SEARLE, J. Proper Names.

8/18/2019 SEARLE, J. Proper Names.

1/9

Proper Names

John R. Searle

Mind , New Series, Vol. 67, No. 266. (Apr., 1958), pp. 166-173.

Stable URL:

http://links.jstor.org/sici?sici=0026-4423%28195804%292%3A67%3A266%3C166%3APN%3E2.0.CO%3B2-%23

Mind is currently published by Oxford University Press.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtainedprior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content inthe JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/journals/oup.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. Formore information regarding JSTOR, please contact [email protected].

http://www.jstor.orgFri May 18 09:04:17 2007

http://links.jstor.org/sici?sici=0026-4423%28195804%292%3A67%3A266%3C166%3APN%3E2.0.CO%3B2-%23http://www.jstor.org/about/terms.htmlhttp://www.jstor.org/journals/oup.htmlhttp://www.jstor.org/journals/oup.htmlhttp://www.jstor.org/about/terms.htmlhttp://links.jstor.org/sici?sici=0026-4423%28195804%292%3A67%3A266%3C166%3APN%3E2.0.CO%3B2-%23


2/9

1 1. P R O P ER N A M E S

DO proper names have senses ? Prege

argues that they must

- have senses, for, he asks, how else can identity statements be

other tha n trivially analytic. How, he asks, can a state m ent of

the form a b, if tru e, differ in cognitive value from a a ?

His answer is th a t though a and b have th e same referent

they have or may havevdifferent senses, in which case the state-

ment is true, though no t analytically so. B u t this solution

. seems more approp riate where a and b are bo th non-

synonymous definite descriptions, or where one is a definite

description and one is a proper name, th an where both are proper

names. Consider, for example, statem ents made with th e follow-

ing sentences :

(a ) Tully Tully is analytic.

But is

(b) Tully Cicero syn thetic ?

If so, then each name m us t have a different sense. which seems a t

first 'sight most implausible, for we do n ot ordinarily th ink of

proper names as having a sense a t all in t he way th a t predicates

do ; we do no t, e.g. give definitions of proper names. B u t of

course (b) gives us information not conveyed by (a ). B u t is this

information abou t words ? The statem ent is no t abo ut words.

For th e mom ent let us consider th e view th a t (b) is, like (a),

analytic. A statement is analytic if and only if it is true in

virtue of linguistic rules alone, without any recourse to empirical

investigation. The linguistic rules for using th e name Cicero

an d th e linguistic rules for using th e name Tully are such th a t

both names refer to, without describing, the same identical

object

;

th us i t seems th e tr u th of th e identity c an be established

solely by recourse to these rules a nd th e s tate m ent is analytic.

The sense in which the statem ent is informative is th e sense in

which any analytic statement is informative ; it illustrates or

exemplifies certain contingent facts about words, though it does

not of course describe these facts. On th is account th e difference

between (a) and (b) above is not as great as might a t first seem.

Both are analytically true, and both illustrate contingent facts

abo ut our use of symbols. Some philosophers claim th a t (a ) is

I11rans2ations rom tihe Philosoplhical W rit ings of Gottlob Frege edited by

Ceach and Black pp 56 ff

1G6


3/9

167

ROPER NAMES

fundamentally different from ( b ) in that a statement using this

form will be true for any arbitrary substitution of symbols

replacing Tully

.I

This,

I

wish to argue, is no t so. The fac t

that the same mark refers to the same object on two different

occasions of its use is a convenient but contingent usage, and

i,ndeed we can easily imagine sit,uations where th is would no t be

th e case. Suppose, e.g. we have a language in which the rules

for using symbols are correlated not simply with a type-word,

but with the order of its token appearances in the discourse.

Some codes are like this. Suppose th e first time a n object is

referred to in our discourse it is referred to by x

,

th e second

time by y

,

etc. Bor anyone who knows this code x = y

is trivially analytic, but x x is senseless. This exam ple is

designed to illustrate the similarity of ( a ) and ( b ) above ; both

are an alytic and both give us information, thoug h each gives us

different information, abo ut th e use of words. The tr u th of the

statements that Tully = Tully and Tully = Cicero both follow

from linguistic rules. B ut the fact th at th e words Tully=

Tully are used to express this iden tity is just a s contingent as,

though more universally conventional in our language than, the

fact th a t th e words Tully = Cicero are used to express th e

iden tity of th e same object.

This analysis enables us t o see how both ( a ) and ( b ) could be

used to make analytic statements an d how in such circumstances

we could acqu ire different information from them, withou t forcing

us t o follow either of Frege s proposed solutions, i.e. t ha t t he

tw o propositions are in some sense abo ut words (Bcgrifsschr t)

or his revised solution, th at th e terms have th e same reference but

different senses

Xinn

un d Bedeu tung) .

B ut though th is analysis

enables us t o see how a sentence like ( b ) could be used to make a n

analytic statement it does not follow that it could not also be

used to make a synthetic statemen t. And indeed some identity

statem ents using two proper names are clearly syn thetic people

who argue that Shakespeare was Bacon are not advancing a

thesis abo ut language. I n wh at follows I hope to examine the

connection between proper names and their referents in such a

manner as to show how both kinds of identity statement are

possible and in so doing to show in w hat sense a proper nam e has

a sense.

I

have so far considered the view th at th e rules governing th e

use of a proper name a re such th a t i t is used to refer to and no t

to describe a p articular object, th a t i t has reference bu t not sense.

B u t now let us ask how i t comes about th a t we are able to refer

W. V Quine, ~ o a Logical Point

o

View,

esp chap

2


4/9

168

J R SEARLE :

to a particular object by using its name. How , for example, do

we learn an d teach th e use of proper names This seems quite

simple-we identify th e object, and , assuming th a t our stu de nt

understands the general conventions governing proper names,

we explain th a t this word is th e name of th a t obiect. B u t unless

our siud ent already knows another proper n a i e of the object,

we can only

identify

the object (the necessary preliminary to

teaching the name) by ostension or description

;

and, in both

cases, we identify th e objec t in virtue of ce rtain of it s character-

istics. So now it seems as if th e rules for a proper nam e mu st

solliehow be logically tied to particular characteristics of the

obiect

n

such a wav that the name has a sense as well as a

reierence indeed, it seems it could no t have a reference unless i t

did hav e a sense, for how, unless th e name has a sense, is it t o b e

correlated with th e obiect

Suppose someone answers this argument as follows : The

characteristics located in teaching the name are no t t he rules for

using the proper name : they are simply pedagogic devices

employed in teaching the name to someone who does not know

how to use it. Once our student has identified th e object to

which the name applies he can forget or ignore these various

descriptions by m eans of which he identified the object, for they

are not p art of th e sense of th e name : th e name does not h ave a

sense kuppose, for example, th a t we'teach t he name ' Aristotle '

by explaining that it refers to a Greek philosopher born in

Stagira, and suppose th a t our student continues to use th e nam e

correctly, that he gathers more information about Aristotle,

and so on. Le t us suppose it is discovered later on th a t Aristotle

was no t born in Stagira at all, bu t in Thebes. We will no t now

say th a t th e meaning of th e name has changed, or th a t Aristotle

did not really exist a t all. I n short, explaining th e use of a name

by citing characteristics of the object is not giving the rules

for the name, for th e rules contain no descriptive con tent a t all.

They simply correlate the name to the object independently of

any descriptions of it.

B ut is th e argument convincing Suppose most or even all

of our present factual knowledge of Aristotle proved to be true

of no one a t all, or of several people living in sca ttered countries

and in different centuries Would we no t say for th is reason

th at A ristotle did not exist after all, and th a t th e name, though

i t has a conventional sense. refers to no one a t all On th e above

account, if anyone said that Aristotle did not exist, this must

simply be another way of saying th a t Aristotle denoted no

objects, a nd nothing more

;

bu t if anyone did say th a t Aristotle


5/9

169

ROPER NAMES

did not exist he might mean much more than simply that the

name does no t denote any0ne.l If, for example, we challenged

his stateme nt by pointing out th a t a m an named Aristotle

lived in Hoboken in 1903, he would no t regard this a s a relevant

countercharge. We sav of Cerberus an d Zeus th a t neither of

them ever G isted , without meaning t h a t no object ev er bore

these names, bu t only th a t certain kinds descriptions) of objects

never existed an d bore these names. So now it looks as thou gh

proper names do have a sense necessarily but have a reference

only contingently. They begin to look more and more like

shorthand and perhaps vague descriptions.

Let us summarise the two conflicting views under consider-

ation th e first asserts th a t proper ni m es have essentially a

reference bu t no t a sense-proper nam es denote but do no t

connote th e second asserts th a t the v hav e essentiallv a sense

an d only contingently a reference-they refer only on th e con-

dition th a t one and only one object satisfies their sense.

These two views are paths leading to divergent and hoary

metaphysical systems. The first leads to ultim ate ob jects of

reference, th e sub stances of t he scholastics an d t he Gegenstande

of the

Tractatus.

The second leads to th e iden tity of indiscern-

i b l e ~ ,and variables of quantification as th e only referential term s

in th e language. The subject-predicate struc ture of th e language

suggests that the first must be right, but the way we use and

teach the use of proper names suggests that it cannot be right

a philosophical problem.

Le t us begin b y examining t h e second.

If

i t is asserted th at

every proper nam e has a sense, it m ust be legitimate to demand

of any name, W hat is i ts sense If i t is asserted th a t a

proper nam e is a kind of sh orthand description th en we ought to

be able to present the description

in

place of the proper name.

B ut how are we to proceed with this If we tr y to present a

complete description of t he object a s th e sense of a proper nam e,

odd consequences would ensue, e g th at any true statement about

the object using the name as subject would be analytic, any

false one self-contradictory, that the meaning of the name

and perhaps th e identity of th e object) would change every time

there was any change a t all

n

th e object, th a t th e name would

have different meanings for different people, etc. So suppose

we ask w ha t are th e necessary a nd sufficient conditions for apply-

ing a particular name to.,a particular object. Suppose for th e

sake of a rgum ent th a t we have independent means for locating

a n object then w ha t are th e conditions for applying a name to

Cf.

Wittgenstein

Philosophical Investigations

para.

79.


6/9

17

J

R

SEARLE :

i t ; wh at are th e conditions for saying, e g This is Aristotle ?

A t first sight these conditions seem to be simply th a t th e object

must be identical with an object originally christened by this

name, so th e sense of th e name would consist in a s tate m en t or

set of statements asserting the characteristics which constitute

this identity . Th e sense of This is Aristo tle might be,

This object is spatio-temporally continuous with an object

originally named ' Aristotle ' . But this will not suffice, for,

as was already suggested, th e force of Aristotle is greater th an

th e force of identical with a n object named ' Aristotle ' , for

no t just an y object named Aristotle will do. Aristotle

here refers to a particular object named Aristotle

,

not t o any.

Named ' Aristotle ' s a universal term , bu t Aristotle , is a

proper name, so Th is is named 'Aristotle' is a t best a necessary

b u t not a sufficient condition for th e t ru th of This is Aristotle

?

Briefly and trivially, i t is not th e identity of this w ith an y object

named Aristotle , but rather its identity with Aristotle that

constitutes t he necessary a nd sufficient conditions for the t ru th

of This is Aristotle .

Perhaps we can resolve the conflict between the two views of

th e nature of proper nam es by asking what is th e unique function

of proper names in our language. To begin with, the y mostly

refer or purport to refer to particular objects

;

but of course

other expressions, definite descriptions and dem onstratives,

perform this function as well. W ha t then is th e difference

between proper names and other singular referring expressions ?

.

Unlike dem onstratives, a proper name refers without pre-

supposing any stage settings or an y special contextual conditions

surrounding th e utte rance of th e expression. Unlike definite

descriptions, they do not in general

specify

any characteristics

a t all of the objects to which they refer. Sc ott refers to th e

same object as does th e autho r of averley , but Scott

specifies none of it s characteristics, whereas th e au thor of

averley refers only in virtue of t he fact th a t it does specify a

charac teristic. Let us exam ine this difference more closely.

Following Strawson we m ay say th a t referring uses of bo th

proper names and definite descriptions presuppose the existence

of one an d only one object referred to. B ut as a proper name

does not in general specify any characteristics of the object

referred to, how th en does it bring th e reference off ? How is a

connection between name and object ever set up ? This, which

seems th e crucial question, wa nt to answer by saying th a t

though proper names do not normally assert or specify any

On

Referring

,

MIND, 1950


7/9

7ROP R

NAMES

characteristics theif referring uses nonetheless presuppose that

th e object to which the y pu rport to refer has certain character-

istics. B u t which ones Suppose we ask th e users of th e nam e

Aristotle to stat e wh at the y regard as certain essential an d

established fac ts abou t him. Their answers would be a set of

uniquely referring descriptive statem ents . Now what I. am

arguing is th a t th e descriptive force of This is Aristotle is to

assert th a t a sufficient bu t so far unspecified num ber of these

sta tem en ts ar e tr ue of th is object. Therefore referring uses of

Aristotle presuppose th e existence of a n object of whom a

sufficient but so far unspecified number of these statements are

true. To use a proper nam e referringly is to presuppose th e

truth of certain uniquely referring descriptive statements but

i t is not ordinarily t o assert these statem ents or even to indicate

which exactly are presupposed. And herein lies most of the

difficulty. The question of what constitutes th e criteria for

Aristotle is generally left open indeed it seldom in fact arises

and when i t does arise i t is we th e users of th e name who decide

more or less arbitr arily w ha t these criteria shall be. If for

example of t he characteristics agreed to be tr ue of Aristotle

half should be discovered to be t ru e of one man and half tru e of

another which would we say was Aristotle Neither

The

auestion is no t decided for us in advance.

But is this imprecision as to what characteristics exactly

con stitute th e necessary and sufficient conditions for applying a

proper name a mere acciden t a product of linguistic slovenli-

.new Or does it derive from th e functions which proper names

L A

perform for us To ask for the criteria for applying th e nam e

Aristotle is to ask n th e formal mode what Aristotle is ; i t is to

ask for a set of iden tity criteria for th e object Aristotle. W ha t

is Aristotle

and W hat are the criteria for applying the name

'

Aristotle

'

ask the same question th e former in th e material

mode an d the latter

in

th e formal mode of speech. So if we

came to agreement in advance of using the name on precisely

what characteristics constituted th e iden tity of Aristotle our

rules for using the nam e would be precise. B u t this precision

would be achieved only a t th e cost of entailing some specific

predicates by an y referring use of th e name. Indeed th e name

itself would become superfluous for it would become logically

equivalent to this set of descriptions. B ut if this were the case

we would be in the position of only being able to refer to an

object by describing it. Whereas in fact this is just w hat th e

institution of proper names enables us to avoid and what dis-

tinguishes proper names from descriptions. If th e criteria for


8/9

72 J R

SE RLE

:

proper names were in all cases quite rigid and specific then

a

proper name would be nothing more than a shorthand for these

criteria, a proper name would function exactly like an elaborate

de fh ite description. B u t th e uniaueness and immense pragmatic

-I

convenience of proper nam es in our language lie precisely in th e

fact th a t th ey enable us to, refer publicly to objects without

being forced to raise issues and come to agreement on what

descriptive characteristics exactlv constitute th e identitv of th e

object:

They function not as deicriptions, bu t as pegs i n which

to hang descriptions. Thus th e looseness of th e criteria for

proper names is a necessary condition for isolating the referring

function from th e describing function of language.

To put the same point differently, suppose we ask, " Why do

we have proper nam es a t all Obviously, to refer to indivi-

duals. " Yes. bu t desc ri~ tio ns ould do th at for us. B ut only

a t the cost of specifying :dentity conditions every tim e referenck

is made : suppose we agree to drop " Aristotle an d use, say,

" the teacher of Alexander ", then it is a necessary truth that

th e man referred to is Alexander's teacher-but i t is

a

contingent

fact th a t Aristotle ever went into pedagogy (though am sug&st-

ing it is a necessary fact that Aristotle has the logical sum,

inclusive disjunction, of properties commonly attr ibu ted to him :

an y individual no t having a t least some of these properties could

no t be Aristotle).

Of course it should n ot be th oug ht t h a t th e only sort of loose-

ness of identitv criteria for individuals is th a t which have

,

described as pe iulia r to proper names. Referring uses of definite

descriptions may raise problems concerning identity of quite

different sorts. This is especially true of pa st tense definite

descriptions. " This is the ma n who taug ht Alexander ma y

be said to entail, e g that this object is spatio-temporally con-

tinuous with the man teaching Alexander at another point in

space-time : but someone might also argue that this man's

spatio-temporal continuity is a con tingent characteristic and not

a n iden tity criterion. And th e logical na ture of the connection

of such characteristics with t he man's identity may again be loose

an d undecided in advance of dispute. B ut th is is quite another

dimension of looseness th an th a t which cited as the looseness of

the criteria for applying proper names and does not affect the

distinction in function be tye en definite descriptions and proper

names, viz. that definite descriptions refer only in virtue of the

fact t h a t th e criteria are no t loose in t h e original sense, for they

refer by telling us what th e object is B u t proper names refer

without so far raising th e issue of what t he object is.


9/9

73ROP R

NAMES

We are now in a position to explain how it is th a t Aristotle

has a reference but does not describe, and yet the statement

Aristotle never existed says more th an th a t Aristotle

was never used to refer to any object.

The statement asserts

that a sufficient number of the conventional presuppositions,

descriptive sta tem en ts, of referring uses of Aristotle ar e false.

Preciselv which statements are asserted to be false is not vet

clear, for what precise conditions constitute the criteria for

applying Aristotle is no t yet laid down by th e language.

We can now resolve our paradox : does a proper name have a

sense

If this asks whether or not proper names are used to

describe or specify characteristics of objects, th e answer is no

.

B ut if i t asks whether or no t proper names a re logically connected

with characteristics of th e object to which th ev refer. t he answer

is yes, in a loose sort of way . (Th is shows fn pa rt'th e poverty

of a rigid sense-reference, denotation-connotation approach to

problems in the theory of meaning.)

We might clarify these points by comparing paradigmatic

proper names with degenerate proper nam es like The Bank of

England

.

Fo r these latte r, i t seems the sense is given

as straightforwardly as in a definite description; the pre-

suppositions, as i t were, rise to th e surface. And a proper

name may acquire a rigid descriptive use without having

th e verbal form of a desc ription: God is just, omnipotent,

omniscient, etc., b y dejinition for believers. Of course th e form

may mislead us ; th e Holy R om an E mpire was neither holy, nor

%oman, etc., but it was nonetheless the Holy Roman Empire.

Again it may be conventional to nam e only girls Martha ,bu t

if

I

name m y son Martha

I

may mislead, bu t

I

do not lie.

Now reconsider our original iden tity , Tully Cicero .

A statement made using this sentence would, I suggest, be

analytic for most people

;

the same descriptive presuppositions

are associated with each nam e. B ut of course if the descriptive

presuppositions were different it might be used to make a

synthetic statement

;

it might even advance a historical dis-

covery of the first importance.

Univ ersit y of xford

Date post:	07-Jul-2018
Category:	Documents
Upload:	isabela-tavares
View:	224 times
Download:	0 times

SEARLE, J. Proper Names.

Documents