Bare Particulars

Author(s): Edwin B. Allaire

Source: Philosophical Studies: An International Journal for Philosophy in the Analytic
Tradition, Vol. 14, No. 1/2 (Jan. - Feb., 1963), pp. 1-8
Published by: Springer
Stable URL: http://www.jstor.org/stable/4318430 .
Accessed: 06/08/2014 03:18
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 forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Springer is collaborating with JSTOR to digitize, preserve and extend access to Philosophical Studies: An
International Journal for Philosophy in the Analytic Tradition.

http://www.jstor.org

This content downloaded from 128.83.63.20 on Wed, 6 Aug 2014 03:18:14 AM

All use subject to JSTOR Terms and Conditions

PHILOSOPH
STUDIES
Edited by WILFRID SELLARS and HERBERT FEIGL with the advice and
of PAUL MEEHL, JOHN HOSPERS, MAY BRODBECK
assistance
VOLUME XIV

ContentsJanuary-February
1963

BareParticularsby Edwin B. Allaire,STATE UNIVERSITY

NUMBERS 1-2
OF IOWA

Does Modal Logic Rest upon a Mistake?by R. M. Martin, UNIVERSITY

and UNIVERSITY OF TEXAS
IndirectSpeech Again by A. N. Prior,MANCHESTER

OF BONN

UNIVERSITY

IndirectSpeech: A Rejoinderto Prof.A. N. Priorby L. JonathanCohen,

QUEEN'S COLLEGE, OXFORD

Baier'sTest for PracticalRules Re-Examinedby David BarryLyons,

HARVARD UNIVERSITY

A Note on Commitment by LennartAqvist, UNIVERSITY

OF UPPSALA, SWEDEN

Lewis' Imperativesof Right by William K. Frankena,UNIVERSITY

OF MICHIGAN

A Note on Austin'sPerformativeTheoryof Knowledgeby BernardMayo,

UNIVERSITY OF BIRMINGHANM

Bare Particulars
by EDWIN B. ALLAIRE
STATE UNIVERSITY OF IOWA
CONSIDER 'this is red,' assertedtruly of a colored disc. Some philosophers
claim that the sentencerefersto a fact consistingof two (kinds of) entities,
an individual (bare particular)and a character(universal),referredto by
'this' and 'red,'respectively.They claim furtherthat the two entities stand
in the exemplificationrelation,representedby 'is.' Currently,that claim is
1

This content downloaded from 128.83.63.20 on Wed, 6 Aug 2014 03:18:14 AM

All use subject to JSTOR Terms and Conditions

PHILOSOPHICAL
STUDIES

widely rejected.Underlyingthe many argumentssupportingthat rejection

is a rathersimple idea which Russell once expressed:"One is tempted to
regard'This is red' as a subject-predicate
proposition;but if one does so,
one finds that 'this' becomes a substance,an unknowablesomething in
which predicatesinhere . .
Though awkwardlyexpressed,Russell's point is clear: The individualcharacteranalysisis at odds with the empiricaltradition.That is, if one
claims that 'this is red' is a subject-predicate
propositionin the sense that
'this' and 'red'referto unanalyzableentities of differentontologicalkinds,
then one has violated the Principleof Acquaintance(PA),2 a basic tenet
of empiricism.One is not acquaintedwith "a something"which could be
construedas an entity of a kind differentfrom red, except in the sense that
this is a "collection"of the samekind of thingsas red,e.g., squareand bright.
The heart of Russell'spoint is thus that the individualsof the individualcharacteranalysisare unknowablein the sense that one is not directlyacquaintedwith them.
I proposeto explorewhether or not one can propoundthe individualcharacteranalysiswithout abandoningthe PA. In particular,I want to discussBergmann'sassertion3that "beingacquaintedwith a redspot,and nothing else, one is presentednot with just one thing but two, a particular. . .
and a character. . ." If I understandBergmann,he is maintainingprecisely
what Russelldenies,namely,that one is on such occasionsacquaintedwith
an individualor, as he prefersto call it, a particular.His motivationis clear.
He attemptsto reconcilethe individual-character
analysiswith the PA.
Beforediscussinghis attempt it will be helpful to examinethe dialectics
which give riseto the problem.I will do that by commentingbrieflyon two
otheranalysesof the disc: (I) The disc is a collectionof what has sometimes
been called "perfectparticulars."'This is red,' assertedtruly of our disc, is
analyzedas follows: 'this' refersto a collection of entities, one of which is
referredto by 'red.'This latterentity is such that if thereweretwo red discs,
there would be two such entities (perfect particulars),each unanalyzable
and numericallydifferentfrom the other. (II) The disc is a collection of
characters(universals).'This is red' is analyzedas in (I), except that the
entities are such that if there were two red discs, the two collectionswould
have one memberin common,i.e., one memberwhich is literallythe same.
Each of these analysesencountersan immediate and, I believe, insurmountableobjection.Considertwo discsof the same (shadeof) color,size,
shapeand so on. The objectionto (I) is that it cannotaccountfor the sameness of the discs since the membersof the two collectionsare all unanalyzable and differentfrom each other. In otherwords,though both collections
contain a memberreferredto by 'red,' (I) providesnothing they have in

This content downloaded from 128.83.63.20 on Wed, 6 Aug 2014 03:18:14 AM

All use subject to JSTOR Terms and Conditions

BAREPARTICULARS

common since the membersare unanalyzableas well as different.The objectionto (II) is that it cannotaccountfor the difference,since each collection has literallythe samemembers.4
The individual-character
analysisencountersneither objection. That is
its strength.The differenceof the discs is accountedfor by each containing
a differentindividual;the sameness,by each containingliterallythe same
characters(as in (II)). The individual-character
analysisthus allows one
to solve the problemsof samenessand difference,at least as they arise in
connection with "things."Speakingmore traditionally,it providesa solution to the nominalism-realism
issue.
Notice that a proponentof the individual-character
analysis,or, as I shall
henceforthcall him, a realist,explicitlygrounds(by meansof entities) the
samenessas well as the differenceof the two discs. There must be entities
to accountfor the discsbeing calledthe same or, less accurately,there must
be a sharedsomethingin orderto accountfor the sameword ('red') being
truly predicatedof 'this' and 'that.' The objection to (I) is in fact sometimes expressedas follows.Since the two red entities of the two collections
are differentand unanalyzable,there is no way of accountingfor theirbeing
referredto by the sameword ('red'). So expressedthe objectionis misleading. Moreover,it temptsa defenderof (I) to appealmistakenlyto a variant
of the meaning-is-usedoctrine.
The demandfor groundingsamenessis at presentsuspect.The suspicion
makes (I) seem attractive.Considertwo discs of differentshadesof red. In
ordinarylanguage,'red'may be correctlyused to referto either. It appears
that a realist,to be consistent,mustalso groundthe samenessof the different
shades.For, if he objects to (I) becauseit does not explainwhy the same
wordis used to referto two "perfectparticulars,"
he must take seriouslythe
same objectionin the case of 'red'when appliedto differentshades.Moreover,sinceeachshadeis simpleand unanalyzablethe ground(i.e., the shared
entity) must be of a differentkind. Thus, a thirdkind makesits appearance.
For those of a Platonicstripe,this third is a "transcendentuniversalor concept."5 Eachshadeis redby virtueof participatingin the sametranscendent
universal,which is the properreferentof 'red.'Hence, either the PA cannot
be maintainedor we must be acquaintedwith ("intuit") concepts.The second alternativehas been unacceptableto those who embracethe PA.
The realistthus seems doomed to Platonism.The proponentof (I) who
denies the need for groundingsamenessdoes not. This seemingadvantage
has had its effects.Recently,it has been arguedthat the root of Platonism
is the mistakenway in which we sometimeslook at language.In particular,
the veryattempt to groundsamenessrevealsthe mistakenbelief that there
must be a referentto justifythe use of a word.Not even the "referringuse"

This content downloaded from 128.83.63.20 on Wed, 6 Aug 2014 03:18:14 AM

All use subject to JSTOR Terms and Conditions

PHILOSOPHICAL
STUDIES

of a wordrequiresa referent,or so it is argued.What holds in generalholds

for 'red.'Hence, its use in referringto differentshadesneed not be grounded.
Thus, we are told, "the samenessof the shades"merelymeansthat they are
referredto by the same word. Moreover,it is claimed,even in the case of
things of the same shade each exemplificationof it is reallydifferentand
unanalyzable.In other words, their samenessalso consists merely in their
being referredto by the same word.6
A defenderof (I) may thus attempt to disposeof the objectionthat he
cannot account for the samenessof "things"by arguingthat the very attempt to accountfor it is mistaken.The attempt,he holds, inevitablyleads
to Platonismwith all its horrorsof transcendententities. That shows why
the use doctrinemaybe attractiveto a defenderof (I).
Consideragaintwo discsof the same (shadeof) color,size, shape,and so
on. Supposethey are shown to you, one to the left of the other. If after a
while you areshownthem again,you will not be able to tell which is which.
In fact, you will not be able to tell whetherthe two you now see arethe two
you saw earlier.You can only tell that the two you now see have all the
propertiesthat the two you saw earlierhad. In other words,taken as such
and in themselves,two perfectparticularsof the same shadecannotbe told
apart.That means that in this case at least the samenessis grounded.We
know how the realistgroundsit. It remainsto be shown that he need not
thereforegroundthe samenessof differentshadesin orderto justifythe use
of 'red'for any of them.
My realisttakesadvantageof what the use doctrinehas taughtus, namely,
(a) that from the fact that the same word is used to referto two things it
does not alwaysfollow that the two sharean entity. (Indeed, we could decide to use just one word to referto greenand red.) On the other hand, it
does not follow (b) that they neverdo. The proponentsof (I) rashlyinfer
(b) from (a). Their rashness,I have argued,may be due to their mistaken
belief that even (b) commitsone to Platonism.Some things, our two discs
for example,are the same shade of color. This is not a linguisticfact; i.e.,
not merelythe fact that they are referredto by the sameword.They are the
same in that they are indistinguishableas such, or, more precisely,one cannot differentiatethem by their coloralone.
Let us take stock. The problemof analyzingsuch things as coloreddiscs
arisesin the context of the realism-nominalism
issue. That issue, we saw,
cannotreallybe handledby accountingfor the samenessin the discsin terms
of the samenessof words. At some point samenessmust be groundedin
entities. To believe otherwiseis to put upon the meaning-is-usedoctrinea
burdengreaterthan it can bear.Thus, in our case, it remainsthe firsttask
of analysisto single out the unanalyzableentities which accountfor the dif-

This content downloaded from 128.83.63.20 on Wed, 6 Aug 2014 03:18:14 AM

All use subject to JSTOR Terms and Conditions

BAREPARTICULARS

ferenceand the samenessof the two discs.The second taskor step, as in all
cases,is to employ the result obtainedin the first step to explicatecertain
philosophicalusesof words,therebydissolvingthe traditionaldialectics.The
realisticanalysisprovidesthe requiredgrounding.Yet it is tainted. Like
Russell, many philosophersclaim that they are not acquaintedwith individuals.The heart of the matter is whether or not Bergmann'sclaim that
one is in fact acquaintedwith individuals(bareparticulars)is defensibleor,
even, whetherit can be made intelligible.
An obviousobjectionis that the claimmerelyspringsfrom the dialectical
needs it satisfiesand is not borne out by a careful inspection of what is
in fact presented.Indeed, Bergmannhimself invites this objection. "I, of
course,have convincedmyselfthat I am actuallypresentedwith two things.
Yet I am loathe to rest the case on this conviction;for I am also convinced
that a verymajorpartof it is dialectical."7
One cannot but wonderhow one does convinceoneself of such matters.
In the light of what he himselfsays,one is indeed tempted to concludethat
his conviction is dialecticalratherthan phenomenological,if I may so expressmyself. If so, then he has abandonedthe PA. Whether or not under
the pressureof the dialectiche has actuallydone that is not my concern.
The point is ratherwhether or not the "description"he proposescan be
defendedindependentlyof the dialectics.
Phenomenologicaldescriptionis promptedby philosophicalpuzzles.One
turnsto the formeras a preludeto the dissolutionof the latter.The dangers
are obvious.One maythink that one still describeswhen in fact one already
argues.That makesall alleged descriptionssuspect.The best one can do is
elaboratethem in severalways,alwayson guardagainstthe variousbiasesthat
might creepin. Once a descriptionhas been accepted,the puzzlesmust be
solvedby speakingcommonsensicallyabout it. In the courseof thus speaking about it, one maywell be led to reconsiderit. Afterall, it was prompted
by the puzzles.Nevertheless,one must not and cannot give awaythe game
by maintainingthat the descriptionis forced upon one by the dialectics.
Bergmannin the quoted passagecomes dangerouslyclose to doing just that.
The most one could say is that the dialecticsdirectsour attention toward
what is presented.But it does not and cannot tell us what actuallyis presented.
A comparisonmay help to makethe point clear.The laterWittgenstein,
believing that the philosophicalpuzzles arise from the misuse of words,
undertookto describetheircorrectuse. The misusesareengenderedby what
he calls misleadinggrammaticalanalogies.Once these latter are recognized
as such, the philosophicalpuzzlesdisappear.This is his basic idea.To speak
as before, Wittgenstein'sdescriptionof correct (and incorrect) use is the

This content downloaded from 128.83.63.20 on Wed, 6 Aug 2014 03:18:14 AM

All use subject to JSTOR Terms and Conditions

PHILOSOPHICAL
STUDIES

preludeto his solution of the philosophicalpuzzles. Are then his descriptions unbiased?There is no guarantee.Nor is it reasonableto demandone
beforehand.Wittgenstein'sdescriptionproceedsdirectlyfromwhat he considersa puzzleand thus indirectlyfromwhathe considersan illicit use. Thus,
the dialecticsmayhave influencedthe description.But once more, the best
one can do is guardagainstthe biasesthat may have been introduced.So,
too, with phenomenologicaldescription.In this regardall philosophersare
in the sameboat. They all startfromwhat they considerunproblematicalor,
as it is sometimesput, fromwhat they hold to be commonsensical.With respect to their startingpoint they must alwaysbe vigilant.One cannot do
more;one must not do less.
I returnto the issue: Can the realisticanalysisbe defendedon phenomenologicalgrounds?It will be well to distinguishbetween two uses of 'know.'
First, there is the use of 'know'in which to know somethingmeans to be
acquaintedwith it. Second, there is the use in which to know something
means to be able to recognizeit. In the second sense individualsas such
or in themselvesareunknowable.Consideragainthe two discsand the situation in which you are shown them twice. Since you can only tell that the
two you now see have all the propertiesthat the two you saw earlierhad, it
follows that if each consists of an individualand the severalcharactersit
exemplifies,the individualsas such or in themselvesare not recognizableor,
as I preferto say,not reidentifiable.The charactersare.To expressthe point
differently,two individualsare merely numericallydifferentwhereastwo
charactersare intrinsicallydifferentas well.
Russellheld that individualsare unknowable.A defenderof the realistic
analysismay take him to have held merelythat they are not recognizable,
i.e., that they cannotbe knownin the secondsense of 'know.'This, though,
is not at all what Russellmeant. Rather,in sayingthat individualsare unknowable,he used 'know'in the firstsense. In turn, I hold with Bergmann
that while they are indeed unknowablein the second sense they are known
in the first.Accordingly,I cannot get awaywith just maintainingthat they
aremerelynumericallydifferent.I must showin whatsenseone is acquainted
with them. Not to recognizethis obligationwould be to confuse again the
two uses of 'know.'Nevertheless,in pointing out that individualsare not
recognizable,i.e., are merely numericallydifferent,one has arrivedat the
heartof the matter.Individualsare just those entities which do groundthe
numericaldifferenceof two things which are the samein all (nonrelational)
respects.
Consideronce more the two discs. When presentedtogether, they are
presentedas numericallydifferent.That differenceis presentedas is their
samenesswith respectto shape, (shadeof) color,and so on. What accounts

This content downloaded from 128.83.63.20 on Wed, 6 Aug 2014 03:18:14 AM

All use subject to JSTOR Terms and Conditions

BAREPARTICULARS

for that differenceis the numericallydifferentindividuals.No character,or

groupof characterscan do that. Thus, to say that there are individualsis to
say that things may be merely numericallydifferent.No matter what descriptionone proposes,the numericaldifferenceof two things which are
alike in all (nonrelational)respectsmust be accountedfor. Consider (IL).
To claim that both discs are but collectionsof literallythe same universals
does not account for the thisnessand thatnesswhich are implicitlyreferred
to in speakingof them as two collections.That is, the two collections of
characters-if one persistsin speakingthat way-are, as presented,numerically different.Clearly,therefore,something other than a charactermust
alsobe presented.That somethingis what proponentsof the realisticanalysis
call a bare particular.Or, perhapsbetter, that is their explicationof 'bare
particular.'
One difficultyremains.Bergmannclaims that in being presentedwith
one redspot one is presentedwith two things.That maybe misleading.The
most prevalentuse of 'two' carriesa spatialconnotation.That is, if thereare
two thingsthen they arespatiallyrelated.Yet exemplification,the "relation"
obtainingbetween a characterand an individual,is obviouslynot spatial.
Thus, the sense in which there are two things is merelythe sense in which
therearetwo characters(redand square)in the presentationof a red square.
For red and squareare not spatiallyrelated.Nor are an individualand the
characteror charactersit exemplifies.If one shouldinsist that the two charactersare in fact spatiallyrelated,arguingthat they are at the same place, I
merelyask him what he means by 'place.'There is only one answerI can
think of whichwould help. A placeas such is itself an entity. But, then, this
answermakes places into bare particulars.This piece of dialecticsis well
known.
Consideragain 'this is red,'assertedtrulyof a coloreddisc. To what does
'this' refer?That it does refer to somethingis obvious. It does not always
referto the entity referredto by 'red.'That is, 'is' in this utterancedoes not
alwaysstand for identity. Hence, 'this' can only referto one of two things:
an individualor a collection of characters,one of which is red.8We know
alreadythat the latter alternativefails, for in the case of two collectionsof
charactersthe membersof which are the same,the numericaldifferencebetween the two collectionsis left unaccountedfor. The realisticanalysisdoes
accountfor it. Moreover,two individualsare presentedin the sense that the
two collectionsare presentedas numericallydifferent.That is not to deny,
though, that individualsare merely numericallydifferentfrom each other
and thus not reidentifiableas such. That they are explainswhy they have
been overlookedso often.
I havearguedthat one can propounda realisticanalysiswithout abandon-

This content downloaded from 128.83.63.20 on Wed, 6 Aug 2014 03:18:14 AM

All use subject to JSTOR Terms and Conditions

PHILOSOPHICALSTUDIES

ing the PA. Moreover,one can single out the bareparticularswithout using
'exist'philosophically,thus avoidingthe dialecticsof the nominalism-realism
issue. In tryingto do this, I described,or triedto describe,the sort of entity
an individualis. Positively,individualsare the carriersof numericaldifference as directlypresentedto us. Negatively,individualsare not rudimentary
Aristoteliansubstances.Thus, they are not the sort of things Russell probably had in mind when he denied that they can be known. To the sort of
thing he had in mind Locke'sphrase"a somethingI know not what"does
indeedapply.To what I mean by an individual,it does not. To one who accepts the PA, Locke's phraseprovidessufficientground for rejectingthe
entities he speaksof. The individualsI want to keep frombeing overlooked
are not such entities.That is why one need not abandonthe PA in orderto
maintainthat we are presentedwith bareparticulars.
ReceivedFebruary10, 1961
NOTES
'Bertrand Russell, An Inquiryinto Meaning and Truth (London: Allen and Unwin,
1948), p. 97.
2 The PA states that the indefinableterms of any "ontological"descriptionmust refer
to entities with which one is directlyacquainted.Furthermore,"an unanalyzableentity"
is explicatedto mean an entity representedby an indefinableterm.
'Gustav Bergmann,"Strawson'sOntology," Journalof Philosophy, 57:616 (September 1960).
' Cf. Gustav Bergmann, "Russell on Particulars,"Philosophical Review, 56:59-72
(1947). Reprintedin The Metaphysicsof LogicalPositivism (London: Longmans,Green,
1954), pp. 197-214.
6 For a discussionof concepts, see ReinhardtGrossman,"Conceptualism,"Review of
Metaphysics,14:243-54 (December 1960).
'This explication of sameness may well explicate the Hatus-vocisdoctrine of some
medievals.
' Bergmann,"Strawson'sOntology,"p. 616.
8 As I have shown, the "perfect-particular"
analysis (i.e., (I) ) is unacceptablebecause
it does not account for sameness.

DoesModalLogicRest upona Mistake?

by R. M. MARTIN
UNIVERSITY OF BONN AND UNIVERSITY OF TEXAS

PHILOSOPHERSdifferenormouslyas to

just what they wish to include under

"formallogic." Some wish to include the whole of mathematics,as developed, say, within the simplifiedtheory of types or within an axiomaticset

This content downloaded from 128.83.63.20 on Wed, 6 Aug 2014 03:18:14 AM

All use subject to JSTOR Terms and Conditions