Chisholm - On Metaphysics PDF

On Metaphysics
This page intentionally left blank
On Metaphysics
Roderick M. Chisholm
University of Minnesota Press Minneapolis
Copyright 1989 by the Regents of the University of Minnesota

All rights reserved. No part of this publication may be
reproduced, stored in a retrieval system, or transmitted, in
any form or by any means, electronic, mechanical, photocopying,
recording, or otherwise, without the prior written
permission of the publisher
Published by the University of Minnesota Press
2037 University Avenue Southeast, Minneapolis, MN 55414.
Printed in the United States of America.
Library of Congress Catalog Card Number 89-40290
ISBN 0-8166-1767-8
The author thanks the publishers of the following journals and collections for permission
to republish essays in this book. New York University Press, for Chap, 1, "Responsibility and
Avoidability," from Sidney Hook (ed.), Determinism and Freedom in the Age of Modern
Science (1958). Department of Philosophy, University of Kansas (Lawrence), for Chap. 2,
"Human Freedom and the Self," from The Lindsay Lecture (1984). Nous, 1 (1967), for Chap.
3, "Identity through Possible Worlds". State University of New York Press (Albany), for
Chap. 4, "Identity through Time," from Howard E. Kiefer and Milton K. Munitz (eds.), Language, Belief and Metaphysics (1970). Midwest Studies in Philosophy, 11. (1986), for Chap.
5, "Possibility without Haecceity", D. Reidel Publishing (Dordrecht), for Chap. 6, "Coming
into Being and Passing Away: Can the Metaphysician Help?" from S. F. Spicker and H. T.
Englehardt, Jr. (eds.), Philosophical Medical Ethics: Its Nature and Significance (1977). Review of Metaphysics, 26 (1973), for Chap. 7, "Parts as Essential to their Wholes," Grazer
Philosophische Studien, 20 (1983), for Chap. 8, "Boundaries" (originally called "Boundaries
as Independent Particulars"), and 25/26 (1985/1986) for Chap. 16, "States and Events" (originally called "On the Positive and Negative States of Things"). MIT Press, for Chap. 9, "Scattered Objects," from Judith Jarvis Thomson (ed.), On Being and Saying: Essays for Richard
Cartwright (1987). Philosophical Studies, 43 (1983), for Chap. 10, "The Nature of the Psychological." The Monist, 69 (1987), for Chap. 11, "Presence in Absence." Holder-PichlerTempsky (Vienna), for Chap. 12, "Questions about Minds," which was the preface to
Proceedings of the Ninth International Wittgenstein Symposium (1985); for Chap. 15, "Properties Intentionally Considered," from Proceedings of the Sixth International Wittgenstein
Symposium (1982); and for Chap. 17, "The Self in Austrian Philosophy," from J. C. Nyiri
(ed.), From Balzano to Wittgenstein: The Tradition of Austrian Philosophy (1986). Philosophical Exchange, 2 (1979), for Chap. 13, "Is There a Mind-Body Problem?" Synthese, 61
(1984), for Chap. 14, "The Primacy of the Intentional."
The University of Minnesota
is an equal-opportunity
educator and employer.
Contents
Introduction
vii
Part I. Freedom and Determinism

1. Responsibility and Avoidability
2. Human Freedom and the Self
3
5
Part
3.
4.
5.
6.
II. Coming into Being, Persisting, and Passing Away

Identity through Possible Worlds
19
Identity through Time
25
Possibility without Haecceity
42
Coming into Being and Passing Away:
Can the Metaphysician Help?
49
Part
7.
8.
9.
III. Parts and Wholes

Parts as Essential to their Wholes
Boundaries
83
Scattered Objects
90
Part
10.
11.
12.
13.
14.
IV. The Mental

The Nature of the Psychological
Presence in Absence
107
Questions about Minds
114
Is There a Mind-Body Problem?
The Primacy of the Intentional
Part
15.
16.
17.
18.
V. An Intentional Approach to Ontology

Properties and States of Affairs Intentionally Considered
States and Events
150
The Self in Austrian Philosophy
156
The Categories
162
Index
171
65
99
119
129
141
Introduction
Two assumptions about the nature of metaphysics are presupposed by this book.
Neither is remarkable in itself. But in contemporary western philosophy each is
usually associated with the rejection of the other. One assumption is that the problems of philosophy are extraordinarily difficult and can be solved only by the
responsible application of what Russell called "honest toil." The other is the view
of Leibniz and Brentano, according to which reflection on the self and on what
it is to think provides us with the key to understanding the fundamental categories
of reality.
The categories discussed here are: substance and attribute; part and whole;
identity, persistence, and change; boundaries and limits; coming into being and
passing away; reference and the nature of the psychological; and the self. The
"intentional" approach to these topics is complemented by an "internal" approach
to the theory of knowledge. The latter is exemplified by my Foundations of Knowing in the present series; it is further developed in the Third Edition of my Theory
of Knowledge (Englewood Cliffs, N.J.: Prentice-Hall, 1989).
Many of the essays appear as they were originally published. Some are corrected in minor respects and certain omissions have been made. "The Primacy
of the Intentional" and "States and Events" have been completely rewritten. "The
Categories" is new; it was written for this book and represents a kind ofZusammenfassung of the ontology here set forth.
vii
Parti
Freedom and Determinism
Responsibility and Avoidability

Edwards and Hospers hold that there is an important sense in which we may be
said not to be morally responsible for any of our acts or choices. I propose the
following as an explicit formulation of their reasoning:
1. If a choice is one we could not have avoided making, then it is one for
which we are not morally responsible.
2. If we make a choice under conditions such that, given those conditions,
it is (causally but not logically) impossible for the choice not to be made,
then the choice is one we could not have avoided making.
3. Every event occurs under conditions such that, given those conditions,
it is (causally but not logically) impossible for that event not to occur.
4. The making of a choice is the occurrence of an event.
.'.5. We are not morally responsible for any of our choices.
If we wish to reject the conclusion (5)and for most of us (5) is difficult to
acceptwe must reject at least one of the premises.
Premise (1), I think, may be interpreted as a logical truth. If a man is responsible for what he did, then we may say, "He could have done otherwise." And
if we may say, "He couldn't help it," then he is not responsible for what he did.
Many philosophers would deny (2), substituting a weaker account of avoidability. A choice is avoidable, they might say, provided only it is such that, if the
agent had reflected further, or had reflected on certain things on which in fact he
did not reflect, he would not have made the choice. To say of a choice that it
"could not have been avoided," in accordance with this account, would be to say
that, even if the agent had reflected further, on anything you like, he would all
the same have made the choice. But such conditional accounts of avoidability
("An act or choice is avoidable provided only it is such that, if the agent were to
do so-and-so, the act or choice would not occur") usually have this serious defect:
3
RESPONSIBILITY AND AVOIDABILITY
the antecedent clause ("if the agent were to do so-and-so") refers to some act or
choice, or to the failure to perform some act or to make some choice; hence we
may ask, concerning the occurrence or nonoccurrence of this act or choice,
whether or not it is avoidable. Thus one who accepted (5) could say that, if the
agent's failure to reflect further was itself unavoidable, his choice was also unavoidable. And no such conditional account of avoidability seems adequate to the
use of "avoidable" and "unavoidable" in questions and statements such as these.
If we accept a conditional account of avoidability, we may be tempted to say,
of course, that it would be a misuse of "avoidable" to ask whether the nonoccurrence of the antecedent event ("the agent does so-and-so") is avoidable. But the
philosopher who accepts (5) may well insist that, since the antecedent clause
refers to an act or a choice, the use of "avoidable" in question is not a misuse.
What, then, if we were to deny (3)? Suppose that some of our choices do not
satisfy (3)that when they are made they are not made under any conditions such
that it is (causally) impossible (though logically possible) for them not to be made.
If there are choices of this sort, then they are merely fortuitous or capricious. And
if they are merely fortuitous or capricious, if they "just happen," then, I think,
we may say with Blanshard that we are not morally responsible for them. Hence
denying (3) is not the way to avoid (5).
We seem confronted, then, with a dilemma: either our choices have sufficient
causal conditions or they do not; if they do have sufficient causal conditions they
are not avoidable; if they do not, they are fortuitous or capricious; and therefore,
our choices are either unavoidable or fortuitous, we are not morally responsible
for them.
There are philosophers who believe that by denying the rather strangesounding premise (4) we can escape the dilemma. Insisting on something like "the
primacy of practical reason," they would say that since we are certain that (5) is
false we must construct a metaphysical theory about the self, a theory denying
(4) and enabling us to reconcile (3) and the denial of (5). I say "metaphysical" because it seems to be necessary for the theory to replace (4) by sentences using
such terms as "active power," "the autonomy of the will," "prime mover," or
"higher levels of causality "-terms designating something to which we apparently
need not refer when expressing the conclusions of physics and the natural
sciences. But I believe we cannot know whether such theories enable us to escape
our dilemma. For it seems impossible to conceive what the relation is that, according to these theories, holds between the "will," "self," "mover," or "active
power," on the one hand, and the bodily events this power is supposed to control,
on the other-the relation between the "activities" of the self and the events described by physics.
I am dissatisfied, then, with what philosophers have proposed as alternatives
to premises (1) through (4) above, but since I feel certain that (5) is false I also
feel certain that at least one of the premises is false.
Human Freedom and the Self

"A staff moves a stone, and is moved by a hand, which is moved by a man."
(Aristotle, Physics, 256a.)
1. The metaphysical problem of human freedom might be summarized in the following way: Human beings are responsible agents; but this fact appears to conflict
with a deterministic view of human action (the view that every event that is involved in an act is caused by some other event); and it also appears to conflict
with an indeterministic view of human action (the view that the act, or some event
that is essential to the act, is not caused at all.) To solve the problem, I believe,
we must make somewhat far-reaching assumptions about the self or the agent
about the man who performs the act.
Perhaps it is needless to remark that, in all likelihood, it is impossible to say
anything significant about this ancient problem that has not been said before.1
2. Let us consider some deed, or misdeed, that may be attributed to a responsible agent: one man, say, shot another. If the man was responsible for what he
did, then, I would urge, what was to happen at the time of the shooting was something that was entirely up to the man himself. There was a moment at which it
was true, both that he could have fired the shot and also that he could have
refrained from firing it. And if this is so, then, even though he did fire it, he could
have done something else instead. (He didn't find himself firing the shot 'against
his will', as we say.) I think we can say, more generally, then, that if a man is
responsible for a certain event or a certain state of affairs (in our example, the
The Lindley Lecture, 1964, pp. 3-15. Copyright 1964 by the Department of Philosophy,
University of Kansas. Reprinted by permission of the author and of the Department of Philosophy
of the University of Kansas, Lawrence.
HUMAN FREEDOM AND THE SELF
shooting of another man), then that event or state of affairs was brought about
by some act of his, and the act was something that was in his power either to perform or not to perform.
But now if the act which he did perform was an act that was also in his power
not to perform, then it could not have been caused or determined by any event
that was not itself within his power either to bring about or not to bring about.
For example, if what we say he did was really something that was brought about
by a second man, one who forced his hand upon the trigger, say, or who, by
means of hypnosis, compelled him to perform the act, then since the act was
caused by the second man it was nothing that was within the power of the first
man to prevent. And precisely the same thing is true, I think, if instead of referring to a second man who compelled the first one, we speak instead of the desires
and beliefs which the first man happens to have had. For if what we say he did
was really something that was brought about by his own beliefs and desires, if
these beliefs and desires in the particular situation in which he happened to have
found himself caused him to do just what it was that we say he did do, then since
they caused it, he was unable to do anything other than just what it was that he
did do. It makes no difference whether the cause of the deed was internal or external; if the cause was some state or event for which the man himself was not
responsible, then he was not responsible for what we have been mistakenly calling his act. If a flood caused the poorly constructed dam to break, then, given the
flood and the constitution of the dam, the break, we may say, had to occur and
nothing could have happened in its place. And if the flood of desire caused the
weak-willed man to give in, then he, too, had to do just what it was that he did
do and he was no more responsible than was the dam for the results that followed.
(It is true, of course, that if the man is responsible for the beliefs and desires that
he happens to have, then he may also be responsible for the things they lead him
to do. But the question now becomes: is he responsible for the beliefs and desires
he happens to have? If he is, then there was a time when they were within his
power either to acquire or not to acquire, and we are left, therefore, with our
general point.)
One may object: But surely if there were such a thing as a man who is really
good, then he would be responsible for things that he would do; yet, he would
be unable to do anything other than just what it is that he does do, since, being
good, he will always choose to do what is best. The answer, I think, is suggested
by a comment that Thomas Reid makes on an ancient author. The author had said
of Cato, 'He was good because he could not be otherwise', and Reid observes:
This saying, if understood literally and strictly, is not the praise of Cato, but of
his constitution, which was no more the work of Cato than his existence'.2 If Cato
was himself responsible for the good things that he did, then Cato, as Reid suggests, was such that, although he had the power to do what was not good, he exercised his power only for that which was good.
All of this, if it is true, may give a certain amount of comfort to those who
are tender-minded. But we should remind them that it also conflicts with a famil-
iar view about the nature of Godwith the view that St. Thomas Aquinas expresses by saying that 'every movement both of the will and of nature proceeds
from God as the Prime Mover'.3 If the act of the sinner did proceed from God
as the Prime Mover, then God was in the position of the second agent we just
discussedthe man who forced the trigger finger, or the hypnotistand the sinner, so-called, was not responsible for what he did. (This may be a bold assertion,
in view of the history of western theology, but I must say that I have never encountered a single good reason for denying it.)
There is one standard objection to all of this and we should consider it briefly.
3. The objection takes the form of a stratagemone designed to show that determinism (and divine providence) is consistent with human responsibility. The
stratagem is one that was used by Jonathan Edwards and by many philosophers
in the present century, most notably, G. E. Moore.4
One proceeds as follows: The expression
(a) He could have done otherwise,
it is argued, means no more nor less than
(b) If he had chosen to do otherwise, then he would have done otherwise.
(In place of 'chosen', one might say 'tried', 'set out', 'decided', 'undertaken', or
'willed'.) The truth of statement (b), it is then pointed out, is consistent with determinism (and with divine providence); for even if all of the man's actions were
causally determined, the man could still be such that, if he had chosen otherwise,
then he would have done otherwise. What the murderer saw, let us suppose, along
with his beliefs and desires, caused him to fire the shot; yet he was such that if,
just then, he had chosen or decided not to fire the shot, then he would not have
fired it. All of this is certainly possible. Similarly, we could say, of the dam, that
the flood caused it to break and also that the dam was such that, //"there had been
no flood or any similar pressure, then the dam would have remained intact. And
therefore, the argument proceeds, if (b) is consistent with determinism, and if (a)
and (b) say the same thing, then (a) is also consistent with determinism; hence
we can say that the agent could have done otherwise even though he was caused
to do what he did do; and therefore determinism and moral responsibility are
compatible.
Is the argument sound? The conclusion follows from the premises, but the
catch, I think, lies in the first premissthe one saying that statement (a) tells us
no more nor less than what statement (b) tells us. For (b), it would seem, could
be true while (a) is false. That is to say, our man might be such that, if he had
chosen to do otherwise, then he would have done otherwise, and yet also such
that he could not have done otherwise. Suppose, after all, that our murderer could
not have chosen, or could not have decided, to do otherwise. Then the fact that
he happens also to be a man such that, if he had chosen not to shoot he would
not have shot, would make no difference. For if he could not have chosen not to
shoot, then he could not have done anything other than just what it was that he
did do. In a word: from our statement (b) above ('If he had chosen to do otherwise,
then he would have done otherwise'), we cannot make an inference to (a) above
('He could have done otherwise') unless we can also assert:
(c) He could have chosen to do otherwise.
And therefore, if we must reject this third statement (c), then, even though we
may be justified in asserting (b), we are not justified in asserting (a). If the man
could not have chosen to do otherwise, then he would not have done otherwise
even //"he was such that, if he had chosen to do otherwise, then he would have
done otherwise.
The stratagem in question, then, seems to me not to work, and I would say,
therefore, that the ascription of responsibility conflicts with a deterministic view
of action.
4. Perhaps there is less need to argue that the ascription of responsibility also
conflicts with an indeterministic view of actionwith the view that the act, or
some event that is essential to the act, is not caused at all. If the act-the firing
of the shotwas not caused at all, if it was fortuitous or capricious, happening
so to speak out of the blue, then, presumably, no oneand nothingwas responsible for the act. Our conception of action, therefore, should be neither deterministic nor indeterministic. Is there any other possibility?
5. We must not say that every event involved in the act is caused by some other
event; and we must not say that the act is something that is not caused at all. The
possibility that remains, therefore, is this: We should say that at least one of the
events that are involved in the act is caused, not by any other events, but by something else instead. And this something else can only be the agentthe man. If
there is an event that is caused, not by other events, but by the man, then there
are some events involved in the act that are not caused by other events. But if the
event in question is caused by the man then it is caused and we are not committed
to saying that there is something involved in the act that is not caused at all.
But this, of course, is a large consequence, implying something of considerable importance about the nature of the agent or the man.
6. If we consider only inanimate natural objects, we may say that causation,
if it occurs, is a relation between events or states of affairs. The dam's breaking
was an event that was caused by a set of other eventsthe dam being weak, the
flood being strong, and so on. But if a man is responsible for a particular deed,
then, if what I have said is true, there is some event, or set of events, that is
caused, not by other events or states of affairs, but by the agent, whatever he may
be.
I shall borrow a pair of medieval terms, using them, perhaps, in a way that
is slightly different from that for which they were originally intended. I shall say
that when one event or state of affairs (or set of events or states of affairs) causes
some other event or state of affairs, then we have an instance of transeunt causation. And I shall say that when an agent, as distinguished from an event, causes
an event or state of affairs, then we have an instance of immanent causation.
The nature of what is intended by the expression 'immanent causation' may be
illustrated by this sentence from Aristotle's physics: "Thus, a staff moves a stone,
and is moved by a hand, which is moved by a man" (VII, 5, 256a, 6-8). If the
man was responsible, then we have in this illustration a number of instances of
causationmost of them transeunt but at least one of them immanent. What the
staff did to the stone was an instance of transeunt causation, and thus we may describe it as a relation between events: "the motion of the staff caused the motion
of the stone." And similarly for what the hand did to the staff: "the motion of the
hand caused the motion of the staff." And, as we know from physiology, there
are still other events which caused the motion of the hand. Hence we need not
introduce the agent at this particular point, as Aristotle doeswe need not,
though we may. We may say that the hand was moved by the man, but we may
also say that the motion of the hand was caused by the motion of certain muscles;
and we may say that the motion of the muscles was caused by certain events that
took place within the brain. But some event, and presumably one of those that
took place within the brain, was caused by the agent and not by any other events.
There are, of course, objections to this way of putting the matter; I shall consider the two that seem to me to be most important.
7. One may object, firstly: "If the man does anything, then, as Aristotle's remark suggests, what he does is to move the hand. But he certainly does not do
anything to his brainhe may not even know that he has a brain. And if he doesn't
do anything to the brain, and if the motion of the hand was caused by something
that happened within the brain, then there is no point in appealing to 'immanent
causation' as being something incompatible with 'transeunt causation'for the
whole thing, after all, is a matter of causal relations among events or states of
affairs."
The answer to this objection, I think, is this: It is true that the agent does not
do anything with his brain, or to his brain, in the sense in which he does something
with his hand and does something to the staff. But from this it does not follow
that the agent was not the immanent cause of something that happened within his
brain.
We should note a useful distinction that has been proposed by Professor A.
I. Meldennamely, the distinction between 'making something A happen' and
'doing A'.5 If I reach for the staff and pick it up, then one of the things that I do
is just that-reach for the staff and pick it up. And if it is something that I do,
then there is a very clear sense in which it may be said to be something that I know
that I do. If you ask me, "Are you doing something, or trying to do something,
with the staff?", I will have no difficulty in finding an answer. But in doing something with the staff, I also make various things happen which are not in this same
10
sense things that I do: I will make various air-particles move; I will free a number
of blades of grass from the pressure that had been upon them; and I may cause
a shadow to move from one place to another. If these are merely things that I make
happen, as distinguished from things that I do, then I may know nothing whatever
about them; I may not have the slightest idea that, in moving the staff, I am bringing about any such thing as the motion of air-particles, shadows, and blades of
grass.
We may say, in answer to the first objection, therefore, that it is true that our
agent does nothing to his brain or with his brain; but from this it does not follow
that the agent is not the immanent cause of some event within his brain; for the
brain event may be something which, like the motion of the air-particles, he made
happen in picking up the staff. The only difference between the two cases is this:
in each case, he made something happen when he picked up the staff; but in the
one casethe motion of the air particles or of the shadowsit was the motion
of the staff that caused the event to happen; and in the other case-the event that
took place in the brainit was this event that caused the motion of the staff.
The point is, in a word, that whenever a man does something A, then (by 'immanent causation') he makes a certain cerebral event happen, and this cerebral
event (by 'transeunt causation') makes A happen.
8. The second objection is more difficult and concerns the very concept of'immanent causation', or causation by an agent, as this concept is to be interpreted
here. The concept is subject to a difficulty which has long been associated with
that of the prime mover removed. We have said that there must be some event
A, presumably some cerebral event, which is caused not by an other event, but
by the agent. Since A was not caused by any other event, then the agent himself
cannot be said to have undergone any change or produced any other event (such
as 'an act of will' or the like) which brought A about. But if, when the agent made
A happen, there was no event involved other than A itself, no event which could
be described as making A happen, what did the agent's causation consist of?
What, for example, is the difference between A's just happening, and the agent's
causing A to happen? We cannot attribute the difference to any event that took
place within the agent. And so far as the event A itself is concerned, there would
seem to be no discernible difference. Thus Aristotle said that the activity of the
prime mover is nothing in addition to the motion that it produces, and Suarez said
that 'the action is in reality nothing but the effect as it flows from the agent'.6 Must
we conclude, then, that there is no more to the man's action in causing event A
than there is to the event A's happening by itself? Here we would seem to have
a distinction without a differencein which case we have failed to find a via media
between a deterministic and an indeterministic view of action.
The only answer, I think, can be this: that the difference between the man's
causing A, on the one hand, and the event A just happening, on the other, lies
in the fact that, in the first case but not the second, the event A was caused and
11
was caused by the man. There was a brain event A; the agent did, in fact, cause
the brain event; but there was nothing that he did to cause it.
This answer may not entirely satisfy and it will be likely to provoke the following question: 'But what are you really adding to the assertion that A happened
when you utter the words "The agent caused A to happen"?' As soon as we have
put the question this way, we see, I think, that whatever difficulty we may have
encountered is one that may be traced to the concept of causation generally
whether 'immanent' or 'transeunt'. The problem, in other words, is not a problem
that is peculiar to our conception of human action. It is a problem that must be
faced by anyone who makes use of the concept of causation at all; and therefore,
I would say, it is a problem for everyone but the complete indeterminist.
For the problem, as we put it, referring just to 'immanent causation', or causation by an agent, was this: 'What is the difference between saying, of an event
A, that A just happened and saying that someone caused A to happen?' The analogous problem, which holds for 'transeunt causation', or causation by an event, is
this: 'What is the difference between saying, of two events A and B, that B happened and then A happened, and saying that B's happening was the cause of A's
happening?' And the only answer that one can give is thisthat in the one case
the agent was the cause of A's happening and in the other case event B was the
cause of A's happening. The nature of transeunt causation is no more clear than
is that of immanent causation.
9. But we may plausibly sayand there is a respectable philosophical tradition
to which we may appealthat the notion of immanent causation; or causation by
an agent, is in fact more clear than that of transeunt causation, or causation by
an event, and that it is only by understanding our own causal efficacy, as agents,
that we can grasp the concept of cause at all. Hume may be said to have shown
that we do not derive the concept of cause from what we perceive of external
things. How, then, do we derive it? The most plausible suggestion, it seems to
me, is that of Reid, once again: namely that 'the conception of an efficient cause
may very probably be derived from the experience we have had . . . of our own
power to produce certain effects'.7 If we did not understand the concept of immanent causation, we would not understand that of transeunt causation.
10. It may have been noted that I have avoided the term 'free will' in all of this.
For even if there is such a faculty as 'the will', which somehow sets our acts agoing, the question of freedom, as John Locke said, is not the question "whether
the will be free"; it is the question "whether a man be free."8 For if there is a 'will',
as a moving faculty, the question is whether the man is free to will to do these
things that he does will to doand also whether he is free not to will any of those
things that he does will to do, and, again, whether he is free to will any of those
things that he does not will to do. Jonathan Edwards tried to restrict himself to
the question"Is the man free to do what it is that he wills?"but the answer to
this question will not tell us whether the man is responsible for what it is that he
12
does will to do. Using still another pair of medieval terms, we may say that the
metaphysical problem of freedom does not concern the actus imperatus; it does
not concern the question whether we are free to accomplish whatever it is that
we will or set out to do; it concerns the actus elicitus, the question whether we
are free to will or to set out to do those things that we do will or set out to do.
11. If we are responsible, and if what I have been trying to say is true, then
we have a prerogative which some would attribute only to God: each of us, when
we act, is a prime mover unmoved. In doing what we do, we cause certain events
to happen, and nothingor no onecauses us to cause those events to happen.
12. If we are thus prime movers unmoved and if our actions, or those for which
we are responsible, are not causally determined, then they are not causally determined by our desires. And this means that the relation between what we want or
what we desire, on the one hand, and what it is that we do, on the other, is not
as simple as most philosophers would have it.
We may distinguish between what we might call the 'Hobbist approach' and
what we might call the 'Kantian approach' to this question. The Hobbist approach
is the one that is generally accepted at the present time, but the Kantian approach,
I believe, is the one that is true. According to Hobbism, if we know, of some man,
what his beliefs and desires happen to he and how strong they are, if we know
what he feels certain of, what he desires more than anything else, and if we know
the state of his body and what stimuli he is being subjected to, then we may deduce, logically, just what it is that he will door, more accurately, just what it
is that he will try, set out, or undertake to do. Thus Professor Melden has said
that "the connection between wanting and doing is logical."9 But according to the
Kantian approach to our problem, and this is the one that I would take, there is
no such logical connection between wanting and doing, nor need there even be
a causal connection. No set of statements about a man's desires, beliefs, and
stimulus situation at any time implies any statement telling us what the man will
try, set out, or undertake to do at that time. As Reid put it, though we may 'reason
from men's motives to their actions and, in many cases, with great probability',
we can never do so 'with absolute certainty'.10
This means that, in one very strict sense of the terms, there can be no science
of man. If we think of science as a matter of finding out what laws happen to hold,
and if the statement of a law tells us what kinds of events are caused by what other
kinds of events, then there will be human actions which we cannot explain by subsuming them under any laws. We cannot say, 'It is causally necessary that, given
such and such desires and beliefs, and being subject to such and such stimuli, the
agent will do so and so'. For at times the agent, if he chooses, may rise above
his desires and do something else instead.
But all of this is consistent with saying that, perhaps more often than not, our
desires do exist under conditions such that those conditions necessitate us to act.
13
And we may also say, with Leibniz, that at other times our desires may 'incline
without necessitating'.
13. Leibniz's phrase presents us with our final philosophical problem. What
does it mean to say that a desire, or a motive, might "incline without necessitating"? There is a temptation, certainly, to say that "to incline" means to cause and
that "not to necessitate" means not to cause, but obviously we cannot have it both
ways.
Nor will Leibniz's own solution do. In his letter to Coste, he puts the problem
as follows: "When a choice is proposed, for example to go out or not to go out,
it is a question whether, with all the circumstances, internal and external, motives, perceptions, dispositions, impressions, passions, inclinations taken together, I am still in a contingent state, or whether I am necessitated to make the
choice, for example, to go out; that is to say, whether this proposition true and
determined in fact, In all these circumstances taken together I shall choose to go
out, is contingent or necessary."11 Leibniz's answer might be put us follows: in
one sense of the terms "necessary" and "contingent", the proposition "In all these
circumstances taken together I shall choose to go out," may be said to be contingent and not necessary, and in another sense of these terms, it may be said to be
necessary and not contingent. But the sense in which the proposition may be said
to be contingent, according to Leibniz, is only this: there is no logical contradiction involved in denying the proposition. And the sense in which it may be said
to be necessary is this: since "nothing ever occurs without cause or determining
reason'," the proposition is causally necessary. "Whenever all the circumstances
taken together are such that the balance of deliberation is heavier on one side than
on the other, it is certain and infallible that that is the side that is going to win
out." But if what we have been saying is true, the proposition 'In all these circumstances taken together I shall choose to go out', may be causally as well as logically contingent. Hence we must find another interpretation for Leibniz's statement that our motives and desires may incline us, or influence us, to choose
without thereby necessitating us to choose.
Let us consider a public official who has some moral scruples but who also,
as one says, could be had. Because of the scruples that he does have, he would
never take any positive steps receive a bribehe would not actively solicit one.
But his morality has its limits and he is also such that, if we were to confront him
with a. fait accompli or to let him see what is about to happen ($10,000 in cash
is being deposited behind the garage), then he would succumb and be unable to
resist. The general situation is a familiar one and this is one reason that people
pray to be delivered from temptation. (It also justifies Kant's remark: "And how
many there are who may have led a long blameless life, who are only fortunate
in having escaped so many temptations."12 Our relation to the misdeed that we
contemplate may not be a matter simply of being able to bring it about or not to
bring it about. As St. Anselm noted, there are at least four possibilities. We may
illustrate them by reference to our public official and the event which is his receiv-
14
ing the bribe, in the following way: (i) he may be able to bring the event about
himself (facere esse), in which case he would actively cause himself to receive
the bribe; (ii) he may be able to refrain from bringing it about himself (nonfacere
esse), in which case he would not himself do anything to insure that he receive
the bribe; (iii) he may be able to do something to prevent the event from occurring
(facere non esse}, in which case he would make sure that the $10,000 was not
left behind the garage; or (iv) he may be unable to do anything to prevent the event
from occuring (nonfacere non esse), in which case, though he may not solicit
the bribe, he would allow himself to keep it.13 We have envisaged our official as
a man who can resist the temptation to (i) but cannot resist the temptation to (iv):
he can refrain from bringing the event about himself, but he cannot bring himself
to do anything to prevent it.
Let us think of'inclination without necessitation', then, in such terms as these.
First we may contrast the two propositions:
(1) He can resist the temptation to do something in order to make A happen;
(2) He can resist the temptation to allow A to happen (i.e. to do nothing to
prevent A from happening).
We may suppose that the man has some desire to have A happen and thus has a
motive for making A happen. His motive for making A happen, I suggest, is one
that necessitates provided that, because of the motive, (1) is false; he cannot resist
the temptation to do something in order to make A happen. His motive for making
A happen is one that inclines provided that, because of the motive, (2) is false;
like our public official, he cannot bring himself to do anything to prevent A from
happening. And therefore we can say that this motive for making A happen is one
that inclines but does not necessitate provided that, because of the motive, (1) is
true and (2) is false; he can resist the temptation to make it happen but he cannot
resist the temptation to allow it to happen.
Notes
1. The general position to be presented here is suggested in the following writings, among others:
Aristotle, Eudemian Ethics, Book II, Chap. 6; Nicomachean Ethics, Book III, Chaps. 1-5; Thomas
Reid, Essays on the Active Powers of Man; C. A. Campbell, "Is 'Free Will' a Pseudo-Problem?" Mind,
N.S. 60 (1951): 441-65; Roderick M. Chisholm, "Responsibility and Avoidability," and Richard
Taylor, "Determination and the Theory of Agency," in Sidney Hook (ed.), Determinism and Freedom
in the Age of Modern Science (New York: New York University Press, 1958).
2. Thomas Reid, Essays on the Active Powers of Man, Essay IV, Chap. 4 (Works, P. 600).
3. Summa Theologia, First Part of the Second Part, Question VI ("On the Voluntary and Involuntary").
4. Jonathan Edwards, Freedom of the Will (New Haven, Conn.: Yale University Press, 1957);
G. E. Moore, Ethics (Home University Library, 1912), Chap. 6.
5. A. I. Melden, Free Action (Oxford: Basil Blackwell, 1961), especially Chap. 3. Mr. Melden's
own views, however, are quite the contrary of those proposed here.
6. Aristotle, Physics, Book III, Chap. 3; Suarez, Disputationes Metaphysicae, Disputation 18,
Sec. 10.
7. Reid, Works, p. 524.
15
8. Essay Concerning Human Understanding, Book II, Chap. 21.

9. Melden, Free Action, p. 166.
10. Reid, Works, pp. 608, 612.
11. "Lettre a Mr. Coste de la Necessite et de la Contingence" (1707), in Johanne E. Erdmann
(ed.), Opera Philosophica (Berlin: Sumtibus G. Eichleri, 1840), pp. 447-49.
12. In the preface to the Metaphysical Elements of Ethics, in T. K. Abbot (ed.), Kant's Critique
of Practical Reason and Other Works on the Theory of Ethics (London: Longman's Green, 1959),
p. 303.
13. Cf. D. P. Henry, "Saint Anselm's De 'Grammatico'," Philosophical Quarterly, 10 (1960):
115-26. St. Anselm noted that (i) and (iii), respectively, may be thought of as forming the upper left
and the upper right corners of a square of opposition, and (ii) and (iv) the lower right and the lower
left.
Part II
Coming into Being, Persisting, and Passing Away
Identity through Possible Worlds

It is now easy to see a simple way of avoiding undesirable existential
generalizations in epistemic contexts. Existential generalization with respect to a term-say b is admissible in such contexts if b refers to one and
the same man in all the "possible worlds" we have to consider.1
In an article on Hintikka's Knowledge and Belief, I suggested that certain difficult
questions come to mind when we consider the thought that an individual in one
possible world might be identical with an individual in another possible world.2
The present paper is written in response to the editor's invitation to be more explicit about these questions.
Let us suppose, then, that the figure of an infinity of possible worlds makes
good sense and let us also suppose, for simplicity of presentation, that we have
a complete description of this one. We may consider some one of the entities of
this world, alter its description slightly, adjust the descriptions of the other entities in the world to fit this alteration, and then ask ourselves whether the entity
in the possible world that we thus arrive at is identical with the entity we started
with in this world. We start with Adam, say; we alter his description slightly and
allow him to live for 931 years instead of for only 930; we then accommodate
our descriptions of the other entities of the world to fit this possibility (Eve, for
example, will now have the property of being married to a man who lives for 931
years instead of that of being married to a man who lives for only 930); and we
thus arrive at a description of another possible world.3
Let us call our present world "W1" and the possible world we have just indicated "W2". Is the Adam of our world W1 the same person as the Adam of the
possible world W2? In other words, is Adam such that he lives for just 930 years
in W1 and for 931 in W2? And how are we to decide?
One's first thought might be that the proposition that Adam is in both worlds
19
20
IDENTITY THROUGH POSSIBLE WORLDS
is incompatible with the principle of the indiscernibility of identicals. How could

our Adam be identical with that one of ours lives for just 930 years and that one
for 931? Possibly this question could be answered in the following way:
"Compare the question: How can Adam at the age of 930 be the same person
as the man who ate the forbidden fruit, if the former is old and the latter is young?
Here the proper reply would be: it is not true that the old Adam has properties
that render him discernible from the young Adam; the truth is, rather, that Adam
has the property of being young when he eats the forbidden fruit and the property
of being old in the year 930, and that these properties, though different, are not
incompatible. And so, too, for the different possible worlds: It is not true that the
Adam of W1 has properties that render him discernible from the Adam of W2;
the truth is, rather, that Adam has the property of living for 930 years in W1 and
the property of living for 931 in W2, and that these properties, though different,
are not incompatible."
I think it is clear that we must deal with the old Adam and the young Adam
in the manner indicated; but in this case, one could argue, we know independently
that the same Adam is involved throughout. But are we justified in dealing in a
similar way with the Adam of W1 and the Adam of W2? In this latter case, one
might say, we do not know independently that the same Adam is involved
throughout. Here, then, is one of the questions that I do not know how to answer.
Let us suppose, however, that we answer it affirmatively.
The Adam of this world, we are assuming, is identical with the Adam of that
one. In other words, Adam is such that he lives for only 930 years in W1 and for
931 in W2. Let us now suppose further that we have arrived at our conception
of W2, not only by introducing alterations in our description of the Adam of W1,
but also by introducing alterations in our description of the Noah of W1. We say;
"Suppose Adam had lived for 931 years instead of 930 and suppose Noah had
lived for 949 years instead of 950." We then arrive at our description of W2 by
accommodating our descriptions of the other entities of W1 in such a way that
these entities will be capable of inhabiting the same possible world as the revised
Noah and the revised Adam. Both Noah and Adam, then, may be found in W2
as well as in W1.
Now let us move from W2 to still another possible world W3. Once again, we
will start by introducing alterations in Adam and Noah and then accommodate
the rest of the world to what we have done. In W3 Adam lives for 932 years and
Noah for 948. Then moving from one possible world to another, but keeping our
fingers, so to speak, on the same two entities, we arrive at a world in which Noah
lives for 930 years and Adam for 950. In that world, therefore, Noah has the age
that Adam has in this one, and Adam has the age that Noah has in this one; the
Adam and Noah that we started with might thus be said to have exchanged their
ages. Now let us continue on to still other possible worlds and allow them to exchange still other properties. We will imagine a possible world in which they have
exchanged the first letters of their names, then one in which they have exchanged
the second, then one in which they have exchanged the fourth, with the result that
21
Adam in this new possible world will be called "Noah" and Noah "Adam."
Proceeding in this way, we arrive finally at a possible world Wn which would
seem to be exactly like our present world W1, except for the fact that the Adam
of Wn may be traced back to the Noah of W1 and the Noah of Wn may be traced
back to the Adam of W 1 .
Should we say of the Adam of W n that he is identical with the Noah of W1
and should we say of the Noah of Wn that he is identical the Adam of W1? In other
words, is there an x such that x is Adam in W1 and x is Noah in W n , and is there
a y such that y is Noah in W1 and y is Adam in Wn? And how are we to decide?
But let us suppose that somehow we have arrived at an affirmative answer.
Now we must ask ourselves: How is one to tell the difference between the two
worlds W1 and Wn? Shall we say that, though they are diverse, they are yet indiscernible from each other-or, at any rate, that the Adam of W1 is indiscernible
from the Adam of Wn (who is in fact the Noah of W1) and that the Noah of W1
is indiscernible from the Noah of Wn (who is in fact the Adam of W1)? There
is a certain ambiguity in "discernible" and in "indiscernible." The two Adams
could be called "discernible" in that the one has the property of being Noah in the
other world and the other does not, and similarly for the two Noahs. But in the
sense of "indiscernible" that allows us to say that "Indiscernibles are identical"
tells us more than merely "Identicals are identical," aren't the two Adams, the two
Noahs, and the two worlds indiscernible? Could God possibly have had a
sufficient reason for creating W1 instead of Wn?
If W1 and Wn are two different possible worlds, then, of course, there are indefinitely many others, equally difficult to distinguish from each other and from
W1 and Wn. For what we have done to Adam and Noah, we can do to any other
pair of entities. Therefore among the possible worlds that would seem to be indiscernible from this one, there are those in which you play the role that I play
in this one and in which I play the role that you play in this one.4 (If this is true,
there may be good ground for the existentialist's Angst; since, it would seem, God
could have had no sufficient reason for choosing the world in which you play your
present role instead of one in which you play mine.)
Is there really a good reason for saying that this Adam and Noah are identical,
respectively, with that Noah and Adam? We opened the door to this conclusion
by assuming that Adam could be found in more than one possible worldby assuming that there is an x such that x is Adam in W1 and lives here for 930 years
and x is also Adam in W2 and lives there for 931. If it is reasonable to assume
that Adam retains his identity through the relatively slight changes involved in
the transition from W1 to W2, and so, too, for Noah, then it would also seem
reasonable to assume that each retains his identity through the equally slight
changes involved in all the other transitions that took us finally to Wn. (These
transitions, of course, may be as gradual as one pleases. Instead of it being a year
that we take away from Noah in our first step and give to Adam, it could be only
a day, or a fraction of a second.) But identity is transitive. And therefore, one
might argue, once we allow Adam to exist in more than one possible world, we
22
commit ourselves to affirmative answers to the puzzling questions we have encountered.

Is there a way, then, in which we might reasonably countenance identity
through possible worlds and yet avoid such extreme conclusions? The only way,
so far as I can see, is to appeal to some version of the doctrine that individual
things have essential properties. One possibility would be this:
For every entity x, there are certain properties N and certain properties E such
that: x has N in some possible worlds and x has non-N in others; but x has E in
every possible world in which x exists; and, moreover, for every y, if y has E
in any possible world, then y is identical with x. (If "being identical with x" refers
to a property of x, then we should add that E includes certain properties other
than that of being identical with x.) The properties E will thus be essential to x
and the properties N nonessential, or accidental.5
To avoid misunderstanding, we should contrast this present use of "essential
property" with two others.
(1) Sometimes the "essential properties" of a thing are said to be just those
properties that the thing has necessarily. But it is not implausible to say that there
are certain properties which are such that everything has those properties necessarily; the properties, the example, of being either red or non-red, of being colored if red, and of being self-identical.6 Thus the Eiffel Tower is necessarily red
or non-red, necessarily colored if red, and necessarily self-identical; and so is
everything else.7
(2) And sometimes it is said (most unfortunately, it seems to me) that each individual thing is such that it has certain properties that are essential or necessary
to it "under certain descriptions of it" and that are not essential or necessary to
it "under certain other descriptions of it." Thus "under one of his descriptions,"
the property of being President is said to be essential to Mr. Johnson whereas "under that description" the property of being the husband of Lady Bird is not; and
"under another one of his descriptions," it is the other way around. Presumably
every property P of every individual thing x is such that, "under some description
of x," P is essential or necessary to x.
But if E is the set of properties that are essential to a given thing x, in the sense
of "essential" that we have defined above, then: E will not be a universal property
(indeed, nothing but x will have E); some of the properties of x will not be included in E; and E will not be such that there are descriptions of x "under which"
E is not, in the sense defined, essential to x.
If we accept this doctrine of essential properties, we may say, perhaps, that
the property of living for just 930 years is essential to Adam and therefore that
he may inhabit other possible worlds without living for just 930 years in each of
them. And so, too, perhaps, for having a name which, in English, ends with the
letter "m". But, we may then go on to say, somewhere in the journey from W1
to Wn, we left the essential properties of Adam (and therefore Adam himself) behind. But where? What are the properties that are essential to Adam? Being the
first man? Having a name which, in English, begins with the first letter of the al-
23
phabet? But why these properties? If we can contemplate Adam with slightly different properties in another possible world, why can't we think of him as having ancestors in some possible worlds and as having a different name in others? And similarly for any other property that might be proposed as being thus essential to Adam.
It seems to me that even if Adam does have such essential properties, there
is no procedure at all for finding out what they are. And it also seems to me that
there is no way of finding out whether he does have any essential properties. Is
there really a good reason, then, for supposing that he does?
The distinction between essential and non-essential properties seems to be involved in one of the traditional ways of dealing with the problem of knowing who*
If this way of dealing with that problem were satisfactory, then the doctrine of essential properties might have a kind of independent confirmation. But I am not sure
that is satisfactory. The problem of knowing who may be illustrated in this way.
I do not know who it was who robbed the bank this morning, but I do know, let
us assume, that there is someone who robbed the bank and I also know that that person is the man who drove off from the bank at 9:20 A.M. in a Buick Sedan. For
me to know who he is, therefore, it is not enough for me to have information enabling me to characterize him uniquely. What kind of information, then, would
entitle me to say that I know who he is? The essentialistic answer would be: "You
know who the bank robber is, provided that there is a certain set of properties E
which are essential to the x such that x robbed the bank and you know that x has
E and x robbed the bank." But if my doubts about essential properties are wellfounded, this solution to the problem of knowing who would imply that the police,
though they may finally "learn the thief s identity," will never know that they do.
For to know that one knows who the thief is (according to the proposed solution)
one must know what properties are essential to the thief; and if what I have said
is correct, we have no way of finding out what they are. How are the police to decide that they know who the thief is if they have no answer to the metaphysical question "What are the essential properties of the man we have arrested?"9
It is assumed, in many writings on modal logic, that "Necessarily, for every
x, x is identical with x" implies "For every x, necessarily x is identical with x,"
and therefore also "For every x and y, if x is identical with y, then necessarily
x is identical with y." But is the assumption reasonable? It leads us to perplexing
conclusions: for example, to the conclusion that every entity exists in every possible world and therefore, presumably, that everything is an ens necessarium.
Why assume that necessarily the evening star is identical with the evening star?
We should remind ourselves that "The evening star is identical with the evening
star" is not a logical truth, for it implies the contingent proposition "There is an
evening star," and that its negation is not "The evening star is diverse from the evening star." Wouldn't it be simpler to deny that "Necessarily, for every x, x is identical with x" implies "For every x, necessarily x is identical with x"? Then we could
deny the principle de dicto, "Necessarily the evening star is identical with the evening star," and also deny the principle, de re, "The evening star is necessarily identical with the evening star."10 We could still do justice to the necessity that is here
24
involved, it seems to me, provided we continued to affirm such principles, de dicto,

as "Necessarily, for every x, x is identical with x" and "Necessarily, for every x
and y, if x is identical with y then y is identical with x," and such principles, de
re, "The evening star, like everything else, is necessarily self-identical."
Notes
1. Jaakko Hintikka, Knowledge and Belief: An Introduction to the Logic of the Two Notions
(Ithaca, N.Y.: Cornell University Press, 1962), p. 152.
2. "The Logic of Knowing," Journal of Philosophy, 60 (1963): 773-95; see esp. 787-95.
3. It should be noted that the possible world in question is not one that Hintikka would call
epistemically possible, for it could be said to contain certain states of affairs (Adam living for 931
years) that are incompatible with what we know to hold of this world; hence it is not one of the worlds
Hintikka is concerned with in the passage quoted above. But it is logically possible, and that is all
that matters for purposes of the present discussion.
4. "She (Ivich) looked at the glass, and Mathieu looked at her. A violent and undefined desire
had taken possession of him; a desire to be for one instant that unconsciousness . . . to feel those
long slender arms from within. . . . To be Ivich and not to cease to be himself." Sartre, The Age
of Reason. Compare N. L. Wilson, "Substance without Substrata," Review of Metaphysics, 12(1959),
and A. N. Prior, "Identifiable Individuals," Review of Metaphysics, 13 (1960).
5. We could put the doctrine more cautiously by saying that the distinction between the two types
of property holds, not for every entity x, but only for some entities x. But what reason could there
be for thinking that it holds of some entities and not of others?
6. Sometimes these properties are called "analytic properties" or "tautological properties"; but
the property of being colored if red should not be so-called if, as some have argued, "Everything that
is red is colored" is not analytic.
7. From the proposition that the Eiffel Tower is red and necessarily colored if red, it would be
fallacious to infer that the Eiffel Tower is necessarily colored; this is the fallacy of inferring necessitate consequents from necessitate consequentiae. And from the proposition that the Eiffel Tower is
necessarily red or non-red, it would be fallacious to infer that the proposition that the Eiffel Tower
is red or non-red is a necessary proposition; the proposition could hardly be necessary, for it implies
the contingent proposition that there is an Eiffel Tower. This latter fallacy might be called the fallacy
of inferring necessitate de dicto from necessitate de re.
8. Compare Aristotle, De Sophisticis Elenchis, 179 b 3; Petrus Hispanus Summulae Logicales,
ed. I. M. Bochenski (Turin: La Scuola, 1947), 7.41; Franz Brentano, Kategorienlehre (Leipzig: Felix
Meiner Verlag, 1933), p. 165.
9. Hintikka says that we know who the thief is provided that there exists an x such that we know
that the thief is identical with x {Knowledge and Belief, p. 153). But under what conditions may it
be said that there exists an x such that we know that the thief is identical with x? Presumably, if ever,
when we catch him in the act-when we see him steal the money. But the teller saw him steal the
money and she doesn't know who he is. I have suggested elsewhere a slightly different way of looking
at these questions; compare "Believing and Intentionality," Philosophy and Phenomenological Research, 25 (1964): 266-69, esp. 268.
10. I have discussed this possibility in "Query on Substitutivity," in Robert S. Cohen and Marx
W. Wartofsky (eds.), Boston Studies in the Philosophy of Science, vol. 2 (New York: The Humanities
Press, 1965), pp. 275-78.
If we deny that "Necessarily, for every x, x is F" implies "For every x, necessarily x is F," then
presumably we should also deny that "It is possible that there exists an x such that x is F" implies
"There exists an x such that it is possible that x is F." But isn't this what we should do? One could
hold quite consistently, it seems to me, that though it is possible that there exists something having
the properties that Christians attribute to God, yet nothing that does exist is such that it is possible
that that thing has the properties that Christians attribute to God.
Identity through Time

According to Bishop Butler, when we say of a physical thing existing at one time
that it is identical with or the same as a physical thing existing at some other time
("this is the same ship we traveled on before"), we are likely to be using the expression "same" or "identical" in a "loose and popular sense." But when we say
of a person existing at one time that he is identical with or the same as a person
existing at some other time ("the ship has the same captain it had before"), we
are likely to be using "same" or "identical" in a "strict and philosophical sense."1
I shall attempt to give an interpretation to these two theses; and I shall suggest
that there is at least an element of truth in each.
To illustrate the first of the two thesesthat it is likely to be only in a loose
and popular sense that we may speak of the identity of a physical thing through
time-let us recall the traditional problem of the ship of Theseus, in a somewhat
updated version. The ship, when it came to be, was made entirely of wood. One
day a wooden plank was replaced by an aluminum one (this is the updating) and
the wooden plank was cast off. But we still had the same ship, it was said, since
the change was only slight. Somewhat later, another wooden plank was cast off
and also replaced by an aluminum one. Still the same ship, of course, since, once
again, the change was only slight. The changes continue, but they are always
sufficiently slight so that the ship on any given day can be said to be the same as
the ship on the day before. Finally, of course, the ship is made entirely of aluminum. Some will feel certain that the aluminum ship is the same ship as the one
that was once made entirely of wood. After all, it preserved its identity from one
change to the next, and identity is transitive. Consider, however, this possibility,
suggested by Thomas Hobbes: " . . . if some man had kept the old planks as
they were taken out, and by putting them afterwards together in the same order,
had again made a ship of them, this, without doubt, had also been the same numerical ship with that which was at the beginning; and so there would have been two
25
26
IDENTITY THROUGH TIME
ships numerically the same, which is absurd."2 To compound the problem, let us
imagine that the captain of the original ship had taken a vow to the effect that if
his ship were ever to go down, then he would go down with it. What now, if the
two ships collide at sea and he sees them start to sink together? Where does his
duty liewith the aluminum ship or with the reassembled wooden ship?
Putting the problem schematically, we may suppose that on Monday a simple
ship, "The U.S.S. South Dakota," came into being, composed of two principle
parts, A and B. On Tuesday, part A is replaced by a new part C. (We may imagine
that the replacement was accomplished with a minimum of disturbance: as A was
eased off, C was pushed on immediately behind and in such a way that one could
not say at any time during the process that there was only half a ship in the harbor.) On Wednesday, there was fission, with B going off to the left and annexing
itself to F as it departed from C, and with C going off to the right and annexing
itself to J as it departed from B. On Thursday, over at the left, B is replaced by
L, while, over at the right, C is replaced by H. And now the captain of the original
U.S.S. South Dakota sees FL and JH in equal distress.
Mon
AB
Tue
EC
Wed
FB CJ
Thu FL
JH
One of his advisers tells him: The ship on the left is the one that took the
maiden voyage on Monday, and the ship on the right, therefore, is not. But another of his advisers tells him: No, it's just the other way around. The ship on
the right is the one that took the maiden voyage on Monday, and the ship on the
left, therefore, is not. Agreeing on the need for philosophical assistance, the two
advisers appeal to a metaphysician who instructs them in the following way: First
of all, he says, we must make a technical distinction between what I shall call an
intactly persisting temporal object and what I shall call a nonintactly persisting
temporal object. A thing is an intactly persisting temporal object if it exists during
a period of time and is such that, at any moment of its existence, it has the same
parts it had at any other moment of its existence. We may suppose that AB, the
object that came into being on Monday and passed away on Tuesday, was such
an intactly persisting object. So, too, for BC, for FB, for CJ, for FL, and for JH.
Thus a nonintactly persisting temporal object will be a temporal object that is
composed of one set of parts at one time and of another set of parts at another
time. If we can say of a ship, that it is composed of A and B on Monday and composed of B and Con Tuesday, then a ship is such a nonintactly persisting temporal
object."3
Appealing now to our diagram, the metaphysician continues: I assume that the
situation you disagree about involves the six intact temporal objects you have labeled. It also involves a number of nonintact temporal objects. Thus (i) there is
that total object, having the temporal shape of an upside down Ythat object is
composed of AB on Monday, of fiCon Tuesday, of FB and CD on Wednesday,
27
and of FL and JHon Thursday; (ii) there is that object composed of the stem and
the left fork of the Ythat object is composed of AB on Monday, of 5Con Tuesday, of FB on Wednesday, and of FL on Thursday; and (iii) there is that object
composed of the stem and of the right fork of the Ythe object that is composed
of AB on Monday, of EC on Tuesday, of CJ on Wednesday, and of JHon Thursday. The second and third of these temporal objects thus have certain parts in
common, and the first one includes both the second and the third among its parts.
Given such distinctions as these, our metaphysician now concludes, you can
see that there is really nothing for you to dispute about. Just consider the question:
Is the ship on the left the one that made the maiden voyage on Monday? If you
are asking whether FL is identical with AB, then the answer is obviously no, for
FL didn't come into being until Thursday and AB ceased to exist on Tuesday. On
the other hand, if you are asking whether FL and AB are both parts of our second
temporal object, the one composed of the stem and of the left fork of the Y, the
answer is clearly yes; and JH is not a part of that object. And if you are asking
whether JH and AB are both parts of our third temporal object, the one that is
composed of the stem and of the right fork of the Y, then the answer, once again,
is clearly yes; and FL is not a part of that object. All you need to do then, is to
distinguish these various objects and make sure you know which ones you are
talking about. Then everything will be clear.
I think we might go along with the metaphysicianup to the very last point.
Consider the reaction that his sort of instructions might produce: You say that
everything will be clear. Things werejar more clear before you entered the picture. We couldn't agree as to which of these two ships was the one that set sail
on Monday. But we were clear, at least, that only two ships were involved. Now,
with all your intact and nonintact temporal objects, we have no idea how many
ships there were. We have learned from Webster that a ship is a structure used
for transportation in water. Your intact temporal objects satisfy that definition;
so they yield at least six ships. What of the nonintact temporal objects? Is the one
having the temporal shape of the Y a ship? That would make seven. The stem
would give us eight, the two forks would bring it up to ten; the stem plus the left
fork makes eleven, and the stem plus the right one makes it twelve. Conceivably
we might countenance the presence of twelve ships in this situation if by so doing
we could solve our problem. But you haven't solved the problem. Consider the
poor captain. He wants to go down with his ship and he still doesn't know which
way to go.
Our metaphysician, I suggest, did not succeed in locating the source of the
dispute.
Consider the problem as it pertained to the relation between FL (the object that
came to be, on the left, on Thursday) and AB (the object that had ceased to be
by Tuesday). It was agreed that Webster's definition of "x is a ship" would do.
It was also agreed that FL, AB, and the other intact objects satisfied that definition.
The question was whether FL constituted the same ship as did AB. And the question whether FL constituted the same ship as did AB must be distinguished from
28
the closely related question whether FL was identical with AB; for, as Locke saw,
at least in principle, "FL constitutes the same ship as does AB" does not imply "FL
is identical with AB."4
Railroad trains may provide a more perspicuous example of the distinction between "jc constitutes the same so-and-so as does y" and "jc is identical with y." Suppose we ask: Is this the same train we rode on last year? We are not concerned
to know whether the set of objects that makes up today's train is identical with
the set of objects that made up the train of a year ago. ("I'm not asking whether
we rode on precisely these same cars a year ago!") The following three statements
tell us three quite different things: (1) This set of cars constitutes a train today
and it also constituted a train a year ago; (2) This set of cars constitutes the same
train as did that set of cars and that set of cars constituted a train a year ago; (3)
This set of cars constitutes the same train that that set of cars constituted a year
ago. By going to the dictionary we may find a definition or criterion of "x is a
train"; but we do not thereby find a definition or criterion of "x constitutes the
same train as does y." A definition of the latter expression would be much more
complex and would doubtless say something about roadbeds, schedules, and cities. Possibly, for example, if we can agree that the present aggregate of cars
leaves Hoboken at 7:30 P.M. for Chicago via Scranton and the Poconos, we may
be willing to concede that this is the same train that we took a year ago, even if
all the cars are different. (We may note, in passing, that in this case applicability
of"* is the same train as y" will presuppose applicability of some such expression
as "jc is the same roadbed as was y" or "x is the same city as was y.")
"The same ship" would seem to require a kind of continuity that "the same railroad train" does not. That is to say, if this is to be the same ship that that was,
then this ship must be evolved in some clear-cut way from that. The requisite
sense of "evolves" is illustrated by our diagram. Thus BC is continuous with AB
in that they have a part in common; we may say, therefore, that the latter object
BC "directly evolved" from the earlier object AB. Analogously for the relation
of FB to BC, of FL to FB, of CJ to BC, ofJHto CJ, and of FB to AB. And since
FL directly evolved from something that directly evolved from AB, we may say
simply that FL evolved from AB.5
What more is needed if this to be the same ship that that was? The best we can
do, I believe, is to formulate various additional criteria which are such that, if
they are satisfied, then this is the same ship that that was. Let us consider only
one such criterionone involving reference to sameness of sailing schedule. Suppose we know, with respect to each object, that it satisfies Webster's definition
of a ship: each object is a structure that is used for transportation in water. Suppose we also know that everything that evolved from that and into this was also
a structure used for transportation in water (none of these things was ever towed
on land and used there as a dwelling-place or as a restaurant). Suppose we know,
moreover, that they all followed the same sailing schedule (they were used, say,
to ferry passengers between Hoboken and lower Manhattan). And suppose we
know, finally, that if at any time one of these objects underwent fission at that
29
time and evolved into more than one structure that was used for transportation
in water, then only one of those structures kept to the original schedule. If we
know all these things, then, I think, we may say with confidence, that this is the
same ship as thator, more accurately, that this constitutes now the same ship
that that constituted then.
Hence one possible criterion (as distinguished from a definition) of "jc constitutes now the same ship that y constituted then" would be this: jc evolved from
y; everything that evolved from y and into jc was a structure used for transportation in water and followed the same sailing schedule that y does; and if at any time
more than one such structure evolved at that time from y, then only one of them
followed the same sailing schedule that y does.
If we should be fortunate enough to find that Wednesday's left hand object followed the same sailing schedule as did those of Monday and Tuesday, and that
Wednesday's right hand object took off on a course of its own, then we may conclude that the one on the left, and not the one on the right constitutes the same
ship as the one that came to be on Monday.
Reverting to the terminology of our metaphysician, we may say that the situation we have been concerned with involved at least six different intactly persisting
objects and at least six different nonintactly persisting objects. Does this mean,
then, that the situation involved at least a dozen ships? No, for if we speak in a
strict and philosophical sense, we will say that counting ships through a given
period of time is not the same as counting structures that are used for transportation in water during that time; it is, rather, to count sets of objects that constitute
the same ship during that time. For example, to say that there is one ship is to
say that there is one set of things all constituting the same ship. To say that there
are two ships is to say that there are two sets of things, all the members of the
one set constituting the same ship, all the members of the other set constituting
the same ship, and no member of the one set constituting the same ship as any
member of the other set. And so on, for any number of ships. If, as we are supposing, the AB, BC, FB, and FL of our example all follow the same sailing schedule,
then they constitute one ship. CJ, we said, took off on its own. Hence ifJH follows the same sailing schedule as did CJ, then the situation will involve at most
two ships.
We could put the matter paradoxically, therefore, by saying that counting
ships is not the same, merely, as counting objects that happen to be ships. But
if we speak strictly and philosophically, we may avoid any such appearance of
paradox. We may say that ships are "logical constructions." The things that they
are constructed upon are things that satisfy Webster's definition of the loose and
popular sense of "ship"they are structures used for transportation in water. We
will not say, therefore, that AB, BC, and the other intact structures we discussed
are ships. We will say, instead, that each of these things constitutes a ship. Given
the concept of ":c constitutes the same ship as does y," we could define "jc constitutes a ship" by saying "x is a member of a set of things all constituting the same
ship." The U.S.S. South Dakota, therefore, would be a logical construction upon
30
one such set of things. If we continue to speak strictly and philosophically, we

will not say of the two different things, AB and FL, that each of them is, on its
particular day, the U.S.S. South Dakota. We will say instead that each of them
constitutes, on its particular day, the U.S.S. South Dakota. The statements we
ordinarily use to describe the ship (e.g., "It weighs more now than it did then")
will be reducible to statements about the things that constitute it ("the thing that
constitutes it now weighs more than the thing that constituted it then").
We now have an obvious interpretation for the first of theses I have attributed
to Bishop Butlernamely, that it is only in a loose and popular sense and not in
a strict and philosophical sense that we may speak of the identity of such things
as ships through time. He could be construed as telling us, first, that the expression "x constitutes at t the same ship that y constitutes at ?'" does not imply jc is
identical with y"; and analogously for "constituting the same tree," "constituting
the same carriage," and so on. Then he could be construed as telling us, second,
that if we express the fact that ;c constitutes at one time the same ship that y constitutes at another time by saying "jc is identical with y" or "jc is the same as y" then
we are speaking only in a loose and popular sense and not in a strict and philosophical sense. And perhaps he could be construed as telling us, finally, that our
criteria for jc constituting the same ship as y are pretty much in our own hands,
after all, and that once we have determined that a given x and y do satisfy our
criteria for constituting the same ship, or that they do not, then no possible ground
for doubt remains.
But there are points of clarification to be made:
(i) In saying that certain uses of language are "loose and popular" rather than
"strict and philosophical," we are not suggesting that those uses are incorrect. Indeed, they may be said to be correct; for it is the loose and popular interpretation
rather than the strict and philosophical one that gives the standard of correctness
(at any rate, in the loose and popular sense of "correct").
(ii) I have said that it is only in a loose and popular sense that the thing which
makes up the U.S.S. South Dakota on Thursday may be said to be identical with,
or the same as, the thing which made up the U.S.S. South Dakota on Monday.
But this is not to say, of that nonintactly persisting four dimensional temporal object depicted by the stem and the left fork of our upside down Y, that it is identical
with itself only in a loose and popular sense; for that object, like any other, is identical with itself in a strict and philosophical sense. So, too, therefore, for that object made up of the stem and the right fork of the Y, as well as for that object
(if our metaphysician was right in assuming that there is such as object) made up
of the stem plus the right and left forks of the Y. Our account of the U.S.S. South
Dakota, therefore, should not be taken to imply that there are no such four dimensional objects, or that, if there are such, they are not strictly self-identical.
If we could formulate an adequate set of criteria for applying the expression
"jc constitutes the same ship as does y" (the one cited above, of course, was only
schematic), and if we found that these criteria were satisfied, say, by the original
intact object of Monday and the left intact object of Tuesday, then, for the ship
31
that is thus constituted ("AB and FL both constitute the U.S.S. South Dakota"),
there would be exactly one nonintact temporal object (namely, the stem and the
left fork of the Y). Therefore an alternative to saying that the U.S.S. South
Dakota is a logical construction upon a set of things all constituting the same ship
would be to say that the U.S.S. South Dakota is a nonintact temporal object having as its parts at different times the members of a set of things all constituting
the same ship. In this case, we could say that the U.S.S. South Dakota is the stem
and the left fork of the Y.
If we speak in this way, then the point that I have attributed to Bishop Butler
will be even more obvious. He would now be telling us that the expression "* is
a part of the same ship at t that y is a part of at t'" does not imply "jc is identical
with /'; hence if we use the expression "x is the same as y" or "jc is identical with
y" merely to express the fact that "x is a part of the same ship that v is a part of,"
then we are speaking in a loose and popular sense and not in a strict and philosophical sense.6
We should remind ourselves, moreover, that merely by referring to AB and
FL as "temporal parts" of one and the same temporal object, we do not thereby
answer the question we had put originally as "Does FL constitute the same ship
as did AB1" and that we might now put alternatively as "Is FL a temporal part of
the same ship as was ABT To simplify the problem even further, consider just
two ships, X and F, and two days, Monday and Tuesday, through which both
ships persist. Let us assume for the moment that there is one set of parts that make
up the ship on Monday and another, entirely different set of parts that make up
the ship on Tuesday. We may refer to the parts that make up X on Mondays as
"the Monday parts of X" and to the parts that make up Xon Tuesday as "the Tuesday parts cf X." We may now distinguish four pairs of successive aggregates of
parts: (1) the Monday parts of Xand the Tuesday parts of X; (2) the Monday parts
of Fand the Tuesday parts of Y; (3) the Monday parts of X and the Tuesday parts
of Y; and (4) the Monday parts of Y and the Tuesday parts of X. We must therefore
choose between two courses. We may say (a) that these four pairs of successive
aggregates of parts constitute four temporal objects; or we may say (b) that only
the first two pairs constitute genuine temporal objects and hence there are only
two such objects. If we take the first course and say that there are four temporal
objects, we will need a criterion for deciding which such temporal objects are to
be counted as ships; for we would be left with our problem if we said that the
third and fourth objects on the list are ships-that the Monday parts of X and the
Tuesday parts of Fmake up one ship, and that the Monday parts of 7 and the Tuesday parts of X make up another ship. If we take the second course and say that
only the first two pairs of objects on our list are "genuine" temporal objects, then
we will need to know what is required for successive temporal parts to be parts
of one and the same temporal object.7
(iii) Finding an acceptable definition of "jc is a ship" is a problem for dictionary
makers. Finding an acceptable definition of "jc constitutes the same ship as does
y" is more likely to be a problem for jurists. It should be noted that we may be
32
in agreement with respect to the proper interpretation of one of those expressions

and in disagreement with respect to the proper interpretation of the other; or we
may be rigid with respect to the one and latitudinarian with respect to the other.
Assuming we have agreed upon our interpretation of "jc is a ship," consider
the latitude that yet remains with respect to the interpretation of "x constitutes the
same ship as does y." According to the particular criterion of constituting the same
ship that was satisfied by our example of the U.S.S. South Dakota, today's object
and tomorrow's object "constitute the same ship" provided, among other things,
that every object that evolves out of today's object and tomorrow's is a ship. And
for there to be such evolution, each object, we said, must have some part in common with the object from which it directly evolved. We could say, quoting Hume,
that with each step it is "in a manner requisite, that the change of parts be
not ... entire",8 but it is very possible that we will find it convenient to relax
these criteria. Thus it may be useful to be able to say, on occasion, that a certain
object of last year constitutes the same ship as does a certain object of this year
even though one of the objects, into which last year's object evolved and out of
which this year's object evolved, was itself not a ship. Perhaps the ship was partially dismantled and used for a while as a tool shed or as a restaurant; yet, when
it was reconverted, we found it convenient, and pleasing to count the result as
the same ship that we had before.9 We may even find it convenient to say on occasion that though a certain object of last year constitutes the same ship as does a
certain object of this year, there was no evolution as defined-the change of parts
at one stage was entire. Switching for the moment from ships to rivers, consider
this situation: We swim in the upper Rio Grande in the early spring; the river dries
up in the summer; new waters then flow in and we swim there once again in the
fall. Surely we will want our criterion of "jc constitutes the same river as does y"
to allow us to say that we swam in the same river twice.
(iv) The expression "jc constitutes the same ship as does y," like "jc is a ship,"
allows for borderline cases. We can readily imagine situations in which the only
appropriate answer to the question "Is that a ship?" is "Yes and no"-or, better,
situations in which "Yes" is no better an answer than "No," and "No" is no better
an answer than "Yes."10 A hydrofoil that is also a hovercraft may serve as an example. We can readily imagine situations in which to the question "Is this the
same ship as that?", i.e., "Does this constitute the same ship that that did?", the
only answer is "Yes and no."
It may well happen that when we encounter such a borderline case, we must
have an answer other than "Yes and no." The captain, as we have seen, may well
need a more definite answer, and we may need a definite answer to the question,
"Is the combination hydrofoil and hovercraft a ship?", for it may be necessary to
decide whether such things are to be subject to the regulations that govern ships
or to the regulations that govern aircraft. Similarly, for the question "Does this
constitute the same ship that that did?"
When the existence of such a borderline case does thus require us to make a
choice between "Yes" and "No" the decision is entirely a pragmatic one, simply
33
a matter of convenience. Which ship is to be called "the Ship of Theseus"the

one that evolved step by step from the original ship, or the one that was assembled
from the discarded planks of the original ship? Here we have such a borderline
question. The question calls for a convention with respect to the interpretation of
"constituting the same ship" (or of "is the same ship as," in its loose and popular
sense). We can have it pretty much as we wish, provided we agree. Which ship
should the captain go down with? Here, too, we have a borderline question. Perhaps you and I cannot decide, but the courts, or the ships' courts, can decide. If
the captain has agreed to go down with the U.S.S. South Dakota, and if the court
decides that the aluminum ship and not the wooden one is the one that constitutes
the U.S.S. South Dakota, then down with the aluminum ship he ought to go. Or
down with it he ought to go unless the authorities decide subsequently (and in
time) to "defeat" the convention they have adopted-for any such convention is
defeasible and may be altered or defeated if unexpected circumstances show that
it will turn out to be inconvenient. The important thing here is this: The convention of the courts, or of the proper authorities will settle the matter. You and I
may object to their decision on the ground that some other decision would have
been more convenient. But it would make no sense for us to say: Well, it just
might be, you know, that they are mistaken. It just might be that, unknown to
them, the wooden ship and not the aluminum one is the U.S.S. South Dakota.
(v) There is also a philosophical point to make about our treatment of the problems of the Ship of Theseus and the U.S.S. South Dakota.
Speaking of identity or persistence through time, Bishop Butler said: "In a
strict and philosophical manner of speech, no man, no being, no mode of being,
no anything, can be the same with that with which it hath indeed nothing the
same."11 We may be certain of at least this much: If there is an individual thing
jc which is such that, through a certain period of time, everything that is part of
jc at any given moment of that time is also a part of jc at any other moment of that
time, then what constitutes jc at any moment of that time may be said to be identical in the strict and philosophical sense with what constitutes jc at any other moment of that time. In such a case, jc would satisfy the concept of intact persistence
that was introduced above. For it was suggested that an individual thing jc could
be said to persist intactly through a given period of time, provided that, at any
subperiod of that time, jc has the same parts that jc has at any other subperiod of
that time. In other words, an individual jc persists intactly through a given period
of time, provided that, for every z, if z is part of jc during any subperiod of that
time, then z is part of jc during every subperiod of that time.12 Thus we may say
that if just one part of our ship is removed or replaced at a certain time, then other
parts of the ship, unlike the ship itself, persist intactly through that time.
We formulated above one possible criterion for saying, of different objects at
different times, that they constitute one and the same ship. This criterion, it should
be noted, presupposes intact persistence, though not intact persistence of the ship;
for if the criterion is applicable in the case of a given ship, then, with each step
of evolution, some part of the ship remains behind, for the change of parts is "not
34
entire." Hence, with each step, some part persists intactly; some part will be such
that it keeps all of its parts. But though the evolution of our ship from Monday
through Thursday involved intact persistence of some part of the ship at some
time during each change that took place, it does not presuppose intact persistence
of any part of the ship from Monday through Thursday. We are thus more liberal
in our interpretation of "jc constitutes the same ship as does y" than we are, say,
in our interpretaion of "jc constitutes the same bar of metal, or the same piece of
wood, or the same hunk of clay as does y." For we are not likley to say of jc that
it constitutes the same bar of metal, or the same piece of wood, or the same hunk
of clay as does y, unless we thing that, throughout the changes from x to y most
of the parts have persisted intactly. But "jc constitutes the same body of water as
does y" need not imply that most of the parts have thus persisted intactly. Indeed
it need not even imply that the body of water has undergone the type of evolution
we described in the case of the ship. Thus, as we have noted, a body of water
jc may constitute in the spring the same river that a body of water y constitutes
in the fall, even though the river has dried up in the summer and y, therefore,
has not evolved in the requisite sense from jc.13 We might say what St. Thomas
said of the river Seine:" . . . the Seine river is not'this particular river'because
of'this flowing water,' but because of'this source' and 'this bed,' and hence is always called the same river, although there may be other water flowing down it."14
Suppose, then, we say that the river of the spring is the same river as the river
of the fall in virtue of the fact that the rivet of the spring flows through the same
river bed as does the river of the fall. What, then, would be our criterion for saying that something jc in the spring is the same river bed as something y in the fall?
It might be the fact that the river bed in the fall has evolved in the manner I have
attempted to describe from the river bed in the spring. Or it might be that the material that constitutes the river bed in the spring is found the same river banks as
is the material that constitutes the river bed in the fall. We might then say that
jc in the spring constitutes the same river bank as does y in the fall if, once again,
y has evolved from jc in the manner I have described.15
In other words, persistence, in the loose and popular sense, through time
would seem to presuppose such evolution; and such evolution, in turn, presupposes persistence, in the strict and philosophical sense, through time. For it
presupposes what I have called intact persistence. It is not implausible to say,
therefore, that if there is anything that persists, in the loose and popular sense,
through any given period of time, then there is something (perhaps not the same
thing) that persists intactly through some subperiod of that time.
What now of Bishop Butler's second thesisthe thesis according to which,
when we say of a person existing at one time that he is identical with a person
existing at another time, we are likely to be using "identical" in a strict and philosophical sense and not merely in a loose and popular sense?
I have suggested a possible interpretation of the expression "loose and popular
sense of the same." Putting the point schematically, we may say that "jc is the same
Fas y" is used in a loose and popular sense if it is used in such a way that it does
35
not imply "jc is identical with y." (The expression "jc constitutes the same Fas does
y" would thus be less misleading for such a use.) I have also suggested that when "jc
is the same Fas/' is used in this loose and popular sense, then it is possible to imagine
conditions under which a question of the form "Is jc the same Fas y?" has no definite
answerconditions under which we may say both "Yes" and "No," for "Yes" will
be as good an answer as "No," and "No" will be as good an answer as "Yes."
Such an interpretation of the expression "loose and popular sense of same" suggests at once a possible interpretation of the expression "strict and philosophical
sense of same." For example, we are using the expression "jc is the same person
as y" in a strict and philosophical sense if we are using it in such a way that it
implies "jc is identical with y." In this case "jc is the same person as y" will be logically equivalent to ux is a person and x is identical with y." I wish to suggest that
"jc is the same person as y," where the expression in the place of "jc" is taken to
designate a certain person at existing at one time and where the expression in the
place of "y" is taken to designate a certain person existing at a different time, does
have this strict and philosophical use.
When we use "the same person" in this strict way, then, although cases may
well arise in which we have no way of deciding whether the person jc is the same
person as the person y, nevertheless the question "Is jc the same person as y?" will
have an answer and that answer will be either "Yes" or "No." If we know that x
is a person and if we also know thaty is a person, then it is not possible to imagine
circumstances under which the question "Is x the same person as y?" is a borderline questiona question admitting only of a "Yes and no" answer.
The latter point may be illustrated in the following way. If jc knows, with respect to some set of properties, that there is or will be a person y who will have
these properties at some future date, then jc may ask himself "Will / be he7" and
to that question the answer must be "Yes" or "No." For either jc is identical with
y or jc is not identical with y.
If it is clear that if x is a person and y is a person, then we cannot answer the
question "Is jc the same person as y?" merely by deciding what would be practically convenient. To be sure, if we lack sufficient evidence for making a decision,
it may yet be necessary for the courts to rule that x is the same person as y, or
that he is not. Perhaps the ruling will have to be based upon practical considerations and conceivably such considerations may lead the court later to "defeat"
their ruling. But if Bishop Butler, as I have interpreted him, is right, then one may
always ask of any such ruling "But is it correct, or trueT For a ruling to the effect
that jc is the same person as y will be correct, or true, only if jc is identical with y.
Here, then, we have one possible interpretation of the thesis that, in one of its
important uses, the expression "jc is the same person as y" must be interpreted in
a strict and philosophical sense. It seems clear to me that "jc is the same person
as y" does have this use. Whenever a person jc asks himself, with respect to some
person y, "Will I be he?" or "Was that person I?" then the answer to his question,
if put in the form "jc is the same person as y," or "jc is not the same person as y,"
must be taken in the strict and philosophical sense.
36
We should remind ourselves, however, that the expression "jt is the same person as y" also has a use which is not this strict and philosophical one. Thus there
are circumstances in which one might say: "Mr. Jones is not at all the same person
he used to be. You will be disappointed. He is not the person that you remember."
We would not say this sort of thing if Mr. Jones had changed only slightly. We
would say it only if he had undergone changes that were quite basic and
thoroughgoingthe kind of changes that might be produced by psychoanalysis
or by a lobotomy, or by a series of personal tragedies. But just how basic and
thoroughgoing must these changes be if we are to say of Mr. Jones that he is a
different person? The proper answer would seem to be: As basic and thoroughgoing as you would like. It's just a matter of convention. It all depends upon how
widely it is convenient for you to construe the expression "He's the same person
he used to be." Insofar as the rules of language are in your own hands, you may
have it any way you would like. (Compare "Jones is not himself today" or "Jones
was not himself when he said that.")
This, however, is only a "loose and popular" sense of identity. When we say,
in this sense, "Jones is no longer the person he used to be," we do not mean that
there is, or was, a certain entity such that Jones was formerly identical with that
entity and is no longer so. We do not mean to imply that there are (or have been)
certain entities, x and y, such that at one time :c is, or was, identical with y, and
at another time ;c is not identical with y. For this, I believe, is incoherent, but
"Jones is no longer the person he used to be" is not.
Nor do we mean, when we say "Jones is no longer the person he used to be,"
that there was a certain entity, the old Jones, which no longer exists, and that there
is a certain different entity, the new Jones, which somehow has taken his place.
We are not describing the kind of change that takes place when one President succeeds another. In the latter case, there is a clear answer to the question "What happened to the old one?" The answer might be "He was shot" or "He retired to Gettysburg." But when we decide to call Jones a new person, we are not confronted
with such questions as: What happened, then, to the old Jones? Did he die, or
was he annihilated, or disassembled, or did he retire to some other place?
The old Jones did not die; he was not annihilated or disassembled; and he did
not retire to any other place. He became the new Jones. To say that he "became"
the new Jones is not to say that he "became identical" with something he hadn't
been identical with before; for it is only when a thing comes into being that it may
be said to become identical with something it hadn't been identical with before.
To say that our man "became the new Jones" is to say only that he, Jones, altered
in a significant way, taking on certain interesting properties he had not had before. (Hence we should contrast the "became" of "Jones then became a married
man," said when Jones ceased to be a bachelor, with that of "The President then
became a Democrat," said when President Eisenhower retired.) When we say of
a thing that it has properties that it did not have before, we are saying that there
is an x such that x formerly had such-and-such properties and jc presently has suchand-such properties. But to say that there is an x, at least one x, such that x was
37
now this and jc is now that, would seem to presuppose the identity of x through
time, in some sense of the term "identity." Is the sense of identity that is presupposed merely that in which we can say of any temporal object, intact or nonintact,
that it is identical with itself? Or are we also presupposing that, in the strict and
philosophical sense, whatever goes to make up that person now is identical with
whatever went to make him up at the earlier time?16
One may well ask: But need we presuppose this? Need we presuppose the persistence of a single subject of change when, as we say, the man becomes "a new
person?" To appreciate the situation, it may be necessary to imagine that the person in question is oneself. Suppose, then, that you were such a personthat you
had undergone basic and thoroughgoing changes and that your friends and acquaintances were in agreement that you are no longer the same person that you
were. What, then, if you remember all the relevant factsthat you had formerly
been a person of such-and-such a sort, that you had undergone certain shattering
experiences, and that these then led to a transformation in your personality, with
the result that you are not the person that you formerly were?
Let us imagine, however, that your friends and acquaintances say to you: "But
you are such a very different person now that henceforth we are going to treat you
like one. We will call you "Smith" instead of "Jones." We will make certain that
you are free from all the obligations that Jones incurred. And if you feel guilty
about some of the wicked things that Jones did, you need no longer have such
feelings, for we can get the highest courts to lay it down that you are two quite
different people. Something, surely, is wrong here.
Some people, I have found, see at once that something is wrong and others
do not. For those who do not, let me propose that we look in a different direction.
What would we think of such talk if we were to hear it before rather than after
the transformation of our personality?
It will be instructive to elaborate upon an example that C. S. Peirce suggests.17
Let us assume that you are about to undergo an operation and that you still have
a decision to make. The utilities involved are, first, financialyou wish to avoid
any needless expenseand, secondly, the avoidance of pain, the avoidance, however, just of your pain, for pain that is other than yours, let us assume, is of no
concern whatever to you. The doctor proposes two operating proceduresone
a very expensive procedure in which you will be subjected to total anaesthesia
and no pain will be felt at all, and the other of a rather different sort. The second
operation will be very inexpensive indeed; there will be no anaesthesia at all and
therefore there will be excruciating pain. But the doctor will give you two drugs:
first, a drug just before the operation which will induce complete amnesia, so that
while you are on the table you will have no memory whatever of your present
life; and secondly, just after the agony is over, a drug that will make you completely forget everything that happened on the table. The question is: Given the
utilities involved, namely the avoidance of needless expense and the avoidance
of pain that you will feel, other pains not mattering, is it reasonable for you to
opt for the less expensive operation?
38
My own belief is that it would not be reasonable, even if you could be completely certain that both amnesia injections would be successful. I think that you
are the one who would undergo that pain, even though you, Jones, would not
know at the time that it is Jones who is undergoing it, and even though you would
never remember it. Consider after all, the hypothesis that it would not be you.
What would be your status, in such a case, during the time of the operation?
Would you be waiting in the wings somewhere for the second injection, and if
so, where? Or would you have passed away? That is to say, would you have
ceased to be, but with the guarantee that youyou, yourself would come into
being once again when the agony was over?18 And what about the person who
would be feeling the pain? Who would he be?
I can appreciate that these things might not seem obvious to you as you ponder
your decision. You may wonder: "I would certainly like to save that money. Will
it really be /who feels that pain? How can it be if I won't know that it's I?" Perhaps
you would have some ground for hesitation. But there is one point, I think, that
ought to be obvious.
Suppose that others come to youfriends, relatives, judges, clergymenand
they offer the following advice and assurance. "Have no fear," they will say,
"Take the cheaper operation and we will take care of everything. We will lay it
down that the man on the table is not you, Jones, but is Smith. We will not allow
this occasion to be mentioned in your biography. And during the time that you
lie there on the table-excuse us (they will interject), we mean to say, during the
time that Smith lies there on the table we will say, 'poor Smith' and we will not
say, even in our hearts, 'poor Jones.' " What ought to be obvious to you, it seems
to me, is that the laying down of this convention should have no effect at all upon
your decision. For you may still ask, "But won't that person be I?" and, it seems
to me, the question has an answer.
Suppose you know that your body, like that of an amoeba, would one day undergo fission and that you would go off, so to speak, in two different directions.
Suppose you also know, somehow, that the one who went off to the left would
experience the most wretched of lives and that the one who went off to the right
would experience a life of great happiness and value. If I am right in saying that
one's question "Will that person be I?" or "Will I be he?" always has a definite
answer, then, I think, we may draw these conclusions. There is no possibility
whatever that you would be both the person on the right and the person on the
left. Moreover, there is a possibility that you would be one or the other of those
two persons. And, finally, you could be one of those persons and yet have no
memory at all of your present existence.19 It follows that it would be reasonable
of you, if you are concerned with your future pleasures and pains, to hope that
you will be the one on the right and not the one on the left-also that it would
be reasonable of you, given such self-concern, to have this hope even if you know
that the one on the right would have no memory of your present existence. Indeed
it would be reasonable of you to have it even if you know that the one on the left
thought he remembered the facts of your present existence.20 And it seems to me
39
to be absolutely certain that no fears that you might have, about being the half
on the left, could reasonably be allayed by the adoption of a convention, or by
the formulation of a criterion, even if our procedure were endorsed by the highest
authorities.
Notes
I am indebted to John Wisdom, Sydney Shoemaker, and Fred Feldman for criticisms of earlier
versions of this paper. Certain paragraphs have been adapted from my "The Loose and Popular and
the Strict and Philosophical Senses of Identity," in Norman S. Care and Robert H. Grimm (eds.), Perception and Personal Identity, by permission of Case Western Reserve University Press, Cleveland.
1. "Of Personal Identity," Dissertation I, in The Whole Works of Joseph Butler, LL. D. (London:
Thomas Tegg, 1839), pp. 263-270. The dissertation is reprinted in Antony Flew, (ed.), Body, Mind
and Death (New York: Macmillan, 1964), pp. 166-72.
2. Thomas Hobbes, Concerning Body, Chap. II ("Of Identity and Difference"), Sec. 7.
3. But a nonintactly persisting temporal object should be distinguished from what I shall call an
"Edwardian" temporal object (after Jonathan Edwards). An Edwardian temporal object would be a
temporal object which is such that, for each moment during which it exists, there is a set of parts which
are what make up that object at that moment and which exist only at that moment. Hence if x is an
Edwardian temporal object, then for any two times, t and /' at which x exists, there is one set of objects which make up x at t, and another set of objects which make up x at t', and no member of the
first set has any part in common with any member of the second set. If, as some philosophers have
supposed, all temporal objects are Edwardian, then no object which persists through a period of time
could be said to persist intactly, as this term was defined by our metaphysician above; for no object
would be such that it has the same parts at any moment of its existence that it has at any other moment
of its existence. This extreme Edwardian view was defended by J. H. Woodger in The Axiomatic
Method in Biology (Cambridge: The University Press, 1937) and by Rudolf Carnap in Introduction
to Symbolic Logic (New York: Dover Publications, 1958), see pp. 213-16. Jonathan Edwards took
this extreme view to be implied by "God's upholding created substance, or causing its existence in
each successive moment." For, he reasoned, "if the existence of created substance, in each successive
moment, be wholly the effect of God's immediate power, in that moment, without any dependence
on prior existence, as much as the first creation out of nothing, then what exists at this moment, by
this power, is a new effect, and simply and absolutely considered, not the same with any past existence. . . . "From this he was able to deduce that it is as reasonable and just to impute Adam's original sin to me now as it is to impute any sin which I may seem to remember having committed myself.
(See the Doctrine of Original Sin Defended, Part 4, Chap. 2.) But this extreme view, when considered
separately from the doctrine of divine re-creation, has at least the disadvantage of multiplying entities
beyond necessity. (Compare: "John is kind toward Mary and unkind toward Alice; therefore there
is something, namely John-toward-Mary, that is kind, and there is something, namely John-towardAlice, that is unkind, and these two different things go to make up John.") It should not be attributed
to our metaphysician above.
4. See Locke's Essay, Book II, Chap. 27, Sec. 5, 6, and 8. The point made above does not, of
course, imply the more extreme thesis, according to which a statement of the form, "jc is identical
with y," is always elliptical for one of the form, "x is the same F as y."
5. These concepts might be defined as follows: x evolves directly from y, provided: either x is
identical with y, or there is no time at which x and y both exist but there is a z such that z is part of
y at one time and z is part of * at a later time. (Possibly we should add that, during any subperiod
between the earlier and the later time, z has the same parts that it has during any other such subperiod.)
And, more generally, x evolves from y, provided: * is a member of every class C such that (i) y is
a member of C and (ii) whatever directly evolves from anything that is a member of C is also a member
of C. (If the definition of "evolves directly" were intended to explicate the ordinary use of this expression, it would doubtless be too broad; but it is not so intended.)
40
6. We should take care not to misinterpret Butler at this point. In saying that "this is the same
ship as that" is to be understood only in a loose and popular sense and not in a strict and philosophical
sense, he is taking the "this" and the "that" to refer to the particular objects that, on their respective
days, constitute (or serve as the parts of) the ship in question on those days; he is not using "this"
or "that" to refer to the set of things that constitute the ship, or to the nonintact temporal object of
which those things may be construed as parts. This is clear, I think, from the following passage: "For
when a man swears to the same tree as having stood fifty years in the same place, he means only the
same as to all the purposes of property and uses of common life, and not that the tree has been all
that time the same in the strict and philosophical sense of the word. For he does not know whether
any one particle of the present tree be the same with any one particle of the tree which stood in the
same place fifty years ago."
7. Our example involves just four pairs of objects. Consider the number of ostensible temporal
objects, or ostensible ships, we would have to choose among, if we added just one more ship and
one more day to our example. Or the objects that would be involved if all temporal objects were "Edwardian," as this term was defined in footnote 3 above (The account that I have given of the history
of the U.S.S. South Dakota presupposes that there are some intactly persisting temporal objects, and
therefore it is inconsistent with the view that all temporal objects are Edwardian.)
8. Treatise, Book I, Part 4, Sec. vi.
9. An Aristotelian who took ships seriously might say that in such a case two "substantial
changes" had occurred.
10.1 owe this way of putting the matter to Professor John Wisdom who criticized an earlier version of this paper at Lewis and Clark College in October 1967.
11. "Of Personal Identity," in Flew, Body, Mind and Death, p. 265.
12. What if the parts of a thing are "simply re-arranged"-say, from ABC to CAB? If we take
the term "part" in its ordinary sense, as I propose that we do, then we must say that the thing will
not have persisted intactly, for it will have lost some parts. If the thing changes from ABC to CAB,
then it will lose BC, as well as that part that consists of the right half of B and the left half of C.
If we could say, however, that compound things are composed of "ultimate particles" which are
not themselves composed of parts, then we could formulate a definition of "intact persistence" which
would allow us to say that a thing may persist intactly through rearrangement of its parts. For we
could say that a thing persists intactly through a given period of time if, at any subperiod of that time,
it has the same "ultimate particles" that it has at any other period of that time. Thus Locke said that,
so long as any group of "atoms" (ultimate particles) "exist united together, the mass, consisting of
the same atoms, must be the same mass, or the same body, let the parts be ever so differently jumbled.
But if one of these atoms be taken away, or one now one added, it is no longer the same mass or the
same body." (Essay, Book II, Chap. 27, Sec. 4. This use of "same," I am sure he would have agreed,
is strict and philosophical and not loose and popular.)
13. For further possibilities, see Helen M. Cartwright, "Heraclitus and the Bath Water," Philosophical Review, 74 (1965): 466-84.
14. De Spiritualibus Creaturis, Article IX, ad. 16; translated as "On Spiritual Creatures," by M.
C. Fitzpatrick and J. J. Wollmuth (Milwaukee, Marquette University Press, 1949). See p. 109 of the
translation.
15. One might be tempted to define sameness of river bank or of riverbed, say, in terms of sameness of place. If sameness of place is not then defined in terms of a relation to things that are said
to exist in space, such a definition would seem to presuppose intact persistence of substantival space
through time. (It may be noted, incidentally, that our account of "evolving" allows us to say that a
thing at a later date evolves from a thing at an earlier date even though there has been no change of
parts.)
16. It should be noted that, although the loose and popular use in question is one in which ";c is
not the same person as y" is consistent with "x is identical with y," it is not a use in which "x is the
same person as y" is consistent with "x is not identical with y." It is difficult to think of any actual
examples of the latter type of use. Perhaps those who takes roles more seriously than they do persons
41
might say such things as "Elizabeth is the same monarch as Victoria was" while aware of the falsehood
of the corresponding identity statement. But such uses, fortunately, are not ordinary.
17." 'If the power to remember dies with the material body, has the question of any single person's
future life after death any particular interest for him?' As you put the question, it is not whether the
matter ought rationally to have an interest but whether as a fact it has; and perhaps this is the proper
question, trusting as it seems to do rather to instinct than to reason. Now if we had a drug which would
abolish memory for a while, and you were going to be cut for the stone, suppose the surgeon were
to say, 'You will suffer damnably, but I will administer this drug so that you will during that suffering
lose all memory of your previous life. Now you have, of course, no particular interest in your suffering as long as you will not remember your present and past life, you know, have you?' " Collected
Papers, vol. 5 (Cambridge, Mass.: Harvard University Press, 1935), p. 355.
18. See Locke's Essay, Book II, Chap, xxvii, Sec. i: "One thing cannot have two beginnings of
existence." Compare Thomas Reid, Essays on the Intellectual Powers of Man, Essay III, Chap. 4.
19. In this case, there might well be no criterion by means of which you or anyone else could
decide which of the two halves was in fact yourself. I would agree with Shoemaker's contention that
"our ability to know first-person psychological statements to be true, or the fact that we make them
(for the most part) only when they are true, cannot possibly be explained on the supposition that we
make them on the basis of criteria": Sydney Shoemaker, Self-Knowledge and Self-Identity (Ithaca,
N.Y.: Cornell University Press, 1963), p. 214. One consequence of this fact, I suggest, is the following: it makes sense to suppose in connection with the above example that you are in fact the half that
goes off to the left and not the one that goes off to the right even though there is no criterion at all
by means of which anyone could decide the matter. I would disagree, incidentally, with what Shoemaker says (pp. 236-38) about the relationship between criteria and necessary truthsat least, if
"necessary" is taken to mean the same as "logically necessary." My own views on this question may
be suggested by Chap. 4 ("The Problem of the Criterion") in my Theory of Knowledge (Englewood
Cliffs, N.J.: Prentice-Hall, 1966).
20. I would endorse, therefore, the following observation that Bayle makes in his article on
Lucretius (see Note Q of "Lucretius," in Pierre Bayle, A General Dictionary, Historical and Critical):
"The same atoms which compose water, are in ice, in vapours, in clouds, in hail and snow; those
which compose wheat, are in the meal, in the bread, the blood, the flesh, the bones etc. Were they
unhappy under the figure or form of water, and under that of ice, it would be the same numerical
substance that would be unhappy in those two conditions; and consequently all the calamities which
are to be dreaded, under the form of meal, concern the atoms which form corn; and nothing ought
to concern itself so much about the state or lot of the meal, as the atoms which form the wheat, though
they are not to suffer these calamities, under the form of wheat." Bayle concludes that "there are but
two methods a man can employ to calm, in a rational manner, the fears of another life. One is, to
promise himself the felicities of Paradise; the other, to be firmly persuaded that he shall be deprived
of sensations of every kind."
Possibility without Haecceity

The Problem of the Merely Possible
The "problem of the merely possible" is that of describing the merely possible
without supposing that there is a realm of possibilia falling between being and
nonbeing and without violating the principles of contradiction and excluded middle. To solve the problem, one may make use of one or the other of the following
two locutions:
(A) x is possible such that he is an F;
(B) There is a (possible) world in which x is an F.
Some philosophers would attempt to explicate (A) in terms of (B). Suppose
I am not a physician but could have been one. Then these philosophers would say
that there are possible worlds other than the actual one and that lama physician
in those worlds. If this explication is to work, then a possible world must be an
entity that is capable of containing me. But what is the relevant sense of "in" or
"contain?" We may give a meaning to this use of such expressions if we can say
that I have an individual essence or haecceitya set of properties that are "essential to me and repugnant to everything else." We could say, more generally, that
an individual thing jc exists in a possible world w provided that the haecceity of
x is exemplified in W.
There is, then, an enormous amount of metaphysics involved in explicating
(A) in terms of (B). And, so, let us try to do things the other way around. I will
suggest that if (B) is to make any sense, it must be explicated in terms of (A).
How are we to choose between these two approaches to the question? We can
ask which one presupposes the simpler ontology. The view that I will set forth
does not presuppose that in addition to individuals, properties and relations, and
states of affairs, there are such things as possible worlds in which particular in42
POSSIBILITY WITHOUT HAECCEITY
43
dividuals such as you and I may be said to exist. Therefore, we do not need to
cope with the problem of "transworld identity," and our view does not require us
to presuppose that individual things have individual essences or haecceities.
I will, then, make use of the undefined de re locution "x is possible such that
it is F" and the other modal locutions that may be defined in terms of it. And I
will make use of the concepts of properties, relations, and states of affairs.
Worlds Defined
I will begin with a conception of "possible world"but I will say just "world"
which is essentially that of Leibniz and other philosophers in the Western tradition. It is unlike that presupposed by most contemporary philosophers.
Dl Wis a world = DfW'vs, a state of affairs; for every state of affairs p, either
Wlogically impliesp or ^logically implies the negation of/?; and there
is no state of affairs q such that W logically implies both q and the negation
of q.
A world, then, is a self-consistent, maximal state of affairs. That it is maximal
is guaranteed by the first clause of our definition; that it is self-consistent is
guaranteed by the second.1 A world is thus a conjunctive state of affairs.
I have defined "a world" not "a possible world." I have avoided "possible
world," since the expression "There are possible worlds" may suggest that there
are certain thingsworldsthat neither exist nor fail to exist and that fall between being and nonbeing. I believe that the concept philosophers have traditionally had in mind when speaking of "possible worlds" can be explicated by reference to those states of affairs that are here called "worlds."2 If this is so, and if,
as I believe, states of affairs are abstract objects existing whether or not they obtain, then all so-called "possible worlds" exist. Hence, I use "world," and not
"possible world."
"But you can't mean to say that all possible worlds are actual worlds. There
isand can beonly one actual world!" The word "actual" here is ambiguous.
If "jc is actual" is taken to mean the same as "jc exists" then all possible worlds
are actual. But when it is said that only one world is actual, then "jc is actual" is
taken to mean the same as "x obtains." There isand can beonly one world that
obtains.
It is well to avoid the temptation to speak of "the real world" or "the actual
world." Let us, rather, speak of "the world that obtains" or "the prevailing world."
What is the Essence of a Thing?

Traditionally, the essence of a thing was said to be a property that is essential to
the thing and necessarily repugnant to everything else. The following definition
may therefore suggest itself:
44
E is the essence of jc = Dfx is necessarily such that it has E; and for all y,
ify is other than x, then y is necessarily such that it does not have E.
But the second clause is not strong enough. Suppose (i) I am necessarily such that
I am a person, (ii) all nonpersons are necessarily such that they are nonpersons,
and (iii) I am the only person there is. Given the proposed definition, it would
follow from these assumptions that the property of being a person is my essence.
But it would not follow that being a person is necessarily repugnant to everything
else.
How, then, are we to strengthen the second clause in the definition? Shall we
say: "It is not possible for there to be a y such that y is other than x and y has E"?
This would have to be reducible to our de re locution, "jc is possibly such that it
is F." What we should say is rather this: "jc is not possibly such that there is a y
such that y is other than x and y has E." Our definition will be:
D2 E is the essence of x = Dfx is necessarily such that it has E; for every
property P, if jc is necessarily such it has P, then E implies P\ and x is not
possibly such that there is a y other than jc such that y has E.
Given our definition of essence, we may now say what an abstract object is:
it is a thing having an essence that is such that everything is necessarily such that
there is something having that essence. The essence of the property blue is the
property of being a thing that is necessarily such that it is exemplified in all and
only those things that are blue. And if, as I believe, the property blue is an abstract
object, then everything is necessarily such that there is something that is necessarily such that it is exemplified in all and only those things that are blue. We will
say, then:
D3 jc is an abstract object = Df There is an E that is such that (a) E is the essence of jc and (b) everything is necessarily such that there is something
that has E.
I will assume that properties, relations, and states of affairs are all abstract objects. I will also assume that individual things-such entities as you and me-are
not abstract objects. It will follow (1) that everything is necessarily such that abstract objects exist and (2) that no abstract object is necessarily such that any individual thing exists.
Existing in a World
The conception of a world that has just been set forth is to be contrasted with that
held by many contemporary philosophers. Worlds, according to them, are things
in whichin some unanalyzed sense of "in"individuals such as you and me are
to be found. But if a world is a state of affairs, and if states of affairs are abstract
or eternal objects, what could it mean to say of an individual thing that it exists
"in a world"? How could you or I exist "in" an abstract object?
45
We may give a meaning to this use of "in." We could say, for example, that
if a world implies the property of being a dog, then at least one dog exists "in"
that world. (A state of affairs may be said to imply a given property provided only
that the state of affairs is necessarily such that, if it obtains, then something has
that property.) Hence, we might say similarly that if a thing jc has an essence, and
if a given world implies that essence, then jc exists "in" that world. And we can
also say that everything exists "in" the world that obtains. Let us say, then:
D4 jc exists in W = Df W is a world; and either (a) jc has an essence E such
that W implies E or (b) H^ obtains and jc exists.
Possibilities without Haecceities

The word "haecceity" has traditionally been used to mean the same as "individual
essence," or "essence of an individual thing." Let us use the word in this way:
D5 H is the haecceity of x = Dfx is an individual thing: and H is the essence
of jc.
Is there any reason for supposing that there are haecceities-that individual things
have essences?
Many believe that the doctrine that there are individual essences follows from
the assumption that individual things have unrealized possibilities. The reasoning
may be illustrated this way: "(1) I am not a physician. Therefore, (2) there is a
world PPthat is other than the world that obtains and that is such that I am a physician in W. But (3) if I exist in a world that does not obtain, then I have an essence.
(This follows from D4 our definition of what it is to exist in a world.) Therefore,
(4) I have an essence."
But (2) does not follow from (1). Let us consider how we might describe my
unrealized possibilities without assuming that I have an individual essence.
Even if I do not have an individual essence, some of my properties are essential
to mei.e., some of my properties are such that I have them necessarily. Suppose
that being a person is such a property. Now there are some worlds that do not
imply the property of being a person ("Some possible worlds don't contain any
persons"). If I am necessarily a person, then I am necessarily such that none of
those impersonal worlds obtain. We may say that I exclude such worlds. (Or we
could put it the other way around and say that I am such that I am excluded by
certain worlds.) For we may say that a thing x excludes a world W, provided only
that jc is necessarily such that Wdoes not obtain.
It is important to note that from the fact that I am not excluded by a certain
world W, it does not follow that I exist in that world W. But if I am not excluded
by W, then I am eligible for W; that is, I am possibly such that W obtains.
If I were to have an essence E such that there is a certain world Wthat implies
E (i.e., a certain world Wthat is necessarily such that if it obtains then something
exemplifies ), then we could say that Wis necessarily such that I exist. But if
46
I have no individual essence, then we cannot say of any world that it is necessarily
such that I exist. And we cannot even say this of "the actual world"i.e., the
world that obtains. The latter point may be put somewhat loosely by saying that
this world could have obtained without me. If the world had obtained without me,
then someone else would have played the role that 1 happen to play in this world.
Indeed, if neither you nor I have individual essences, then the prevailing world
could have obtained with you playing my role and me playing yours. One might
say, paradoxically, that you and I would have been very different but the world
would have been the same; this gives us a use for the label "existentialism."3
We cannot say, therefore, that "x exists in W'means the same as "If W were
to obtain, then jc would exist." For I exist in this world, but this world could have
obtained without me.
An Objection Considered
Consider now an objection to what we have been saying: "(1) Your theory implies
this: that world Wl is identical with a world W2 if and only if Wl implies the same
states of affairs that W2 implies. But (2) it is possible (a) that there is a world W1
implying exactly the same properties as are implied by a world W2 and yet (b)
W1 may be such that a certain personsay, Jonesis in it, and W2 may be such
that Jones is not in it. Therefore (3) your theory is false."
We are now in a position to see that even if the existentialistic intuition is true,
the second premise of the argument is false. The point is not that there are two
worlds, W1 and W2, such that Jones is "in" one of them and not "in" the other.
It is, rather, that there is a world W1 (= W2) that is such that Jones is eligible
for W1 but does not have an essence that W1 implies. One might also put this by
saying: W1 is possibly such that it obtains and Jones does not exist. Jones is "in"
W1, but W1 could have obtained without him.
To say that you and I "could have changed places" in W1 is to say this: there
is a certain set A of properties such that I have the members of A; there is another
set B of properties such that you have the members of B\ I am possibly such that
I have the members of B and W1 obtains; and you are possibly such that you have
the members of A and W1 obtains.
We should distinguish the following locutions:
(p) It is possible there are no persons;
(q) I am such that possibly there are no persons;
(r) I am possibly such that there are no persons;
(s) I am possibly such that I exist and there are no persons.
There is an ambiguity in (r), for it may be taken to say the same thing as (q), or
it may be taken to say the same thing as (s). But although (s) implies (q), it is not
the case the (q) implies (s). For if I am essentially a person, then although (p) and
(q) will be true, (s) will be false.
Suppose a thing has a certain property necessarily. What does this imply with
47
respect to those characteristics that the prevailing world has necessarily? Next to
nothing, I would say. For example, Socrates may be necessarily such that he is
a person. It does not follow from this either (a) that Socrates is necessarily such
that he is a person in this world, or (b) that this world is necessarily such that Socrates exists and is a person.
The fact that I exist only in the prevailing worldif it is a factdoes not restrict my possibilities. The unrealized possibilities of a given individual are not
to be explicated in terms of the different worlds in which that individual might
be said to exist. We may speak of such possibilities, using the undefined de re
modal locution with which we began: "x is possibly such that it is F." And so we
may say of a person who is not a physician that "he could have been a physician."
This does not tell us that he is a physician "in some possible world." It tells us
no more nor less than that he is possibly such that he is a physicianthat none
of his essential properties are incompatible with being a physician. "He is possibly
such that he is a physician" does not tell us that he has a nature that is compatible
with being a physician. For what would this "nature" be if the person does not
have an individual essence? "He is possibly such that he is a physician" tells us,
rather, that he does not have an essence that is incompatible with his being a physician.
"If he is possibly such that he is a physician, and if he's not a physician in the
prevailing world, then isn't it the case that, if he were a physician, he would be
a physician in another world?" This may be correct. But from this fact it does
not follow that he is a physician in any other world. For unless he has an individual essence, any other world could obtain without him.
An Argument for the Doctrine of Haecceities

Alvin Plantinga writes: "Is there any reason to suppose that 'being identical with
Socrates' names a property? Well, is there any reason to suppose that it does
not? . . . Surely it is true of Socrates that he is Socrates and that he is identical
with Socrates. If these are true of him, then being Socrates and being identical
with Socrates characterize him; they are among his properties or attributes".4 One
could argue similarly, as I did in the book Person and Object, that the expression
"being me" has the property being me as its sense and that the property of being
me is essential to me and repugnant to everything else. Then one could conclude
that being me is my haecceity.
But why assume that terms such as "Socrates" and "I" have senses? One can
describe their use adequately without such an assumption. The function of proper
names in a language may be described without presupposing that they have
senses. The same applies to pronouns and demonstratives. We may say that the
primary use of the locution "I am F" in English is that of conveying that one has
attributed to oneself the property of being F. The primary use of "That is F" is
that of conveying that the thing one is calling attention to has the property of being
F.5
48
Of course, it is true that Socrates is necessarily identical with Socrates and that
I am necessarily me. But I see no reason for supposing that these statements commit us to anything other than Socrates, me, and the property of being selfidentical.6 What if one were to say: "That thing is necessarily identical with that
thing; therefore, there is the property of being identical with that thing that constitutes the sense of 'that thing' "? How do I distinguish this property from the one
I may refer to later when, pointing to a different object, I say: "And that thing
is necessarily identical with that thing"? Surely I don't contemplate two essences,
one for the first occurrence of "that thing" and another for the second.
It has even been suggested that those haecceities that constitute the senses of
"Socrates," "that thing," and "me," respectively, are ontologically dependent upon
Socrates, that thing, and me. To say that one thing is "ontologically dependent"
upon another is, I suppose, to say that the first thing is necessarily such that the
second exists. But if what I have suggested is correct, haecceities are abstract objects, and no abstract object is dependent for its being upon any particular thing.
I see no reason, then, to assume that the fact of unrealized possibilities presupposes that individual things have essences or haecceities.
Notes
1. If we wanted to say that there are also "impossible worlds," we should remove the final clause
from the definiens. But no useful purpose is served by speaking in this way.
2. Compare Leibniz, Theodicy (London: Routledge and Kegan Paul, 1952), part III, sec. 414:
"These worlds are all here, that is, in ideas [Ces Mondes sont tous id, c'est-d-dire en idees}. I will
show you some, wherein shall be found, not absolutely the same Sextus as you have seen (that is not
possible, he carries with him always that which he shall be) but several Sextuses resembling him."
Compare C. I. Lewis, An Analysis of Knowledge and Valuation (La Salle, 111.: Open Court, 1946),
Chap. Ill ("The Modes of Meaning"), esp. p. 56.
3. See Alvin Plantinga, "De Essentia," Grazer Philosophische Studien 7/8 (1979); 101-21.
4. "World and Essence," in M. J. Loux, Universals and Particulars (Notre Dame, Ind.: Notre
Dame University Press, 1976), pp. 369-70. Given (i) that Socrates has a haecceity and (ii) that he
exists in a world other than the world that obtains, Plantinga can go on to argue, as he does, that if
Socrates is snubnosed in W, then he is necessarily such that he is snubnosed in W. But if what I have
said is correct, then it is not the case that individuals have such "world-indexed" properties necessarily. I don't exist in any worlds other than the world that obtains; the world that obtains is not necessarily such that I exist if it obtains; and I am not necessarily such that I exist in it.
5.1 have tried to show this in detail in The First Person (Minneapolis and Brighton: University
of Minnesota Press and Harvester Press, 1981), Chap. 6, and in "The Primacy of the Intentional,"
Synthese (1984): 89-109.
6. Some philosophers have been unduly impressed by the fact that quantified modal logic happens to have been developed in such a way as to contain the theorem "(x) N (x = x)." The philosophical
question is whether there is anything for this theorem to express other than the proposition that everything is self-identical.
Coming into Being

and Passing Away:
Can the Metaphysician Help?
What We Have a Right to Believe about Ourselves
I assume that, in our theoretical thinking, we should be guided by those propositions we presuppose in our ordinary activity. They are propositions we have a
right to believe. Or, somewhat more exactly, they are propositions that should
be regarded as innocent, epistemically, until there is positive reason for thinking
them guilty.
A list of such propositions would be very much like the list of propositions with
which G. E. Moore began his celebrated essay, "In Defence of Common Sense."1
The list I might make for myself may be suggested by the following: "(1)1 am
now thinking such and such things and have such and such beliefs, feelings, and
desires. I now see people, for example, and I don't see any unicorns. And I have
thought such and such other things in the past and then had such and such other
beliefs, feelings, and desires. (2) I now have a body of such and such a sort and
I had a body of such and such a different sort in the past. And (3) I am now acting
with the intention of bringing about such and such thingsand I could instead
have acted with the intention of bringing about such and such other things. In the
past I have acted with the intention of bringing about such and such other things,
and I could then have acted with the intention of bringing about such and such
other things instead."
There is such a list that each of us might make. The items on the list are obvious
and, one might think, so trivial as not to be worth mentioning at all. Yet some
of those who now say that these things are trivially true may say, at the end, that
strictly speaking they are false. But one can't have it both ways.
A sceptic may ask: "Might it not be that I'm mistaken in suggesting there are
these things we know about ourselves? Isn't it possible that I'm deluded about
these things?" Of course, it's possible that I am deluded about these things. It's
49
50
COMING INTO BEING AND PASSING AWAY
possible that I will wake up in a few minutes and find myself in a hospital. But
from the fact that it is thus theoretically possible that I am deluded it hardly follows that it is now reasonable for me to think that I am in fact deluded. Until you
give me some very good positive reason to think the contrary, it is now reasonable
for me to assume that I am in a room with other people and not suffering from
hallucination or delusion.
These are some obvious truths about myself, then, which it is now reasonable
for me to accept. But these truths, if we take them at their face value, imply that
I am an ens per se, that is to say, they imply that, in the strictest sense of the word
"is," there is a certain thing which is I.
Now some philosophers have held that the word "I" is a logical construction,
a mere/afon de parler, like the expressions "5" and "the average plumber." One
can show that "5" and "the average plumber" are logical constructions. For one
can take sentences in which these ostensible terms occur ("5 is 7 less 2" and
"the average plumber has 2.6 children") and translate them into other sentences
no longer containing terms ostensibly referring to 5 and the average plumber.
If I, too, am simply a logical construction, orfagon de parler, then the various
truths on the list that I have made could be reformulated without reference to me.
They could be re-expressed in new sentences which contain no terms, such as the
word "I", which ostensibly designate me. The new sentences might contain terms
designating what I now call my sensations, as Ernst Mach and Bertrand Russell
once thought, or they might contain terms designating my body or certain parts
of my body, as other philosophers have thought. But these various philosophers
do not know that the word "I" is thus a logical construction, a mere faqon de
parler. No one can take the sentences I have citedthe truisms about myselfand
translate them without loss of meaning into sentences referring only to things
other than me. If you think that I am mistaken about this, just consider the truth
which I can now express by saying, "I don't see any unicorns," and try to put exactly what that says in sentences which don't refer to me.
Perhaps you will say on reflection: "Well, I can't do it now, but maybe some
day somebody will do it." This would be like the sceptic we just considered. It's
possible you can show I'm a mere^fon de parler. And it's possible that I am now
lying in a hospital bed somewhere. But what have these mere possibilities got to
do with what is going on in fact? They certainly do not mean that I am deluded
with respect to the truisms that I began. No one has been able to show that these
truths can be paraphrased as truths about some entity or entities other than myself.
And therefore no one has been able to show that I am not an ens per se.
I say, then, that we have a right to assume that persons are entia per se, that
there are persons, in the strict and philosophical sense of the expression "there
are." You and I, in short, are real things and the terms that designate us are not
linguistic fictions. But if there are persons, in the strictest sense of the expression
"there are," then persons are such that either (i) they exist forever or (ii) they come
into being but will not pass away or (iii) they will pass away but never come into
being or (iv) each is such that it came into being and will pass away. I assume
51
that the last of these four possibilities is the one that is most likely. You and I exist
now but there was a time before which we did not exist and there will be a time
after which we will no longer exist.
We should be clear, at the outset, about one very simple point. The concepts
of coming into being and passing away are not merely physiological concepts.
Consider the relation, for example, between the concepts of passing away and
dying and assume (what, of course, is doubtful) that the latter concept is pretty
clearly fixed. If by "passing away," we mean, as I do, ceasing to be, then if we
say that the body dies we cannot say that it passes awayfor the body continues
to exist but in such a way that it is no longer alive. It may well be that, when the
body dies, then the person whose body it is passes awaythat the person ceases
to be. But the two concepts are different. It's logically possible that, when the person's body dies, then the person does not cease to be. And it's also logically possible that the person ceases to be before his body dies. To admit this distinction
is not to say that there is any likelihood that these possibilities are actual. The
point is only that it is one thing to say that a person's body has died and it is another
thing to say that the person has ceased to be, even if in fact the two events
coincide.
And, similarly, the concept of coming into being is not the same as any physiological concept. Conceivably there is some physiological event which coincides
uniformly with the coming into being of a person, but to say that this physiological event occurs is not the same as saying that a person comes into being.
The points I have just made are typically philosophical. I know they will bring
forth two quite different reactions. One reaction will be: "But why insist upon
what is trivial and obvious?" And the other reaction will be: "What you say is obviously false." As long as there are people who react in the second way, and I
know that there are such people, it is worthwhile to insist upon what is obvious,
even if it is trivial. For, as Aristotle said, if you deny what is trivial, then there
is no hope for your investigation.
Of course there may be philosophers or there may be people practicing medicine who don't think there are any persons. And this means, if they are consistent,
that they don't believe with respect to themselves that they ever came into being
or that they will ever pass away. For people who do really believe that, I have
no messageexcept to urge them to think again.
Alteration
One of the ways in which a metaphysician can help a nonmetaphysician is to protect him from bad metaphysics.
People are sometimes led to think that nothing persists through any period of
time and hence that all things are constantly ceasing to be and new things are constantly coming into being to replace them. This was the view of Heraclitus who
said "You cannot step into the same river twice." (One of Heraclitus's followers,
according to Aristotle, held that things are in such constant flux that you can't even
52
step into the same river once.) If this view is true, then it would be incorrect to
say that you and I have existed for any period of time. The things that bore our
names at any given moment yesterday have since then ceased to be and you and
I are no more the same people as those people of yesterday than we are identical
with each other. This view is a disastrous beginning, if our aim is to understand
coming into being and passing away.
Why would anyone think that such a thing is true? Respectable philosophers
I regret to say, have accepted this view. When philosophers don't simply pick
their theories out of the air, they arrive at them in attempting to deal with philosophical puzzles. The kind of puzzle that has led philosophers to think that everything is in flux, in the sense in question, may be illustrated as follows.
You say to me: "I see you have a new fence in your back yard." I say: "No,
it's the same fence I've always had." You say: "But your fence is red; the fence
you used to have was white." I say: "No, it's the same fence; I painted it, that's
all." And you say: "But it couldn't be the same fence. If something A is identical
with something B, then whatever is true of A is true of B. But if today's fence
is identical with yesterday's, how can it be that the old one is red and the other
is white?"
Very great philosophers, I'm afraid, have stumbled over that one. (Some have
been led to conclude not that everything is in flux, but that things can be identical
with each other even though they don't have all their properties in common.) What
went wrong in the dialogue we have just imagined?
Consider the sentence: "Today's fence is red and yesterday's fence was white."
One trouble with it is that the dates are in the wrong place. For what we know
is not merely that there was something that was yesterday's fence and that was
white. It is rather that there is something that is a fence and that was white yesterday. And it's not merely that there is a thing that is today's fence and is red. It's
rather that there is a fence that is red today. The fence I have now and the fence
I had yesterday have all their properties in common. I have had just one fence
one that is red today and that was white yesterday.
If you don't see the error involved in using the expressions "today's fence" and
"yesterday's fence," perhaps this analogy will help. Consider someone who reasons as follows: "Mr. Jones the husband is very meek and submissive. Yet Mr.
Jones the father is extremely authoritative and overbearing. But one and the same
thing can't be meek and submissive and also authoritative and overbearing. Therefore there are two Mr. JonesesMr. Jones the husband and Mr. Jones the father."
Saying what went wrong in this case is like explaining a joke. But perhaps we
should risk it. It's not that Mr. Jones the husband has properties that are different
from those that Mr. Jones the father has. It's rather that Mr. Jones is such that
he is meek and submissive toward his wife and overbearing and authoritative toward his children.
All this is to spell out, once again, what ought to be obvious. But let us keep
the moral in mind: The fact that a thing has altered in a certain way does not imply
that the thing has ceased to be and that some new thing has come into being.
53
Coming into Being and Passing Away

Secundum Quid
We are assuming, then, that persons you and Iare real things, entia realia.
And we are also assuming that they come into being and pass away. But this
means that the coming into being and the passing away of persons is also the real
thing. That is to say, it's not a pseudo kind of coming into being and passing away;
it's not that kind of merely apparent coming into being and passing away that some
of the Scholastics called coming into being and passing away secundum quid. Let
us consider for a moment this pseudo kind of coming into being and passing away.
In his book Generation and Corruption, Aristotle considers a man who had
a talent for music and who then lost this talent but continued to exist for some
time thereafter. Aristotle described this fact by saying: "The musical man passed
away and an unmusical man came to be, but the man persists as identically the
same" (I, 4, 319b). One is inclined to ask: "Why on earth did Aristotle express
himself that way? After all he was just talking about a certain alteration. Why
didn't he say that a man ceased to be musical, instead of saying that a musical
man ceased to beT
It should be noted that we, too, sometimes talk that way. That is to say, we
sometimes use such expressions as "coming into being" and "passing away" to describe what is in fact the mere alteration of a persisting subject. Thus you may
ask me: "How is our old friend Jones?" And I may reply by saying: "The Jones
you knew doesn't exist any more. He's just dull and bitter now." A reporter once
wrote, after visiting one of our great comedians in a nursing home in California:
"Alas, the great comedian is no more." But the manthe man who had been a
comedianpersisted for some time after that.
It was a result of Aristotle's way of talking ("the musical man ceased to be")
that medieval philosophersSt. Thomas, for examplecame to distinguish between (1) coming into being and passing away per se and (2) coming into being
and passing away per accidens or secundum quid.
We have coming into being per accidens, or secundum quid, when a thing
alters in some way or otherwhen there is something which so changes that it
first has a certain property and subsequently has a certain other property instead.
This is what happens if a musical man is said to cease to be and an unmusical man
to come into being when in fact one and the same man persists through the change.
Let us consider some other examples.
It is sometimes said that, when one becomes aware of a feeling or of a sensation, then the feeling or sensation is something that comes into being ex nihilo.
When British and American philosophers, in the first third of the century, were
concerned about the status of what they called appearances or sense-data, they
took very seriously the possibility that these are things that come into being, ex
nihilo, when the appropriate physiological and psychological conditions obtain.2
But isn't the fact of the matter that, when a feeling or sensation is thus said to
"come into being," what actually happens is that the person or subject is simply
54
altered in a certain way? In making me feel sad, for example, what you do is not
to cause a feeling of sadness to come into being ex nihilo, but to cause me to have
a certain propertythat of feeling in a certain way. An analogously for making
me aware of an "appearance" or "sense-datum." What you do is simply to cause
me to sense in a certain way. But if this is true, then the so-called coming into
being and passing away of feelings and sensations is simply coming into being
and passing away per accidens or secundum quid and not coming into being and
passing away per se.
Let us consider another type of case. One might say: "If I turn the light on over
our heads, I will make a shadow come into being out of nothing. And if I then
turn the light out again, I will cause the shadow to go out of existencewithout
leaving any traces behind. And so isn't this genuine coming into being and passing
awaycoming into being and passing away per seT
I think the answer is no. But it is instructive to consider the case somewhat
further. What we conveniently describe as a shadow coming into being and passing away can also be described, somewhat less conveniently, as an alteration in
what we might call the shadowed object, or the shadowed objects. When I create
a shadow on the floor, what I do is merely cause a certain part of the floor to be
darker, to reflect less illumination than it had before. And when I make the
shadow cease to beto disappear without remainderall that I do is to cause the
relevant parts of the shadowed object to reflect light once again. So we don't have
a coming into being and passing away per se of shadows. All we have is a coming
into being and passing away per accidens or secundum quida. mere alteration
in the shadowed object.
Let us note that a shadow is a paradigm case of what some medieval philosophers called an ens in alioand what we might call an "ontological parasite." Entia in alio were thought of as things that got all their being, so to speak, from other
things. Thus a shadow has no being of its own. Anything we seem to be able to
say about it is something that really is a truth just about some shadowed object
or other. The shadow is entirely parasitical upon its object. And this is really to
say that there aren't such things as shadows.
Whatever thus comes into being or passes away secundum quid is not a real
thing; it is an ontological parasite, at best a mere faqon de parler.
Sometimes we say that a certain thing x became a certain other thing y. And
we take our statement to imply that the first thing x then ceased to be and the second thing y then came into being. In such a case, we are speaking of coming into
being and passing away secundum quid; x and y are ontological parasites and not
entia per se. This may be seen as follows.
If the first thing became the second thing, then we may say:
(1) There exists a z such that z once was x and z now is y.
Suppose now we add
(2) x has now ceased to be and y hadn't yet come into being when z was x.
55
If we taking "coming into being" and "passing away" literally and thus mean coming into being and passing away per se, then our two statements will imply
(3) There exists a z such that (i) z was once identical with x, (ii) z is now identical with y, and (iii) x but not y no longer exists.
But (3) is absurd. Therefore, if (1) and (2) are true, they must be taken to refer
to coming into being and passing away secundum quid. The fact of the matter was
simply that z was altered in a certain way.
If at a certain time, a thing literally becomes identical with something it hadn't
been identical with before, then the thing came into being at that time and it wasn't
identical with anything before.3
Leibniz's Law implies that nothing can change its identity. That is to say, it
is impossible for there to be anything that is identical with a certain thing at one
time and diverse from that thing at another time. ("The G is diverse from the H,"
unlike "The G is such that it is not identical with the H," implies "There is something which is identical with the H.") For suppose that there is something, say,
the G, that is identical with the H today and will be diverse from the H tomorrow.
If the H is now identical with the G, then anything that is true of the G is also
true of the H. Therefore if the G will be diverse from the H tomorrow, then the
H will be diverse from the H tomorrow. But this consequence is absurd.
"Last year the President of the United States was identical with Mr. So-and-so.
This year the President of the United States is identical with Mr. Such-and-such.
And Mr. Such-and-such is diverse from Mr. So-and-so. Therefore things can
change their identity and be identical at one time with what they are diverse from
at another time."
Surely the argument does not merit refutation. If we restrict the use of "there
is" to its strict and proper sense, we will not take the premises of the argument
to imply that there isor wasa certain thing x such that x was once identical
with Mr. Reagan and x is now identical with Mr. Bush. That x is not even an ens
per accidens.4 Leibniz's law implies, then, that there cannot be substantial
change. An individual thing or substance may come into being; it may be altered
in various ways and it may pass away; but it cannot become another individual
thing or substance.
Elanguescence
Reflection on the coming into being and passing away of sensations, feelings, and
shadows may bring to mind a monstrous hypothesis proposed by Kant in the Critique of Pure Reason, Different things, he said, may have different degrees of
reality. It is possible, he thought, for the degree of reality of a thing to increase
or to decrease in a continuous manner. And so, he said, a thing "may be changed
into nothing, not indeed by dissolution, but by gradual loss (remissio) of its
powers, and so, if I may be permitted the use of the term, by elanguescence."5
Kant is to be taken literally here. He was clear that existence is not a predicate.
56
Yet he thought that some things could have more existence than others. It is as
though he thought that there is a path between being and nonbeing, so that one
day you may set out from nonbeing and head in the direction toward being with
the result that the farther you go in that direction the more being you will have.
But surely there is no mean between being and nonbeing. If something is on a
certain path, then that something is. Or if it isn't yet, then it can't be on the path
between being and nonbeing.6
Of course things may be more or less endowed. But things cannot be more or
less endowed with respect to being. What is poorly endowed is poorly endowed
and therefore is.
One might object: "Consider an intense pain that becomes less and less intense
and finally fades away. Doesn't it become less and less real and thus gradually
cease to be?" The objection would ignore the point we have just made about ceasing to be secundum quid. When we say that the pain gradually faded away, we
are talking about the alteration of a person; we are speaking about the way in
which a person felt or the way he experienced something. Thus one might say,
similarly, that the feeling of sadness faded away and finally ceased to be altogether. But the fact of the matter is only this: a person felt less and less sad until
he finally reached the point where he didn't feel sad at all. And we should remind
ourselves, moreover, that even if we do reify pains and feelings of sadness, we
have no ground whatever for saying that the feeling that is less intense is less real
than the feeling that is more intense.
"But if one thing has more properties than another, isn't it more real than the
other?" No thing has any more or any less properties than does any other thing.
Every property and every thing is such that either the thing has that property or
the thing has the negation of that property. If you can play the viola and I cannot,
you don't have more properties thereby than I.
To be sure, you have the property of being someone who can play the viola
and I don't have that property. But I have the property of being someone who cannot play the viola and you don't have that property.
Coming into Being and Passing Away Per Se

What, then, is the strict sense of coming into being and of passing away? I suggest
the following tensed definitions:
Dl x comes into being = Df There is a property which is such that x has it,
and there is no property which is such that x had it
D2 x has just passed away = Df Something that was such that x exists begins
to be such that x does not exist
One question that was discussed at length by medieval philosophers is: "When
a thing comes into being, does the thing existor is it the case merely that the
thing is going to exist. And when it passes away, does the thing existor is it
57
the case merely that the thing once existed.7 These definitions presuppose that the
thing exists when it comes into being and also that it exists when it passes away.
The Coming into Being and Passing Away of Persons

If persons are real things or entiaper se, then the coming into being and passing
away of persons is not a matter merely of something or other being altered in a
certain way. When my body dies, then it is altered in a certain way. And if it happens to be the case that I then cease to be, as it well may be, then my ceasing to
be is not just the fact that my body has been altered in a certain way. And when
I came into being, this may well have been at the time of a certain alteration of
the fetus, or of a certain alteration of matter that was going to become a part of
the fetus. But whatever alteration that may have been, that alteration was not the
same event as my coming into being. For our assumptions imply that persons are
entia per se and not entia in altonot ontological parasites like shadows.
What does all this have to do with the facts of biology and physiology, with
the questions about when human life begins and ends? Not very much, I'm afraid.
What I've said so far won't help anyone in dealing with those questions. But if
you begin at the point at which I've begun, you will want to put the question first
the other way around. What do the facts of biology and physiology, the things
we know about the beginning and ending of human life, tell us about the coming
into being and passing away of persons? Here, too, I'm afraid, the answer is not
very much-or not very much as far as anyone can possibly know. But there may
be some relevant points that the metaphysician can make.
The United States Supreme Court decreed, in the case of Roe v. Wade, in
1973, that the fetus before a certain stage of development is not a "person in the
whole sense." Possibly the ruling presupposes Kant's absurd hypothesis about
degrees of reality. Then it would be telling us that, in its fetal stage, the person
is somewhere between being and not being ("On the one hand, it doesn't really
exist, and on the other hand, it doesn't really not exist"). But it would be more
charitable not to assume that the court was presupposing bad metaphysics. And
it is more likely that the court meant only that becoming a human being is a
gradual process: the fetus is on the way to becoming a human being but, at its
early stages at any rate, hasn't gotten there yet. One could take a similar view
about the process of ceasing to be a human being. The one who is moribund is
gradually ceasing to be a human being; in the early stages of his illness he is still
a human being, but in the later stages not.
This view has recently been set forth by Lawrence C. Becker in Philosophy
and Public Affairs.8 It may be summarized in the two theses: "Entry into the class
of human beings is a process" and "Exit from the class of human beings is a process."9 The expression "human being" is certainly a proper term of biology and
physiology; one cannot quarrel with these theses on terminological grounds. But
I am not at all convinced that this gradualistic theory, even if it is true, will help
us very much in dealing with the philosophical and ethical questions involved in
58
the coming into being and passing away of those things that may thus gradually
become or cease to be human.
To see that these theses are probably not helpful, let us consider the consequence of assuming that they are true. We may do this by relating them to what
we have already said.
Consider the process of becoming a human being. (As Professor Becker makes
clear, much of what we can say about the process of becoming a human being
can also be said, mutatis mutandis, about the process of ceasing to be a human
being.) Let us consider this thesis, that entrance into the class of human beings
is a gradual process, and take it together with what we have already assumed.
Thus we have:
(1) I am one of the members of the class of human beings.
(2) There was a time at which I did not exist.
(3) Entrance into the class of human beings is a gradual process.
Let us now consider our three premises together. I am as certain as I am of
anything that the first of these premises is true. And I don't think there are many
of us who are prepared to challenge the second. The third premise is the statement
of the biological hypothesis we are now considering.
Our premises, quite clearly, have these two consequences:
(4) There was a time at which I was not one of the members of the class of
human beings.
(5) My entrance into the class of human beings was a gradual process.
The second of these consequences"My entrance into the class of human beings
was a gradual process"may suggest the process of entering a room. If we consider a man who is entering a room, we may say that his entrance is gradual in
this sense: it begins with the entrance of the front part of one of his feet and this
is followed by the entrance of more and more parts of his body. Then, when he
gets them all in, he has entered the room. But perhaps a more accurate figure
would be that of a sober man who becomes drunk: his entry into the class of the
people who are drunk might be thought to be gradual.
But consider this further consequence:
(6) There was a time at which I existed but had not yet entered the class of
human beings.
If I went through the process of becoming a human being as (5) tells us, then I
wasn't already a human being when I started to go through this process.
What (6) tells us can be rephrased this way:
(7) My corning into being antedated my entry into the class of human beings.
Consider Aristotle's conception of the musical man once again. Aristotle might
have said: "A musical man came into being but the man himself had existed long
59
before." And then he could have said that the man's coming into being antedated
his entrance into the class of musicians. For the man can become more and more
musical without thereby coming into being, just as a man can become more and
more drunk without thereby coming into being. Analogously, one could say that
I came to be more and more human but without thereby coming into being. In
each case, the thing that went through the process of gradual entrance is assumed
to have antedated that process.
This is a consequence, then, if we take what is obvious and combine it with
the thesis that "entrance into the class of human beings is a process."
If entrance into the class of human beings is a process, then my coming into
being antedated my entrance into the class of human beings. This means that the
event that is my coming into being is not the same as that event which is my entrance into the class of human beings. There was a time, before I entered the class
of human beings, when I existed. And so, if someone at that time could have
caused me to cease to be, my ceasing to be as well as my coming into being would
have antedated my entrance into the class of human beings.
If in the future someone causes me gradually to leave the class of human beings, then, while he's doing this, while I am gradually leaving the class of human
beings, I will be there to make the exit, and the man will not yet have caused me
to cease to be. When I'm half way out of the room, I'm somewhere, partly in the
room, partly outside, and partly in the doorway. Perhaps, once you've gotten me
all the way out of the class of human beings, then you will have caused me to cease
to be. But I suppose no one knows.
Would it help if we replaced the concept of entrance by the pair of concepts,
full and partial entrance? Then we could distinguish between^// and partial entrance into the class of human beings. And we could also do this in the case of
entering a room. As soon as I get a part of my body in the room, then, however
small the part may be, I have partially entered the room. And it is not until I have
all the parts of my body in the room that I can say that I have fully entered the
room. But I don't think this will help. For if we replace "entrance" by the two concepts "full entrance" and "partial entrance," then we have to give up the process
theory. One has only to reflect just a little to see that both partial and full entrance
can only be instantaneous.10
I am certain, then, that this much is true: if I'm a real thing and not just afaqon
de parler, then neither my coming into being nor my passing away is a gradual
processhowever gradual may be my entrance into and my exit from the class
of human beings.
Now if we give the biologist and physiologist the term "human," perhaps we
have a right to use the term "person" for the sort of thing that you and I are. Suppose now we define a person in terms of what it could become. We might say,
for example, that a person is a thing that is such that it is physically possible (it
is not contrary to the laws of nature) that there is a time at which that thing consciously thinks.11
If we thus define a personas that which is necessarily such that it is physically
60
possible that there is a time at which it consciously thinksthen we cannot say

that anything gradually becomes a person or gradually ceases to be a person. For
if a thing has the property of being necessarily such that it is not contrary to the
laws of nature that there is a time at which it consciously thinks, then it has that
property from the moment it comes into being until the moment it passes away.12
And so the questions we thought we escaped with our gradualistic concept of being a human may arise once again with the concept of a person.
The Moral of the Story

If all of this is right, as it seems to be, then no one could have known just when
it was that I came into being. And no one will know just when it is that I will pass
away. Or perhaps the latter point should be put more cautiously: the present state
of our knowledge is such that, if I have the misfortune to be one of those people
who, as Lucretius put it, "leave the light dying piecemeal," then no one will know
just when it is that I will pass away.
Hence it may be, for all anyone knows, that by terminating my mother's pregnancy at a certain very early stage, one could have caused me then to cease to
be. (But it may also be that I didn't come into being until after that stage.) And
it may be, for all anyone knows, that, by disconnecting a life-sustaining device
at a very late state in my gradual exit from the class of human beings, you will
then cause me to cease to be. (But it may also be that I already ceased to be, sometime before that.)
Analogous things may be said about you and about everyone else.
So where does this leave us with respect to the moral problems that are involved in causing someone to cease to be? Surely it is right, sometimes, to terminate a pregnancy or to disconnect a life-sustaining device. Doubtless such acts
always call for an excuse.13 But let's not pretend that, when we perform them,
probably we are not causing anyone to cease to be. Let's have the courage to face
the moral facts of the matter: occasionally it is right for one person to annihilate
another.
Notes
1. See G. E. Moore, Philosophical Papers (London: George Allen & Unwin, 1959), pp. 32-59.
2. Compare C. D. Broad's discussion of "Causation and Creation" in Scientific Thought (London:
Kegan Paul, Trench, Trubner, 1925), pp.535-43.
3. Aristotle attempts to circumvent this conclusion with his doctrine of prime matter and substantial change. I cannot believe that he was successful. But I think it is clear that, if one does wish to
circumvent this conclusion, then one must appeal to a concept that is very much like that of prime
matter.
4. According to Aristotle, accidents cannot be transferred from one subject to another.
5. See the "Refutation of Mendelssohn's Proof of the Permanence of the Soul," in Critique of
Pure Reason, B414; the passage appears on p. 373, of the Norman Smith translation (London: Macmillan, 1933). Compare also Kant's discussion of the "degrees of reality" in the "Anticipation of Perception," B207-9.
61
6. Compare the criticism of Kant's doctrine in Franz Brentano, The Theory of Categories (Th
Hague: Martinus Nijhoff, 1981), pp. 74-77.
7. If all beginnings are instantaneous, then no ending is instantaneous. For suppose (i) there was
an instant at which Socrates ceased to be and (ii) an instant at which his survivors begin to be such
that he no longer exists. Then either these instants are the same (in which case the instant is one at
which Socrates both exists and does not exist) or the second instant is later than the first (in which
case, since between any two instants there are other instants, Socrates would neither exist nor not exist
during the intervening instants). But to say that there is thus no instant at which a thing passes away
is not to say that the thing passes away gradually. See Norman Kretzmann, "Incipit/Desinit," in Paul
K. Machamer and Robert G. Turnbull (eds.), Motion and Time, Space and Matter (Columbus: The
Ohio State University Press, 1976), pp. 101-36.
8. Lawrence C. Becker, "Human Being: The Boundaries of the Concept," Philosophy and Public
Affairs, 4 (1975): 334-59. Compare W. R. Carter, "Do Zygotes Become People?" Mind, 91 (1982):
77-85.
9. Becker, "Human Being," 335, 336.
10. We can, of course, retain proposition (6) abovei.e., "There was a time at which I existed
but had not yet entered the class of human beings"if we replace "entered" by "fully entered." What
if we replace it by "partially entered"? For all we know, the result might be a proposition that is false.
It may be that, from the time I did come into being, whenever that was, I already had one foot in
the door, so to speak, and was pan way into the class of human beings.
11. The moral philosopher might insist upon defining a person as a thing having rights of a certain
sort. If now we should give him the term "person," then we might appropriate the term "self and consider our definition as a definition of a self.
12. And so we are saying more than that persons are things that are "potentially thinkers." For
if we take "potential" in its ordinary sense, then we may say that our potentialities are variable and
dependent on our circumstances at any particular time. But our potentialities, in this sense of the term,
are a function of what it is physically possible for us to bea function of what the laws of nature do
not preclude us from being. And physical possibilities, in this latter sense, are invariable. I have attempted to distinguish these various senses of possibility in more detail in Person and Object: A
Metaphysical Study (La Salle, 111.: Open Court, 1933), Chap.2.
13. Can part of the excuse be that the persons involved aren't then humans in the complete sense?
This moral question falls outside the scope of the present paper.
Part III
Parts and Wholes
Parts as Essential to Their Wholes

1
One kind of philosophical puzzlement arises when we have an apparent conflict
of intuitions. If we are philosophers, we then try to show that the apparent conflict
of intuitions is only an apparent conflict and not a real one. If we fail, we may have
to say that what we took to be an apparent conflict of intuitions was in fact a conflict
of apparent intuitions, and then we must decide which of the conflicting apparent
intuitions is only an apparent intuition. But if we succeed, then both of the intuitions will be preserved. Since there was an apparent conflict, we will have to conclude that the formulation of at least one of the intuitions was defective. And though
the formulation may be embedded in our ordinary language, we will have to say
that, strictly and philosophically, a different formulation is to be preferred. But to
make it clear that we are not rejecting the intuition we are reformulating, we must
show systematically how to interpret the ordinary formulation into the philosophical one. The extent to which we can show this will be one mark of our success
in dealing with the philosophical puzzle. Another will be the extent to which our
proposed solution contributes to the solution of still other philosophical puzzles.
I shall consider a philosophical puzzle pertaining to the concepts of whole and
part. The proper solution, I believe, will throw light on some of the most important
questions of metaphysics.
The Presidential Address delivered at the twenty-fourth annual meeting of the Metaphysical Society
of America, Tulane University, March 16, 1973.
65
66
PARTS AS ESSENTIAL TO THEIR WHOLES

II
The puzzle pertains to what I shall call the principle of mereological essentialism.
The principle may be formulated by saying that, for any whole x, if x has y as one
if its parts then y is part of x in every possible world in which x exists. The principle
may also be put by saying that every whole has the parts that it has necessarily,
or by saying that if y is part of x then the property of having y as one of its parts
is essential to x. If the principle is true, then if y is ever part of x, y will be part
of x as long as x exists.
Abelard held that "no thing has more or less parts at one time than at another."1
Leibniz said, "We cannot say, speaking according to the great truth of things, that
the same whole is preserved when a part is lost."2 And G. E. Moore gave us this
example:
Let us take as an example the relational property which we assert to belong
to a visual sense-datum when we say of it that it has another visual sensedatum as a spatial part: the assertion, for instance, with regard to a colored
patch half of which is red and half yellow: "This whole patch contains this
patch" (where "this patch" is a proper name for the red half). It is here, I
think, quite plain that, in a perfectly clear and intelligible sense, we can say
that any whole, which had not contained that red patch, could not have been
identical with the whole in question: that from the proposition with regard
to any term whatever that it does not contain that particular patch it follows
that that term is other than the whole in questionthough not necessarily that
it is qualitatively different from it. That particular whole could not have existed without having that particular patch for a part. But . . . it seems quite
clear that, though the whole could not have existed without having the red
patch for a part, the red patch might perfectly well have existed without being
part of that particular whole.3
Instead of considering such things as sense-data and visual patches, let us consider physical things. Let us picture to ourselves a very simple table, improvised
from a stump and a board. Now one might have constructed a very similar table by
using the same stump and a different board, or by using the same board and a different stump. But the only way of constructing precisely that table is to use that particular stump and that particular board. It would seem, therefore, that that particular
table is necessarily made up of that particular stump and that particular board.
But to say of the table that it is necessarily made up of the stump and the board
is not to say of the stump and the board that they are such that they are necessarily
parts of the table. And it is not to say that the stump is necessarily joined with
the board. God could have created the stump without creating the board; he could
have created the board without creating the stump; and he could have created the
stump and the board without creating the table. But he could not have created that
particular table without using the stump and the board.
Let us be clear about the view that is here set forth. It is no spurious essential-
67
ism. (That is to say, it is not the kind of essentialism that is arrived at in such arguments as these: "Szigeti was a violinist; necessarily all violinists are musicians;
therefore Szigeti was necessarily a musician"; and "The word 'Homer', as we use
it, connotes or intends being a person who wrote the Iliad and the Odyssey; therefore Homer, if he existed, was such that he necessarily wrote the Iliad and the
Odyssey.") We are saying, in application to our example of the table, that there
exists and x, a y, and a z such that: x is identical with this table, y is identical
with this stump, z is identical with this board, and x is such that, in every possible
world in which x exists, it is made up of y and z. Our statement says nothing whatever about the way in which human beings may happen to conceive or to look
upon such things as this table. And, a fortiori, it says nothing whatever about the
way in which we may happen to describe this table or use the language we do.
Its subject matter is no more nor less than this table, the parts of this table, and
the possible worlds in which this table exists.
Considered in the abstract and considered in application to such simple examples as these, the principle of mereological essentialism may seem obvious. Indeed, I would say that it ought to seem obvious. Yet the principle appears to
conflict with certain other truths that perhaps from a somewhat different point of
view,would also seem to be obvious. I will indicate these other truths by formulating two objections to the principle of mereological essentialism.
(A) "(i) My automobile had parts last week that it does not have this week and
it will have parts next week that it never had before. But (ii) the principle of
mereological essentialism implies that, if anything is ever a part of my automobile, then that thing is a part of my automobile as long as the automobile exists.
And therefore (iii) the principle of mereological essentialism is false."
(B) "(i) I could have bought different tires for my automobile, (ii) If I had
bought different tires for my automobile, then it would have had different parts
from those it has now. Therefore (iii) my automobile could have had different
parts from those it has now. Hence (iv) my automobile is such that, in some possible worlds, it has parts it does not have in this one. But (v) the principle of
mereological essentialism implies that in every world in which my automobile exists it has exactly the same parts it has in this one. And therefore (vi) the principle
of mereological essentialism is false."
Philosophers who are interested in the ways in which people ordinarily talk
may wish to multiply examples at this point. But I believe that our two examples
are enough.
I would say, then, that we have he.re a typical philosophical puzzlean apparent conflict of intuitions.
Ill
Before we try to solve the puzzle, let us consider the antithesis of extreme mereological essentialism. This would be what we might call complete, unbridled
mereological inessentialism.
68
Complete, unbridled mereological inessentialism would seem to be manifestly

absurd. This would be the view that, for any whole w, w could be made up of
any two things whatever. For, given such a view, one could say, of this table,
that it could have been made up of the number 36 and the property blue.
Perhaps it will be conceded that the set of things that are capable of being parts
of this table must be restricted in at least a general waysay, to things of the same
ontological category as the table. Suppose, then, one says that, for any two physical objects, this table could have been made up of those two objects.
If the view is true, then this table, this physical thing that is before us now,
is such that it could have been made up of my left foot and the Grand Central Station. Or, to be more exact, if extreme mereological inessentialism is true, then
this table, my left foot, and the Grand Central Station are three things that are
such that there is a possible world in which the first is made up of the second and
thirdin which this table is made up of what, in this world, are my left foot and
the Grand Central Station.
Indeed, there would be indefinitely many such possible worlds. In trying to
imagine this table being made up of my foot and the station, perhaps we thought
of my foot and the station as they now are, with all the particular parts that they
now happen to have. But if extreme mereological inessentialism is true, then the
foot and the station could have had parts entirely other than those that they have
in fact. The foot could have been made up of Mt. Monadnock and Mr. Robinson's
necktie and the station could have been made up of a certain horse and a certain
fish. So, of the indefinitely many possible worlds in which this table is made up
of the foot and the station, some of those will be such that in them the foot is made
up of the mountain and the necktie while the station is made up of the horse and
the fish, but others will be such that in them the station is made up of the horse
and the necktie while the foot is made up of the mountain and the fish.
It is difficult to imagine how even God could tell these worlds apart. Which
are the ones in which the necktie is made up of the horse and the station and which
are the ones in which the mountain is made up of the fish and the foot? One would
have to say, of the mountain and the necktie and the horse and the fish, that they
could have been made up of other things, too. Hence, of those worlds in which
the foot is made up of the mountain and the fish, there will be those in which the
fish is made up of the necktie and the station. . . .
But we need not formulate such extreme examples. Consider just two tables,
x and y, and suppose, what from one point of view would seem to be reasonable,
that these tables are such that they could survive replacement of any of their
smaller parts. We consider, then, the consequences of exchanging certain of their
smaller parts; then there will be a world possible in respect to this one in which
x has one of the parts that y has in this world and y has one of the parts that x
has in this world; then there will be a world possible in respect to that world, and
therefore also in respect to this one, in which x and y will have exchanged still
other smaller parts. We can imagine the process continued in such a way that it
will remind us of the ancient problems of The Ship of Theseus and The Carriage.
69
There will be a possible world that is like this one except for the fact that in that
one x has the parts that y has in this one and y has the parts that x has in this one.
We have only to reflect a moment to see that there will be indefinitely many such
possible worlds. Thus of those possible worlds W, which are such that the thing
u which is one of the legs of x in this world is the corresponding leg of y in W
and the thing v which is one of the legs of y in this world is the corresponding
leg of x in W, there will be those worlds W which are such that the things that
are parts of u in this world will be parts of v in W and there will those worlds
W" which are such that the things that are parts of v in this world will be parts
of u in W", and so on, ad indefinitum.
These reflections, on the consequences of extreme mereological inessentialism, may suggest to us that some version of mereological essentialism must be
trueeven if it is not the extreme principle we have set forth. But instead of trying
to formulate plausible alternatives to the extreme principle (a task which I have
found to be extraordinarily difficult), let us return to our philosophical puzzle and
see whether the extreme principle might not be defended.
IV
Let us begin by introducing some mereological definitions and axioms, taking as
undefined "x is part of y" where "part" is understood in the sense sometimes expressed by "proper part." Now it is possible that the term "part" is taken in one
way in our formulation of the principle of mereological essentialism and in another way in our formulation of the objections to it. In the principles that follow,
we will use the term "S-part" instead of "part." Use of "S-part" will indicate that
we are speaking strictly and philosophically. Then we may formulate, without
ambiguity, certain questions about the relation of "part" in its ordinary, or loose
and popular, sense, and "S-part" or "part" in its strict and philosophical sense.
Of the three axioms and the three definitions that follow, the first two in each
group were set forth, though in a somewhat different terminology, by Whitehead
in The Organisation of Thought.4
(Al) If x is an S-part of y and y is an S-part of z, then x is an S-part of z.
(A2) If x is an S-part of y, then y is not an S-part of x.
(A3) If x is an S-part of y, then y is such that in every possible world in which
y exists x is an S-part of y
We are suggesting, then, that the principle of mereological essentialism be
taken as a basic principle of the theory of part and whole. We add these definitions:
(Dl) x is discrete from y = Df (i) x is other than y and (ii) there is no z such
that z is an S-part of x and z is an S-part of y.
70
(D2) w is strictly made up of x and y = Df (i) x is an S-part of w, (ii) y is

an S-part of w, (iii) x is discrete from y, and (iv) no S-part of w is discrete
both from x and y.
(D3) x is strictly joined with y = Df There is a w such that w is strictly made
up of x and y.
Making use of the above terminology and principles, as well as the concepts
of spatial and temporal location, let us now consider the make-up of certain ordinary and familiar things.
V
Consider the history of a very simple table. On Monday it came into being when
Mon AB a certain thing A was joined with a certain other thing B. On
Tue BC Tuesday A was detached from B and C was joined to B, these
Wed CD things occurring in such a way that B remained throughout as a part of a table. And on Wednesday B was detached from C and D was
joined with C, these things occurring in such a way that C remained throughout
as a part of a table. Let us suppose that no other separating or joining occurred.
I suggest that in this situation there are the following three wholes among
others: AB, that is, the thing made up of A and B; BC, the thing made up of B
and C; and CD, the thing made up of C and D. I will say that AB "constituted"
our table on Monday, that BC "constituted" our table on Tuesday, and CD "constituted" our table of Wednesday. Although AB, BC, and CD are three different
things, they all constitute the same table. We thus have an illustration of what
Hume called "a succession of objects."5
One might also say, of each of the three wholes, AB, BC, and CD, that it
"stands in for" or "does duty for" our table on one of the three successive days.
Thus if we consider the spatial location of the three wholes, we see that on Monday AB occupied the same place that our table did, on Tuesday BC occupied the
same place that our table did, and on Wednesday CD occupied the same place
that our table did. And so we might define "constitutes" in the following way:
(D4) x constitutes y at t = Df There is a certain place such that x occupies
that place at t and y occupies that place at t.
The final clause could also be read as "y occupies exactly that same place at t."
We add this obvious definition:
(D5) x constitutes at t the same physical object that y constitutes at t' = Df
There is a z such that x constitutes z at t and y constitutes z at t ' .
Every physical object will, of course, constitute itself. But, according to our present suggestion, some things may constitute, and be constituted by, things other
71
than themselves. (Thus AB constituted our table on Monday; but AB, unlike our
table, also ceased to be on Monday.)
What if our table should undergo fission with the result that on Thursday there
Mon
AB
were two different tables, CE and EF? We cannot say that
Tue
BC
CE and EF both constitute the same table as does CD, BC,
Wed
CD
and AB. For our definitions imply that, if two different
Thu CE EF things constitute the same table at the same time, then those
two things are in the same place at that time; and CE and EF are not in the same
place on Thursday.
It is possible, however, that one or the other, CE or EF, constitutes the same
table as does CD. Which one, then? To answer this question, we would have to
turn to the philosophy of tables, or to the philosophy of furniture, and attempt
to set forth criteria which a pair of things must satisfy if one of them at one time
is to constitute the same table as does the other of them at another time. But this
project is not relevant to our present discussion.
Similar remarks will apply to fusionto what happens when two tables are
joined to make a single table. The fused table will not constitute the same table
as both of the original tablesbut it may, or may not, constitute the same table
as one of them.
And so we have described one possible way of looking upon what happens
when, as we would ordinarily put it, a thing such as a table undergoes a change
of parts. I propose that we consider our philosophical problem from this perspective. Before doing so, however, we should consider two objections to this way
of looking at the matter.
(1) "You are committed to saying that AB, BC, CD and our table are four
different things. It may well be, however, that each of the three things AB,
BC,CD, satisfies the conditions of any acceptable definition of the term 'table'.
Hence you are committed to saying that, in the situation described, there are four
tables. But this is absurd; for actually you have described only one table."
We will find the answer, I think, if we distinguish the strict and philosophical
sense of such expressions as "There are four tables" from their ordinary, or loose
and popular, sense. To say that there are four tables, in the strict and philosophical sense, is to say that there are four different things, each of them a table. But
from the fact that there are four tables, in this strict and philosophical sense, it
will not follow that there are four tables in the ordinary, or loose and popular
sense. For there to be four tables in the ordinary, or loose and popular, sense,
it must be the case that there are four things, not only such that each constitutes
a table, but also such that no two of them constitute the same table.
We may, therefore, explicate the ordinary, or loose and popular, sense of
"There are n so-and-so's at t" in the following way:
(D6) There are, in the loose and popular sense, n so-and-so's at t = DF
There are n things each of which constitutes a so-and-so at t and no two
of which constitute the same so-and-so at t.
72
The term "so-and-so" in this schematic definition may be replaced by any more
specific count-term, e.g., "table" or "ship."
And so the answer to the above objection is this: In saying that there are four
tables in the situation described one is speaking in the strict and philosophical
sense and not in the loose and popular sense; and in saying that there is just one
table one is speaking in the loose and popular sense and not in the strict and philosophical sense. The two assertions, therefore, are not incompatible.6
(2) The second objection to our way of looking at the simple table-or tables above may be put as follows.
"You say that the thing constituting our table at a given time may be something
other than the table itself. Yet you say that it occupies exactly the same place that
the table does. Therefore what you say is incompatible with the principle according to which it is impossible for two things to occupy exactly the same place at
the same time."
The expression "It is impossible for two things to be in the same place at the
same time" may be taken either in a strict and philosophical sense or in a loose
and popular sense.
If we take it in a strict and philosophical sense it tells us that it is impossible
for there to be an x and a y such that x is diverse from y and x occupies at a certain
time exactly the same place that y occupies at that time. If we take the principle
in this sense, then we must say that it is false. For a shadow and a part of the surface of a physical object may occupy exactly the same place at the same time; so,
too, for a hole in a shadow and a part of the surface of an object; so, too, perhaps,
for a person and a part of his body; and so, too, for any two things one of which
constitutes the other. Thus the AB of our example is other than the persisting table
x; yet on Monday AB occupies the place that x does; but in so doing AB does
not get into the way of x, for on Monday it is AB that does duty, so to speak,
for x. Hence it is no objection to say that our way of viewing the table is incompatible with the strict and philosophical sense of the principle, "It is impossible for
two things to be in the same place at the same time."
And what we have said is not incompatible with the loose and popular sense
of the principle. Taken in that sense, the principle tells us that it is impossible for
there to be two physical objects, in the loose and popular sense of "There are two
physical objects," occupying the same place at the same time. What we have said
does not imply that there are two physical objects, in this loose and popular sense,
occupying the same place at the same time. Looking back to D6, we see that there
cannot be two physical objects in this sense unless there are two physical objects
neither of which constitutes the other.
VI
We are now in a position to reply to the two objections to our version of mereological essentialism.
The first objection was this: (A) "(i) My automobile had parts last week that
73
it does not have this week and it will have parts next week that it has never had
before. But (ii) the principle of mereological essentialism implies that, if anything
is ever a part of my automobile, then that thing is a part of my automobile as long
as the automobile exists. And therefore (iii) the principle of mereological essentialism is false."
In reply to this objection one may observe that the term "part" is used in one
way in the first premise and in another way in the second and hence that the conclusion rests upon an equivocation. But if the reply is to be taken seriously, one
must state what the two uses of the term "part" are and how they are related to
each other.
In formulating the principle of mereological essentialism, we used the expression "S-part," suggesting that this might be read as "part in the strict and philosophical sense." (Perhaps the reader would prefer to read it as "part in the philosopher's sense.") We proposed three axioms in the attempt to explicate "S-part."
This is the sense in which "part" should be taken in premise (ii) of the above objection.
What of premise (i)? Here, I suggest, "part" must be taken in the loose and
popular sense. (Perhaps the reader would prefer to say: "Here 'part' must be taken
in its ordinary sense.") How, then, are we to relate this loose and popular sense
of "part" to the strict and philosophical concept of S-part?
To say, for example, that a certain tire is now a part of my automobile is to
say that what now constitutes that tire is a part, in the strict and philosophical
sense, of what now constitutes my automobile. And to say of a certain other tire
that it was a part of my automobile yesterday is to say that something that constituted that tire yesterday was a part of something that constituted my automobile
yesterday. I propose, then, this definition of the ordinary sense of "part" in terms
of the vocabulary we have here introduced:
(D7) x has y as a part at t = Df Something that constitutes y at t is an S-part
of something that constitutes x at t.
Taking "part" in this ordinary, or loose and popular sense, we may now say
of a physical thing, such as my automobile, that it may have one part at one time
and another part at another time. And saying this will be quite consistent with saying, as our principle of mereological essentialism requires us to say, that in a strict
and philosophical sense if a thing y is ever a part of a thing x then that thing y
is a part of x at any time that x exists.
The second objection was this:
"(B) (i) I could have bought different tires for my automobile, (ii) If I had
bought different tires for my automobile, then it would have had different parts
from those it now has. Therefore (iii) my automobile could have had different
parts from those it has now. Hence (iv) my automobile is such that, in some possible worlds, it has parts it does not have in this one. But (v) the principle of
mereological essentialism implies that in every world in which my automobile
74
it has exactly the same parts it has in this one. And therefore (vi) the principle
of mereological essentialism is false."
Here, too, we may observe that the term "part" is used equivocallyin the
loose and popular sense in premises (ii) and (iii) and in the strict and philosophical
sense in premise (v). But now we must show how the "could have" of premises
(ii) and (iii) is to be explicated in the strict and philosophical vocabulary. And
when we have done that, we may consider the status of premise (iv)the premise
according to which my automobile is such that in some possible worlds it has parts
it does not have in this one.
The statement, "My automobile could now have a certain thing as one of its
parts," even when restricted to its ordinary or loose and popular sense, has a certain ambiguity. On the one hand, it could be taken in a somewhat narrow sense
to mean the same as (A) "My automobile could have O as one of its parts and remain an automobile while having O as a part." One the other hand, it could be
taken more broadly to mean the same as (B) "My automobile could become a thing
that has O as a part," where there is no implication that the thing which is my automobile remains an automobile after it has taken on O as a part. Let us define
"x could have y as a part oft" in this second, broader sense. For given this broader
sense of "could" one can then readily express in terms of it what is intended by
the narrow sense (in our example, "x is an automobile and x could be at t an automobile having O as a part").
If something w is strictly made up of two things x and y, then x is strictly joined
with y (see D2 and D3). Our principles imply that, in such a case, w is necessarily
such that it has x as a part, in the strict and philosophical sense of the term "part."
But they do not imply that x is necessarily such that it is a part of w. And they
do not imply that x is necessarily such that it is joined with y. Returning to our
very simple table which, we supposed, was strictly made up of a stump and a
board, we may recall that, although the table is necessarily such that it has the
stump as a part, in the strict and philosophical sense of the term "part," the stump
is not necessarily such that it is a part of the table and it is not necessarily such
that it is joined to the board.
To say, then, in the loose and popular sense, that my automobile could now
be a thing having a certain tire will be to say that something that now constitutes
a part of my automobile could be joined with something that now constitutes the
tire.
Let us say, then:
(D8) x could have y as a part at t = Df There is a w and a v such that (i) w
is an S-part of something that constitutes x at t, (ii) there is a time at
which v constitutes y, and (iii) there is a possible world in which w is
strictly joined with v.
If we say, then, in this loose and popular sense, that my automobile could have
a certain tire as one of its parts, we are not saying that there is a possible world
in which that automobile does have that tire as one of its parts. We are saying,
75
rather, that something that constitutes a part of my automobile and something that
constitutes the tire are such that there is a possible world in which they are joined
together.
And so now we see that the fourth proposition in our objection does not follow
from the second and third. From that fact that my automobile could, in this loose
and popular sense, have a certain tire as a part, it does not follow that my automobile is such that in some possible world it has that tire as a part.
If, for any reason, we should persuade ourselves that this table could have been
made up of my left foot and the Grand Central Station, we need not be led to the
infinity of indiscernible possible worlds discussed earlier. We need not suppose
that, in some of the worlds in which this table is made up of the foot and the station, some are such that the foot is made up of the mountain and the horse and
others are such that it is made up of the necktie and the fish. For we may say what
we like about the possible makeup of the table, the foot, and the station, without
committing ourselves to the thesis that any of these things exist in any possible
world other than this one.
The theory of possibility does not require us to say, of any of these common
sense objects-the automobile, the table, the station, the mountain, the horse, the
foot, the necktie, and the fish-that they exist in any other possible worlds. But
it does require us to say, of the strict and philosophical wholes that constitute
these commonsense objects, that they exist in other possible worlds.
This last point, however, must be put more precisely.
VII
Let us consider two ordinary tables, x and y, that evolved in the way depicted
x
y
on the accompanying diagram. We are supposing that on
Mon AB CD Monday there were two things, each of them a table, one
Tue BC DE made up of A and B and constituting x, and the other made
Wed CD EB up of C and D and constituting y; that these two things
Thu DF AB "evolved" into BC and DE, respectively, on Tuesday, then
on Wednesday into CD and EB, and finally on Thursday into DF and AB. We
will suppose further that no additional joinings or separatings took place.
Our present question is: can we put precisely the difference between the two
kinds of things that are here involvedthe difference between such things as x
and y on the one hand and such things as AB, BC, CD on the other?
It is tempting to say that the ordinary or vulgar things, x and y, differ from
the strict and philosophical things AB, BC, and CD in that, whereas the ordinary
things are constituted by different things at different times, the philosophical
things are never constituted by different things at different times. We cannot say
this, however, although we will say something very much like it.
The relation of constituting, as we have defined it in D4, is symmetrical.
Hence, not only does AB constitute x on Monday, but x also constitutes AB on
76
Monday. But AB constitutes y on Thursday and therefore y constitutes AB on

Thursday. Therefore the philosophical object like the vulgar object is constituted
by different things at different times.
Let us say that our diagram depicts two object serieswhere the term "object
series" is an alternative for Hume's "succession of related objects." An object series will be a set of objects related to each other as the constituents of our ordinary
table x are related to each other and as the constituents of the ordinary table y
are related to each other. The mark of an object series will be that it is a set of
things related by succession: AB was succeeded by BC, BC was succeeded by
CD, and so on. But instead of saying that the individual thing AB was succeeded
by the individual thing CD, let us think of succession as relating sets of things.
We will say that the set consisting of AB and Monday was succeeded by the set
consisting of BC and Tuesday; and so on. (Thus, although we might be able to
say that the individual thing CD was succeeded by DE on Tuesday and by DF
on Thursday, we cannot say anything comparable of the set consisting of CD and
Monday.) We will introduce, then, the concept of an object-pair.
(D9) C is an object-pair = Df C is a class containing just a thing and a time
such that the thing constitutes an object during the time.
In this definition and in those that follow, the term "object" may be replaced
throughout by any count-termfor example, "table" or "ship."
Let us introduce the notation "[x,t]" as an abbreviation for the locution "the
object-pair having as members the individual thing x and the time t."
To define succession, we first define direct succession, construing the latter
concept in such a way that: [BC,Tue] directly succeeds [AB,Mon]; [CD,Wed]
directly succeeds [BC,Tue]; and [DF,Thu] directly succeeds [CD,Wed]. We will
also be able to say that: [DE,Tue] directly succeeds [CD,Mon]; [EB,Wed]
directly succeeds [DE,Tue]; and [AB,Thu] directly succeeds [EB,Wed]. Each
object-pair may also be said to be its own direct successor. Direct succession is
not otherwise exemplified in the things depicted in our diagram. I propose this
definition:
(D10) [x,t] is a direct object successor of [y,t'] = Df (i) t does not begin before t'; (ii) x constitutes at t the same object that y constitutes at t'; and
(iii) either x is identical with y, or there is a z such that z is an S-part
of x, z is an S-part of y, and at any time between t and t' inclusive there
is some w that then constitutes the same object that y constitutes at t',
and z is an S-part of w.
Our definition enables us to say, then, that [BC,Tue] is a direct object successor
of [AB,Mon]. Or, replacing "object" in our definition by "table" throughout, we
may say that [BC,Tue] is a direct table successor of [AB,Mon]. The definition
assures us that B persisted throughout the period from Monday to Tuesday and
that, at all times within that period, B was an S-part of a tablean S-part of something that constitutes the same table that AB constituted on Monday.
77
We should note that, given our definition, we may say of any object-pair that
it directly succeeds itself. And this will be true not only of direct succession, but
of succession more generally.
Of the "table-pairs" depicted in the x column of our diagram, the Wednesday
and Thursday pairs were not direct successors of the Monday pair, but they were
successors of the Monday pair, just as the Thursday pair was a successor of the
Tuesday pair. Similarly for the table-pairs depicted in the y column. Succession
is related to direct succession in the following way: u is a successor of v, if and
only if, it is true either that u is a direct successor of v, or u is a direct successor
of a direct successor of a direct successor of v, or u is a direct successor of a direct
successor of v, or . . . and so on. Hence we may define "successor" in the way
that was suggested by Frege. Let us say:
(Dll) [x,t] is an object successor of [y,f] = Df t does not begin before t';
and (ii) [x,t] belong to every class C which contains [y,f] and everything that is a direct object successor of any member of C
We may now say, of each of the "table pairs" depicted in our two columns, that
it is a "table successor" of itself and of each of the "table pairs" depicted above
it in the column.
Our two columns may be said to depict two "table series"if we think of x
and y as coming into being on Monday and as ceasing to be on Thursday. The
more general concept of "object series" may be defined this way:
(D12) C is an object series = Df C is a class having as its members an objectpair x, all the object successors of x, everything of which x is an object
successor, and nothing which is unrelated to x by object succession.
The final clause is, of course, short for: "nothing which is such that neither it is
an object successor of x nor x is an object successor of it." Hence, any two members of an object series will be such that one of them succeeds the other. We add
this definition:
(D13) C is an object series corresponding to x = Df C is an object series, and
every member of C contains a thing and a time such that that thing constitutes x at that time
If x should pass away and then come into being again at a later time (assuming
for the moment that this is possible) then there will be more than one object series
corresponding to x.
We are now in a position to state the difference between ordinary things such
as x and y, on the one hand, and the stricter things such as AB, BC, CD, on the
x
y
other. We have noted that it is not enough to say that,
Mon AB CD whereas the ordinary things are constituted at different
Tue BC DE times by different things, the stricter things are always
Wed CD EB constituted (when they exist) by the same thing. For the orThu DF AB dinary things are always constituted (when they exist) by
78
themselves, and the stricter things may be constituted at different times by different thingsas the stricter thing AB is constituted by x on Monday and by y on
Thursday, and the stricter thing CD is constituted by y on Monday and by x on
Wednesday. But we may now characterize the difference between the two types
of thing by reference to their corresponding object-series.
The object-series corresponding to the stricter things will be more constant
than those corresponding to the ordinary things. Let us call the stricter things primary objects and define the concept as follows, by reference to the constancy of
object-series:
(D14) z is a primary object = Df No object-series corresponding to z has two
members which are such that nothing belongs to both and z belongs to
neither
Thus neither the x nor the y of our diagram is a primary object. The object
series corresponding to x has at least two memberse.g., [AB,Mon] and
[BC,Tue]which are such that nothing belongs to both and x belongs to neither.
Similarly for y.
But AB, BC, CD, DF, and EB will be primary objects. It is true that AB is
constituted by x on Monday and by y on Thursday. But [x,Mon] and [y,Thu] are
not members of the same object series; for neither one is a successor of the other.
Analogously for CD and [y,Mon] and [x,Wed].
The S-parts of AB, BC, CD, DF, DE, and EB will also be primary objects.
For any S-part of a primary object is itself a primary object.
Consider now the following objection to what has been said. "You say that,
in the strict and philosophical sense, there were two different tables in one and
the same place on Mondayone of them, AB, which ceased to be when A was
disjoined from B, and the other of them, x, which was constituted by BC on Tuesday and by CD on Wednesday and which, therefore, persisted for at least three
days. Now suppose that the world had been destroyed late on Monday. Would
there still have been two tables-AB and x? Or would there have been just one?
And if the latter, which one?"
In describing the situation, we supposed that there occurred no joinings or disjoinings other than the ones that were mentioned. Hence the situation involved
three primary tables and one nonprimary table. Had the world been destroyed late
on Monday, then the situation would have involved just one primary table and
no nonprimary one. In short, there would have been just table AB and no table
x. "Does your answer imply, then, that x did not come into being until Tuesday?"
No, for a nonprimary object comes into being with the earliest members of its
object pairs. And if an object pair is such that it is going to have a direct object
successor which is other than it is, then the thing which belongs to it constitutes
an object which is other than w. "But had there been just AB and no x on Monday,
then our table would not have been such that it could have had parts other than
those that it does have. For, on your account, only nonprimary objects are such
that they could have parts which are other than they do have." The latter assertion
79
is mistaken, Given D8, our definition of "x could have y as a part at t," we may
say, in the loose and popular sense of "part," that primary objects are such that
they could have parts other than those they have in fact, even though, in the strict
and philosophical sense of "part," in the sense we have expressed by the term "Spart," they have exactly the same parts in every possible world in which they
exist.
According to the principle of mereological essentialism if a thing loses any of
its parts, then it ceases to be. In describing the history of table x, we said that
on Tuesday A was detached from B. This means, therefore, that AB ceased to
be on Tuesday. But now we find that what constitutes y on Thursday is an object
made up of A and B. Is this the same AB as the one that constituted x on Monday
or is it a different one? I have assumed that it is the same ABand, more generally, that if a primary object u is made up of the same things as is a primary object
v, then u is identical with v.
What I have just said, however, is contrary to the opinion of Thomas Reid,
who argues as follows: "I see evidently that identity supposes an uninterrupted
continuance of existence. That which hath ceased to exist, cannot be the same
with that which afterwards begins to exist; for this would be to suppose a being
to exist after it ceased to exist, and to have had existence before it was produced,
which are manifest contradictions."7 But it seems clear to me that the propositions
in question are not manifest contradictions. It would be contradictory to suppose
a being to exist after it had ceased to exist for the last time, and to have had existence before it was produced for the first time. But these things are not what we
are supposing when we say that a thing can come into being after it has ceased
to be.
We may now put more precisely the point that was formulated above as follows: "The theory of possibility does not require us to say, of any of these commonsense objectsthe automobile, the table, the station, the mountain, the horse,
the neckties, and the fishthat they exist in any other possible worlds. But it does
require us to say, of the strict and philosophical wholes that constitute these common sense objects, that they exist in other possible worlds."
The theory of possibility does not require us to say of any nonprimary object
that it exists in any possible world other than this one. But it does require us to
say \hatprimary objects exist in possible worlds other than this one. What we can
truly say about the unrealized possibilities of nonprimary things may be reformulated more precisely in terms of the unrealized possibilities of primary things. We
do not need to suppose, therefore, that there are possible worlds which are indiscernable except for the fact that some nonprimary things are constituted by one
set of primary things in one of them and by another set in another. And what we
say is entirely compatible with the principle of mereological essentialism: if x has
y as one of its parts, in the strict and philosophical sense of the term "part," then
in every possible world in which x exists, x has y as one of its parts.
80
VIII
Finally, let us note briefly how these suggestions relate to certain other philosophical questions and puzzles.
(1) Consider first what has been called "the Paradox of Increase." "It is impossible for anything to increase by the addition of parts, since when further parts
are adjoined to a thing, neither that to which the parts are adjoined, nor the adjoined parts themselves, increase in the sense that they have more parts than they
had before. . . . What then can be made of the way in which both ordinary usage and logic appear to countenance increase?"8
We think we can make things bigger just by adding parts to things. But what
are the things that we then make bigger? Suppose we have a certain thing A and
then attach to it a certain other thing B. We then have a bigger object than we
had before (assuming that neither A nor B shrunk or contracted during the process). But what object became bigger? It was not either A or B, for both of these
remained the same size they were before. And it was not AB for AB did not exist
until A was joined with B. That is to say, AB did not have two different sizes,
a smaller one at one time and a larger one at another.
If, in the situation I have described, there is something that became bigger,
then there is something which was constituted by A or by B at one time and by
AB at a later time and the thing that constituted it at the later time is bigger than
the thing that constituted it at the earlier time.
(2) Consider secondly a puzzle about identity.
"Suppose that on Monday we cast a certain bar of metal into a statue. Then
on Tuesday we melt the statue down and recast the metal into a vase. And on Wednesday we melt the vase and are left with just the piece of metal. Surely the statue
was the piece of metal on Monday and the vase was the piece of metal on Tuesday.
But the vase was not the statue and neither one of these was the piece of metal
on Wednesday. Therefore we must say either that one and the same thing can be
identical with one thing at one time and with another thing at another time or else
that two things can be identical with the same thing. But both of these conclusions
are absurd."9
Both of the conclusions are, of course, absurd. But if we describe the situation
accurately, we will not be led to either. Thus we may say that what constituted
the statue on Monday was identical with what constituted the piece of metal on
Monday, that what constituted the vase on Tuesday was identical with what constituted the piece of metal on Tuesday but not with what constituted the statue on
Monday, and that what constituted the piece of metal on Wednesday constituted
neither the statue nor the vase on Wednesday. From the fact that the piece of metal
and the statue constituted each other on Monday, we may not infer that they were
identical with each other on Monday. For x is not identical with y unless x and
y constitute each other during the entire time that either x or y exists. The statue,
therefore, was not identical with the vase and neither of these was identical with
the piece of metal.
81
"But if the statue was other than the piece of metal, then two different physical
objects - the statue and the piece of metal - both occupied the same place on Mo
day. And that is impossible."
From the fact that the statue is other than the piece of metal and that they both
occupied the same place on Monday, it does not follow that two different physical
objects occupied the same place on Monday. For to say that two different physical
objects both occupy the same place on Monday would be to say that there are two
things, neither of which constitute the same object on Monday, and both of which
are in the same place. See D6.
One could also deal with the problem, of course, by denying that there were
two things, a vase and a statue, in addition to the piece of metal. One could say
that there was just the piece of metal which had the property of being statuesque
on Monday and that of being vase-shaped on Thursday. But if we allow tables
and automobiles to count as things that come into being and pass away, why not
also vases and statues? The view that has been proposed here does allow us to
say that there are the three things in the situation described.
Indeed, we could revive the traditional term mode and say that the statue and
the vase were at different times modes of the piece of metal. A thing x could be
called a mere mode of a thing y provided only (i) x is necessarily such that everything that constitutes it at any time also constitutes y at that time and (ii) y is not
necessarily such that what constitutes it at any time also constitutes x at that time.
Notes
1. See D. P. Henry, Medieval Logic and Metaphysics (London: Hutchinson University Library,
1962), p. 120.
2. New Essays concerning Human Understanding, Book II, Chap, xxvii, Sec. 11 (Open Court
ed., p. 247). Compare Hume, Treatise of Human Nature, Book I, Part 4, Sec. 6.
3. Philosophical Studies (London: Kegan Paul, Trench, Trubner, 1922), pp. 287-88. Compare
also J. M. E. McTaggart: "For if a whole is a combination it is built up of parts which could exist
without being combined in that way, while the combination could not exist without them." Some
Dogmas of Religion (London: Edward Arnold, 1906), p. 108.
4. The Organisation of Thought (London: Williams and Norgate, 1917), pp. 158-62. Whitehead
adds another axiom, to the effect that, if x is part of y, then there is a z such that z is part of x. Whitehead applies his theory of part and whole to events. I believe it is accurate to say that he conceives
events in such a way that they may be said to have their parts necessarily.
5. See A Treatise of Human Nature, Book I, Part4, Sec. 6(Selby-Biggeed., p. 255): "all objects,
to which we ascribe identity, without observing their invariableness and uninterruptedness, are such
as consist of a succession of related objects." In this same section, Hume affirms a version of the principle of mereological essentialism.
6. It may be noted that we have defined the loose and popular sense of the expression, "There
are n so-and-so's at t" and not the more general, "The number of so-and-so's that there ever will have
been in ." For the loose and popular sense of this latter expression is not sufficiently fixed to be explicated in any strict and philosophical sense. The following example may make this clear. In the infantry
of the United States Army during World War II each private carried materials for half a tentsomething like one piece of canvas, a pole, and ropes. Two privates could then assemble their
materials and create a tent which would be disassembled in the morning. On another night the two
privates might find different tent companions. Occasionally when the company was in camp the various tent parts were collected, stored away, and then reissued but with no attempt to assign particular
82
parts to their former holders. Supposing, to simplify the matter considerably, that all the tents that
there ever will have been were those that were created by the members of a certain infantry company,
how, making use of our ordinary criteria, would we go about answering the question "Just how many
tents have there been?" Would an accounting of the history of the joinings of the various tent parts
be sufficient to give us the answer?
7. Thomas Reid, Essays on the Intellectual Powers of Man, Essay III, Chap. 4. Compare Locke's
Essay, Book I, Chap. 27, Sec.l: "one thing cannot have two beginnings of existence."
8. D. P. Henry, Medieval Logic, p. 120.
9. A slightly different version of this puzzle is set forth by Hugh Chandler in "Essence and Accident." Analysis, 26 (1966): 185-88.
Boundaries
Introduction
Stephan Korner has noted that one way of drawing up a theory of categories will
divide all particulars "into (a) a class of independent particulars, i.e., particulars
that are ontologically fundamental, and (b) a class of dependent particulars, i.e.,
particulars that are not ontologically fundamental."1 The dependent particulars
might be said to be "parasitical upon" the fundamental particulars.
I shall here discuss the nature of spatial boundaries, viewing them as dependent particulars.
What Are Boundaries?
Frege observes: "One often calls the equator an imaginary (gedachte) line, but
it would be wrong to call it a line that has merely been thought up (erdachf). It
was not created by thought as the result of a psychological process, but is only
apprehended or grasped by thought. If its being apprehended were a matter of
its coming into being, then we could not say anything positive about the equator
for any time prior to this supposed coming into being."2 Suarez had said, of the
outer surfaces of a body, that they are genuine entities distinct from the body itself. And evidently he held that God could preserve the boundaries of a thing in
separation from the thing (and also that God could preserve the thing in separation
from its boundaries).3
Are boundaries parts of things? To avoid a mere verbal question, we will introduce the word "constituent" and say that things may have two types of
constituentparts and boundaries. And we will say that a part of a thing is a constituent which is not a boundary.
83
84
BOUNDARIES
Why assume, then, that there are boundaries? The concept is needed for the
description of physical continuity.
Contiguity and Continuity
What is it for two things to be continuous with each other?
Let us recall an ancient problem. "Consider two discrete physical bodies
thought to be continuous with each other; the east side of body A, say, is continuous with the west side of body B. How is this possible? Either (i) the eastmost
part of A is in the same place as is the westmost part of B or (ii) no part of A
occupies the same place as does any part of B. In the case of (i), we would have
two discrete things in the same place. But this is impossible. In the case of (ii),
since A and B occupy different places, there is a place between the place where
A is and the place where B is. But if there is a place between A and B, then A
and B are not continuous."
Shall we say that, if two things are continuous with each other, then nothing
can be put between them unless at least one of the two things is moved? This
would be true, but it is too broad to capture the concept of continuity. For it holds
of things that are merely contiguous with each other but which are not continuous
with each other (for example, two blocks pushed together). A similar objection
applies to the suggestion that, if two bodies are continuous with each other, then
there is no space between them.
The problem requires that we make reference to the boundaries of things.
Aristotle had said:
The continuous is a species of the contiguous. I call two things continuous
when the limits of each, with which they touch and by which they are kept
together, become one and the same, so that plainly the continuous is found
in the things out of which a unity naturally arises in virtue of their contact.4
If the continuous object is cut in half, then does the one boundary become two
boundaries, one thing thus becoming two things? This is suggested by the passage
from Aristotle. But how can one thingeven if it is only a boundarybecome
two things. And does this mean that when two things become continuous, then
two things that had been diverse become identical with each other, two things thus
becoming one thing?
Or should we say that when two things become continuous, then one of the
outer boundaries ceases to bein nihilum. This view has been attributed to Bolzano.5 If we took this view, then we would have to say, of the thing that is cut
in half, that one of the two severed halves keeps the boundary and that a new
boundary comes into being which is then the boundary of the other half. This
would seem to be a clear case of coming into being ex nihilo. And what is to determine which half gets the new boundary and which half keeps the old one?
Or could it be that one of the halves retains the old boundary and the other half
is open-ended, having a side without a boundarythough not a side that is bound-
BOUNDARIES
85
less? But what determines which side is to be the one without the boundary? If
it is possible for a thing to exist without a boundary, why assume that either half
has a boundary? And why assume that there is a boundary separating the two
halves of the continuous object?
Or could it be that, if two things are in contact, then their boundaries coincide
or overlap? Descartes, in speaking of the relation between a surrounding body
and the body that it surrounds, speaks of "the common surface which is a surface
that is not a part of one body rather than of the other."6 This would mean that
distinct boundaries can occupy precisely the same place at the same time. And
it would also mean that, strictly speaking, more than one straight line can be extended between two points. This is the view that Brentano suggests.
Let us try to develop the suggestion further.
Boundaries, Parts, and Constituents

We will make use of the concepts of constituent, ofde re necessity and of coincidence.1 We presuppose that there is no actual concrete infiniteand hence that
there cannot be lines that are infinitely long or bodies that extend infinitely in
space. The expression "x is discrete from y" is an abbreviation for "there is nothing that is a constituent both of x and y."
There are two ways of defining boundary. We could appeal to the fact that a
boundary is a dependent particulara thing which is necessarily such that it is
a constituent of something. Or we could appeal to the fact that a boundary is a
thing that is capable of coinciding with something that is discrete from it. One
of these should be a definition and the other an axiom.
Let us take the first course and say that a boundary is a dependent particular:
Dl x is a boundary in y =Df x is a constituent of y; and every constituent of
x is necessarily such that there is something of which it is a constituent
Why not say simply that a boundary is a thing which is necessarily such that it
is a constituent of something? In such a case, we would have to count as a boundary such a hybrid object as the sum or heap consisting of Venus and the top surface of a certain table.
It should be noted that we have defined "x is a boundary in y," and not "x is
a boundary o/y." The latter expression would normally be taken in such a way
that it applies only to the outer boundaries of y. Thus, the expression "x bounds
y" would normally be taken to abbreviate "x is a boundary of y." Any boundary
in y is a boundary of a proper part of y. 8
I have said, following Brentano, that the concept of a boundary is closely
related to that of total coincidence. I suggest that the relation is this:
Al For every x, x is a boundary, if and only if, x is possibly such that there
is something with which it wholly coincides.
86
BOUNDARIES
We will consider the logic of total coincidence below.

If we take as our primitive concept, that of being a (proper) constituent, then,
as I have noted, we may define the concept of a (proper) part as follows:
D2 x is a part of y = Df x is a constituent of y; and x is not a boundary in y
If we replace "constituent" for "part" in the usual axioms for the concept of
(proper) part, we can retain the axioms of transitivity and irreflexivity, as well
as that of mereological essentialism (for every x and y, if x is a constituent of y,
then y is necessarily such that x is a constituent of it). But although we can say
that every part has a (proper) part, we cannot say that every constituent has a
(proper) constituent. For a point may be a constituent without having a constituent.
The following axiom reflects the fact that boundaries are essentially dependent
entities:
A2 For every x, y and z, if x is a boundary in y, and if z is a part of y in which
x is not a boundary, then there is a part of y discrete from z in which x
is a boundary
This principle implies that every constituent of every boundary is a constituent
of something that is not a boundary. It is thus inconsistent with Suarez' suggestion, referred to above, according to which God could remove just the surface
of a three dimensional object. The principle does not preclude our saying that
there are hybrid objects such as the sum of Venus and a certain surface which
is not a surface of Venus; but it does imply that, if there are such objects, then
there are individual things which are discrete from such objects.
Could God preserve any of the boundaries of a thing apart from the thing? We
could say that, for any thing having boundaries, God could destroy the thing and
preserve the boundariesby destroying some part of the thing such that the part
did not contain any of those boundaries. But he couldn't preserve the boundaries
except by retaining some part of the original thing.
Coincidence
We have taken as primitive the concept of total coincidence. Let us now express
this concept by "x wholly coincides with y" ("xWy").
One axiom for this locution is Al above:
Al For every x, x is a boundary, if and only if, x is possibly such that there
is something with which it wholly coincides
In formulating additional axioms, we will use the abbreviations: "xCy" ("x is
a constituent of y"); "xDy" ("x is discrete from y") and "xBy" ("x is a boundary
in y").
The relation of total coincidence is symmetrical and irreflexive:
BOUNDARIES
87
Al.l xWy -> yWx

A1.2 xWy. -> D(xWx)
Hence "xWy" is not transitive. But if one thing wholly coincides with a second
thing and if the second thing wholly coincides with still a third thing, then the first
thing wholly coincides with the third thing. In other words:
[xWy & yWz & D(x = z)] -> xWz

We may also affirm that, if x wholly coincides with y, then every constituent of
x wholly coincides with a constituent of y:
A1.3 xWy - {(z) (zCx) -> [(Ev)(vCy & vWz)]]
And if two boundaries wholly coincide, then they are constituents, respectively,
of two things having no parts in common:
A1.4 xWy -+ [(Eu) (Ev) (uDv & xBu & yBv)]

Total coincidence may hold between surfaces, or between lines, or between
points, but it may not hold between two things of different dimensions or between
solids. In terms of the undefined concept of total coincidence, we may now define
the broader concept of coincidence:
D3 x coincides with y = Df either (a) x wholly coincides either with y or with
a constituent of y or (b) a constituent of x wholly coincides with a constituent of y
The first clause of this definition insures that coincidence, unlike total coincidence, may obtain between points and lines, between points and surfaces, and between lines and surfaces; the second clause insures that coincidencebut not total
coincidencemay obtain between three-dimensional things.
What, then, of dimensionality?
Dimensionality
We now define dimensionalityassuming that things may have either no spatial
dimensions, or one such dimension, or two, or three. Thus we take it to be a
necessary truth that there are exactly three spatial dimensions.9
We will not equate solids with what is 3-dimensional. That sum consisting of
Venus and the top surface of the table is 3-dimensional, but not a solid. We will
say, analogously, that there are no "broken surfaces"; hence that 2-dimensional
object which is the sum of the front and back surfaces of a certain cube will not
be a surface. And analogously there will be no "broken lines." Surfaces are like
solids and unlike lines in that they may have holes.
We now set forth the following definitions:
D4 x is O-dimensional (a point) = Dfx is a boundary and x has no constituents
88
BOUNDARIES
D5 x is 3-dimensional (a point) = Dfx has constituents and is not a boundary

We have thus defined O-dimensionality and 3-dimensionality. Let us now consider 2-dimensionality.10
Shall we say that a 2-dimensional object is a boundary which is possibly such
that the only things of which it is a constituent are 3-dimensional? No; for, if there
cannot be things that are infinitely extended, then any part of the surface of a thing
must also be a part of that 2-dimensional sum composed of it and some other part
of the surface of the thing.11 Thus the front surface of a cube must be a part of
that 2-dimensional sum consisting of its front surface and back surface. Our conception of coincidence, moreover, requires that solids have "inner surfaces";
hence the entire outer surface of a solid is necessarily such that it is a constituent
of 2-dimensional wholes consisting of it and an inner surface.
The concept of coincidence, however, presents us with one feature that is peculiar to 2-dimensional boundaries. A point is capable of coinciding with any
number of points at a time; and a line is capable of coinciding with any number
of lines at a time; but a surface can coincide only with one surface at a time. Let
us say, then:
D6 x is 2-dimensional = Df x is a boundary; and for all y and for all z, if
x wholly coincides with y and x wholly coincides with z, then y = z
(This definition, as well as the others here, should be taken to be in the present
tense. If the definitions are interpreted "tenselessly", then a temporal variable
should be taken to be implicit throughout.)
And now we may define 1-dimensionality:
D7 x is 1-dimensional = Df x is neither O-dimensional, 2-dimensional nor
3-dimensional; and x is necessarily such that it is a (proper) constituent
of something that is 2-dimensional
Continuity
Two three-dimensional things are said to be discrete when they have no parts in
common. It is convenient to introduce analogous concepts for 2-dimensional
things and for 1-dimensional things. And so let us distinguish among three
subspecies of discreteness: as a relation between 3-dimensional things; as a relation between 2-dimensional things; and as a relation between 1-dimensional
things.
D8 x and y are 3-dimensionally discrete = Df x and y are 3-dimensional;
and nothing 3-dimensional is a part of both
and nothing 2-dimensional is a constituent of both
and x is other than y
BOUNDARIES
89
We may now distinguish three types of contact: that between 3-dimensional

things ("touching"); that between 2-dimensional things; and that between 1dimensional things.
Dll x is in 3-dimensional contact with y (x touches y) = Df is 3dimensionally discrete from y; and x coincides with y
D12 x is in 2-dimensional contact with y = Df x is 2-dimensionally discrete
from y; and x coincides with y
D13 x is in 1-dimensional contact with y = Df x is 1-dimensionally discrete
from y; and x coincides with y
We may say, then, that a thing x is in contact with a thing y, provided only
that x is either in 1-dimensional, 2-dimensional or 3-dimensional contact with y.
And now we may say that x is continuous with y, provided only that x is in contact
with y. 12
Notes
1. Stephan Korner, Categorial Frameworks (Oxford: Basil Blackwell, 1970), p.4.
2. Gottlob Frege, Foundations of Arithmetic (Oxford: Basil Blackwell, 1950), p.35.
3. See Hugo Bergmann, Das Philosophische Werk Bernard Bolzano (Halle: Max Niemeyer,
1909), p. 207. Bergmann refers to Suarez's Metaphysicae Disputationes, XL, 5, Sees. 37 and 41.
4. Aristotle, Metaphysics, 1069a.
5. Compare Bernard Bolzano, Paradoxes of the Infinite (New Haven: Yale University Press,
1950). See paragraph 67 (p. 168). Bolzano speaks here of certain things but not others "being devoid
of limiting atoms."
6. Descartes, Principles of Philosophy, Part 2, Principle XV; in E. S. Haldane and R. T. Ross,
Philosophical Works of Descartes, vol. 1 (Cambridge: Cambridge University Press, 1931), p. 261.
7. The concept of coincidence is used by Franz Brentano, in Stephan Korner and Roderick M.
Chisholm (eds.), Philosophische Untersuchungen zu Raum, Zeit und Kontinuum (Hamburg: Felix
Meiner Verlag, 1976). Compare also Brentano's Psychology from an Empirical Standpoint (London:
Routledge & Kegan Paul, 1973), pp. 351-58; Psychologic vom empirischen Standpunkt, Band II
(Hamburg: Felix Meiner Verlag, 1971), pp. 259-62.
8. Brentano discusses inner and outer boundaries, Raum, Zeit und Kontinuum, p. 15.
9. One can, of course, construe time as a "fourth dimension." But it is not a fourth spatial dimension. The "four dimensional" things that relativity theory speaks about are events of a certain sort- not
bodies having four spatial dimensions.
10. Brentano discusses dimension in Philosophische Untersuchungen zu Raum, Zeit und Kontinuum, pp. 13-16. He appears to assume, mistakenly, that the only boundaries found in surfaces are
lines (but smaller surfaces will also be boundaries in any surface); and analogously for lines and
points.
11. This point was made by Michael Zimmerman.
12.1 am indebted to Stephan Korner, who introduced me to this topic when we were preparing
Brentano's Philosophische Untersuchungen zu Raum, Zeit und Kontinuum for publication. I am also
indebted to Robert Frederick, Richard Potter, James Van Cleve, and Michael Zimmerman.
Scattered Objects
1
The classic paper on scattered objects was written by Richard Cartwright.1 What
I present here may be thought of as a commentary on that paper. Like Cartwright,
I believe that there are scattered material objects. But my views differ from his
in several respects: (1) Where Cartwright makes use of such absolute spatial concepts as point and region, I make use of the relational concept of touching (or direct spatial contact). This alternative approach may throw a different light on
some of the metaphysical questions that the problem of scattered objects involves.
(2) I consider a distinction between two fundamentally different types of scattered
objectsa distinction that Cartwright does not discuss. (3) I express some doubts
about the relevance of "temporal parts" to the metaphysical problems that scattered objects involve.
II
I take the relation of proper part of (written henceforth as "part of) as undefined
and assume that it is transitive and assymmetric:
(Al) For every x,y, and z, if x is part of y and if y is part of z, then jc is part of z.
(A2) For every jc, if jc is part of y, then y is not part of*.
To abbreviate a further principle, I add two further definitions:
(Dl) jc is discrete from y = Df (a) x is other than y and (b) there is no z such
that z is part of jc and z is part of y.
(Without the first clause we would have to say of points and monads that they are
discrete from themselves.)
90
SCATTERED OBJECTS
91
(D2) jc is composed of y and z = Df (a) y is part of jc, (b) z is part of x, (c)

y is discrete from z, and (d) no part of x is discrete from both y and z.
(This definition was first proposed by Whitehead.2) Consider a table of the following sort: It has a top T; it has four legs, A, B, C, and D; and it has no part
that is discrete from each of the parts, A,B,C,D, and T. What is the table composed of? The question has many answers. We may say that it is composed of
A and TBCD, also that it is composed of B and TACD, that it is composed of C
and TABD, and that it is composed of D and TABC. (I use the locution "WX" as
short for "that object that is composed of Wand X" and "WXYZ" as short for "that
object that is composed of WX and YZ." ) If there are scattered objects, we may
also say that the table is composed of Tand ABCD, that it is composed of TA and
BCD, that it is composed of TABC and D, and that it is composed of such entities
as the lower half of B and the remainder of the table.
Ill
We may define the concept of a scattered object if we allow ourselves the additional concept of touching, or direct spatial contact. (If a book is on the table and
the table is on a rug, then, although the book may be said in some sense to be
in spatial contact with the rug, it is not in direct spatial contact with the rug.)
(D3) jc is a scattered object = Df there is a y and there is a z such that (a) jc
is composed of y and z and (b) no part of y is in direct spatial contact
with any part of z.
The present sense of "scattered," it should be noted, does not preclude the possibility of order. "Scattered," therefore, should not be taken to imply "haphazard,"
"chance," or "random." The essential mark of a scattered object is the absence of
a certain type of spatial contact.
IV
"Is there a material object composed of the Eiffel Tower and the Old North
Church?"3 If there is such a material object, then, we may agree, there are scattered material objects.
As Cartwright notes (p. 155), such entities as the following would seem to be
quite respectable ontologically: the United States, the solar system, a suite of furniture, a pile of coal, a watch that is spread out on the watch repairer's workbench,
printed words, the lowercase letters i and j, the constellation Cassiopeia.
But there are two extreme views about the existence of such objects.
One, which we could call conjunctivism, is the view that for any two individual
things there is a third thing that is composed of both.4 (Or we could qualify the
doctrine by saying: "For any two individual things that are discrete from each
other, there is a third thing composed of both.")
92
SCATTERED OBJECTS
The other extreme could be put this way: All genuine individuals are compact;
that is, they are nonscattered.
Sometimes the defender of the second view seems to presuppose the contradictory of that view. S/He may say: "Scattered objects are mere heaps or aggregates;
and mere heaps or aggregates are not genuine things." This way of talking seems
to imply that there are mere heaps or aggregatesand that such things are not
"genuine." But if there are mere heaps or aggregates and if scattered objects are
mere heaps or aggregates, then there are scattered objects.5
Consider again the example of Cassiopeia. One might argue as follows: Those
particular stars have just been associated by human beings; other creatures might
have associated some other group of starssay, some group of stars that has
never in fact been thought of as a constellationand they might have called that
constellation "Alcibiades." Now surely there is no constellation Alcibiades. But
by similar reasoning there really is no constellation Cassiopeia.
The objection says, correctly, that we focused on Cassiopeia and not on that
other group of stars that could have been called "Alcibiades." But the only relevant difference between the two heaps of stars is that we did focus on the former
and not on the latter. The heap that might have been called "Alcibiades," unlike
Cassiopeia, has no name in our astronomy, and therefore we do not call it a constellation. (If one insists that, although there is a heap, the constituent stars "do
not form a constellation," then one is using "constellation" to mean a "heap of stars
for which we have a name.").
V
Cartwright makes it abundantly clear that scattered objects play an essential role
in the problem of identity through time.
Consider a whole Wthat persists through timetaking on new parts and shedding old ones during the course of its existence: I have depicted "the Ship of
Theseus" in the following way:
W
Mon
Tue
Wed
Thur
X Y Z
X Y
X
[ABC]
[ABZ\
[AYZ\
[XYZ]
B C
ABC
Column W depicts some of the parts that constituted Theseus on the four days
Monday, Tuesday, Wednesday, and Thursday. The letters immediately to the left
of the column depict parts of the flotsam that were later to constitute the ship of
Theseus. And the letters to the right depict parts of the jetsam that was composed
of the discarded part of Theseus.
The diagram could also be taken to depict a human body. Let us consider it
that way and replace W by "Charlie." In doing this, we will be using the name
SCATTERED OBJECTS
93
"Charlie" to refer to the type of thing that Cartwright used it for-an individual
thing that may lose some of its parts and take on others.
If we look at our diagram, we can find a number of objects that are of a very
different nature from Charlie. One of them is that object having parts A, B, and
C, which occupied the place of Charlie on Monday. Let us call this object
"Harry," thus using a name that Cartwright introduces at a similar point.
Consider the relations between Charlie and Harry. We can say, as Cartwright
would, (1) that Harry and Charlie occupy exactly the same place on Monday, (2)
that Harry is not identical with Charlie, and (3) that things never "change their
identities" (roughly, you cannot be diverse from something today and identical
with it tomorrow). Charlie and Harry are intimately related, then, but not identical. What more can we say about the relations between them?
I assume that Cartwright would want to say of our Charlie and Harry what he
says of his Charlie and Harry:
We have treated Charlie as a continuant, an object that endures for a period
of time during which it undergoes change. It would seem only fair to treat
Harry in the same way. Like Charlie, Harry underwent a certain change.
He occupied a connected receptacle at t [Monday] and a disconnected one
att' [Thursday]. ("Scattered Objects," p. 166)
I believe that, in proposing to treat Harry "as a continuant" like Charlie, Cartwright supposes that, like Charlie, Harry is capable of losing partshe says that
his Harry does lose certain parts (see p. 169). But is it fair to our Harry to treat
him this way? If Harry is that object that has parts A, fl, and C and that occupies
the place that Charlie occupies on Monday, doesn't Harry exist with precisely the
same parts on the next three days? He becomes somewhat scattered on Tuesday,
more widely scattered on Wednesday, and still more widely scattered on Thursday when he becomes a mass of jetsam.
Charliethe human body occupying Harry's place on Mondaychanged parts
several times during the period depicted. But Harry did not change any parts at
all. Indeed, Harry's parts would seem to be essential to him. Unless some of his
parts cease to be in nihilum, it looks as though Harry will exist forever. What
kind of a thing is he, then?
We might define the concept of a substance this way:
(D4) x is an individual substance = Df If x has parts, then for every y, if y
is part of jc, x is necessarily such that y is part of it.
The definition is put as it is to allow for the possibility that a monad is a substance.
It also allows us to say, with Platonists, that abstract objects are substances (but
this defect, if it is a defect, is easily remedied).
Those individuals that may survive the loss of their parts may be called nonsubstantive individuals. (I hesitate to use "nonsubstantial individuals.") Thus the
Ship of Theseus and Charlie are nonsubstantive individuals. The metaphysical
problem that Cartwright discusses in terms of what he calls the fusion principle
94
SCATTERED OBJECTS
now becomes, What is the relation between nonsubstantive individuals and substances?
By our definition, Harry is a substance and Charlie is not. And so we are left
with the metaphysical question, What is the relation between Charlie and Harry?
VI
Cartwright introduces the philosophical concept of a temporal pan at this point.
He suggests that
although Charlie and Harry are distinct objects, as is revealed by their
divergent careers, a certain temporal part of Charlie is identical with a certain temporal part of Harry: Charlie's f-stage [Monday-stage], as we might
call it, is identical with Harry's t stage. ("Scattered Objects," p. 169)
What are temporal parts, and what is the reason for thinking that such things
as Charlie and Harry have temporal parts?
If we take "temporal part" in the technical sense in which, I believe, Cartwright
understands it, then it will not do to say merely that a temporal part of a thing
is a part that the thing has at a certain time and fails to have at other times, for
some parts of the latter sort may be shed and then taken on again ("Let's try it
once more with the old carburetor"). A temporal part, however, cannot be taken
on again once it has been shed; whatever has it is necessarily such that it has it
only once.6
There is a reasonably clear sense in which such entities as states, processes,
or careers may be said to have temporal partsparts that are unique to the times
at which they are had and that cannot be taken on and off. Your second year and
your twentieth year are different temporal parts of your life history. But our problem has to do with individual things and not with processes, states, or careers of
individual things.
The question is, Why assume that individual thingssuch things as people,
chairs, and matchboxes have temporal parts? The assumption does give us a
kind of answer to our metaphysical question about Harry and Charlie, but this
fact does not seem to me to be by itself a sufficient reason for thinking that the
assumption is true. Is there some other reason for thinking that it is true?
Evidently it is useful in theoretical physics to abstract from individual things
and to investigate those processes that are the histories of such things. Hence
physics may be said to investigate things that do have temporal parts. But it hardly
follows from this that individual things have temporal parts.
To be sure, there are "process"-philosophers who say that such things as human bodies and matchbooks are really processes. But, so far as I know, no one
has ever devoted any philosophical toil to showing how to reduce such things to
processes. In the absence of such a reduction, I would agree with Broad, whom
Cartwright quotes: "It is plainly contrary to common sense to say that the phases
in the history of a thing are parts of the thing."7 Harry is an individual who once
SCATTERED OBJECTS
95
had the shape and size of a man, but no process or career can have the shape and
size of a man.
Could we modify Cartwright's suggestion, then, and say something similar to
"a certain temporal part of Charlie's history is identical with a certain temporal
part of Harry's history"? This seems to me to leave us with our problem. How
are Charlie and Harry related if they are diverse and such that parts of their histories overlap?
I feel, therefore, that the appeal to temporal parts will not help.
Harry seems to make out better than he ought to. And, strangely, the one
whose status is now unclear is Charlie. I am not really satisfied with any of the
proposed solutions to the problem.8 And so I hope very much that Cartwright will
continue to work on it.
Notes
I am indebted to Judith Jarvis Thomson for a number of helpful suggestions.
1. Richard Cartwright, "Scattered Objects," in Keith Lehrer (ed.), Analysis and Metaphysics,
(Dordrecht: Reidel, 1975); reprinted in Richard Cartwright, Philosophical Essays (Cambridge,
Mass.: MIT Press, 1987). All references to Cartwright are to this paper.
2. A. N. Whitehead, The Organization of Thought (London: Williams and Norgate, 1917). p.
165. Whitehead used "separated from," whereas I have used "discrete from."
3. Cartwright, "Scattered Objects." p. 155.
4. S. Lesniewski proposed as an axiom of mereology: "For every non-empty class A of individuals there exists exactly one individual x which is a sum of all the members of A." And he said
that an individual x is the sum of the members of a class A provided only that every member of A
is a part of x and that no part of x is discrete from every member of A. See the exposition of Lesniewski's mereology in A. Tarski, Logic, Semantics, Metamathematics, (Oxford: Clarendon Press,
1956), pp. 24-29.
5. Compare Cartwright, "Scattered Objects," 158-59.
6. Compare Rudolf Carnap, Introduction to Logic and Its Applications (New York: Dover,
1958), pp. 197-210.
7. C. D. Broad, An Examination of McTaggart's Philosophy, vol. 1 (Cambridge: The University
Press, 1933), pp. 349-50. Quoted by Cartwright in "Scattered Objects," p. 171.
8. My own most recent attempt to deal with the problem may be found in Radu G. Bogdan (ed.),
Profiles: Roderick M. Chisholm (Dordrecht: Reidel, 1986), pp. 65-77. I suggest there that I may
resemble Harry more than I do Charlie.
Part IV
The Mental
10
The Nature of the Psychological

I shall here formulate a definition of a psychological attribute. What I will say
presupposes the common sense distinction between the psychological, or mental,
and the nonpsychological.
The following may be taken as being paradigmatic cases of psychological attributes: judging; being sad about something; being pleased about something; wondering about something; feeling depressed; seeming to oneself to have a headache; and being appeared to redly. And the following will be paradigmatic cases
of nonpsychological attributes: being extended; wearing a hat; being green;
weighing 7 pounds; being the successor of 9; being such that all men are mortal.
There is a simple formula that tells us what the psychological attributes we
have listed have in common and what distinguishes them from everything that is
nonpsychological. It is this:
Any property which is possibly such that it is exemplified by just one thing and
which includes every property it implies or involves is psychological
This formula provides us with an interpretation of one traditional thesisnamely,
that whatever is 'purely qualitative' is psychological. We will define the psychological by reference to that which is purely qualitative. But we will not define the
psychological as that which is purely qualitative, since certain psychological
attributes-for example, thinking about one's brother-are not purely qualitative.
I will make the following metaphysical presuppositions. (1) It is necessarily
the case that there are entities that are not individual things; examples are such
abstract objects as propositions, attributes and numbers. (2) The ontological the-
Philosophical Studies 43 (1983) 155-164. Copyright 1983 by D. Reidel Publishing Co., Do

drecht, Holland, and Boston, U.S.A.
99
100
THE NATURE OF THE PSYCHOLOGICAL
sis of materialism ("Every individual thing is a material thing"), even if it is true,

is not necessarily true. And (3) whatever is a proper part has a proper part.
These presuppositions seem to me to be clearly true, but it is not the point of
the present paper to defend them. It is enough to note that, without them, our analysis of the psychological would have to be considerably more complex.
Our first task is to explicate the four expressions we have used in the formula
above: 'property', 'implies', 'includes' and 'involves'. In doing this, I shall make
use of four philosophical concepts. These are: (1) the concept of an individual
thing; (2) the concept of exemplification at a time (as in 'jc exemplifies y at f or
'jc has y at f); (3) the concept of de re modality (as in lx is necessarily such that
it is F); and, finally, (4) the intentional concept of conceiving (as in lx conceives
the property being red').
Properties
Let us think of an attribute as being anything that is capable of being exemplified:
D1 P is an attribute = Df. p is possibly such that there is something that exemplifies it.
I assume that there are attributes and that some of them are exemplified and others
not. I also assume that every attribute is capable of being conceivedi.e., that
every attribute is possibly such that there is someone who conceives it.
I will next single out a certain subspecies of attribute I will call a 'property'.
This use of'property' is entirely arbitrary; it is introduced only to avoid circumlocution. The definition of property will be so formulated that each of the psychological attributes on our list above will count as properties. The definition is this:
D2 P is a property = Df. P is an attribute which is such that: (a) only individual things can have it; (b) anything that can have it can have it, or
fail to have it, at any time it exists; and (c) it can be such that some individuals have it and some do not.
Condition (a) tells us that if an attribute may be exemplified by a nonindividual
(say, by an attribute or by a number), then that attribute is not a property. Hence
being an even number and being exemplified are not properties. And no attribute
that is capable of being universal is a property.
Condition (b) tells us that if P is a property, then whatever can have it can have
it, or fail to have it, at any period of its existence-that is to say, at the time at
which the thing comes into being and at any time thereafter up to and including
the time it passes away. Hence coming into being and passing away will not themselves be properties. And what we might call 'past-tensed attributes' (for example,
being such that it did walk} and 'future tensed attributes' (for example, being such
that it will walk) will not be properties.
All the psychological attributes on our list above, as I have said, are properties, in the present sense of the term.
101
Implication and Inclusion

We have now explicated one of the four terms that appear in our general formula.
We turn to implication and inclusion.
D3 P implies Q = Df. P is necessarily such that if anything has it then something has Q.
The attribute of being a man implies the attribute of being rational. And the attribute of being a person who owns a boat implies the attribute of being a boat.
Inclusion in a subspecies of implication:
D4 P includes Q = Df. P is necessarily such that whatever has it has Q.
If P includes Q, then P also implies Q. But P may imply Q without including Q.
The attribute of being a person who owns a boat implies but does not include the
attribute of being a boat. And the attribute of being populated implies but does
not include the attribute of being a living thing.
Nonrelational Properties
There is one noteworthy feature that has been traditionally associated with psychological or mental properties. It is that of being internal or nonrelational.
We are now in a position to specify one sense in which certain psychological
properties may thus be said to be 'nonrelational'. For one mark of a relational
property is this: a relational property implies a property it does not include. (Note
I have said "implies a property it does not include".) The property of being a biped
will be relational according to this account, since it implies but does not include
the property of being a foot. But the property of being angry will be nonrelational;
it includes every property that it implies. So, too, for the psychological properties
with which we began: judging; being sad about something; being pleased about
something; wondering about something; feeling depressed; seeming to oneself to
have a headache; and being appeared to redly.
To be sure, judging implies but does not include the attribute of being judged
about or that of being an object of judgement; and wondering implies but does
not include the attribute of being wondered about or that of being an object of
wonder. But being judged about and being wondered about are not properties,
in our restricted sense of this term. They may be exemplified by such entities as
propositions, attributes and numbers; hence they are not such that only individual
things can have them. ("What, then, of being an individual that is judged abouf!
No abstract object can have that attribute." This is true; but that attribute, unlike
our psychological attributes, implies a property that it does not includenamely,
that of judging.)
Psychological properties, however, are not the only properties that are nonrelational by our definition. The properties on the following list are both nonpsychological and nonrelational:
102
being
being
being
being
either angry or two-legged;

intoxicating;
potentially dangerous;
an individual thing that is nongreen.
But these attributes, unlike our paradigmatic psychological attributes, are all such
that they involve properties they do not imply.
Involvement
What, then, is the requisite sense of involvement? Consider these four attributes:
(i) being either red or round, (ii) being nonred, (iii) being possibly red and (iv)
wanting something that is red. These attributes all involve the property red, yet
they do not include or imply it.
Using the intentional expression 'conceives', we can say exactly what the requisite sense of'involvement' is. Each of the attributes may be said in the following
sense to involve the property red: the attribute is such that one cannot conceive
it without conceiving the property red.
So let us define involvement this way:
D5 P involves Q = Df. P is necessarily such that whoever conceives it conceives Q.
We may note, incidentally, that these concepts of inclusion and involvement yield
an identity condition for attributes:
An attribute P is identical with an attribute Q, if and only if: P includes Q,
Q includes P, P involves Q, and Q involves P.
In other words: an attribute P is identical with an attribute Q, if and only if, they
are necessarily such that (i) whatever has the one has the other and (ii) whoever
conceives the one conceives the other. Hence attributes that imply and include
each other need not be identical. Being an equilateral triangle and being an equiangular triangle are distinct attributes, since neither involves the other. And being sad is not identical with being sad and such that seven and five are twelve,
since, although the latter attribute involves the former, the former does not involve the latter. Killing and being killed are distinct attributes, since, although
they involve each other, they do not include each other.
A Mark of the Psychological

We have now defined the terms that occur in our formula:
Any property which is possibly such that it is exemplified by just one thing and
which includes every property it implies or involves is psychological.
It is clear that the nonrelational nonpsychological properties we have listed above
do not satisfy this formula. Thus the disjunctive property of being either angry
103
or two-legged involves but does not include the property of being angry. One can't
conceive the disjunctive property without conceiving the property of being angry;
but one can have the disjunctive property without having the property of being
angry. Again, a thing can have the property of being intoxicating without having
the property of being intoxicated; but one can't conceive the property of being intoxicating without conceiving the property of being intoxicated. Thus we can say
of dispositional properties generally that they involve but do not include the property or properties toward which they dispose their bearers.
But our paradigmatic psychological properties are all such that they include
every property that they imply or involve. And, it would seem, only psychological properties satisfy this condition. Hence the formula gives us a sufficient condition of the psychological.
Some Test Cases

We now consider certain nonpsychological attributes which, at first consideration, may seem to satisfy our formula; that is to say, each of them may seem to
be a property which is such that it includes every attribute it implies or involves.
But a closer examination will show that these attributes do not satisfy our formula.
And in most such cases, we will find that the attribute in question does not satisfy
our definition of property.
Being a proper pan of an individual thing does not satisfy our formula, for
it is not possibly such that there is just one thing that has it. So, too, for having
a proper part, being extended, and being green. Anything having any one of these
properties has a proper part that also has that property. Therefore these properties
are not possibly such that they are exemplified by just one thing.
What of such semantical attributes as referring to something, designating oneself, and being heterologicall These attributes involve but do not include such
properties as using a language and interpreting a language.
What of moving] The attribute of moving may be defined as follows: being
one of several things that are changing their spatial relations to each other. Hence
moving does not satisfy our formula, for it cannot be exemplified by just one individual.
The attribute undergoing change is not a property. It is necessarily the case
that if anything undergoes change then everything undergoes change. Suppose,
for example, that a certain individual thing changes from being red to being
green. Then each thing changes in the following way: certain relations it had formerly born just to something red become relations that it now bears to something
green. Hence undergoing change is capable of being universal. But no
property-in our present sense of the term-is capable of being universal.
"What of changing internally! If eternal objects change (say, from being such
that there are dinosaurs to being such that there are no dinosaurs), this change
is only external and is a function of internal changes in individual things. Eternal
objects would not change at all if there were no individual things. Therefore the
104
attribute of changing internally would seem to be a property that includes every

property it implies or involves. But it is not a psychological attribute."
What would be an example of an 'internal change'? I can think only of four
types of such change, (i) There are psychological changes-say, the change from
being happy to being sad. But these are not counter to our thesis, (ii) Possibly
we can say that coming into being and passing away are internal changes. But,
even if we can say this, we cannot say that they are properties, for they are not
attributes a thing can have at any period of its existence, (iii) There is the attribute
of undergoing rearrangement of parts. We could say that a material thing changes
internally if it has this attributei.e., if its proper parts are changing their spatial
relations to each other. Thus a watch, if it is running, is undergoing internal
change. But this attributehaving proper parts that are changing their relations
to each otherimplies but does not include being a proper part, (iv) There is undergoing a change of causal properties. But this is not an attribute that can be
exemplified by just one individual. For, if an individual undergoes a change of
its causal properties, then it is a member of a set of individuals that are changing
their relations to each other. Hence this attribute includes the attribute of being
an individual thing which is such that there is an individual discrete from it. (Two
individual things may be said to be discrete from each other if they have no parts
in common.) But this attribute cannot be exemplified by just one individual. And
therefore undergoing a change of causal attributes is not a property.
What of going, functioning, working? If these are not purely dispositional, or
if they are not to be thought of as species of moving, then they imply the property
of being expected to accomplish certain things or that of being designed for a certain purpose. Therefore they involve but do not include the properties of expecting or of designing or intending something.
Useful and harmful, if they are not purely dispositional, involve but do not include such properties as benefiting from something and being harmed by
something.
Good and bad, if taken as species of intrinsic value, are attributes that are
necessary to the things that have them and are therefore not properties. But if they
are taken as species of instrumental value, then they are analogous to useful and
harmful.
Beautiful and ugly, if they are not merely dispositional, involve but do not include such properties as taking pleasure in perceiving something and taking displeasure in perceiving something.
What of being alivel It would seem that the word may be taken in two ways.
In one sense, 'being alive' connotes a physiological property. And therefore, in
this sense, it implies but does not include the property of being a physiological
organ. And, in another sense, it is essentially dispositional and thus involves certain properties (physiological or psychological) that it does not imply.
What of being a person! It may be plausibly maintained that being a person
is necessary to whatever has it and is therefore not a property.
105
What of sick and healthy! The words may be taken in one or the other of three
different waysbut on no interpretation do they provide us with a counterexample to our thesis, (i) They may refer to certain physiological properties; in this
case the properties will imply but not include the property of being a physiological
organ, (ii) They may be used, psychologically, to mean the same, respectively,
as 'feeling ill' and 'feeling well'; in this case, the properties that they connote will
satisfy our thesis. Or (iii) the words may be used in a way that is tantamount to
a disjunction of the first two usesto connote the attribute of either being in such
and such a physiological state or feeling in such and such a way. In this case, the
attribute in question would involve certain attributes it does not include.
We have been unable to find any nonpsychological attribute that satisfies our
formula. I conclude, therefore, that the formula gives us a sufficient condition of
the psychological.
The Definition
We will now propose a definition which will give a necessary as well as a
sufficient condition of the psychological.
We could say that any attribute satisfying our formula is 'purely qualitative'.
D6 P is a purely qualitative attribute = Df. P is a property which (a) is possibly such that it is exemplified by just one thing and (b) includes every
property it implies or involves.
We will now introduce a somewhat broader concept of qualitative which will enable us to say that any disjunction of purely qualitative attributes is a qualitative
attribute.1
D7 P is a qualitative attribute = Df. Either (a) P is a purely qualitative attribute or (b) P is equivalent to a disjunction of attributes each of which is
purely qualitative.
Judging and wanting will be qualitative by this definition. But judging that there
are unicorns and wanting a sloop will not be qualitative, for each involves a property it does not include. Hence, the above concept is not identical with that of the
psychological, since judging that there are unicorns and wanting a sloop are both
psychological attributes.
We can now define a broad sense of the psychological:
D8 P is a psychological attribute = Df. P includes an attribute that is
qualitative.
Some attributes that are thus psychological are not purely psychological-e.g.,
believing truly that it is raining and wishing in vain for a sloop. What, then, would
a purely psychological attribute be? I suggest this:
D9 P is purely psychological = Df. P is psychological and every property it
implies involves something qualitative.
106
Our definition applies only to those psychological attributes that are nondispositional. But one may wish to say of certain dispositionsbeing irascible, for
examplethat they, too, are psychological. Given the concept of a disposition,
we could readily extend our definition to such cases. We could say that, in an extended sense of'psychological attribute', a disposition to have a psychological attribute is itself a psychological attribute.
Concerning those attributes that are purely psychological, we may affirm the
following material epistemic principle:
If the attribute of being F is a purely psychological attribute, then, for every
x, if x has the attribute of being F, then it is certain for x that he is F.
This principle could be said to tell us that every purely psychological attribute is
'self-presenting'. But the principle does not hold of those attributes that are only
'partly psychological'. A person may have the attribute of thinking while wearing
a hat or that of thinking about his brother without thereby being certain that he
is thinking while wearing a hat or that he is thinking about his brother.2
Notes
1. An attribute D may be said to be a disjunction of two attributes, P and Q, provided only D
involves P and D involves Q, and D is necessarily such that, for every x, x has D if and only if either
x has P or jc has Q.
2.1 am indebted to Fred Feldman, Richard Foley, Richard Potter, Philip Quinn, Allen Renear,
Bruno Schuwey, Robert Shope, Ernest Sosa, and Michael Zimmerman.
11
Presence in Absence
An account of intentionality should be adequate to the following two theses: (I)
"Mental phenomena can succeed in achieving objective reference" and (II) "Mental phenomena are distinguished by the fact that they may be directed upon objects
that do not exist."1 The second thesis is sometimes said to involve "the problem
of error" or "the problem of presence in absence."2 The first, therefore, might be
said to involve "the problem of truth" or "the problem of presence in presence."
Just what are these problems? I will discuss them by reference only to judging.
But what I will say about judging may also be applied, mutatis mutandis, to other
intentional attitudes.
The Problem
I will first put the matter somewhat loosely. If a person x judges a thing y to be
F, then x makes a judgment which is directed upon y but which is such that one
could make that judgment even if y did not exist. For example, if you now judge
your neighbor to be a philosopher, then your neighbor may be said to be such
that you judge him to be a philosopher. Yet you make a judgment which you could
make even if your neighbor didn't exist. How is that possible?
One might say that there is a world which is exactly like what you find this
one to be at the time you are judging and which does not contain your neighbor.
In such a case, you would be making the judgment you are making now, but your
neighbor would not be such that you are making a judgment about him. Since,
however, your neighbor does exist, your judgment is a judgment about him. It
is a judgment, with respect to him, that he is a philosopher. And so one asks,
"What is there about the judgment, then, that makes it a judgment about your
neighbor!"* The problem is to exhibit the nature of the judging in such a way that
we can see both (1) how it is that, your neighbor having the properties he does
107
108
PRESENCE IN ABSENCE
have, your judging can be said to be directed upon him, and also (2) how this same
judging could be made even if it were not directed upon anything.
An Attempt at a More Precise Formulation

If a person x judges, with respect to an individual thing y, that y is F, then there
is a certain property P which is such that: (1) y has P; and (2) x judges in such
a way that (a) necessarily if there is just one thing that has P, then x judges with
respect to that thing that it is F and (b) it is possible to judge in that way even
if there is not just one thing that has P. We may add (c) that the way of judging
is necessarily such that anyone who judges in that way can know directly that he
judges in that way.
One way of putting the problem of objective reference, then, would be to ask
this: Consider someone x, who judges with respect to something y that it is F;
describe x's way of judging so that we may see that a judging of that sort is such
that (a) it is directed upon something having P, (b) it could be made even if nothing has P, and (c) it is such that anyone who judges in that way can know directly
that he judges in that way.
A partial answer would be the following. The judging is of such a nature that,
if there happens to be an object having the property P, then the judgment is
directed upon that objectit is a judgment, with respect to that object, that it is
F. But unless the object in question is an abstract object, the judging is of such
a sort that it could be made even if nothing had the property P. Let us give a name,
then, to the relation between the judging and the object that the judging is directed
upon. Then we will try to say just what the relation in question is and to explicate
what it means to say that a judgment is "directed upon" an object.
Let us say this: "x makes a judgment which implies that whatever has P is F."
And let us assume, for the moment, that we will be able to specify more exactly
the relevant sense of "implies that." Given this much, we could say that x's judgment is directed upon a thing y if and only if y is the sole thing having the property
P. Then we could answer our philosophical question in the following way.
If you judge your neighbor to be a philosopher and if you have just one neighbor, then you make a judgment which implies him to be a philosopher. The judgment may be said, therefore, to be directed upon him. But you could make the
judgment even if he didn't exist and even if you had no neighbor. For it is possible
for it to be the case both (1) that one makes a judgment which implies that whatever has P is F and also (2) that nothing has P. (Strictly speaking, we should
specify that P not be a property that is essential to some abstract object.)
What is it, then, to make a judgment which implies that whatever has P is F?
I know of at most four general types of such answer.
The First Type of Theory: Abstract Propositions

The "abstract proposition" theory tells us this:
PRESENCE IN ABSENCE
(1)
109
x makes a judgment which implies that whatever has P is F = Df x accepts an abstract proposition which is necessarily such that (a) it is true
if and only if whatever has P is F and (b) whoever conceives it conceives
the property of being-F.
If x judges his neighbor to be a philosopher and if y is x's neighbor, then the abstract proposition that x accepts might be one that x could express by saying: "My
neighbor is a philosopher."
Why clause (b) in our definition? Without clause (b), the definition would require us to say that whoever judges someone to be a philosopher also judges that
person to be such that either he is a philosopher and a Greek or he is a philosopher
and a non-Greek. But this consequence would be false.
The theory presupposes, then, that there are abstract objects which are nameable by "that"clauses and sentential gerundives. And it also presupposes that these
entities can be grasped or conceived by believing subjects.4
If we view intentionality this way, we may say that it is a mistake to suppose
that "presence in absence" is the mark of intentionality. For we must distinguish
two senses of the expression "object of thought." (1) If someone x judges with
respect to something y, that it is F, then we may, if we choose, say that y is the
"object" of x's thought. In such a case we may go on to say that x could make the
judgment he does make even if the judgment had no objectin this sense of "object." (2) But if we say that the "object" of one's judgment is the abstract proposition that one accepts, then, according to the abstract proposition theory, the judgment is necessarily such that it has an object. It is clear that Brentano uses 'object'
in the first way in the following passage: "We can, therefore, define mental
phenomena by saying that they are those phenomena which contain an object intentionally within themselves."5 And it is clear that Russell used the word 'object'
in the second sense when, in "Meinong's Theory of Complexes and Assumptions"
(1904), he wrote: "every presentation and every belief must have an object other
than itself and, except in certain cases where mental existents happen to be concerned, extra-mental."6
But the abstract proposition theory has its difficulties. These may be seen if
we ask how the theory would deal with those judgments we express by means of
demonstratives-such words as T, 'you', 'this', and 'that'.7 One possibility is to
interpret the sense of demonstrative expressions by reference to "individual essences" or "haecceities." What abstract proposition would one express in English
by saying "I am a musician"? This proposition, according to one suggestion, is
like that expressed by saying "The tallest man is a musician," except that where
the latter involves the property of being the tallest man, the proposition expressed
by saying "I am a musician" involves the speaker's individual essence or haecceity. One might be tempted to argue: "My haecceity is the property of being me;
it is appropriately called my individual essence since it is a property that I have
necessarily and that no other thing could possibly have." But is there a property
expressed by "being me"? How does mine differ from yours? Can we really find
110
PRESENCE IN ABSENCE
a difference? If there are no such properties, then, given the abstract proposition
theory, how are we to interpret those judgments we express by means of demonstratives?8
The Second Type of Theory: Singular Propositions

Instead of appealing to abstract propositions, one may appeal to concreteor
singularpropositions. Then we would have:
(2)
x makes a judgment which implies that whatever has P is F = DF x accepts a singular proposition which is necessarily such that (a) it is true
if and only if whatever has P is F and (b) whoever conceives it conceives
the property being-F.
The presupposition of this theory is that, if there is an individual thing x which

has the property being-F, then there is that singular proposition which is x-beingF an entity which contains x as a "constituent" and which is to be distinguished
both from x and from the property being-F.9 If the wealthiest man is such that
you believe him to be generous, then, according to the singular proposition theory, there is a singular proposition, the wealthiest man being generous, into
which the wealthiest man enters as a constituent and which you accept.
Suppose, then, that there exists an x such that x is the wealthiest man: what
of that entity which is x being generousl If the wealthiest man is generous, then
there may be that fact or event which is his being generous and he could be said
to be a "constituent" of that fact or event or to be "directly involved" in it. Singular
propositions, however, are not to be identified with facts or events-unless (what
seems to me to be very problematic) there are events that do not occur. For the
singular proposition theory requires us to say that even if the wealthiest man is
not generous, there is that singular proposition which is his being generous and
he-the wealthiest man himself-is an actual constituent of it. Surely, we should
not feel comfortable about positing such entities.
Moreover,the theory does not appear to be adequate to Castaneda's distinction
between what is expressed by (i) "x judges x to be F" and (ii) "x judges himself
to be F." The second locution implies the first, but not conversely.10 Suppose that
I am looking at a certain photograph which, unsuspected by me, is a photograph
of me. I may conclude that the person depicted was a politician without thereby
thinking that / was a politician. I would be an x such that x judges x to have been
a politician, but I would not be an x such that x judges himself to be a politician.
It is clear that, if I am x, there is no distinction between that singular proposition which is x being a politician and that singular proposition which is my being
a politician. It looks, therefore, as though the singular proposition theory requires
us to say, incorrectly, that "x judges x to be a politician" and "x judges himself
to be a politician" express the same singular proposition.
PRESENCE IN ABSENCE
111
The Third Type of Theory: Properties

The third type of explication is this:
(3)
x makes a judgment which implies that whatever has P is F = Df x

directly attributes to x a property which is necessarily such that (a) whatever has it bears a certain relation R just to a thing which has P and to
a thing which is F and (b) whoever thus directly attributes that property
conceives the relation R.
The property that x directly attributes to himself might be this: having just one
neighbor and a neighbor who is a philosopher.
Judging, according to this conception, is fundamentally a matter of attributing
a property to something. Every judgment includes a direct attribution of a property to oneself. I presuppose that one can directly attribute properties only to oneself. In other words, direct attribution is necessarily such that, for every x and
y, if x directly attributes anything to y, then x is identical with y.
I have defended this conception elsewhere.11 My present concern is not to
define it in detail, but to determine what the alternatives are.
In attributing a property to himself, a person may also happen to attribute a
property to another thing; in such a case, the person indirectly attributes a property to the other thing. If you have just one neighbor and if you directly attribute
to yourself the attribute of being a person whose neighbor is a philosopher, then
you have indirectly attributed to your neighbor the property of being a philosopher. It is not difficult to find, with respect to anything to which we attribute a
property, some relation which is such that we bear that relation only to the thing
to which we attribute the property.
According to this conception, then, the "he, himself locution comes to this:
x judges himself to be F = DF x directly attributes to x the property of
being-F.
Contrast now this locution:
x judges x to be F = Df Either (1) x judges himself to be F; or (2) there is
a relation R such that (a) x bears R to y and only to y and (b) x directly attributes
to x a property which is necessarily such that (i) whatever has it bears R to
something that is F and (ii) whoever conceives it conceives being-F.
Hence we see that, of the two locutions "x judges himself to be F" and "x judges
x to be F," the first implies the second and the second does not imply the first.
Is There an "Inner Sentence" Theory?
We may attempt to formulate a fourth type of theory by replacing the reference
to "propositions" in our first two formulae by a reference to sentences. Then we
would have:
112
PRESENCE IN ABSENCE
(4.1) x makes a judgment which implies that whatever has P is F = DF x accepts a sentence which is necessarily such that (a) it is true if and only
if whatever has P is F and (b) whoever conceives it conceives the property being-F.
But this will not do as it stands. The English sentence, "My neighbor is a philosopher" is not necessarily such that it is true if and only if my neighbor is a philosopher. For our language could have developed in such a way that "My neighbor
is a philosopher" means something very different.
Perhaps, therefore, one might say something of this sort:
(4.2) x makes a judgment which implies that whatever has P is F = Df S accepts in a language L a sentence which in L is necessarily such that (a)
it is true in L if and only if whatever has P is F and (b) whoever understands it in L conceives the property being F.
But what is it "to accept a sentence S in a language L"? One has the feeling
that this comes to no more than: "to accept what it is that S expresses in L." In
this case the inner sentence theory is not an alternative to the other theories.
One may, of course, make a judgment about a sentencea judgment which
is directed upon a sentence. And some philosophers would replace "x accepts a
sentence S in L" by something like "x recognizes S in L as expressing his attitude."12 The latter expression, however, implies "x takes p in L to express his attitude," which, in turn, implies the intentional locution, "x takes y to be F." Hence
this move leaves us with our problem.
It is difficult, therefore, to avoid the conclusion that, despite all that has been
written about the relevance of language to intentionality, there really is no linguistic interpretation of intentionality.
These considerations confirm a traditional view that may be expressed by saying "the intentional is prior to the semantical."
Are There Alternatives?

During the past few years there has been an enormous amount of literature
devoted to these problemsmuch of it containing such expressions as "noemata"
and "inner systems of representation." It seems to be thought that there are alternatives which are much more profound than anything suggested by the present
analyses. If there really are such alternatives, one has only to produce formulae
comparable to the four just cited. But I have no idea what the formulae might be.
Notes
1. Similar theses are formulated by Maurita J. Harney, in Intentionality, Sense and the Mind
(The Hague: Martinus Nijhoff, 1984), p. 2.
2. Josiah Royce discusses "the problem of error" in Chapter 40 ("The Possibility of Error") in
The Religious Aspect of Philosophy (Boston: Houghton Mifflin, 1898), pp. 384-435. William James
uses "presence in absence" in The Meaning of Truth (New York: Longmans Green, 1900), p. 44.
PRESENCE IN ABSENCE
113
James asks what it is for a person to believe that there are tigers in India and observes: "A great mystery is usually made of this peculiar presence in absence; and the scholastic philosophy . . . would
explain it as a peculiar kind of existence, called intentional inexistence, of the tigers in our mind."
3. Compare Wittgenstein: "What makes my idea of him an idea ofhimT Philosophical Investigations (Oxford: Basil Blackwell, 1953), p. 177. Compare also his Last Writings on the Philosophy of
Psychology, vol. 1 (Chicago: University of Chicago Press, 1982): "What makes this sentence a sentence that has to do with himT (p. 43).
4. Frege seems to have accepted the abstract proposition theory; see his essay "The Thought:
A Logical Inquiry" (1918-19), translated in Mind, 65 (1956): 289-311. I believe that Husserl also
accepted the abstract proposition theory. At any rate, this is the only way I can interpret what he says
about noemata in the Ideen zu einer Phdnomenologie und phdnomenologischen Philosophic (Turbingen: Max Niemeyer, 1980), pp. 265-75. Compare Dagfinn Follesdal, "Husserl's Notion of the
Noema," Journal of Philosophy, 66 (1969): 680-87. Russell, in the article referred to below, also
seems to accept the abstract proposition theory. The expressions "states of affairs" and "event-types"
are sometimes used instead of "abstract propositions" in expressing this theory.
5. Psychology from an Empirical Standpoint (London: Routledge & Kegan Paul, 1973), pp.
88-89; Psychologic vom empirischen Standpunkt, Zweiter Band, (Hamburg: Felix Meiner Verlag,
1971), pp. 124-25. Brentano seldom used such expressions as "intentional inexistence." In The Psychology of Aristotle, he used "exists objectively, i.e., as cognized object within us [objektiv, d.h. als
Erkanntes in uns existiert]." Psychology of Aristotle (Berkeley: University of California Press, 1977),
p. 54; compare pp. 77, 210 (n6), and 229 (n23); Die Psychologic des Aristoteles (Mainz: Franz Vlircheim, 1867), pp. 80, 120.
6. Bertrand Russell, Essays in Analysis, ed. Douglas Lackey (New York: George Brazilier,
1973), p. 21.
7. See John Perry, "Frege on Demonstratives," Philosophical Review, 86 (1977): 474-97.
8. Ernest Sosa has suggested that there are two types of propositionthose which are nonperspectival and those which are perspectival; and possibly his distinction may help with some of these
difficulties. See Ernest Sosa, "Propositions and Indexical Attitudes," in Herman Paret (ed.), On Believing: Epistemological andSemiotic Approaches (Berlin and New York: Walter de Gruyter, 1983),
pp 316-32; see esp. pp. 322-23. Compare also his "Consciousness of the Self and of the Present,"
in James Tomberlin (ed.), Agent, Language and the Structure of the World (Indianapolis: Hackett
Publishing, 1983), pp. 131-43.
9. Compare David Kaplan, Draft Two of his privately circulated monograph, An Essay on the
Semantics, Logic, Metaphysics and Epistemology of Demonstratives and Other Indexicals (1977).
Kaplan observes that "free variables under an assignment of values are paradigms of what I have been
calling directly referential terms" (p. 2).
10. See Hector-Neri Castaneda, "He: A Study in the Logic of Self-Consciousness," Ratio, 8
(1966): 130-57. Compare also the same author's Thinking and Doing (Dordrecht: D. Reidel, 1975),
Parti.
11. See The First Person: An Essay on Reference and Intentionality (Brighton and Minneapolis:
Harvester Press and University of Minnesota Press, 1981); and "The Primacy of the Intentional," Synthese, 61 (1984): 89-108.
12. A recent example is provided by Lynne Rudder Baker, "De Re Belief In Action," Philosophical Review, 91 (1982): 363-87; see in particular, 380.
12
Questions about Minds

The Mind-Body Problem
What is the mind-body problem? The word "mind" has a number of rather different uses and this fact has led to some confusion in recent discussions. We may
distinguish at least five such uses and therefore at least five senses of the "mindbody problem."
(1) We could use the term "mind," as Descartes had used the term "mens" to
refer to that which has psychological propertiesto that which thinks, senses, believes, desires. In this case, we would be using "mind" to mean the same as "person" and hence to designate such entities as you and me. If we use "mind" to mean
the same as "person," one form of the mind-body problem would be the question:
What is the relation between persons and their bodies? What is the relation, for
example, between me and my body? There are two broad possibilities: Either I
am identical with my body or I am not identical with it. If I am not identical with
my body, then once again there are two possibilities: Either I am identical with
something that includes a part of my body or I am not identical with anything that
includes a part of my body. In the latter case, what kind of thing am I?
It has been suggested in recent years that "the mind" is related to the body in
the way in which the abstract diagram of a computer is related to the computer.
But this suggestion cannot be true if "mind" is taken, in the sense just distinguished, to refer to that which has psychological properties. For the abstract diagram of a computer is itself a property, an abstract object; but that which has psychological propertiesthat which senses, thinks, feels, and desiresis an
individual thing.
(2) We could also use the term "mind," as many now do, to refer to a person's
intellectual capacities. The word is being used in this way when one says, "That
person has a good mind," meaning that the person is intelligent. An "investigation
114
QUESTIONS ABOUT MINDS
115
of the mind," in this second sense of the word "mind," would be a psychological
investigation of intellectual capacities. In this case, the substantive "mind" may
be misleading, since it might be taken to designate an individual thing that exists
along with the person and his body. In such a case one is mislead into supposing
that there are three individual things to be related-the person, the person's mind,
and the person's body.
Using "mind" in this second way, may we say that there is a mind-body problem? Perhaps there is the question "How can a body have intellectual capacities?"
If we may assume that there is something that has intellectual capacities, then this
version of the mind-body problem reduces to the first: What is the relation between a person and his body? For it is presumably the person who has intellectual
capacities.
(3) We may also speak of a person's mind as being that by means of which he
thinks. In this case the term "mind" does designate an individual thing. That by
means of which one thinks is quite obviously the brainor at least something that
includes a part of the brain. If we use "mind" in this way (to mean that by means
of which one thinks) and not in the first way (to mean the person) and not in the
second way (to refer to intellectual capacities), then what we would call an "investigation of the mind" would be a neurophysiological investigation of the brain.
And the expression "mind-body problem" would refer to certain questions of neurophysiology. Answers to these neurological questions do not, as such, give us
a solution to the "mind-body problem," where this expression is interpreted in accordance with the first sense of "mind" distinguished above.
(4) The term "mind" is sometimes used to designate a spiritual substance, an
individual thing of a nonmaterial nature. (Here, perhaps, the word "soul" is more
appropriate.) If, in dealing with the person-body problem, we were to decide that
the person is not identical with his body or with anything that includes a part of
his body, then we might want to consider the possibility that the person is a soul
or mind, in the present sense of the word "mind." (If we conclude that the person
is a soul or mind, then we should take care not to express this conclusion by saying
that the person has a soul or mind. For then we might be misled, once again, into
thinking that we are dealing with three types of individual thingpersons, minds,
and bodies.)
(5) Finally, "the mind-body problem" is sometimes taken to be that of studying
the relation between psychological and physical properties and thereby getting a
better understanding, not only of the nature of the psychological, but also that of
the physical universe. The concern is with such question as: Is each psychological
property identical with some physical property? Is each psychological property dependent in some special sense upon physical property or upon some set of physical
properties? Here, once again, the expression "mind-body problem" may be misleading once again; I suggest that "psychophysical problem" would be better.
Most contemporary discussions that purport to be about the mind-body problem are concerned primarily with the psychophysical problem.
116
Descriptive Psychology
Much of what has been called "philosophy of mind" is concerned with describing
the psychological states and properties of persons and describing them as they are
experienced by those who are in those states or have those properties. One purpose of doing this is to resolve the philosophical perplexity to which some of these
states and properties give rise. The hope is thereby better to understand the person
who is the subject of such states and properties. This type of philosophy of mind
includes, not only the descriptive psychology of Brentano, but also the "phenomenology" of Husserl and the later existentialists, as well as the type of study to be
found in Wittgenstein's Philosophical Investigations.
Some of the philosophers who have called themselves "phenomenologists"
have said that their descriptions are designed to catch the essences of such
phenomenathe essence, say, of such a phenomenon as being sad. The essence
of an individual thing would be a set of properties that that thing and only that
thing has necessarily and that nothing else could possibly have. Now sadness is
a property and not an individual thing. Shall we say, then, that the essence of the
property sadness is a property of the property sadnessa property that is essential
to it and only it and repugnant to every thing else? Then there would be no
difficulty in saying what the essence of sadness is: sadness is a property which
is necessarily such that it is exemplified in all and only those things that are sad.
When philosophers have said they were describing the essence of sadness,
what they may have meant is that they were providing with an analysis of sadness.
What would that be?
Perhaps this: the philosophers are concerned with a familiar phenomenon and
that they are pointing out certain essential properties of it that are revealing but
often go unnoticed. If you say, "Sadness is to be analysed as being so-and-so and
such-and-such," and if I understand what you are saying, then either I am already
acquainted with sadness or I take you to be saying that that which you are acquainted with as sadness is so-and-so and such-and-such. In either case, I am able
to identify the analysandum independently. More generally, if one professes to
give an informative statement about the nature of a state or property, then one
presupposes that that property can be independently identified. But some philosophers who profess to tell us about the "true nature" of the psychological do not
presuppose this.
Such philosophers say that the data to which the descriptive psychologists appeal are themselves doubtful, being comparable to the old wives' tales of "folk
psychology." These philosophers are apparently concerned with what psychological states and psychological properties would be if (1) psychology were a science
in the sense in which physics and chemistry may be said to be science and if,
nevertheless, (2) psychology were to investigate psychological states and properties. The concern, then, is with such questions as "What sorts of things might appropriately be called 'believing,' 'feeling,' 'sensing,' and the like if psychology
were a natural science?" Sometimes it is suggested that what might then be called
117
"psychological" or "mental" would be very different from the psychological states

and properties that we are in fact acquainted with. In this case, the philosophers
may have no theory about the nature of those psychological properties that we are
directly acquainted with, since they may profess not to believe that there are such
properties. It is difficult, therefore, to evaluate their statements since it is not clear
what they take as their data or what they would accept as confirmation or as falsification of what they are trying to say.
The Psychophysical Problem
The authors of the Wissenschaftliche Weltauffassung and those who followed
them in the tradition of "physicalism" hoped to be able to identify psychological
states and properties with physical states and properties. But with what physical
states and properties? The program seemed promising in the case of sensations
where, it was thought, there is a physical correlate that is unique to every type
of sensation. One could then propose as a general hypothesis that each sensation
is identical with its physical correlate. But there was difficulty finding such correlates for intentional phenomena (e.g., believing, hoping, questioning, desiring).
With what physical state, for example, are we to identify the desire to own a 23foot sloop? There would seem to be no identifiable physical state that is uniquely
correlated with such a desire.
It came to be thought, therefore, that such intentional phenomena might be
identified, not with physical states or properties, but with dispositions to behave.
A crude form of this view would be that you desire to own a 23-foot sloop if and
only if you would respond affirmatively when asked whether you desire to own
a 23-foot sloop. But this biconditional needs to be qualified. Thus it could be false
if you do not desire such a sloop but wish to deceive someone into thinking that
you do. And it could be false if you do desire such a sloop but do not understand
the language of the one who is questioning you. Can we find suitable qualifications, then, to add to the biconditional? The difficulty is that the qualifications we
must add to deal with such examples will refer to other intentional phenomena
for example, to other things that you desire or do not desire, to what you believe
or do not believe, to what you perceive, and to what you remember. Such a theory
could hardly be said to reduce psychological phenomena to dispositions to
behave-since the relevant dispositions can be described only by reference to
other psychological phenomena.
Could one defend a psychophysical identity thesis without appeal to a psychophysical correlation thesis? One cannot say, of course, that a thing may be
identical with one thing at one time and with another thing at another time. But
it has been suggested that the particular occurrences ("tokens") of any given psychological property may be such that some are identical with the particular occurrences of one physical property and others are identical with the particular occurrences of some other physical property. But what ground is there for the
metaphysical thesis according to which there are such particular occurrences
118
(tokens) of properties? So far as I have been able to see, the thesis has no basis
at all other than the possibility that it might help us with the psychophysical
problem. The metaphysical price of accepting this metaphysical doctrine in order
to preserve the psychophysical identity theory seems very high, especially when
we consider that this "token-token" version of the theory is par excellence, a theory which is neither verifiable nor falsifiable.
Some "functionalists" have said, in effect, that even though a piecemeal reduction of the psychological to the physical is not possible, a wholesale reduction is
possible. The point is, not to identify particular psychological states (such as
desiring a 23-foot sloop) with neurological states or with dispositions to behave
in certain particular ways, but to identify one's total psychological state at any
time one's total physical state at that time. Such a theory is no longer in the spirit
of the Wissenschaftliche Weltauffassung, for, like the "token-token" theory just
considered, it would seem to be a paradigm case of a theory that is neither verifiable nor falsifiable.
13
Is There a Mind-Body Problem?

The title"Is there a Mind-Body Problem?"will suggest that I have doubts as to
whether there is a mind-body problem. And I do have doubts as to whether there
is a special problem concerning the relation between the mind and the body. You
may say: "Well, plenty of people have worried about the problem of the relation
of the mind and the body. And so therefore there is a problem." And of course that
is true enough: people have been concerned with it. But what I wish to suggest is
that they shouldn't have been concerned with it: there is no evidence to suggest that
I have something to be called a mind which we must relate somehow to the body.
Now in saying thisthere is no reason to suggest that I have something to be
called a mindI do not mean to say that there is no person-body problem. If we
use "person" to designate such entities as you and me, then there is no question
but that there are such things as persons. And obviously there is no question but
that there are such things as our bodies. There is a problem about the relation between those entities. Thus there is the question: What is the relation between me
and my body? There are two broad possibilities: Either I am identical with my
body or I am not identical with it. And if we decide that I am not identical with
it, then once again there are two possibilities: Either I am identical with some part
of my body or I'm not. And If I'm not, then just what kind of thing am I? So there
is a person-body problem.
But I want to urge that we multiply problems beyond necessity if we suppose,
that in addition to the person-body problem, there is also a mind-body problem.
In suggesting that there is no mind-body problem, then, I am suggesting this: if
the substantival expression "mind" is taken to designate some individual thing
which is other than the person, something that person may be said to have, just
as it has a hand and a foot, then there is no reason to suppose that there is such
a thing as the mind; and if there is no reason to suppose that there is such a thing
as the mind, then there is no problem about how it may be related to the body.
119
120
IS THERE A MIND-BODY PROBLEM?
One Case for Minds

So let us ask, then, "Why assume that there is such a thing as my mind?" This
is different from asking "Why assume that there is such a thing as me?" And it
is also different from asking "Why assume that I have various mental properties
and potentialities, such as the the ability to think or to think in such ways?" For
the assumption that / have these mental properties and potentialities doesn't imply
that I have a mind which has them.
Why should one suppose that there is a nonmaterial thing which is the mind?
Aristotle had argued that "that in the soul which is called mind (by mind I mean
that whereby the soul thinks and judges) . . . cannot reasonably be regarded as
blended with body."1 The mind, he said, must be "capable of receiving the form
of an object" but without thereby becoming that object.2 And this would be impossible if the mind were itself a material thing.
Aristotle's reasoning was essentially this: (1) If you apprehend a thingsay,
a dogthen you do it by means of something which bears a certain intimate relation to the form or nature of a dog. But (2) a material thing couldn't bear the requisite relation to the form or nature of a dog unless the material thing were thereby
to become itself a dog. On the other hand (3) a nonmaterial thing could bear the
requisite relation without thereby becoming a dog. Hence if you and I can apprehend dogs, and of course we can, then it is by means of a certain nonmaterial
thing which is our mind.
What are we to say of this argument? The argument requires a more specific
characterization of the relation in questionthe relation that must be born to the
form or nature of a dog if one is to be able to apprehend a dog. Until we have
such an account, I think we must say that both premise (1) and premise (2) are
problematic.3
Perhaps the most important consideration which may make us wonder whether
there is a nonmaterial substance which is a mind is the nature of our immediate
experienceour experience of what are sometimes called "sense-data" or "appearances." Let us consider one twentieth-century conception of appearances, for
this was thought by many to demonstrate an irreducible dualism between mind
and body. I am referring to the view set forth by A. O. Lovejoy in his book The
Revolt Against Dualism (1930).
"No man doubts," Lovejoy wrote, "that when he brings to mind the look of
a dog he owned when a boy, there is something of a canine sort immediately present to and therefore compresent with his consciousness, but that it is quite certainly not that dog in the flesh" (p. 305). The thing that is there-the something
of a canine sort that is immediately before the mindis not itself a physical object, Lovejoy said; it is a private, psychological object, conditioned by a series
of physiological and psychological events, reaching back to the earlier dog which
it now reveals.
If the man now looks at his desk, then, according to Lovejoy, there is another
series of physiological and psychological events, this time involving the activity
121
of sense organs, but resulting as before in a private, psychological objecta sensation, this time something of a desk sort, a "visible desk" which in certain
respects serves to duplicate the real, external, physical desk which it makes
known to us.
Both of these examplesthe earlier dog and the external desk being presented
by an inner visual deskprovide us with the essentials of two philosophical theories, which Lovejoy had referred to as "epistemological dualism" and "psychophysical dualism." According to "epistemological dualism," which is a thesis
about our knowledge, we have direct or imemdiate knowledge only of certain private or subjective states; some external objects, past or present, are "duplicated"
in these private or subjective states and it is in virtue of this duplication that we
know what we do about the rest of the world. Our knowledge of external things
and of past events involves a "cleavage" between the object of our knowing and
the subjective vehicle which makes that object known. And according to "psychophysical dualism," which is a thesis about reality, the world is constituted out
of at least two fundamentally different kinds of stuff-the physical or material
things that are studied by physics, and the psychical or mental things that are objects of our private or subjective states. When asserted in conjunction, as they
were by Lovejoy, and in the seventeenth century by Descartes and Locke, these
two forms of dualism imply that our knowledge of physical or material things is
derived from our knowledge of the mental or psychical duplicates of these things.
Our present interest is in the second of these types of dualism -psychophysical
dualism, the view that there is a set of mental or psychical entities, which are appearances or sense-data, and that these psychical entities are housed in a psychical
place, known as "the mind."
The Sense-Datum Fallacy

Let us begin by considering those strange entities which are sense-data or appearances.
It was supposed, for example, that if a man were to walk around a table, while
focusing on the white tablecloth on the top, he could experience a great variety
of sense-data or appearances. Some of these entities would be rectangular like the
tabletop itself; they would be the ones he would sense if he were to get his head
directly above the table and then look down. Most of them, however, would be
rhomboids of various sorts. If the lighting conditions were good and the man's
eyes in proper order, most of the appearances would be white, like the tablecloth.
But if the man were wearing rose-colored glasses, he might sense appearances
that were pink, or if he were a victim of jaundice, he might sense appearances
that were yellow. The other senses, as well as imagination, were thought to bring
us into relation with still other types of appearances or sense-datum.
It was assumed that, if a physical thing appears white or rhomboidal or bitter
to a man, then the man may be said to sense or to be aware of an appearance that
is white, or an appearance that is rhomboidal, or an appearance that is bitter. It
122
was assumed that if a dog presents a canine appearance, then the dog presents an
appearance that is canine.4 And it was assumed, more generally, that whenever
we have a true statement of the form "Such-and-such a physical thing appears,
or looks, or seems so-and-so to Mr. Jones," we can derive a true statement of the
form "Mr. Jones is aware of an appearance which is in fact so-and-so." But this
assumption is quite obviously false.5
Consider the following reasoning, which would be quite sound if the assumption were true: "That dog looks vicious and more than 10 years old. Therefore
he presents an appearance which is vicious and he presents an appearance which
is more than 10 years old." It is absurd to suppose that an appearance, like a dog
or a man, may be vicious or more than 10 years old. It is also absurd to suppose
that an appearance may be a dogi.e., something of a "canine sort." And, I think,
it is equally absurd to suppose that an appearance, like a tablecloth, may be rectangular, or pink, or white.
We should compare the grammer of (a) "I sense a red appearance," (b) "I have
a depressed feeling," and (c) "I have a green Chevrolet." The sense-datum philosopher interprets (a) as though it resembled (c) more than (b). But I suggest that
it should be taken in such a way that it resembles (b) more than (c). Thus "I have
a depressed feeling" should not be taken to say that I have a feeling that is itself
depressed. It doesn't predicate being depressed of a feeling; it predicates feeling
depressed of a person. And "I sense a canine appearance" doesn't predicate caninity of an appearance. It predicates being appeared to in-a-certain-way to me. I'm
appeared to in a way that is optimal for the perception of dogs. Being appeared
to is an undergoinga nonrelational quality of the person.
And so, if what we have said is correct, then one of Lovejoy's arguments for
psychophysical dualismthe dualism of mind and bodyis inconclusive. For
Lovejoy had argued: (1) We see desks and stars and other objects by means of
internal desks and stars which are not identical with the objects they enable us
to perceive; but (2) no place among physical objects can be found for such internal
desks and stars; therefore (3) the latter objects inhabit "the world of the mind"
and not "the world of matter." But if premise (1) is false, this argument for psychophysical dualism is no longer available. Since there are no internal desks and
stars, the materialists need not be asked to find a place for them. (But he must,
of course, fit the fact of appearing into his scheme of things.)
But does this settle the matter? What if being sad and being appeared red to
are undergoingsand not relations between persons and sense-data? Isn't there
something special about these undergoings? After all, they seem to give the world
a "qualitative dimension" it might otherwise not have.6 And isn't this qualitative
dimension a mental or psychical aspect of the world?
Let us try to do justice to this. In particular, let us consider what has sometimes
been called the "double aspect" theory. I think that those who spoke this way may
have been on the right track.
123
The Daylight View of Matter

The property of being depressed, I have said, is not a property of a feeling and
the property of being red is not a property of a sense-datum. Rut feeling depressed
and being appeared to redly are properties of the personthey are nonrelational
qualities of the person. So, too, for those other properties that present us with "a
qualitative dimension of being." They are all "modifications" of the subject of experience.
But if the subject of experience thus exhibits a "qualitative dimension of being," then doesn't the subject have certain mental properties thereby? And doesn't
this mean that the subjectthe personis a mindl
It may be natural to say that such a qualitative dimension is "mental," and this
statement is harmless enough if "mental" is taken to mean the same as "that which
is known immediately." For being appeared to is mental in that respect: roughly
speaking, it can't happen to you unless you know that it is happening to you. But
this use of the adjective "mental" should not be taken to suggest the substantive
"mind." There is no reason to suppose that only minds can have mental properties
in this sense of the word "mental." It is possible that, in this sense of "mental,"
physical things may have properties that are "mental." Or to put the matter more
carefully, it is possible that there are things having properties that are physical
and also having properties that are mental.
These remarks will recall what has sometimes been called "the double aspect
theory." The theory may be put by saying that persons have "inner" and "outer"
aspects. The "inner" aspects are "mental"; that is to say, they are those subjective
and intentional properties which are necessarily such that, if a person has them,
then it is evident to the person that he has them. And the "outer" aspects are certain
physical properties.
It is essential not to confuse this use of the expression "double aspect" with certain other uses that the expression has been given in recent philosophy. Thus we
are using it for the view according to which certain individual thingsnamely
personshave both "mental" and "physical" properties. But it has also been used
for the view according to which there are certain events or activities having both
mental and physical properties.7 The latter view is quite different from the one
that I have suggested. (Obviously, it would be a category mistake of the most
egregious kind to identify persons with events or activities.) The expression "double aspect" theory has also been used for the so-called identity-theory, according
to which "mental events" are to be identified with "physical events."8 But "the double aspect theory," as I use this term, refers to a theory about the nature of persons
and not to a theory about events.
The double aspect view has been clearly set forth by Gustav Theodor Fechner.9 Thus he held that we are "to ourselves" psychical and "to others" material.
The important point of his doctrine, it seems to me, is the assertion that we are
both mental and material. For the assertion implies that what is material can also
be mental. A material thing can be "intrinsically psychical."
124
Perhaps we could accept this conclusion without holding, as Fechner did, that
all matter is intrinsically psychical. But if some material things are intrinsically
psychical, what about the others? What could their intrinsic properties be? Do we
know of any intrinsic properties other than those we have been calling psychical?
Or is it possible that only some individual things have intrinsic properties and
hence that others have no intrinsic properties?
The word "intrinsic" has two senses here. One is suggested by the concept of
"self-presentation"; Fechner says that certain things are "psychical to themselves," and this means that they have certain states that "present themselves." The
other sense of "intrinsic" is that of nonrelational: an intrinsic property of a thing
would be a property not entailing relations of the thing to other things.
Fechner called this the "daylight view" (die Tagesansichf) of matter and contrasted it with the "night view" of matter.10
The "double aspect theory" tells us this: There are certain things which have
physical properties and therefore physical objects; some of these things also have
certain mental or intentional properties; and personsyou and Iare such things
as these.
C.A. Strong put this last point clearly. He wrote:
7 am to outer appearance physical but to inner perception psychical; there
is therefore no contradiction in a thing being at once physical, that is, extended, composed of parts, productive of effects, and psychical, that is of
the nature of feeling.11
Strong is not here saying that "my mind" is an aspect of a physical thing, much
less that / am an aspect of a physical thing. What he says is that there is a certain
physical thing which has inner and outer aspects and that that physical thing is
identical with me.
"Which Physical Thing Am I?"
If we were to accept this theory, then we could ask: "Which physical thing am
I?" I am afraid we could not provide a precise answer to this question.
If I am in fact a physical thing, then, it should be obvious, that physical thing
is either this gross physical body now standing before you or it is some proper
part of this gross physical body. There are, of course, many philosophical arguments professing to show that the person cannot be identical with his gross macroscopic physical body. Some of these arguments, I think, are soundin particular
those appealing to certain facts about persistence through time.
The body that persists through timethe one I have been carrying with me,
so to speakis an ens successivum. That is to say, it is an entity made up of different things at different times. The set of things that make it up today is not identical
with the set of things that made it up yesterday or with the set of things that made
it up the day before. Now one could say that an ens successivum has different
"stand-ins" at different times and that these stand-ins do duty for the successive
125
entity at the different times. Thus the thing that does duty for my body today is
other than the thing that did duty for it yesterday and other than the thing that will
do duty for it tomorrow. But what of me?
Am / an entity such that different things do duty for me at different days? Is
it one thing that does my feeling depressed for me today and another thing that
did it yesterday and still another thing that will do it tomorrow? If I happen to
be feeling sad, then, surely, there is no other thing that is doing my feeling sad
for me. We must reject the view that persons are thus entia successiva.
Our reasoning can be summarized. Suppose (i) that I am now sad. Now (ii)
if there is an ens successivum that bears my name and is now sad, then it is sad
in virtue of the fact that one of its stand-ins is now sad. But (iii) I am not sad in
virtue of the fact that some other thing is doing my feeling sad for me. Therefore
(iv) I am not an ens successivum.
What would be an ens nonsuccessivuml If an individual thing were a nonsuccessive entity, what would it be like? If an ens successivum is an individual thing
that is made up of different things at different times, then an ens nonsuccessivum
would be an individual thing that is not made up of different things at different
times. This means that, at any moment of its existence, it has precisely the same
parts it has at any other moment of its existence; at no time during which it exists,
does it have a part it does not have at any other time during which it exists.
It is tempting to reason, in Leibnizian fashion: "There are entia successiva.
Therefore there are entia nonsuccessiva.n I believe this reasoning is sound. I
would add, moreover, that every extended period of time, however short, is such
that some ens nonsuccessivum exists during some part of that time. For I believe
it is only by presupposing this thesis that we can make sense of the identity or
persistence of any individual thing through time.
Might I not be, then, such an ens nonsuccessivum^! Leibniz mentionsand
rejectsa theory which is similar to this. "The soul," he says, "does not dwell
in certain atoms appropriated to itself, nor in a little incorruptible bone such as
the Luz of the Rabbis."12 Of course, the hypothesis I have suggested, if filled in
by reference to such a material thing as the Luz bone, would not imply that "the
soul" dwells thereif the soul is understood to be something other than the person, still another thing that the person "has." We would be saying rather that the
person dwells there. And to say that he "dwells" there would be to say that the
person is the Luz bone or some proper part of it.
If we accept this theory, then, of course, we part company with personalism.
The doctrine that persons are physical thingseven intactly persisting physical
things-would not have been taken seriously by Borden Parker Bowne and his
followers. Yet, if we view the person in the way I have suggested, we may go
on to affirm many of the other philosophical theses that the personalists felt to be
important. Thus we could say, as Bishop Butler did, that "our gross organized
bodies with which we perceive the objects of sense, and with which we act, are
no part of ourselves. . . . We see with our eyes in the same way we see with
our glasses."13 The eyes are the organs of sight, not the subject of sight. We could
126
say, as Butler and the personalists did, that the destruction of the gross physical
body does not logically imply the destruction of the person. And we could accept
the view that St. Thomas attributes to Plato: the person is "in a body in somewhat
the same way as a sailor is in a ship."14
Some Objections Considered

To understand the view that is being proposed, let us formulate certain objections
that readily come to mind and then attempt to reply to them. I will consider four
such objections.
(1) "The hypothesis you are considering implies, then, that there is a kind of
matter that is incorruptible and that the person is a material thing of that sort? But
this is hardly adequate to the facts of physics."
The reply is that the theory does not imply that there is certain matter that is
incorruptible. It implies rather that there are certain material thingsin all probability, certain material particles or subparticlesthat are incorrupted and remain
incorrupted as long as the person survives.
The theory would be, then, that I am literally identical with some proper part
of this macroscopic body, some intact, nonsuccessive part that has been in this
larger body all along. This part is hardly likely to be the Luz bone, of course;
more likely, it would be something of a microscopic nature, and presumably
something that is located within the brain.
(2) "Persons, being thinking things, must have a complex structure. But no
microscopic entity that is known to physics has the equipment that is necessary
for thinking. After all, you can't think unless you have a brain. Andthose little
things don't have brains!"
The hypothesis being criticized is the hypothesis that / am such a microscopic
entity. But note that I do have a brain. And therefore, according to the hypothesis
in question, the microscopic entity has one, toothe same one that I have, the
one that is inside my head. It is only a confusion to suppose that the microscopic
entitywhich may in fact be inside my brainhas another brain which is in fact
inside of it.15
The brain is the organ of consciousness, not the subject of consciousness
unless I am myself my brain.16 The nose, similarly, is the organ of smell and not
the subject of smell-unless I am myself my nose. But if I am one or the other the brain or the nosethen, I the subject, will have some organs that are spatially
outside me.
The hypothesis in question, then, is that I am a certain proper part of my brain.
This would imply that the subject of consciousness is a proper part of the organ
of consciousness.
(3) "You say I'm identical with some microscopic particle or some subparticle.
But I am 6 feet tall and weigh 175 pounds. Therefore your theory would imply
that there is a certain microscopic particle which is 6 feet tall and weighs 175
pounds. But this is absurd and therefore your theory is absurd."
127
The argument, of course, errs in taking too literally the premise expressed by
saying "I am 6 feet tall and weigh 175 pounds." For what the premise actually
tells us is that I have a body which is 6 feet tall and weighs 175 pounds.
(4) "Do you mean to suggest seriously, then, that instead of weighing 175
pounds, you may weigh less than a milligram?" The answer has to be yes. We
must be ready, therefore, to be ridiculed, for, in this case, even those who know
better may be unable to resist the temptation. But those who do know better will
realize that a person can truly say, in one sense, that he weighs 175 pounds, and
in another sense, that he weighs less than a milligram. The formulation of the first
statement would be more nearly accurate (I say "more nearly accurate" not
"more nearly correct") if it read: "I have a body that weighs 175 pounds."
Speaking in a loose and popular sense, I may attribute to myself certain properties of my gross macroscopic body. (And speaking to a filling station attendant
I may attribute certain properties of my automobile to myself: "I'm down there
on the corner of Jay Street without any gasoline." The response needn't be: "How,
then, can you be standing here?" One might say that the property of being down
there is one I have "borrowed" from my automobile.) But if I am a microscopic
part of my gross body, then, strictly and philosophically, one cannot attribute to
me the properties of it. The properties of weighing 175 pounds and being 6 feet
tall are properties I "borrow" from my body. Strictly and philosophically, it has
them and I do not.17
Conclusion
What are the possibilities, after all? There are persons. Therefore either the person is a physical thing or, as Lovejoy suggests, the person is a nonphysical thing.
But does anything we know about persons justify us in assuming that persons are
nonphysical individual things?
If I am a physical thing, then the most plausible hypothesis would seem to be
that I am a proper part of this gross microscopic body, even if there is no way
of telling from the "outside" which proper part I happen to be.
If I am a nonphysical thing, then either I am composed of some nonphysical
stuff or I am a monad. But what kind of "stuff" could be nonphysical?
Notes
1. De Anima, 429a.
2. Ibid.
3. "In the picture and the pictured there must be something identical in order that the one can
be a picture of the other at all." L. Wittgenstein, Tractatus Logico-Philosophicus, 2.161.
4. A. O. Lovejoy, The Revolt Against Dualism (La Salle, 111.: Open Court, 1930), p. 305.
5. "The general rule which one may derive from these examples is that the propositions we ordinarily express by saying that a person A is perceiving a material thing M, which appears to him to
have the quality x, may be expressed in the sense-datum terminology by saying that A is sensing a
sense-datum s, which really has the quality of x, and which belongs to M." A. J. Ayer, The Foundations of Empirical Knowledge (New York: MacMillan, 1940), p. 58.
128
6. This expression is used by Roy Wood Sellars, in Evolutionary Naturalism (Chicago: University of Chicago Press, 1922), pp. 306-7.
7. For example, by R. J. Hirst, The Problems of Perception (London: Allen & Unwin, 1959),
p. 189.
. 8. Compare Herbert Feigl, "Physicalism, Unity of Science and the Foundations of Psychology,"
in P. A. Schilpp (ed.), The Philosophy of Rudolf Carnap (La Salle, 111., Open Court, 1963), pp.
227-267; see p. 254.
9. See in particular his Uber die Physikaalische und die Philosophische Atomlehre (1855) and
The Little Book of Life After Death (English translation, 1912).
10. Die Tagesansicht Gegenueber der Nachtansicht, 1879.
11. C. A. Strong, "Final Observations," Journal of Philosophy, 38 (1941): 233-43; the quotation
is on p. 237.
12. New Essays Concerning Human Understanding, Book II, Chap. 28 (La Salle, 111., Open
Court, 1916). p. 242. Alfred Langley, editor of this edition of Leibniz, quotes an ancient discussion
of the Luz bone: "The old Rabbis of blessed memory have not only seen this bone, but have found
it actually so strong and hard that their hammer and rock flew in pieces before this bone was injured
in the least" (p. 242n).
13. Joseph Butler, The Analogy of Religion, Part I, Chap. 1 ("Of a Future Life"); See The Whole
Works of Joseph Butler. LL.D. (London: Thomas Tegg, 1839), p. 7.
14. St. Thomas Aquinas, On Spiritual Creatures: Article II (Milwaukee: Marquette University
Press, 1949), p. 35.
15. I have illustrated this confusion in Richard Taylor's Action and Purpose (Englewood Cliffs,
N.J., Prentice-Hall, 1966), p. 137.
16. Compare Franz Brentano, Religion Und Philosophie (Bern: A. Francke Verlag, 1954).
17. Strawson emphasizes that persons have both psychological and physical properties. But, if
what I say is true, most of the physical properties that we ordinarily attribute to the person are "borrowed" in this sense from the person's body.
14
The Primacy of the Intentional

According to the thesis of the primacy of the intentional, the reference of language is to be explicated in terms of the intentionality of thought. The word
"Pferd," for example, refers to horses insofar as it is used to express thoughts that
are directed on horses. But most contemporary philosophers of language, until
recently at least, have held that the intentionality of thought is to be explicated
in terms of the reference of language. But no such explication is at hand.
I will suggest what one can say about the reference of language if one presupposes the primacy of the intentional. I will not use any undefined semantical concepts, since all such concepts, I believe, can be explicated only by reference to
intentional concepts. Therefore, I will not speak of "inner systems of representation," "inner speech acts," or "inner language." But I will make use of a number
of intentional concepts. In terms of these and certain other familiar concepts, I
will formulate intentional definitions of such semantic concepts as sense and
reference. I will try to show how this approach will throw light upon a number
of philosophical questions (e.g., "How are we to interpret the 'he, himself locution?" and "Do demonstratives and proper names have senses?").
Language Intentionally Considered
Two assumptions underlie what might be called the orthodox approach to the
analysis of sense and reference. One of these is ontological and the other is psychological. Each seems to me questionable.
(1) The first assumption is this: For each use of any well-formed indicative
sentence in our language, there is a proposition which is the meaning of that sentence in that use.
I would say that this is not an assumption with which we should begin our investigations. It is, at best, a conclusion we might draw at the end of our investiga129
130
THE PRIMACY OF THE INTENTIONAL
tions. For the "propositions" that are thus presupposed will not be restricted to
the abstract objects now commonly called "states of affairs." Those "singular
propositions" that would constitute the meanings of sentences containing demonstratives and proper names would be contingent things, dependent for their being
on those individual things that are thought to "enter into them." And these singular
propositions are strange entities.
There is a certain plausibility in the assumption that there are what might be
called "concrete states or events." It is reasonable to assume, after all, that if a
contingent thing x has a certain property P, then there is also that concrete state
or event that is x having P. But concrete states or events will not do the work that
is required of singular propositions. For false or nonoccurent singular propositions would be needed to constitute the meanings of sentences that are false. If
you believe that there are 12 planets, then, even if there are only 9 planets, the
singular proposition theory requires the existence of that contingent thing which
is there being 12 planets (or that there are 12 planets).
(2) Since, presumably, we can believe only what we can grasp, or conceive,
the singular proposition theory presupposes further that the believer is able to
grasp or conceive not only those abstract objects which are properties and states
of affairs, but also those contingent "singular propositions" constituting the meanings of sentences containing demonstratives and proper names. It is presupposed
that demonstrative terms and proper names, like such definite descriptions as "the
tallest spy," have what might be called a Fregean sense. A Fregean sense is a
property that is such that a term having that sense designates a thing only if the
thing has that property. One then looks in vain for those properties constituting
the Fregean senses of demonstratives ("this," "I," "you," "now") and of proper
names ("Tom," "Cicero").
The conception of intentionality that I shall set forth is considerably simpler
than the foregoing and it has a number of advantages as a basis for understanding
language.
De Re Belief
I will assume that intentional attitudes involve a relation between a person and
a property or attribute rather than between a person and a. proposition. I shall set
forth this view in some detail in application to occurrent belief or believingand
somewhat more sketchily in connection with endeavoring and perceiving. What
is said about these attitudes may also be applied, mutatis mutandis, to thinking,
or considering.1
We begin, then, with believing.
I presuppose that believing is essentially a matter of believing certain properties ^directly of oneself." (We could also say: "attributing certain properties
directly to oneself.") The fundamental doxastic concept may be expressed by the
locution
131
jc believes directly ofy the property ofbeing-F

a locution that is taken to imply
x is identical with y
The letter "F" in "being-F" is schematic and may be replaced by any English
predicate-say, "wise." The result of such replacement will be the English term
"being-wise"a term designating a property.
What would it be, then, for one to have a belief about something other than
oneself? Suppose, for example, that I believe you to be wise. In this case, there
is a property I believe directly of me; this property relates me just to you; and
because of this relation, I can be said to believe being-wise indirectly of you. Such
a belief might come about under the following circumstances. You are the person
I am talking with; and the property of talking with just one person and with a person who is wise is a property that I attribute directly to myself.
We may be tempted, then, to characterize indirect attribution this way:
x believes being-F indirectly of y = Df There is a relation R such that (a) x
bears R only to y and (b) the property of bearing R to just one thing and to
a thing that is F is one that x believes directly of x
Then we could say that x believes y to be F provided only that x believes being-F
either directly or indirectly of y. But this preliminary account may be more
latitudinarian than it should be.
Consider the following example, adapted from one suggested by Keith Donnellan.2
(1)1 believe that the person I am looking at is a member of the Temperance
Union; (2) I believe, mistakenly, that the person I am looking at is the only one
at the party who is drinking a martini; (3) putting two and two together, I judge
that the only one at the party who is drinking a martini is a member of the Temperance Union; and (4) as it happens, the only one at the party who is drinking a martini is not the person I am looking at, but the person who is standing behind me.
Application of our formula above to this example would require us to say not
only that the person I'm looking at, but also that the person who is standing behind
me, is such that I believe that he is a member of the Temperance Union. My belief, so to speak, "points to" both individuals. But if it is a belief that is directed
on one of them, then it is not also a belief that is directed on the other.
If my belief is directed on a specific individual, say, you, then I have some
conception of just who you are. But in what sense of "conceiving who you are"?
First I will put the answer somewhat loosely. Then I will attempt a more precise formulation.
If my belief is directed on you, then there are two things to be said about the
justification that I have for that belief: (1) the property by means of which I relate
myself uniquely to you is one such that I am justified in believing that there is just
one thing having it; and (2) at the moment I have no better way of identifying you.
132
In putting the account of indirect attribution more exactly, I will use "x believes
himself to be F" as short for "the property of being F is such that x believes it
directly of x":
Dl x believes indirectly with respect to y that it is F = DF There is a relation
R such that:
(1) x bears R just to y;
(2) x judges himself to bear R to just one thing and to a thing that is F;
(3) x is more justified in judging himself to bear R to just one thing than
in not judging himself to bear to just one thing; and
(4) if x judges himself to bear R to the thing he bears S to, then he is
at least as justified in judging himself to bear R to just one thing as he is
in judging himself to bear S to just one thing
If, in judging that he is talking with a member of the Temperance Union, x has
a belief that is directed upon y, then: (1) x is more justified in believing that there
is one and only person he is talking with than in having such a belief; and (2) if
x also judges that the person he is talking with is, say, the Chairman of the Temperance Union, then he, x, is at least as justified in believing that there is just one
person he is talking with as in believing that there is just one person who is the
Chairman of the Temperance Union.
And now we may set forth our definition of de re belief:
D2 x believes with respect to y that it is F = Df The property of being F is
such that x believs it either directly or indirectly of y
On this account, the de dicto locution, "x accepts the proposition that p" could
be defined as: "There is a y such that x believes with respect to it that it is true."
We should note, in passing, that this epistemic conception of de re belief may
also be adapted to the prepositional conception of intentionality. One could say:
x believes with respect to y that it is F =DF
(1) x believes that the G is F;
(2) y is the G;
(3) x is more justified in believing that there is just one thing that is G than
in not believing that there is just one thing that is G; and
(4) if x believes that the G is the H, then x is at least as justified in believing
that there is just one thing that is G as he is in believing that there is just one
thing that is H
Object and Content
The distinction between object and content is essential to understanding language.
In saying something to you, my concern may be to get you to believe something.
Or it may be to get you to do something. Or it may be merely to get you to think
o/something. In each case, there is a distinction between object and content. Thus
133
there is the object, or there are the objects, that I want you to believe something
about, or to do something to, or to think of in a certain way. And there is what
it is that I want you to believe about the object, or what it is I want you to do to
the object, or what it is I want you to think of the object as being.
Suppose you are driving a car in which I am a passenger and I say to you urgently: "That green Chevrolet over there is out of control!" If you reply, "That's
not a Chevrolet," you may correct what I say, but you will have misinterpreted
my message. I might reply, if there is time, "But whatever it is it's out of control!"3
The words "That green Chevrolet is out of control" in my original utterance did
not express any part of the content I meant to convey. I used them only so that
you would pick out the object I want to convey something about. The content of
my message was expressed by the words "out of control."
Since, in believing, one directly attributes a property to oneself and in so doing
one may indirectly attribute another property to some other thing, we may distinguish between the direct and the indirect object and content of belief. And analogously for the other intentional attitudes. Thus, we may distinguish between the
object and the content of thinking of and between the object and content of endeavor.
Perceiving and Endeavoring
The concept of meaning to convey, which is central to understanding the intentionality of language, presupposes the further intentional concepts of perceiving
and endeavoring.
Perceiving is a paradigm case of what we have called "indirectly believing,"
or "indirect attribution." In perceiving the car, say, to be moving, one directly
attributes to oneself the property of being appeared to by just one thing and to
a thing that is a car and moving. The intentional element is perceiving is this:
D3 x perceptually takes y to be F = Df y and only y appears G to x; and x
directly believes of x the property of being appeared G to by just one thing
and by a thing that is F
The ordinary concept of perception includes the further concept of being evident:
D4 x perceives y to be F = DF x perceptually takes y to be F; and it is evident
to x that y is F
The intentional element in the concept of endeavor may be expressed by "The
property of being F is such that x endeavors to have it." In some contexts, "acts
with the intention of" will be used instead of "endeavors to." Indirect endeavor
is analogous to indirect belief. (Once again, I will use "x believes himself to be
F" as short for "x believes being-F directly of x.")
D5 x indirectly endeavors that y be F = Df There is a relation R such that:
(1) x will bear R just to y;
134
(2) the property of bearing R to just one thing and to a thing that is F
is such that x endeavors to have it;
(3) x is more justified in believing himself to be such that he will bear
R to just one thing than in not believing himself to be such that he will bear
R to just one thing; and
(4) if x believes himself to be such that the thing he will bear R to is
the thing he will bear S to, then he is at least as justified in believing himself to be such that he will bear R to just one thing as he is in believing
himself to be such that he will bear S to just one thing
Our definition of de re endeavor is now this:
D6 x endeavors that y be F = Df Either (a) x is identical with y and endeavors
to have the property being-F or (b) x indirectly endeavors that y be F
We need also the concept of endeavoring to do one thing for the purpose of
bringing about another thing (or in order to bring about another thing). For present purposes this simplification will do:
D7 x endeavors to bring about P and does so/or the purpose of bringing about
q = Df x endeavors to bring it about that his endeavor to bring about P
cause Q
This concept thus presupposes the concept of causation.
Meaning to Convey
The relation between thought and language may now be described by reference
to the concept of meaning to convey. Meaning to convey is endeavoring to convey. What, then, is conveying! If I want to convey something to you, then I have
a certain thought I want to communicate to you; and this means, in part, that there
is something I want to cause you to think of. (I here take "cause to think of"
broadly-to cover both the case where one is caused to begin to think of a certain
thing and also the case where one is caused to continue to think of that thing.)
But conveying is more than merely causing to think of. Let us consider two
cases of causing to think of that are not cases of conveying.
(1)1 inject you with a certain drug that makes people paranoiac and then I present you to Mr. Jones. The result is that you believe that Mr. Jones desires to persecute you. But even if it had been my intention to cause you to believe this, we
cannot say, taking "convey" in its present sense, that I have conveyed this to you.
(2) Kant cites the following case of intended deception that is not a case of lying: "I may wish people to think that I am off on a journey, and so I pack my lug
gage; people draw the conclusion I want them to draw. . . . I have not lied to
them, for I have not stated that I am expressing my opinion." 4
Kant's remark may suggest that, in order to be able to tell you anything, I must
first tell you that I am going to tell you something, and in order to to be able to
135
convey anything, I must first convey that I'm going to convey something. This
type of regress would hardly be acceptable.
What, then, does conveying involve that mere causing to think o/does not involve? I suggest that there are three marks that are characteristic of conveying
but need not hold of mere causing to think of.
One mark of conveying may be illustrated by this: If I convey something to
you, I do so by causing you to believe that /am thinking of that something. This
is not what happens when, merely by injecting a drug, I cause you to have a certain belief.
Second, if I convey something to you, then my purpose in causing you to believe that I am thinking of the thing in question is that of causing you to think of
that thing.
And a third mark of conveying pertains to the fact that one is addressing someone. If I mean to convey to you the thought that so-and-so, then I intend to cause
you to believe that I intend to cause you to have the thought that so-and-so. Hence,
conveying presupposes a complex belief situation. We must be able to say such
things as: "x endeavors to cause y to believe that he, x, believes with respect to
z that it is F."
Meaning to convey, then, presupposes the following concepts: making an utterance; causal contribution; perception; endeavor: thinking of; and belief. I will
now say what it is to address an utterance for the purpose of conveying something. For simplicity, I will restrict myself to the situation wherein the speaker
is addressing only one person; and I will use abbreviations introduced by previous
definitions.
D8 x addresses an utterance to z to convey the thought to z that y if F = Df
(1) x makes an utterance so that z will perceive that utterance and thereby
believe that x thinks of y as F; (2) x does this to cause z to believe that
he, x, intends to cause z to think of y as F; and (3) his purpose in doing
this is to cause z to think of y as F
When the conditions of the above definition are fulfilled, we may say that y
is the object concerning which x means to convey something, and that the property of being F is the content of what it is that x means to convey with respect to y.
A broader concept may be obtained by revising the first clause of the definiens.
Instead of saying that x intends to cause z to "believe that he, x, thinks of y as
being F," we could say that x intends to cause z to "believe that there is a y such
that he, x, thinks of y as being F." This would accommodate the situation (for
example, a hallucination) wherein the belief has no intentional object.
Sense and Reference

If a person x believes, with respect to a thing y, that y has a certain property P,
then y is the object of the belief in question and the property P is the content. In
terms of this intentional distinction between content and object, we may now spell
136
out the linguistic distinction between sense and reference.5 The analysis of sense
makes use of the concept of the content of thought, and the analysis of reference,
or designation, makes use of the concept of the object of thought.
I will use the expression "attributive sense" in order to distinguish the concept
in question from the kind of sense that is sometimes attributed to proper names
and demonstratives.
We begin with the "speaker's sense" of a predicate:
D9 x uses P with the attributive sense S = DF x addresses an utterance for
the purpose of conveying something, and P is that part of x's utterance
which is intended to bring it about that S is the content of the thought he
thus endeavors to cause
"Part" is here to be read as "proper part"; hence, if P is "that part of the utterance"
which is intended to bring about so-and-so, then there will be another part of the
utterance, discrete from P, which is not intended to bring about so-and-so. Hence,
P cannot be identified with the entire utterance.
The concept of the "hearer's sense" is this:
D10 z interprets x's use of P as having the attributive sense S = Df z perceives
that part of x's utterance which is P and believes of it that x meant to use
it with the attributive sense S
I have spoken of the sense of predicative expressions those utterances that
are used to convey the content of one's thought. May we also ascribe a sense to
such designative expressions as demonstratives and proper names?
We could ascribe a sense to proper names and demonstratives. For example,
if I use a proper name in speaking to another person, then the demonstrative sense
of that name on that occasion could be said to pertain to the relation or relations
by means of which I then single out the objector objectsof the belief I am expressing to the other person. Any such identifying relation will be a relation such
that the bearer of the name is the thing to which the user of the name bears that
relation. Consider again my statement "That green car is out of control." One
could say that the demonstrative sense of my expression "that green car" might
be the property expressed by "the green car I have just been watching." In this
case, it would be the relational property of being someone who has just been
watching one and only one green automobile.
The demonstrative sense of a name, then, would not be a property of the bearer
of the name. It would be, rather, a relational property that the user of the name
attributes to himself. The property, therefore, would be a property of the user
of the nameprovided there is a bearer of the name. And the corresponding relation would be one that the user of the name bears only to the bearer of the name.
But it is not clear that anything is to be gained by the introduction of this concept of demonstrative sense. For we do not need to use it in explicating the designative function of proper names and demonstratives.
137
What, then, of the concept of designation?

We first consider what may be called the "speaker's designation" of a word:
Dll x uses N to designate y = Df x makes an utterance for the purpose of
thereby conveying something about y; and N is that part of x's utterance
which is intended to bring it about that y is the object of the thought that
x thus endeavors to cause
We have distinguished the case where one's thought has an object from the case
where one's thought only purports to have an object. It was suggested that the latter situation may arise when I believe that there is one and only one person who
is a devil and that that person is following me. Now it might be that I use the name
"Satan" for this person. But we cannot say that I use "Satan" to designate him,
for the person doesn't exist. Let us say that, in such a case, I "mean to use 'Satan'
to designate something." The relevant concept is this:
D12 x means to use N to designate something = Df x makes an utterance for
the purpose of thereby conveying something about a certain thing; and
N is that part of x's utterance which is intended to bring it about that the
thing, with respect to which he endeavors to convey this something, is
the object of the thought he thus endeavors to convey
The concept defined in Dll implies that defined in D12, but not conversely.
We now turn to what might be called the "hearer's designation" of a word:
D13 z interprets x's use of N as designating y = Df z perceives that part of
x's utterance which is N and believes of it that x meant to use it to designate y
Or z may simply interpret x's use of N as purporting to designate something. In
such a case, z does not believe with respect to a thing y that x meant to use N
to designate y; z believes only that x meant to use N to designate something.
Looking back to our account of designation, one may ask, "How does it happen
that x's utterance of N can bring about the desired effect?" If x's utterance is successful, then there will be someone z who is caused to perceive N and in consequence to think of y. There is, therefore, a causal factor that is involved: x's utterance of N causes z to think of y. Hence there can be a "causal theory of meaning":
that is to say, there can be a causal explanation of the fact that a person's perception of an utterance N, or part of such an utterance, causes him to think of a certain
thing y. But this is not a causal theory of what it is for one to think of another
thingmuch less a causal theory of what it is for one thing to designate another
thing.
I have, then, attempted to single out certain intentional concepts in terms of
which we may define sense and reference. We must distinguish (1) the sense in
which we can say of a certain word in English that it has a fixed sense and that,
at any time, it can be used in English to designate only one specific type of object
at that time; and (2) the sense in which we can of a certain word, say "John," that
138
it is now used to designate John and now used to designate Mary. We have been
concerned here with (2), with "speaker's meaning," and not with (1), not with "linguistic meaning." Linguistic meaning, so conceived, should be thought of as
idealized speaker's meaning.
In this way, then, I would defend the principle of the primacy of the intentional.
Notes
1.1 have defended a version of this view in The First Person: An Essay on Reference and Intentionality (Brighton and Minneapolis: Harvester Press and University of Minnesota Press, 1981).
2. This example is adapted from one suggested by Keith Donnellan in "Reference and Definite
Descriptions," Philosophical Review, 75 (1966): 281-304.
3. Compare A. N. Whitehead's example of "That college building is commodius," in The Concept of Nature (Cambridge: The University Press, 1930), pp. 6-7.
4. Lecture on Ethics (New York: Harper & Row, 1963), p. 226. The German reads:
" . . . dann habe ich ihn nicht belogen, denn ich habe nicht deklariert, meine Gesinnung zu a'ussern." See Paul Menzer (ed.), Eine Vorlesung Kants uber Ethik (Berlin: Rolf Heise, 1925), p. 286.
Compare Roderick M. Chisholm and Thomas D. Feehan, "The Intent to Deceive," Journal of Philosophy, 74 (1977): 143-60.
5. Compare the distinction between "referential" and "attributive" in Keith Donnellan, "Reference and Definite Descriptions."
Part V
An Intentional Approach to Ontology
15
Properties and States of Affairs

Intentionally Considered
Platonism
Are there such things as properties? There are many things that we seem to know
about properties.
Some properties (for example, the property of being a horse) are exemplified;
and some of them (for example, the property of being a unicorn) are not exemplified. Some of them (for example, the property of being both round and square)
cannot be exemplified by anything. And some of them (for example, the property
of being self-identical) must be exemplified by everything. There are properties
P and Q that are necessarily so related that if P is exemplified, then Q is also exemplified. There are properties P and Q that are necessarily so related that if P is
exemplified, then Q is not exemplified. Some properties P and Q are are so related
that one cannot conceive P without also conceiving Q. And there are many other
such a priori truths.
Is this knowledge really just a kind of knowledge about individual things and,
therefore, strictly speaking, not about properties at all? And how are we to decide? If our ostensible knowledge of properties were just a kind of knowledge
about individual things, then we could paraphrase our ostensible property statements into statements in which the individual terms and variables can be interpreted as taking only individual things as their values.
Certainly some of the statements expressing what we know about properties
can be thus paraphrased. For example, "The property of being a horse is exemplified" can be put as "There exists an x such that x is a horse"; and "The property
of being a unicorn is not exemplified" can be put as "It is false that there exists
an x such that x is a unicorn."
But such paraphrase is not possible in every case. Consider the following statements: "There are shapes that are not exemplified"; "Certain pairs of properties
141
142
PROPERTIES AND STATES OF AFFAIRS
are necessarily such that if one of the members is exemplified then the other is
also exemplified"; and "Certain pairs of properties are necessarily such that it is
impossible to conceive one of them without also conceiving the other." Such statements as these, so far as I can see, cannot be paraphrased into statements referring
only to individuals.
This fact, I would say, constitutes at least a prima facie justification for accepting extreme realism: there are properties, some of which are exemplified and
some of which are not exemplified.
The Question of the Structure of Properties
Properties have an intentional structure. Some are compound and some are noncompound. And of those that are compound, some are conjuctive, some are disjunctive, and some are negative. Thus red and round is compound and conjunctive; red or round is compound and disjunctive; and nonred is compound and
negative.
Such a conception readily provokes the following objection. "You are mislead
by the language we use to talk about properties. If a propertysay, what you call
'red or round'can be said to be 'disjunctive', then this simply reflects an incidental linguistic fact, the fact that the property can be expressed by means of a disjunctive predicate (a predicate made up of two noncompound predicates with a
disjunction sign between them). And analogously for 'conjunctive', 'negative',
and 'compound'. What you call the 'structure of properties' is nothing more than
the structure of the language we use in formulating our predicates."
More particularly, one may raise such questions as these:
Is brother a noncompound propertyor is it the conjunctive property of being male and a sibling?
What of the property of being a sibling? Is it a noncompound propertyor
is it the disjunctive property of being either a brother or a sister?
Is being human the disjunctive property of being either a man or a woman?
Or is being a man the conjunctive property of being male and human, and is
being a woman the conjunctive property of being female and human?
Is being mortal negative or affirmative? What, then, of being immortal?
You say that red is noncompound. But isn't it the same as the conjunctive
property, red and redl And isn't it also the same as the disjunctive property,
red or red, not to mention the disjunctive property of being either red and
round or red and nonrouruft
One may be tempted to conclude, therefore, that the so-called structure of properties is simply a reflection of the kind of language we happen to use in expressing
our thoughts about properties.
But let us look more closely at the nature of properties.
143
Basic Relations between Properties

We will first note certain ways in which properties may be related to each other.
(A) Implication:
One property may be said, in the following sense, to imply another;
D I P implies Q = Df P is necessarily such that if it is exemplified then Q is
exemplified
The property being a wife may be said, in this sense, to imply the property being
a husband, and conversely. Each of the two properties is necessarily such that,
if it is exemplified, then the other is exemplified. But the thing that exemplifies
the one is not identical with the thing that exemplifies the other. Hence, we may
single out a subspecies of property implication.
(B) Inclusion:
D2 P includes Q = Df P is necessarily such that whatever exemplifies it exemplifies Q
The property of being a dog includes that of being an animal: a thing cannot have
the first property without also having the second. If one property includes another, then the first property also implies the second; but, as we have just seen,
one property may imply another without including the other.
(C) Involvement:
We next introduce an important intentional relation that obtains between various properties: one property may be said to involve another.
Consider these four properties: (i) being either red or round; (ii) being nonred;
(iii) being possibly red; and (iv) wanting something that is red. All these properties are intimately related to the property being red, yet they neither include nor
imply it. They involve the property red in the following sense: each is necessarily
such that it is impossible for one to conceive it without conceiving the property
red. We will say, then:
D3 P involves Q = Df P is necessarily such that whoever conceives it conceives Q
The property of being both red and round involves the property of being round
(for one cannot conceive it without conceiving the property of being round). And
it also implies the property of being round. The property of being either red or
round also involves the property of being round, but it does not imply the property
of being round. So, too, for the property of being nonround: it involves but does
not imply the property of being round. Again, the intentional property of wanting
something that is red involves but does not imply the property of being red. Further, the dispositional property of being potentially red involves but does not imply the property of being red. And, finally, the causal property, preventing something from becoming red, involves but does not imply the property of being red.
Thus many properties involve properties they do not imply.
144
And many properties imply properties they do not involve. Being red, for example, impliesbut does not involvebeing either red or round; it also implies
but does not involve the property of being either red and round or red and nonround, as well as that of being self-identical and that of being such that 7 and 5
are 12.
(D) Entailmenf.
We define the relation of property entailment'm terms of attribution and
necessity:
D4 P entails Q = Df P is necessarily such that, for every x and every y, if
y attributes P to x, then y attributes Q to x
We assume that, if P entails Q, then P includes and involves Q. And we have
noted that, if P includes Q, then P implies Q.
But it is possible for a property P both to include and to involve a property
Q without thereby entailing Q (in the present sense of "entail"). The property of
being either both round and square or such that 7 and 5 are 13 includes and involves the property of being both round and square, but it does not entail that
property. In other words: one cannot conceive the first property without conceiving the second; but one can attribute the first property to a thing without attributing the second property to anything. Again, the property being red and either not
colored or round implies and involves the property of being round, but it does
not entail that property.1
One property thus entails another only if the attribution of the first property
implies the attribution of the second.
Identity Criteria for Properties

If, in contemporary philosophy, one discusses a type of entity that is not an individual thing, one is confronted with the question: "What criteria of identity are
there for the type of entity you are discussing?"
Let us consider four possible criteria of property identity.
(1) The first is this: A property P is identical with a property Q if and only
if whatever has P has Q and conversely. Adoption of this criterion would have
the consequence that the property of being a rational animal is identical with that
of being a featherless biped, and that the property of being a unicorn is identical
with that of being a mermaid. It would also have the consequence that the property of being such that all men are mortal is identical with the property of being
such that 7 and 5 are 12.
(2) A second possible criterion of property identity is this: P is identical with
Q if and only if each of P and Q is necessarily such that whatever has it has the
other. Or, in other words, P is identical with Q if and only if P and Q are logically
equivalent. Adoption of this criterion would have the consequence that the property being an equilateral triangle is the same as the property of being an equiangular triangle. It would also have the consequence that the property of being red is
145
identical with the property of being either red and spherical or red and nonspherical.
(3) A third possible criterion of property identity is intentional: P is identical
with Q if and only if P and Q both imply and involve each other. In such a case,
they are necessarily such that if the one is exemplified then the other is exemplified, and also necessarily such that whoever conceives the one conceives the
other. But this would require us to say that the property of striking is identical
with the property of being struck and that the property of killing is identical with
the property of being killed.
(4) Hence a more satisfactory intentional criterion would be one using mutual
inclusion instead of mutual implication: P is identical with Q if and only if P and
Q include and involve each other.
We will adopt this final intentional criterion of property identity. Consider
now the following objection to our criterion.
"Surely the statement 'Red is the color of ripe strawberries' is a paradigmatic
contingent statement of property identity. And yet it is one thing to think of a
thing as being red and it is another thing to think of a thing as having the color
of ripe strawberries. A person who does not know what color ripe strawberries
have may be able to conceive the property red-but will not be able to conceive
the color of ripe strawberries. Therefore your criterion of property identity is
too strong."
We should distinguish (A) that property which, as it happens, is the color
property that is exemplified by all ripe strawberies and (B) that property which
is being that color property that is exemplified by all ripe strawberries. Property
(A) is the property red; property (B) is not the property red, but a property that,
as it happens, is exemplified by the property red. It is true that a person cannot
conceive the property red unless he can conceive (A) that property which is the
color property exemplified by all ripe strawberries. But the above objection
presupposes, mistakenly, that a person cannot conceive the property red unless
he can conceive (B) that property which is being the color property that is exemplified by all ripe strawberries. One can conceive the color property exemplified
by all ripe strawberries without having any conception at all of a ripe
strawberry.
It is one thing, then, to know that something has the color red, but it is another
thing to know that something has the color of ripe strawberries. If you know that
a thing has the color red, then you may deduce that the thing has a color that is
necessarily such that, for every possible world, if a thing has that color in that
world, then it is red in that world. But if you know that the thing has the color
of ripe strawberries, then you may not deduce that the thing has a color that is
necessarily such that for every possible world, if a thing has that color in that
world, then it has the color of ripe strawberries in that world.2
I turn now to the intentional structure of properties.
146
Conjunctive and Disjunctive Properties

"Being red and being red" is a conjunctive phrase. But there is no point in saying
that, in addition to the property being red, there is also the conjunctive property,
being red and being red. We cannot distinguish conceiving being red from conceiving being red and being red. And, similarly, there is no point in saying that,
in addition to the conjunctive property, being red and round, there is also the conjunctive property, being round and red; nor is there reason to say that there is
also the property being round and red and round.
We have said that a property P entails a property Q, provided only that for
every x and y, if x attributes P to y, then x attributes Q to y.
I will say that a conjunctive property is a property that entails two properties
and that is implied by everything that implies those two properties. (Since we
speak of two properties, neither will entail the other.) The two properties are the
conjuncts of the conjunction.
The definition is:
D5 C is the conjunction of P and Q = Df (1) C entails P; (2) C entails Q; (3)
P does not entail Q; (4) Q does not entail P; and (5) everything entailed
by C entails something that either P or Q entails
The property red and round will thus be a conjunction of the properties red and
round. But it will not be a conjunction of red and both round and self-identical,
for the property red does not entail the latter property.
I assume that for any two properties not related by entailment, there is that
property which is their conjunction.
A disjunctive property is a property that is related in the following way to two
other properties: it involves them but does not entail them, and it is necessarily
such that it is exemplified by a thing only if the thing has one or the other of the
two properties. The two properties are its disjuncts.
D6 D is a disjunction of G and H = Df (i) D involves but does not entail G,
and D involves but does not entail H; (ii) G does not entail H, and H does
not entail G; and (iii) D is necessarily such that, for every x, x has D, if
and only if either x has G or x has H
The reference to entailment in this definition has as a consequence that such expressions as the following do not connote disjunctive properties: "either red or
both red and round."
I assume that for any two properties that are not related by entailment there
is that property which is their disjunction.
Negations of Properties
We turn now to the following two concepts: that of the negation of a property
and also that of a negative property. The definitions I shall propose have the fol-
147
lowing two consequences: (1) the property of being red has as its (only) negation
the property of being nonred; and (2) the property of being nonred is a negative
property and the property of being red is not a negative property.
Let us first mark off the notion of property contradiction:
D7 P contradicts Q = Df P is necessarily such that, for every x, x exemplifies
P if and only if x does not exemplify Q
Red thus stands in the relation of contradiction with its negation, nonred. But it
also stands in the relation of contradiction with many properties other than its negation; these are such complex properties as nonred and self-identical, and either
(nonred and round) or (nonred and not round). Given our criterion of property
identity, we cannot say that these more complex properties are identical with the
property nonred.
What, then, is distinctive about the relation between nonred and real Not only
does nonred contradict red; it also "properly involves" itthat is to say, nonred
involves red, but red does not involve nonred. Moreover, nonred, unlike the
complex properties cited above, does not properly involve anything that both contradicts and properly involves red. Let us use these marks to define a negative
contradictory of a property:
D8 P is the negation of Q = Df (1) P contradicts Q; (2) P involves Q; and
(3) Q does not involve P
We may now say that a negative property is a property that is the negation of
a property.
We may note still another mark of negative properties: A negative property
does not entail any property other than itself. And if a negative property (say,
nonred) is one such that its regation (red) is not necessary to anything, then the
negative property is possibly such that everything has it.
Negative properties are sometimes referred to as infinite propertiesthe
thought being, I think, that a negative property is a property that is possibly such
that everything has it. This is true of nonred, but it is not true ofnonself-identical,
which would be negative according to our definition.
Earlier Questions about Structure

We now return to those questions which expressed doubts about the concept of
the intentional structure of properties. The answers are clear.
Is the property of being a brother a conjunctive property? It is. One of its conjuncts is being male; the other conjunct is having the same parents that someone
else has.
Is the property of being a sibling a disjunction property? No. It is identical with
that of having the same parents that someone else has. Hence, it is noncompound
and therefore not a disjunctive property.
The property of being a man is the conjunctive property of being a male and
148
being human. Being a sibling, therefore, is noncompound and, hence, not a disjunctive property.
The property of being mortal is the property of being such that one is going
to die. And the property of being immortal is the conjunction of being alive and
not being mortal. Hence, being mortal is a noncompound affirmative property;
and being immortal is a conjunction of a noncomound affirmative property and
a noncompound negative property.
States of Affairs
The term "states of affairs" in one of its various uses refers to a type of abstract
object that is at least analogous in many respects to properties. Properties may
be divided into those that are exemplified and those that are not exemplified. There
are, therefore, properties that are not exemplified. And states of affairs may be
divided, analogously, into those that obtain and those that do not obtain. There
are, therefore, states of affairs that do not obtain.
Is it possible, then, that states of affairs, so conceived, constitute a subspecies
of property? We might be tempted to say that the state of affairs that-p is to be
identified with the property ofbeing-such-that-p. But is there such a property as
"being such that p"? What we should say is, rather, this:
D9 That-p is a state of affairs = Df There is a property which is necessarily
such that it is exemplified if and only if p
This definition guarantees that states of affairs are abstract things and not contingent events. For properties are necessary things and therefore if a property can
be said to be "necessarily such that p," then the sentence replacing "p" in the expression "that-p is a state of affairs" cannot express a contingent event.
Since states of affairs are thus reducible to properties, the expression "the state
of affairs that-p obtains" is reducible to a statement about exemplification:
DIG The state of affairs that-p obtains = Df Something has a property which
is necessarily such that it is exemplified only if p
What of propositions'? If we use "proposition" to refer to a type of abstract object and not to a type of contingent thing (such as those "singular propositions"
that are thought to contain contingent things as their constituents), then there
would seem to be no ground for distinguishing propositions from states of
affairs-unless we say that propositions are those states of affairs which are necessarily such that either they are always exemplified or they are never exemplified.
And the concept of the truth of a proposition would be explicated by reference
to exemplification in the way in which obtaining is explicated above.
And so our analyses of the intentional structures of properties may also be interpreted as analyses of the intentional structures of states of affairs or abstract
propositions. The schematic letters in the definitions of property conjunction
149
(D5), property disjunction (D6), and property negation (D7) may be replaced by
terms for states of affairs.
Frege raises the question whether there can be a distinction between affirmative and negative states of affairs (in his terminology, affirmative and negative
thoughts) and he concludes: "For logic at any rate such a distinction is wholly
necessary: its ground must be sought outside logic. I know of no logical principle
where verbal expression makes it necessary to use these terms."3 Yet there is a
valid distinction to be made. If we assume that every state of affairs has one and
only one negation, then we may say this: a state of affairs is affirmative if and only
if it does not involve its own negation; otherwise it is negative. Hence, for every
state of affairs and its negation, one will be negative and the other affirmative.
Notes
1. Compare the distinction between the "entertainment" and the "doxastic" senses of prepositional entailment in Roderick M. Chisholm, "Events, Propositons, and States of Affairs," in P. Weingartner and E. Morscher (eds.), Ontologie und Logik (Berlin: Duncker und Humblot, 1979), pp.
27-47; see esp. p. 31.
2. Compare Wittgenstein's observation: "It is easy to see that not all colour concepts are logically
of the same sort, e.g., the difference between the concepts 'colour of gold' or 'colour of silver' and
'yellow' or 'grey.' " Remarks on Colour (Berkeley: University of California Press, 1978), p. 9e.
3. Gottlob Frege, in Peter Geach and Max Black (eds.), Translations from the Philosophical
Writings of Gottlob Frege (Oxford: Basil Blackwell, 1952). The quotation is from p. 125 of "Negation."
16
States and Events

Aristotle and Bolzano had said that some things are beings of other things.1 We
will say that if one thing is a being of another thing, then the first thing is a state
o/the second. And we will take "x is a state of y" as undefined. In this way, we
will be able to set forth an explication of events and to show how events are related
to attributes and to individual things.
The Logical Properties of States

Our assumption that there are states of things may be put this way:
Al For every x, there is the state x-being-F if and only if x exemplifies
being-F
We thus use "being-F" as a term to designate attributes and "x-being-F" as a term
to designate states. The letter "F" may be replaced by any well-formed English
predicate; and "x" in "x-being-F" designates the entity having the state that is
designated by "x-being-F."
As a being of another thing, a state is not an ens per se. We may express this
fact by saying that states are ontologically dependent on the things of which they
are states:
A2 For every x and y, if x is a state of y, then x is necessarily such that it
is a state of y
Even the contingent states of a thing are necessarily such that they are states of
that thing. A person who is sad is not necessarily such that he is sad; but his-beingsad is necessarily such that it is a state of him.
Those contingent things that are not states of other things may be called individual things.
150
STATES AND EVENTS
151
"Particular States"
A thing may be said to enter into temporal or causal relations via its states. For
example, we may say of a person that his falling contributes causally to his being
injured. The cause of the injury was a "particular fall" that the person had and
the effect was a "particular injury." How are we to describe this situation? It was
not just the attribute of falling that contributed to the injury; it was a certain particular fall and a certain particular injurya fall that can be individuated and an injury that can be individuated.
Must we distinguish, then, as Stout did, between universals as abstract objects
and universals as particulars? Nofor, making use of the concept of a state and
expressing ourselves in a language that is tensed, we may individuate the fall and
the injury in another way.
I will not assume, as many contemporary philosophers do, that there are such
things as "times." And, therefore, I will not assume that there are things designated by those linguistic expressions that aredatesfor example, "June 3,1988."
I do assume, of course, that things persist: some things had attributes that they
no longer have; and some things will have attributes that they do not have. And
from this it follows that some states are temporally preceded by other states and
that some states are temporally followed by other states.
I will now describe a way of individuating states without presupposing the concept of time. I will single out certain attributes that may be called "temporally
denumerable."
A temporally denumerable attribute has two distinct features: (1) it is an attribute that is necessarily such that whatever has it has it only once; yet (2) it has
a content that is repeatable. Hence, we must make clear what is here intended by
the expressions "only once" and "repeatable content."
We presuppose that there are things that have come into being and that there
are things that will pass away. In "Coming into Being and Passing Away," we discussed what it is for a thing to come into being and for a thing to pass away. Using
tense, we said that a thing is coming into being provided only it is such that there
is nothing it did exemplify; and we said that a thing has just passed away provided
that something that was such that the thing exists is beginning to be such that it
does not exist. We add, then, this principle:
A3 There exists an x which is such that either x is coming into being or did
come into being; and there exists or will exist a y which is or will be such
that x has just passed away
Now we may give a sense to such expressions as "the first time," "the second
time," and "the n-th time," and do so without reference to times.
Dl x is F for the first time = Df (1) x is F; and (2) x is not such that it was
both non-F and such that it had been F
152
STATES AND EVENTS
x is F for at least the n-th time = Df (1) x is F; and (2) x was both non-F
and such that it had been F for the (n-1) time
x is F for not more than the n-th time = Df (1) x is F; and (2) it is false
that x was both non-F and such that it had been F for at least the n-th time
x is F for the n-th time = Df (1) x is F for at least the n-th time; and (2)
x is F for not more than the n-th time
And analogously for "the last time," "at least one more time after this," "at most
one more time after this," and so on.
Now we may say what a "temporally denumerable" attribute is:
D2 P is a temporally denumerable attribute of x = Df x has P; there is a finite
number n and an attribute Q which are necessarily such that: for every y,
y has P if and only if, either y has Q for the n-th time or y will have Q
n more times
The attribute Q may be said to be the "repeatable content" of the attribute P.
For example, if I am lecturing for the 700th time, then the attribute of lecturing
for the 700th time is a temporally denumerable attribute that has as its content
the repeatable attribute of lecturing. Or if I will lecture just 37 times after this,
then the attribute of being as to give that lecture which is my 38th lecture before
my final one is a temporally denumerable attribute that has the attribute of lecturing as its repeatable content.
Definition of Event
We have assumed that for every x, x is F if and only if there is the attribute beingF and there is the state x-being-F. Now we may say what it is for a state to be
an event.
D3 x-being-F is an event = Df (1) x is F; and (2) the attribute being-F is a
temporally denumerable attribute that only individuals can have and that
nothing has necessarily
One may object: "Your definition is adequate to those events involving just one
thing. But what of events involving a great multiplicity of thingsevents such as
hurricanes, wars, and revolutions?" If we think of aggregates of individuals as
being themselves individuals, then we may say that such events are states of aggregates of individuals.
I make use, therefore, of the concept "x is part of y," and add the following
principle, which has the same content as Lesniewski's principle concerning
"sums" of individuals:
A5 If x and y are contingent individuals and have no parts in common, then
there is a contingent individual composed of x and y
This tells us, in effect, that heaps, aggregates, or sums of individuals are them-
STATES AND EVENTS
153
selves individuals. Hurricanes, wars, and revolutions may thus be thought of as

states of sums of individuals.2
Recurrence
Events are entities that are contingent, for they are terms of the relation of causation and terms of such temporal relations as before and after. Yet the theory of
causation and the theory of probability require us to say that events may recur.
What would it mean to say, of a contingent thing, that it may recur?
The question is especially difficult if we accept Locke's dictum according to
which nothing can "have two beginnings of existence." Recurrence, one might
suppose, is a property of attributes; attributes, we have said, are noncontingent
things.
What, then, of recurrence! I suggest the following:
D4 x-being-F is recurring = Df x is F; and x was such that it was both non-F
and such as to have been F
Strictly speaking, therefore, what recurs are not events but rather those attributes
that constitute the repeatable contents of events.
Negative Attributes and Negative States

Given our assumption that there are states, we need not also assume that there
are "facts." For states perform all the theoretical functions commonly assigned
to facts. Hence, we do not have the problem of explicating "negative facts."
But there are negative attributes for example, the attribute nongreen. And
therefore there are negative statesfor example, this piece of paper being nongreen. Every negative state is a state of some entity and, like every other state,
presupposes an entity that is not a state.
A negative state, then, has as its content a negative attribute. And what is a
negative attribute? I am convinced that, to answer such a question, we must consider attributes from an intentional point of view.
Viewing attributes intentionally, we may say that the defining mark of an attribute is this: it is a thing that is possibly such that there is someone who attributes
it to something. And we should note that attributes may be related in several
different ways.
We may say that an attribute implies another attribute if the first attribute is
necessarily such that if anything has it, then something has the second attribute.
And we may say that an attribute includes another attribute if the first attribute
is necessarily such that whatever has it also has the second.
Let us speak again of the "involvement" of attributes. Consider the four attributes: (i) being either red or round; (ii) being nonred; (iii) being possibly red;
and (iv) wanting something that is red. All these attributes are intimately related
to the attribute being red, yet they neither include nor imply it. Let us say that
154
STATES AND EVENTS
they involve the attribute red and try to say what involvement is. Once again we
find the general idea suggested by Bolzano. He cites these examples: the concept
of a land without mountains and that of a book without engravings. Bolzano does
not use the term "involves" but puts the relationship by saying that "the parts of
the idea [die Teile der Vorstellung]" need not be "parts of the object [Teile des
Gegenstandes]."3
The attribute of being a land without mountains may be said, in the following
sense, to involve the attribute being a mountain: it is impossible for anyone to conceive it without also conceiving the attribute being a mountain. And the four attributes just cited, which involve the attribute being red, do so in the following
sense: each is such that it is impossible for anyone to conceive it without also conceiving the attribute red.
Now we are in a position to distinguish between attributes that are positive and
attributes that are negative. The distinction between positive and negative is not
a function of the fact that we use negative expressions such as "non" or "not" in
connection with just one of the two expressions; it is not a linguistic distinction
at all. It has to do, rather, with the structure or inner nature of the attributes themselves.
Let us distinguish attributes that exclude each other from attributes that contradict each other. Red may be said to exclude yellow in that it is impossible for
anything to be both red and yellow. But it is possible for a thing to be neither red
nor yellow. Red contradicts nonred in that it is necessary that everything is either
red or nonred and it is impossible that anything is both. Hence, if two attributes
contradict each other, then they also exclude each other; but they may exclude
each other without contradicting each other.
Now we may say what it is for an attribute to be a negative attribute.
D5 Being-F is a negative attribute = Df One cannot conceive an attribute that
excludes being-F without conceiving an attribute that contradicts being-F
In other words, if being-F is a negative attribute, then a contradictory of being-F
is involved in every attribute that excludes being-F. Red is involved in every attribute that excludes nonred; but nonred is not involved in every attribute that excludes red. (Yellow excludes being red, and a person who cannot conceive red
may be able to conceive yellow.}
Are There Negative Events?

The expression "negative event" may be taken in two ways. (1) It may be taken
to refer to an event "that does not occur"; or (2) it may be taken to refer to an
event having a negative content. If we take it in the first way, we need not say
that there are negative events. But if we take it in the second way, we should say
that there are negative events.
(1) Why would one think that there are "events that do not occur"? The reason
for thinking this may be suggested by the following argument:
STATES AND EVENTS
155
(1) The dam in the river prevented a serious flood

Therefore
(2) There was something that the dam prevented
Hence
(3) This something was an event
But
(4) That event did not occur
Therefore
(5) There was an event that did not occur
Step (2) of the argument is unjustified. For (1) does not tell us that there is, or
was, something that the dam prevented. It tells us only that the dam caused that
negative state which was the river not being such as to flood. This negative state
is not "an event that does not occur"; it is an event with a negative content.
(2) We should say, then, that there are events having negative contents. There
is, for example, that event which is this piece of paper having the negative attribute of nonblue.
"But do you need to speak of such events? This paper would not have the negative attribute nonblue unless it had some positive attributein the present case,
the attribute of being whitethat logically excludes the negative attribute.
Doesn't the positive eventthis piece of paper being whiteserve all the theoretical purposes of the negative event?"
The objection presupposes that, if an individual thing has a negative attribute,
then it has some positive attribute that logically excludes the contradictory of the
negative attribute. But this presupposition is false. This may be seen most clearly
in the case of psychological events. A person may have the negative attribute of
not believing that it is raining in Graz; and the person may have that negative attribute without thereby having some positive attribute that excludes the contradictory of the negative attribute. Indeed, being such as not to be thinking is not a
state that one has in virtue of any positive state that logically excludes the attribute
of thinking.
Notes
1. "Every thing that there is is of one or the other of the following two types: either it is an entity
which is of another thing [an etwas Anderem] or it exists, as one is accustomed to saying, in itself
\fiir sich]." Bernard Bolzano, Athanasia oder Griindeflir die Unsterblichkeit der Seele (Sulzbach: J.
G. v. Seidelschen Buchhandlung, 1838), p. 21.
2. The present approach to events is obviously very close to that developed by Jaegwon Kim.
See, for example, his "Events as Property Exemplifications," in Myles Brand and D. Walton (eds.),
Action Theory (Dordrecht: D. Reidel, 1976), pp. 159-77. Events, Kim says, are "exemplifications
by substances of properties at a time"; an event is a "structure consisting of a substance (an n-tuple
of substances, a property (an n-adic relational property) and a time" (p. 160). But our account differs
from that of Kim in that we do not presuppose that there are such entities as times.
3. Theory of Science, ed. Rolf George (Oxford: Basil Blackwell, 1972), p. 79. Compare Hugo
Bergmann's Das philosophische Werk Bernard Bolzanos (Halle: Max Niemeyer, 1909), p. 40.
17
The Self in Austrian Philosophy

Bolzano's definition of substance provides us with a kind of key to the conceptions
of the self in Austrian philosophy. His definition is as clear as anyone could possibly wish. He says that there are two kinds of things: (1) those things that are states
or conditions of other things ("Beschaffenheiten von anderen Dingen"); and (2)
those things that are not states or conditions of other things: "the latter are what
I call substances."1 Examples of things that are states or conditions of other things
are "the color, smell and weight of a body," the beliefs that a particular person
has, the sensations that he has, and the actions that he performs. Examples of
substancesof things that are not states or conditions of other thingsare physical bodies and selves.
Bolzano says, in Leibnizian fashion, that, if there are things that are states or
conditions of other things, then there are things that are not states or conditions
of other things.2 If we use the term "substance" in the way he suggests, then we
need not ask whether a given philosopher believes in substances; we need ask only
what the things are that function for him as substances.
Sometimes Bolzano uses the word "property [Eigenschaft]" instead of "state
or condition [Beschaffenheit]". But as he realizes, this is somewhat misleading.
Acting, believing, and sensing are properties of me; but the actions that I perform
and the beliefs and sensations that I have are Beschaffenheiten and not properties
of me. One may say, for example, "Jones's sadness caused him to weep." This
tells us, of course, that Jones does have the property of being sad. But it should
not be taken to say that the property of being sad is what caused him to weep.
It tells us rather that that state or Beschaffenheit which is his being sadhis having
the property of being sadis what caused him to weep. I would say that properties, as distinguished from states and conditions, are abstract objects and that the
Platonist is right in calling them substances-provided he takes "substance" in
156
THE SELF IN AUSTRIAN PHILOSOPHY
157
Bolzano's sense. A person's states or conditionshis Beschaffenheitenare contingent things, dependent for their existence upon the person who is in them.
Such distinctions are obviously fundamental to the theory of the self.
Thus we may distinguish four quite different moments in Austrian philosophy.
(1) There is the view of the metaphysicians that the self is a substance. (2) There
is the view of early Austrian positivism that the self is not a substance but only
a Beschaffenheit. (3) There is that phase of Austrian philosophy in which the self
and its Beschaffenheiten are almost completely ignored. And finally, (4) there is
the somewhat embarrassed Wiederbekanntmachen of Austrian philosophy and
the self.
The Self as Substance

The self is a simple substance according to Bolzano.3 He speaks of "our soul
[unsre Seele]" but I believe that, strictly speaking, he would say, not that I have
a soul, but that I am a soul.
Franz Brentano's view is very much like that of Bolzano. There is no problem
in transforming Brentano's terminology of "substance and accident" into that of
"substance and Beschaffenheit." He says that a substance is in its accidents, and
it is natural to say, of course, that a thing is in its states. For Brentano and Marty,
as we know, the psychological properties of the self constitute the key to all of
metaphysics. I would say that their view has at least this much plausibility: the
self is the only contingent substance with respect to which we have nontrivial certainty.
Meinong provides a variation upon this theme which recurs in subsequent Austrian philosophy. Exaggerating only a little, we could say that, according to
Meinong, the Beschaffenheiten of the self provide us with the primary objects of
philosophical investigation, but the selfthat which has those Beschaffenheitenis of no importance at all. Psychological propertiesjudging, assuming,
sensing, perceiving, abstracting, comparingprovide Meinong with most of his
subject matter and constitute his principal philosophical interest. And selves are,
by definition, the things that have such properties. But Meinong shows no interest
whatever in what it is that has the properties that only selves can have. Unlike
Mach, he is not concerned to explain the self away or to reduce it to other entities;
he simply ignores it.
We now turn to the second moment.
The Self as Beschaffenheit

Ernst Mach claims to dispense with the concept of substance. But if we take "substance" in Bolzano's sense, it would be more accurate to say that, according to
Mach, the substances of the world are the things that Mach calls "elements"
namely, sensations. Physical things are complexes of sensations and so is the self.
What does it mean to say that the self is a complex of sensations? Mach ex-
158
presses himself this way: "The elements form the self. To say that / sense green
is only to say that the element green occurs in a certain complex of other elements."4 His view is that any truth that may seem to be a truth about me is really
a truth about certain sensations.
There are two serious difficulties with this version of the "bundle theory."
The first difficulty is that it is more of a promissory note than a theory; and
Mach gives us no directions for cashing it in. He seems to be saying that statements about physical bodies and statements about selves can be derived from
statements about sensations or complexes of sensations. What, then, of such simple statements as "There is a brown dog" and "Some people are philosophers"?
How are they to be derived from statements about sensations? Mach gives us no
answer and leaves us completely in the dark. We now know, what may have been
less obvious a hundred years ago, that the program of reducing statements about
physical things to statements about sensations involves difficulties in principle.
No one has found a way of overcoming these difficulties.
The second difficulty with Mach's version of the bundle theory of the self is
of a rather different sort. It is that of providing an answer to the question, "Which
sensations constitute the elements?" There are three ways of interpreting Mach
and none of them is satisfactory.
(1) According to one possible interpretation, your sensations and mine are
simply "fictions" or "thought objects"; they are two construction steps away from
those sensations that Mach finds to be the elements of the world. But you and I
know that this view is false. At least /know that my sensations are not constructions upon Mach's sensations.
(2) The second way of interpreting Mach is solipsistic: there is only one self.
So far as this possibility is concerned, the most reasonable comment is still that
of Thomas Reid: "A traveller of good judgment may mistake his way, and be unawares led into a wrong track, and while the road is fair before him, he may go
on without suspicion and be followed by others; but when it ends in a coal-pit,
it requires no great judgment to know that he hath gone wrong, nor perhaps to
find out what misled him."5
(3) According to the third way of interpreting Mach, your sensations are, for
him, fictions or thought objects, and his sensations are, for you, fictions or thought
objects. But this answer-with its "for him" and "for you"-reintroduces the concept of the self.
The problem, in a word, is this. We are provided with no way of deriving ordinary statements about the self from statements about the elements. And if we are
told that, all the same, selves are bundles of sensations, then we have to have a
way of distinguishing that bundle which is you from that bundle which is me. As
Brentano had said in his lectures on descriptive psychology, the concept of a bundle requires that of "a cord or wire, or the like, that ties things together."6 Without
such a cord or wire, the contents of your bundle would spill over into mine. And
no such cord or wire is at hand.
It is interesting to note that Ehrenfels, who was very much concerned with the
159
self and its Beschaffenheiten, also flirts with the bundle theory. He said in his Kosmogonie that "substantialism" leads to serious philosophical problems. His
thought seems to have been that, if we can emancipate ourselves from substantialism, then we can avoid the doctrine of the complete ineffectuality of the psychical.7 Where Mach thinks of the self as a bundle of sensations, Ehrenfels says that
it is a bundle of properties [Eigenschaften]. Like Mach, he doesn't tell us how
the different bundles are to be distinguished from each other.
Ignoring the Self and Returning to Its Beschaffenheiten

With the advent of physicalism, the self and its Beschaffenheiten seem to be ignored. But it does not take long for the latter to return to the center of the stage.
Consider Schlick, for example. He ridicules the Cartesian cogito and suggests
that the first-person pronoun cannot be used in expressing what is given in immediate experience. Yet when he tells us more about the nature of this experience, he makes essential use of the first-person pronoun.
In his essay "Uber das Fundament der Erkenntnis" (1934), he compares what
he calls "observation statements" with analytic statements. Observation statements and analytic statements, he says.
have in common that the occasion of understanding them is the same as that
of verifying them [daft bei beiden der Vorgang des Verstehens zuleich der
Vorgang der Verifikation isf\; I grasp their meaning at the same time as I
grasp their truth [mit dem Sinne erfasse ich zugleich die Wahrheit]....
Both are absolutely valid. However, while the analytic, tautological statement is empty of content, the observation statement supplies us with the
satisfaction of genuine knowledge of reality [echter Wirlichkeitserkenntnis].*
He goes on to say that an observation statement might be expressed by saying
something like "Here now blue" or "Here now yellow." But there was this objection: "Here now blue" and "Here now yellow" are not well-formed sentences; they
contain no verb, copula or quantifier; and the predicates "blue" and "yellow" seem
to function as terms. The year after the publication of his paper, Schlick deals
with this objection. An observation statement, he says, tells us something like
this: "There is something which is such that I believe I have used 'yellow' in the
past to refer to that sort of thing and which is here now."9 Further refinements
are necessary.10 But it is now clear that Schlick's observation statements are statements about the self.
So Austrian philosophy stumbles on the self once again.
Once we formulate our observation statements in the way Schlick proposes,
we see that there is a large class of psychological statements all having the
epistemological character associated with his observation statements. These are
the statements expressing what the members of the Brentano school had called
160
"inner perception"statements beginning with such expressions as "I wish that,"

"I wonder whether," "I seem to see that," "I feel that."
Combining the terminologies of Meinong and Bolzano, we may say that such
statements express self-presenting Beschaffenheiten of the self. It is these
Beschaffenheiten and not sense-data or sense-qualities that constitute the objects
of direct presentation or acquaintance. And when they are presented, then, as
Brentano holds, the self is presented nebenbeias that of which they are
Beschaffenheiten.
Whatever the ontological elements may be, the epistemological elements are
certain facts about the self.
And what of Wittgenstein? I would say that he gets us back on the road. The
Philosophical Investigations is, of course, a treatise on certain Beschaffenheiten
of the self. And a positive theory of consciousness is essential to the doctrine of
the Tractatus. Wittgenstein does discuss the self there, but he doesn't say enough.
I am thinking of this passage (5.641):
There is therefore really a sense in which in philosophy one can speak nonpsychologically of the I \nicht-psychologisch vom Ich die Rede sein kann].
The I occurs in philosophy through the fact that "the world is my world".
The philosophical I is not the man, not the human body or the human soul
[Seele] of which psychology treats, but the metaphysical subject, the
limitnot a part of the world.
Wittgenstein does not elaborate. We don't know how this view is to be compared
with that of Mach.
I would not say it to Wittgenstein myself, but if I were Bolzano I would speak
to him this way: "Herr Kollege, Sie wissen ja, daB woriiber man nicht schweigen
kann, daruber muB man sprechen."
I wish that Wittgenstein had said more. For any plausible interpretation of
what he did say has the consequence that the selfdas Ichis not one of those
things that are Beschaffenheiten of other things.
I look forward to seeing where Austrian philosophy goes next.
Notes
1. Bolzano, Athanasia oder Grilnde fur die Unsterblichkeit der Seele (Sulzbach: J. G. v. Seidleschen Buchhandlung, 1838), p. 283.
2. Bolzano (1827), p. 22. He holds that Beschaffenheiten may themselves have Beschaffenheiten
and that such things as numbers also have Beschaffenheiten (p. 22), and he seems to hold that God
has a Beschaffenheit (p. 22).
3. Bolzano (1827), pp. 22, 53. He also argues, convincingly, that there is no problem involved
in attributing spatial location to a simple substance (pp. 53, 287). He discusses this question further
in the Paradoxes of the Infinite, Sees. 55-57. There is a brief reference to the distinction between
substance and Beschaffenheit in Wissenschaftslehre, Sec. 142.
4. "Die Elemente bilden das Ich. Ich empfinde Grun, will, sagen dafi das Element Grim in einem
gewissen Complex von anderen Elementen (Empfindungen, Erinnerungen) vorkommt." Beitrage zur
Analyse der Empfindungen (Jena: G. Fischer, 1886), Sec. 17.
5. An Inquiry into the Human Mind, Chap. 1, Sec. 8.
161
6. "Zum 'Bunde!', wenn man es genau nimmt, gehort ein Strick oder Draht oder sonst etwas,
was zusammenschniirt." Deskriptive Psychologic (Hamburg, Felix Meiner Verlag, 1982), p. 11.
7. Christian von Ehrenfels, Kosmogonie (Jena: Diedrichs, 1916), pp. 59ff, 77ff.
8. Moritz Schlick, "The Foundations of Knowledge," in A. J. Ayer (ed.), Logical Positivism
(Glencoe, 111.: Free Press, 1959), pp. 209-27; the quotation appears on p. 225. The original version
of the paper, "Uber das Fundament der Erkenntnis," may be found in Schlick's Gesammelte Aufsatze
(Vienna: Gerold, 1938), pp. 289-310; the German version of the quotation appears on pp. 308-9.
9. "In our example, if the sentence 'Yellow here' stands for an affirmation [observation statement], then 'yellow' means 'the colour I remember always to have called "yellow" '. If so, there [may
indeed be] a deception of memory, but even in this case the affirmation remains true (so long as a
lie is not in question). Its truth does not depend on how I have otherwise really employed the words,
but only on how I think at this moment that I have employed them. But I cannot be mistaken about
that; as shown earlier, it is impossible for me not to have known this." From "Introduction and on
'Affirmations'," in M. Schlick, Philosophical Papers, vol. 2 (Dordrecht: D. Reidel, 1972), p. 412.
This article is a selection from Schlick's Sur le Fondement de la Connaissance (Paris: Herman & cie,
1935).
10. I have discussed the question in more detail in "Schlick on the Foundations of Empirical
Knowing," Grazer Philosophische Studien, 16/17 (1982): 149-57.
18
The Categories
A substancethat which is called a substance most strictly, primarily, and
most of allis that which is neither said of a subject nor in a subject, e.g.,
the individual man or the individual horse. (Aristotle, Categories, Chap. 5.)
We present finally a classification of the most fundamental ontological categories.
The classification will refine some of the concepts introduced in earlier essays.
There will be four dichotomiesfour ways of dividing sets of things into exclusive and exhaustive subsets. In each case, one of the two subsets will be the negation of the other. I will also attempt, so far as possible, to characterize each subset
in positive terms.
The dichotomies are these:
(1) Things that are contingent and things that are noncontingent or necessary;
(2) contingent things that are states and those that are nonstates or contingent individuals; (3) contingent individuals that are limits and those that are nonlimits
or contingent substances; and (4) noncontingent things that are abstracta and
those that are nonabstracta or noncontingent substances.
A Table of Categories
'EntityContingent
States
Events
Necessary
Individuals
Limits
Substances
162
Abstracta
Substances
THE CATEGORIES
163
According to this way of looking at the world, then, there are contingent substances along with their states and their boundaries or limits; and there are necessary things, each of which is either an abtractum or a substance. I believe we have
no good reason to affirm the existence of any other type of thing.
The Basic Concepts

Among the principal desiderata in thus setting forth a theory of categories are (1)
economy with respect to the types of entity which are countenanced and (2) simplicity with respect to the types of concept which are used.
I will make use of these undefined concepts: (1) x exemplifies (has) y; (2) x
is necessarily such that it is F; (3) x is a state of y; (4) x is a constituent of y;
and (5) x attributes y to z. I will introduce three types of term: (1) "being-F"
wherein the letter "F" may be replaced by any well-formed English predicate; (2)
"x-being-F" which will be used to designate states of the entity designated by "x";
and (3) "that-p" in which the letter "p" may be replaced by any well-formed English sentence. And I will make essential use of tense. For I assume that there are
truths that can be adequately expressed only in a language that, like our ordinary
language, is tensed.
Since I am using tensed language, I will say that whatever exists exists now.
And since the language is tensed, the "now" in "Whatever exists exists now" is
redundant and the statement is logically true. But there was a philosopher who
drank the hemlock. And this means that there is something-for example, the
property of being blue-that was such as to have the property of being such that
there is a philosopher who is drinking the hemlock.
I now turn to the four dichotomies.
(1) Contingent and Necessary Things

How are we to distinguish between those things that are contingent and those
things that are not?
We have the locution "x is necessarily such that it is F." But unfortunately we
cannot use this locution to make the distinction between necessary and contingent
things. For "is necessarily such that it is F," at least as I interpret it, is equivalent
to "is necessarily such that it exists if and only if it is F." And if the locution is
taken in this way, then everything may be said to "exist necessarily"for everything is necessarily such that it exists if and only if it exists. So "exists necessarily"
does not yield the distinction between necessary and contingent things.
A contingent thing, unlike a noncontingent thing, is a thing that might not have
been-a thing that is possibly such that it came into being and is possibly such
that it will pass away. And so I will return to the concepts of coming into being
and passing away. (Our definitions of these concepts, like all our definitions, are
in the present tense.)
164
THE CATEGORIES
Dl x is coming into being = Dfx is such that there is nothing it did exemplify
D2 x has just passed away = Df Something that was such that x exists begins
to be such that x does not exist
There are things that you and I did exemplifysay, the property of being a
childand, therefore, we are not now coming into being. And there are, I trust,
properties that we will exemplifysay, walking somewhere later today.
If we interpret "x is such that it is F" correctly, we will see that, if a thing is
not possibly such that it is coming into being or passing away, then it never was
and never will be possibly such that it is coming into being or passing away.
The distinction between contingent and necessary things, then, is this:
D3 x is a contingent entity = Df x is possibly such that it is coming into being
or has just passed away
A necessary thing is a thing that is not contingent.
(2) States and Individuals

Contingent things may be divided into those that are states of things and those that
are not states of things. In "States and Events," we put our assumption that there
are states of things this way:
Al For every x, there is the state x-being-F if and only if x exemplifies
being-F
Bolzano had said that a state is an entity that is o/something. We may express
this fact by saying that states are ontologically dependent upon the things of which
they are states:
A2 For every x and y, if x is a state of y, then x is necessarily such that it
is a state of y
A stone that is warm is not necessarily such that it is warm; but its-being-warm
is necessarily such that it is a state of the stone.
Those contingent things that are not states of other things may be called individual things.
D4 x is a contingent individual = Df x is a contingent thing that is not a state
In saying that an individual thing is something that is not a state, we are saying
of it that it is not "in a subject." But we are not assuming that "the individual alone
is real," for we are assuming that there are states and that states are not individual
things.
In the essay "States and Events," I have characterized events as constituting
a subset of states.
THE CATEGORIES
165
(3) Limits and Contingent Substances

An adequate theory of categories should enable us to distinguish between those
contingent individual things that may properly be called substances from those
contingent things that are the limits or boundaries of substances.
To make this distinction, I use the locution "x is a constituent of y." I assume that
the relation of being-a-constituent-of is asymmetrical and transitive: for every x
and y, if x is a constituent of y, then y is not a constituent of x; and for every x, y,
and z, if x is a constituent of y and if y is a constituent of z, then x is a constituent of z.
A limit or boundary is, in the following sense, a dependent individual:
D5 x is a limit (boundary) = Df (1) x is a contingent individual; and (2) every
constituent of x is necessarily such that it is a constituent
Now we may say what a contingent substance is:
D6 x is a contingent substance = Df x is a contingent individual that is not
a limit
The parts of a contingent substance are those of its constituents that are not
boundaries or limits. It follows, therefore, that the parts of a contingent substance
are themselves contingent substances. Hence, we should reject the view of
Aristotle, according to which the parts of actual substances are not themselves
actual substances. And we should reject the view of Leibniz, according to which
actual substances cannot be parts of actual substances.
(4) Abstracta
It is often said that there is just one ens necessariumnamely, God. But if, as
I believe, extreme realism, or Platonism, is true, and if the distinction between
contingent and necessary things is to be drawn in the way that I have suggested,
then it follows that there are indefinitely many necessary things. All so-called abstracta are necessary thingsthings incapable of coming into being or passing
away. These include not only exemplified attributes, such as the attribute of being
a dog, but also unexemplified attributes, such as the that of being a unicorn or
that of being a round square.
Is there any reason to believe that there are abstracta that are not capable of
being attributed?
What of classes, or sets! Russell showed how the principles of set theory may
be construed as being principles about attributes. To say that x is a member of
the class of F's is to say that x is F; to say that the class of F's includes the class
of G's is to say that everything that is G is F; and more generally, to say that the
class of F's is so-and-so is to say that the attribute of being-an-F is exemplified
166
THE CATEGORIES
by exactly the same things as is an attribute that is so-and-so. Following Russell,

then, we will say this:
D7 The class of F's is G = Df There is an attribute which is such that (a) it is
G and (b) it is exemplified by all and only those things that exemplify
being-F
Given that we have countenanced the being of attributes, there is no seed to
assume, therefore, that in addition to attributes there are also such things as
classes or sets. Our definition, however, enables us to use the convenient terminology of classes or sets. (Since we do not suppose that there are sets in addition
to attributes, we need not face such difficult questions as: "Do sets have their
members necessarily?" and "Can sets change their members?")
Aren't states of affairs to be counted among abstract objects? To be sure, there
is that abstract state of affairs which is all men being mortal, as well as that state
of affairs which is some men not being mortal. But we may characterize these
states of affairs by reference to being such that all men are mortal and being such
that no men are mortal. More generally, the state of affairs that-p may be characterized by reference to being such that p. Our definition is this:
D8 That-p is a state of affairs = Df There is an attribute which is necessarily
such that it is exemplified only if p
This definition guarantees that states of affairs are abstract things and not contingent events. For attributes are necessary things and, therefore, if an attribute can
be said to be "necessarily such that p," then the sentence replacing "p" cannot express a contingent event.
Since states of affairs are thus reducible to attributes, the expression "the state
of affairs that-p obtains" is reducible to a statement about exemplification:
D9 The state of affairs that-p obtains [is true] = Df Something has an attribute
which is necessarily such that it is exemplified if and only if p
What of propositions! If we use "proposition" to refer to a type of abstract object and not to a type of contingent thing (such as those "singular propositions"
that are thought to contain contingent things as their constituents), then there
would seem to be no ground for distinguishing propositions from states of
affairs-unless we say that propositions are those states of affairs which are necessarily such that either they are always exemplified or they are never exemplified.
And the concept of the truth of a proposition would be explicated by reference
to exemplification in the way suggested by D9 above.
If there is reason to think that there are such things as possible worlds, then
such entities may be identified with a certain type of state of affairs.
DIG W is a world = Df W is a state of affairs such that: for every state of
affairs p, either W logically implies p or W logically implies the negation
of p; and there is no state of affairs q such that W implies both q and the
negation of q
THE CATEGORIES
167
In other words, a word is a self-consistent, maximal state of affairs. That it is

maximal is guaranteed by the first clause of the definition; and that it is selfconsistent is guaranteed by the second.
If possible worlds are thus reducible to states of affairs, and if states of affairs
are reducible to attributes, then possible worlds are reducible to attributes.
What of relations'! In contemporary logic it is usual to assimilate relations to
sets of a certain sortnamely, those sets that are ordered-pairs. For example, following Kuratowski, one may construe the ordered-pair, x-paired-with-y, as the
set whose sole members are (a) the set whose sole member is x and (b) the set
whose sole members are x and y. Hence, we are able to distinguish the set xpaired-with-y from the set y-paired-with-x. This conception is readily carried
over into the theory of attributes.
An ordered attribute, John-to-Mary, will be any attribute whose sole instances
are: (1) an attribute whose sole instance is John and (2) an attribute whose sole
instances are John and Mary. If John is, say, the tallest man in town and if John
and Mary are the only people in the green Chevrolet, then an instance of the ordered attribute John-to-Mary would be the attribute of being exemplified only by
the tallest man in town and by the only people in the green Chevrolet. This attribute, then, is one of those things that are exemplified by the relation being taller
than. So, too, for any of the other attributes that are thus ordered from John to
Mary (say, the attribute of being exemplified only by the local television repairman and by the 17th and the 284th persons to have registered in the last local election). All these ordered attributes exemplify the relation being-taller-than. Hence,
we may say that John is paired to Mary by the relation being taller than.
Relations, then, will be characterized as follows:
Dl 1 R is ordered from x to y = Df R is an attribute whose sole instances are
(a) an attribute whose sole instance is x and (b) an attribute whose sole
instances are x and y
D12 R is a relation = Df R is necessarily such that, if it is exemplified then
there exists an x and a y such that R is ordered from x to y
This conception of relations may be extended to relations of any number of
terms. If a relation has three terms, then one of its instances has two terms and
the other just one; if a relation has four terms, then either (a) each of its instances
has two terms or (b) one of them has three and the other just one; and so on. Every
relation, no matter how many terms it has, is an attributean attribute that is exemplified by other attributes.
(5) Noncontingent Substances
We will assume, then, that there are two types of noncontingent beingthose that
are attributes and those that are not attributes. We will say that if there is a necessary being that is not an attribute, then it is a necessary substance:
168
D13
THE CATEGORIES
is a necessary substance = Df x is not contingent and x is not an attribute
A necessary substance, then, is an eternal object. And since a necessary substance

is not possibly such that there is anyone who attributes it to anything, we may
say with Aristotle that it is something that is "not said of a. subject."
Such a characterization of necessary substance is essentially negative. We will
not consider here the question whether there is such a being or whether, (/"there
is such a being, we may characterize it in positive terms.
Index
Index
Abelard, 60
Abstract objects, 99, 101, 165-67
Aggregates, 91-92, 152-53
Anselm, St., 13-15
Appearances, 54, 91-92, 120-24
Aquinas, St. Thomas, 7, 34, 53-54, 126, 128
Aristotle, 5, 9, 10, 53, 58-94, 120, 150, 162,
165
Attributes. See Properties
Attribution, direct and indirect, 111-12
Austrian philosophy, 156-61
Ayer, A. J., 127
Baker, Lynn Rudder, 113
Bayle, Pierre, 41
Becker, Lawrence C., 57-58, 61
Bergmann, Hugo, 89, 165
Bochenski, J. M., 24
Bogdan, RaduG., 95
Bolzano, B., 89, 150, 156-57, 160
Boundaries, 9, 165
Bowne, Border Parker, 125
Brentano, Franz, 61, 85, 89, 109, 113,
116-17, 128, 157, 160
Broad, C. D., 60, 94
"Bundle theory," the, 50, 158-59
Butler, Bishop Joseph, 25, 33, 40, 125, 128
Campbell, C. A., 14
Carnap, R., 39, 95
Cartwright, Helen, 40
Cartwright, Richard, 90-95
Castaneda, H. N., 113

Categories, 162-68
Causal theory of meaning, 137-38
Causation, 8-9, 153
Chandler, Hugh, 82
Classes, 165-67
Coming into being, 49-56, 151, 164
Common sense, 49-50
Conceiving, 130-32, 143
Constituents, 83-86
Contact, 89, 90
Content and object, 132-33, 135
Contiguity, 84, 88-89
Contingent things, 162-68
Continuity, 84
Conveying, 132-35
"Daylight view" of matter, 123-24
De dicto attitudes, 132
De re attitudes, 130-38
Demonstratives, 109, 129-40
Dependence, ontological, 83, 150, 164
Descartes, R., 85
Descriptive psychology, 116-17
Designation, 137-38
Determinism, 3-15
Dimensions, 87-89
Donnellan, Keith, 131, 138
Double-aspect theory, 123-24
Dualism, 120-21
Edwards, Jonathan, 11-12, 39
171
172
Ehrenfels, Christian von, 158-59, 161

Elanguescence, 55-56
Endeavor, 123-24
Entailment, 144
Entia per accidens, 53-55
Entia per se, 150
Entia successiva, 125
Essence, individual. See Haecceity
Essential properties, 116-17. See also Mereological essentialism; Pseudo-essentialism
Events, 152-55, 164-65
Exclusion, 154
Fechner, Gustav, 123-24
Feehan, Thomas, 138
Feigl, Herbert, 128
Feldman, Fred, 39, 106
Fission, 71
Foley, Richard, 106
Follesdal, Dagfin, 113
Frederick, Robert, 89
Free will, 3-15
Frege, G., 89, 112, 149
Fusion, 71
God, 165, 168

Haecceity, 47-48, 109-10
Harney, M. J., 112
Henry, D. P., 15, 81, 82
Heraclitus, 51-52
Hintikka, Jaakko, 19, 24
Hirst, R. J., 128
Hispanus, Petrus, 24
Hobbes, Thomas, 12
Hume, D., 81
Husserl, Edmund, 112, 116
Identity, criteria of, 144-45
"Identity theory," the, 117-19
Identity through time, 124-25
Implication, property, 101-2, 143-44, 153
Inclusion, property, 101-2, 143-44
Increase, "the paradox of," 80
Individual things, 164-65
Individuation, 151
Intentionality, 107-12
Involvement, property, 143, 153-54
James, William, 112-13
Justification, epistemic, 131-32
INDEX
Kant, I., 55-56, 60
Kaplan, David, 113
Kim, Jaegwon, 155
Korner, Stephan, 83, 89
Kretzmann, Norman, 61
Leibniz, G. W., 12, 48, 66, 135, 138, 165
Lesniewski, S., 95
Lewis, C. I., 48
Linguistic theories of reference, 112-13
Locke, John, 11, 39, 82, 121
Lovejoy, A. O., 120-22, 127
Luz bone, 135
Mach, Ernst, 50, 157-59
McTaggart, J. M. E., 81
Marty, Anton, 157
Materialism, 100
Meinong, A., 157
Melden, A. I., 12, 15
Mendelssohn, Moses, 60
Mereological essentialism, 65-82, 86
Minds, 114-28
Modes, 81
Monads, 93
Moore, G. E., 49-50, 66
Names, 136-38
Necessary things, 162-64, 168
Necessity and possibility, 42-48
Negative events, 162-64, 168
Negative properties, 146-47, 153-55
Noemata, 112, 113
Occurring, 166
Ordered properties, 167
Parts, 69-75, 83, 152-53
Passing away, 49-50, 151, 164
Peirce, C. S., 37, 41
Perceiving, 133-34
Perry, John, 113
Personalism, 125
Persons, 59-60, 156-61
Phenomenology. See Descriptive psychology
Plantinga, Alvin, 47-48
Plato, 126
Platonism, 93, 141-42
Possible worlds, 166-67
Possibility. See Necessity and possibility
Potter, Richard, 89, 106
Primacy of the intentional, 112, 129-38
INDEX
173
Quinn, Philip, 106
Sets. See Classes

Shoemaker, Sidney, 39, 41
Shope, Robert, 106
Sosa, Ernest, 106, 133
Souls, 115, 157
States, 94, 130, 156, 164-65
States of affairs, 148-49, 166
Stout, G. F., 151
Strawson, P. F., 128
Strong, C. A., 124, 128
Suarez, F., 9, 83, 86
Substance, 93, 156, 168
Sums. See Aggregates
Recurrence, 153
Reid, Thomas, 6, 79
Relations, 117
Renear, Allen, 106
Representation, inner systems of, 112
Roe versus Wade, 57-58
Royce, Josiah, 112
Russell, Bertrand, 50, 109, 113, 165-66
Tarski, Alfred, 95
Taylor, Richard, 14, 128
Temporal parts, 90, 94-95
Temporally denumerable states, 151-52
Tense, 88, 163
Thomson, Judith Jarvis, 95
Tomberlin, James, 113
Truth, 163
Sartre, Jean Paul, 24

Scattered objects, 90-95
Schlick, Moritz, 159, 161
Schuwey, Bruno, 166
Self. See Persons
Self-attribution, 160
Sellars, Roy Wood, 128
Sense and reference, 132, 135-38
Sense-data. See Appearances
Sense-datum fallacy, 121-23
Van Cleve, James, 89
Prior, A. N., 24
Probability theory, 183
Process philosophy, 94-95
Processes, 94
Properties, 88-106, 141-49; conjunctive, 142,
146; disjunctive, 102-3; internal, 101-4;
negative, 142, 146-47; nonrelational,
101-2; psychological, 99-106
Propositions, 129-30; abstract, 108-10,
148-49; singular, 109-10
Pseudo-essentialism, 66-67
Whitehead, A. N., 69, 81, 91, 95, 138

Wilson, N. L., 24
Wisdom, John, 40
Wittgenstein, L., 112, 116, 127, 149, 160
Woodger, J. H., 39
Zimmerman, Michael, 89, 106
Roderick Chisholm is author of Theory of Knowledge (third edition, 1989),

Brentano and Intrinsic Value (1986), The Foundations of Knowing (Minnesota,
1982), The First Person (Minnesota, 1981), Person and Object (1976), and Perceiving: A Philosophical Study (1957). He has served as president of the Eastern
Division of the American Philosophical Association, of the Metaphysical Society
of America, and of the International Phenomenological Society. He formerly was
editor of Philosophy and Phenomenological Research and is professor of philosophy at Brown University. He is also the editor of numerous posthumous books
by Franz Brentano.

Chisholm - On Metaphysics PDF

Uploaded by

Copyright:

Available Formats

Chisholm - On Metaphysics PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chisholm - On Metaphysics PDF

Uploaded by

Copyright:

Available Formats

On Metaphysics

This page intentionally left blank

University of Minnesota Press Minneapolis

Copyright 1989 by the Regents of the University of Minnesota

Part I. Freedom and Determinism

II. Coming into Being, Persisting, and Passing Away

III. Parts and Wholes

IV. The Mental

V. An Intentional Approach to Ontology

This page intentionally left blank

This page intentionally left blank

This page intentionally left blank

Responsibility and Avoidability

RESPONSIBILITY AND AVOIDABILITY

Human Freedom and the Self

HUMAN FREEDOM AND THE SELF

HUMAN FREEDOM AND THE SELF

HUMAN FREEDOM AND THE SELF

HUMAN FREEDOM AND THE SELF

HUMAN FREEDOM AND THE SELF

HUMAN FREEDOM AND THE SELF

HUMAN FREEDOM AND THE SELF

HUMAN FREEDOM AND THE SELF

HUMAN FREEDOM AND THE SELF

HUMAN FREEDOM AND THE SELF

8. Essay Concerning Human Understanding, Book II, Chap. 21.

This page intentionally left blank

This page intentionally left blank

Identity through Possible Worlds

IDENTITY THROUGH POSSIBLE WORLDS

is incompatible with the principle of the indiscernibility of identicals. How could

IDENTITY THROUGH POSSIBLE WORLDS

IDENTITY THROUGH POSSIBLE WORLDS

commit ourselves to affirmative answers to the puzzling questions we have encountered.

IDENTITY THROUGH POSSIBLE WORLDS

IDENTITY THROUGH POSSIBLE WORLDS

involved, it seems to me, provided we continued to affirm such principles, de dicto,

Identity through Time

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

one such set of things. If we continue to speak strictly and philosophically, we

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

in agreement with respect to the proper interpretation of one of those expressions

IDENTITY THROUGH TIME

a matter of convenience. Which ship is to be called "the Ship of Theseus"the

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

IDENTITY THROUGH TIME

Possibility without Haecceity

POSSIBILITY WITHOUT HAECCEITY