T H E JOURNAL OF PHILOSOPHY

V O L U M E LXIV, NO. 2 0 , OCTOBER 2 6 , I 9 6 7

4- •

INTENTIONAL IDENTITY *
A W E L L - K N O W N epigram might serve both as a motto for this
l \ paper and (at least arguably) as a serious example of its topic.
J L X . For if a speaker asserts:
(1) 1 saw a man on the stair yesterday at time t^, and I saw him (the same
man) on the stair again today at time ^2
in such a sense that it would be compatible with his also saying:
(2) No man but me was on the stair either yesterday at time or today
at ume
then the identity he means to ascribe in (1) to the man on the stair
is what I am calling intentional identity, rather than actual identity.
I am sure that in fact this is a proper example of what I shall be dis-
cussing; but the intentionality of the verb *to see', though I think it
can be established, is bound up with matters of controversy that I do
not wish to touch upon here; so I pass quickly on from seeing to
believing.
Etymology is more often a hindrance than a help in philosophy,
but in this case it may be a help to remember the metaphor that under-
lies the words 'intention' and 'intentional': 'intendo arcum in . . .',
'I draw a bow at . . .' For a number of archers may all point their
arrows at one actual target, a deer or a man (real identity); but we
may also be able to verify that they are all pointing their arrows the
same way, regardless of finding out whether there is any shootable
object at the point where the lines of fire meet (intentional identity).
We have intentional identity when a number of people, or one person
on different occasions, have attitudes with a common focus, whether
or not there actually is something at that focus.
Suppose a reporter is describing an outbreak of witch mania, let us
say in Gotham village:
* To be presented in an APA symposium of the same title, December 29, 1967.
CoERraentatois will be L. Jonathan Cohen (Columbia University) and James F. Ross
(I'niversity of Pennsylvania).
627
628 T H E JOURNAL OF PHILOSOPHY

(3) Hob thinks a witch has blighted Bob's mare, and Nob wonders whedier
she (the same witch) killed Cob's sow.
Quine has distinguished opaque and transparent ways of constru-
ing indirect-speech clauses, but neither sort of construction will give
an appropriate sense to (3). For if the indirect-speech clauses in (3)
are construed opaquely, then each clause must stand on its own syn-
tactically; this is graphically shown by Quine's way of enclosing such
clauses in square brackets; and Quine forbids syntactical liaisons, like
the binding of variables, to cross this barrier. But on the face of it we
have in (3) a pronoun, 'she' or *the same', bound to an antecedent, *a
witch', that lies outside the clause containing the pronoun; so unless
this prima facies can be discounted (and I shall presently consider
some possible ways of doing that), the clauses in (3) cannot be con-
strued opaquely.
(In speaking, as I have just done, of a pronoun as being bound to
its antecedent, I am deliberately adopting a jargon that serves to ex-
press an important syntactical insight of Quine's: that certain uses of
pronouns correspond very closely to the bound variables of symbolic
logic and that the relation of these pronouns to their antecedents cor-
responds to the binding of variables by quantifiers and other oper-
ators. I know that some philosophers and some linguists are opposed
to Quine's view, but I am disposed to attach far more weight to the
way the view is confirmed by the detailed working out of numerous
and varied examples, such as occur in Quine's works and mine. In pre-
vious writing I have used the medieval term 'relative pronouns'; but I
now think it is awkward, except in writings on the history of logic, to
divest this term of its familiar grammatical sense; and I propose that
the familiar jargon of binding, scope, and so on should be extended
from symbolic language to appropriate pieces of the vernacular.)
On the other hand, there is no obvious way of construing the indi-
rect speech in (3) transparently. We might try:
(4) As regards some witch. Hob thinks she has blighted Bob's mare, and
Nob wonders whether she killed Cob's sow.
But (4) would express (what the speaker took to be) the real not the
intentional, identity of a witch; and (unlike the wise men of Gotham)
our reporter might mistakenly believe that there are no witches (not
just that spells against livestock are ineffectual, but that nobody ever
casts them); in that case he might, and could with consistency, assert
(3) and deny (4). Nor do we fare any better with something like:
(5) As regards somebody. Hob thinks that she is a witch and has blighted
Bob's mare, and Nob wonders whether she killed Cob's sow.
 INTENTIONAL IDENTITY 629

For (5) would imply that Hob and Nob had some one person in mind
as a suspected witch; whereas it might be the case, to the knowledge of
our reporter, that Hob and Nob merely thought there was a witch
aroimd and their suspicions had not yet settled on a particular person.
The difference between (3) and (4) corresponds to a difference that
was much discussed by Jean Buridan and illustrated with numerous
examples; (3) and (4) correspond respectively to the first and the
second member of such pairs as the following:
(6) I owe John a horse
(7) There is some horse that I owe John
(8) John wants a stamp
(9) There is some stamp that John wants
The difference is gropingly brought out by saying *Just a so-and-so,
not necessarily a definite so-and-so' for examples like (6) and (8), and T
mean a definite so-and-so' for examples like (7) and (9). Similarly, it
would be quite natural to insert the adjective 'definite' before
'witch' in (4), or again after 'somebody' in (5). No clue is to be found
here to the real logical difference; for of course witches, horses, and
stamps do not come in two species, the definite ones and the indefinite
ones. Frege indeed remarked that 'referring to an indefinite so-and-so'
often really means 'referring indefinitely to a so-and-so', and we might
try to interpret similarly the phrase 'a definite so-and-so'; but there is
no definite reference made by the "some" phrase in (4), (7), or (9), so
this is no clue either. Buridan's own attempt to characterize the differ-
ence is made in hopelessly obscure terminology. He speaks of the ob-
ject phrase in propositions like (6) and (8) as calling up, appellans, a
certain ratio. But the criterion of identity for a ratio remains quite ob-
scure; and the semantical term 'appellatio' neither has a consistent
technical use in medieval logic generally (different authors use it dif-
ferently), nor is carefully explained in Buridan's own text. The most
one can say is that appellatio of a ratio in Buridan is something like
a term's having ungerade Bedeutung in Frege—but of course Frege's
theory is very obscure too.
A l l the same, I am strongly inclined to think that Buridan was here
on to an important logical distinction—and that the same difference
of logical structure is involved in all these contrasted pairs; for it is
easy to catch on to what the intended difference is and transfer the
learning to a fresh example; we have not got to learn it anew for each
intentional verb or context. Only at present we (or I at least) are not
able to say what the difference is. We are in the same position as the
medievals were in about the contrasted pairs they used to show the
630 T H E JOURNAL OF PHILOSOPHY

differences between suppositio confusa (the first members of the fol-

lowing pairs) and suppositio determinata (the second members):
(10) In order to see, I need an eye
(11) There is an eye that I need to see with
(12) There always has been a man alive
(13) There is a man who has always been alive
This difference can be clearly explained in modern logic as a differ-
ence in the order of application of two quantifiers or other operators.
We have no such clear view of the other problem. Even the difference
between this pair:
(14) Hob thinks some women are witches
(15) There are some women Hob thinks are witches
is not clearly explicable in terms of a difference in scope for the quan-
tified phrase 'some women':
(16) Hob thinks that, as regards some women x and y, x and y are witches
(17) As regards some women x and y, Hob thinks x and y are witches
For Quine has raised some difficulties about the quantification in (17)
and the like; though others (Hintikka, Prior, and myself) argue that
they need not be insuperable. But (3) raises an even worse difficulty:
a pronoun in one indirect-speech clause is on the face of it bound to
a quantified phrase in another such oblique context; the scope of the
quantified phrase thus seems both to lie wholly within the earlier
oblique context and to cover something in the later context. I can-
not even sketch a structure of operators that would make good logical
sense of this; I go on to consider a couple of ways of evading the diffi-
culty.
First, it might be suggested that 'she' in (3) ought not to be glossed
as 'the same witch', but should rather be regarded as an anaphoric
substitute (what I have called a pronoun of laziness) to avoid repeti-
tious language; (3) would then be a substitute for something like this:
(18) Hob thinks a witch has blighted Bob's mare, and Hob wonders
whether the witch who blighted Bob's mare killed Cob's sow
This suggestion is easily dismissed: for our reporter might be justi-
fied in asserting (3) if he had heard Hob say 'The witch has blighted
Bob's mare' and heard Nob say 'Maybe the witch killed Cob's sow',
even if Hob had not thought or said anything about Cob's sow nor
Nob about Bob's mare. Of course our reporter would somehow have
to know that when they used the words 'the witch' Hob and Nob
meant to refer to the same person. But this would not necessarily
 INTENTIONAL IDENTITY 631

mean his knowing that there was some ("definite") person to whom
they both meant to refer; rather, the italicized phrase is itself yet
another example of our problematic intentional identity.
This in fact points up the importance of intentional identity as a
problem in the philosophy of logic. We very often take ourselves to
know, when we hear the discourse of others, that they are meaning to
refer to some one person or thing—and that, without ourselves being
able to identify this person or thing, without our even being certain
that there really is such a person or thing to identify. What we are
claiming to know in such cases—let alone, whether the claim is
justified—must remain obscure so long as intentional identity is
obscure.
There is another objection to this proposed method of analyzing
away an intentional identity in (3). Although I see no reason to
doubt that:
(19) The witch who blighted Bob's mare killed Cob's sow
is analyzable as:
(20) Just one witch blighted Bob's mare and she killed Cob's sow
it seems doubtful whether these two are mutually replaceable salva
vertitate in a context like *Nob wonders whether*. If we prefix *Nob
wonder whether' to (19), the result seems to be analyzable, not as:
(21) Nob wonders whether (the following is the case:) just one witcli
blighted Bob's mare, and she killed Cob's sow
but rather in some such way as this:
(22) Nob assumes that just one witch blighted Bob's mare, and Nob
wonders whether she (that same witch) killed Cob's sow
It is not easy to be sure about this, because it is easy to confound the
analysis (21), which I have just rejected, with the following:
(23) Nob wonders whether just one witch blighted Bob's mare and (Nob
wonders whether) she killed Cob's sow
Objections to (23) as an analysis of 'Nob wonders whether the witch
who blighted Bob's mare killed Cob's sow' naturally do not carry over
to (21). (If the words in parentheses are omitted, as they might easily
be, from (21) and (23), we get a verbally identical sentence that could
mean either; hence the likelihood of confusing the two.) But even
allowing for this, I think (22) is the right sort of analysis rather than
(21). Now if so, the analysis of the second conjunct in (18) would intro-
duce intentional identity over again; for (22) manifestly does so—'she'
or 'that same' is bound to an antecedent, 'just one witch', in another
632 T H E JOURNAL OF PHILOSOPHY

oblique clause. In that case (18) is quite useless as a way of getting rid
of intentional identity.
This objection to (18), if it is sound, also serves against another pos-
sible way of trying to deal with (3): namely, to say that (3) is true
just in case the following is true for some suitable interpretation
of'F:
(24) Hob thinks that the (one and only) witch that is F has blighted Bob's
mare, and Nob wonders whether the witch that is F killed Cob's sow
For supposing we have such a reading of *F, we should have exactly
the same difl&culty over the second conjunct of (24) as over the second
conjunct of (18): it looks as though intentional identity, which was
to be analyzed away, recurs in the analysis.
There is further trouble about (24). On the face of it, if (24) is true
for some suitable reading of *F', then (3) is true; there are, I think,
some diflSculties about this, but I waive them. Anyhow, it is very
doubtful whether the converse holds; whether, if (3) is true, (24) is
true for some suitable reading of For, as I said, the truth of (3)
could be established (in a suitable set of background circumstances)
if Hob said "The witch has blighted Bob's mare" and Nob said
"Maybe the witch killed Cob's sow"—provided that, in using the
phrase *the witch'. Hob and Nob meant to refer to the same person.
Now is it in truth necessary, if Hob and Nob are to mean to refer to
the same person as "the witch," that they should both have some one
definite description actually in mind, or even, one producible from
each of them by a suitable technique of questioning? This appears to
me to stand or fall with the corresponding theory, held by Russell and
by Frege (cf. his article "Der Gedanke"), that any ordinary proper
name is used equivocally if it does not go proxy for some one definite
description; and in spite of these great names, such a theory seems to
me extremely ill-founded and implausible.
I have only stated a problem, not tried to solve it; it seems to me
interesting and important. It surely brings out how much is obscure
in the logic of quite simple constructions of ordinary language. Straw-
son has spoken of ordinary language as a maze whose paths we tread
unhesitatingly. 'Maze' suggests something in a gentleman's formal
garden, with neat box hedges and a discoverable plan; to my mind it
would be better to speak of a clearing in a jungle, whose paths are only
kept free if logicians work hard with the machete, and where he who
does not hesitate may none the less be lost.
p. T . GEACH
The University of Leeds