Frank-Manfred Self-Conssciousness PDF

CH 5_f4_53-136 11/23/04 9:28 AM Page 53
Manfred Frank
Fragments of a History of the Theory of
Self-Consciousness from Kant to Kierkegaard
ABSTRACT
In the development of modern philosophy self-consciousness was not generally or unanimously given important consideration. This was because philosophers such
as Descartes, Kant and Fichte thought it served as the
highest principle from which we can deduce all propositions that rightly claimed validity. However, the Romantics
thought that the consideration of self-consciousness was
of the highest importance even when any claim to foundationalism was abandoned. In this respect, Hlderlin and
his circle, as well as Novalis and Schleiermacher, thought
that self-consciousness, itself, was not a principle but must
be ranked on a minor or dependent level, and presupposed the Absolute as a superior but inaccessible condition or ground. This reservation did not hinder them
from recognising that the foundationalist Fichte was the
first to have shown conclusively that from Descartes, via
German Rationalism and British Empiricism, up to Kant,
self-consciousness was misconceived of as the result of
an act of reflection by which a second-order act bent
back upon a first-order act that is identical to itself. This
conception entailed circular entanglements and infinite
regresses, and was too high a price to pay. Whereas
Fichte thought pre-reflexive self-awareness was a philosophical principle, the Romantics and their vehement critic
Kierkegaard, abandoned the idea of self-consciousness
as a foundational starting point of philosophy. Instead,
they founded self-consciousness on transcendent Being,
a prior non-conceptual consciousness (feeling) and
reproached Fichte for having fallen back into the repudiated reflection model of self-consciousness.
CH 5_f4_53-136 11/23/04 9:28 AM Page 54
KEYWORDS: Self-consciousness, Enlightenment, Romanticism, Kant, Fichte, Hlderlin, Sinclair,

Herbart, Schleiermacher, Kierkegaard
If one were asked to specify the lowest common denominator of modern philosophy from Descartes to Sartre, it would not take much time to decide on
the reply: this common denominator is self-consciousness. Not only did
Descartes rely on self-consciousness to guarantee ultimate certainty in the
form of the fundamentum inconcussum for an epistemic self-orientation threatened by doubt; he also considered it to be a principle of deduction for all
potentially true propositions. Leibniz followed him in this respect. David Hume
was prevented by his empiricist premises (which only allow isolated sense
experiences and the reflection on them as sources of knowledge) from recognising a self-consciousness which could remain identical with itself over time.
However, in the appendix to his major work, he confessed that in this domain
he was not only uncertain, but perplexed. The connectedness of the bundles
of perception required a unitary self-consciousness, which he could only
reject on the basis of his own principles. All my hopes vanish, he admitted,
[. . .] when I proceed to explain the principle of connexion, which binds
them [sc. the successive perceptions as distinct existences] together, and
makes us attribute to them real simplicity and identity; I am sensible, that
my account is very defective, and that nothing but the seeming evidence of
the precedent reasonings coud have inducd me to receive it. [. . .] / [. . . In
fact,] I cannot discover any theory, which gives me satisfaction on this head.
In short there are two principles, which I cannot render consistent; nor is it
in my power to renounce either of them, viz. that all our distinct perceptions
are distinct existences, and that the mind never perceives any real connexion among
distinct existences [. . .].
For my part, I must plead the privilege of a sceptic, and confess, that this
difficulty is too hard for my understanding. I pretend not, however, to
pronounce it absolutely insuperable. Others perhaps, or myself, upon
more mature reflexions, may discover some hypothesis, that will reconcile
those contradictions.1
This was Kants achievement. Like Descartes and Leibniz before him, he
treated the knowledge possessed by conscious beings, not only of their mental states, but of the coherence of these states in the unity of conscious life,
as a principle of deduction for all true propositionsin Kants terminology,
54 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 55
Judgements. He also undertook to derive the thought of possible objectivity from the application of this principle to sense experience. Like Descartes,
Kant was so preoccupied with the foundational function which he attributed
to the I in his philosophy that he never really paid attention to the structure of this principle itself. To his surprise, this was to become a chief preoccupation of his pupils and successors. And that will be the topic of the
following discussion.
I
Kant described self-consciousness as the highest point of (theoretical)
philosophy: all employment of the understanding, even the whole of logic,
and transcendental philosophy, depend on it.2 A little later, Kant presents it
again, this time in the traditional form borrowed from Leibniz as the unity
of apperception . . . which is the highest principle in the whole sphere of
human knowledge.3
Kant was by no means alone in this unbending belief that the consciousness
which thinking beings have of themselves is apt to serve as the highest principle of philosophy. Rather, he stands in the history of a tradition, whose
great figures are Descartes, Leibniz, and Rousseau. Kant was perfectly aware
of this. His originality is not to be found here, but in the function which he
attributes to the principle of self-consciousness with regard to the explanation of the objectivity of our representations. For Kant, speculation about the
nature of the self was in no sense an end in itself (as it was later to be, in a
certain sense, for Fichte and the early romantics). He was rather concerned
with drawing wide-ranging conclusions from a certain characteristic of the
I think which is known to itself with Cartesian self-evidence: this feature is
its identity. Identity is, of course, something different from analytical unity.
The latter term indicates that property of the I which is a common characteristic of all representations accompanied by the I (namely, that they all
have the same feature of being able to be accompanied by the one, constant
I think). Identity (Kant also says synthetic unity), however, indicates the
property of the I by virtue of which it can not only be connected transversally, as it were, to all representations, as in the previous use, but also links
these representations together horizontally.4 To achieve this, a finite set of
rules is required for connecting the individual representations: these are the
A History of the Theory of Self-Consciousness 55
CH 5_f4_53-136 11/23/04 9:28 AM Page 56
categories. Categories are predicates with such a wide extension that something cannot both be an object, and not be determined by them. In other
words, categories are predicates of objects in general. When attributed to
complexes of intuitions, they can be seen as constituting condensed judgements, and there will be as many of them as there are forms of judgement.
Kant derived this idea from the Savoyard vicar, at the beginning of Book IV
of Rousseaus Emile.5 The basic idea can be sketched rapidly. While I feel
myself to be passive as a sensible being, I experience myself as active insofar
as I think. Thinking is judging. In the act of judgement, different representations deriving from various viewpoints are grasped as a unity, and this unity
is recognised through the veritative is, common to all statements. Whoever
uses the word is meaningfully in relation to representations connects them
together through a common feature: their concept, which captures what is
common to them. If this concept is applied to something which existsthat
is, if the judgement concerning the underlying state of affairs is rightthen
this something is constituted as an object. Its objectivity consists in, and can
be proved by, the fact that, at any time, it can be translated into a set of true
judgements. Thus, the objectivity of the individual things is a function of the
truth of judgements concerning states of affairs.6 At the origin of objectivity
there stands the identity of the I, which is operative in the is of judgement.7
In this wayKant merely presented the argument in more detailthere arises
an indissoluble connection between the identity of the I, truth (as a property of statements), and objectivity (as a property of representations). To
demonstrate the necessary character of this connection with the requisite precision was the sole ambition of Kants transcendental deduction of the categories. The principle of this deduction, the I itself, was important to Kant
only because of itsadmittedly indispensablefunction as a principle; in
other words, because of the consequences that arise for the explication of the
ground of objectivity. It seems to have been one of the greatest surprises Kant
ever received that his pupils began to argue, above all, about the structure
of this I, and strove to demonstrate that Kants philosophy failed to provide
an adequate description of it.
If this objection were sound, it would be a serious one. For it concerns nothing less than what Kant himself described as the highest point of his philosophy. If the self-evidence of this highest point were put in question, then
56 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 57
the transcendental deductionwhich was itself never questioned by the

Kantianswould lose its demonstrative force. The objection in fact runs as
follows: entirely absorbed in this task of a deduction of the categories (as valid
for all appearances), Kant neglected to show the self-evident character of his
highest principle; indeed, he was obliged to deny that it could be known.
This is becauseaccording to Kants own directives for the use of the word
knowledge (Erkenntnis)only objects can be known. Objectivity is the result
of the intervention of the unity of the self into the chaotic manifold which is
delivered to us by the senses. This intervention occurs through the categories
that allow me to move from one representation to another, according to rules
that have a priori evidence, so that everything representable fits together into
a continuous conception of the world. A manifold of intuition determined by
the categories is precisely what Kant terms cognition. It is obvious, however, that the principle in whose name this determination takes place, namely
pure self-consciousness, cannot become an object of knowledge, because it is
not sensible. Kant, not without some embarrassment, terms it an intellectual representation of the spontaneity of a thinking subject.8 Conversely, only
what can be cognised can claim to be an object, that is, a secured item of
extant knowledge (component of our conception of the world). In fact, Kant
does speak of the objective unity of the I.9 Since, at the same time, he denies
the possibility of an intellectual intuition (this would be a synthesis of the
understanding in which the intuitional content would be not only unified,
but produced), the question arises of how self-consciousness can be given
the status of an objective existent.
In other words: insofar as only the act of cognition (Erkenntnis) can give one
objective knowledge (Wissen), the pure I cannot be known as an objective
existent. As soon as this conclusion is accepted, however, it is clear that it
has disastrous consequences for the self-evidence of the highest point of theoretical philosophy.
Kant draws a strict distinction between the being (Sein) of self-consciousness
and its appearance to itself (Sich-Erscheinen) in time. The latter is only thinkable as an empirical fact (and therefore as an object of knowledge). By contrast, the former, the naked being of the self, as a condition of possibility of
its self-appearance, remains a mere presupposition.10 What is presupposed
in this way is an existent bereft of any property or quality;11 all I know about
CH 5_f4_53-136 11/23/04 9:28 AM Page 58
it is that it is not: what it is or how it appears to itself.12 All I grasp of

it is the actus purus of its naked being.13 Since every appearance of the I has
its being as a presupposition, one might say that its existence precedes its
essence. Only the appearancethe essenceis accessible to knowledge. Kant
remarks, in this context, that the I only has access to its being via predicates
through which it both determines itself and disguises its being.14 This being
itself, which is pre-predicative and pre-categorical (thus, non-propositional,
since categories are condensed propositional forms), is inaccessible to knowledge. And yet there must be a consciousness of it, if the highest point of philosophy is not to lose its self-evidence. It might be objected that talk of the
being of consciousness (in contrast to its self-appearing) refers only to its
mode of being, and thus to one of its qualitative determinations, much in the
same way as does the formulation I exist as intelligence,15 which does not
mean that intelligence, above and beyond the fact of appearing as intelligence, also has a subsistent being external to this determination.16 And yet it
is transphenomenal existencewhich alone could be described in a radical
sense as naked or featurelesswith which Kant is concerned.
Fortunately, to demonstrate this, I can refer to a short, but highly significant
note which Kant added to the second edition of the chapter on the paralogisms.17 Firstly, Kant here distinguishes clearly, as he does elsewhere in this
context, between the existence expressed by Descartes statement Cogito sum,
and concepts such as reality (Realitt) and being (Dasein). These latter concepts refer to two different types of categories, those of quality and those of
modality. Furthermore they require completion through an intentional content, and are therefore not suitable for characterising the essence of the purely
determining I. Secondly, Kant distinguishes the bare existence of the pure
I from any actual relation to sensation, which is provided by the being of
(sensible) objects. Under these circumstances, what mode of consciousness
could correspond to the pure being of self-consciousness? Certainly not intuition, for intuition is directed towards the sensible world, and the being of
the pure I bears no trace of the sensible. But neither can it be thought, since
thinking is never immediate, but reaches its object by means of a concept, an
analytically isolated characteristic which this object has in common with many
others.18 The existence grasped in pure apperception therefore cannot be
accounted for in either the sensible or the conceptual components of our faculty of cognition (Erkenntnisvermgen). It is definitively situated this side of
58 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 59
the threshold beyond which the distinction between the intuitional and the
conceptual becomes possible, and takes place. Kant suggests that this existence in pure apperception corresponds to a weird construction, an indeterminate empirical intuition or perception,19 which he had earlier qualified
as inner experience or inner perception.20 This inner and yet empirical
perception is entirely distinct from, and has nothing to do with, what Kant,
in other contexts, calls inner perception, where the term is synonymous
with the inner sense, through which sensuous appearances of the (empirical) I are experienced.21 This can be easily demonstrated by the fact that
Kant characterises this inner self-perception or self-intuition as purely intellectual. He adds that it bears in itself the source of a pure spontaneity,22
which the empirical I obviously lacks, but which also includes existence.
Regarding this, Kant remarks that it is given to apperception [as] something
real, and indeed for thought in general, thus not as appearance.23
In Kants view it is clear that pure apperception includes the immediate consciousness of its own existence, and that this consciousness, although preintuitional, nonetheless includes the perception of an existent. This is because
existence cannot be attained by thought alone; it must be given if there is to
be consciousness of it. I pass over a lot of parallel quotations, because Kants
problem, though enigmatic in its matter, is clearly posed in its terms. As Kant
himself states:
The I think is . . . an empirical proposition, and contains in itself the proposition, I exist . . . [It] expresses an indeterminate empirical intuition i.e.
perception (thus shows that sensation, which as such belongs to sensibility,
lies at the basis of this existential proposition) but the I think precedes the
experience, which is required to determine the object of perception through
categories with respect to time; and the existence here is not a category.24
This experiential aspect, however, does not affect the fact that the sensation
here referred to precedes the sort of experience through which our intuitional
faculty receives sensuous material from outside, and passes it on to the understanding to do the categorical work. The existence of the pure cogito is
neither intuition nor category. It is epistemically classified as an inner perception, which must be strictly distinguished from the perception of psychic
objectivities, as these appear to inner sense. In the Metaphysical Principles of
Natural Science, Kant notes in the same way:
CH 5_f4_53-136 11/23/04 9:28 AM Page 60
Since the thought I is in no sense a concept, but only inner perception,

nothing can be deduced from it (except the utter difference of an object of
inner sense from what we think of as an object of outer sense), and consequently not the continuous existence of the soul as substance.25
In order to grasp the peculiar aporia of Kants basic argument, it is worth

examining Kants conviction that Existence (Being) is not a real predicate.26
The property of reality constitutes one of the three sub-divisions of the category of quality; one of the qualities of the object concerned is judged by
means of it, its realitas, its thinghood or mode of being.27 Existence is not a
real predicate means then: existential judgements provide no judgement concerning the mode of being (qualitas, realitas, quidditas): being is not a property in this sense. If I say I exist as intelligence, then intelligence is a real
predicate that belongs to the quality of the cogito, and the judgement itself is
analytic.28 Whether such a cogito also exists (beyond its being-thought-of),
ndependently of its property as thinking, is left undetermined. And yet
according to Kant the proposition cogito implies the empirical state of affairs,
I exist. The question is: in virtue of what does the claim of the existential
judgement exceed the claim of the judgement about reality? Answer: in that
existential judgements depend on sense-experience, where the latter judgements dont.
In a short text dating from 1763, The Only Possible Ground for a Proof of the
Existence of God, Kant first presented in a systematic form his thesis that the
indefinite verbal expression, to be, has two different meanings. Kant argues
here that being (Sein) is the object of a relative positing, whereas existence
(Dasein) is the result of an absolute positing. The positing of something is relative when this something is posited in relation to (relative to) something
else, for example, in the statement a is B. Here a is only posited in relation to its being B, but not absolutely. This is the type of positing which is
operative in the statement, I am as intelligence. By contrast, the positing of
a would be absolute, if it were in relation not to B, but to a itself. In this
case, as Kant says, existence (Dasein) is attributed to a.29 If I say this a
exists, then I do not refer to anything else, or to any determination of a;
rather, I posit a as existing, independently of any relation. Relative and
absolute positing thus occur in the form of a judgement. Usually judgements
link together different classes of representations. In the simple singular state60 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 61
ment, a content of intuition is linked with a concept (a is B); bothin Kants

terminologyare real determinations. In the judgement a exists, by contrast, no concept is imposed upon the intuitional content, but rather being
is attributed to the thing concerned. It is posited as such, without any additional characterisation. This attribution of existence, which oversteps the
purely conceptual determination of the thing, requires sense-perception. Indeed, only sense experience can convince me or dissuade me whether a concept has an existing content, or whether I am merely thinking it. According
to Kant, the categories of modality (under which existence falls) have the
peculiar feature that they in no way enlarge the concept to which they are added
as predicates, but merely express its relation to our faculty of cognition.30
Now once more according to Kant, sense-perception is the exclusive character of existence (Wirklichkeit). The aporetical result is: our knowledge of the
existence of the I must somehow rely on perception. And thats exactly what
Kant is asserting. In more detail: Dasein, Wirklichkeit, and Existenz (terms
which Kant usually employs synonymously), merely bear on the question
whether such a thing is so given to us, that the perception of it can, in all
cases, precede the concept. For the fact that the concept precedes the perception signifies the concepts mere possibility; the perception, however, which
delivers the material content to the concept, provides the exclusive character of existence.31 If this is so, then the mark of being absolutely posited coincides with that of being experienced through the senses (only in this way can
our cognitive faculties take up material which derives from a source that is
independent of them).
I do not intend, in this context, to investigate the problematic and ambiguous character of this thesis. I am only concerned to make clear the reasons
which motivated Kant to correlate the existential judgement I amin which,
what is judged, is pure, not empirical, apperceptionwith an indeterminate,
empirical intuition, or perception (and thereby with sensation). The only
way in which consciousness can make contact with existence is through that
specification of intuition which Kant calls sensation (Empfindung). Only sensation can testify to the absolute positing of the cogito. If this is the case
and Kants premises exclude any other explanationthen there must be an
element of intuition correlated with the auto-perception (or: ad-perception)
of the cogito, its pure spontaneity notwithstanding. In other words, Kant
CH 5_f4_53-136 11/23/04 9:28 AM Page 62
cannot avoid bringing into play a possibility that he passionately rejects

the possibility of an intellectual intuition.32 It is a question of intuition here,
because only receptivity, as opposed to the spontaneity of intelligence,
could testify to existence (absolute positing). The intuition is at the same time
intellectual, because it is grounded in the pure spontaneity of the understanding and brings with it the idea of a complete determination. Thus, the
pure I exists, and we have an immediate (in other words, preconceptual and
also prereflexive) awareness of it.33 In other words: in order to justify the existence and at the same time the Cartesian transparence of the cogito, Kant has
to admit an intellectual intuition as a fact of our conscious life.
Obviously, this is a conclusion that stands in an irresolvable contradiction to
Kants conviction that the two sources of our cognitive faculty, intellect and
sensibility, are entirely independent of one another. For Kant, the thought
that I think is entirely free of any element of intuition, while intuition in
no way requires the function of thought;34 for without any function of
the understanding appearances can be given to us.35 Furthermore, in other
contexts, Kant stubbornly opposed the possibility of immediate self-apperception (in this he follows Leibniz, who renders aperception as rflexion,
for example in his Principles of Nature and of Grace). The transcendental I,
says Kant in a famous passage, is a completely empty representation, which
is only known through the thoughts which are its predicates, and of which,
in itself, we can never have the slightest concept . . . but can only revolve
in a perceptual circle, since any judgement upon it has already made use of
its representation.36
Thus, the prepredicative being of pure apperception remains unelucidated; or
rather, since we cannot renounce it, we have no other recourse than always
to pre-suppose it.37 A corresponding formulation in the A-Version of the chapter on the paralogisms had already indicated this:
Henceforth we can no longer derive [the concept of a substantially enduring
self] from [the mere concept of the identical self], since this concept constantly
revolves around itself, and brings us no further in any question concerning
synthetic knowledge [. . .]. When I strive to observe the pure I amidst the
changing representations, I have no other basis for my comparisons than
once again myself, with the general conditions of my consciousness. Consequently, I can only give tautological answers to all questions, by attribut62 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 63
ing to the properties which belong to myself qua object the concept of myself
and of my unity, and thereby presupposing what I wanted to make known.38
In other words, if I want to explain who I am, I must declare certain perceivable qualities to be mine. But as soon as I consider the legitimacy of this
self-attribution, it obtains that I can only undertake it if I was already familiar with the meaning of I or mine. Thus, I find myself compelled to take
myself (or the concept I) as the point of comparison which is to enlighten
me as to what property defines me, and in this way I presupposes what I
was supposed to learn through experience. So, the explanation, Kant suggests, is a completely circular one.
But the highest point of philosophy can scarcely be admitted to be an
unfounded presuppositionin another context Kant suggests that it could be
called the subreption of hypostatised consciousnesseven less its description by means of a circulus vitiosus. And yet the description which Kant himself gave of his principle bears the unmistakable structure of such a circle. It
refers backperhaps against the original intentions of its authorto the
untenable reflection model of self-consciousness, which always presupposes
that which it is to demonstrate, and which stands at the centre of the Fichtean
critique of Kant.
One must bear in mind that Kants aporia has two dimensions, which are
closely related, and yet distinguishable: one epistemological, and the other
ontological. In the first instance, the issue is that of how a pure, non-objective subject of consciousness can acquire knowledge of itself, without objectifying itself (which would presuppose, in a circular manner, that the
self-objectifying process is precisely that of a subject).39 In the second instance,
the question raised is how a pure, and thus non-sensible spontaneity can
acquire a consciousness of its own being, given thatby definitionbeing
can only be authenticated by sensation. It appears that both problems can,
and must, be solved at the same time.
To my knowledge, there is only one single remark in the rest of Kants work
that unambiguously shows that he was aware of the dimension and of the
dual nature of this problem-syndrome. I am referring to a posthumous reflection,40 which relates to practical rather than theoretical philosophy. Indeed,
the need to account for the intelligibility of the principle of practical philosophy, on the one hand, and its existence, on the other, runs up against the
CH 5_f4_53-136 11/23/04 9:28 AM Page 64
same problems which we have already encountered with respect to the principle of theoretical philosophy. The principle is freedom, and both its intelligibility41 and its reality are in question (reality means existence in this
context; Kants terminology is not entirely consistent). I will attempt to reconstruct the general context within which this set of problems appears.
In the Critique of Practical Reason, Kant is concerned to show that the objective validity and universality of the moral law cannot be the outcome of an
individual empirical maxim of only regional applicability. Here I shall pass
over the various formulations which Kant gives of the categorical imperative (categorical here meaning unconditionally valid), and consider only
the fundamental idea, which justifies the supra personal and general character of commandments. The principle of morality is applied in such a way
that all those norms, which cannot be agreed upon by those affected by them,
must be excluded as invalid. The principle that intervenes, in order to make
a general agreement possible, must thus ensure that only those norms that
correspond to a universal will have validity. The categorical imperative can
be understood as a principle that demands the strict universalisability of all
actions and the maxims that prime them. In this way, all norms that contradict this requirement can be excluded. Kant has in mind those internal contradictions that arise in the maxim of an agent, when he or she attempts to
reach a goal with means that are incompatible with the universalisation of
his or her behaviour.42
It is astonishing for the reader who only knows the first Critique that Kant
grounds the universal will prescribed by the categorical imperative on a fact
of reason, for which he claims a priori evidence.43 For in fact only theoretical reason is endowed with a priori valid, and thus universal and objective,
concepts. In contrast to the categories which ground the objectivity of knowledge, a categorical imperative only prescribes an ought, and cannot be
demonstrated in the form of knowledge that can be checked against empirical facts. In other words, it is part of the structure of practical reason, that,
despite its objectivity, its claims to validity can only be raised counterfactually, and this in contradiction to empirical reality, which can never be adequate to it.
We are confronted here with an aporetic structure that is analogous to that
of the cogito as principle of theoretical reason. No knowledge can be adequate
64 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 65
to this principle either, for then it would collapse into the sphere of the sensible. On the other hand, neither the theoretical nor the practical principle
can be inaccessible to any knowledge whatever, since in this case they would
be unintelligible. Neither can existence be denied them, for then there would
be no such principle.
Once again, I can use Kants own formulations to show that he was aware
of the extent of this problem. Although Kant categorically denies to pure
apperception the status of an intellectual intuition in the Critique of Pure Reason,
he does at least consider this status for freedom. Kant argues that practical
reason guarantees the reality of a supersensible object, namely freedom.44
However, this reality (once more in the sense of existence) cannot be that of
an empirical object in space and time, since it is supersensible. On the other
hand, freedom cannot be merely required, it must really exist, if I am to make
the least moral demand on my fellow human beings. We find ourselves in a
situation which is comparable theoretically to that in which we were placed
by the remark on page 422/23 of the B version of the Critique of Pure Reason:
we must explain the existence of a precategorical and supersensible entity,
which functions as the principle of that which has the character of an appearance in the I, and is accessible to knowledge.45 Only the intellectual intuition could have access to such a supersensible reality, which definitely
remains this side of the threshold beyond which the sphere of possible cognition begins. In the posthumous Reflexion to which I have already referred,
Kantat an early datehad already stated this conclusion:
We cannot establish the existence/actuality of freedom on the basis of experience. But we nevertheless have a concept of it throughout intellectual inner
intuition (not inner sense) of our activity, which can be initiated through
motiva intellectualia, and through which practical laws and rules of the good
will itself are possible for us. This freedom is a necessary practical presupposition. Neither does it contradict theoretical reason. For, as appearances,
our actions are always in the field of experience, while as objective data
they are in the field of reason and are approved or disapproved of. Sensibility
is under the laws of the understanding and departs [manuscript ends].46
Similarlyalthough now in a more problematised formulationthis

conclusion can be found in the note to paragraph 7 of the Critique of Practical Reason:
CH 5_f4_53-136 11/23/04 9:28 AM Page 66
The consciousness of this fundamental law may be called a fact of reason,

since one cannot ferret it out from antecedent data of reason, such as the
consciousness of freedom (for this is not antecedently given), and since it
forces itself upon us as a synthetic proposition a priori based on no pure or
empirical intuition. It would be analytic if the freedom of the will was presupposed, but for this, as a positive concept, an intellectual intuition would
be needed, and here we cannot assume it. Nevertheless, in order to regard
these laws as given, and to avoid misconstruing them, it must be stressed
that what announces itself to be originally legislative (sic volo, sic iubeo), is
nothing empirical, but the sole fact of pure reason.47
Towards the end of the second version of the Critique of Pure Reason, Kant
had already given some indications of the connection between the being of
the pure spontaneity of the I, which precedes all appearances, and the subject of freedom.48 Without setting off along this path, Kant considers carefully
the possible discovery
of a spontaneity [. . .] through which our reality would be determinable,
without requiring the conditions of empirical intuition. And hence we would
become aware, that there is something contained a priori in the consciousness of our existence, which can serve to determine our existence (the complete determination of which is possible only in sensible terms) as being
related, by dint of a certain inner faculty, to an intelligible world (which is
of course only thought).49
Here it would be a matter of a self-determination of the I of such a kind

that the appearing I or the will would receive its instructions from the purely
intellectual I.50 However this may be, the mere assumption of a purely intellectual and determining I requires recourse to intellectual intuition (which
alone makes possible an immediate acquaintance with the subject and is able
to attain its being), and which in one stroke secures the intelligibility of the
principles of theoretical and practical philosophy by referring both back to
this act. It is in just this way that Fichte will conceptualise what he terms the
absolute I.
II
Before I turn to Fichtes revolutionary theories of self-consciousness and his
departure from Kant, I will first look at some of the traditional rationalist
66 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 67
and empiricist approaches to self-consciousness that led to Kants explanatory model. Following on from Fichte, I will then attempt to give an outline
of romantic theories of self-consciousness that build on Fichte and will end
with a glance at Schleiermacher and Kierkegaard.
Naturally, Fichtes point of departure is Kant, although his attitude to Kant
is critical. We have seen that Kant was so absorbed in the task of deducing
the thought of objectivity from the identity of consciousness that he paid
hardly any attention to the internal structure of his deductive principle, namely
self-consciousness. Even the question of how the self acquires epistemic access
to itself generates great difficulties for him. He must deny the self the status
of a possible object of cognition, because the term cognition (Erkenntnis)
is reserved by him to those operations in which the intellectual enters into
synthesis with the sensible. At the same time he indicates the circle which
arises from the fact that, in order to grasp itself reflexively, the Ego must
always already have been acquainted with its own objectivity prior to any
self-conception. And that way self-awareness turns out to be a mere presuppositionwhich it cant be if its existence is to be guaranteed.
However, we were forced to conclude that, should this implication prove to
be unavoidable, then the highest point of Kantian philosophy would collapse. Indeed, Kant never questioned the Cartesian self-evidence of the cogito, even though his metaphors are more cautious than those of Descartes. If
this self-evidence is acknowledged, and if the existence of the phenomenon
is thereby guaranteed, then the error must be with the explanatory model
that is employed to make it comprehensible. (A phenomenon can either exist
or not exist, but it cannot be true or false: truth and falsehood are properties
of statements, thus elements of theory).
Despite this, the explanatory model employed by Kant derives from an honourable tradition. Lets call it the representation model of consciousness.
This model assumes that (apart from a few intransitive states such as pain)
consciousness is always the representation of an object, whichto retain the
image of the German Vor-stellungis placed before the eyes of the subject of consciousness, as it were, standing over against it. Every (or rather:
almost every) consciousness, Husserl will later assert, is consciousness of
something (which is transcendent to consciousness). Thus the relation of being
conscious is divided into a subject-pole and an object-pole of representation.
CH 5_f4_53-136 11/23/04 9:28 AM Page 68
Every representation requires someone who represents, and something that

is represented (in the revolutionary year 1789, Reinhold developed from this
structural feature the conception of an entirely New Theory of our Faculty
of Representation).51 Leibniz seems to have been the first to have referred
to the subject of this representation with the nominalised form ce moy.52
With this step the subject of consciousness is identified as the same I of
which Kant and Fichte will speak. It appears as soon as the place of the
object of representation is occupied by the objectified representing agency
itself: as soon as the subject of representation re-presents itself, rather than
something else.
Once again, it is Leibniz who takes the first step when he defines the subject
or the I of consciousness as the result of a connaissance rflexive de cet tat
intrieur.53 Kant, who always equated I-hood (or egoity) with the self-reflexivity of representation, took over this definition, and also the term which
Leibniz provided for it: Apperception. This borrowing may have been mediated by C.A. Crusius, who spoke of (self-) consciousness as a representation
of representations.54 Kant noted in a reflexion dating from 1769: in fact the
representation of all things is the representation of our own state.55 Or:
Consciousness is a knowledge of what comes before me. It is a representation of my representations, it is a self-perception.56
Thus, Kants circular definition of the nature of the transcendental I summarises a whole tradition. Lets give it the label reflection-model of self-consciousness. Its essential feature is that it interprets the consciousness that we
have of ourselves on the model of representation: as the result of a turningback of a representation onto itself, which transforms the representation in
question into an object. Every reflection occurs between two distinct terms;
its paradoxical character consists in the fact that it must then deny this difference, otherwise the goal reached by my turning back on myself would be
something, or someone, else.
Lack of time permits me only to illustrate the mechanisms of this model by
casting two spotlights on the work of two important precursors of German
Idealism, Descartes and Leibniz. In addition, I will cast a glance at the use
of the same model in Anglo-Saxon empiricism.
Descartes describes the cogito as a self-reaction that arises between thinking
in general and a specific form of thinking.57 Thinking splits into an indiffer68 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 69
ent and general consciousness, on the one hand, and a more closely specified consciousness on the other: for example, as willing, perceiving, loving,
comprehending, acting. However, these two (as pense dtermine dune certaine manire) sides of thinking are indissolubly connected: a doubt could
never arise without being supported by a consciousness in general, which
takes cognisance of it by means of that tmoignage intrieur.58 The converse
is also true: in order to doubt, to love, to think, and so on, thought-in-general must specify itself. Thus, doubttaken simply as an example of a specific form of thought (penser dune certaine manire)would be one mode
of being amongst others in which thought-in-general presents itself; of course,
it is only in relation of the latter that the various modes of thought acquire
their peculiar certainty: although I can doubt my love, I cannot doubt the
immediate consciousness which I have of it. The relation is a classic example of the structure of reflection; and thus it has been handed down to us in
Burmans notes of his conversation with Descartes:
Conscium esse est quiddam cogitare et reflectere supra suam cogitationem.59
Elsewhere Descartes speaks of the idea of the cogito as that qui me reprsente
moi-mme.60 Here we have a reflectionan auto-representation of thinkingwhose structure presupposes the identity of the two moments, but cannot ground it.61
This aporia becomes even more acute when one passes from Descartes work
to that of Leibniz. Leibniz distinguishes, even more explicitly than his predecessor, entre la perception qui est un tat intrieur [mais souvent insensible] de
la Monade reprsentant les choses externes, et lAperception, qui est la Conscience
ou la connaissance rflexive de cet tat intrieur.62 It is these actes rflexifs he
explains, qui nous font penser ce qui sappelle moi, et considrer que ceci ou
cela est en nous.63
Thus the problem arises from the difficulty of explaining how a perception
that is described as insensible can become conscious (perceptible, sensible) by virtue of its being reflected (ds quon saperoit de ses perceptions) if
a consciousness of it (although not one based on apperception) did not already
exist.64 In other words, if I had to wait for the light of reflection, in order to
know that I had just perceived something, then I would never perceive anything at all. For either I am perceiving, in which case there can be no question
CH 5_f4_53-136 11/23/04 9:28 AM Page 70
of unconsciousness, or I am not perceiving, in which case unconsciousness

means simply that the corresponding act has simply not taken place. (Reflection
can only discover what was already there: if it finds consciousness, then the
reflected act cannot have been unconscious). Leibniz is of course correct when
he remarks that I do not need to pay reflexive attention to my perception in
order to perceive. But the reason for this is that perception is conscious in
itself and does not need a second, accompanying act to be imposed on it,
in order to acquire this consciousness. When there is consciousness, it is immediate; of course acts of reflection can connect up mediately with this immediate consciousness and raise it to the level of knowledge. However, what is
originally presented is consciousness itself, which is clearly single and does not
appear as the object-pole of a subject of consciousness directed towards it.
In broad outline, this was the situation that confronted Kant in his own
attempts to clarify the phenomenon of self-consciousness insofar as his gaze
was directed towards the predominantly rationalistic tradition of continental Europe. The empirical tradition (above all in Scotland and England) was
no less productive for our theme. But as this tradition only admitted individual sensations as sources of knowledge and considered each of those to
be temporally and numerically distinct, the problem of the unity of consciousness over time was posed with particular acuteness. (The combining
of singular perceptions into bundles of perceptions is carried out in Hume
by the imaginationthus objects, as syntheses of perceptions, are unfounded,
literally imaginary constructsand thus his theory becomes entangled in
the performative contradiction of its own validity/objectivity.) Hume claimed
only ever to stumble on particular perceptions, and never on someone who
like the Cartesian egomight unite them all.65 Since he confuses self-consciousness with consciousness of a substantial ego, and since in fact there is
no (sensuous) perception of the ego, he believes himself to have convicted
the entire Cartesian tradition of incoherence. He overlooked the fact that the
expressions sensation, impression, or perception must analytically contain the property of consciousness (even without an ego), if such psychic
processes are to be distinguished from their physical, or rather their intentional objects. And when Hume speaks of such conscious perceptions, he
makes use of an awareness of consciousness that he can only deny at the cost
of further self-contradiction. Fichtes critique of Kant also applies to Hume:
the latter had self-consciousness, but without reflecting on it. And in partic70 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 71
ular Hume failed to realise that self-consciousness and ego-consciousness are

not the same, so that the critique of Descartes substantialisation of the ego
in no way shows the possibility of a knowledge of consciousness, and thus
of self-consciousness, to be incoherent.
However, Humes radicalism was not characteristic of the Anglo-Saxon/empiricist tradition of the early eighteenth century in general, as we can learn from
the Essay on Consciousness.66 This tradition either assumed an inner sense
which directed itself towards objectified conscious experiences (and thus fell
foul of the same critique as Descartes theory of inspection or tmoignage
intrieur); or (like Ralph Cudworth) it asserted an immediate awareness of
sense-experience, although without making any enduring contribution to
its clarification (Cudworth has recourse to metaphors derived from reflection theory to characterise the immediate awareness of sensation: . . . that
duplication, that is included in the Nature of synaisthesis, Con-Sense and
Consciousness, which makes a Being Present with itself; wherefore that
which is thus conscious of itself, and reflexive upon itself . . .).67 For John
Locke all thinking implies consciousness of thinking (thinking consists in
being conscious that one thinks)68 and indeed is a mode which is distinguished from the reflexive-objectifying use of inner sense, a pre-reflexive or
intuitive mode which Locke attributes not to animals, but only to rational
beings. Here, however, the notion of thinking is employed as broadly as in
the Cartesian cogitatis, which refers to all modes of representation. Lockes
thesis is surprising when one looks at it more closely:
Consciousness . . . is inseparable from thinking, and it seems to me essential to it: it being impossible for any one to perceive, without perceiving,
that he does perceive. When we see, hear, smell, taste, feel, mediate, or will
anything, we know that we do so. Thus it is always as to our present
Sensations and Perceptions.69
Even Locke cannot escape the idea that self-consciousness accompanies acts.
Neither does he avoid the assumption that self-consciousness is comparable
to a perception that makes other perceptions conscious. But when I feel pain,
this is the required (self-conscious) perception; I do not need to perceive it
in its turn. The logic of this model implies that the (perceived or accompanied)
acts are not themselves necessarily conscious (which appears to be the case
with animals according to Lockes view, since they have inner experiences
CH 5_f4_53-136 11/23/04 9:28 AM Page 72
without being reflexively conscious of them). One may conclude that Locke
considered self-consciousness to be an additional activity, modelled on reflection, which constantly accompanies the thought-processes of higher organisms. However, he made no contribution to the clarification of its structure.
Nevertheless, Lockes thesis that every act of thought is accompanied by a
co-consciousness of thinking was strongly opposed by John Sergeant. In his,
Reflexions on Mr. Lockes Essay Concerning Human Understanding,70
Sergeant asserts, We may think, without being conscious that we Think71 (and
illustrates this, unfortunately, with the case of a memory which is not at our
immediate disposal, but which later returns, and thus precisely shows the
continuity of consciousness, if not of reflection; Sergeant assimilates reminiscence and reflexion).72 Then he defends the counter-thesis to that of
Locke, namely that Tis impossible to be conscious, or know we know, without a new Act of Reflexion,73 which would follow the primary, object-directed
act of consciousness. Finally, he subjects the awareness of the act of reflection to the same condition: Tis impossible to be Conscious of, or know
our present Reflex Act, but by a new Reflex one.74 He explains this in the
following way:
The same Argument demonstrates that we cannot be Conscious of our Reflex
Acts at the very time we produce them. For, my First Reflex Act has for its
sole Object that Operation of Mind, which I had immediately before by a
Direct one; and my Second Reflex Act has for its object the First; and in the
same manner, each succeeding Reflexion has for its Objectthat Act which
immediately Proceeded. Wherefore, if the First Reflex Act had for its Object,
at the same time both the Direct and itself too; that is, did we, when we first
Reflected, know by that very Act itself that we did reflect, then the Second
Reflex Act would be forestalld, and have no Proper Object left for it. To clear
this better, let us assign one Reflexion to be the Last: it were not the Last
Reflexion, unless the Object of it were that Reflexion which was the last but
one. Wherefore, unless that Reflexion that went last of all remained unknown,
the Last would have two Objects, viz. The Preceding Reflexion and itself too.75
There we have an almost caricatural, and thus ideal, illustration of the circular theory of self-consciousness, based on the model of reflection. Sergeant
assumes that every consciousness is objective, and consequently considers
the consciousness of consciousness as the objectification of a foregoing con72 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 73
sciousness. The illusion of the immediate acquaintance of consciousness with

itself is explained in terms of the imperceptible rapidity with which the
Reflex Act follows the first one. Sergeant also assumes that the preceding
consciousness acquires the quality of being conscious only under the gaze of
the subsequent reflection directed upon it, which in its turn remains unconscious, until a further subsequent reflection raises it into consciousness. Thus,
the last reflection, directed towards the penultimate one, remains non-objective, and this means: unconscious. The consequence is clear: Hence we can
never come to know our last Reflexion.76 But since the consciousness of all the
preceding acts depends upon this last act, they too are unconscious, and the
thesis that there is something like self-consciousnesseven if mediated by
reflectiondestroys itself. These are my reasons, Sergeant summarises his
Certain and Evident Corollary, why I recede from Mr. Locke in his Opinion,
that A Man cannot think without being Conscious of it.77
The theory of an unknown author (Pseudo-Mayne) fares better. This theory
attributes to human beings a fundamental non-sensible knowledge of their
own consciousness, and indeed sets the knowledge of objects in a relation of
transcendental dependence on self-consciousness.
The Mind, in its several Acts of Thinking and Perceiving, of Imaging,
Remembering, Willing, or Affecting, is Conscious of any of its Faculties, it is
conscious of it, as its Objects, i.e. Something which is perceived by it self . . .
The Notion of Object (as I may here take the occasion to observe) is entirely
owing to Consciousness; it being plainly impossible that I should be able to
consider or regard anything, as having such an Appearance to me in my
Act of Perceiving it, . . . any otherwise than by being conscious of my own
Perception, and of the Appearance to which it refers. And whereupon it follows, that consciousness is indeed the Basis and Foundation of all Knowledge
whatsoever; inasmuch as whatever I can know or Apprehend of a Thing
by observing it, and reflecting on it with my Understanding, depends altogether on my first considering and regarding it as an Object, or something
which hath such a certain Appearance to me in my perceiving it.78
This (self-) consciousness is said to be clear and distinct, truthful and certain,
present to the mind, and always adequate; it implies a feeling of existence and
does not first arise through reflection, but is already at work in all reflection
CH 5_f4_53-136 11/23/04 9:28 AM Page 74
(for example, that carried out by Descartes).79 Nevertheless, Pseudo-Mayne

presents it as accompanying all cogitationes, thus assuming an analytical separation of the accompanying from the accompanied, which is not abolished
by the addition of the term immediate (does immediately accompany).80
Since Pseudo-Mayne does not set aside the common premises of all empiricists, according to which sense experience is the only source of knowledge,
he also assumes that self-consciousness can only occur as a feature of an
object-related act of perception in respect of which it (self-consciousness) is
subsequent and dependent:
For tho we are as well conscious of every thing we do or act, as of our own
Beings or selves; and it is absolutely requisite and necessary that some other
should precede that of Self-Consciousness, for we are conscious of our selves
only from our Acting, or because we act, and Self-Consciousness must of
course depend therein for its Existence; yet is it impossible to be conscious
of any Act whatever, without being sensible of, or perceiving ones Self to
be that which Does it.81
This thesis is incompatible with the assumption of the pre-reflexive character of self-consciousness (which indeed is not clearly developed by PseudoMayne). It shows clearly that our author belongs to a rich tradition of reflection
theory, represented by names such as Sherlock and Robert South. The latter
invites, in his considerations on a text of the former,82 that self-consciousness
presupposes not only a person as its bearer (suppositum), but first and
foremost:
another Act Antecendent to it self. For it is properly and formally a Reflex
Act upon the Acts, Passions, or Motions of the Person whom it belongs
to . . . [and] being a Reflex Act, must needs in Order of Nature be posterior
to the Act reflected upon by it.
Here self-consciousness is clearly explained as the subsequent reflection of a

preceding (object-directed) representation and is given a temporal index (posterior). Another example is provided by Bishop Peter Brownes text, The
Procedure, Extent and Limits of Human Understanding,83 which is significant in
general for the theory of consciousness.
. . . so that all the Operations of the Mind necessarily presuppose Ideas of
Sensations as prior Materials for them to work upon; and without which
74 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 75
the Mind could not have worked at all; no, nor have even a Consciousness
of itself, or of its own Being: Insomuch that it never could have exerted an
Act of Thinking, if it had not been first provided with some of these to think
upon; and this the compound word of consciousness plainly imports.84
Pseudo-Mayne takes over this view, and thus becomes entangled in the following circle: on the one hand, every object-consciousness must have its condition of possibility in a non-sensible (purely intellectual) self-consciousness,
whose bearer is a spiritual I-substance; on the other hand, self-consciousness
can only appear as an epiphenomenon of, and subsequent to, an objectrelated act.
Neither does Pseudo-Mayne hesitate to attribute an object to self-consciousness: it has an (inner) object. The correct insight that all object-consciousness presupposes self-consciousness85 does not prevent the author from
interpreting the kind of consciousness in which self-consciousness occurs as
a special case of object-consciousness:
In narrowly inspecting and examining into Conscious Knowledge and Perception,
we shall find that Self, or ones own proper Being, is its Principal and most
proper object.86
As a result of this recourse to the model of representation, Pseudo-Mayne

does not hesitate to ascribe the mode of (inner) perception (perceiving ones
Self) to self-consciousness. But the fact that perceptions are always conscious
does not entail that they themselves can be perceived as objects.87 (When
I have pains, consciousness is implicit in having the pains; I do not need
in addition to feel or sense my pain, as people sometimes misleadingly
say; furthermore, as self-observation, self-consciousness would lose its adequacy and distinctness.) No more helpful is the thesis that in self-perception
the perceived object (the self) is given in an incomparably inward way;
perceiver and perceived are here the same.88 Once subject-object duality is
introduced into the relation of self-consciousness, its ideality can become
epistemically uncertain; in reality, however, we encounter no identification
problem in the sphere of self-consciousness.
The problematic character of this separation becomes more acute when PseudoMayne denies consciousness to dreamers (as to animals).89 If my dreams were
unconscious, they could not be attributed to me as part of my biography as
CH 5_f4_53-136 11/23/04 9:28 AM Page 76
a conscious being. Statements such as I remember that yesterday I dreamt

of my first day in school would be inadmissible. It is enough to say that the
dream is not originally reflexively known; however, when I remember my
dream of last night in a reflexive attitude, I can only do so because there is
a continuity of consciousness between my present act of contemplation and
that which was dreamt, which is itself known. Since the author grounds the
unity of the person (and of all objects in general) on self-consciousness, this
expulsion of the unconscious becomes all the more difficult to achieve.
Kant shares with Browne, Sherlock, Robert South, and Pseudo-Mayne the
conviction that self-consciousness only appears in object-related consciousness (as the conscious unification of its manifold under categories).90 Kant
also becomes caught in the contradiction between his conviction that the
thought I think has Cartesian self-evidence (and is thus appropriate as the
principle of deduction for the categories), and his inability to explain this
self-evidence without recourse to the representational model, according to
which consciousnessincluding that of the Istands over against its object.
In reality, Kant interpreted the I as the activity in which the subject of consciousness turns away from all particular objects, and, turning back upon
itself, grasps its continuous identity with itself. Thereby this identity splits
into the pure, non-objective spontaneity of the subject of knowledge and a
phenomenal I opposed to it, whose identification with the former rests on
an unproven presupposition, and whose faultiness Kant perceived, but had
no means of avoiding.91 This inadequacy implies the failure of the explanatory model employed by Kant. For if there is I-hood (egoity), and if Cartesian
self-evidence is claimed for it, then it cannot be based on a petitio principii.
The model must therefore be false.
Thats the way the problem arisesbut no modern philosopher seems to
have recognised it as such. The first to become aware of the full ramifications of the fundamental problem with the reflection theory of self-consciousness was Fichte, and I shall now deal in more detail, first with his
suggested solution, then with why it doesnt satisfy either. Fichte wrote:
But there is consciousness; so that claim must be false. If it is false, then its
opposite is true. Hence the following claim is true [you may ask: which
claim? The one, according to which self-awareness is the result of a subjective
act turned back onto itself as an object]: there is a consciousness in which
76 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 77
the subjective and the objective are not to be separated, but are absolutely
[and without any mediation] one and the same. Thus it is this kind of consciousness that would be needed to explain consciousness at all.92
The bogus claim to which Fichte refers runs something like this: I come to
know the I through reflection, which is to say that the I enters into a selfrelation and consequently sets eyes upon itself.
But how can the subject recognise itself, if it is to be true that it is nothing
but pure subject? Kant drew a definite distinction between pure apperception and the I think and even went so far as to claim that the former produces the latter.93 It corresponded to his distinction between pure, non-objective
being and the objectified self-appearance [das gegenstndliche Sich-Erscheinen]
of the I.94 If it is true that knowledge proper is only ever knowledge of phenomena which stand over and against a knowing subject (and Kants conviction on this matter never wavered), then there can be no knowledge of
the Subject-I, which thus remains an unfounded assumption.
The theory of reflection as Kant inherited it from Descartes and Leibniz (as
well as numerous other thinkers from the British empiricist tradition) has
therefore to presuppose the very phenomenon whose structure it took upon
itself to explain. That is why Fichte repudiated the sophistry of reflection
theory in his lectures on the Wissenschaftslehre nova methodo (1797-1799).
I shall briefly sketch the main thrust of Fichtes argument. If our experience
of being conscious of ourselves were the result of a concatenation of several
states of consciousness, whereby the penultimate state of consciousness were
always witnessed by the last one in the series, then there would be no selfconsciousness. This is because the same conditions hold for the final consciousness, namely, that in order to become conscious of itself, it would have
to be made the object of yet another consciousness. But this other state of
consciousness needs to be unconscious, attaining self-consciousness only
through being objectified it its turn, and so on ad infinitum. But there is consciousness; so this model must be wrong. If wrong, then consciousness must
have been immediately acquainted with itself, that is, prior to any objectification by means of a succeeding consciousness. Fichte accounts for this immediate self-acquaintance as the complete indiscernibility of subject and object
in self-consciousness. Now in Kantian terminology an immediate consciousness
CH 5_f4_53-136 11/23/04 9:28 AM Page 78
is an intuition. But what is intuited here, is not a spatio-temporal entity, like

in sensible intuition, but rather the being of the sheer spontaneity of apperception: hence the intuition is deemed to be intellectual.
Once we do know this so well, the question arises again of how we found
it out. Obviously we intuited the very intuition of the internally acting I.
Hence it must be possible to have an intuition of the internally acting I,
and such an intuition is intellectual. This does not refute the Kantian system. Kant only denies the possibility of a sensible intellectual intuition and
rightly so. However, the intuition of the I is not something fixed, quiescent,
it is an acting I. In his system Kant does not explicitly consider this type of
intellectual intuition, the type that results from our representations being
the product of our own self-active mind. Kants system nevertheless presupposes the result of this type of intellectual intuition.95
If the I must indeed be explained by intellectual intuition, then several consequences ensue. Firstly, self-consciousness can no longer be considered as a
result of deliberate, purposive action (one can only posit oneself, if one already
knows what self means). Being coincides here (and only here) with being
known. Fichte puts this another way, namely: all consciousness of something
(including its own psychological states) presupposes an immediate consciousness of the initial consciousness itself.96 Secondly, self-consciousness is
not an instance of knowledge, because all knowledge is conceptual, and concepts are related mediately to objects, through the analysis of a feature or features common to several representations. Thirdly, self-consciousness is not
an instance of genuine (informative) identification. Every identification equates
semantically discernible elements according to certain criteria. But in selfconsciousness there is no such difference of polarity, and no need for such
criteria. Identity is a form of relation (indeed it is the most precise of all forms
of relation, in the eyes of contemporary logicians), whereas self-consciousness is unitary and acquainted with itself in the absence of any detour via a
second term.
III
The figure of speech with which tradition likes to sum up the early romantic post-Fichteana is that of outdoing (or even out-Fichteing) Fichte, and,
78 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 79
provided one can make clear what it means to out-Fichte Fichte, this judgement is not inappropriate. The accepted version is that, what is outdone, is
Fichtes alleged subjectivism. This view of romanticism predominates in the
tradition which spans such thinkers as Hegel, Rosenkranz, Kierkegaard, Haym
and even Heidegger. It is, however, not just erroneous, but actually reverses
the main current of the early romantic continuation of the Fichtean project.
If Fichte was indeed outdone, then it was not by virtue of ever more extravagant subjectivism, but, on the one hand, by an increasing concern as to the
meaning of the term identity, to which Fichte had allotted the central position in his grounding principle, and on the other, by a radicalisation of his
critique of the reflection model of self-consciousness.
Both concerns are closely allied. I shall begin with the first, and then proceed to the second. In what follows I shall take Hlderlin and Novalis as
representative of the whole of the early romantic movement, because their
thought alone attained a sufficient degree of thoroughness and clarity. They
levelled the following accusation at Fichte; namely, that he had been lucid
enough to spot the shortcomings of the reflection model of self-consciousness, but had ultimately failed to find a way around them. This criticism
seems harsh and above all unfair. The point is nonetheless not unfounded,
and from about 1800 Fichtewith the early Romantic critique evidently in
mindhad himself toyed with the idea of making certain improvements to
the formulation of his principle, improvements which show a similar train
of thought to that of his critics, and which have been definitively presented
elsewhere by Dieter Henrich.97
These improvements pertain to the notion of intellectual intuition that in what
follows was to be suppressed. When, contrary to Kants own intentions, Fichte
took up the claim of intellectual intuition, as the presupposition of all method,
including Kantian criticism, he meant to vindicate this with the thought that
self-consciousness could not legitimately be thought as the opposition of a
subject and an object. Intuition alone could vouchsafe the indiscernibility of
both poles, and this intuition, due to the intellectual nature of the I, certainly had to be understood as non-sensible.
On further analysis, however, it becomes clear that the formula of intellectual intuition is not up to the task of explaining the complete lack of differentiation between that which has consciousness, and that of which consciousness
CH 5_f4_53-136 11/23/04 9:28 AM Page 80
is had. The formula does draw a clear distinction between an object and a
subject of consciousness, and, furthermore, between an intuition and a concept. In other texts Fichte distinguishes, with no less conviction, between the
original, intuitively observed act of self-positing and its result, the thereby
obtained concept I. Of course he swiftly re-emphasises their indiscernibility and moreover, he emphatically repudiates the notion that the two moments
could be separated in time; (the rather hackneyed phrase in one fell swoop
is supposed to make the paradox disappear). Hlderlin and Novalis are not
that gullible. They point out that, as soon as a duality of moments is introduced into the sphere of self-consciousness, then its prereflexivity must remain
in question. No binary structure could possibly furnish the grounds of a strict
identity. Underlying this critique is a radicalisation of the meaning of the
term identity, for identity has traditionally always been defined as a relation.
In fact Fichte does define the concept I further as activity returning into
itself.98 He even believed both terms to be synonymous:
I prefer to use the term egoity rather than the word intelligence: since the
latter is the most direct description of the return of activity into itself, for
those capable of only minimal attention.99
Within the defining term egoity the activity can be discerned from the process
of its return into itself. Fichte goes one step further and assigns to this activity the cognitive mode of intuition. Only such a cognitive intuition, he thinks,
can establish both immediate consciousness and the lack of all distinction
between what posits and what is posited. The intuition intuits the act of selfpositing, even before it comes to the light of conceptual differentiation. Fichte
uses the term consciousness to mean distinct consciousness, (true to the tradition of Leibniz and Kant in which consciousness is defined as explicit, distinct
or reflected consciousness).100 Yet at the same time he holds self-consciousness
to be completely immediate, and moreover, immediately conscious.101 Hence
he slides into terminological ambiguity. One moment he stresses that intuition is not merely immediate, but also conscious, the next he claims that consciousness presupposes conceptual differentiation (and thus the possibility of
mediation). At times he only disputes that intuition has distinct consciousness (borrowing the Cartesian/Leibnizian distinction), but grants it clarity
of consciousness,102 which, as Leibniz and Wolff use the term, does not exclude
confusion [confusionem].103 Whatever the case, according to Fichte the intuition
80 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 81
ensures the immediacy of consciousness, but does not suffice to establish the
distinctness of the concept I. Hence what is needed is a concept to establish
the identity of the intuited contents. But concepts are mediate, they refer to
a feature common to several representations; in this case the feature which
belongs to, or better still, which constitutes an I. Representations [Vorstellungen]
are distinguished by means of concepts, and distinction presupposes opposition; omnis determinatio est negatio. Here are Fichtes own words.
Only the state of opposition can make clear to us what it is to act, (for an
act strictly speaking cannot be defined); [it is only] through fixity and repose
that we can think activity, and hence conversely we can only think repose
through activity.104
There is a law at work here, one that Fichte later, with great lucidity, was to
call The law of reflection that governs all our cognition:
namely: we cannot know what something is, without our thinking at the
same time that which it is not.105
To know something, to know something conceptually by opposing it to something else, means determining something, and determination (conversely)
in the Doctrine of Science means delimination, narrowing something down to
a particular region or sphere of our knowledge.106
In order to identify myself as me (and to differentiate myself from all that is
not I)107 I have to distinguish myself from everything else, that is, to limit
myself to an extension, which still allows for some otherness on my side,
against which I can define myself. Distinction rests essentially upon a relation of differentiation (determined is to say limited, or confined to a particular sphere by means of what is opposite).108 However, differentiation gainsays
both the claim to simultaneity and immediacy and the subject-object identity
in intellectual intuition. Fichte in fact speaks of a law of reflection without
which (distinct) consciousness would not take place. It infiltrates the innermost articulation of the cogito and destroys its pretensions to prereflexivity.
In this manner an unbridgeable dualism is engendered in the structure of
egoity; in order to describe this duality, the positing activity which intuition
lays hold of, has to be distinguished from its own resultthe concept. The
concept is the product of intuition; a state of repose in contrast to the intuition itself, which is characterised by agility.109
CH 5_f4_53-136 11/23/04 9:28 AM Page 82
In this repose we can observe that the positing of activity turns into the
posited. This is to say that this activity is first thought of as non-action, i.e.
in a state of fixity and quiescence. It is perceived and intuited thereafter as
quiescent, for otherwise we could not actively intuit the activity as active.
In this manner a product, or rather, the concept of the I comes to be. This
I can be thought but not intuited, since acting activity alone constitutes
intuition, and acting activity cannot be thought without simultaneously
thinking its opposite, namely the same activity in its previous state of repose,
i.e. without a concept. Concept and intuitionwhich are always immediately and simultaneously combinedcoincide, collapse into one.110
However, Fichte only asserts the coincidence of concept and intuition. So,
what seems to happen to Fichte, in spite of his intentions, is that he lapses
back into the reflection model of consciousness. He draws in fact a distinction (in almost Kantian terms) between what are intrinsically blind intuitions
on the one hand, and empty thoughts on the other. Their difference is clear, not
so their collaps[ing] into one. From 1800 onwards Fichte attempted to locate
identity beyond the sphere of reflection. But this is not my current topic.
Instead I shall try to show the precise meaning that is given to the term of
identity in the Doctrine of Science. We have just seen how the notion of identity contradicts the immediacy of self-acquaintance, and how Fichte would
have done better not to employ this term in such a context. It is understandable, however, that he should want to hold on to the term. For the
Leibnizian tradition had defined identity as a relation holding between semantically different entities which concur in all essential features. Hume had
added that simplicity (the characteristic of somethings being itself, so as to
be capable of being predicated of itself tautologically and without contradiction) was not to be confused with identity. For, whether or not something
is identical to something else cannot be decided analytically; an identity cannot be inferred from semantic features and the mere application of the law
of non-contradiction alone. On the contrary, judgements of identity only make
sense in Humes view when that which is identified, and that by means of
which it is identified, can be indicated by two different verbal expressions
(or two different modes of being of an object); with the result that the identification forms a synthetic judgement that adds to my knowledge: (simplicity is tautological, whereas identity is informative).
82 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 83
Fichte does not want to relinquish the distinctness of either of the two terms
that coalesce in the identity-judgement of self-consciousness, since this would
rule out a conceptual knowledge of the self. On the other hand, this distinctness
cannot be relied upon, or we risk losing the immediacy of self-consciousness.
In the third opening paragraph of the first Doctrine of Science (1794-5) Fichte,
following in the footsteps of Reinhold and Maimon, tries to defend the view
that difference and sameness can and do coexist in the ultimate grounding
principle of philosophy. His argument runs as follows:
Every opposite is like its opponent in one respect = X; and each like is
opposed to its like in one respect = X. Such a respect = X is called the ground;
in the first case of conjunction, and in the second, of distinction: for to liken
or compare opposites is to conjoin them, and to set like things in opposition is to distinguish them.111
This claim is proven in the following way. A and B are the two terms to be
conjoined with and distinguished from each other. If I then oppose A with B
(in the sense of positing the one in place of the other), A must nevertheless
remain partially intact, otherwise B would have no opposing term, and the
relation would fall apart. Thus A is only partially annulled in favour of B, as
something of A remains. Fichte uses the symbol X to show that A and B, in
order to oppose each other, have at the same time to share a common domain.
The formula A is not B can therefore be replaced by, there is an X which is
partly A and partly B.
The same holds for the relation of identity between the two. If I judge that
A is the same as B I do not necessarily assert that A, insofar as it is A, is
also B, or that B, qua B is at the same time A. That would be absurd, for in
this case I would not have identified one entity with another, but would only
have uttered a tautology; I would merely have said the same thing twice.
Identity cannot take place between one term and itself, for the very good reason that here there would then be no second term to be related by identity
to the first. Moreover, if the semantic distinction of A and B (I and not I)
is a necessary condition for a possible relation of identity to obtain between
the terms, then it seems to be the case that an identification presupposes a
prior non-self-sameness112 of the relata. Instead of writing A = B we have to
try and describe this relation in another way: there is an X, such that this X
CH 5_f4_53-136 11/23/04 9:28 AM Page 84
is A on the one hand and B on the other. But A is not B, and B is not Aqua
A and qua B respectively. In this manner Fichte can declare the proposition
A = B to be commensurate with the opposite proposition A B. The actual
identity, that is, the strict identity in the sense of complete unison, takes place
not between the terms A and B, but between the X and itself (here in the
sense of Humean simplicity). This identity alone is absolute, the one between
A and B presupposes difference, and is thus only relative. Now, the question
arises, however, of whether an identity (defined that way) can be known of
in self-awareness. And the romantic thinkers deny this with emphasis. (Schelling
developed this idea of Fichte in illuminating fashion, without actually adding
anything new. He speaks pertinently of a union that is the unity of the subject, not the predicates which is fully compatible with the so-called principle of non-contradiction.)113 Anyhow, Fichte quite clearly understands the
third principle of the first Wissenschaftslehre not as the principle of contradiction, but merely as the principle of opposition. I and not-I are diametrically opposed, and thereby, as the extreme members of a sub-category, fall
under the higher concept of the absolute I, andby quantifying themselveshave to divide themselves up into reality.
IV
The early romantic thinkers, most of whom (including Johann Friedrich
Herbart) attended Fichtes lectures on the Doctrine of Science, were well
acquainted with the Fichtean theory of the self. Following, indeed furthering Fichtes critique of Kant, they took up the (epistemic) question of how to
understand this identity which, like the identity of the Kantian subject of consciousness, is condemned to remain a mere presupposition of any relation,
be it one of unification or opposition. If strict identity takes place only between
X and itself, then although identity can become manifest in consciousness
(whose being is conditioned by opposition and distinctness, according to
Fichte, and the divisive form of judgement according to Novalis and Hlderlin),
identity cannot be understood through its function there. Strict identity (or
absolute identity to the early Romantics) would in this case, so to speak,
emigrate out of consciousness and occupy a position that is not merely prereflexive but wholly irreflexive with regard to consciousness. This is in fact
the very conclusion reached by Friedrich Hlderlin and Friedrich von
84 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 85
Hardenberg (Novalis) along their different but related paths of thought. This
conclusion is a logically consistent development of the Fichtean critique of
tradition, but it has certain ramifications that can no longer be reconciled
with consciousness-immanent idealism.114
Hlderlin attended Fichtes first lecture series at Jena on the Doctrine of Science,
as can be seen from his first letters to his friends Schelling and Hegel. In late
January of 1795, Hegel writes to Schelling:
Hlderlin writes to me from Jena every so often (. . .) Hes listening to Fichte,
and is full of enthusiasm for him, likening him to a Titan fighting for mankind,
whose sphere of influence will certainly not remain confined to the four
walls of the auditorium.115
On the 19th of January 1795, Hlderlin had written to Neuffer: I am now

working full time every day, and only in the evening do I manage to attend
Fichtes collegium. He relates his first impressions in a letter to Hegel on
the 26th of January 1795, and they give a foretaste of his later critique of
Fichte:
Fichte seems, if I may make so bold, to have stood at the crossroads, [between
criticism and dogmatism] indeed he is still theremuch of what he says
shows that he wants to go beyond the fact of consciousness within theory,
that much is certain, and it strikes me as even more transcendent than the
desire of all previous metaphysicians to go beyond worldly existencehis
absolute I (= Spinozean substance) contains all reality. It is everything;
there is nothing outside it; this absolute I thus has no object, for otherwise
it would not contain all reality within it; but consciousness without an object
is unthinkable, and if I myself am to be this object, then I am as such necessarily limited, if only in time, but I am therefore not absolute; hence consciousness cannot be thought in the absolute I; as absolute I I do not have
consciousness, and insofar as I have no consciousness, I am nothing (for
myself), hence the absolute I is nothing (for me).116
It is noticeable that in this argument there is a certain hesitancy, indecisiveness, indeed a tendency to recoil as it develops. In solidarity with the spirit
of Kantian criticism, Hlderlin begins with the observation that the search
for an I which is prior to all relation, and which grounds all our knowledge,
is an overly ambitious enterprise which takes no heed of the limits of our
CH 5_f4_53-136 11/23/04 9:28 AM Page 86
faculty of cognition. A little later he remarks that an I cannot, in any case,

be called absolute, if it is conceived as an object upon which cognition is
brought to bear, since objectification calls forth limitation. And his conclusion is equally aporetical. An absolute I, beyond the bounds of my understanding, would be unthinkable for us, and hence nothing.
The conviction that consciousness presupposes opposition and that the absolute
identity must therefore be unconscious pervades the preliminary draughts
of Hyperion, which in part predate, and which are in part contemporaneous
with Hlderlins attendance of Fichtes lectures. The preface to the Fragment
of Hyperion, written as early as mid-1794, distinguishes between a state of
ultimate simplicity based upon the mere organisation of nature, and a
state of highest education (. . .) by means of the organisation which we are
capable of giving ourselves.117 Between the two points there runs the eccentric path (exzentrische Bahn). I shall not comment upon this phrase, which
has attracted so much interpretive attention. Contextual correlates clearly
show that both the natural unity, and the unity achieved by means of education, their transreflexive status notwithstanding, are composite and articulated. With this unity there are in general and in particular essential
orientations, which other texts gloss as antagonistic tendencies proper to
being, within the domain of the unconditioned or of love, (the prose version
and the metrical version of winter 1794/5 are particularly good examples of
this). With this exciting thought, Hlderlin, doubtless under Niethammers
influence, departs abruptly from the Jacobian idea, whereby the unconditioned has to be thought of, even for the most obvious semantic reasons, as
unencumbered by any opposition which would situate it in a relationin
other words, the unconditioned is bereft of all opposition which would condition it. Perhaps the reader is familiar with the beautiful iambic pentameters in which Hyperion (in the metrical version of winter 1794/5) is delivered
by a wise man (line 27) from his reliance on the subject-object schema of
the modern age. He then draws a distinction between a state of unconscious
purity, intimacy, and freedom on the one hand (corresponding directly to the
natural state of ultimate simplicity mentioned in the preface to the Fragment
of Hyperion), and a state of consciousness on the other:
The day on which the beauteous world began,
Began for us the indigence of life,
86 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 87
and we exchanged our state of consciousness,

for our state of purity and freedom.118
At this point Hlderlin comes up with a conception of essential unity as a

structure articulated through opposition, a conception which is incompatible
with the dualistic intuition of Descartes and Kant, and even that of Fichte,
and one which, although not widely known, heralds a turning point of modern thought.
Pure spirit, free of passion, embraces
Not the stuff of life, and is not conscious
Of any thing, not even of itself,
Pure spirit has no world, beyond its bounds
Is nothing.Yet these words are merely thoughts.
We feel the limitations of our being
And the hindered power strives impatient
Against its shackles and the spirit longs
Back into the distant, undimmd ether.
And yet within us there is something, that
Wants to keep its chains, for if the divine
In us encountered no resistance, then
Wed not feel one another or ourselves.
But not to feel oneself, amounts to death,
To know nought, to be obliterated,
Are for us the same.Were we to deny
The drive to stride forth tward the infinite,
To purify, ennoble, liberate
Ourselves: that would be brutish. But neither
Must we proudly set ourselves above
The drive to limitation, to passion:
that would not be human, twere almost death.
So love unites the conflict of the drives,
None of which can rest unsatisfied.119
The first lines basically reproduce the position of the preface to the Hyperion
fragment. This position can also be found in the aforementioned letter to
Hegel of January the 26th, which contains Hlderlins thoughts on Fichtes
lectures. The unity, which remains forever presupposed by the self-relation
CH 5_f4_53-136 11/23/04 9:28 AM Page 88
of consciousness, can itself not be thought as conscious, and hence it is nothing for us. Therefore this unity is no longer just the prior ground of reflection (used synonymously with consciousness), but rather the transcendent
ground. Schleiermacher, in his lectures on Dialectics, will make the characteristic move of replacing the phrase transcendental ground with the alternative transcendent ground. 120
To return to the above quotation from the metrical version, the second half
moves beyond the demand for a suprareflective unity. The ultimateunconsciousunity is nevertheless not opaque, it has an internal structure. It opens
the space for two antagonistic drives in which Schellings notion of a reciprocity of hindrance and striving is prefigured.121 There interrelation is articulated as the opposition between a real activity, heading forth into the infinite,
and an ideal activity, working retroactively against the first, driving it back
into itself. If the unconditioned were to be represented on the model of infinite striving, then it would remain unconscious. But if it showed itself to be
limited, it would contradict its own concept (determination presupposes negation, hence limitation, and hence conditions; whereas the infinite is completudo realitatis). Thus the unconditioned is discursively represented as hindered
or inhibited striving ( gehemmtes Streben) (a solution we also find in Novalis,
in Friedrich Schlegel, and in Schelling). For the sake of conceptual clarity, the
unconditioned binds itself, albeit transiently, to limitation, but in virtue of its
infinity, it constantly transgresses its own limitations. In a word, the unconditioned is made manifest as excentricity or ecstasis, as the temporality of consciousness, whereby temporal is understood according to its celebrated
definition as the being that is, what it is not, and that is not, what it is.
The discordance between the drive to stride forth to infinity (which Hlderlin
calls real activity) and the drive to limitation (or ideal activity) does not
on this account destroy the structure of the unconditioned. It is rather its
most proper articulation: Den Widerstreit der Triebe, deren keiner / Entbehrlich
ist, vereinigt die Liebe. (The combat between the [two] drives, neither of
which / Can be missed, is unified through love.)
Love is usually understood as a consubstantial relation between beings,
equal in status and autonomy, which leaves no room for bondage or coercion. Schelling puts it beautifully:
88 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 89
It is the mystery of love, that it bonds in such a way, that each can be for
itself, but neither is nor can be without the other.122
Through love a being transcends the sphere of its individuality whose centre of gravity seems to lie outside itself. A person who is, as was once said,
inflamed by the love for another, seeks his own worth outside of himself,
he seeks to reclaim his own essence in a heightened form, from where his/her
beloved lies. The lover, says Schiller, does not desire the other, like he or she
desires to possess a thing, but values the other, as one respects a person.123
So, in defiance of all dualistic intuition la Kant and Fichte, love calls forth
a principle that surpasses the dichotomy of self and other, a principle which
embraces two related terms equiprimordially, without one having to be sacrificed to the God that reigns within us [dem in uns waltenden Gott],124
though the lovers do indeed experience the bond that unites them as the
God within us.125
This speculative conception of love brings into play a completely new conception of what identity is.
Since the conception was first developed in the analysis of the structure of
self-regulating entities, that is, organisms, it is not difficult to understand
why, in this context, so much weight is attached to the concept of nature, as
a being, bearing the highest degree of organisation. Hlderlin takes the idea
one step further: not only nature, as a whole, but spirit itself is organically
structured. Spirit consists in the absolute identity of the real and the ideal,
an identity that is articulated as the complete equiprimordiality of identity
and difference. This formula is often incorrectly associated with the name of
Hegel, and contains the following thought: unlike tautology (whereby one
and the same thing is merely repeated, A = A), identity is not trivial, it is a
real relation. The model of identity is A = B. It shows, how, to use a fairly
mundane example, a man who for instance has two names, can nevertheless
be one and the same person.126 This example bears a striking resemblance
to Freges Venus, which is differently determined as the evening and the
morning star, but not at the cost of its identity, and qua evening star is the
same as the morning star, and not trivially so. (It took thousands of years for
mankind to discover this identity. Schelling would say that it took thousands
of years for mankind to grasp their own identity with nature, in a non-reductive way, that is, neither materialistically nor idealistically). In much the same
CH 5_f4_53-136 11/23/04 9:28 AM Page 90
way absolute identity identifies two semantically discernible termsthe ideal

and the real limiting activity, an activity that strives towards infinity. However,
this difference remains only a virtual difference in the matrix of the absolute,
and only becomes actual when I disregard the bond that unites the two. This
means to say that two things which are only virtually opposed can still coexist; since that which can only be real, but is not, is powerless to banish
from its domain that which can only be ideal, but is not. Only when one
of the moments is realised, does it have to eject the other from its place, and
determine this other as its predecessor or successor. Once we disregard the
bond of substantial unity, then the real opposes itself to the ideal, and
only in virtue of this actuated relativity does the whole succeed in realising
itself as finite and temporal reality that is opposed to the absolute.127
It follows that Hlderlins philosophy of love could not be deemed a philosophy of absolute identity, if its concept of the absolute did not also include
what it is notthe relativity, the difference of the separate essential tendencies. Relativity proves, on the one hand, to be a moment within the structure
of the absolute, since whatever manner of Being persists in relativity, it is
nothing but the presence of the whole in the part, and the whole here means
precisely the absence of all distinction between difference and identity. In this
manner the structure of the absolute is analogous to that of the organism,
which similarly includes within itself its own oppositemechanism.
However, amongst all these poetic drafts of a philosophy of love, one thing
remains ambivalent and undecided. It is the seamlessness of the identity
which guarantees that the relation between infinite striving, and limiting
activity is indeed a true self-relation. But this identity cannot be evinced from
the duality of reflecting and reflected moments alone; it is still presupposed,
rather than posited by reflection. For this very reason Hlderlin comes up
with a solution, a few months after the metrical versionsupposedly in May
1795 (though, as will be shown, this supposition is perhaps unjustified)a
solution which he scribbled down upon the fly leaf of a copy of a book
we ignore which one.128 I shall now attempt to sketch out the bones of his
argument.
Kant sees judgement as the activity of thinking, an idea that stems from the
Profession de foi du Vicaire Savoyard in Rousseaus Emile. Every judgement
works towards cementing a synthetic unity, for a judgement is nothing but
90 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 91
the combination of a subject and a predicate. Now if the resultant proposition is true, then what is known is the type of relation that holds between
the proposition and the object it denotes (analytic judgements either presuppose a prior synthesis, or they are tautological, like judgements of logic,
which are based exclusively upon the principle of non-contradiction). Hlderlin,
under the aegis of Kant, but drawing upon a voguish though bogus etymology, interprets the German word Urteil (judgement), as the index of an
originary separation Ur-teilung.129 In the act of judgement a prior unity is split
into two members or two relata, whose relation at once conceals and reveals
the original unity: reveals, because two different representations are combined
in the judgement, hence they are both referred back to a grounding unity,
but conceals, because this unity never appears as such but only as the differentiation of two mutually dependent types of representation (that is, the unity
is articulated through grammatical subjects and concepts). Hlderlin then
applies this general principle to the judgement I = I. Even here there is a
differentiation: the judgement divides the relata, otherwise the determinacy
of what was judged would be occluded. But in this case the differentiation
occurs because the content of the judgement contradicts its own form. What
the judgement states is that the relata are undifferentiated. But formally what
happens is that the judgement differentiates the undifferentiated terms. From
this observation Hlderlin draws the following conclusion. On the one hand,
I can have no knowledge about something without my forming a judgement
upon it, that is to say, without my depriving it of its absolute identity. On
the other hand, the judgement, as a relation (of two things, for instance the
I and itself), is now dependent upon a fundamental and non-relative identity. It follows quite clearly that the synthesis of judgement has to be distinguished from a pre-judgemental, non-relative unity. Hlderlin joins Spinoza
and Jacobi in naming the latter unity Being (Seyn). Being is of a higher
order than the relative or predicative identity of which Fichte speaks. Being
cannot be thought, (since to think is to judge, and to judge is to differentiate) yet I cannot simply do without being, since the actual and evident experience of the I = I qua ego-identity remains a mystery without the postulate
of a unity that grounds the terms of the relation.
Strictly speaking it is not only the pre-identitarian unity that cannot be derived
from relations within judgements. Hlderlin supposes that self-consciousness
is only made possible by opposing myself to myself, by separating myself
CH 5_f4_53-136 11/23/04 9:28 AM Page 92
from myself, but by recognising the identity of the opposed self in spite of
the separation.130 Put differently: not only is it the case that the reflection
model of self-consciousness cannot explain the fact that subject and object
are identical; it also fails to explain how my awareness that in dealing with
the other of myself I am indeed dealing with myself (as the same self). (For
it could always happen that a subject and an object as conceived in our selfconsciousness are identical without our conscious awareness of this identity;
thus it is quite conceivable that someone correctly identifies a person as X,
and knows of X that he is such-and-so, without, at the same time, necessarily knowing that his knowing actually amounts to an epistemic self-relation).131
This initial idea represents in my opinion a deep conviction underlying all
early Romantic thought. It is the assumption that Being, qua simple, seamless unity, in contrast to the Kantian or Fichtean cogito, cannot be epistemically retrieved from either relations of judgement or reflection, which all
perform an originary separation on what they were supposed to unite, and
only ever manage to presuppose the original, simple unity. Hlderlins critique of Fichte is to be found in the way he emphatically opposes intellektuale Anschauung (as he calls it) to the act of Ur-teilung: judging, or originary
separation (that is, the determinate consciousness of something.)
The radicality of this move takes it a step beyond Fichtes conception of intellectual intuition, which articulates the claim to immediate unity only mediately, that is, by means of the conceptual pair of intuition and intellection. Of
course a conceptual pair could betoken unity, but only with the circular presupposition, that immediate knowledge of this unity already existed, prior
to the act of originary separation. If, on the contrary, knowledge is bound to
consciousness, then we are forced to conclude that there can in principle be
no knowledge of the absolute unity, which is only mediately available to us
in the play of reflection.
Hlderlins objection to Fichte is more extensively expounded and more
clearly thought out in a lengthy footnote to another of his essays not intended
for publication, On the Operations of the Poetic Spirit.132 Again I shall only summarise Hlderlins thoughts.133 Hlderlin argues in the following manner:
The two defining characteristics of the representation Ithat it is simultaneously absolute and self-referentialare mutually exclusive. If egoity were
subject to the condition of having to refer to external reality in the form of a
92 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 93
synthetic judgement, then its claim to unconditionality would a fortiori be

null and void, since it would depend upon relation, and relation means quite
the opposite of absoluteness. On the other hand, we cannot, for the sake of
expounding the I, do away altogether with its claim to unconditionality, for
then we would be left with two discrete and non-identical parts, and the
moment of self-possessionthe evident feeling of identity and oneness in my
consciousness of myselfwould remain unexplained and inexplicable.
Therefore we cannot afford to drop either viewpoint. We must stress the point
that the active self-relation of the I does not give rise to knowledge of the
absolute identity of what is combined in this relation (but not by it). Yet I do
have this knowledge in an utterly pellucid way. Hence Hlderlin concludes
that within the infinite oneness of the self there is revealed an ultimately
united and unifying power which is not an I as such.134
Sometimes Hlderlin calls this power the One, and sometimes Being, but
always in contrast to identity, which only creates relative (synthetic) or judgemental links between related terms, thus laying claim to a criterion that cannot be discursively retrieved from the relation as such, but a criterion that
always has to be presupposed. Being does found consciousness, but strictly
speaking it is not itself conscious (insofar as consciousness is synonymous
with reflection, as it is for the whole of the post-Leibnizian tradition). In this
sense we can speak of the primacy of Being over consciousness. The light in
which consciousness basks, does not radiate from consciousness itself but
comes from a ground which is not causal and which consciousness can never
quite illuminate. Only the inexhaustible wealth of meaning in aesthetic representation manages to depict it as such, that is, as reflexively non-representable; herein lies the superiority of aesthetic means of expression over
speculative ones. This is the consequence that Hlderlin draws from the aporia that concludes the above-mentioned footnote.
Even if the footnote was not written down before 1800, we can still get an
accurate picture of the state and content of Hlderlins thought of around
1795 by looking at the so-called Philosophische Raisonnements of his friend
Isaak von Sinclair. This will also enlighten us as to the aesthetic consequences
that Hlderlin draws from the failure of reflection epistemically to retrieve
the absolute. Unlike his friend, Sinclair had come to Jena as early as 1794 and
was amongst those present at the reading of the first (theoretical) part of
CH 5_f4_53-136 11/23/04 9:28 AM Page 94
Fichtes Doctrine of Science. Moreover, Sinclair came into contact with Karl
Christian Erhard Schmid, an early intellectual ally (and later opponent) of
Fichtes, a man who was close to Novalis, having formerly been his tutor,
and whom he had kept informed of the fate of the Doctrine of Science and the
surrounding debate. Sinclair valued Fichte and Schmid above all for political reasons:
Their cold inspection, their reasoning which proceeds from the depths of
reason, and which spreads to all branches of human action, will vindicate
the rights of man and overthrow the thrones.135
In the winter term Hlderlin came to Jena along with Jakob Zwilling, another
friend whom Hlderlin will meet again in Bad Homburg.136 Of the three Sinclair,
who had been part of the intellectual circle formed by the former students
of Reinhold and the friends of Niethammer, was by far the most capable of
giving an informed judgement about Fichtes own thought. As far as the dating of the notes is concerned, we can be fairly precise. The notes are written
on the back of a programme for a concert which took place on the 6th of
December 1795. Sinclair later twice reworked the Raisonnements, which were
hastily and rhapsodically written, but the main ideas of the two subsequent
versions did not substantially alter. Dieter Henrich and Hannelore Hegel suggest that it is likely that the theory sketched out by Sinclair assumes knowledge
of Hlderlins Urtheil und Seyn. On this account it is all the more instructive,
all the more conceptually acute, and all the more delightfully argued.
The Raisonnements presents four theses. 1) Reflection (which in everyday language is termed judgement) is a separation, in which the demand for unity
lives on, and is co-posited as such a demand. 2) Fichtes I is not a substance.
3) Praxis cannot be evinced from theory: and finally 4) Beings resistance to
discursive articulation by reflection raises art to the emblem ne plus ultra of
the downfall of philosophy. Of these four theses, only the first interests us
here. As stated above, the position corresponds to Hlderlins conviction that
the relation of one to another (for us or from the third-person perspective)
can always be a self-relation, but that the self can only apprehend its own
act in the other by means of its knowledge of a unity that survives within
the separation and yet goes beyond it. Insofar as this (material) unity cannot
make itself manifest as such in the form of separation, it becomes a mere postulate (or a demand). But this demand must be rationally motivated within
94 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 95
the structure of self-consciousness, and this motive is the factical knowledge

of myself as a single and united (not as a divided) being.
Sinclair (along with Novalis) was the first to hold consciousness to be a positing [Setzung] or thesis.137 What is posited by consciousness is the other of
itself (the object or itself as object). Since judgementas the originary expression of the self-sundering of Being, or of the Yeow138consists in just such a
thesis, then Being proper has to be thought as immaterial, non-objective. Being
corresponds to the mode of consciousness pertaining to the non-positing cogito, the athesis.139 Aisthesis, which Sinclair, by dint of an adventurous etymology, renders as Aeisthesis (or continual positing), makes the content of
what is not posited and what cannot be posited comprehensible to the positing consciousness. In this sense the aesthetic reality (. . .) [is] a self-denial of
the I, a repudiation of the pure thesis; it is the return of dirempt Being into
athetic unity.140 This (positive) sense of aesthetic, however, stands over and
against the (Kantian) aesthetic of the sensible world, whereby Aeisthesis
takes on yet another meaning, namely the persistence of the divisions of finite
human understanding: Knowledge is always incomplete for the aesthetic,
because knowledge, as the product of reflection, always presupposes the
judgement (originary separation), that renders it incapable of thinking the
existence of unity, incapable of thinking the aesthetic ideal.141 In this quotation it is noticeable that the second (negative) meaning of aesthetic slides
into the first (positive) meaning, whereby the aesthetic becomes an ideal of
the understanding, the utopia of a sensible representation of unity.
The moment one tries to know or to posit the Yeow (the athetic unity, or
athetic essence), one makes it into the I (into Fichtes absolute I). By reflecting upon and positing his ultimate nature, this nature is divided, and then
after the division its undivided character is restored by unification; a process
in which, as it were, the Being of division is presupposed; it is the most
incomplete concept. En kai Pan.142
If Being (that is, not the relative unity of synthesis, according to Sinclair, but
absolute oneness)143 were not presupposed by the division, then it could not
be read as the proposition I am I.144 In order to find myself and nothing
else in the other related term, then the unity that is negated by the form of
judgement has to persist in the form of a postulate, (reflection upon limits
is only possible under the condition of unity as an ought).145 Sinclair declines
CH 5_f4_53-136 11/23/04 9:28 AM Page 96
to follow the path of Jacobi, who relied upon a feeling to which absolute
unity as such is supposed to be given.146 That would amount to abandoning
consciousness for transcendent explanations.147 Philosophy cannot go beyond
reflection, or beyond whatever can discursively be redeemed within reflection as a reaction against it. This does not mean that philosophys discovery of the relative unity of self-consciousness cannot rationally motivate its
demand for absolute unity: hence the warrant for the postulate of unity lies
in the fact that reflection (. . .) reflects upon itself;148 hence philosophy comes
to know the formal rules which prevent it from grasping the thought of the
absolute, at the same time as it comes to comprehend the indispensability of
the absolute, as a warranted presupposition.149
Sinclairs philosophic and aesthetic raisonnements (ratiocinations) concur in a
most surprising manner with the deliberations of Friedrich von HardenbergNovalis, which were written either simultaneously or only a few months earlier, and which are known to us as the Studies on Fichte. They date back to
just after Hlderlins notes on Judgement and Being, that is to say between
autumn 1795 and early 1796. Once again I can only offer a rsum of the work.
Along with Hlderlin and Sinclair, the first independent thought of Novalis
begins with a reflection upon the form of judgement. As with the Savoyard
vicar, and with the author of The Critique of Pure Reason,150 what is at issue is
the meaning of the copulative is. The is identifies one thing with another,
if only relatively. The verb to be in this case really means to be identical.
In order to explicate the identity, as expressed in a judgement, we must step
outside it, explains Novalis. We leave the identical [,] in order to explicate
it.151 In other words the Being of the original identity is transformed or rather
transfigured in the act of synthesis. This act transmits the identity to consciousness (by means of judgements or concepts, which latter are just condensed judgements) but in so doing conceals the identity it transmits. The
act of judgements does reveal a kind of identity, but does so by illusion:
what already is, happens. The act of synthesis produces a result that already
was, prior to the act. The forms of judgement can only ever impart relative
identities to particular contents; the Being of absolute identity can only find
expression as Non-being, Non-identity, Index that is to say in forms which
never fit properly, and which in fact denote the very opposite, since they substitute for, and thus fail to grasp, what is actually intended.152
96 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 97
The same holds for the reflection by means of which consciousness represents itself to itself. This self-representation, however, in no way produces
the self, but rather brings to light what already was: What reflection finds,
seems to be there already.153 Any self-reflective knowledge that articulates selfacquaintance as an explicit self-relation must presuppose a prior unity that
has nothing to do with any relation. Novalis calls this non-relational acquaintance feeling.154 Its epistemic status is non-positing or non-knowledge of
what it is conscious of, and of what constitutes it as consciousness, in contrast to reflective consciousness which posits (and which knows) these same
things. If all knowledge is positing, then it follows that feelingor rather the
spirit revealed in feelingcannot amount to knowledge. Hence Novalis
gives it the epistemic status of belief.155 We believe what we cannot know,
but what must be presupposed by all knowledge. Thus belief is anything but
unfounded.156 If it is true, however, that the supreme Being overtaxes our
faculty or cognition, then how is it that we can be conscious of it? Still true
to the deepest conviction of Kantian and Fichtean criticism, Novalis devotes
a whole series of deliberations to this very question, deliberations that may
properly be called works of genius, given their originality and their effective
history. These deliberations inaugurated a new and independent avenue of
idealist speculation that led ultimately to the overcoming of idealism. Let me
once again just sketch out his train of thought.
What triggered off Novalis thought experiment was a consideration of the
meaning of the word reflection. Reflection means mirroring, and every image
which is mirrored is inverted. When I hold an object in front of a mirror, right
is reflected as left and left as right; moreover the rays of light that fall onto
the surface of the mirror seem to be coming out of the mirror, in the opposite direction.
Now, Novalis asks himself: Does the reflection that we call self-consciousness
behave differently? In fact, despite Fichtes protestations to the contrary, even
intellectual intuition consists in consciousness return into itself, since what
is in the final analysis one and the same, appears as the duality of intuition
and concept. On the other hand, there is something similar to intellectual intuition, and what is more, it constitutes the highest form of consciousness that
we can achieve. But then the identity appears to give way to the manifest
relation between intuition and concept; the identity is no longer a content of
CH 5_f4_53-136 11/23/04 9:28 AM Page 98
consciousness. In intellectual intuition (Novalis like Hlderlin uses the neologism intellektuale instead of intellektuelle) we experience a perpetual failure
to grasp the absolute that we aim for. But our aim, or rather our longing for
the absolute, is itself the essence of intellectual intuition. What is intended
has to be held separate from what is in fact achieved. Intellectual intuition
aims to depict a seamless unity, with no distinction between subjective and
objective poles, but this representation can only be brought about as the reciprocal play between two reflexes, as reflection. Intellectual intuition is thus
characterised by a tension: when it aims for the absolute, as that which it can
never grasp, then the absolute becomes the point of departure and the intentional object of its involuntary reflective movement. Novalis characterises this
aim as the illusory striding from the finite to the infinite, hence from the I
(qua determinate intellectual intuition) to that in the I which exceeds the I
itself; the true One, the prereflexive unity that it failed to secure.157 The stride
is illusory. The illusion that in the consummation of intellectual intuition we
managed to step from the finite to the infinite lies in the way in which all
reflected relations are turned around (the mirror image of the reflection throws
the original relations back to us in reverse, ordine inverso). But reflection does
possess the means to return this false appearance to truth by reflecting the
reflection onto itself, or by doubling up the reflection. A reflected reflection
inverts the reversed relations once again, restoring the original order, the one
that obtained prior to all the mirroring. What appeared to tend from the finite
towards the infinite is now revealed in the light of the dual reflection as the
illusory striding from the infinite to the finite.158
The point of what Novalis calls the Deduction of Philosophy is to prove
that under certain conditions we are justified in speaking of an absolute which
intellectual intuition is still incapable of representing. The object of the first
reflection is certainly not the absolute itself, and the wish to represent or to
explicate it only leads to its polarisation into the play between two reflexes,
(Fichte spoke of the fundamental reflex [Grund-Reflex]): one which has consciousness, and another, of which consciousness is had. This is the classical
model of reflection which is unable to explain the original unity of the self.
Novalis calls the object that the first reflection fails to capture feelingas
soon as it is observed, its spirit vanishes.159 To feel the absolute, or rather,
to possess a prereflexive intuition of the absolute, is not at all akin to objectifying,
knowing, hence representing (or positing) the absolute. Feeling, for Novalis
98 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 99
is a type of receptive consciousness to which something must be given.160

The limits of feeling are the limits of philosophy.161 It is towards this kind
of consciousness that the view of the first of the two reflections is directed.
Such reflection, taken as a cognitive mode of positive or positional knowledge, would have to be termed a reflection upon non-knowledge, and this
reflection is only achieved at the price of the retreat or withdrawal of the
absolute within the feeling (of the original intuition). But the second reflection is directed towards just this lack of Being that opens up within originary intuition at the very moment of its objectification. The second reflection
makes clear that the mode of consciousness proper to the initial one was
in fact not-knowing. What distinguishes the second from the first is that it is
no longer merely reflection upon a state of original not-knowing, but notknowing that reflects upon itself, that is, that knows itself as such. Philosophy
must be content with this docta ignorantia. Only the inexplicable fecundity of
meaning, proper to a work of art, can show positively what cannot be definitively resolved by knowledge. In this way the work of art remains the only
available medium of representing the unrepresentable.162
Even within the context of contemporary theories of reflection and self-consciousness Novalis attempted solution to the problem is unusual. Those who
are suspicious of Novalis highly speculative imagination would do well to
remember that his ideas were not intended for publication, and that he was
a young man, who had just turned twenty-three at the time, and who had
studied law and geology, and not philosophy. But we can ignore the form of
his deliberations and turn instead to the question that Novalis addressed,
namely: how can we speak of the absolute, when awareness of the impossibility of achieving a concrete knowledge of the absolute is built into its
very concept? Knowing is a cognitive operation which in the understanding
of the early romantics proceeds by means of relations, and in the case of selfknowing the relation assumes the form of reflection. In a reflexive relation two
different entities are related to a unity. But this unity is lost in the moment
of its representation, or strictly speaking remains beyond all representation.
We could endorse the Kantian conclusion that this unity, as a transcendent
unity, can never become a theme for our finite faculty of cognition. However,
this conclusion would be premature, since we cannot do without this absolute
unity, even if we cannot explicitly represent it. In actual fact reflection does
CH 5_f4_53-136 11/23/04 9:28 AM Page 100
not reveal to us a consciousness of endless diremption, but, in the medium

of the specular play of two reflected images, evinces an acquaintance with a
complete unity whose completeness eludes us. Fichte was hard on the heels
of Novalis finding, but the expression intellectual intuition with which he
hoped to grasp it was still caught up in the reflection model of consciousness, still marked by the latters unreconciled dichotomies. What is needed
to escape the reflection model is a call for a pre-identical simple unity, that is
binding ex negativo in the identity of reflexive self-consciousness, and that can
be rationally postulated from the existence or the fact of the latter, that is, as
a presupposition which cannot be known. The postulate of unity is rational,
since the structure of reflexive self-consciousness leaves us with no alternative but to make such a postulate, although this does not mean that the object
of the postulate could ever come into our purview in the form of an object.
The fact that our acquaintance with ourselves is based upon a pre-reflective
feeling finally carries us beyond the bounds of intellectual intuition, and
gives a presentiment of the opening up of a completely new dimension, which
no longer enters into our consciousness, but upon which our consciousness
essentially depends. Instead of being understood as intentional self-positing,
consciousness comes to grasp itself as the opening up of an anonymous
dimension, a dimension in which it gains access to itself, but for which it is
no longer responsible. Novalis calls this other dimension Being.163 In every
self-acquaintance there is an opening up of Being whose circumambient obscurity far outmeasures what it can bring to light. But the subject does not
renounce its enquiry here. Disabused of the idea of its sovereign self-authorship, and fully cognisant of its radica dependence upon its Being,164 the subject is nonetheless still responsible for the way in which this Being comes
into appearance.
Though early Romantic thought does constitute a turning point in modern

debate about the foundational nature of self-consciousness, the theory of
Hlderlin is far away from the hysterical thesis of the death of the subject.
This thesis will find succour neither in the early Romantic speculation, which
took upon itself the task of explaining our irrefutable acquaintance with ourselves, nor anywhere else, except perhaps in the decisionist self-renunciation
of argumentative thought. Such an attitude may well be post-modern, but
is definitely not romantic.
100 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 101
V
Johann Friedrich Herbart, like Hlderlin, was also a student of Fichtes and
attended his lectures during his Jena period. He, like other democratic students, belonged to the Association of Free Men [Bund der freien Mnner].
Although his intellectual temperament was less well disposed towards the
influence of Jacobi and although he was less prone to romantic escapades,
Herbart, as Kants successor in the chair of philosophy in Knigsberg, attempted
to prove, in no less radical a way, how Fichtes model of reflection remains
caught up in the reflection model of consciousness.
Herbart argues that Fichtes assumption that there is a subject-object identity
in self-consciousness leads to insoluble contradictions both in the subject
and in the object.165 This thought of identity, from either the perspective of
the subject or that of the object, would lead to an endless iteration of the Self
[Sich], and therefore an infinite regressHence we have gone no way towards
answering the age old question concerning the Self.166
If one starts from the side of the object, then the subject for which the object
is given and through which the object is known, is presupposed. But then,
how can the subject for its part be known, in this conception of the object
[Objekt-Auffassung]? Whenever the subject is changed into an object from the
side of a thematised subject, the thematised subject itself, remains unobjectifiable, and therefore unknown. This is so, until it is represented by another
subject. If this occurs, then the same process will only be carried out on a
higher level. Herbart writes:
Who or What is the object of self-consciousness? The answer must be found
in the proposition: the I represents Itself [Sich]. This Self [Sich] is in fact
the I. If you replace the concept I, then the first sentence changes in the
following way: the I represents the representing Self [Sich Vorstellende]. If
you repeat the same substitution with the term Self you will find the following: the I represents that, which represents the representing Self. But here
the expression Self [Sich] is merely repeated. It therefore requires the same
substitution. If you raise the question again what does this Self mean? Who
is ultimately represented? There is no other answer but that the Self is dissolved into its I, and the I into the representing Self. This circle repeats
itself into infinity without ever giving any detail about the actual object in
the representation of the I.167
CH 5_f4_53-136 11/23/04 9:28 AM Page 102
There is a corollary of this type of infinite regress on the side of the subject.168 From this point of view, what is represented by the Subject in the
position of the object, is always the Subject itself as representing. In this way
the represented object will always dissolve into the act of representing it:
If you therefore assume that the I is objectively given, then it is given to
Itself [Sich selbst] and to no other. It is represented through itself. However,
you must not neglect the act of this representing itself. What the I is, the
I must also know, according to its concept of itself. What it does not know,
it cannot be. It is really the representing of itself. As such a representing
Self [Sich Vorstellendes] it must therefore be represented. However, that
which is represented afresh, which was necessary to this end, insofar as it
is certain that it is a real act of the I, must become again an Object for a
higher-level knowledge. And this knowledge demands, in order to become
known, the same act. Again we have an infinite regress, and again we ought
not to have. For here too self-consciousness knows these albeit rare cases
where certain repetitions of reflections have succeeded in working knowledge
into the object [Gegenstand] of a new observation [Betrachtung], it knows nothing about why such a repetition is necessary if we are to talk about ourselves
at all and it knows even less about the endless continuation of the series.
Additionally, the repeated return to ourselves, where we always again become an Object of consciousness, takes time. The concept of the I, however,
does not allow for time. This concept, if it can ever be thought, must include in
itself all this thinking of thinking. Otherwise, it would not be an I because
at some point it would be lacking knowledge about itself. We can therefore
see that the I, according to this interpretation, even if it had truly found
its object, would remain for itself infinite and therefore an incomplete and
uncompletable task.169
Indeed the I would have to wait for the end of an infinite regress, in order
to become what its own definition claims that it is, that is, self-knowledge.
This would happen only on Doomsday, hence never. Its definition would
remain unrealised. One would go from what can be thought to the unthinkable.170 This, however, contradicts the phenomenon of the I, which is factually properly known to itself and, as Herbart shows us in the above quotation,
although in principle it can be known in individual cases of finite self-reflection, nothing can be known about a complicated infinite reflection: because
102 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 103
self-consciousness does not know about this development into many parts,
or about such a plurality of immanent interpolations.171 And anyway, such
an iterated self-reflection can only be thought of as a process which takes
place in time, and as such Fichtes conception of Immediate-Identity [Im-NuIdentitt] of the subject and the object in self-consciousness cannot be made
compatible with it, but is, on the contrary, destroyed by it.
Herbart, like Hlderlin and Novalis, looks for an alternative explanation for
the existence of the I, which he doubts as little as they do: since it is selfevident, that a contradictory concept, if it cannot be totally dismissed, must
at least be changed.172 With this in mind, Herbart distances himself from
thePlurality of Faculties of Kantian philosophy, which also Hegel, Herbarts
contemporary (and later Nietzsche) have rejected as tautological: a phenomenon is explained by searching for a faculty which belongs to it and makes
it possible. (In this way a cold is explained through a sneezing faculty [Fhigkeit]
and love through the capability of lovingvis amoris, and so on.) In the place
of this Faculty Psychology, Herbart wished to posit a mathematics of the
functions of presentations which Fechner, Mach and Franz Brentano still entertained (according to which even self-consciousness is a function of representations). Not only Hegel, but even Fichte, Herbarts teacher, rejected the
Kantian Plurality of Faculties.
The ways in which Fichte and Herbart realise the identical programme are, however, in their result diametrically opposed. While Fichte searches for a totality of representations deduced from a single principle that is the I, Herbart
attempts to illuminate and deduce the I from the concrete traits of the inner
(mental) life [Seelenleben]. He regards even Fichtes intellectual intuition as
a faculty, and an unjustifiable one, since Fichte cannot show us a way in
which we can capture ourselves epistemically within a finite series of steps.
We must come to Ourselves from the direction of the objects [representations]
and be guided by them, since without them self-consciousness is incoherent,
and undoubtedly cannot be a concern of freedom. He who finds himself in
pain and distress, and admits his weakness, and despairs with himself, he
certainly finds himself [Sich], but in a way he did not want to, and would
not want to, if he had a choice. Here there is even no room for deceptions,
a characteristic that one generally tends to associate with the consciousness
of the will.173
CH 5_f4_53-136 11/23/04 9:28 AM Page 104
Unlike Hlderlin and Novalis (though in some sense compatible with them)
Herbart designates the phenomenon of the I not as something sovereign
which determines, but as an I which is always already determined. We find
ourselves in self-consciousness, a position in which we have not situated ourselves. The states of mind that the I goes through are even less its own work.
(The judgements: I am ashamed, I am sad, I am happy are altogether synthetic, since their predicates are not regarded as inherent in the subject.)174
It is now a contradiction that any determinate A that is represented would
even be able to modify or diminish the act of the As representation. In this
way A would have to be opposed to itself.175
Herbarts criticism of Fichte attempted to show that the object cannot be

found in the I itself, that it therefore has to be looked for in the area of representations, which is placed independently.176 We must therefore firstly
attach to the concept I an unknown object which still remains to be determined; and then we will have to see what follows from that.177 Has Fichte
not shown definitely, however, that the I as self-consciousness cannot be
found amongst objects in the world? He showed, for example, in the
Wissenschaftslehre nova methodo that (. . .) an immediate consciousness never
appears as an object.178 Herbart admits, we should not neglect this (. . .).
That is why the consequence that has just been cited is in need of the following self-critical limitation: (. . .) we can therefore only attach an object to
the I under the condition that it [the object] will disappear again from the
conception of the self.179 But then again in want of an object although not
necessarily the same one, which we forwarded at first.180 In this way one has
to understand self-consciousness as representations [note the plural!] that are
successively absorbed-into-themselves and expelled-from-themselves,181 and
through which many and different objects are alternately represented.182
Thought would not begin, and the solution of the problem would not be in
the least advanced if we wished to remain within the circle of these two
reflections: first, the I needs a differentiating object; and secondly, the I could
never regard a differentiated object as itself. These observations would lead us
to separate the borrowed object, introduce it, and take it away repeatedly;
and oscillation without end, and purpose. If the successive character of
reflection is to be superseded and the result were to be demanded, then the
104 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 105
following obvious contradiction would arise: For the I, there is an external

object that belongs and does not belong to it. This contradiction before us cannot be solved as it stands, by any distinction. As long as we only deal with
a single external object, there is not room for a modification to arise, whereby,
on the one hand, the object would belong to, and on the other hand, would
be expelled by, the I.183
The object of self-consciousness is replaced by a manifold of representations

that remain totally external to it:
What is revealed is that the egoity [Ichheit] is based on an objective manifold.
In this manifold every part is contingent because the other parts still support
the I when something is taken away from it. I posit myself as this or that,
but I am in no way bound to either, as long as I can change. For example
a table can stand which has many feet. In principle, it stands on all of them
at the same time. However, it could do without one of them at a time,
because the other legs would still support it.184
Herbart seems to assume that even if the object of self-consciousness can

never be located in a single representation, without muddying the waters of
self-knowing, epistemic transparency of consciousness is nonetheless possible for itself within a plurality of presentations.
Only if a number of objects are represented could Something in them belong
to the representing agent, namely their unification within a single representative process [ein Vorstellen] and what will later spring from this. This must
provide for the requisite modification in virtue of which the various objects
can be recognised. It, however, does not characterise a single one of them,
and for this very reason it might precisely belong to us. In this way, my
representation of myself would remain dependent on the representations
of objects,it refers to themit does not, however, coincide with them.185
The logic of Herbarts approach leads him to regard this unity not as a transcendental synthesis of apperception, but as a rendezvous for anonymous
(and also unconscious) representations which he attempts to explain by means
of his factical mechanics. We can link this up with a more recent position which Russell characterised as neutral monism when he described
Ernst Mach, William James, and Stout. Consciousness is here explained as
CH 5_f4_53-136 11/23/04 9:28 AM Page 106
ontologically neutral (that is without bias towards either realism or idealism)

based on constellations and configurations of anonymous representations.
First, we must obtain certain objective predicates. These, however, should
not be of a kind which exist in themselves since in the end this would lead
us into the humiliating necessity of merely sticking our knowledge-aboutsomething onto it, as it were externally. Rather, this wonderful knowledge
which returns into itself must emerge by itself out of this objective foundation. That is to say, in a way where objectivity withdraws in the face of
this knowledge so that the I does not encounter the Self as any kind of
determined other, but as its own self.186
We come to ourselves from and through objects and are guided by them.
Without them, self-consciousness remains inconsistent and has certainly
nothing to do with freedom.187
Herbarts approach is not uninterestingand with his psychomechanical

results, he is only apparently at the opposite end of the early Romantics and
Schleiermacher. Both types of approaches refer self-consciousness to something in a non-reductive manner (that is, without undermining the Fichtean
discovery of the problem), indeed even explain it from the standpoint of
something which it itself no longer is: (. . .) at last we can see clearly now
that this knowing-about-oneself [Von-sich-Wissen] refers to something which
is presupposed and has until now been omitted and that we must correct the
mistake by supplementing [die Ergnzung] the omission.188 We know what
Herbart is thinking about: the objective (innerwordly) basis-of-representation, whose primed mechanism should take us out of the regression and circularity of the Fichtean theory of reflection in which the self is continuously
encased anew in its object.
It is doubtful, however, whether self-consciousness can spring out of a configuration of elements, where none of them contains self-consciousness as a
predicate. If it were to contain it, then self-consciousness would only once
again be presupposed and has not been explained by something which is
itself not conscious.
Herbart is convinced that all consciousness is linked to an object and that the
thought of a consciousness without an object is, literally, groundless:189
106 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 107
Without any doubt, when one talks about oneself, everybody (even if it is
only vaguely represented) has something in mind, for a representation bereft
of any object cannot be the true expression of the I. First of all, we must
therefore give the concept I to an unknown object which remains to be
determined and then see what follows from that.190
In this way, our search for an object of self-consciousness is thought according to the model of representation. Every representation [Vorstellung], according to this, is contrasted with something which is posed before us [Vor-gestelltes].
Fichte and his early romantic successors have made the eventual failure of
the search plausible, even if one does not want to accept their suggested positive alternatives. It is impossible to ground the transparency of Self-Knowledge
in a duality of reflections: how is it possible to find oneself in the other without the existence of a trans-reflective unity, which only manifests itself in the
interplay of object and subject within monadic consciousness, but cannot
be explained by it?
Herbart then maintains that through the ingesting and expelling of the
manifold of representations the I manages to contrast its own unity to the
variety-of-the-object. Since what we mean when we refer to the I is something different, something more comprehensive than what is just felt [das
Gefhlte], such as desire, pain, and so on, which supersede each other in the
objective succession of representations:
It might be more difficult to explain what it means to say I find myself feeling [ fhlend ]. However, what is apparent is that feeling (the objective in its
own quality) such as this desire or that pain, in no way provides what we
would regard as our own I.191
Herbart admits that representation as mere sum or aggregate is not anymore illuminating than intuition in its unrelated isolation when we link it to
the thought I.192 It is said, on the other hand, that we can only attain this
thought through and by the objects themselves. The process of representing,
according to this, is first and foremost an Iand non-conscious.193 We gain a
consciousness of the I by tearing ourselves away from the (objective) stream
of representations. Since these are certainly not the product of our freedom,
it is obvious why self-discovery [Selbstfindung] is always associated with a
feeling of involuntariness.194 This involuntariness of the self-representation
CH 5_f4_53-136 11/23/04 9:28 AM Page 108
does not mean that it must not be distinguished from the stream of representations. We emerge from being ensconced in the objects of [our] representations. The predicates [we ourselves] ascribe to the states, mentioned
above, [pain and despair, complacency and pride], are still something objective, although the subject to whom we ascribe them is already presupposed
as known.195 We have already seen that Herbart regards statements such as
I am happy as synthetic. He therefore ascribes something in the predicate to
the representation of the subject which the latter did not contain before. Furthermore, we can only experience the representation of mourning through the
condition that it is ours and that we have already become familiar with it.
Therefore the egoity does not lie in the articulation of what is objective, and
this is in conformity with its own concept (the I). It rather forms an opposite pole even to the predicates attributed to the I. In fact, it is these predicates that allow us to distinguish between the egoity and those predicates,
despite their being conjoined with it.196
Is there not a hidden contradiction here? On the one hand, we are supposed
to obtain our self-consciousness from mere objects (representations), whilst, on
the other hand, one should be able to distinguish self-consciousness from the
aggregate of representations. The first premisethe objective or the unconscious as the point of departureseems to be the stronger one for Herbart.
Thus he says that only this objective content can provide the reason [Grund]
for why we stand apart from the process of its representation.197 The manifold of representations themselves must be constituted in such a way that
it can loosen the chains that would shackle a subject which only knows objects
and never itself.198 Herbart suggests the following solution to the problem:
The demand that our representations should lift us beyond them and bring
us to ourselves is a particular demand which is contained within a more
general one, namely that: in a certain way, what we represent places us outside of the process of representing it. So it is a contradiction to say that any
determined A which is represented would alone be up to the task of modifying or diminishing the act of representing this A. In that way, A would
have to be opposed to itself.
Now no process of representation, taken in isolation, as the process of representing a determinate A or B or C and so forth, can place us outside of
108 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 109
itself. The only possible alternative is to conclude that the different representations, insofar as they are determined as one thing or another by the
various represented elements, are mutually diminishing or modifying. This
is the case, as long as it is determined as this one or that one through its
differing representations. So that one can place us outside of the other.
For the being of I to be possible, the manifold of representations therefore must
supersede one another. This proposition is the result which we will stand by.
It is easy to show that experience proves this proposition to be true. That
it is also extremely fruitful will be shown below.199
Herbarts thought is clear. If self-consciousness is constructed by the object

alone, that is to say, if it is constructed from the material of a manifold of
successive representations, and moreover if each of them were to be differentiated from one another, then this differentiation must itself be a
characteristic of the succession of representations themselves. One representation withdraws in favour of another, and this self-negation tour de
rle of the whole series then lifts the I out as that which does not drown
in the stream.
Certainly someone will ask how this representation of the I comes about
if there was no primary knowledge of oneself that was independent of any
representation. Either these representations are already consciouseach considered by itself (this is what Herbart denies); or they only become known
in their interference (this is what Herbart assumes). In the first instance, consciousness was not explained, but only presupposed. In the latter, one cannot
see how consciousness could arise from the conjunction (or disjunction) of
elements, where none of them were characterised with the predicate of
consciousness. Nor can one see how consciousness succeeded in withdrawing
itself from the series as a whole: Neutral monism falls at the very same hurdle.
Herbart, however, takes his conclusions so far as to categorically deny consciousness the possibility of any self-acquaintance. If the consciousness of
consciousness were to arise out of the fact that each successive consciousness
makes conscious its previous (though in itself unconscious) consciousness,
then, at each moment, the final consciousness in the seriesand in this way
the whole series itselfwould remain unconscious. This is exactly what
Herbart assumes:
CH 5_f4_53-136 11/23/04 9:28 AM Page 110
Amongst the many aggregates of representations, where each successive

apperceives the preceding one, or where a third takes as its object the conjunction or conflict between the first and the secondout of all of them,
then, one must be last. This ultimate apperception will not be apperceived
again.200
The early romantics who tried to overcome Fichtecompared with this absurd
consequence which has reproduced the worst lapses of the empiricist authors
of the seventeenth and eighteenth centuriesappear to be more loyal to
Fichtes original insight and more conscious of the problematic.
VI
Finally, I would like to outline the continuity between the early romantic context of the debate and the speculations of Friedrich Schleiermacher and Sren
Kierkegaard on the subject. This appears to be relatively unproblematic for
Schleiermacher. For although his decisive thought on this theme only came
to light in his lectures on Dialektik in the first decade of the nineteenth century, especially in the introductory paragraphs of his Glaubenslehre,201 they
contain, nonetheless, the conception of self-consciousness as a non-reflective feeling in opposition to knowledge about something202 and as subordinated to the absolute. And these ideas have quite clearly been developed
in the wake of the early Romantics.
Can the same be said about Kierkegaard, the bitter critic of the Romantics?
I think it can. This consequence is not as striking as it might seem, given that
the picture of romanticism outlined in the previous section is significantly
different from the clich vulgar romantic scenario with posthorns sounding,
and moonlight glimmering and with its unwarranted moral and religious
overtones; precisely the kind of scenario against which Kierkegaard directed
his polemic.
In principle little is yet known about the early Romantics. It has only been a
few years since Novalis authentic speculations have become available in a
critical edition. The same is true for the placing of Hlderlins reflections,
which we could not put into context were it not for the reconstruction of the
speculations by the Homburg circleand we have not been able to obtain
Sinclairs and Zwillings reasoning until recently.203
110 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 111
An overview of Friedrich Schlegels philosophical development has only

recently become possible. The vast text, however, still makes it a tricky undertaking to form an overall judgement, since we have to keep in mind that the
critical edition is still far from completion. Just as unsatisfactory is the situation
of Friedrich Schleiermachers work. Most of it is recorded in a wholly inadequate and uncritical manner compiled from many transcripts of his lectures.
And the critical edition has only recently been published in its initial volumes,
among which there are not yet any specifically philosophical textsespecially
not ones that relate to our theme. We, however, have to start with the authentic records of the authors concerned if we want to deal with romantic thought.
These offer a very different picture than the one which Kierkegaard was able
to or wanted to construct, who in any case was not neutral in these matters.
In my discussion on their respective approaches, I wish to bring to light the
convergence of Kierkegaards and Schleiermachers speculations. They agree
already to the extent thatin contrast to Hlderlin and Novalisthey both
have a religious answer to the question of self-consciousness dependence on
whence [Woher] it comes. Kierkegaard emphasised even more resolutely than
the liberal Schleiermacher who maintained that:
Here the word God is articulated as meaning nothing else in our field of
language than what is posited along with the original, absolute accessible
feeling of dependence. Hence, all further determinations must be developed
from this. The accepted view is the converse; that the feeling of dependency
arises from a knowledge of God which is given elsewhere. This, however,
is wrong.204
God is therefore only one of the many possible interpretations into which
self-consciousness gets caught so as to give a name to its feeling of not-beingits-own-origin. With this, a principal right was conceded to non-Christian worldviews, which the religious Anti-Climacus was in no way disposed to admit.
But let us first slow down and see how Schleiermachers analysis of subjectivity
works itself through to the thought of faith. In the handwritten marginalia of
his first edition of the Glaubenslehre, he defined faith as das im Selbstbewutsein
mitgesetzte Bewutsein vom Mitgesetzten [a consciousness of co-positing which
is co-posited in self-consciousness], a formula that almost reminds us of
Kierkegaards reflection on the self at the beginning of The Sickness into Death.
CH 5_f4_53-136 11/23/04 9:28 AM Page 112
Schleiermachers Dialektik, unlike early Idealism, does not treat the subject
as a philosophical starting-point. Within the course of an unanchored experience that unconsciousness forms from the simultaneous differences and
interconnections of its internal and external functions, the subject itself becomes
a theme precisely at the moment in which it seeks the ground that will explain
the respective relations and differentiations. Such an experience cannot be
assuaged by the discovery of a ground that is still determined solely as a
higher order function, with respect to its subordinate functions; a ground that
at the end of an even deeper reflection will eventually be revealed as a relation of a relation which has hitherto remained invisible. On the contrary, such
an experience can only be assuaged by a ground that ultimately explains the
fact of differentiation and unification.
In relation to this ultimate ground, Schleiermacher now realises that its likelihood increases successively with experience which consciousness forms
from the dissonant yet uniform organisation of its functions. Indeed, he
realised that any relative syntheses which are achieved on the way only
become understandable through an absolute presupposition of a unity which
simultaneously manifests itself in and withdraws itself from them, without
it being clear (due to the bipolarity of consciousness) how this unity could
catch up with its endpoint at any point along the path that is followed. It
therefore makes sense to call this endpoint transcendent in relation to the
level of relations in which consciousness is active. The essential law, inscribed
in consciousness, is that of its orientation towards the transcendent ground,
wherein the difference between dispute and counterdispute would be annulled
once and for all. This seems to contradict the very structure of consciousness,
since even the most stable experience of which it is capable cannot get around
the fact that none of the functions that have been isolated in the analysis can
be maintained in this methodical abstraction without spilling over into the
next, such that all the oppositions arising in the process of consciousness
self-understanding can only be described as the temporary prevalence or disappearance of one opposition in favour of another.205 The unity which is searched
for its own sake can only be glimpsed at the point of the interval of the transition between differences [Differenten], without being able to escape the
dimension of the relation and emptiness.
112 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 113
In this way the organic function, which is itself already a generic term for
diverging and converging moments, and the intellectual function for which
the same holds, is united in the concept of thought.
What this formulation omits, namely, that the proof that the ultimately founding instance of any self understanding can be found neither in the syntheses
of thinking nor of willing but only in their common origin: This is furnished
by the fact that both thought and will exhibit a twofold lack which is carried
over into their respective spheres in the form of a relation (thinking-thought
or freedom-necessity and willing-willed or purpose and material resistance).
Moreover, these pairs are defined as mutually completing moments which
are opposed to one another and hence are again in a relation (towards each
other). This relation cannot avoid the law of the object:206 That is, the transition between thoughts is mediated by free productivity (even receptivity
would require it in order to be recognised as sensitivity [Empfnglichkeit]). In
just the same way the acts of volition require intervention of cognitive acts,
so as to be able to model themselves in accordance to aims. The syntheses of
thought and will have not failed because they did not bring about any hint
of a transrelative unity, but because they were not able to ground this unity
in their own sphere.207
Immediate (or immaterial) self-consciousness and feeling208 are the terms
that Schleiermacher gives to this function, a function that surpasses the duplicity of willing and thinking and tends towards the unity manifested in their
intertwinement. The attribute immediate is supposed to suggest that we
are dealing with a form of consciousness in which the relata of what is reflected
upon, and of what performs its reflection, no longer divergeas was the case
with all previous syntheses. Both functions should be thought in unison as
mutually self-negating moments of a single and integral reality.209 That is to
say, as moments of an action which becomes transparent to itself in its own
realisation, of a form of being that always manifests its own appearance
or in whatever way one may wish to express the coexistence of the deed
and its reflex. In the choice of the appropriate conceptual form what is paramount is to do justice to the fact that here the opposition between subject
and object (. . .) (remains) utterly impossible and inapplicable.210 This enables
Schleiermacher to insist repeatedly upon calling immediate self-consciousnessin opposition to the mediated oneimmaterial [ungegenstndlich].
CH 5_f4_53-136 11/23/04 9:28 AM Page 114
Here we encounter a theory of the subject which is strongly reminiscent of

the one we became acquainted with earlier in the context of the speculative
thought of the early Romantics and Hlderlins, a theory whose historical
roots lie in Fichte. As we know, Fichtes philosophy achieved its breakthrough
with the discovery of the impossibility of explaining self-consciousness with
the model of reflection (in order to be able to judge oneself as the reflected
term, the one who reflects must have a prior knowledge about himself). Few
of his contemporaries show such depth and dexterity in their understanding
of this theory as Schleiermacher. The relevant passages from the Glaubenslehre
and the Dialektik actually read as if Schleiermacher had been familiar with
all the most standard formulations of the recent literature on Fichte,211 which,
by the way, has hardly been influenced at all by Schleiermacher.
In immediate self-consciousness we are concerned with a matter whose
mode of being cannot be described in propositions of the same grammar with
which one forms statements [Aussagen] about the world of objects. Immediate
self-consciousness is so radically opposed to objectified consciousness or
knowing about something, 212 that any attempt to confront it with itself by
mediating it via a representation which it can then observe fails to do justice to its own peculiar mode of being [Seinsweise].
The reflection theory of consciousness does just this. It attempts to grasp immediate self-consciousness as representation of itself, as reflection213 or to grasp
it as objective consciousness,214 whereby the object is supposed to be the
subject itself. A theory of reflection could not explain the indisputable fact of
the self-possessing [Sich-selbst-Haben].215 On the other hand, Schleiermacher
was aware that heightened pure feeling could not avoid the discourse of the
theory of reflection (which he denied).
Nobody has been more perspicuous than Schleiermacher in unravelling this
aporia, upon which Fichtes approach failed. I summarise his complicated
thought fragments that can be found scattered through his texts in the following way: one can neither think the reflection without presupposing a simple self-conscious identity (otherwise one limb of the relation could not be
certain of seeing in the other itself rather than just the other limb). Nor can
one disregard the fact that this identity is not immediately present itself, but
has to call upon the otherthe other linked relataas a witness for it to be
identical to itself. As we have seen, Hlderlin had depicted the aporia in
114 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 115
1794/5. Fichte, to whom he refers, did indeed discover that the evidence of
the other must be authenticated by a pre-reflexive knowledge of ones own
identity. But he also became entangled in a circle. He explains in his lectures
Wissenschaftslehre nova methodo that the determinacy of the thought I (its concept) is linked to the difference between at least two mutually differentiated
expressions (You think the I, and hence you think nothing else; you therefore do not think the Not-I), for a thought can only have conceptual distinction when it can delimit itself negatively against everything that it is not.
Fichte called this differential basis of all conceptual knowledge [Erkenntnis]
the law of reflection in all our cognition. On the other hand, we have to get
around the diremption of the two links by means of an immediate intuition
[Anschauung] of their non-separation, otherwise the other is no longer the same
as the one, and the indispensable identity of the thought I has been lost.
This explanation gets us nowhere. The fact that the I is exhausted by the other
becomes its own condition of possibility. Even if it denies the ground of its
own differentiation, the self is still split into two. The path from the reflected
term to the reflecting leads through a mediation that cannot be circumvented.
The question now becomes: how can the thought of insuperable (and conceptually unavoidable) mediation be reconciled with that of necessary immediate familiarity? Schleiermacher has an ingenious answer to this question.
What is made clear to immediate self-consciousness, when it flickers to and
fro between the two poles of the reflexive split, is not the consciousness of
the perfection of superreflexive identity, but rather the negative consciousness,
that is, its lack.
He always says that self-consciousness crosses the empty space of an absent
unity216 in the moment of transition217 from what is reflected to that which
reflects. Since the self cannot ascribe this lack to its own activity, it must recognise it as an effect caused by a transcendent certainty [that is beyond its
power].218 This again can only be expressed by means of a transcendent
ground. It is (positively) determined even before it starts to determine
itself (actively). In other words, (and here I will combine various citations
from Schleiermacher), feeling has access to itself precisely because it reads
the imprint [Prgemal] of its transcendent determination219 as an indication
of an identity which supplements the defect220 inscribed in reflection. Roughly
speaking this is the conclusion reached by the Analysis of Self-Consciousness
CH 5_f4_53-136 11/23/04 9:28 AM Page 116
in Relation to the co-positedness [Mitgesetztsein] of an Other,221 and which

leads it into the religious disposition of assuming a feeling of dependency
as such [schlechthinnig] with respect to a whence [woher] to which it owes
its determination as immediate self-consciousness, a determination which it
cannot attribute to itself as its own achievement. Its power is broken upon
this facticity of unavailable self-mediation.222
Schleiermacher attempted to capture this crisis of the subject with the
following formulation. In self-consciousness the Ideal- and the Real [Real]
foundation diverge. That means that the self is the ground for its self-cognition [Sich-Erkennen], but not however, the ground of its being. It has not
brought itself to a state of immediate self-intimacy. However, once it is there,
it is just as free, responsible and spontaneous in its thoughts as Fichte and
Kant had allowed.
Schleiermacher has emphasised over and over again that the predicate
schlechthinnig [as such] is the German translation of the word absolute. Selfconsciousness is only dependent absolutely, or in other words, on the absolute.
It is not the ground for its being. This does not mean that it depends on an
innerworldly or natural authority. Such a dependency, as mentioned for example by the materialists, would have to be called relative.
I hope I have managed to make apparent the complete analogy of Schleiermachers construction of self-consciousness, in particular with that of Novalis.
Both authors have one and the same problem in front of them. For conveniences sake we must think of the self as a relation between two relata.
If we wish, however, to avoid the circularity of the theory of reflection, this
relation must be circumvented through a pre-reflective self-acquaintance. This
acquaintance alone could ground an identity where being-itself is maintained irrespective of its articulation as relation, and hence as difference. Since
this identity is no longer within the reach of (conceptual) knowledge, it can
only be postulated as a necessary condition for feeling in which two complementary, seemingly mutually exclusive experiences are required: 1) that
the identity must exist in order to explain the existence of feeling; 2) that
this identity does not itself enter into this feeling. In this way, feeling shows
up a fundamental lack in experienceand it interprets this lack as the absolute
which is longed for. In feeling the whence of its inaccessible determinacy
is disclosed.
116 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 117
VII
This is the fundamental experience of the early Romantics par excellence. We
can find it again with Friedrich Schlegel and Karl Wilhelm Solger, whose
theories of self-consciousness ought to be covered by this paper but which
I must omit due to lack of space. Is it still this same experience for Sren
Kierkegaard?
I must, again due to lack of space, draw only on one single passage from one
text; namely, the beginning of The Sickness unto Death.223 Kierkegaards interpreters are confronted with some characteristic problems. On the one hand,
its author, the edified Anti-Climacus, writes with a Christian ardour. A devotional writing has no arguments. On the other hand, nowhere in Kierkegaards
complete work (as far as I am aware) can you find a similarly condensed
speculative remark on the theme we are concerned with here; that is to say,
the topic of self-consciousness.
The book starts in the first paragraph with the famous but obscure words:
A human being is spirit. But what is spirit? Spirit is the self. But what is
the self? The self is a relation that relates itself to itself or is the relations
relating itself to itself in the relation; the self is not the relation [Verhltnis],
but the relations relating itself to itself.224
This structural sketch appears unnecessarily complex. Obviously the term

relation [Verhltnis], the understanding of which is difficult, has different
meanings in different contexts. We can distinguish three ways through which
the self appears as relation [Verhltnis], or to use the language of the early
romanticsas relationship [Relation] and moreover as reflection.
1) Kierkegaard explains this in the statement that follows. The self is a relationship [Relation] between infinity and finitude, temporality and eternity,
freedom and necessity. If I had more space, I would and should have to show
that these determinations, just like the determination of relation [Verhltnis],
have already appeared in Schleiermachers Dialektik and in his Glaubenslehre, and with the same function. Obviously, Kierkegaard has studied this
essential work of protestant dogmatics thoroughly, as, for example, one can
discover in his diaries.
2) The relation of relationship is the self in its comportment [Verhalten] towards
the two moments, or better still: as self-comportment in bipolarity. The self
CH 5_f4_53-136 11/23/04 9:28 AM Page 118
and this ambiguity leads to what Kierkegaard calls his possible despairis,
as we say in German (and as Adorno in particular was fond of putting
it) forced to be [ist dazu verhalten] infinite and finite, subject and object,
eternal and temporal. Yet it is never able to be only the one at the cost of
the other.
How can one explain the confusing duplication in the formulation by which
the self is a relation [Verhltnis] that relates to itself? Obviously only by differentiating between two aspects in the subject-object relationship [Relation],
which I take as a representation of all other semantic renderings of the same
phenomena. First, the relationship [Relation] itself and as such, and secondly,
the composites out of which it has been composed, (or better still, to which
the relation leads), that is, subject and object, freedom and necessity. In a certain sense one must therefore say that the relationship [Relation] as such is
not relative. Only the endpoints subject and object are relative. In this way
it is impossible to represent the relation [Verhltnis] of both, in whose middle the self exists, the non-relative (or in idealist terms: absolute) identity of
the poles as such. The identity would be the whence of the relationship
[Relation] just as it was for Schleiermacher. Kierkegaard expresses this by saying that man is only a synthesis of both relata but not the self as such
because this self is the relationship and is therefore not in the relationship.
(In this way man is not yet a self).225
3) In this manner we have also secured the third aspect. The self is a relation
as well as a comportment [Verhalten] its whence, its ground. I quote Kierkegaard:
If the relation that relates itself to itself has been established by another,
then the relation is indeed the third, but this relation, the third, is yet again
a relation and relates itself to that which established the entire relation. The
human self is such a derived, established relation, a relation that relates
itself to itself and in relating itself to itself relates itself to another.226
Here we are confronted with the same confusing formula of a doubled relation [Verhltnis]. This time we are dealing with the relation [Verhltnis] of the
subject-object relationship [Relation] towards that which allows it to exist as
unity. This unity, which cannot be represented in the relationship [Relation],
is its whence or its ground. As we can remember, Schleiermacher talked
118 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 119
about the co-positedness [Mitgesetztsein] of the other in reflexivity or as

Kierkegaard says in the relation [Verhltnis], that is, in the synthesis of the
self as a link [Beziehung] between its subject and object pole. We can express
this more simply: Every relation [Verhltnis] presupposes two moments. What
cannot be explained by the relation tout court, is the fact that this duality in
no way prevents the thought that what can be determined as either infinitive or finite comportment [verhalten] is nonetheless one and the same thing,
that is, the self. To explain this fact a third is needed, which can only be
alluded to ex negativo in the syntheses, and which allows it to grasp itself,
despite its relativity, as related [bezogen] (or comported [verhalten]) to an
absolute. This is expressed by Kierkegaard when he says that the self only
relates [verhalten] to itself by relating itself to another; since only this other
permits it to exist in a finite relation as a self-relation, and not an external
relation between different things.
I can see, at least up until now, no fundamental difference here from
Schleiermachers model. Both, as the early Romantics already had done,
acknowledge self-consciousness relativity as an uncircumventable condition
of our (philosophical) description of it. And both add that this relativity
demands that we transcend it to a trans-relative third or other which founds
it and without which the relation could not be experienced as a self-relation.
We have to say that the epistemological [erkenntnistheoretisch] aspect is peculiarly under-investigated in Kierkegaard. We have supplemented his text making use of the knowledge that we have of the texts already discussed, rather
than because he has dealt with this aspect explicitly.
In contrast, we can find in Kierkegaard a quite peculiar interest in the freedom of the self, in opposition to the early Romantics. It is the foundation of
his possible despair. Of course the relationship knows itself to be grounded
in a third, that is, not itself. This foundation, however, does not rid the relationship of the undecidability of the two poles, between which it is swaying
to and fro as if it were in a frenzy of possibility. Kierkegaard says that this
despair is a negativity;227 that is to say, it exists because the counterpart, the
opposing half, or the other pole, has always been withdrawn from the comportment [Verhalten] in a determinate situation. Hlderlin knew this, when he
despaired about the fact that while I have the one, the other is absent. This
despair about comportment is therefore the experience of an absent unity: an
CH 5_f4_53-136 11/23/04 9:28 AM Page 120
absolute identity which escapes the play of appearance and counter-appearance and which would not be merely relative.
The self is essentially an unhappy consciousnesswithout being able to
transport itself into the realm which has escaped it [das Verfehlte], as is the
case in Hegels Phenomenology of Spirit. Or perhaps we should rather say: this
transportation no longer occurs in knowledge but in and as faith. At the same
time, faith has got a speculatively accessible basis, as in Schleiermacher. Only
he who can contrast the being torn between two comportments with a condition of successful reconciliation can truly despair, just as only he who is
aware of the idol of a happiness that has escaped him can be unhappy.
Thus the absolute, a state from which anything relative and negative would
have been removed and which is an object of faith, remains for ever present
in spite of its withdrawal. Kierkegaard alludes to this when he refers to the
peace and the equilibrium which comportment cannot achieve by itself.228 It
is the point of transition or indifference of Schleiermachers Dialectic, which
has also here aided Kierkegaards thinking.
Now one might say: it is this which forms the actual despair of the Kierkegaardian self, since it cannot rid itself of the condition of negativity and can
therefore not move across into the absolute. Like Tantalos, in the way in which
his food is constantly withdrawing from his outstretched hands, the self is
condemned to remain in a passion inutile.
Two things can be said about this: Firstly, Kierkegaardor at least the edifying author Anti-Climacusis Christ, and despair is a test that should not
be interpreted as an insurmountable conditio humana as did his later (atheist)
pupil Sartre. The leap from despair into belief lends a new meaning to the
negativity of the relation. It only lacked this meaning as long as it remained
encapsulated within the duality of reflection. Secondly, it would be wrong,
at least superficially, to oppose Kierkegaard to Schleiermacher and the early
romantics by saying that the one remains in negativity while the others happily slid into a certainty about the absolute that has only just been gained.
Friedrich Schlegel defined the romantic as yearning for the eternal,229 and
he added that this yearning is never satisfied in a fulfilment: Something
higher [than the yearning for eternity] does not exist in man.230 Reflection
has to be aware of the idea of an absolute in order to explain its relative unity
120 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 121
in which it sustains itself. To want to reach this unity (which Novalis, in a

very Kantian way, has called regulative)231 as an absolute is a senseless
undertaking since reflection would have to destroy its goal by the means
through which it wishes to achieve it. That is why our state is essentially one
of a yearning; and yearning can only be experienced by a being which lacks
that which it is yearning for. If one wishes to call this state despair, then the
romantics would be more despairing than the believing Christ Anti-Climacus.
Translated by Peter Dews and Simon Critchley.
Prepared by Dieter Freundlieb.
* Manfred Frank, Department of Philosophy, University of Tbingen, Bursagasse 1,
D-72070, Germany.
Notes
1
David Hume, A Treatise of Human Nature, ed. A.L. Selby-Bigge, Oxford, Oxford
University Press, 1978, pp. 635-6.
Kant, Kritik der reinen Vernunft, B134, footnote.
Ibid., 135.
To show this is the object of Dieter Henrichs work, Identitt und Objektivitt. Eine
Untersuchung ber Kants transzendentale Deduktion, Heidelberg, Winter, 1976. Cf.
idem, Kant und Hegel. Versuch der Vereinigung ihrer Grundgedanken, in
Selbstverhltnisse, Stuttgart, Reclam, 1982, pp. 173-208, esp. pp. 176 ff.; and idem,
Die Identitt des Subjekts in der transzendentalen Deduktion, in eds. Hariolf
Oberer and Gerhard Seel, Kant. AnalysenProblemeKritik, Wrzburg, Knigshausen & Neumann, 1988, pp. 39-70.
Kant himself also attributes numerical identity to self-consciousness, for example
in the chapter on the paralogisms in Kant, Kritik der reinen Vernunft, A344 = B402,
A361 ff. This identity can only be called sequential [durchgngig] if it is not relationless, but proves itself to be one and the same in relation to a manifold of changing representations (for example in the temporal sequence). Thus in the schema of
A344 Kant defines the unity of self-consciousness as one which is attributed to it,
according to the different times at which it is present. Since this sequential unity
is one of self-consciousness, it is also known to itself; or in other words the subject is one and the same at different times and knows itself to be so. Its unity is a
knowing-itself-as-identical-with-itself at any given point in time or more generally: at any given point in an intuition (Karen Gloy, Die Kantische Theorie der
Naturwissenschaft. Eine Strukturanalyse ihrer Mglichkeit, ihres Umfangs und ihrer
Grenzen, Berlin & New York, de Gruyter, 1976, p. 117).
CH 5_f4_53-136 11/23/04 9:28 AM Page 122
I have examined this in more detail in my Einfhrung in Schellings Philosophie,

Frankfurt am Main, Suhrkamp, 1985, pp. 32 ff.
Dieter Henrich has shown the limits of the legitimacy of this model in Ding an
sich. Ein Prolegomenon zur Metaphysik des Endlichen, in Vernunft des Glaubens.
Wissenschaftliche Theologie und Kirchliche Lehre (Festschrift for Wolfhart Pannenberg),
eds. Jan Rouls und Gunther Wenz, Gttingen, 1989, pp. 42-92.
Kant, Kritik der reinen Vernunft, B141/2.
Ibid., B278.
For example, ibid., B139 ff.
10
Ibid., B402: Now it is clear that I cannot know what I must presuppose in order
to know an object, and that the determining self (thinking) is distinct from the
determinable self (the thinking subject), as knowledge is from objects.
11
Ibid., B404, A355.
12
Ibid., B155-157.
13
Ibid., cf. note B157 ff.
14
Ibid., A366 and B404.
15
Ibid., B158.
16
Schelling accused Descartes of a false deduction of this kind (from veritative to

existential being) in his Munich Lectures On the History of Modern Philosophy: The
sum comprehended in the cogito thus only means: sum qua cogitans, I am as thinking, that is, in that specific kind of being which is known as thinking, and which
is only another kind of being than, for example, that of the body whose kind of
being consists in filling space, in other words excluding every other body from the
space which it occupies. The sum included in the cogito thus does not have the
meaning of an unconditional I am, but only of I am in a certain way namely as
thinking, in this mode of being which is called thinking. (Schellings Smmtliche
Werke, ed. K.F.S. Schelling, Stuttgart, 1856-61, 1/10, p. 10.)
17
Kant, Kritik der reinen Vernunft, B422/3.
18
Ibid., A68/9 = B93/4.
19
Ibid., B422.
20
Ibid., A342 ff.-B400 ff.
21
Ibid., cf. BXL/XLI, A107, B68, B156, B277, B430.
22
Ibid., B428, 430 ff.
23
Ibid., B430, with B423, (real [real] in this context obviously means existent/actual).
24
Ibid., B422/3N.
25
Kant, Metaphysical Principles of Natural Science, Kants gesammelte Schriften,

Akademieausgabe, Berlin & Leipzig, 1900 ff., vol. IV, p. 543.
26
Kant, Kritik der reinen Vernunft, A598/9 = B626/7.
27
Ibid., A598/9 = B625.
122 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 123
28
Ibid., A354/5. Kant also says immediate, or tautological.
29
Cf. also ibid., A598 ff. = B626 ff.
30
Ibid., A219 = B266.
31
Ibid., A255 = B272/3; cf. A374 ff.: perception is the representation of an existence;
what . . . is represented through perception is . . . also existent.
32
See also W. Ltterfelds, Zum undialektischen Begriff des Selbstbewutseins bei

Kant und Fichte, in Wiener Jahrbuch fr Philosophie 8, Vienna, 1975, pp. 7-38, here
p. 19. Along with most other interpreters, Wolfgang Becker (Selbstbewutsein und
Erfahrung. Kants transzendentale Deduktion und ihre argumentative Rekonstruktion,
Freiburg & Mnchen, Alber, 1984, pp. 239 ff.) interprets talk of the experiential
character of the proposition I think, or of the indeterminate inner perception,
in terms of the emptiness, in other words depending on experience, of the pure
cogito (in a similar manner to Sartres ontological proof of the intrinsic nothingness of conscious, on the basis of reference to being). But the emptiness of the I
merely shows that it only appears in connection with the sensible (for example
with the pure forms of intuition); it does not imply that what it is related to must
be something experienceable (empirical). Thus the necessary co-givenness of sensation does not follow from the emptiness of the I think.
33
Schelling had already accused Kant of this unintended consequence, with reference to the footnote to B422/3 (Schellings Smmtliche Werke, 1/1, p. 401 ft). He
was entirely clear that this kind of intellectual intuition does not come into contradiction with that which Kant explicitly rejects, in which the spontaneity of
understanding autonomously generates the sensory material (ibid., p. 181 ff.). I
have set out and interpreted the relevant references in my Eine Einfhrung in
Schellings Philosophie, p. 42.
34
Kant, Kritik der reinen Vernunft, A91 = B123.
35
Ibid., A90 = B122.
36
Ibid., A345/6 = B404.
37
Ibid., A402.
38
Ibid., A366.
39
I cannot know as an object, that which I must presuppose in order to know an

object at all . . . (ibid., A402). In thinking the categories, the subject of the categories cannot acquire a concept of itself as an object of the categories; for, in order
to think them it must take its pure self-consciousness, which was to be explained,
as the ground (ibid., B422).
40
Kant, Reflexion (of uncertain date: 1770/1?, 1769?, 1773-5?, in Kants gesammelte
Schriften, vol. XVII, 509/10).
41
There is also a self-consciousness corresponding to freedom; one can readily foresee the consequences that would be entailed by the assumption that the I of theA History of the Theory of Self-Consciousness 123
CH 5_f4_53-136 11/23/04 9:28 AM Page 124
ory is other than the I of action (with action falling into the unconscious). Cf.
Jrgen Stolzenberg, Das Selbstbewutsein einer reinen praktischen Vernunft, in
Metaphysik nach Kant?, eds. Dieter Henrich and Rolf-Peter Horstmann, Stuttgart,
Klett-Cotta, 1988, pp. 181-208, esp. pp. 183 ff. Stolzenberg refers, besides the Kritik
der praktischen Venunft, to the Grundlegung zur Metaphysik der Sitten. The unconditional law as self-consciousness of freedom cannot appear in the form of knowledge. For it is evident, that the concept of the self-consciousness of a pure practical
reason as considered by Kant must be termed the concept of a self-consciousness
a priori. Thus it is the concept of a self-reaction in which reason does not possess
a knowledge of itself, and therefore fundamentally unsuited to provide an answer to
the question of the ground of the knowledge of freedom, which is essential for Kant.
42
I have explained this in more detail in the special issue, edited by me, of the Revue
Internationale de Philosophie, vol. 42, no. 166, 3/1988, pp. 361-82 (Comment fonder
une morale aujourdhui?).
43
Kant, Kritik der praktischen Venunft, p. 56. See, on this concept, the still unsurpassed
early text of Dieter Henrich, Der Begriff der Sittlichen Einsicht und Kants Lehre
vom Faktum der Vernunft, in Die Gegenwart der Griechen im neueren Denken
(Festschrift for H.G. Gadamer on his 60th Birthday), Tbingen, Mohr (Siebeck),
1960, pp. 77-115.
44
Kant, Kritik der praktischen Venunft, p. 9.
45
Cf. ibid., pp. 9/10.
46
Kant, Reflexion no. 4336 in Kants gesammelte Schriften, vol. XVII, 509 f. Further evidence concerning Kants view of freedom as intuited intellectually are to be
found in Reflexion no. 4228, ibid., vol. XVII, 467, no. 4224, ibid., 470.
47
Kant, Kritik der praktischen Venunft, 55 f., (the first emphasis is mine; M.F.).
48
Kant, Kritik der reinen Vernunft, B430 ff.
49
Ibid., B430/1.
50
Kant identifies to a large extent the concepts of Spontaneity and of the intellectual: The intellectual is that whose concept is an activity (Reflexion no. 4182, Kants
gesammelte Schriften, vol. XVIII, p. 447). Thus the activity which is intuited in intellectual intuition is that of the intelligence itself: the I combine Cf. also Kant, Kritik
der reinen Vernunft, BXL f., N.
51
Cf. Dieter Henrich, Die Anfnge der Theorie des Subjekts (1789), in Zwischenbetrachtungen. Im Proze der Aufklrung, Jrgen Habermas zum 60. Geburtstag, eds.
Axel Honneth, Thomas McCarthy, Claus Offe & Albrecht Wellmer, Frankfurt am
Main, Suhrkamp Verlag, 1989, pp. 106-170, esp. pp. 139 ff. The fragmentary character of the present sketch in part derives from the exclusion of the lesser Kantians
(Jacobi, Reinhold, Maimon, AenesidemusSchulze, Beck).
52
Leibniz, for example, in his Discours de Mtaphysique, 34. In an even more Kantian
124 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 125
manner in the Philosophische Schriften, ed. C.I. Gerhardt, Berlin, 1875-1890, Vol. 2,
p. 53, Leibniz writes: la notion de ce moy lie ou comprenne les diffrens estats
[du sujet].
53
Leibniz, ( 4 of Principles of Nature and of Grace; cf. Monadology, 23 and 30: nous
sommes levs aux actes rflexits, qui nous font penser ce qui sappelle moi).
54
C.A. Crusius, (Entwurf der nothwendigen Vernunft-Wahrheiten, wiefem sie den zuflligen entgegen gesetzet werden, Leipzig, 1745, p. 863).
55
Kant, Reflexion dating from 1769 (No. 3929, Kants gesammelte Schriften).
56
Kant, Vorlesungen ber die Metaphysik, ed. K.H.L. Plitz, Erfurt, 1821, p. 135.
57
Descartes, uvres et lettres, ed. Andr Bridoux, Paris, 1953 [= Bibl. la Pliade], pp.
879, 284, 287, 289 passim.
58
Ibid., p. 899.
59
Ibid., 1359.
60
Ibid., p. 291.
61
Descartes saw the problem, but did not perceive its disastrous consequences for
his explanation:
Cest une chose trs assure que personne ne peut tre certain sil pense et sil existe, si
premirement, il ne connat la nature de la pense et de lexistence. Non que pour cela il
soit besoin dune science rflchie, ou acquise par une dmonstration, et beaucoup moins
de la science de cette science, par laquelle il connaisse quil sait, et derechef quil sait quil
sait, et ainsi jusqu linfini, tant impossible quon puisse jamais avoir une telle daucune chose que ce soit. ibid., pp. 526/7.
However, Descartes does not draw from this aporetic observation the obvious conclusion that self-consciousness cannot be thought on the model of a representation of ones own representation. Rather, he has recourse to this model in what
follows: mais il suffit quil sache cera par cette sorte de connaissance interieure
qui prcde toujours lacquise . . . Every form of knowledge (connaissance) is
the representation of an object distinct from it. Once the object is separated from
it, no technical term will be able to show that, at the same time, it should not be
separated from it, in order to be, in the radical sense, the subject of knowledge.
Descartes obliviousness to the disastrous consequences of the reflection model
can also be seen in Burmans notes on the conversation which I have already
quoted. Burman had asked if the reflection model did not imply that the reflected
subject is no longer in the same temporal location as the reflecting subject, thereby
requiring a consciousness of having thought to be brought to light. Descartes replies:
the consciousness of representing (cogitare) does in fact arise from a reflection on
this representing. He simply denies the temporal gap between the act and the
noticing of the act, and suggests that in general the soul can represent several
things at once. If this answer is authentic, it simply shows that Descartes does not
CH 5_f4_53-136 11/23/04 9:28 AM Page 126
perceive the specific difference between a (successful) self-representation and the

representation of an object. The former can never draw its certainty of identity
from a numerical doubling into representing and represented unless the represented is already recognised as subject before the representation, and simply repeats
this implicitly in and for knowledge.
62
Leibniz, Principles of Nature and of Grace, 4.
63
Leibniz, Monadology, 30.
64
Ibid., 23.
65
Hume, Treatise of Human Nature, p. 252: For my part, when I enter most intimately
into what I call myself, I always stumble on some particular perception or other,
of heat or cold, light or shade, love or hatred, pain or pleasure. I never catch myself
at any time without a perception, and never observe anything but the perception.
On the definition of identity, cf. ibid., 14, pp. 200 ff.; for the (sceptical) application
of this concept to the person, pp. 253 ft. and pp. 633 ft.
66
Of an unknown English author, dating from 1728, Pseudo-Mayne, ber das

Bewutsein, translated and with an introduction and notes by Reinhard Brandt,
Hamburg, 1983 (cf. esp. pp. XXVII ft). Brandt rightly states: the history of the
philosophy of consciousness will in future no longer be able to pass over [this
first monograph on consciousness.]
67
Ibid., p. XXXI.
68
Ibid.
69
Ibid., p. XXXII.
70
Thus runs the second sub-title of his Solid Philosophy Asserted, Against the Fancies
of the Ideist: Or, The Method of Science Farther Illustrated, London, R. Clavil, 1697,
reprinted as a facsimile by Garland Publishing Inc., New York and London, 1984,
with John Lockes hand-written marginalia. I would like to thank my colleague
Richard Glauser for pointing out to me this remarkable author and his objections
to Locke, derived from reflection theory. For further information I refer the reader
to his essay John Sergeants Argument against Descartes and the Way of Ideas,
in The Monist, Oct. 1988, vol. 71, no. 4, pp. 585-595.
71
John Sergeant, Reflexions on Mr. Lockes Essay Concerning Human Understanding, Solid Philosophy Asserted, Against the Fancies of the Ideist: Or, The Method
of Science Farther Illustrated, London, R. Clavil, 1697, p. 121.
72
Ibid., p. 122.
73
Ibid.
74
Ibid., p. 124.
75
Ibid.
76
Ibid., p. 125.
77
Ibid.
126 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 127
78
Pseudo-Mayne, ber das Bewutsein, pp. 6 and 8.
79
Ibid., pp. 6-7; with reference to practical self-consciousness, without which responsible and purposeful action would be inexplicable, cf. pp. 64-5.
80
Ibid., pp. 40-1.
81
Ibid., pp. 8-10.
82
Animadversions upon Dr. Sherlocks Book, Entitled A Vindication of the Holy and
Everblessed Trinity, etc. London, 1693, p. 71, cited in Pseudo-Mayne, ber das
Bewutsein, ed. Reinhard Brandt, Hamburg, 1983 p. 111, n. 19.
83
Bishop Peter Browne, The Procedure, Extent and Limits of Human Understanding,
London, W. Innys, 1728.
84
Ibid., p. 67; cf. ed. Brandt, Pseudo-Mayne, ber das Bewutsein.
85
Cf. esp. ibid., p. 100.
86
Ibid., p. 8u.
87
Consciousness is the Minds immediate Perception of itself, considered as now in the

Act of perceiving itself: which is saying, that Self-Perception is its own proper Object;
or that the Mind, in perceiving itself, is sensible of itself, as that which perceives, and
is Conscious of itself. ibid., p. 42.
88
Ibid., p. 10 n. Bishop Browne goes further, in the treatise already cited, The Procedure . . ., when he strictly distinguishes self-consciousness, as a type, from objectconsciousness (and thus also from perception), although this does not prevent him
from speaking of inner Perception or inner Feeling: Another Kind of Knowledge
is that which we have from self-Consciousness. As we came to the knowledge of
things without us by the Mediation of their Ideas; so on the contrary we have an
Immediate Feeling or Consciousness of what is transacted in our Mind, without the
Intervention of any Ideas whatsoever . . . This kind of Perception some have not
inaptly called Internal Sensation, in order to distinguish it from the Perception we
have of External objects by their Ideas. (Browne, The Procedure, pp. 124-5). However,
the predicate internal is no more sufficient to repair the damage done by the
reflection model, than the assertion that self-consciousness does not follow the act
(as Browne himself had maintained a page earlier ([ibid., p. 125]), but falls indistinguishably within it (ibid., p. 126). Talk of the immediacy of self-consciousness
always indicates insight into a set of problems, even when no suitable conceptual
tools are available for their solution.
89
Pseudo-Mayne, ber das Bewutsein, pp. 48-9 ff.
90
Kant, Kritik der reinen Vernunft, B133.
91
Kant, Kritik der reinen Vernunft, A366, B404.
92
Fichte, Werke, herausgegeben von Immanuel Hermann Fichte, Berlin, 1845, Neudruck
93
Berlin, de Gruyter, 1971, vol. I, p. 527.
CH 5_f4_53-136 11/23/04 9:28 AM Page 128
94
95
Ibid., 24/35 of the Transcendental Deduction.

Fichte, Wissenschaftslehre nova methodo, in Nachgelassene Schriften, ed. Hans Jacob,
vol. 1, Berlin, 1937, p. 357.
96
Fichte, Werke, vol. I, p. 521, p. 528.
97
Cf. Fichtes ursprngliche Einsicht, Frankfurt am Main, 1967, and La dcouverte de

Fichte, in Revue de mtaphysique et de morale, Anne 72, 1967, no 2, 154-169, especially 159 ff.
98
Fichte, Wissenschaftslehre nova methodo, pp. 350, 358 passim.
99
Fichte, Werke, vol. I, p. 530, 2. Fichte in general does not distinguish between
egoity and consciousness for our self, or that which represents, or our consciousness are all idem. Our self is nothing other than consciousness itself (Fichte,
Wissenschaftslehre nova methodo, p. 350). The dangers of this dual identification
have been noticed early on (e.g. by Novalis and Schleiermacher) and have stimulated attempts at a non-egological explication of self-consciousness (for example in the work of the early Husserl and of Sartre as well as of Henrich and his
followers.
100
Leibniz, cf. la conscience o la connaissance rflexive [= 4 of Principes de la
101
For example, Fichte, Werke, vol. I, p. 528.
102
Fichte (e.g. ibid., p. 525).
103
A concept is clear when I can distinguish it from others, affirm or deny it cor-
Nature et de la Grce].
rectly, re-identify it in different contexts and over time, and so on. A concept is
distinct [deutlich] when I can give an account of all its individual constitutive features.
[cf. Descartes, Principia philosophiae I, 45/6; Leibniz, Die Philosophischen Schriften,
ed. C.I. Gerhardt, Berlin, 1875-80, reprinted Hildesheim 1983, vol. IV, 422.]
104
Fichte, Wissenschaftslehre nova methodo, pp. 357/8.
105
Ibid., p. 368.
106
Ibid., p. 359.
107
Fichte, Werke, vol. I, p. 151.
108
Fichte, Wissenschaftslehre nova methodo, p. 361.
109
Ibid., pp. 359/60; cf. the conclusion to the whole of paragraph 1, pp. 360-7.
110
Ibid., p. 358.
111
Fichte, Werke, vol. I, p. 111/tr. Science of Knowledge, eds. & trans. Peter Heath and
John Lachs, New York, Appleton-Century-Crofts, 1970, p. 110.
112
Schelling says: non-self-sameness (Nicht-Einerleiheit), Die Weltalter. Fragmente. In der

Urfassung von 1811 und 1813, ed. Manfred Schrter, Munich, 1946, pp. 26/7, 128/9.
113
Ibid., p. 27.
114
By hypostatising an immediate unity that could not be explicated in terms of

reflexive relations the Early Romantics were able to link up with Jacobi. We owe
128 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 129
to Dieter Henrich a first systematic reconstruction, using the original sources, of

Jacobis arguments that were of such relevance to Early Romanticism. Cf. his Die
Anfnge der Theorie des Subjekts (1789), in Zwischenbetrachtungen. Im Proze der
Aufklrung, eds. Axel Honneth et al., pp. 106-170, esp. 123 ff. and 159 ff.
115
Hegels letter to Schelling, end-January, in eds. Manfred Frank and Gerhard Kurz,
Materialien zu Schellings philosophischen nfangen, Frankfurt am Main, Suhrkamp,
1975, p. 122.
116
Hlderlin Letter to Hegel, 26.1.1795, ibid., p. 124.
117
Friedrich Hlderlin, Smtliche Werke. Kritische Textausgabe, ed. D.E. Sattler, Darmstadt
and Neuwied, Luchterhand, 1984, vol. 10, p. 33.
118
Hlderlin, Fragment of Hyperion (or draught), in Smtliche Werke, 86 f., Lines 127130.
119
Ibid., Lines 131-154.
120
Schleiermacher, Smmtliche Werke, III. Abtheilung, Vierten Bandes zweiter Theil,

Berlin, 1839, esp. 422 ff.
121
Vom ersten Ringen dunkler Krfte/ Bis zum Ergu der ersten Lebenssfte,/ Wo
Kraft in Kraft, und Stoff in Stoff verquillt,/ Die erste Blt, die erste Knospe
schwillt,/ Zum ersten Strahl von neu gebornem Licht,/ Das durch die Nacht wie
zweite Schpfung bricht/ Und aus den tausend Augen der Welt/ Den Himmel
so Tag wie Nacht erhellt.[sic!]/ Hinauf zu des Gedankens Jugendkraft,/ Wodurch
Natur verjngt sich wieder schafft,/ Ist Eine Kraft, Ein Pulsschlag nur, Ein Leben,/
Ein Wechselspiel von Hemmen und von Streben (From the first wrestling of
dark forces/ To the flowing forth of the first vital fluids,/ Where force with force
and matter with matter are joined,/ Where the first blossom and the first bud are
bursting/ In the first ray of newborn light/ That pierces the night like a second
creation/ And which, through the thousand eyes of the world,/ Illuminates heaven
as well as day and night. [sic!]/ Upwards to the youthful force of thought,/
Whereby nature, rejuvenated, recreates itself/ There is a force, a single heartbeat, a life/ a reciprocity of hindrance and of striving). Schelling, Epikurisch
Glaubensbekenntnis Heinz Widerporstens, in eds. M. Frank and G. Kurz, Materialien zu Schellings philosophischen Anfngen, Frankfurt am Main, Suhrkamp, 1975,
p. 151, [6].
122
Schelling, Smmtliche Werke, F.K.A. Schelling, Stuttgart, Cotta, 1856-61, 1/7, p. 408;
123
Friedrich Schiller, Smtliche Werke, eds. Gerhard Fricke and Herbert G. Gpfert,
see 174.
(dtv-Gesamtausgabe), Munich, 1966, vol. 18, p. 46, 2.
124
Hlderlin, Der Abschied, (Ach! wir kennen uns wenig,/ Denn es waltet ein Gott
in uns) (Ah! We know each other so little,/ for a God reigns within us).
125
Die Liebe allein (. . .) ist das absolut Groe selbst, was in der Anmut und Schnheit
CH 5_f4_53-136 11/23/04 9:28 AM Page 130
sich nachgeahmt und in der Sittlichkeit sich befriedigt findet, es ist der Gesetzgeber
selbst, der Gott in uns, der mit seinem eigenen Bilde in der Sinnenwelt spielt
(Schiller, Smtliche Werke, vol. 18, 49 f.). (Love alone is what we call the absolutely
magnificent, that which imitates each other in grace and beauty and finds itself
satisfied in ethical life; it is the law giver himself, the God within us, who toys
with his own image in the sensory world).
126
Schelling, Smmtliche Werke, I/6, p. 501, para. 270.
127
I have discussed in more detail the relation between virtual and actual identity/difference in my Die Grenzen der Verstndigung. Ein Geistergesprch zwischen
Lyotard und Habermas, Frankfurt am Main, Suhrkamp, 1988, 85 ff.
128
This short manuscript was eventually published as Urtheil und Seyn in Friedrich
Hlderlin, Smtliche Werke. Groe Stuttgarter Ausgabe, ed. Friedrich Beiner, Stuttgart,
1961, vol. 4, p. 216 f. Cf. also Dieter Henrich, Der Grund im Bewutsein. Untersuchungen zu Hlderlins Denken (1794-1795), Stuttgart, Klett-Cotta, 1992.
129
The incorrect etymology of Ur-teil can also be found in Sinclair and Hegel. In
his popular lecture series on logic and metaphysics (according to Platner s
Philosophische Aphorismen) which Fichte, prompted by student request and with Hlderlin
and Sinclair among his students, gave for the first time in the winter semester of 1794/5,
Fichte emphasised not just the synthetic character of a judgement but also its differentiating nature: To judge means: to posit a relation between different concepts. [. . .] This
relation becomes obvious through opposition (Nachgelassene Schriften, ed. by Hans
Jacob, Berlin, 1937, vol. II, 469, p. 126). Ibid., 508, p. 129: In the act of making a judgment concepts are set alongside each other. In the originary act to which
this relates they may have been set alongside each other or separated from each
other. The division of particular things into general concepts Fichte calls the
fundamentum divisionis ( 462, p. 124). Violetta Waibel, in her Masters Thesis on
traces of Fichte in the development of Hlderlins writings (Munich, 1986, 54),
has discovered a direct precursor of Hlderlins etymology: Urtheilen, ursprnglich
theilen; (. . .) es liegt ein ursprngliches Theilen ihm zum Grunde (Judging,
originary separation; [. . .] it is grounded in an originary act of separating) Fichte,
Nachgelassene Schriften zu Platners Aphorismen 1794-1812, eds. Reinhard
Lauth and Hans Gliwitzky, in Gesamtausgabe der Bayerischen Akademie der
Wissenschaften, vol. II, 4, p. 182.
130
Hlderlin, Groe Struttgarter Ausgabe, ed. Friedrich Beissner, vol. 1, p. 217.
131
Cf. Hector-Neri Castaeda, He. A Study in the Logic of Self-Consciousness.

Ratio, vol. 8, 1966, pp. 130-157.
132
133
Hlderlin, Groe Struttgarter Ausgabe, vol. IV, pp. 253/4.

For a more detailed discussion see my Eine Einfhrung in Schellings Philosophie,
61 ff.
130 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 131
134
Hlderlin, Letter to his Brother, mid-1801, Groe Struttgarter Ausgabe, vol. V,

p. 419.
135
Karl Christian Erhard Schmid, Letter to Brhl 29.6.1794 in Jamme, 136, p. 9.
136
Cf. Christoph Jamme, Isaak von Sinclair. Politiker, Philosoph und Dichter zwischen
Revolution und Restauration, Bonn 1988, 9 ff., 48 ff.; idem, Isaak von Sinclairs Philosophische Raisonnements. Zur Wiederfindung ihrer Originale, in Hegel-Studien,
18, 1983, pp. 240-244. The Raisonnements have been published as an appendix
to Hannelore Hegels doctoral thesis titled: Isaak von Sinclair zwischen Fichte,
Hlderlin und Hegel. Ein Beitrag zur Entstehungsgeschichte der idealistischen Philosophie,
Frankfurt am Main, 1971, pp. 289-291 (page numbers in brackets above refer to
this publication). Zwillings contribution to the discussions at Jena and Bad
Homburg has been investigated most thoroughly by Dieter Henrich and Christoph
Jamme in their book Jakob Zwillings Nachla. Eine Rekonstruktion, in Hegel-Studien,
Beiheft, 28, Bonn 1986, pp. 9-99.
137
Isaak von Sinclair, Philosophische Raisonnements, in Hannelore Hegel, Isaak

von Sinclair zwischen Fichte, Hlderlin und Hegel. Ein Beitrag zur Entstehungsgeschichte
der idealistischen Philosophie, (doctoral thesis), Frankfurt am Main, 1971, p. 268.
138
Ibid., p. 271.
139
Ibid., p. 268.
140
Ibid., p. 269.
141
Ibid., p. 269.
142
Ibid., pp. 268-9.
143
Ibid., for example, p. 271.
144
Ibid., p. 282.
145
Ibid., pp. 269, 272.
146
Ibid., p. 270/10.
147
Ibid., p. 273.
148
Ibid., p. 273.
149
According to what is currently known, the only remaining item of Jakob Zwillings
posthumous philosophical works is a fragmentary draft, some three pages in
length, entitled ber das Alles (in eds. Henrich/Jamme, Jakob Zwillings Nachla
in Rekonstruktion, pp. 63-65). Unlike in the case of Sinclair and Hlderlin, Zwilling
seems to have believed that the defect of separation could be cured by reflection
since [according to the law of cognition-through-counterposition] a relation contains within itself a non-relation. Thus what is related must be counter-posited
as non-related, or else we must posit as absolute the relation between a statement
and its contradiction (ibid., 64 f.). This thought, however rudimentary and only
hinted at by Zwilling, leads us, in the end, to Hegels positing, as absolute, the
self-referential negation (or reflection). The category of relation us such as an
CH 5_f4_53-136 11/23/04 9:28 AM Page 132
autonomous posit is claimed by Zwilling as the truly Infinite, indeed, as Infinity

itself (65). The opposition between unity in terms of content and separation
in terms of form that is characteristic of Hlderlin and Sinclair can also be found
throughout Zwillings fragment.
150
151
Novalis, Schriften, eds. P. Kluckhohn and R. Samuel, 2nd ed., Stuttgart, 1960 ff.
vol. II, p. 104, no. 1.
152
Ibid.
153
The proviso contained in the word seems here does not pertain to all terms of
the relation, but only to the reflection which is nevertheless required, in order to
gain epistemic access to what there is as such, prior to all reflection, ibid., p. 112,
No. 14.
154
We might be justified in attributing Novalis choice of the term feeling to his

fondness of Jacobi, a choice that will come to full fruition in Schleiermachers theory of Gefhl. In his Jacobi und die Philosophie seiner Zeit. Ein Versuch, das wissenschaftliche Fundament der Philosophie historisch zu errtern, Mainz, 1834. J. Kuhn,
a disciple of Jacobi, has given us an excellent exposition and elaboration of Jacobis
pathbreaking but conceptually little developed theory of non-objectifying selfconsciousness or feeling. Kuhn saw very clearly the infinite regress that emerges
if one wants to support definitively the fact of Self-observation through an
already reflexive consciousness (p. 19). For the derived or reflected consciousness presupposes an irreflexive, primitive or originary consciousness in which
no separation between representing and represented content can be found in the
way this is possible in derived (i.e. higher-level) consciousness (35 f.) Since every
explanation moves within the realm of reflection, the originary consciousness
just like imaginationremains inexplicable (38 f., footnote). On pages 409 ff.
Kuhn analyses how both unmediated feeling and mediated knowledge-of-oneself-as-object relate to each other; in other words, he investigates the structural
unity of consciousness as such. He seeks to interpret the unmediated consciousness as a merely potential difference between the ideal and the real, a difference
that does not contradict the actual difference that we find in mediated self-knowledge (411 f., 514). I am grateful to Alexander Weber for alerting me to the work
of J. Kuhn.
155
This term is again taken from Jacobi by Novalis.
156
Novalis, Schriften, Die Werke Friedrich von Hardenbergs, eds. Paul Kluckhohn
157
Ibid., pp. 115 and 117, No. 19.
158
Ibid.
159
Ibid., p. 114, line 7.
and Richard Samuel, Stuttgart, Kohlhammer, 1960 ff., vol. 2, 106, No. 3.
132 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 133
160
Ibid., p. 114, lines 3/4.
161
Ibid., p. 114, lines 3/4.
162
Cf. M. Frank, Einfhrung in die frhromantische sthetik, Vorlesungen, Frankfurt

am Main, Suhrkamp, 1989. Novalis theory is presented in detail in lectures 15
and 16.
163
Novalis, Schriften, p. 107, No. 3.
164
Cf. Novalis, Schriften, p. 259, nos. 508; p. 528, no. 21.
165
Johann Friedrich Herbart, Psychologie als Wissenschaft, neu gegrndet aus Erfahrung,
Metaphysik und Mathematik, Erster, synthetischer Theil, Knigsberg, Unzer, 1824,
mainly pp. 93-112; also in J.F. Herbart: Smtliche Werke, ed. by G. Hartenstein, Vol.
5, Schriften zur Psychologie, Erster Teil, Leipzig, 1850 ff. (All my citations refer to
the latter edition).
166
Herbart, Smtliche Werke, pp. 27, 275.
167
Ibid., p. 274.
168
Ibid., p. 275.
169
Ibid., pp. 275/6.
170
Ibid., 29, p. 284.
171
Ibid., p. 275.
172
Ibid., 28, p. 281.
173
Ibid., pp. 284/5.
174
Ibid., p. 285.
175
Ibid., p. 286.
176
Ibid., 28.
177
Ibid., p. 281.
178
Fichte, Wissenschaftslehre nova methodo, p. 362.
179
Herbart, Smtliche Werke.
180
Ibid., p. 282.
181
Ibid., p. 283, Middle. The objects to which Herbart here refers are, as he warns
not real [reale] objects, but merely represented ones, as such [sondern bloe Vorgestellte,
als solche], 29, p. 284.
182
Ibid.
183
Ibid.
184
Ibid.
185
Ibid., p. 283.
186
Ibid., 27, p. 278.
187
Ibid., 29, p. 284.
188
Ibid., p. 278.
189
Ibid., 29, p. 284.
190
Ibid., 28, p. 281.

CH 5_f4_53-136 11/23/04 9:28 AM Page 134
191
Ibid., 28, p. 284.
192
Ibid., 29, p. 284.
193
Ibid.
194
Ibid., pp. 284/5.
195
Ibid., p. 285.
196
Ibid.
197
Ibid.
198
Ibid.
199
Ibid., pp. 285/6.
200
Ibid., Section II, part. II, chapter 5, 199. Franz Brentano quotes this passage in
order to exemplify the absurdities one encounters if one believes that consciousness is originarily unconscious, and therefore maintains that our knowledge of
it is based on a later act of reflection [Reflexions-Akt] (Psychologie vom empirischen
Standpunkt, Erster Band, 2. Buch, Kap. II [About inner consciousness], p. 7
[= p. 175]). This absurd theory which we have already encountered with John
Sergeant, in a very caricatured form, is nowadays promoted by Niklas Luhmann.
Like John Sergeantand indeed already Thomas von Aquin (Summ. teol. P. I.,
Q. 78, A.4, ad 2; Q. 87. A.3 and ad 3; Q 87, A.3,2 and ad 2.)he interprets the act
of making conscious the (originally unconscious) consciousness as an observation: A consciousness relates to another (previous) mode of observation and does
not only make it conscious, but also encapsulates it as an object of a self-reference [Selbstreferenz] (Niklas Luhmann, Die Autopoesis des Bewutseins in eds.
A. Hahn and V. Knapp, Selbstthematisierung und Selbstzeugnis: Bekenntnis und
Gestndnis, Frankfurt am Main, Suhrkamp, 1987, pp. 25-94). The obvious circularities of such an explanation have been brought to light and criticised by Vronique
Zanetti, Kann man ohne Krper denken? ber das Verhltnis von Leib und
Bewutsein bei Luhmann und Kant, in eds. Hans-Ulbrich Gumbrecht und Ludwig
Pfeiffer, Materialitten der Kommunikation, Frankfurt am Main, Suhrkamp, 1988,
pp. 280-294.
201
Schleiermacher, Glaubenslehre, 1st edition 1821, 2nd edition 1830.
202
Der christliche Glaube, nach den Grundstzen der evangelischen Kirche im

Zusammenhange dargestellt. 7. Auflage. Aufgrund der 2. Auflage und kritischer
Prfung des Textes neu herausgegeben von Martin Redeker. II Bnde. Berlin, 1960
[cited Glaubenslehre], I. 16 bottom.
203
Cf. Dieter Henrich, Hlderlin ber Urteil und Sein. Eine Studie zur Entstehungsgeschichte des Idealismus, in Hlderlin-Jahrbuch, 14, 1965/66, pp. 73-96; Hannelore
Hegel, Isaak von Sinclair zwischen Fichte, Hlderlin und Hegel. Ein Beitrag zur
Entstehungsgeschichte der idealistischen Philosophie, Frankfurt am Main, 1971 (Sinclairs
Philosophische Raisonnements, pp. 243-283); Dieter Henrich and Christoph
134 Manfred Frank
CH 5_f4_53-136 11/23/04 9:28 AM Page 135
Jamme, eds., Jakob Zwillings Nachla. Eine Rekonstruktion, in Hegel Studien, Beiheft,
28, Bonn 1986, pp. 9-99, and Isaak von Sinclair. Politiker und Dichter zwischen Revolution
und Restauration, Anhand von Originaldokumenten dargestellt von Christoph
Jamme, Bonn, 1988, esp. 23 ff., 48 ff.
204
Schleiermacher, Glaubenslehre, I, pp. 28/29.
205
This is not a fundamental theorem of Schleiermachers thought that can be found

everywhere in his work.
206
Cf. F. Schleiermacher, Dialektik, ed. by Rudolf Odebrecht, Leipzig, Hinrichs, 1942,

pp. 265-267 (XLVI) and pp. 283/4 (L). I quote from this edition which is based
on rewritten lecture notes of the Dialektikkollegs from 1822 by referring to Dial 0.
I use the abbreviation Dial J when citing from Schleiermachers handwritten
Nachla of all Dialektikkollegs (Dialektik, III/4,2, ed. L. Jonas, Berlin, Reimer,
1839).
207
Schleiermacher, Dial 0, 284 (= L): When we refer to the degree of failure, then
we have, however, not failed to experience the transcendent ground, indeed we
have achieved this through both functions. However, we have failed to bring this
transcendent ground to a unity of real consciousness. We gain it, however, only
insofar as we recognize the inadequacy of the onesided and divided forms.
208
Ibid., 286 ff., Glaubenslehre, 14 ff. The two terms do not mean exactly the same,
although Schleiermacher does not always emphasise their difference. I will neglect
the difference in this context.
209
Ibid., p. 286.
210
Ibid., p. 287.
211
I am here thinking especially of Dieter Henrichs already cited work Fichtes

ursprngliche Einsicht. Nearly all the important comments about the problematic
of self-consciousness in Fichte are present in this little book.
212
Schleiermacher, Glaubenslehre, 3, p. 2, with a hand-written note.
213
Schleiermacher, Dial 0, p. 288, Glaubenslehre, 3, p. 2, hand-written note.
214
Schleiermacher, Glaubenslehre, I, 8, p. 52.
215
Schleiermacher, Dial 0, p. 288.
216
Ibid., p. 290.
217
Ibid., p. 286.
218
Ibid.
219
Schleiermacher, Dial 0, p. 290, = Dial J, p. 420 = SLI.
220
Schleiermacher, Dial 0, p. 287; cf. also pp. 290 and 295/6.
221
Schleiermacher, Glaubenslehre, I, p. 24.
222
Ibid., p. 27.
223
Kierkegaard, The Sickness unto Death (published 1849). I am quoting from the dtvedition, Munich, 1976 (hence from the translation by W. Rest) cited here under
CH 5_f4_53-136 11/23/04 9:28 AM Page 136
the title Die Krankheit zum Tode. My interpretation adopts gratefully the consequence which Gnter Figal was first to point out, Die Freiheit der Verzweiflung und
die Freiheit des Glaubens. Zu Kierkegaards Konzeption des Selbstseins in der Krankheit
zum Tode in Kierkegaardiana XIII, Kbenhavn, 1984, 11-23. Obviously I emphasise, in contrast to Figal, the structural homology of Kierkegaards account of the
independent self with that of the early romantics.
224
Sren Kierkegaard, Die Krankheit zum Tode, p. 31; The Sickness unto Death, p. 13.
For a complete reference: This translation has been taken from: The Sickness unto
Death, a Christian Psychological Exposition for Upbuilding and Awakening, by Sren
Kierkegaard, eds. and trans. Howard V. Hong and Edna H. Hong, Princeton, 1980.
225
Sren Kierkegaard, ibid.
226
Sren Kierkegaard, Die Krankheit zum Tode, 31 f., The Sickness Unto Death, pp. 13/14.
227
Kierkegaard, Die Krankheit zum Tode, p. 69.
228
Kierkegaard, Die Krankheit zum Tode, p. 32.
229
Friedrich Schlegel, (critical edition of his works, eds. Ernst Behler, et al., MunichPaderborn-Vienna, Schningh, 1958 ff., vol. XVIII, m 418, No. 1168, already XII,
7/8.
230
Ibid., vol. XII, p. 7.
231
Novalis, Schriften, II, 451 f., 454 f.
136 Manfred Frank

Frank-Manfred Self-Conssciousness PDF

Uploaded by

Copyright:

Available Formats

Frank-Manfred Self-Conssciousness PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Frank-Manfred Self-Conssciousness PDF

Uploaded by

Copyright:

Available Formats

CH 5_f4_53-136 11/23/04 9:28 AM Page 53

CH 5_f4_53-136 11/23/04 9:28 AM Page 54

KEYWORDS: Self-consciousness, Enlightenment, Romanticism, Kant, Fichte, Hlderlin, Sinclair,

CH 5_f4_53-136 11/23/04 9:28 AM Page 55

CH 5_f4_53-136 11/23/04 9:28 AM Page 56

CH 5_f4_53-136 11/23/04 9:28 AM Page 57

the transcendental deductionwhich was itself never questioned by the

CH 5_f4_53-136 11/23/04 9:28 AM Page 58

it is that it is not: what it is or how it appears to itself.12 All I grasp of

CH 5_f4_53-136 11/23/04 9:28 AM Page 59

CH 5_f4_53-136 11/23/04 9:28 AM Page 60

Since the thought I is in no sense a concept, but only inner perception,

In order to grasp the peculiar aporia of Kants basic argument, it is worth

CH 5_f4_53-136 11/23/04 9:28 AM Page 61

ment, a content of intuition is linked with a concept (a is B); bothin Kants

CH 5_f4_53-136 11/23/04 9:28 AM Page 62

cannot avoid bringing into play a possibility that he passionately rejects

CH 5_f4_53-136 11/23/04 9:28 AM Page 63

CH 5_f4_53-136 11/23/04 9:28 AM Page 64

CH 5_f4_53-136 11/23/04 9:28 AM Page 65

Similarlyalthough now in a more problematised formulationthis

CH 5_f4_53-136 11/23/04 9:28 AM Page 66

The consciousness of this fundamental law may be called a fact of reason,

Here it would be a matter of a self-determination of the I of such a kind

CH 5_f4_53-136 11/23/04 9:28 AM Page 67

CH 5_f4_53-136 11/23/04 9:28 AM Page 68

Every representation requires someone who represents, and something that

CH 5_f4_53-136 11/23/04 9:28 AM Page 69

CH 5_f4_53-136 11/23/04 9:28 AM Page 70

of unconsciousness, or I am not perceiving, in which case unconsciousness

CH 5_f4_53-136 11/23/04 9:28 AM Page 71

ular Hume failed to realise that self-consciousness and ego-consciousness are

CH 5_f4_53-136 11/23/04 9:28 AM Page 72

CH 5_f4_53-136 11/23/04 9:28 AM Page 73

sciousness. The illusion of the immediate acquaintance of consciousness with

CH 5_f4_53-136 11/23/04 9:28 AM Page 74

(for example, that carried out by Descartes).79 Nevertheless, Pseudo-Mayne

Here self-consciousness is clearly explained as the subsequent reflection of a

CH 5_f4_53-136 11/23/04 9:28 AM Page 75

As a result of this recourse to the model of representation, Pseudo-Mayne

CH 5_f4_53-136 11/23/04 9:28 AM Page 76

a conscious being. Statements such as I remember that yesterday I dreamt

CH 5_f4_53-136 11/23/04 9:28 AM Page 77

CH 5_f4_53-136 11/23/04 9:28 AM Page 78

is an intuition. But what is intuited here, is not a spatio-temporal entity, like

CH 5_f4_53-136 11/23/04 9:28 AM Page 79

CH 5_f4_53-136 11/23/04 9:28 AM Page 80

CH 5_f4_53-136 11/23/04 9:28 AM Page 81

CH 5_f4_53-136 11/23/04 9:28 AM Page 82

CH 5_f4_53-136 11/23/04 9:28 AM Page 83

CH 5_f4_53-136 11/23/04 9:28 AM Page 84

CH 5_f4_53-136 11/23/04 9:28 AM Page 85

On the 19th of January 1795, Hlderlin had written to Neuffer: I am now

CH 5_f4_53-136 11/23/04 9:28 AM Page 86

faculty of cognition. A little later he remarks that an I cannot, in any case,

CH 5_f4_53-136 11/23/04 9:28 AM Page 87

and we exchanged our state of consciousness,

At this point Hlderlin comes up with a conception of essential unity as a

CH 5_f4_53-136 11/23/04 9:28 AM Page 88