F I C H T E S MODIFICATION OF KANT'S

1
TRANSCENDENTAL
IDEALISM IN THE WISSENSCHAFTSLEHRE OF 179 4
AND INTRODUCTIONS OF 1797

Charles Griswold

In this essay I study Fichte's modification of

Kant's Transcendental Idealism. 1
Since the Gründlage
der gesamten Wissenschaftslehre of 1794 and two Intro-
ductions of 1797 present this modification more e x -
plicitly and clearly than do Fichte's later w o r k s , and
since these earlier works were studied by Hegel in the
Differenz des Fichteschen und Schellingschen Systems
der Philosophie/ I shall confine myself to them.^ Fichte
is one of the central figures, if not the central figure,
in the transition from Kant to Hegel, and thus stands at
the centre of "German Idealism." I hope that the follow-
ing study will suggest, however, that Fichte's thought
is of philosophical, as well as historical, interest.
I w i l l begin by stating the ways in which Fichte sees
himself as modifying Kant's teaching. I will then
state the Kantian criticisms of this modification.
Having outlined the controversy between Kant and
Fichte, I will discuss parts of the GWL, especially
the three Grundsatze der gesamten Wissenshaftslehre.
I w i l l , in conclusion, make several remarks about the
relationship o f Fichte to Kant, and of Transcendental
Idealism as a whole to Hegel's Absolute Idealism.

Throughout the GWL of 1794, and the two Introductions

of 1797, Fichte characterizes his position as the m e d i a -
tion of "dogmatic realism" and "dogmatic idealism." Kant
offers an almost identical formulation of his own p o s i -
tion. 3 The two kinds of dogmatism are based on the
fundamental distinction between mind and body, i n t e l -
ligence and matter, or better yet, between subject and
object. The dogmatic idealist insists on the primacy
of the free activity of the self, intelligence, the
subject. The object is merely chimerical. The dogmatic
realist insists on the primacy of the body, matter, the
object. The self is merely chimerical. Kant and Fichte
take the "middle way" in this "gigantomachia."

Fichte's explicit criticism of the dogmatists (one •

that is implicit in Kant's Refutation of Idealism: CPR
B274-5 and context) is that they cannot explain the fact
 134

of self-consciousness, or more accurately, the unity of

self-consciousness and consciousness. The dogmatic
realist will try either to deny self-consciousness alto-
gether, or to explain it as the effect of the object on
intelligence. The former claim leads to the absurd r e -
sult that if it were true, its advocate could not say
that it is his claim or explain why it is t r u e . The
latter claim might show how the self has an impression
of the object, but not how the self is conscious that
it has such an impression (2nd Intro., p p . 7 4 - 8 0 ) . The
dogmatic realist stays on the level of consciousness.
The dogmatic idealist, on the other hand, also loses
self-consciousness but by doing away with consciousness.
Yet one essential component of self-consciousness is
the object of consciousness. The object cannot be
assimilated into the subject without destroying the
subject's consciousness of himself. Fichte's mediation
between the two dogmatic positions may be summarized as
a ". . . complete deduction of all experience from the
possibility of self-consciousness" (2nd Intro., p . 3 7 ) .
While self-consciousness is possible only if both self
and object are preserved, the deduction from the p o s -
sibility of self-consciousness, Transcendental Idealism,
is still a kind of idealism. The reason for this is
that it is impossible to "abstract from the self" (2nd
Intro., p . 71; GWL pp. 98, 202, 2 5 2 ) . The object is
given, but given for us; the dispute between dogmatists
exists only for us as self-conscious selves; and so on.

Fichte 's actual presentation of Transcendental

Idealism is nonetheless very different from that of
Kant. Fichte states that he is developing the "spirit"
of Kant's philosophy (2nd Intro., p . 5 2 ; GWL p . 1 7 1 ,
note 3 ) , building on foundations laid by Kant (2nd Intro.,
p . 5 1 ) , making explicit what Kant left implicit (GWL p .
1 0 0 ) , or simply understanding Kant properly (1st Intro.,
p . 3; 2nd Intro., p . 4 3 ) . All of this was vehemently
denied by the Kantians of the time (Reinhold, Schulze,
E b e r h a r d ) , and finally by Kant himself.4 Fichte's
development of Kant's teaching is based on three d i s c r e -
pancies between its "spirit" and its "word." First,
Fichte believes that the basic outlines of Kant's
"Copernican turn" are correct. Kant sees that we d e t e r -
mine appearances through the spontaneous synthesis of
thought in the categories, and so that "concepts rest
on functions" (CPR B 9 3 ) . But Kant does not deduce those
categories from a single principle. Kant merely finds
the categories, so to speak. Nor does he deduce the
forms of intuition, or successfully state their relation
to the categories (2nd Intro., p . 5 1 ) . Kant does not
possess a system; the parts are there, but in a heap.5

Second, Fichte believes that Kant correctly sees

135

that the transcendental unity of apperception, the pure

I, must lie at the basis of all consciousness. In Kant's
terms, "it must be possible for the 'I think' to a c -
company all my representations" (CPR B 1 3 1 ) . The "I
think" is original apperception, the act of spontaneity,
the "transcendental unity of self-consciousness" (CPR
Bl32)~ This unity is the "highest principle" of all
logic and transcendental philosophy (CPR B135; note,
B134). Hence Fichte observes that for Kant the p o s -
sibility of consciousness w i l l be conditioned by the
possibility of pure self-consciousness (2nd Intro,, p .
5 0 ) . However, Kant fails to demonstrate this "condi-
tioning" coherently, i.e., he does not give a systematic
derivation of all consciousness. To do so, Kant w o u l d
have had to begin with the "pure I" and derive the cat-
egories and the forms of intuition, instead of doing the
opposite. This systematic derivation in Fichte h a s , as
we shall see, crucial consequences for the "Ding an
sich." In short, for Fichte the "thing in itself" is a
concept.

Third, Fichte accuses Kant of omitting a crucial

element of the explanation of self-consciousness. Kant
does not explain how the ' I ' "accompanies" every repre-
sentation. Had he done so, he would have accepted,
rather than rejected, intellectual intuition (intel-
lektuelle A n s c h a u u n g ) . This faculty is indispensable,
in Fichte's view, Tn explaining our consciousness of
identity through time, as w e l l as our consciousness of
the moral law.& Fichte does not mention intellectual
intuition in the GWL of 1794, however, but it is hinted
at in the preface of 1795 (GWL p . 91) and defended
vigorously in the Introductions of 1797.7

The major Kantian criticisms of the modification of

Transcendental Idealism are twofold. First, and most
obviously, Kant had simply rejected intellectual in-
tuition. The reason for this rejection is crucial.
Kant argues in the Transcendental Aesthetic (CPR, B
edition) and the Paralogisms (CPR) that the subject has
n o immediate intuition of himself because all self-
consciousness is mediated by inner sense, and finally
through outer sense as well (Refutation of Idealism,
CPR B x l , B 2 7 7 ) . Fichte replies that intellectual i n -
tuition is always accompanied by concepts and sensory
intuition (2nd Intro., pp. 3 8 - 9 ) , and that we are not
ever conscious of this intellectual intuition. Intel-
lectual intuition cannot be reduced to concepts and so
become an object of analysis. In other w o r d s , Fichte
agrees with Kant that self-consciousness is possible
only by the subject's mediation of subject and object.
Nevertheless, intellectual intuition must exist if w e
are to explain certain facts of which we are conscious
 136

— i . e . , identity through time and space, and the moral

law (2nd Intro., p p . 4 5 - 6 ) . To say that w e are mediately
conscious of this identity is to say that w e are not in
fact conscious of it.

The second major Kantian objection to Fichte's m o d -

ifications is that Kant's system requires a "thing in
itself." Sensations must be initiated by the "thing in
itself" even though that thing is unknowable. There are
appearances which we know, so there must b e something b e -
hind what appears (CPR A 2 5 1 - 2 ) . Fichte's x e p l y is that
the few passages in the CPR to that effect must be inter-
preted in the light of the "CPR as a w h o l e . The notion
of causality, which first gives rise to the view that
there are "things in themselves" (the "giveness" being
an effect on us) is nothing but a subjective category
of the understanding. Hence its use cannot show that
there are such things. Fichte agrees with Kant that
noumena may be necessary as "limiting concepts" (CPR
A255). But the Critique as a whole shows that this
necessity is only subjective, i.e., a necessity of
thought. In Fichte's terms, the n o t - I , the pure "give-
n e s s " of the "thing in itself," is "posited" (setzen)
by the absolute I (the transcendental unity of appercep-
tion) . For Fichte this thesis is the core of T r a n -
scendental Idealism. To insist on an independent "thing
in itself" is to regress to dogmatic realism, and so t o
lose seif-consciousness.

Yet this seems to turn Fichte into a dogmatic

idealist, because it seems to destroy the "affection"
and "receptivity" of the Transcendental Aesthetic ( C P R ) .
Kant accused Fichte of trying to deduce existence and
affection from a concept, and so of uniting the two
"stems of human knowledge" (CPR B 2 9 ) . 8
Fichte denied
this accusation; he too insists on the "facticity" of
experience, on the feeling of being necessitated o r
affected by the object: "The object is not a priori,
but is first given to that science (GWL) in experience;
objective validity is furnished of everyone by his own
consciousness of the object, which consciousness can
only be postulated a priori, but not deduced" (GWL p .
224; see also pp. 105-6, 223, 242, 246 f f . ) . However,
Fichte argues that reflection on this experience cannot
self-consciously postulate a "thing in itself," though
it may have to postulate an ideal "thing in itself."
The GWL is, Fichte says, an empirical realism and a
transcendental idealsim (2nd Intro., p . 61; GWL pp. 147,
174, 2 4 6 - 7 ) . This is equivalent to the teaching o f the
Transcendental Aesthetic (GWL p . 62) . Time and space
are transcendentally ideal because they do not belong
to "things in themselves." They are empirically real
because they are "objectively" valid, i.e., necessary
 137

conditions for any object being an appearance for us

(CPR B 4 4 ) . Kant e r r s , however, in trying to demonstrate
"the ideality of objects from the presupposed ideality
of space and time" (GWL p . 171, n o t e ) . By beginning
with the Transcendental Aesthetic, Kant can only assert
the ideality of space and time, and so suggest the tran-
scendental reality of "things in themselves." This con-
fusion is avoided if the procedure is reversed; the
ideality of objects, and that in turn from the "highest
principle" under which anything can be an object of
thought for us at all, the Absolute 1.9

Let me now begin with the GWL itself. The "architec-

tonic" is very different from that of the C P R . 1 0
All of
Kant's rigid distinctions are, so to speak, collapsed
and set into m o t i o n .
1 1
Perhaps the most important change,
for our purposes, is that of method. Fichte's "deductive"
method is t r i p a r t i t e — t h e famous thesis, antithesis, and
synthesis. 1 2
Antithesis is the act of seeking in things
equated the respect in which they are opposed. Fichte
says that this is what Kant calls "analysis." Synthesis
consists of discovering in opposites the respect in
which they are alike. Antithesis and synthesis are
equivalent to dividing and collecting according to genus
and difference. Antithesis shows the difference which
separates species from genus, while synthesis shows why
the species belongs to the given genus (GWL p . 1 1 4 ) .
Both procedures are based on the thesis, in which nothing
is opposed or conjoined. There is only one thesis in
the GWL; the Absolute I, the "I am" as such (2nd Intro.,
p . 73; GWL p . 1 1 4 ) .

What Kant calls "analysis" seems to include both

the antithesis and the synthesis of Fichte.i3 Analysis
for Kant is simply conceptual. Kant's "synthesis," h o w -
ever, is not equivalent to that of Fichte. For Kant
synthetic judgements are "judgements of experience" (CPR A
154 , 1 5 8 ) . In these judgements it is necessary to
"advance beyond a given concept," i.e., to include
sensible intuitions. The a priori synthetic unity of
concepts must be exhibited in experience or in a "pos-
sible experience." There can be a priori synthetic
judgements because "the conditions of the possibility of
experience in general are likewise the conditions of the
possibility of objects of experience" (CPR A 1 5 8 , 1 1 1 ) .
This is hardly reducible to the "discovering in opposites
the respect in which they are alike" (Fichte's "syn-
t h e s i s " ) . The net effect is that in Fichte the criterion
of "possible experience" is dropped. Alternatively, the
meaning of "experience" has changed. For Kant, "exper-
ience" and "knowledge" are almost synonymous terms. For
Fichte, while experience is the unity of concepts and
intuitions, knowledge of that experience requires only
 138

that concepts of intuitions be joined to concepts of the

understanding. "Knowledge" for Fichte seems to be more
like what Kant calls the "principles of pure reason"
(CPR A 2 9 9 - 3 0 2 ) . I shall return to this point below.
In sum, Fichte alters the relation between concepts and
intuitions which Kant appears to advocate. Whether o r
not this alteration makes Kant consistent has yet to be
decided.

Fichte's first principle is the 1=1, the Absolute I

(absolute I c h ) , i.e., the transcendental unity of apper-
ception. I have already discussed why Fichte starts
with the I, and so with s e l f - c o n s c i o u s n e s s — i t is the
"highest principle." But how does he get to the 1=1?
Fichte begins with a "fact of consciousness," the p r i n -
ciple of identity, A=A. This principle is accepted by
everyone; but why is it true? The principle says that
if A is given, then A must follow; nothing is said about
the existence of A. Its truth is that necessary con-
nection, the "=" itself, which Fichte entitles "X." X
is the "form" of A=A, while the two A's are the "con-
tent." This form is equivalent to the 1 = 1 , which Fichte
also calls "pure self-reverting activity." "pure s p o n -
taneity." The Absolute I posits A=A, and only for that
reason does A=A seem so certain. The ordo cognoscendi .
(the principles of speech) images, so to speak, the
ordo essendi (the pure "I think").14

The relationship between these two orders is not,

however, symmetrical. The 1=1 and A=A have the same
logical form, but a different, meaning. The two state-
ments are thus "wholly different" (GWL p . 9 6 ) . The I
is the perfect unity of form and content, subject and
object, positing and existence. Unlike A=A, the p r i n -
ciple 1=1 is posited " . . . not conditionally, but
absolutely, with the predicate of equivalence to itself"
(GWL p . 9 6 ) . The self-identity of the Absolute, which
sustains the logical principle A=A or "posits" the two
A's as identical, is itself " . . . valid not merely in
form but also in content" (GWL p . 9 6 ) . But we can d i f -
ferentiate form and content, positing and existence, in
both 1=1 and A=A. Although the terms in each equation
are the "same," they are still numerically "different"
(in that respect one A or I is_ not the other A or I_.
Thus while 1=1 (which shares the same logical form as
A=A) might mean "Absolute I," it cannot accurately say
it. The Absolute I reveals itself in discourse, but in
so doing conceals or alters itself.15 The source of
Activity and the products of Activity are discontinuous:
". ... initially I am neither the reflecting subject nor
the reflected object, and neither of the two is c o n d i -
tioned by the other, since I am both in combination;
though this union I cannot indeed think; since in the
act of doing so I separate the reflected from that which
 139

reflects" (GWL pp. 6 0 - 6 1 ) . The reflecting subject is

the Absolute I, the reflected object is the Absolute as
articulated in discourse; the articulation is not fully
adequate to the articulated and as such is separate from
it. This problem, which I will return to below, is
fundamental to the GWL. Fichte's attempts to resolve
it lead to the "ought" (Sollen) of practice (GWL p p . 102,
229, 230 n o t e ) , and to the doctrine of intellectual i n -
tuition (above). In Fichte's view, the problem is intrin-
sic to Transcendental Idealism, though it was not recog-
nized as such by Kant.

There is another aspect to the difference between

the Absolute I and its articulation in discourse. The
Absolute I is "permanently uniform, forever one and the
same" (GWL p . 96; also 2nd Intro., p p . 4 0 , 63, 7 5 ) . D i s -
course, on the other hand, is in flux. In some sense,
the Absolute I is eternal. The same point is implicit
in Kant. For Kant, all self-consciousness is in time;
but the transcendental unity of apperception does not,
presumably, change. Kant insists that this unity is a
"logical" condition (cf. the Paralogisms, B edition, C P R ) ,
but leaves the exact status of this condition undermined
with respect to its permanence. In his effort to deduce
Critical Philosophy from the transcendental unity of
apperception, Fichte brings to light several problems
concerning the relation of that unity to what it unifies.
While Fichte denies that the transcendental unity of
apperception is transcendent (the I is not a noumenon;
GWL p p . 101, 117, 140, 226, 2 4 6 - 4 7 ) , that unity, now
translated as the "Absolute I," looks more and more like
God. In the GWL, at least, Fichte would resist the
identification of the Absolute I with God.16

Fichte's second principle is another "certain"

fact of consciousness, Aj*-A ("A is not equal to not A " ) .
This principle is the basis for the principles of excluded
middle and contradiction. While it means the same thing
as the principle of identity, it introduces a new term;
-A, o r negation. Correspondingly, in the ordo essendi
-I ("not-I") has been introduced. This second principle
is based on the first; logically, because -A cannot be
stated without A, while the reverse is not true; ontolog-
ically, because the I is assumed by the - I , not vice versa,
and because in order to explain self-consciousness and so
consciousness of the - I , we must begin with the I (the
"highest p r i n c i p l e " ) . This means that the I posits i t -
self as 1=1, and so posits A=A; then the I posits - I , or
I^-I, and A^-A. However, in order to posit I^-I, the I
must first posit - I ; but in doing so the I contradicts
itself. That contradiction is not visible in the state-
ments "A=A and "A5*-A" since "A" is not an activity which
first posits itself and then its opposite.
 140

Let me try to clarify further this transition from

the first to the second principle. Fichte says that
" . . . even the possibility of counter-positing (i.e..
of positing the not-I) itself presupposes the identity
of consciousness" (GWL p . 1 0 3 ) . Hence we begin with
the I and then move to the n o t - I . In Kant's terms, we
begin with original apperception and then move to the
object, which is by definition what original appercep-
tion is_ not. In Fichte's terms, we move from reality
to negation (note that these are two of Kant's categor-
ies) , since there is nowhere else to move to if we begin
with reality.

But this introduces a decisive problem. The -I or

-A, being negations of I and A, are materially condition-
ed by I and A respectively. The negation itself, i.e.,
the "form" of the proposition, cannot be derived from
the I or A. Reality and negation are formally discon-
tinuous . Thus the -I must have been "absolutely posited"
just as the I w a s . Since only the I can posit, the -I
must have been posited by the I; yet this is impossible,
since there is no negation in the I to begin w i t h . Hence
there is no strict deduction of the principle of negation
from that of reality, i.e., no deduction of opposition
from identity. As Fichte says over and over again, that
the -I should be posited cannot be theoretically d e t e r -
mined (above) ; but it is determinable that if_ something
is to be posited after the I, it must be the not-I
(GWL p p . 144, 145, 153, 154, 171; and part I, section 2 ) .
Yet this does not explain how there can be any positing
of negation after the I. The I is absolute, the totality,
reality; how can something be other than the totality?
In sum, "factic" experience shows that there is a n o t - I ;
but reflection cannot systematically account for it, as
it ought to. The discontinuity between the source o f
synthetic activity (Absolute I) and its product (not-I)
means that the Kantian split between the transcendental
unity of apperception and the thing in itself cannot be
overcome. However, it cannot be affirmed either.

To restate these last few points "theologically," one

could say that for Fichte the reasons for God's creation
are unknowable. What can be known is that if God creates,
He must create the negation of Himself, namely the finite
w o r l d — H i s image. But in this creation God contradicts
himself, since He is limited by His creation and so is
no longer God, i.e.. Totality. Moreover, even God's
thought of creating before the actual creation contradicts
His nature; thinking is self-consciousness, and depends
on images and negativity. God contains n o negativity,
and so cannot be self-conscious.17 But there is a c r e -
ation and we are compelled to think of it as created.18
Yet we cannot conceive of how God could have created
 141

the world, i.e./ we cannot deduce the one from the other.
This is closely linked to another problem; w e cannot
accurately articulate God as He is "in Himself." We can
recognize ourselves as images of God and so have an i n -
tuition of the original (God), but any effort to bespeak
the original simply produces more (verbal) images. To
translate this back into Fichte's terminology, we can
intuit but not discursively analyze the Absolute I,
and we can experience the not-I, but we cannot deduce
the latter from the former. The third principle of the
GWL shows that these two difficulties are really two
sides of the same coin.

The third principle supplies the category of limita-

tion as the resolution to the contradiction between
reality and negation. I and not-I limit each other; to
limit is to negate in part. Something can be limited in
part only if it is divisible; limitation turns what is
limited into quantities. This means that the limitation
of reality by negation has produced a "quantitative"
reality which is opposed to a "quantitative" negation.
Yet the Absolute I is unconditioned, and thus "qualita-
tively" real (GWL pp- 109, 1 1 0 ) . That is, the third
principle lowers us from the level of infinitude to the
level of finitude. The I and the not-I which limit each
other are finite; neither is equivalent to the Absolute
I. Finite I and finite not-I contradict each other, and
both contradict the Absolute I. In some sense the finite
not-I is posited by the I, and in some sense both finite
I and not-I are posited by the Absolute I. The situation,
needless to say, has by now become exceedingly complex,
and the rest of the GWL is devoted to its analysis.

Roughly put, the theoretical section of the GWL deals

with the contradiction between finite I and n o t - I . The
thesis of the theoretical section is that "the I posits
itself as limited by the not-I." This corresponds to
Kant's Transcendental Analytic (CPR) , i.e., to the effort
to unite concepts of the understanding (finite I) with
intuitions (not-I) within the unity of transcendental
apperception (Absolute I ) . The unity of concepts and
intuitions takes place at the level of finitude; in-
tuitions "affect" the self but are also themselves
affected by the syntheses of the self. The problem is
to see how original apperception, the Absolute I, syn-
thesizes or posits both affections and activities w i t h -
out itself being affected, i.e., how the identity of
self-consciousness is preserved alongside the contra-
dictions it produces and retains in itself. Reflection
cannot solve this problem because it is unable to effect
the "descent" from infinitude to finitude. Nor can it
effect the "ascent" from finitude to infinitude. The
practical section of the GWL, which corresponds partly
 142

to Kant's Transcendental Dialectic (CPR), but more gen-

erally to the Critique of Practical Reason, "solves"
this contradiction between infinitude and finitude
through the "ought" (Sollen). The thesis of practical
philosophy is that "the I posits itself as limiting the
n o t - I . " This is the principle of the infinitude o f the
I, legislating autonoumously over nature. It is the
principle that nature (the not-I) ought to conform to
the Absolute I.

The theoretical section of the GWL attempts to r e -

solve the contradiction between finite I and not-I by the
"grounding principle" (Satz des G r u n d e s ) , a logical v e r -
sion of the principle of sufficient reason. It consists
of applying antithesis and synthesis in the m a n n e r d e -
scribed above, i.e., of finding the ground of disjunction
and conjunction of the I and n o t - I . But this ground can-
not be found precisely because it is impossible to ascend
discursively to the grounding Absolute I or to the summum
genus (the ascent is accomplished by intellectual i n -
tuition and abstraction). Hence reflection cannot termi-
nate itself; it progresses ad infinitum (GWL p p . 1 0 6 , 113,
etc.). For theoretical philosophy, the Absolute I will
possess only "regulative validity" (GWL p . 119; see also
p . 1 6 4 ) . For practical philosophy, the Absolute I w i l l
possess constitutive validity; but the constitution of
nature by the Absolute I is an "ought." The "ought" is
accompanied by "infinite striving;" the desire, n e v e r to
be fulfilled, to make the "ought" actual. The Absolute
I is given in intellectual intuition, but cannot be made
actual in practice, just as previously it could not be
made actual in theory (concepts and d i s c o u r s e ) .

As opposed to the Absolute I, the not-I is "abso-

lutely nothing." As opposed to the finite I, the not-I
is a negative quantity, insofar as it is opposed to the
n o t - I , the finite I too would seem to be "absolutely
nothing" in relation to the Absolute I. We seem to have
here a confirmation of Hegel's point that Verstand.(under-
standing, or the finite I) cut off from Being (infinitude)
rests on Nothingness (Nichts). From Hegel's point of
view, both Kant and Fichte failed to overcome this
Nothingness, and so failed to explain the A b s o l u t e .
Fichte's cure for this Nothing is the "ought" accompanied
by intellectual intuition; hence his idealism is "practi-
cal" (GWL p . 1 4 7 ) . Morality is the antidote for the
weakness of the understanding. From Hegel's point of
view, however, Fichte's reliance on intellectual intuition
is contentless subjective feeling.20 The corresponding
subjectivity of the Absolute I prevents Fichte from e x -
plaining the externality of the not-I; Fichte is a s u b -
jective idealist.21 in Fichte, the Absolute becomes an
ideal, which ". . . floats as a vision before u s , and is
 143

rooted in our innermost nature" (GWL p . 2 3 8 ) . The vision

cannot be actualized; the finite I cannot be absolute,
but it can unceasingly strive to be absolute. Hegel
claimed to actualize the vision of the Absolute in d i s -
course; thus it is not surprising that he dispenses with
intellectual intuition, and perhaps with the "ought" as
well.

In sum, the Absolute I has two aspects. On the one

hand, it is needed in order to explain self-consciousness.
Self-consciousness must be understood in the light of the
fact that it is always 1^ who am self-conscious (the "I
think" must accompany all my representations), and that
the unity of consciousness is a synthesis spontaneously
effected by the I. Even the object of consciousness,
while retaining all of its "factic" empirical reality,
must be explained as a product of the I. On the other
hand, the Absolute I makes morality possible, since it
is the source of our legislation over nature. Thus one
might say that Fichte's Absolute I combines the tran-
scendental ego of Kant's first Critique with the homo
noumenon of the second Critique. In any event, the im-
possibility of denying the actuality of the - I , and
finally combined with the demands of theory and practice
that the -I be determined by the I, define the contra-
dictoriness of human existence. "This fact, that the
finite spirit must necessarily posit something absolute
outside itself (a thing in i t s e l f ) , and yet must recog-
nize, from the other side, that the latter exists only
for it (as a necessary n o u m e n o n ) , is that circle which
it is able to extend into infinity, but can never escape
(GWL p . 2 4 7 ) .

The result is that for Fichte the n o t - I , or what

Kant calls the "thing in itself," cannot be proved or
denied. Every effort to prove that it exists either
loses self-consciousness, or shows only that we must
think "as if" it exists. Every effort to prove that it
does not exist must do so by.deriving it from the A b s o -
lute I. Yet that is impossible, because the Absolute I
and the not-I are formally discontinuous, and the A b s o -
lute I cannot in any case be articulated as it is in i t -
self, (i.e., as undifferentiated pure a c t i v i t y ) . The
"thing in itself," as Fichte strikingly puts it, is
"nowhere and everywhere at once." Any attempt to change
this status of the "thing in itself" " . . . would entail
the elimination of all consciousness, and with it of all
existence" (GWL p . 2 4 9 ) . Thus in the final analysis,
Fichte's position on the "thing in itself" is very close
to that of Kant, or at least to one of Kant's characteri-
zations of it as a "problematic" concept (CPR B 3 1 0 ) .

Fichte, like Kant, insists on human finitude. On

 144

the one hand, it is impossible to articulate the A b s o -

lute I, the first principle of all knowledge. On the
other hand it is equally impossible to define the not-I
as it ought to be defined, i.e., as a product of the
Absolute I. But Fichte's demonstration of this finitude
seems to be superior to that of Kant's in exactly the
ways which Fichte indicates; it is more coherent, s y s -
tematic, and presents the real thrust of Transcendental
Idealism more consistently. At the same time, Fichte's
presentation o f Transcendental Idealism clarifies some o f
the problems of that d o c t r i n e — i n particular, the problem
of discursively uniting reality and negation in the a c -
tivity of the Absolute I. Fichte achieves this clarifi-
cation in great part because of his main modification of
Transcendental Idealism; the demand that everything be
deduced from the Absolute I. Hegel, in turn, comple-
ments Fichte for attempting to carry out this m o d i f i c a -
tion, but criticizes him for failing to complete the
deduction.22

Fichte seems also to be superior to Kant in that he

is more self-conscious than Kant about the status of
philosophical reflection in the "Science of Knowledge."
As noted above, Kant seems to equate knowledge and e x -
perience; both are the result of the work of understand-
ing (Verstand) together with intuitions. Pure reason
(Vernunft) seems to have a solely "dialectical," and so
negative, function. And yet the Critique of Pure Reason
is both a critique about pure reason (as Fichte notes;
GWL p . 216) and a critique carried out by_ pure reason
(cf. CPR A x i - x i i ) . Hence the definition of "knowledge"
in the CPR does not seem to include the definition i t -
self. The faculty of producing the definition (reason)
seems to be inferior to the faculty it defines (under-
standing) . This split between meta-level (or critical
and transcendental level) and object-level considerably
weakens Kant's "criticism" as a whole, and seems to b e -
tray a loss of self-consciousness about the meaning of
"criticism." Unless the conditions for the possibility
of knowledge are themselves objects of knowledge, it is
impossible to defend those conditions as the true o n e s .
Hegel criticizes Kant on these grounds; in effect, Kant
lost self-consciousness about his own project.23

Fichte avoids this criticism of Kant, at least in

part. In short, Fichte unites Kant's Transcendental
Analytic with his Transcendental Dialectic in a continuous
deduction by_ pure reason from the Absolute I, i.e., from
pure reason itself (GWL p . 2 1 6 ) . While Fichte still d i s -
tinguishes understanding from reason (GWL p p . 2 0 7 , 2 1 4 - 1 6 ) ,
as well as "natural" reflection from philosophical r e -
flection (2nd Intro., pp. 30, 4 7 ; GWL p p . 196-220, 2 0 8 ,
 145

2 1 6 - 1 7 ) . he insists that reason's investigation of itself

produces knowledge (hence the title "Wissenschaftslehre") .
Fichte is very much aware of the "circle" involved in
this self-investigation: but he is equally aware of the
impossibility of avoiding it without losing self-conscious-
ness and so an accurate account of "the possibility of
self-consciousness" (2nd Intro./ p . 3 7 ) . For the self to
try to abstract from itself ". . . is a contradiction/
since it is impossible that what does the abstracting
should ahstract from itself" (2nd Intro., p . 71; GWL p p .
98, 202, 2 5 2 ) . By beginning with itself, the "highest
principle" or the "Absolute thesis" (2nd Intro., p . 73;
GWL pp. 1 1 4 - 1 5 ) , pure reason can deduce everything
lower" while not losing itself in the p r o c e s s . A c o n -
sequence of this procedure is that Fichte seems to deduce
both the a priori and the a posteriori; they are " . . .
merely two points of view, to be distinguished solely
by the mode of o u r approach" (1st Intro., p . 2 6 ) . Most
importantly, intuitions seem to be deduced, i.e., e q u i v a -
lent to concepts of intuitions. Yet that is not entirely
possible; the "thing in itself" behind intuitions is
finally a "problematic concept." If this signifies a
failure in Fichte's deduction, however, at least it is
a fully self-conscious failure.

In conclusion, Fichte's modifications of, and advance

over, Kant can be restated as follows. Kant's T r a n s c e n -
dental Analytic (CPR) ends with a chapter entitled "The
Ground of the Distinction of all Objects in General into
Phenomena and Noumena," a chapter which serves as the
transition to the Transcendental Dialectic. This of
course is no accident; the distinction between phenomena
and noumena, first suggested in the Transcendental
Aesthetic, and then established by the Transcendental
Analytic, also serves to define the dialectical use o f
reason. From Fichte's point of view, however, the d i s -
tinction is not only a transition to dialectic, it is_
dialectical, insofar as it is a distinction of pure reason.
The distinction cannot be established in the Transcendental
Analytic, let alone in the Transcendental Aesthetic,
before Tt is deduced from reason by reason itself. Fichte
in effect deduces it in the three Grundsätze, i.e., in
part I of the GWL, before the sections on theory and
practice. In Kant's Transcendental Dialectic, the d i s -
tinction between phenomena and noumena is invoked in
order to resolve an antinomy between freedom and nature.
At the end of part II of the GWL, Fichte analyzes this
antinomy as a contradiction between infinitude and
finitude. In Fichte's terms, the ground of the third
antinomy (and so of the Categorical Imperative) lies in
the I's reflection on itself. If on the one hand the I
reflects upon itself and so determines itself, the not-I
(nature) is infinite and unbounded (the realist posi-
tion) . If on the other hand the I reflects on the not-I
 146

and so determines it, the I (freedom) is infinite (the

idealsit p o s i t i o n ) . Kant stopped at this point, and
"solved" the conflict through the distinction between
phenomena and noumena.

However, Fichte points out that there is still a

higher reflection to be accomplished, namely the r e -
flection that the antinomy is itself a product of reason
(GWL p . 2 1 7 ) . The same is true f o r t h e distinction b e -
tween phenomena and noumena. T o perform this higher
reflection is to become fully self-conscious, and so t o
go beyond the antinomy, though it does not so much remove
the antinomy as to place it on a more complex level.
For Fichte, this more complex level must in turn be
transposed by reflection onto still another level, and
so on ad infinitum. Hegel's task is to close the circle
of reflection and so to transform self-consciousness into
complete, and discursive, self-knowledge.

Pennsylvania State university

 147

NOTES

^"This essay was presented, on October 23, 1976, to

the American Graduate Student Conference in Philosophy,
at the American University. In an expanded form, it
was submitted to the Department of Philosophy at the
Pennsylvania State University in completion of the Master
of Arts degree. I am indebted to Dr. Thomas Seebohm for
his helpful comments and criticisms. The present essay
is intended solely as an introduction to Fichte and not
as a thorough study of his thought.
2
J. G. Fichte, Grundlage der gesamten Wissenschaft-
slehre (1794) and Erste und Zweite Einleitungen in die
Wissenschaftslehre, trans. P. Heath and J. Lachs (New
York: Appleton, 1 9 7 0 ) . All references to the W l s s e n -
schaftslehre (abbreviated GWL) and to the Introductions
advert to this translation and are included directly in
the text. Concerning the title "GWL," W. G. Jacobs notes
in his intoduction to the Meiner edition (Hamburg: 1970),
p. viii: "Fichte schrieb hier, wie der Titel sagt, kein
Werk der ersten Philosophie, sondern sozusagen Meditationes
de prima philosophia: er Uberlegt nicht metaphysische
Probleme, sondern wie überhaupt Probleme der Metaphysik
zu begründen sind." It is also worth noting that Hegel's
criticisms of Fichte in the "Differenzschrift" remain
substantially unchanged in Hegel's later w o r k s ; thus
Fichte's Jena period is the most important for our p u r -
poses. Cf. H. Girndt, Differenz des Fichteschen und
Hegeischen Systems (Bonn! Bouvier u. Co., 1 9 6 5 ) , p . x:
"Hegel hat seine in der "Differenzschrift" dargelegte
Auffasung von der Philosophie Fichtes später nie
revidiert. . . . "
3
I . Kant, "The History of Pure Reason," Kritik der
Reinen Vernunft, trans. N. K. Smith (New York: S t . MartTn's
Press, 1965) . See also B 274 ff. All references to the
Critique of Pure Reason (abbreviated CPR) advert to this
translation and are included directly in the text. Cf.
M. Guerolt, "L'Antidogmatisme de Kant et de Fichte," in
Etudes sur Fichte (New York: Georg Olms Verlag, 1 9 7 4 ) ,
II, p . 1 7 : "L'antidogmatisme de K a n t et celui de Fichte
se reveilent bien identiques dans leur esprit: ce qu'ils
affirment tous les deux, en face du dogmatisme, c'est
l'absoluite de l'acte du sujet, auqüel ils subordonnent
l'objet."
4
Kant, "Erklärung in Beziehung auf Fichtes W i s s e n -
schaftslehre, " Ak. Bd. X I I , p . 370, cited by Ä. Philonenko,
 148

La Liberte dans la Philosophie de Fichte (Paris: Librarie

Philosophique J. Vrin, 1 9 6 6 ) , pp. 105-6. It should be
noted that Fichte distinguished the Kantians from Kant;
the .Kantians misinterpreted Kant, in Fichte's view. For
Fichte's reply to Kant's disclaimer, see 2nd Intro., p .
43, n. 4.

5 k

Fichte cites several passages in which Kant him-

self seems to say that the CPR is merely propadeutical:
2nd Intro., p . 5 1 , n. 7.

^Intellectual intuition and the moral law are

closely related; the moral law warrants belief in the
reality of the object of intellectual intuition. See
2nd Intro., p p . 40-41. One might also say that since
intellectual intuition cannot be reduced to concepts,
there is for Fichte a "giveness" for consciousness
analogous to the "giveness" in Kant's sensible intuitions.
7
For a full discussion of intellectual intuition in
Fichte, see A. Philonenko, ibid., part I, ch. 5. Philonen-
ko opposes Hegel's interpretation of intellectual intuition
in Fichte, and finds the GWL and the two Introductions
reconcilable. I believe that the correct solution lies
between the positions of Philonenko and Hegel. In some
sense the Absolute I is reached by "abstraction" as well
as by intellectual intuition (cf. GWL p p . 2 1 5 - 1 6 ) .
g
See note (f 4 above.
g
Fichte barely mentions space and time as deduced
categories in the GWL. See p . 171, n. 3; p p . 1 9 4 , 201,
275.

T h e GWL is divided into three parts; the "Funda-

1 0

mental Principles. . . ," the "Foundation of Theoretical

Knowledge," and the "Foundation o f Knowledge of the
Practical." The first part is neither theoretical nor
practical, but rather the basis for both theory and p r a c -
tice. Part II corresponds roughly to the Transcendental
Analytic (CPR); part III combines elements of the T r a n - ~
scendental Dialectic (CPR), Transcendenta1 Doctrine of
Method, and the Critique of Practical Reason. The T r a n -
scendental Aesthetic (CPRl^Ls hardly present at all, and
morality is barely mentioned (see pp. 2 3 0 , 259, 2 8 5 ) .
Fichte does not distinguish "analytic" from "dialectic;"
the latter term is not used at all in the GWL.

C£.
1X
A. Philonenko, ibid., pp. 1 6 4 , 307-9, 334.
Philonenko tries to interpret the course of the GWL
as the opposite of that of the CPR. There is some truth
to this, but part II of the GWL also follows the order
 149

of the Transcendental Deduction (CPR. B e d i t i o n ) . It

should be noted also that part III of the GWL includes
a discussion of feelings and drives, which explains the
"facticity" of experience. Fichte attempts to produce
a holistic analysis of man (GWL p p . 259, 266, 280.)
12
In a sense Fichte's method is reminiscent of
Leibniz, insofar as it runs together concepts and i n t u i -
tions. In another sense, Fichte's "deduction" is r e m i -
niscent of Spinoza's method, although the similarity
between the two is not very deep. For a discussion of
Fichte's method and its relation to Spinoza, Leibniz,
and others, see M . Guerolt, L'Evolution et la Structure
de la Doctrine de la Science chez Fichte, (P"ä~ns:
Societe D'Edition: Les Belles Lettres, 1930) , I, p p .
18-51; and R. Adamson, Fichte (New York: 1903), p. 134.

*" For Kant, analysis splits up two concepts, one of

3

which is "contained in" the other. Relations of species

to genus, as w e l l as tautologies, fall within the range
of analytic judgements (CPR, Intro., i v ) . Analysis is
the explication of a concept according to the principle
of contradiction, and thus is the work of Verstand (CPR
B 1 9 1 - 9 2 ) . Kant also allows for negative and positive
judgements in analysis (CPR A 1 5 4 ) , but Fichte associates
negative judgements with antithesis, and positive judge-
ments with synthesis.
14
One could object that in beginning with a logical
principle, Fichte assumes a special organon of logic.
But Fichte is only assuming the principles of any d i s -
course whatever; he assumes what he must if he is going
to speak, and then shows how such principles, and so
speech in general, are the result of the activity of the
Absolute I. The principles of discourse are the same
as, or at least modeled on, the principles o f self-
consciousness (GWL pp. 2 6 - 2 8 ) . Hegel criticizes Fichte's
"beginning" in the Wissenschaft der Logik, trans. W . H.
Johnston and L. G. Strüthers (New Y o r k : Humanities Press,
1 9 6 6 ) , I, pp. 87-89. For Hegel, Fichte's starting point
is arbitrary and subjective, since it is not derived
adequately. An adequate derivation must begin from
"ordinary consciousness" and proceed through its "inner
necessity."
15
This formulation of the issue is taken from S.
Rosen's discussion of the GWL in G. W . F. H e g e l : an
Introduction to the Science of Wisdom, (New Haven: Yale
University Press, 1974) , p . 10*4.

C f . W. G. Jacobs, ibid., p . x i :
1 6
"Näher liegt e s ,
das absolute Ich mit Gott gleichzusetzen; aber wenn auch
Fichte in der GWL nicht vom Unterschied zwischen Gott
 150

und absolutem Ich redet, ist deren Gleichsetzung ein

Missverständnis." Jacobs correctly notes that "setzen"
is not equivalent to "schaffen." The Absolute I becomes
God in some of Fichte's. later w o r k s , such as Die Bestim-
mung des Menschen, part I I I . In that w o r k , however, God
is the moral order of the world. Note also that Fichte
separates the personal I from the Absolute I; e.g., in
the 2nd Intro., pp. 73-75, 83-84.
17
Fichte makes this explicit in the GWL p . 2 4 2 . This,
together with Fichte's teaching about the immanence of God
(as the moral order of the w o r l d ) , led t o charges of a t h e -
ism against him. For Fichte's two references to atheism,
see GWL p p . 220, n. 1, and 245, n. 4. Kant implies
several times that the divine understanding, consisting
solely of intellectual intuition, cannot be self-conscious
(CPR, B 71-72, 148 f f . ) .
18
For Fichte, to assert that the finite does not
depend on the infinite, is to assert "absolute finitude"
(GWL p . 1 6 9 ) , a self-contradictory concept. To think the
concept is to transcend it; to become self-conscious about
a thought is to go beyond that thought. In thinking
"absolute finitude" one proves that it is not absolute
(GWL pp. 1 9 2 - 9 3 ) . Note that the I is "infinite" in
Fichte in the sense that it is limited by itself only;
the same is true in Hegel (Phänomenologie des Geistes-
[Hamburg: F. Meiner Verlag, 1 9 5 2 ] , ch. V I , p p . 3 1 3 - 1 6 ) .
19
Hegel, Differenz des Fichteschen und Schelling-
schen Systems d e r Philosophie^ (Frankfurt am M a i n :
Suhrkamp Verlag, 1 9 7 0 ) , p . 2 6 .
20
Hegel, Science of Logic, ibid., p p . 88-89. Intel-
lectual intuition in Fichte does seem to be more like a
feeling than a thought, since what one "sees," i.e., the
Absolute I, is pure activity, and so has no form o r d e -
termination.
21
Hegel, Differenz. . . , ibid., p . 56: "Das Wesen
des Ich und sein Setzen fallen nicht zusammen: Ich wird
sich^nicht objectiv"; and p . 6 5 . See also the V o r l e s u n -
gen über die Geschichte der Philosophie, t r a n s . E , S.
Haidane and F. H. Simson (London: Routledge and Kegan
Paul, 1 9 5 5 ) , III, p . 494; and part I of the Enzyklopädie,
trans. W. Wallace (London: Oxford university P r e s s , rpt.
1 9 0 4 ) , p p . 90, 120: Kant is a subjective idealist, and
Fichte does not go beyond Kant in this respect.
22
Hegel, Differenz. . . , ibid., p . 6 7 : "das System,
das vom Sich-selbst-Setzen ausgeht, führt die Intelligenz
su einer bedingten Bedingung in ein Endloses von E n d l i c h -
 151

keiten fort, ohne sie in ihnen und aus ihnen w i e d e r h e r -

zustellen"; Vorlesungen. . . , ibid.. pp. 49 4-95;
Enzyklopädie, ibid.., p . 87.
2 3
H e g e l , "Einleitung," Phänomenologie. . . , ibid.;
Enzyklopädie, ibid., pp. 84-85, 116-17; Science of Logic,
ibid., pp. 73-75. For a discussion of Hegel's criticisms
of Kant and Kant's failure to be self-reflexive, see J.
Smith, "Hegel's Critique of Kant," Review of Metaphysics,
26, N o . 3 (1973), 438-60; and R. Pippin, "Hegel's
Phenomenological Criticism," Man and World. 8, N o . 3
(1975), 296-314.
 152

CORRECTION

AUSLEGUNG IV, 1, p . 35 ("Transcendental Arguments:

Verificationism or Parasitism?" by Douglas E h r i n g ) : The
sentence, "On the other hand, the correct2 use of "in-
ner experience" concepts requires the use of "physical
object" concepts" should read as follows:
On the other hand, the correct-use of an e x -
pression would imply its correct^ u s e . The
"parasitism" argument might show that the
correct, use of "inner experience" concepts
requires the use of "physical object" concepts.