1/21

Gary Banham
Apperception and Spontaneity (2008)

Apperception and
Spontaneity

Gary Banham, Editor, Kant Studies Online

n 16 of the B-Deduction Kant distinguishes between
two forms of representation indicating that whilst
intuition is given prior to all thought that its manifold is
nonetheless necessarily connected to the I think. The
representation of the I think is however then determined in
two ways, which are presented as having a relation to each
other. On the one hand, this representation is determined as
an act of spontaneity although this determination is not here
clearly defined with Kant simply stating that the repre-
sentation of the I think cannot be regarded as something
that belongs to sensibility. So this initial connection of the I
think to spontaneity appears to involve only a negative
characterization of spontaneity. On the other hand, this
representation of the I think is connected to a form of self-
consciousness that is explicitly distinguished from anything
empirical. This form of self-consciousness is termed pure or
original apperception as it is a self-consciousness that is said
to be responsible for generating the representation of the I
think. So, on the one hand, the I think is related to
spontaneity and described as an expression of it whilst on
the other it is presented as connected to pure apperception
where the latter is said to be what generates it. Now, whilst
the understanding of Kants account of apperception has
been a subject of some interest for philosophers, its
relationship to spontaneity and the I think, as explicitly
made clear at B132, has been rarely commented on. One of
the reasons why this larger agenda has been rarely attempted
are well expressed by Robert Pippin. As Pippin writes:
relying on Kants discussion of apperception to explain his
characterization of thinking as spontaneous can appear a
I
2/21
Gary Banham
Apperception and Spontaneity (2008)

classic case of the obscure through the more obscure.
1
My
investigation will differ from Pippins however as what
interests me is not the attempt to explicate spontaneity with
the help of apperception but rather the relationship both have
to the representations of the I and the I think, in the
process thinking through more carefully the question of how
to view the connection between acts of spontaneous
representation and the manifold of intuition.
2

In his revision of the Paralogisms in the second edition of
the Critique Kant states in response to the claim that the I
is a simple subject that apperception is something real and
that its simplicity is already given in the mere fact of its
possibility (B419). Kants suggestion that the simplicity of
apperception is a consequence of its possibility requires
more elaboration. In his parallel treatment of the Second
Paralogism in the first edition Kant uses this point to frame a
clear logical understanding of apperception: I am simple
means nothing more than that this representation, I, does
not contain in itself the least manifoldness and that it is
absolute (although merely logical) unity (A355).
3
In the A-
version of this discussion Kant speaks of the I here as
referring only to something in general and thus as being
indeterminate. On this basis he dismisses the claim that we
have a sense of the I as a simple subject stating instead that

1
Robert Pippin Kant on the Spontaneity of Mind (1987) in Robert Pippin
Idealism As Modernism: Hegelian Variations (Cambridge and New York:
Cambridge University Press, 1997), pp. 41-2.
2
I should add however that the main point of this piece is to focus on
precisely how much we can say about the relationship between apperception,
spontaneity and the I think independently of the relation either has to
intuition. Because of this focus the discussion of the connection to the
manifold of intuition will come late and be only presented in the form of a
sketch. This paper is the first of a group and only at a subsequent stage will
the fuller sense of the relationship of apperception to intuition become the
main focus when elaborating on the sense of the objective unity of
apperception that is discussed in 18 of the B-Deduction. This would be the
topic of a separate paper as would concentration on the necessary unity of
apperception.
3
Kant at the same point adds a connection between this logical understanding
of apperception and the Cartesian cogito, a connection based on his view that
the cogito states a tautology, a view earlier presented by Gassendi, but always
repudiated by Descartes.
3/21
Gary Banham
Apperception and Spontaneity (2008)

we have a simple representation, which does not entail the
simplicity of what is represented. Whilst connecting the
statement from the first edition here to that from the second
does clarify the latter it also complicates the question of how
to understand the claim that apperception is something
real. If we understand apperception as involving a merely
logical unity then in what sense is it something real?
One solution to this problem is to distinguish between
empirical and transcendental apperception in the manner
pioneered by Henry Allison.
4
In the A-Deduction for
example, after completing the preliminary statement of the
synthetic discussion given there, Kant describes
apperception as one of the three subjective sources of
knowledge. There is then a statement to the effect that these
sources can be viewed as both empirical (in application to
appearances) and as a priori elements of knowledge, which
make experience possible. However, after distinguishing
these senses Kant proceeds to explain each of the three
empirically, giving the following characterization of apper-
ception: apperception in the empirical consciousness of the
identity of the reproduced representations with the appear-
ances whereby they were given, that is, in recognition
(A115).
On the basis of this description of the empirical
application of apperception to appearances Allison makes
the following distinction between transcendental and
empirical apperception: the contrast Kant really needs to
draw is between a consciousness of the activity as it
functions determinately with a given content and a thought
of the same activity, considered in abstraction from all
content. To regard apperception in the first way is to
consider it empirically, and thus as something real; to
consider it in the second way is to consider it transcenden-

4
Henry E. Allison (1983) Kants Transcendental Idealism: An Interpretation
and Defense (Yale University Press: New Haven and London), Chapter 13.
The reference to viewing transcendental apperception as involving
abstraction from all content is clearly based on A355.
4/21
Gary Banham
Apperception and Spontaneity (2008)

tally, and thus as a transcendental condition of experience.
5

This reading thus squares the two statements from B419 by
viewing the simplicity that arises from the mere possibility
of apperception as a consequence of the logical unity that we
term the transcendental unity of apperception whilst the
reality of apperception is taken instead to refer to the
awareness of determinate content that we call empirical
apperception. Whilst this latter is often confused by Kant
with inner sense it can be distinguished from it as
including a reflective act of introspection.
6

Whilst Allisons account of the distinction between
transcendental and empirical apperception appears to
provide a response to the passage from B419 we shall
discover that the distinction advanced is more difficult to
sustain when it is related to our opening passage from B132
in which Kant connects transcendental apperception to both
spontaneity and the generation of the representation of the I
think. Prior to making the claims with which I opened my
discussion Kant opens 16 by the famous declaration
concerning the necessity of the I think accompanying all
my representations which Allison views as constituting the
real starting point of the B-Deduction as it states the
principle on which the first part of its argument is said by
him to be based.
7
Regardless of views on this matter
however the point of interest for this investigation turns
rather on the way in which the statement of the principle said
here to be introduced is assessed.
Allison works on the assumption that the I think is
identical with what he terms the apperception principle, an

5
Allison (1983) p. 274..
6
Udo Thiel, in contrast to Allison, attempts to make the distinction between
transcendental and empirical apperception on the basis of the B-Deduction
though in the B-Deduction Kants claim concerning empirical apperception is
merely that it accompanies different representations (B133) and hence
lacks the reflective component that Allison generates from A115. See U.
Thiel (2006) The Critique of Rational Psychology in Graham Bird (ed.)
(2006) A Companion to Kant (Blackwell: Oxford and New York), p. 210
7
There are a number of reasons for finding this claim less obvious than
Allison seems to take it to be. Interestingly this statement is added in Allison
(2004) (p. 163) and is not in the first (1983) edition of his book.
5/21
Gary Banham
Apperception and Spontaneity (2008)

identification we shall later question. Firstly, however, I
wish to look at the manner in which the principle, as viewed
by Allison, is determined. The key point is that whilst
Allison views Kant as claiming that the principle in question
is a single principle, Allison himself does not take it to be so.
It is, rather, on Allisons view, multiple. The statement that it
must be possible for the I think to accompany all my
representations is taken to apply to each of my
representations taken individually
8
. Any given represent-
ation, to be taken to be mine, would thus have to meet the
condition of being able to be thought as mine. Allison views
this claim as equivalent to the possibility of reflectively
attaching the I think to the representation in question.
This point is distinguished from the way that the I think
relates to the collective unity of a subjects representations as
in the case of a complex thought.
9

This point about the unity of a complex thought is what
requires an elaboration of Allisons conception of the
transcendental unity of apperception in such a way as to
make his contrast between transcendental and empirical
apperception more difficult to sustain than appears at first
sight. The account of empirical apperception on which
Allison relies is that stated at A115 in which Kant describes
it as involving an empirical consciousness of the identity of
reproduced representations with the appearances whereby
they were given. Since the reference to appearances here
explicitly requires intuitions it would appear that empirical
apperception has a manner of individuating representations
by means of them. By contrast the indeterminate
representations to which transcendental apperception is
directed would seem to include no evident principle of
individuation. In response to this point Allisons account
moves from the discussion of B132 to that of A108 in order
to present a discussion of the manner of recognition that

8
Allison (1983) (p. 137), slightly emended in Allison (2004) to each of a
subjects representations taken individually (p. 163).
9
This is a point where Allison (1983) (p. 138) explicitly invokes B407 which
is the second edition statement of the 2
nd
Paralogism.
6/21
Gary Banham
Apperception and Spontaneity (2008)

reflective awareness of transcendental apperception gives
rise to.
Allison addresses this in the following way:

because of the contentlessness of the I think, there is
literally nothing, apart from the consciousness of the
identity of its action (in thinking a complex thought),
through which the thinking subject, considered as such,
could become aware of its own identity. Expressed
schematically, the consciousness of the identity of the I
that thinks A with the I that thinks B can only consist in
the consciousness of the identity of its action in thinking
together A and B as its representations. That is why a
consciousness of synthesis (considered as activity as well
as product) is a necessary condition of apperception, even
though the latter requires merely the possibility of the self-
ascription of ones representations.
10


Allison has here combined together a number of different
conceptions of transcendental apperception. Firstly he
presents the view that consciousness of the I think is
consciousness of the identity of action in thinking a complex
thought. This consciousness of identity of action as the basis
of awareness of thinking a complex thought could be
assimilated to his earlier view of it as involving merely
logical unity. However he here adds to this characterisation
two others that are not obviously connected to this one,
namely the sense of synthesis (which is itself presented in
two distinct ways in this quote) and the view of apperception
that we can term a possibility conception of it.
The first question to raise about Allisons view concerns
the manner of connection he is suggesting here between the
contentless I think and the consciousness of synthesis of

10
Henry E. Allison (2004) Kants Transcendental Idealism: An
Interpretation and Defense Revised & Enlarged Edition (Yale University
Press: New Haven & London), p. 171. In this second edition Allison is here
clearly expanding on the earlier account of transcendental apperception
though interestingly his discussion of empirical apperception is very much
briefer.
7/21
Gary Banham
Apperception and Spontaneity (2008)

different thoughts together in an awareness of identity. The
emphasis on the contentlessness of the I think is connected
to the simplicity of its representation, in as Kant put it, the
thought of a bare I. However this representation of a bare
unity that Kant takes to be the basis of the conception of the
subject as simple has to be distinguished from the principle
of identity that would be the ground of any analytic unity of
consciousness. Allison, in moving from the contentless I
think to the consciousness of the identity of the action of
the I think elides the distinction between simplicity and
identity. Whereas the simple representation of the I think
requires no more than a presentation of its bare unity at any
given point in time, the assertion of an awareness of identity
is, by contrast, related to a grasp of distinct acts of awareness
at different moments of time and it is this latter that we can
see to be at work with empirical apperception. Hence in
moving from the simple representation to the awareness of
identity that requires temporal indexing Allison fails to
maintain his distinction between transcendental and
empirical apperception.
11

Not only does Allisons discussion require a movement
from the sense of simple representation to the identity of
intra-temporal awareness there is also something curious
about the nature of the connection he suggests between the
consciousness of synthesis and the understanding of
transcendental apperception. Consciousness of synthesis is
described in two different ways, as an activity and as a
product. Whilst the activity sense of synthesis involves the
notion of combination of discrete elements into a unity, the
product sense requires the sense of this unity to be graspable

11
It is necessary to be careful with this point. I am suggesting two distinct
things here. Firstly, Allisons earlier distinction between empirical and
transcendental apperception only allowed temporal indexing with regard to
empirical apperception so that the reference here to the need for some form
of it at the transcendental level introduces complexity that is not included in
Allisons basic account. Secondly, the relationship to temporality would have
to be different in the two cases and the nature of the difference is not
described and this means there is an implicit elision of transcendental and
empirical apperception.
8/21
Gary Banham
Apperception and Spontaneity (2008)

in principle. However Allison suggests that both these
senses of consciousness of synthesis are a condition of
apperception despite the fact that apperception is itself the
basis of the mere possibility of self-ascription of
representations. If the combination that is at work in the
synthesis is, even in the act sense of synthesis, something
that requires a relation to the possible representation of an I
think then it becomes unclear how this act, let alone its
product, can also be a condition of the I think as it must
surely rather be an effect of the I think.
We might also raise the question of how the awareness of
the act of thinking a complex thought can itself be given?
This implies a description of a kind of reflective act in which
the combination of elements is itself made available to some
possible consciousness. That would imply a reflexive
conception not merely of empirical apperception but also of
transcendental apperception and such a reflexive interpret-
ation of transcendental apperception is quite frequently
presented. This implication of Allisons account is not
drawn out by him and one reason for this would be that on
its elementary description it appears insufficient to provide
the sense of unity required for transcendental apperception.
Simply reflecting upon awareness of discreet particulars
does not provide in itself a connection between them and
certainly gives no sense of necessary connection.
12

Kant however would appear to have at times thought
otherwise and there are grounds for thinking that one of the
senses of transcendental apperception that he occasionally at
least endorsed involved a view of reflective consciousness of
identity such as is implicitly referred to by Allison and
which appears to be given in two passages he cites. These
would be A108 where Kant states: the mind could never
think its identity in the manifoldness of its representations,
and indeed think this identity a priori, if it did not have
before its eyes the identity of its act and we could also refer

12
Allison is indeed clearly aware of this point as is evidenced in the
aforecited reference to the second edition account of the second paralogism.
9/21
Gary Banham
Apperception and Spontaneity (2008)

here to the similar claim at A116 where we have the
following: We are conscious a priori of the complete
identity of the self in respect of all representations which can
ever belong to our knowledge. The first citation suggests
that the identity of the manifold of representations depends
on a prior sense of the identity of the act that is required for
synthetic combination itself to be possible and hence
provides the reversal of Allisons position on the relation
between act of synthesis and transcendental unity of
apperception that I intimated earlier. The second citation, by
contrast, indicates a dependence on representations that
merit the title of knowledge on a sense of the identity of
the self. These passages suggest a kind of conflation
understanding of apperception whereby distinction of
elements within a synthetic whole are not merely dependent
on the sense of something prior existent that could
conceivably combine them together into one representation
but also that an awareness of the activity of the combination
is required for a sense that the combination is occurring and
capable of giving knowledge as its product.
Whilst Allisons reading is not committed to such an
explicit conflation account of apperception it is reliant on the
implication that some form of reflective awareness of
transcendental apperception is necessary.
13
The nature of
this could however be weaker than the implications of a
conflation reading of apperception would suggest. Peter
Strawson, for example, did not view the reflexive element of
apperception in terms of a conflation model as he
understood that this would collapse the distinction between
transcendental and empirical apperception but suggested
instead that empirical self-ascription of experiences depend-
ed for their possessive possibility on the a priori unity of
transcendental self-consciousness where the latter involved

13
The peculiarity of this point in Allisons account is worth emphasis as in
the original (1983) edition of his work he explicitly emphasized Reflexion
5661 where Kant distinguishes transcendental apperception from experience
(Allison 1983 pp. 275-7) so that the implication of a reflexive reading in the
citation given above from the re-edition of his work (Allison (2004) p. 171 is
at odds with this point in a way that is singularly surprising.
10/21
Gary Banham
Apperception and Spontaneity (2008)

as a minimum the distinction between how things are in the
world which experience is of and how they are experienced
as being, between the order of the world and the order of
experience.
14
Such a weaker type of reflexivity requires
only an implicit appeal to self-consciousness within the
experience of consciousness. In response to this however
there are two related problems. The first is that despite the
points made above concerning the suggested dependence of
transcendental apperception on the activity of synthesis in
Allisons account there is some warrant for this dependence
in the B-Deduction where Kant argues that the relation of
different representations to the subject requires more than
simple accompaniment of representations with conscious-
ness and adds: Only in so faras I can unite a manifold of
given representations in one consciousness, is it possible for
me to represent to myself the identity of the consciousness in
[i.e. throughout] these representations (B133). This
suggests a dependence of the identity of apperception on a
prior unification of synthesis that enables it to be produced
as implied in Allisons account. The second problem is that
the implied reference to self-consciousness can seem too
weak for there to be any sense given to the suggestion that
what is here involved merits in any sense the title of self-
consciousness. As Susan Hurley puts this objection: if the
activity itself, as opposed to its product, need not be
something we are conscious of in apperception, then in what
sense it is a matter of agency, something we do, rather than a
manner of passive happenings, mere events?
15

These problems are distinct but clearly related. On the
one hand, the problem with an implicitly reflexive reading of
the claim of transcendental apperception could be taken to
be that it first requires an act of unification before identity

14
Peter Strawson (1966) The Bounds of Sense: An Essay on Kants Critique
of Pure Reason (Routledge: London and New York), p. 107 where he also
states that what is meant by self-reflexiveness of experience is that
experience must be such as to provide room for the thought of experience
itself.
15
Susan Hurley (1994) Kant on Spontaneity and the Myth of the Giving,
Proceedings of the Aristotelian Society Vol XCIV Part 2: p. 145n.
11/21
Gary Banham
Apperception and Spontaneity (2008)

can be stated to be attained. On the other hand, it may seem
that if awareness of apperception does not require awareness
of acts of combination, that the implication of self-
consciousness is simply too weak for the claim to really be
one about any act of ourselves. Susan Hurley has in fact
pressed the latter point as a way of advancing her problem
with the suggestion that Kant has achieved, in any fully
satisfactory sense, a recognition of the place of subjectivity
in the examination of mind when she writes: Kant
ultimately appeals to activity to explain unity, as if unity
could not be a feature of the data of sensibility on their own,
and more importantly still, as if agency did not already
depend on unity.
16

Here we arrive at two connected problems that show the
need for a further examination of the connection of
spontaneity to apperception. On the one hand we need some
sense of transcendental apperception that enables us to
respond to the complicated question of whether Kant views
it as leading from unity to identity or the reverse and on the
other hand we need some sense of the way in which
transcendental apperception really merits description in
terms of self-consciousness rather than being understood in a
purely impersonal sense that will remove all possessive and
subjective connotations.
17

The relationship between unity and identity is one I will
return to as I wish first to assess the question of the nature

16
Susan Hurley (1996) Myth Upon Myth, Proceedings of the Aristotelian
Society 96: p. 163.
17
I dont here intend to simply run together subjectivity and possessive
relationships. Subjectivity seems at minimum to require the simple
representation of the I think with which we began as Descartes suggested.
For the I think to state in some sense an I however does seem to involve
something possessive being claimed. Attempts to eliminate this possessive
reference tend to produce functionalist accounts of apperception of which
there are many in the current literature. See for example Andrew Brook
(1994) Kant and the Mind (Cambridge University Press), Patricia Kitcher
(1990) Kants Transcendental Psychology (Oxford University Press) and
Pierre Keller (1998) Kant and the Demands of Self-Consciousness
(Cambridge University Press). It is impossible to respond to these accounts
here but see Gary Banham (2006) Kants Transcendental Imagination
(Palgrave Macmillan: London and New York), pp. 79-90.
12/21
Gary Banham
Apperception and Spontaneity (2008)

and degree of possessive connotation involved in the
transcendental unity of apperception. Since Kant repeatedly
stresses that acts of spontaneity are carried out by us and
indicates that one of the ways of understanding
transcendental apperception is as a form of transcendental
self-consciousness it seems natural to assume that there is
something possessive involved here and the notion of such a
possessive sense is surely connected to the idealistic
connotations of the transcendental unity of apperception.
When describing the manner in which transcendental
idealism provides a key to the solution of the cosmological
dialectic Kant suggests the subjective claim involved here is
one that relates to the noumenal self when he states: Even
the inner and sensible intuition of our mind (as object of
consciousness) which is represented as being determined by
the succession of different states in time, is not the self
proper, as it exists in itselfthat is, is not the transcendental
subjectbut only an appearance that has been given to the
sensibility of this, to us unknown, being. (A492/B520) The
identification suggested here between the transcendental
subject and the noumenal self is not an isolated one
18
but it is
not repeated in the Critique and it is in any case possible to
view it as one of the occasions where Kant writes
transcendental when he should have written
transcendent, not least because that makes most sense of
the noumenal neutrality thesis that is repeated in this very
quote. However, if we follow this suggestion through, it
follows that the subjective reference included in the notion
of transcendental self-consciousness is not only, (as befits
something declared transcendental) not empirical, but not
noumenal either. On those grounds it appears that there is a
form of subjective reference that is not involved with
anything like the notion of a subject.
In order to begin to make sense of this point I would like
to turn to two specific passages of the Critique, both from

18
See Reflexion 60001, Ak. 18: 420-1 where transcendental apperception is
equated with noumenal substance.
13/21
Gary Banham
Apperception and Spontaneity (2008)

the second edition. The first passage that is of interest here
concerns the manner in which Kant attempts a distinction
between inner sense and transcendental apperception, a
distinction that will turn out to be important for viewing how
the second passage from the emended Paralogisms chapter
should be viewed. At B153 Kant claims that transcendental
synthesis, when viewed by itself alone contains only the
unity of the act, of which, as an act, it is conscious to itself.
The liminal claim that when, viewed purely alone, such
synthesis is contained purely in a conscious act of
unification requires, if the term conscious is here to be
used, some sense of the combination being sensed
spontaneously by the kind of being that can be characterized
in transcendental terms and this would appear central to
Kants transcendental idealism. Kant speaks here directly of
the transcendental synthesis of imagination as the means by
which understanding determines sensibility and in the
process performs an act on the passive subject whose faculty
it is. An illustration that is subsequently given, in apparent
amplification of this point, concerns attention (B156-7n)
though the example appears confusing since attention would
necessarily be an example of an activity performed
empirically.
19
The more useful point that follows the initial
discussion of the transcendental synthesis of imagination as
involving self-affection concerns the distinction that Kant
subsequently gives of the type of representation given of the
I in transcendental apperception. Kant now claims that in
transcendental synthesis generally and hence in the
transcendental unity of apperception: I am conscious to
myself, not as I appear to myself, nor as I am in myself, but
only that I am (B157).
20

The consciousness that was claimed to belong to
transcendental synthesis strictly at B153 concerned the unity

19
Allison (1983) p. 268 notes this point and suggests a different reading of
the passage in question but the reading given certainly requires quite an
extension of Kants text beyond what is given in it.
20
This substantially corrects the impression of A492/B520 of an identity of
reference between the transcendental unity of apperception and the noumenal
self.
14/21
Gary Banham
Apperception and Spontaneity (2008)

of the act of synthesis and this consciousness of unity is now
explicated in terms of a mere consciousness that I am. The
representation of the I am is claimed to be a thought, not
an intuition and so something is required to connect it to
intuition. Rather than next directly saying more about the
statement that I am Kant returns instead to the
implications of the I think and claims that the I think
involves an act of determination of existence. In order to
claim I think he now says, I must first determine my
existence which implies that the awareness that I am is
logically prior to the awareness that I think.
21

Rather than expand on this question at this stage of the
argument of the Critique however Kant merely points out
that for determination of the manifold to be given some form
of intuition is required and that there is only one type of
manifold that is given to awareness of the self, namely
awareness of inner sense, or determination of time.
22
The
representation of my thought as a determination is a
representation that entitles me to claim that I am
spontaneous or that I am an intelligence.
These claims certainly allow for a sense that whilst the
function of transcendental apperception is, in the largest
terms, impersonal (in terms of being the foundation of
objective judgments)
23
, that it remains important to also
focus attention on the basis of the claim that I am

21
I am here drawing on B157-8n where the I think is described as an act
of determination of my existence. Kant here clearly states: Existence is
already given thereby, i.e., in my reading, it is there for the I think to be able
to determine it.
22
At B153 Kant points out that in order to avoid the contradiction of putting
us in a passive state with regard to our own action systems of psychology
tend to equate inner sense with apperception. It is clear however from B158n
that since I also cannot determine my existence as a self-active being that
there is here another source of the tendency to assume that the awareness of
inner sense in empirical apperception is sufficient to ground the basic sense
of the self.
23
This element of apperception is the most emphasised in standard accounts
of transcendental apperception and is based on 19 of the B-Deduction,
B140-2. However, important as it doubtless is to emphasize what we might
term the objective unity of apperception, the discussion of the ineliminable
subjectivity involved in the apperception claim is obscured by an exclusive
focus on it.
15/21
Gary Banham
Apperception and Spontaneity (2008)

spontaneous. Kant summarizes the claim made in the B-
Deduction simply when he writes: I exist as an intelligence
which is conscious solely of its power of combination
(B158-9). The point that I am conscious of this power is here
directly claimed and that I am assured of my existence as an
intelligence is due solely to this consciousness that I possess
the power of combination.
Now to understand these claims it is necessary to turn to
the discussion of the I think in the Paralogisms chapter.
This discussion is faceted since Kant here returns to the
implication of the argument of the transcendental deduction
of a connection between the representations I am and I
think. Kant states that the claim that I think is one that we
should view as analytically being a proposition that already
includes existence as given. Not only does he claim this but,
he subsequently adds, that with the I think the unity of
consciousness is also given, but that it is given only in
thought. The point of stressing that the unity of
consciousness that comes with the I think is given only in
thought is to deny the status of object to the unity in
question. The kind of point Kant is making here is that to
make the subject into a determinable object presupposes the
activity of the subject that is thereby apparently being
explained. The point I would prefer to pursue however
concerns the basis of the claim that the I think analytically
includes a reference to existence.
In making this claim Kant explicitly states that the I
think is an empirical proposition which contains within
itself the proposition I exist. If thought is somehow related
to existence this does not in itself affect the status of the bare
representation I. Kant indeed acknowledges this stating
that the I is not an empirical representation but a purely
intellectual one. Hence it precedes the determination that is
involved in thinking or existing. The conjunction of the I
with the determination that is involved in thinking is
apparently what produces an empirical representation. Kant
describes the combination of the bare representation I with
the determination thinking however as still only giving
16/21
Gary Banham
Apperception and Spontaneity (2008)

what he calls an indeterminate empirical intuition or
something like perception in general. For there to be a
perception, even in the most general sense, is to be affected
by a sensation so that if the I think refers to empirical
intuition in this most general sense then the statement of the
I think is conjoined with a sense that there is something
being sensed. What would be being sensed precedes
determination of any object of perception or as Kant puts
this point: An indeterminate perception here signifies only
something real that is given, given indeed to thought in
general, and so not as appearance, nor as thing in itself
(noumenon), but as something which actually exists, and
which in the proposition, I think, is denoted as such
(B423n). So there is here something like the barest
minimum of intensive magnitude, the barest sense that
something real is given and this sense is so bare that we can
express it only in terms of a thought, not directly as
appearance though it is by reference to something prompting
the thought that we arrive at the sense of it.
The suggestion in general terms however concerns the
fact that this awareness of thought involves a sense that what
is thinking is, in the very fact of thinking, evidently and
immediately given as also existent. However what we have
here is first the pure intellectual representation that is given
as the bare I in combination with a determination and this
determination is one in which the I becomes conscious of
itself as such, a consciousness that we can term transcend-
ental self-consciousness. For the act of unity that we have
seen to belong to the transcendental synthesis when it is
viewed in pure isolation to be made manifest is for the
unification of representations to be presented as given not in
a pure intellectual form alone but in combination with the
direct sense of existence: something is given to thought and
this something is first of all the spontaneous sense that there
is thinking as such. However this spontaneous sense that
there is thinking as such is viewed by Kant as part of an
empirical proposition as it relates the I to sensation. At
B423n the description of the indeterminate perception that
17/21
Gary Banham
Apperception and Spontaneity (2008)

something real is given is described as only given to thought
and thought, taken by itself alone, is stated at B428 to
involve pure spontaneity only. The consciousness of thought
by itself gives the sense that there is a being that thinks but
nothing concerning this being is directly given in the thought
in question.
We can move from this sense of thought by itself to the
thinking of such thought in stating the proposition which is
not a mere logical function, namely that I exist thinking at
which point we must invoke, Kant states, inner sense
(B429). The understanding that goes beyond the mere
intellectual representation and gives an application of it to a
manifold occurs with the sense that what is thinking is more
than a general being but is something actual whose activity
we are aware of. However whilst this gives content to the
representation the content in question is described by means
of inner empirical intuition which is sensible. Thus the only
element that attaches to pure consciousness is the sense of
combination, the determination of this combination in
relation to the manifold already moves beyond the
intellectual element to the empirical and in so doing takes
the representation of unity to be the basis of the combination
of the intuition of the manifold.
24

The relationship between the intellectual representation of
the I and the combination of the manifold in intuition is
likewise set out in the Transcendental Aesthetic where Kant
remarks again that what is a simple representation is the
consciousness of the I with its connection to a manifold
again taken to require relation to sensible receptivity in
which the intellectual representation is said to affect the
mind in order for an intuition of the self to arise. (B68-9)
Going back to the argument of the transcendental deduction
we also find that Kant there asserts that the principle of the

24
The subsequent discussion in the concluding part of the B Paralogisms
chapter moves over to the consciousness of the moral law in which Kant
clearly applies a sense of the fact of reason in terms of a consciousness
that involves activity in a more general sense (B430-1). Exploration of this
discussion would require in a different place thinking about the relationship
between spontaneity, freedom and autonomy.
18/21
Gary Banham
Apperception and Spontaneity (2008)

necessary unity of apperception is, as stated in the
Paralogisms, an analytic proposition which therefore states
something purely identical. The I is again taken to be a
simple representation, which is distinct from the manifold of
intuition but again the combination of the manifold requires
consciousness of the self as identical in the activity of
combination. The nature of the type of consciousness is
however stated famously at A346/B404 to be of only a
transcendental subject which is known only through the
thoughts that are its predicates and is distinguished thereby
from the I think that is individuated.
The relationship between the identity claim that attaches
to the bare representation I and the unity claim that we
require for the manifold of intuition to be able to represent
anything like an object and even give the sense of existence
to the I that thinks is one that still remains to be explored.
In Kants observation on the antithesis of the Second
Antinomy however he writes the following: Self-
consciousness is of such a nature that since the subject
which thinks is at the same time its own object, it cannot
divide itself, though it can divide the determinations which
inhere in it; for in regard to itself every object is absolute
unity (A443/B471). Here the logical point is made that for
a representation to be a representation of something there
must be a unity given that enables us to determine the
distinction between this representation and some other.
25

When we have a representation of self-consciousness the
basis of distinction is given through the predicates that are
attached to the sense of thinking which is the manner in
which the purely intellectual representation of the I is
connected to determinations although these determinations
have their basis only in intuition. There is hence required
both the sense of absolute unity as attaching to the subject
and yet also the determination of perceptions by means of a

25
As I have argued elsewhere this implies the presence of the concepts of
reflection as the basis of the most elementary possibility of representations,
as the effective genetic source of experience as an experience of anything at
all. See Gary Banham (2006) pp. 139-44.
19/21
Gary Banham
Apperception and Spontaneity (2008)

relation not to the purely intellectual representation but
instead merely to appearances that are themselves
represented according to the pattern of the I itself, namely,
by means of unification. Thus the Is simplicity is at the
same time, in any given act of synthesis and any act of
awareness of combination not merely a sense of identity but
an awareness of this identity as a combinatory power.
The division between arguments in which Kant appears
to go from a sense of the identity of the I to its unity by
contrast to arguments in which he appears to go from unity
to identity has been noted and questioned by some
commentators.
26
As stated earlier the difference between
claims of unity and claims of identity are that whilst unity
alone does not require reference to combination over time,
identity does imply differential states that are related to each
other, a point that is made clear in the difference between the
claims examined in the 3
rd
Paralogism by contrast to the 2
nd
.
The reference to the unity that is inherent within
apperception itself can however be grounded in the bare
simple representation of the I alone so that unification can
be grasped in a singular intellectual act whilst identification
involves a determination. In a sense Kants argument should
then go from unity to identity as without unity already
existent as such it is impossible to see how it could ever
emerge as a product. Since identity also requires
differentiation then it implies a sense of particulars and if
there is a possessive sense to apperception it would seem to
reside in the awareness, as Kant himself suggests, of a real
particular that is given.
There are two problems with Allisons description of this
awareness as empirical apperception: the first concerns the
question as to whether this distinction is enough whilst the
second concerns the problem of whether possessive claims
can be left only at the level of the empirical. The first point

26
It is central to the reading given by Dieter Henrich (1976) though the
question of whether the reading advanced here can respond to the questions
put by Henrich would require much more work than can be undertaken in this
piece so the answer given above will be of a provisional nature.
20/21
Gary Banham
Apperception and Spontaneity (2008)

can be made clearer as whilst Kant does interpret the I
think as an empirical proposition, the basis of this claim
concerns two distinct elements, firstly the sense of an
indeterminate empirical intuition and subsequently the sense
of a relation to inner sense with this latter bringing us to
the manifold. Rather than it being the case that we
distinguish only between transcendental apperception,
empirical apperception and inner sense, I suggest we also
need a sense of the indeterminate empirical intuition as
spontaneity. The grasp of this indeterminate empirical
intuition does not yet involve appearances directly but does
require a sense of affection and this sense of affection would
be the generation of inner sense, a generation that would
permit the subsequent awareness of relation to inner sense
that we could term empirical apperception. The point that
would follow from this is a rescue of the possessive
connotations that would seem to be required for a relation to
be sustained between empirical apperception and
transcendental apperception. This possessive sense is that it
is not merely affection that is occurring but that this affection
is something that is happening to a given particular,
something real as Kant puts it. The actuality of this
particular as the ground of the move from transcendental
apperception to empirical apperception would be the basis
also of a type of possessive relation to the merely possible
consciousness given in the bare representation of the I.
Whilst this I is not equivalent to a personal sense of
identity it is the ground of such a sense as without it the
general claim of affection would not relate to a sense that the
affection in question is touching something that I can
identity as a particular. This sense of the particular as
affected is indeterminate, given prior to appearances being
distinguished as such and yet no longer purely intellectual as
the affection in question still possesses a barest intensive
magnitude.
In conclusion I have not so much disagreed with the
positions of Allison that I presented in beginning as
underscored them and in the process amplified their sense so
21/21
Gary Banham
Apperception and Spontaneity (2008)

that further distinctions have become necessary, distinctions
that enable a clearer sense of the relation between
spontaneity and apperception and also in the process rescue
a possessive sense of apperception without compromising
the purity of its transcendental unity.