of personal or bodily unity. t is tempting odo the same with the counterpart elations, forming he conceptof a superperson r superbody consisting f persons or bodies indifferent orlds, inked together y a personal or bodily counterpart

relation. But this cannot be done in any straightforward ay be-cause counterpart elations are not equivalence relations. Like allsimilarity elations n a sufficiently ariegated domain, they fail tobe transitive because chains of little differences dd up to bigdifferences.

Why should I think t plausible to employ multiple ounterpartrelations to translate (1') as (1**)rather than (1*)? Precisely be-cause by doing so I escape the refutation f (T), and I am convincedof (T). I am offered trade: instead of a multiplicity f kinds ofthing can have a multiplicity f counterpart elations. A reductiorefutes he whole combination f assumptions hat led to contradic-tion; if all but one of those assumptions re highly lausible, which-ever remains s the refuted ne. And in addition, f contemplate hepropositions express by means of (1') and (1**), it seems to methat they are the same.

I would like to present he translation f (1') by (1**)and (2') by(2**)as instances f a general chemefor ranslating nglish modal

predications nto sentences of counterpart theory with multiplecounterpart elations. do not know how to do this. Roughly, theidea is that the sense of a term omehow elects the counterpart ela-tion that is to be used to find he counterparts f the thing denotedby that term. The terms I', 'you', that person', the lady I saw youwith ast night', George', all select the personal ounterpart elation.'This thing' pointing t myself), this body', my body', that whichwill be my corpse after die', all select the bodily counterpart ela-tion. Similarly or ndefinite erms phrases of restricted uantifica-tion): 'everybody' elects the personal ounterpart elation,whereas'every body' selects the bodily counterpart elation. Even if every-body is his body, and conversely, Everybody s such that he mighthave beena disembodied pirit s true, whereas Every body ssuchthat it might have been a disembodied pirit is false. The formermeans that each of those things which are both persons and bodies


9/10

210 THE JOURNALOF PHILOSOPHY

has a disembodied pirit s personal counterpart, whereas the attermeans that each of those same things has a disembodied pirit asbodily counterpart.

In certain modal predications, he appropriate counterpart ela-tion is selected not by the subject term but by a special clause. Tosay that something, egarded as a such-and-such, s such that itmight have done so-and-so s to say that n some world t has a such-and-such-counterpart hat does so-and-so. With these regarded sclauses in mind, might ay that I translate 1') as (1**) because Itake it to be synonymous with (1 ):

(1 ) I, regarded s a person, nd my body, egarded s a body, re suchthat without ny duplication f either) hey might othave beenidentical oday.

Likewise I translated 2') as if it had contained two regarded asa body clauses.

If we are to have multiple counterpart relations, we may wellwonder how many to have. One for every sortal? One for everynatural kind? One for ny common noun phrase whatever hat can

grammatically e inserted nto regarded s a ', even the phrase'yellowpig or prime number'? One for ny kind of entity, ven kindsthat cannot be specified n our language?5 do not know. Nor do Iknow whether ne of the counterpart elations, orresponding er-haps to the clause regarded s an entity', an be identified ith thesingle counterpart elation of my original ounterpart heory.

It is customary o distinguish eal essences of things from heirnominal essences under descriptions. Now, however, we have athird, ntermediate, ind of essence. My real essence consists of theproperties ommon to all my counterparts. Here I use the originalsingle counterpart elation.) My nominal ssence under the descrip-tion 'person' consists of the properties common to all possiblepersons. My intermediate ssence under the description person'consists of the properties ommon to all my personal counterparts.I have no reason to think that any two of these sets of propertiesare the same. It may even be that none of the three s properly n-cluded in any other, if my personal counterparts nclude some

entities (robots, say) which are almost persons but not quite.Counterpart relations are vague, being dependent on the relativeweights ssigned to respects f similarity r dissimilarity. ence real

6We could lso put this uestion nother ay: given three-placeelationif is a al counterpart f.. , which inds re appropriate iddlearguments?


10/10

COUNTERPARTS OF PERSONS ANDTHEIR BODIES 2 I

essencesare vague in a way nominal essencesare not. Intermediateessences under descriptions share this vagueness, for the newmultiple counterpart elations are no less vague than the originalcounterpart elation.

In my original ounterpart heory, ny de re modalpredication sreferentially ransparent. omething as the same counterparts ow-ever we may choose to refer o it. Given a de re modal predication,we find he thing denoted by the subject term n the actual world;then we consider what befalls that thing-or rather, ts counter-parts-in other worlds. Only the denotation of the subject termmatters. We can substitute another subject term with the samedenotation but different ense, and the truth value of the modalpredication will not change.

But in the present revision of counterpart heory, de re modalpredications re not n general ransparent. Not only the denotationof the subject term matters, but also the counterpart elation itselects. f we substitute nother ubject term with the same denota-tion but different ense, t may change the truth value of the modal

predication by selecting different ounterpart elation.Then eventhough the denoted thing here in our world remains the same, wehave a different ay of following he fortunes f that thing n otherworlds.

Nevertheless, hese modal predications re still dere, not dedicto.We still find the denoted thing n our actual world and then findcounterparts f that thing lsewhere. We do not at all consider hethings denoted by the subject term n other worlds, s we would nthe case of a de dicto modal predication.

Transparency of modal predications an fail whenever he senseof the subject term s used to do anything eyond determining heactual denotation f the subject term. One further hing t might dois determine hedenotation f the subject term n other worlds; thatis the de dicto ase. Another, nd altogether ifferent, urther hingit might do is select a counterpart elation. These two are not theonly alternatives.) t is the latter, suggest, that happens in theargument we are considering. Therefore we can accept (1') as a

consequence of the possibility hat I might have switched bodies,reject (2') as self-contradictory, nd yet accept (T) and its conse-quence that if I occupy the same body whenever or it exist thenI am my body.

DAVID LEWIS

St. Catherine's College, Oxford

Documents

Counterparts of Persons and Their Bodies - David Lewis