<< . .

. 18
( : 30)



. . >>

selves as absolutely unitary (as simple), that is, as individual egos. For we
±·
The pseudo-discipline of transcendental psychology
are conscious of ourselves as subjects of self-consciousness, as egos, that
cannot be a mere collection of di¬erent subjects. () We are conscious of
ourselves as being numerically identical selves in all our experiences of
which we are conscious. For this is what it means to say of all these
experiences that we are conscious of them as our own. Finally, () we are
conscious of ourselves as being in a cognitive relation to possible objects
that are external to us. For it is only in relation to these external objects
that we think of ourselves as distinct individuals.
Kant argues that there are four aspects of self-consciousness that give
rise to four natural fallacies concerning the nature of the self. It is these
fallacies that underlie traditional metaphysics of the soul (self ). It seems
possible to infer the substantiality, simplicity, identity over all time, and
independence of the self, from an analysis of the proposition ˜˜I think,™™
since this proposition is a condition under which we ascribe thoughts to
ourselves and others, and thus think of each other as thinking beings
endowed with a self.
Kant diagnoses the four basic paralogisms of rational psychology as
fallacies of ambiguity (sophismata ¬gurae dictionis:  °;  ±±) that are
based on a confusion of the problematic use of the proposition ˜˜I think™™
with the assertoric use of that proposition. The major premise of the
syllogism lays out a de¬nition of what it is to be a substance, a simple
thing, a person, and a thing the existence of which is doubtful. Now, in a
certain sense ˜˜the proposition that expresses self-consciousness: I think™™
( ) refers to something that satis¬es the de¬nition of what a sub-
stance, etc. is. For Kant insists that ˜˜the mere apperception (I) [is]
substance in concept, simple in concept, etc.™™ ( °°). In other words,
apperception is the expression of the proposition ˜˜I think.™™ And this
proposition ˜˜I think™™ analytically entails that I am the absolute subject
of my thoughts, that I am a single subject that has an intrinsic unity to it
that I can call simplicity, that I am numerically identical over my
di¬erent thoughts, and that I can distinguish myself as thinker from
other things outside of my thought (see especially  °·¬.). Thus, the
minor premise of the syllogistic inferences that Kant ascribes to rational
psychology seems to be able to draw on the self-verifying and empirical
claim that I am thinking in its claim that I have an existence that
corresponds to the de¬nition of substance, etc. The conclusion is then
the claim that I am a substance, etc. However, the minor premise draws
its support for the claim that I am a substance, etc. from the way I must
think of myself and think of others in order to be able to think of us as
thinkers at all. In other words, the minor premise depends on the
± Kant and the demands of self-consciousness
problematic use of the expression ˜˜I think,™™ which does not support
bona ¬de existential claims, while presenting itself as based on the
self-verifying status of the assertoric use of the expression ˜˜I think.™™ The
subtle mistake that underlies transcendental psychology is to conclude
from the fact that the formal de¬nition of substance, etc. is satis¬ed by
the way in which each of us is compelled to think of him- or herself and
others that each of us therefore knows that he or she and other thinkers
are substances, etc.
The second edition of the Paralogisms clari¬es the nature of the
fallacy of ambiguity involved in the paralogistic inferences of transcen-
dental psychology:
Thought is taken in both premises in a totally di¬erent meaning: in the major
proposition as it applies to objects in general (hence as it may be given in
intuition); but in the minor proposition as it subsists in relation to self-conscious-
ness, whereby no object is thought, but only the relation to oneself :Sich
[sic]9, as subject (as the form of thought) is represented. ( ±±)

The major premise applies to all thinking beings regardless of how
they are thought to be. It is explicitly said to include thinking beings that
are given to us as objects of knowledge through intuition. It is a premise
that is analytically true, since it merely analyzes the meaning of ˜˜sub-
stance,™™ ˜˜simple things,™™ ˜˜personal identity,™™ or ˜˜external things.™™ In
the minor premise, thinking beings are taken only as they are available
to us subjectively through our thoughts of ourselves. Due to the in-
eliminability of the ¬rst-person point of view, it is then tempting to infer
that this description applies to oneself and others not just as we must think
of ourselves, but without restriction.
In e¬ect, the Paralogisms are based on running together two di¬erent
senses in which thoughts about thinkers may be abstract. In the ¬rst
case, thoughts are abstract in the sense that they apply to objects in
general, absolutely all objects. This is the sense required in order to
make substantive claims about thinking beings. This is a ˜˜transcenden-
tal use™™ of the notion of thought, that is, a use of something that serves as
a condition for our experience as if the object that conformed to that
condition could be given independently of our experience. It is an
illegitimate extension from a de¬nitional or purely logical use of
thought. But, in ascribing thoughts, we are abstracting from facts about
particular thinkers in a di¬erent way. We are thinking of them in terms
of the conditions under which we ascribe thoughts to them. This does
not license us to make claims about the way thinkers are in general, and
±
The pseudo-discipline of transcendental psychology
in abstraction from the way we must think of them. Thus, it does not
underwrite claims about how thinkers are in themselves.
The subject as represented in self-consciousness is not an object in the
proper sense at all. It makes no sense to say that one might fail to be
conscious of oneself, since this self is an internal accusative of self-
consciousness. It does, however, make sense to say that one might fail to
know who one is, as one can see from cases of amnesia. When I am
conscious of myself merely as subject of self-consciousness, there is no
room for reference failure. It is as some particular individual (as an
object) that ˜˜I™™ can fail to refer to myself. Since such an object must be
identi¬able, the identi¬cation of who the subject is, as a certain particu-
lar (object), may fail. The ˜˜I™™ serves as its own formal object, but when I
think of myself as this formal object I am actually abstracting from all
objects in the normal sense of the word. Although ˜˜I™™ or rather ˜˜me™™ (its
accusative form) serves as the grammatical object of consciousness, it
need not correspond to any publicly identi¬able or reidenti¬able ob-
ject. Non-identi¬catory self-consciousness does not provide access to
the self under any particular description. If one is self-conscious, one
must refer to oneself. One also has a belief about a particular object. But
the object of one™s belief may not be the actual bearer of self-conscious-
ness and thus it may not be the self to which one is referring in
self-consciousness.

  ¦©   °  ¬  §©  
The second edition of the Critique provides only one explicit example of a
paralogism, which apparently serves as a paradigm for how the others
are to be reconstructed. It involves the notion of substance: (±) ˜˜That
which cannot be thought other than as subject, exists in no other way
than as subject and is therefore substance.™™ () ˜˜Now a thinking being
merely regarded as such cannot be thought of other than as subject.™™ ()
˜˜Therefore it exists also only as such, i.e. as substance™™ ( ±°“±±). The
example displays the shift in scope that Kant takes to be characteristic of
a paralogism in the second edition. We begin with an analytic principle
that is little more than a nominal de¬nition of substance. In the minor
premise, we move to a principle expressing what is involved in our
conception of a thinking being as such. However, in the conclusion, the
restriction in the minor premise to the thought of a thinker as a thinker is
ignored. Ignoring the restriction leads to the assertion of a synthetic
proposition a priori on the basis of a de¬nitional claim about the nature
±·° Kant and the demands of self-consciousness
of substance and a claim about what the conditions are under which we
think of ourselves and others as thinkers.
The basic problem with the argument is that it involves a fallacy of
ambiguity. We begin with the idea that something that can be made
sense of only as a subject, that is as a bearer of properties, is a substance.
Then the Cartesian premise is introduced that a thinking being thought
of as a thinking being can only be thought of as a bearer of thoughts. The
restriction in the minor premise to thinking beings thought of as thinking
beings is then dropped. Dropping the restriction leads to the fallacious
conclusion that a thinking being can only be regarded as a subject of
thought or bearer of thoughts. The problem is that a being might be a
thinker without being essentially a thinker. Thinking might be only a
derivative and non-essential property of the actual individual who
thinks. The basicness of I representations to thought encourages one to
assume that I am also ontologically basic as a thinker. The inference to
my being ontologically basic is based on confounding the way I must
represent myself as subject of judgment and of thought with the actual
bearer of I thoughts.
Kant notes that ˜˜I™™ is a singular term that cannot be used as a
predicative term. The fact that ˜˜I™™ can only be used as a singular term
suggests that ˜˜I™™ refers to a substance, since a substance is precisely what
one refers to by means of a singular term that is not predicated of other
things, as general predicative terms may be. The ˜˜I™™ as subject of
thought is a basic singular representation which cannot serve as a
general representation that could be predicated of some more basic
particular. Since the logical subject of the judgment is, in this case, a
thinker, ˜˜I as thinking being,™™ it seems to be licit to move from the
logical basicness of the I representation to an assertion of the ontological
basicness of me as a thinker. For here it is not just a linguistic expression
which is the subject of the judgment, but a thinker, the subject of
self-consciousness.
The inference in question is based on a use-mention fallacy. For it is
still not acceptable to move from what is true of a representation of self
in self-consciousness and thought to a claim about the bearer of that
self-consciousness. I thoughts are logically basic as representations of
whatever it is that is doing the thinking, but one cannot legitimately infer
from this logical basicness of I thoughts that thinkers qua thinkers are
basic particulars. Thus, while it is true that I must represent myself as a
substance or basic particular when I think of myself as someone that
thinks, it does not follow from the fact that I must represent myself as a
±·±
The pseudo-discipline of transcendental psychology
subject that I know any substantive facts about the nature of the
individual to which I refer in thought.
There is a trivial sense in which the self or thinker qua thinker may be
said to be a substance. The self is an absolute subject of thought in the
sense that it is whatever is represented as the bearer of thought. But this
sense in which the self or thinker ¬ts the de¬nition of a substance, is an
artefact of the ¬rst-person point of view:
Now in all our thought the I is the subject, in which thoughts only inhere as
determinations, and this I cannot be used as the determination of another
thing. Therefore everyone must necessarily regard himself :Sich [sic] selbst9
as substance, and thought only as accidents of his existence, and determinations
of his state. ( )

Since I represent the ultimate bearer of thought by means of the
expression ˜˜I think,™™ it is tempting to think that I as thinker must be the
ultimate bearer of my thoughts. From this it would seem to follow that I
as substance must be a thinking being. According to this non-trivial or
substantive sense of substance, the real bearer, represented by ˜˜I,™™ is
essentially characterized as an ego or thinking substance. The important
thing to note is that ˜˜I ™™ who I think must be valid as subject and be
regarded as something that cannot belong to thought merely as predi-
cate is an apodeictic and even identical proposition; but it does not mean
that I as object am a being subsisting for myself alone, or substance™™ ( °).
In his Second Meditation, Descartes notoriously infers that the soul is
a substance in the metaphysically interesting sense from the fact that it is
a ˜˜substance™™ in the trivial one of being a self-conscious representer that
is certain of its existence. Descartes argues from the certainty of cogito
statements that thought is an essential property of me as a thinking
being. He also argues independently that thought is not an essential
property of body. From this he deduces that I as a thinking being have a
property that my body lacks. This is the property of being essentially a
thinker. And from this he concludes that there is a real distinction
between me and my soul on the one hand, and my body, on the other.
In the Search for Truth, Descartes develops the argument as follows:
Indeed, I do not even know whether I have a body; you have shown me that it is
possible to doubt it. I might add that I cannot deny absolutely that I have a
body. Yet even if we keep all these suppositions intact, this will not prevent me
from being certain that I exist. On the contrary, these suppositions simply
strengthen the certainty of my conviction that I exist and am not a body.
Otherwise, if I had doubts about my body, I would also have doubts about
±· Kant and the demands of self-consciousness
myself, and I cannot have doubts about that. I am absolutely convinced that I
exist, so convinced that it is totally impossible for me to doubt it.

Kant™s objection to the Cartesian claim about the nature of the soul is
that it moves from the way one ˜˜must necessarily regard™™ oneself to the
way one must be independently of being so regarded, that is, it shifts
between a subjective necessity concerning the way things must be
described by us to a de re necessity about the way things themselves must
be. The self must be regarded as a substance, but in the sense that the
ego to which one ultimately refers must be regarded as the real bearer of
one™s thoughts. Self-consciousness per se tells us nothing about the nature
of the metaphysical relation between the ultimate bearer of thought and
the representation ˜˜I™™ which we use to refer to that bearer of thought.
For, in self-consciousness, I abstract from any knowledge of who the real
bearer of my thoughts is, and regard myself purely as the point of view to
which I ascribe di¬erent representations. The temptation, however, is to
think that because one can think of a subject of thought in abstraction
from its particular representations that one therefore has a grip on some
real bearer that exists independently of all the representations that we
may ascribe to it:
I think of myself for the sake of a possible experience by abstracting from all real
experience and infer that I can be conscious of my existence also outside of
experience and its empirical conditions. Therefore I con¬‚ate the possible
abstraction of my empirically determined existence with the alleged conscious-
ness of a separate existence of my thinking self, and believe that I know the
substantial in me as the transcendental subject, while I merely have the unity of
consciousness in thought that underlies all determination as mere form of
cognition. ( ·)



     ©  ¤ °  ¬ § © 
In the Third Paralogism, Kant makes use of the relation of the mental to
the physical as it presents itself to our experience from the ¬rst- and the
third-person points of view to call into question traditional arguments
for the persistence of the soul.µ The ¬rst-person point of view of self-
consciousness presents one to oneself as a thinker endowed with numeri-
cal identity over the times of which one is conscious. This seems to
satisfy the de¬nition of a person endowed with numerical identity. It is
an analytic truth that, if one ascribes a sequence of beliefs or other
representational states to oneself, one also believes that one is ascribing
±·
The pseudo-discipline of transcendental psychology
those representations to one and the same self. This is a feature of
¬rst-person self-reference, but it is dependent on taking a ¬rst-person
point of view. Certainly, one must take a ¬rst-person point of view, but
there may well be competing ¬rst-person points of view. The fact that
these points of view compete with one another is disclosed by di¬erences
in the content of what they represent. Each of us represents the world
from a di¬erent sequence of spatio-temporal positions. The di¬erent
ways in which things appear to those positions give rise to competing
conceptions of the unity of experience.
To think of oneself as a self, one must take oneself to be the same self
through whatever set of experiences one ascribes to oneself. There has
to be a single subject that represents a series of experiences to itself if
those experiences are to be represented as part of a single experience:
˜˜On this basis the personality of the soul would have to be [mußte] ¨
regarded not even as inferred, but as a completely identical proposition
of self-consciousness in time™™ ( ). The basis in question is the way I
relate all my inner states to my self.
With rationalist philosophers, Kant de¬nes the soul as that which is
conscious of the numerical identity of its self through di¬erent times.
This provides the major premise for a syllogism. The minor premise of
the syllogism then takes the soul to be a person as de¬ned in the ¬rst
premise, but the soul must be a person in this sense only from the
¬rst-person point of view. The fallacy of ambiguity involved in the Third
Paralogism is again to take what is true from the ¬rst-person point of
view and treat it as if it were true from any point of view, that is, as if it
were true absolutely. This leads to the conclusion that the soul is a
person in the sense of a particular that persists through di¬erent times
and is conscious of that persistence.
The analytic truth that I must be conscious of my identity through all
the states of which I am conscious is not clearly distinguished by Kant
from another idea. This is the idea that one must be able to conceive of a
standpoint that is not itself being successively replaced by some other
standpoint in order to have a conception of succession itself. It is one
thing to say that I am conscious of my identity through all the times in
which I am conscious of myself. It is quite another for me to assume that
there is a numerically identical self in all time: ˜˜I relate any and all of my
successive determinations to the numerically identical self in all time,
that is in the form of the inner intuition of myself™™ ( ). Kant o¬ers
no argument for this strong claim.
There is an argument available to Kant for the strong claim that
±· Kant and the demands of self-consciousness
suggests itself from the First Analogy and its connection with the
Transcendental Deduction. Succession requires one item to succeed or
replace another item relative to something else that is not itself undergo-
ing replacement. Thus, if one is to be conscious of succession, one needs
to be conscious of a point of view that is not replaced in that succession.
If I am to be conscious of successive states as successive, I must be able to
order these states in time. But, in order to be able to set up a temporal
ordering, I must be able to represent the temporal relations that consti-
tute the form of inner representation in a way that purports to be
independent of the way things merely appear to me now. I must be able
to represent temporal relations in tenseless terms. To do this, I must be
able to think of myself as being able to shift from one possible temporal
position to the other.
To perform the kind of transference of the ¬rst perspective required
to ascribe rationality, my consciousness of myself must, in a certain
sense, be immune to reference failure by virtue of misidenti¬cation, that
is, in being conscious of myself it cannot happen that I fail to refer to
myself, and instead refer to someone else. This immunity to reference
failure allows me to contemplate alternative possibilities of how my life
might have gone or even of who else I might have become. But this
immunity to reference failure extends only to the use of ˜˜I™™ to refer to
my representation of myself as a thinker, it does not guarantee my
personal identity in a sense that satis¬es third-person criteria.
I may be constrained to think of myself as identical through the
experiences that I ascribe to myself, but my individual personal identity
as it exists through a series of representational states is clearly something
that can only be known from experience, indeed, through knowledge of
my inner states underwritten by the identity of my human body over
time: ˜˜the persistence of the soul, as a mere object of inner sense is
unproved, and even unprovable, although its persistence in life is for
itself clear, since the thinking being (as human being) is also an object of
outer sense™™ ( ±µ). Personal identity is not something that can be
inferred from the kind of self-consciousness expressed in the statement
˜˜I think™™ alone, even though I can legitimately claim that I am con-
scious of my identity as a subject through all those experiences of which
I am conscious of myself ( °):

The proposition of the identity of myself in respect to all manifolds of which I
am conscious is also a proposition that lies in the concepts themselves, that is, an
analytic proposition; but this identity of the subject of which I can become
±·µ
The pseudo-discipline of transcendental psychology
conscious in all its representations does not pertain to the intuition of that
subject as it is given as an object, it can therefore not also mean the identity of
the person for which to prove it the mere analysis of the proposition I think is
not enough, but di¬erent synthetic judgments that are based on given intuition
would be required. ( °“°)

The necessity of the possibility of the kind of self-transferral involved
in my being able to think of myself from di¬erent temporal positions can
easily seem like an argument for the necessity of my actually existing at
all of these di¬erent times. As an experiencer, my numerical identity
would have to be taken as a basic given of experience. This leads to the
temptation to argue from the ¬rst-person phenomenology of personal
identity and survival to claims about the survival of my soul. From the
¬rst-person point of view of my time-consciousness, it makes no di¬er-
ence whether I think of the time of my experience as belonging to the
unity of my experience or whether I think of the unity of my experience
as belonging to the unitary time in question:
For it really says nothing more than that in the whole time in which I am
conscious of myself, I am conscious of this time as belonging to the unity of my
self and it is the same thing whether I say: this whole time is in Me as an
individual unity or I am to be found in all this time with numerical identity.
Personal identity is therefore unfailingly to be found in my own conscious-
ness. ( ±“)

As long as we restrict ourselves to the ¬rst-person point of view, there
is no distinction to be drawn between the way the di¬erent states of my
life present themselves to me and the way that they really are. There is
therefore no distinction to be drawn between the temporal order in
which my di¬erent experiences present themselves to me and the actual
temporal order of those di¬erent experiences. I am conscious of my
numerical identity through the time in question because that time is the
time of my time-consciousness. No distinction has yet been drawn
between subjective and objective time, or between my subjective time
and the subjective time constituted by another point of view.
Kant argues plausibly that the possibility of a second- or third-person
perspective must become apparent to me before I can have a grip on the
distinction between the temporal order in which things present them-
selves to my self-consciousness and the temporal order in which they
actually occur. It is only if one can conceive of a perspective di¬erent
from the one which one actually has that one can understand the
relativity of one™s own perspective. One™s very understanding of one™s
±· Kant and the demands of self-consciousness
own personal identity thus depends on one™s ability to perform the kind
of feat of transference in terms of which one understands other persons.
From the purely ¬rst-person perspective, there does not seem to be
any way of drawing a distinction between one™s representation of one™s
identity over time and one™s actual identity over time. In order to get a
feel for this distinction, one must be able to shift from the ¬rst- to the
second- or to the third-person perspective. Inner experience is, for
Kant, essentially temporal. Mental states are constantly being replaced
by other mental states. In immediate experience, no distinction is
needed or drawn between a successive consciousness of inner states and
a consciousness of inner states in succession. Although I relate all my
successive states to a numerically identical self, it is not clear what is
numerically identical here, nor can it be until I have been able to draw
the distinction between the perspective of a representer and what is
represented by the representer. I do not have full-blown consciousness
of myself as an object of temporal experience until I am able to draw
that distinction.
This is why I need to be able to take the perspective of an external
observer in order to place myself in time. The vantage-point of the
external observer is thus one which I myself must occupy in order to be
aware of myself as an object of inner sense: ˜˜If I regard myself from the
perspective of another (as the object of his external intuition), then this
external observer considers me for the ¬rst time in time, for in apperception
time is only actually represented in me™™ ( ).· Kant makes the point
somewhat less explicitly in the second edition: ˜˜Neither can the subject in
which the representation of time has its original ground, determine its
own existence through that [pure self-consciousness]™™ ( ).
By taking up the perspective of a second person I am forced to
represent myself in a di¬erent way. I have ¬rst-person access to my own
representations, but another person™s access to my inner states depends
on inference. Such inferences are based on my outer states. This is why
the states in question are considered to be inner and outer respectively.
We represent outer states spatially, just as we represent inner states

<< . .

. 18
( : 30)



. . >>