Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
Joseph Almog
OXFORD
U N I V E R S I T Y PRESS
Oxford University Press, Inc., publishes works that further
Oxford University's objective of excellence
in research, scholarship, and education.
Oxford New Y o r k
Auckland Cape T o w n D a r es Salaam Hong Kong Karachi
Kuala Lumpur Madrid Melbourne Mexico City Nairobi
New Delhi Shanghai Taipei Toronto
With offices in
Argentina Austria Brazil Chile Czech Republic France Greece
Guatemala Hungary Italy Japan Poland Portugal Singapore
South Korea Switzerland Thailand Turkey Ukraine Vietnam
www.oup.com
ι 3 5 7 9 8 6 4 2
for me, it is the other way round: we have many clear answers
but no fundamental question.
Perhaps we are overly wedded to the standard philosophical
languages. A t one point in his correspondence with Princess
Elizabeth, Descartes, after he himself so ingeniously used the
standard philosophical languages of his time, says to her that
to understand the mind-body union, one needs to bracket away
metaphysical analysis and fall back on "ordinary life and con-
versation." Well, I tried. Such is the aforementioned graphic
language o f the movie Blade Runner. Recently, I have so de-
ployed ih teaching the language of an imaginative novel, The
Golden Compass Qjy Philip Pullman). The author there connects
in a most beguiling way each person—seemingly a kind-mate
of you and me—with what he calls his (her) daemon. Such
graphic aids are invaluable but one still feels the problem has
not been articulated.
The philosopher by nose he was, I do think that Descartes
smelled things right—the insight that is needed will come not
from, but in spite of, technical philosophical languages, by at-
tending to our "ordinary life and conversation." A t this time, I
don't know how to do it.
2. Integrative Dualism, 59
2.1. The Prove Too Much/Prove Too Little Dilemma, 59
2.2. Separatist Dualism: Four Basic Annotations, 61
2.3. Integrative Dualism: The Quest for Symmetry, 71
2.4. Subjects in Time: T w o Cartesian Frameworks, 81
Index, 125
Synopsis: The Project and the Challenge
xvii
xviii W H A T A M I?
ι. The readings I take to be relevant to this issue and which are re-
ferred to in detail below are the exchange with Arnauld over "complete
ideas" in the fourth replies (CSM I I , 156-160), related remarks to Caterus
in the first replies (CSM I I , 85-86), and the letters to Mesland (May 2,
1644, PL, 152) and Gibieuf (PL, 24).
3
4 W H A T A M I?
(1) D M bears F.
(3) D M Ψ DB.
6 W H A T A M I?
Hence,
(3C) D M Ф DB.
Hence,
(3P) D M Ф DB.
(3W) D M Ф DB.
Arnauld's Objection
ing Kripke's own use, "possibility tout court." Here we encounter one of
the advantages of lecturing about Descartes to immaculate beginners; they
do not ask questions of this kind. More seriously, I flnd it hard to make
such comparisons. The examples used by Descartes and Arnauld concern
mathematics. In contrast, a favorite stock of examples in our age concerns
the natural sciences (e.g., water is made in part of oxygen; gold is an
atomic element; etc.). Nonetheless, I happen to believe that the modern no-
tion of Kripke is more continuous with the Descartes-Arnauld discussion
than the 300-year gap might lead us to think. Still, in what follows I make
an effort not to indulge in time travel from Descartes's period to ours. Our
discussion—inevitably one run by a contemporary mind shaped by contem-
porary texts—aims at understanding Descartes's text in his own terms.
THE REAL D I S T I N C T I O N 17
now that, in fact, it turns out that it is not merely actually true
that my mind is connected to a (this) body but also that of
necessity it is so connected. I f so, I stand corrected in my con-
ceivability claim. I have not really conceived of my mind in
this exercise; I have merely conceived of a mind, or of an epi-
stemically equivalent subject that from the inside is indistin-
guishable from my (our kind of) mind. O f that item, I have
conceived—really conceived—that it exists without bodily con-
nection. This does not show that I can so conceive o f my (our
kind of) mind. A l l in all, concludes Arnauld, D M and D B may
well be necessarily connected. They may or may not be two
items (this much is left open). But there is no real possibility
of D M without D B . I n this sense of "can," there is no real
distinction between D M and D B . Descartes refuted.
see.
This view sacrifices the reliability of conceivability as a
guide to possibility only to regain the reliability of our own
self-ascriptions of what we conceive. Regarding Arnauld's t r i -
angle, this view proposes that we did manage to conceive T
without P (though it is not possible for T to exist iMess);
furthermore, we do conceive of water even i f we make i t — i n
the story—oxygenless; and we may well conceive of Elizabeth
even i f we misplace her in time, flouting a necessary truth, and
place her in the eighteenth century. I n the mind-body case, the
response asserts that we are successful in conceiving the mind
without any body on hand, even i f it turned out to be impos-
sible for the mind to exist without (this specific) embodiment.
The question of whether we manage to conceive of the mind-
bearing property F is simply prior to the question of whether
that mind could have really been F. I f it could not, fine: We
have already conceived of it as bearing F.
In sum, the present response suggests that we seem to con-
ceive of D M without D B ; thus, we really do. I n our conception
(or " i n the understanding"), D M is distinct from D B . We
would now like to conclude as follows: and so, in reality, D M
is distinct from D B . What is needed for this last step is a certain
kind of conceivability-reflecting-reality projection principle.
Exactly what kind o f principle? A general reality-projection
principle cannot be correct. Just because I conceive of you on
the beach in Rio does not make you be there. But there may
22 W H A T A M I?
directly are the С-ideas. We thus really say, " I n our concep
tion, C ( D M ) Ф C ( D B ) . " But now, by the reflecting reality
principle, we may infer that in reality D M is not D B .
I have at this point two things to say about this line of
argument. First, even granting the foregoing, we would still
have no result of real distinction. We would have the numerical
distinction of D M and D B , very much consistent with the ex
clusion of any sense of "can" according to which the one can
exist without the other. Second, the argument critically depends
on developing Descartes's notion of complete idea, which I do
in the next few pages. Then I return briefly to an evaluation of
the present defense of a conceivability-based yet possibility-free
proof of the numerical distinction of D M and D B . 7
7. See the notion of "conceptual fix" in I.5 and the analysis of conceiv
ability illusions in ch. 3, section 3.1.
2 4 W H A T A M I?
about the piece of wax (on which much more later) and about
the mind and the body, and it raises our primal question: What
am I? I n all, Descartes is after the kind of item involved, not
just the listing of its accidents, necessary though they may be.
The theory about whatness-specifying features is part of the
metaphysics of a given subject. But we see that it soon leads to
epistemic corollaries. I n the quest for natural constraints on
what it takes to have a conceptual fix on a given object x, we
look for metaphysically distinguished features of x. Arnauld
proposes the necessary features. But they fail twice over. They
fail epistemically because intuitively this demands too much.
We all feel that arbitrary, hidden necessities about x cannot undo
our mental hold on x. Arnauld's modal answer fails a second time
in not distinguishing, now in the metaphysics of x, the premodal
articulation of just what kind of thing this subject is from a subse-
quent business—how that subject (now already given) might
have been, which ways it could have gone on to be.
I submit that both basic facts must be respected by a theory
of complete ideas or conceivings about x. I t now seems that a
proper treatment of the epistemic basic fact might be a byprod-
uct of respecting the metaphysical basic fact. I n other words, a
theory of genuinely thinking about x may rely on a theory of
what χ is. I believe that Descartes's remarks about complete
ideas supply such a theory, and it is to this theory that I now
turn.9
12. This much applies as well to fictional characters like Hamlet. Shake-
speare did not generate Hamlet by gluing properties together. Rather, the
world (here certain historical acts of the author, Shakespeare) provided
Hamlet. We may now try to conceptually subtract some of Hamlet's fea-
tures, for example, being melancholic. Of course, with fictional items we
have to be careful about what Hamlet is in the story (a man and melan-
cholic) and what he is outside the story (a fictional character). Various com-
plications arise from this duality but they do not take away the main point:
Real subjects—fictional or not—are not created by gluing properties to-
gether.
34 W H A T A M I?
ally false about sets and (ii) as such, necessarily false. Nonethe-
less, the story is a coherent story about sets, preserving their
very essence.
Here is a summary o f such a story. The theory of sets is
taken to describe a "real" universe of sets V . By one of set
R
of sets that is not itself collectible into one more set. When set
theoreticians initially thought about the fundamental structural
operations that generate the universe o f sets, they articulated
the following: We generate sets in stages. We start, in stage
zero, by taking as our basis the empty set, the unique set with-
out members. Given this basis, we go on to generate, at later
stages, more sets. We have two such generative operations for
the two kinds of stages we encounter. I n any successor stage—
the one after the first, the one after that, and so on—we take
the power set, the set of all previously generated subsets, and
make that power set the new universe. The second operation
concerns limit stages of construction—the stage beyond the
first, second, and so on—a stage we think of as the first inhnite
stage, dubbed the CO stage. A t such a limit stage, we take the
union o f all previously available sets. Applying these operations
on and on, we generate the universe of sets.
The question now is this: How many stages does it take to
generate the full universe? Suppose we think that there are only
CO many stages o f generation. We get a universe with infinitely
many sets, each of which is merely a finite set. I n this universe
of finite sets, no single set is itself infinite. We may call this the
universe of finite sets, V . This story casts in modern set theo-
F
retic terms (as it were, with hindsight) what was probably the
received view for generations of mathematicians who believed
in the merely "potential infinite," a universe with infinitely
many items but no single infinite item. This common view was
questioned in the nineteenth century by the work o f mathemati-
cians like Bolzano, Dedekind, and Cantor.
Under the impulse of these nineteenth-century pioneers, it
has become a received truth of modern set theory—by the first
36 W H A T A M I?
there are "larger" sets. Modern set theory asserts not only that
the universe of hitherto generated finite sets is itself infinite but
also that there exists at least one infinite set; for example, ω
itself is such a set. Indeed, it goes on to assert that many such
infinite sets exist. But the existence of at least one such set, set
I , is sufficient for our purposes. I will assume it is true—in the
real universe of sets V —that there is this set I . Since this is a
R
13. (For the technically minded reader only): Of course, this example is
presented here merely as an illustration. I do not mean to delve into the
philosophy of set theory. What is more, many might think that the opera-
tions under which the universe of finite sets is closed do not exhaust what
sets are. They may think so (i) because V is wrong about how many sets
F
a way to number such infinite sets according to size, thus providing for the
second infinite size, the third, and so on. But what about the 00 place in
this sequence, what is standardly referred to as K ? Is it also a set? Zer-
m
melo's own theory "forgot" to state principles that guarantee the existence
of this set. The theory is consistent with there not being such a set. But
given its actual existence in V , it is impossible that this set does not exist.
R
Similar remarks apply to yet larger sets, for example, those called in set the-
ory "the (strong) inaccessible" cardinals, whose theory was articulated by
Zermelo later, in the 1930s. The objector who resists the present line of re-
sponse will have to suggest that each such (existential) truth about sets is
ipso facto pertaining to the essence of sets. I regard this as much too strong
a claim. It does not distinguish arbitrary truths about sets from truths about
what they are.
38 W H A T A M I?
14. See Saul Kripke, Naming and Necessity (Cambridge, Mass.: Harvard
University Press, 1980), pp. 144—152.
52 W H A T A M I?
"Pain is distinct from FCF." I have taken the extra step here of making
Kripke assert that given the failure to expose an illusion, the seeming possi
bility is a real possibility.
I should point out that I stick to the role of an expounder of Kripke's
views on pain. I take Kripke to be proposing that its phenomenological
quality exhausts its list of necessary features. For myself, I have doubts
about this on Kripkean grounds. It is Kripke who inspires us, in Naming
and Necessity, to depart from Hume's famous doctrine that there are no nec
essary connections between distinct existences. To the contrary, everything
real is necessarily connected to, or dependent on, other real things. Notable
THE REAL D I S T I N C T I O N 53
Integrative Dualism
59
6o W H A T A M I?
ι. Let me add here a general remark, going beyond our strict Descartes
agenda. The separation between personal identity problems and lifetime
metaphysics questions strikes me as of general use. Personal (ship, bicycle,
etc.) identity theories fuse in my view (i) atemporal (pseudo) issues about
individuation and defining criteria with (ii) genuine lifetime (endurance)
problems for individuals (with a given, primitive identity). In many of the
genuine questions raised in personal identity puzzles, I flnd a quest for an
explanation of the limits in the change potential of a specific item (Rene
Descartes, the Titanic, etc.) of a specific kind (a person, a ship, etc.). More
on the methodology of personal identity theories arises in chapter 3, in our
discussions of illusions of conceivability.
INTEGRATIVE DUALISM 67
E v e r y t h i n g i n w h i c h there exists i m m e d i a t e l y , as i n a
subject, or t h r o u g h w h i c h exists a n y t h i n g w e perceive
INTEGRATIVE DUALISM 69
Separatist m i n d ?
fundamental theses:
6. CSM I I , 17.
INTEGRATIVE DUALISM 75
T h e p r i m a l i t y o f m a n encoded i n t o ( I D B i ) and ( I D B 2 ) i m -
pinges o n existence questions broached i n ( I D B 3 ) . T h e exis-
tence o f the b o d y is subordinated t o that o f the m a n : W h e n the
m a n dies, its l i v i n g h u m a n b o d y is dead and gone; the r e m a i n -
i n g alignment o f molecules o n the pathologist's table is—despite
its b e i n g the same material extension as before—another k i n d
o f t h i n g , a corpse. T h u s , as u r g e d i n ( I D B 4 ) , D B has been
transformed f r o m a generic abstraction—an instantaneous dis-
t r i b u t i o n o f matter i n space—into a historical subject, one that
is generated i n h i s t o r y , undergoes change, survives material a l -
terations i n c o m p o s i t i o n , and eventually corrupts w h e n the m a n
whose b o d y i t is has come t o an end. F i n a l l y , ( I D B 5 ) and
2.4. Subjects i n T i m e : T w o
Cartesian Frameworks
10. In what follows, when I speak of "the wax passage" I refer to CSM
I I , 20—21.
86 W H A T A M I?
и . CSM I I , 59-60.
92 W H A T A M I?
(Of-ness) A m i n d is o f a g i v e n k i n d i f f i t is the m i n d o f a
b e i n g o f that k i n d .
3.1. T h e Real D i s t i n c t i o n R e v i e w e d
99
ioo W H A T A M I?
w i t h w h a t D M that i t w o u l d be l o d g e d i n another h u m a n b o d y
and, for that matter, another h u m a n b e i n g .
T h e v i e w just o u t l i n e d is certainly appealing. A m o n g m a n y
critics o f classical separate dualism, the v i e w just expounded is
a natural balance p o i n t . T h e y a d m i t that t o assert the conceiv-
a b i l i t y o f full disembodiment is a mistake. B u t equally mistaken
is the attempt t o deny o u r a b i l i t y t o conceive o f D M w i t h o u t
this particular b o d y , D B , and this particular h u m a n , R D , or to
deny o u r a b i l i t y to conceive o f this h u m a n b e i n g w i t h o u t this
v e r y specific m i n d , D M . I call this f o r m o f integrative dualism
generic integrative dualism because i t assigns t o D M and D B the
still somewhat generic essence—the m i n d ( b o d y ) o f a h u m a n
b e i n g , o f some h u m a n or other. 3
with the city of Venice, is implanted with "memory traces" from to a man
called Paul and gotten by him through direct perception in the piazzas and
by the canals. I see these as sheer logical constructions tagged misleadingly
to suggest kinship with an existing species of phenomena. None of these fa-
miliar constructions has anything—verbal excesses aside—to do with the
real, cognitive process of memory. I am not merely expressing doubt about
the genuineness of the possibility of applying such a supposedly coherent no-
tion of memory to beings like us. I question the very coherence of the idea
as a notion of memory for any historically real, existing species and/ or for
human beings. I do not pursue here these grave reservations. In the text, I
focus on the use of this one technology in the current literature that is (i)
relatively informal and (ii) pertinent to the sustenance of separatist dualist
theses.
THE REAL M A N 115
3.2. T h e P r i m a l Q u e s t i o n a n d the P r i m a l A n s w e r
POLYANDER: Y o u d i d n o t w a n t t o ask a n y t h i n g w h i c h
c o u l d n o t be answered v e r y easily. So I shall say I am
a man.
E U D O X U S : Y o u are n o t p a y i n g attention t o m y ques-
t i o n and the r e p l y y o u g i v e , h o w e v e r simple i t m a y
seem t o y o u , w o u l d plunge y o u i n t o v e r y difficult and
complicated problems, were I t o press y o u even a l i t -
tle. I f , for example, I were t o ask even Epistemon
w h a t a m a n is, and he gave the stock r e p l y o f the
scholastics, that a m a n is a " r a t i o n a l a n i m a l " and, if,
i n order to explain these t w o terms ( w h i c h are just as
obscure as the f o r m e r ) , he were t o ask us further,
t h r o u g h all the levels w h i c h are called " m e t a p h y s i c a l " ,
w e should be dragged i n t o a maze f r o m w h i c h i t
w o u l d be impossible t o escape. F o r t w o questions
arise f r o m this one. First, w h a t is an animal} second,
w h a t is rational? i f i n order t o explain w h a t an animal
is, he were t o reply that i t is a " l i v i n g and sentient be-
i n g , " that a l i v i n g b e i n g is an "animate b o d y " and
that a b o d y is a " c o r p o r e a l substance", y o u see i m m e -
diately that the questions, l i k e the branches o f a f a m -
i l y tree, w o u l d r a p i d l y increase and m u l t i p l y . Q u i t e
clearly the result o f all these admirable questions
w o u l d be pure verbiage, w h i c h w o u l d elucidate n o t h -
i n g and leave us i n o u r o r i g i n a l state o f ignorance.
(CSM I I , 410) 7
125
i26 INDEX
prove too much/prove too little existence i n , 4, 25, 39, 60, 106
dilemma as apart, 99—103
integrative interpretation of, existential separability of, 5-8,
115-123 70, 101, 115
in separatist dualism, 59—61, generic versus tight integration
66 of, 109—112
Pythagorean theorem, 14, 17, 19 man-mind swaps i n , 103—104,
whatness and, 28, 34 107-109, 112, n3n.5
of mind, 80, 98
quasi-demonstrative representa- necessary connections and,
tions, i n de dicto prem- 52n.15 *"
ise, 56n.18 observations of, 52n.15,56, 101
projection method for, 10—13,
rational animal, man as, 117—123 11n.2, 13n.4, 18, 39
real conceivability, seeming con- simpliciter, 70
ceivability versus, 19, 62, of specific human mind, 112—
66n.1 115, 113n.5
real conceivability projection, real imaginability, seeming imagi-
10—13, 11n.2, 13n.4, 18 nability versus, 49—51,
real distinction 52n.15, 53, 58, 114
Arnauld's notion of, 12n.3, real man
13n.4 primal answer of, 116—123
basis for, 3—7, 70 primal question of, 115—116
of bodies primality of, xviii, 60—61, 66,
generic, 75 71, 76-80, 96-97, I I I
given kind, 77 real possibility-imaginability,
categorical, 70 49-51
coherent stories i n , 8—10, 12¬ real possibility projection, 10—13,
13, 12n.3, 98n.16 12n.3, 13n.4, 39
complete subjecthood i n , reality
99-103 conceivability of, 21—23
conceivable separability i n , necessary connections and,
5-6, 10-11, 59, 101, 52n.15
103-104, 107 in real distinction, 12—13,
Descartes's meditations on, 4, 12n.3
8-10, 22, 28-29, 38, reality-bound representations, o f
90-91 pain, 47-48
disembodiment illusion i n , reality-projection principle,
23n.7, 103—107, 109 21—22
INDEX 137