Kant's Transcendental Ideal as Fiction 
Peter Klepec 
Le fictif, en effet, n'est pas par essence ce qui est trompeur, 
mais, à proprement parler, ce que nous appelons le 
symbolique. 
Jacques Lacan 
What can be said, from the point of view of Kant's Critique of Pure Reason, about Kant's relation to fiction? At first glance, the answer is 
quite ambiguous; if the Transcendental Analytic is taken as our starting point, 
it seems that Kant is one of the greatest opponents of fiction, but if we, 
contrariwise, accept the point of view of Transcendental Dialectic it seems that 
the answer should be precisely the opposite. Do we need to choose between 
the two apparently irreconcilable answers or is there something wrong with 
this option as such? In our opinion, the latter is true. One of the differences 
between Transcendental Analytic and Transcendental Dialectic is, that they 
are based on a different theory of truth: whereas Analytic is based on classical 
theory of truth which can be shortly summed up as adequatio rei et intellectu, 
the Dialectic does not concern the truth conceived as adequation of concepts to 
objects. In Dialectics the relation to objects has no role at all, because main 
relation is the relation of concepts to concepts. In other words, the basic 
premise of Analytic is that concepts, which are not related to sensible intuition, 
are »without sense, that is, without meaning« (B 299)1, for, as famous Kant's 
dictum goes, »thoughts without content are empty, intuitions without concepts 
are blind.« (B 75) Strawson in his Bounds of Sense summarized this thesis with 
the so-called principle of significance. »This is the principle that there can be 
no legitimate, or even meaningful, employment of ideas or concepts which 
does not relate them to empirical or experiential conditions of their applica-
tion. If we wish to use a concept in a certain way, but are unable to specify the 
kind of experience-situation to which the concept, used in that way, would 
apply, then we are not really envisaging any legitimate use of that concept at 
all.«2 In short, if the concept transcends the bounds of experience the illusion 
arises. Can we, from this point of view, give a straight and plain answer 
1 Kant's works are cited from: Immanuel Kant, Werkausgabe in zwölf Bänden, ed. Wilhelm 
Weischedel, Frankfurt am Main 1989. In the parentheses, the second edition of the Critique of 
Pure Reason is cited. All translations of Kant are mine. 
2 P. F. Strawson: The Bounds of Sense, Routledge, London 1975, p. 16. 
Fil. vest. /Acta Phil., XV (2/1994), 97-126. 
98 Peter Klepec 
concerning Kant's conception of fiction? In other words, is for Kant the term 
fiction equivalent to illusion? Our answer is negative. It is true that Kant uses 
numerous expressions for illusion (besides Illusion and Schein, illusion and 
semblance, there are also expressions like Wahn, Blendwerk, Hirngespinst, 
Täuschung, Geschöpf, Betrug, Chimäre, Erdichtung, Einbildung etc.), but he 
never identifies, in the very strict sense of a word, illusion with fiction. Even 
more. He obviously differentiates between both terms since he reserves the 
term fiction for a very special concept, i.e., the idea. The only Kant's utiliza-
tion of the term fiction in the Critique of Pure Reason links fiction with ideas, 
ideas, which are for Kant nothing but heuristic fictions, heuristische Fiktionen. 
(B 799) 
Consideration of ideas, as is well known, belongs to the Transcendental 
Dialectic, which, incidentally, comprises almost a half of Critique of Pure 
Reason. But this half could be, from the point of view of Strawsonian principle 
of significance, easily cut off, since Dialectics contains, as it seems, nothing 
valuable for Strawson. This is in fact Strawson's authentic conviction, for he 
claims: »After construction, demolition; after the Transcendental Analytic, the 
Transcendental Dialectic. (...) The primary aim of the Dialectic is the exposure 
of metaphysical illusion; the primary instrument of exposure is the principle of 
significance.«3 
Although it is accompanied by »natural, but inevitable illusion« (B 354), 
Transcendental Dialectic is, at least in our opinion, anything but pure exposure 
of error. One cannot just cut off the ideas and the dialectic of pure reason, as 
Strawson recommends. The problem is that »it is possible to tell reality from 
fictions (...) the legitimate use of transcendental categories in the constitution 
of reality from their illegitimate use which brings about 'transcendental illu-
sion'; however, as soon as we renounce fiction and illusion, we lose reality 
itself; the moment we subtract fictions from reality, reality itself loses its 
discursive-logical consistency. Kant's name for these fictions, of course, is 
»transcendental Ideas«, whose status is merely regulative and not constitutive: 
Ideas do not simply add themselves to reality, they literally supplement it; our 
knowledge of objective reality can be made consistent and meaningful only by 
way of reference to Ideas. In short, Ideas are indispensable to the effective 
functioning of our reason.«4 
To see more concretely what it means that for Kant ideas are something 
indispensable and that ideas are heuristic fictions, we will turn in this paper to 
the third chapter of Transcendental Dialectic, the chapter entitled Ideal of Pure 
'Ibid., p. 33. 
4 Slavoj Žižek, Tarrying with the Negative, Duke University Press, Durham 1993, p. 88-89. 
Kant's Transcendental Ideal as Fiction 99 
Reason, which is divided into seven sections. These can be, for the conve-
nience of interpretation, separated further into three main parts: the first part 
deals with Kant's conception of ideal in general and the transcendental ideal, 
comprising the first and the second section, the second includes sections from 
three to six and it contains Kant's critique of traditional proofs for the existence 
of God, the third part is identical with the seventh section and it concerns some 
consequences of Kant's dealing with the idea of the supreme being for the 
rational and natural theology. 
The only »part« which interests us here is the first one. It contains Kant's 
treatment of transcendental ideal5 and it also represents the clue for the second 
and the third part. The reason for that being the fact that the first part deals with 
the idea of omnitudo realitatis, whereas the other two parts deal with the link 
of this idea with the idea of absolutely necessary being. The reason for the link 
of both ideas lies in the fact that there is, of course, no difficulty in giving a 
verbal definition of absolutely necessary being, »namely, that it is something 
the non-existence of which is impossible. But this yields no insight into the 
conditions which make it necessary to regard the non existence of a thing as 
utterly unthinkable. It is precisely these conditions that we desire to know, in 
order that we may determine whether or not, in resorting to this concept, we 
are thinking anything at all.« (B 621) These conditions are another expression 
for the idea of omnitudo realitatis and Kant has convincingly shown that all 
proofs for the existence of God necessarily contain the following step: »The 
necessary being can be determined in one way only, that is, by one out of each 
possible pair of opposed predicates. It must therefore be thoroughly deter-
mined through its own concept. Now there is only one possible concept which 
thoroughly determines a priori, namely, the concept of ens realissimum. The 
concept of the most real being is therefore the only one through which a 
necessary being can be thought.« (B 633-634) In other words, all speculative 
proofs for the existence of the supreme being are based for Kant only on 
reciprocity of two concepts or ideas: the idea of the most real being and the 
idea of absolutely necessary being. (See B 816-817) But this reciprocity, we 
would like to add, is not mutual - although the concept of the most real being is 
the only concept through which a necessary being could be, but in fact cannot 
be thought, Kant tries to show that we can think the idea of omnitudo realitatis 
without linking it with ens necessarium. Namely, Kant's basic task is to 
develop the concept of transcendental ideal and to separate it from idea of 
5 This section is closely related to Kant's pre-critical work The Only Possible Basis for a 
Demonstration of the Existence of God (1763). For basic parallels and differences between the 
two works see: Dieter Henrich: Der ontologische Gottesbeweis, Sein Problem und seine 
Geschichte in der Neuzeit, J. C. B. Mohr (Paul Siebeck), Tübingen 19672, p. 137 ff. 
100 Peter Klepec 
absolutely necessary being to prevent every, even the slightest possible link 
between these two concepts. Furthermore, not only that there is no necessary 
link between the above mentioned concepts, there is also no necessary link, at 
least in our opinion, between transcendental ideal and theological idea of God. 
Although we cannot think the idea of God without the transcendental ideal, we 
can perfectly think, as we will at least try to show in our paper, the transcen-
dental ideal without theological consequences. It is true that even the transcen-
dental ideal itself originates some sort of illusion, but Kant emphasizes that 
this illusion is, although natural, not inevitable. This proposition of Kant is in 
apparent opposition to the usual description of illusion in Critique of Pure 
Reason, but this is due to the fact that for Kant, transcendental ideal is »based 
on a natural, not on a merely arbitrary idea«. (B 609) 
But what is in fact Kant's conception of the ideal? Kant begins the section 
entitled About the Ideal in general with the distinction between concepts of 
understanding and concepts of reason, i.e. ideas. Basic difference between the 
first and the second concepts is that, to the latter, no corresponding object can 
ever be given in sense-experience, which is the primary reason why there can 
be no transcendental deduction of the ideas. But in spite of the fact that no 
object adequate to any transcendental idea can ever be found within experi-
ence, ideas are not arbitrarily invented. They are imposed by the very nature of 
reason itself. Reason needs ideas because they contain a certain completeness 
to which no possible empirical knowledge can ever attain. »In them reason 
aims at a systematic unity, to which it seeks to approximate the unity that is 
empirically possible, without ever completely reaching it.« (B 596) After this 
brief review of the term idea, Kant introduces6 the concept of the ideal: »By 
the ideal I understand the idea, not merely in concreto, but in individuo, that is, 
as an individual thing, determinable or even determined by the idea alone.« (B 
596) It means that the ideal seems to be even further removed from objective 
reality than ideas and that is one reason more for Kant to prevent possible 
misconceptions. In order to do that, he emphasizes that: 1. we cannot realize 
any ideal; 2. ideal as such does not possess objective reality, however, it is not 
to be identified with the illusion and with products of imagination; 3. ideal 
does not have creative, but merely a certain practical power. But why reason 
needs ideals at all? There are two needs of reason, the first is a practical one: as 
the idea gives the rule, the ideal »serves as the archetype for the thorough 
determination of the copy; and we have no other standard for our actions than 
the conduct of this divine man within us, with which we compare and judge 
ourselves, and so reform ourselves, although we can never attain to the 
6 The term ideal is, strictly speaking, used by Kant already before (see: B 398, 434-435), but it 
is here that Kant gives the definition of the term. 
Kant's Transcendental Ideal as Fiction 101 
perfection thereby prescribed.« (B 597) The second need is a speculative one: 
»Reason, in its ideal, aims, on the contrary, at thorough determination in 
accordance with a priori rules. Accordingly it thinks for itself an object which 
it regards as being thoroughly determinable in accordance with principles. The 
conditions that are required for such determination are not, however, to be 
found in experience, and the concept itself is therefore transcendent.« (B 599) 
The introduction of the ideal in general is followed by the introduction of the 
»The transcendental Ideal« or »prototypon transcendentale«, as the second 
section is formally entitled. The section begins with two principles which both 
consider possibility. The first one is called the principle of determinability, the 
second one the principle of thorough determination. The first principle is more 
or less self-evident and unproblematic. According to it »every concept is, in 
respect of what is not contained in it, undetermined, and is subject to the 
principle of determinability.« (B 599) Though the expression contain1 is very 
indefinite and at least disputable, it is quite clear that according to this prin-
ciple, which abstracts itself from the entire content of knowledge and is 
concerned merely with its logical form, of every two contradictorily opposed 
predicates only one can belong to a concept. In other words, either a or non-a 
can belong to a concept, tertium non datur. This principle is based on the 
principle of contradiction, and is therefore a purely logical principle. 
Although it would not seem so at first sight, the more problematic one is the 
second principle, the principle of thorough determination which Kant intro-
duces here for the first time. The place alone of this introduction - Kant 
introduces it in the middle of the transcendental Dialectic, in the middle of the 
logic of apparition, Schein - and the fact that in spite of the place of its 
introduction the principle is something what is for Kant indubitably true, 
causes to the interpretation of this section - as far as interpretation finds it 
worthy enough to involve with it8 - some troubles. Principle is the following: 
»Every thing, as regards its possibility, is likewise subject to the principle of 
thorough determination, according to which if all the possible predicates of 
things be taken together with their contradictory opposites, then one of each 
pair of contradictory opposites must belong to it.« (B 599-600) This principle 
is not based, as the principle of determinability, merely on the principle of 
'Kantian vague expression »contain« is undoubtedly problematical. For the sketch of the 
problem see: Henry Allison, Kant's Transcendental Idealism. An Interpretation and Defense, 
Yale University Press, New Haven & London 1983, p. 74. 
8 Bennett, for instance, says: »This is an unconvincing tale.« (Kant's Dialectic, Cambridge 
University Press, Cambridge 1974, p. 282), Strawson (op. cit.) devotes a page to it in his book, 
etc. There are only two useful interpretations: Peter Rohs, »Kants Prinzip der durchgängigen 
Bestimmung alles Seinden«, Kant-Studien 69/1978, p. 170-180; Svend Andersen, Ideal und 
Singularität, Kantstudien Ergänzungshefte 116, Walterde Gruyter, Berlin & New York 1983. 
102 Peter Klepec 
contradiction, where we can chose between two contradictory predicates, but 
involves something more. Every thing is in relation to something third, »to the 
sum-total [Inbegriff] of all possibilities., that is, to the sum-total of all predi-
cates of things«. (B 600) For that reason the principle of thorough determina-
tion needs »a synthesis of all predicates which are intended to constitute the 
complete concept of a thing, and not simply a principle of analytic representa-
tion in reference merely to one of two contradictory predicates.« (Ibid.) The 
principle of thorough determination concerns, therefore, the content, and »not 
merely the logical form«. (B 600) 
Now the difference between the two principles seems to be more or less clear: 
the first one involves both concepts and things, whereas the second one is 
reserved for things only. In other words, the first principle involves logical and 
the second one real possibility9. But - what is supposed to be the specificity of 
the second principle compared to the first one? The central point of the second 
principle is undoubtedly the claim that every thing presupposes the sum-total 
of all predicates. The relation of things and us to this sum-total is said to be that 
what assures to the second principle the character of syntheticity. But - what 
relation precisely has Kant in mind? It is said that this principle is »the 
principle of synthesis of all predicates which are intended to constitute the 
complete concept of a thing, and not simply a principle of analytic representa-
tion« (B 600) To put it differently, if we would have to deal with a complete 
concept, concept given as some sort of a Whole from which everything would 
then be derived, this would be nothing but a mere analytic representation. 
However, this sum-total, as Kant warns, can never be completed or even 
given, existent. The reason for that is that »the conditions that are required for 
such determination are not to be found in experience« (B 599). It means that 
we must add new predicates to the sum-total always anew and this process 
never ever ends. Maybe we should stop here for a moment since it seems that 
we have to deal with the same old Kantian song, which Hegel called bad 
infinity (schlechte Unendlichkeit). It seems that Kant regards the concept of 
sum-total as something somehow completed and perfect which stays forever 
somewhere beyond, or to use precise Kantian term - transcendent. The clue to 
this undoubtedly paradoxical concept of sum-total of all predicates brings up 
the following question: where are all this predicates which are supposed to 
determine the concept of sum-total taken from? Obviously from our experi-
ence. And it is experience or synthesis which assures the principle of thorough 
* In order to clarify the matter, Kant introduces another distinction: Whereas the determinability 
of every concept is subordinate to the universality (universalitas) of the principle of excluded 
middle, the determination of a thing is subordinate to the totality (universitas) or the sum of all 
possible predicates. (See: B 600) 
Kant's Transcendental Ideal as Fiction 103 
determination the relation to the sum-total of all predicates and the character 
of syntheticity. In other words, »the existence of this synthesis is identical to 
the existence of multitude of predicates«.10 
Although synthesis plays decisive role in the second principle, the difference 
between the two principles is not to be simply understood as the difference 
between the logical and the transcendental principle. It is true, of course, that 
the first principle concerns only the form and is therefore analytical, logical 
principle, and it is also true that the second principle involves something more, 
things, reality, content. It would be a mistake to think that the only reason for 
Kant's introduction of these two principles is to emphasize how separated are 
form and content, concept and reality. That there is a certain difference 
between them is as plain and self-evident as the claim that a thing is a thing, 
and a concept - a mere concept. But, is all that really so self-evident? It is true, 
of course, that a thing is a thing, a reality, and concept, fiction something 
different. But it nevertheless does not mean that there should be nothing but a 
sharp distinction between them. So, purely hypothetically, let us pose some 
questions. First, is it really so necessary that the two principles are simply 
separated? Is it really so clear what is the distinction between the two prin-
ciples based upon, in other words, if the decisive mark of the principle of 
thorough determination is the character of syntheticity, it is still not clear what 
is this sum-total of all predicates and how to think it? Furthermore, is there any 
possibility that one single concept would fulfill the conditions of the second 
principle? In other words, is a thoroughly determined concept possible or only 
things and empirical intuitions can be thoroughly determined? 
It seems that Kant has already given the answer for he introduces first prin-
ciple in the following manner: »every concept is, in respect of what is not 
contained in it, undetermined.« (B 599) A concept can - as Kant numerously 
emphasizes - never be thoroughly determined, since, as Kant very exactly 
posits above, concept as such is always undetermined. There always exists 
something what is not contained in it. Even more - not only that a concept is 
not, but for Kant also cannot be thoroughly determined - otherwise it would 
cease to be a concept and it would become an empirical intuition. It is the 
empirical intuition, and not the concept, which always relates to a single 
object, to a certain thing. A concept cannot relate to a singular object, for it is a 
general representation or a representation of what is common to several 
objects, in other words, a concept contains common marks of several objects. 
Their generality enables that they can be repeatedly used. From their general-
ity follows their irreality, that is, the more one concept is general, the lesser 
Peter Rohs, op. cit., p. 171-172. 
104 Peter Klepec 
number of marks or features it contains, for it leaves open - this is the 
consequence of its general validity - what are further features and attributes of 
these things. The more a concept is general, the wider is its sphere, the lesser 
amount of reality it contains. And for all these reasons, no concept can be 
thoroughly determined. But, we could ask, if this is true for the concepts of 
reason, concepts which literally demand their relatedness to objects, what is 
then true for concepts which are not and cannot be related to any object of 
experience, i.e. with ideas and ideals? Since it is the ideal which interests us 
here, let us recall Kant's definition of it: ideal is not idea »merely in concreto, 
but in individuo, that is, as an individual thing, determinable or even deter-
mined by the idea alone.« (B 596) From this definition follows that ideal is 
determinable, not with objects, but with an idea and it can even be determined 
by the idea. Again, this Kant's definition raises some questions. Does it mean 
that ideal as such can be thoroughly determined? And if the answer is negative, 
if the ideal as such, ideal in general, cannot be thoroughly determined - is there 
at least one ideal which can be? Does at least one mean only one? And, if this 
single thoroughly determined ideal is possible, which ideal is it and under what 
conditions this may be true? If such ideal is possible, is it in any relation to the 
concept of thorough determination and consequently with the sum-total of all 
possible predicates? Last but not least - does it mean that our two principles 
are not so strictly separated as it seemed at first glance? 
If we want an answer to some of these numerous questions, we will first have 
to specify what is Kant's conception of thorough determination. Let us start 
with Kant's proposition which represents some sort of example for the prin-
ciple of thorough determination: »Everything which exists is thoroughly deter-
mined.«. (B 601) This Kant's example is, certainly, anything but coincidentally 
chosen. As Dieter Henrich" points out, in Wolfs school existence is a supple-
ment of essence or inner possibility of things. And maybe this possibility was 
alluded to in Kant's claim that »every thing, as regards its possibility, is 
likewise subject to the principle of thorough determination ...« (B 599) How-
ever, there is a crucial difference between Kant and Wolf. For Kant, thorough 
determination is not supplement of essence but »the criterion of existence«.12 
A certain number of predicates belongs to everything what exists, to every 
existent thing. But it does not mean merely that only one of every two 
contradictorily opposed predicates can belong to a certain concept, it means 
that one of each pair of all contradictory opposites must belong to it. For 
example, our task is to determine a certain object. The object - and it does not 
need to be always an object or a sensual object at all - can either be red or not 
11 Henrich, op. cit., p. 155-156. 
12 Ibid.. 
Kant's Transcendental Ideal as Fiction 105 
red, either blue or not blue, either black or not black etc. It can also be either 
round or not round, either square or not square, etc. etc. In other words, we are 
not comparing the predicates only logically, but also transcendentally, with 
regard to the sum-total of all possible predicates. Kant's proposition that 
everything what exists is thoroughly determined in fact meant the following: 
in order to know one thing completely, volständig, we need to know all and 
everything that is possible and determine it either positively or negatively. 
»The thorough determination is thus a concept, which, in its totality, can never 
be exhibited in concreto. It is based upon an idea, which has its seat solely in 
the faculty of reason - the faculty which prescribes to the understanding the 
rule of its complete employment.« (B 601) It seems to be the proper reason, as 
Rohs emphasizes13, for Kant's placement of the principle of thorough determi-
nation in the Transcendental Analytic. This principle requires, as it seems, 
complete knowledge and not knowledge as such. But complete knowledge is 
for Kant something impossible since only things in themselves are thoroughly 
determined, whereas we, we as human beings, have nothing but appearances at 
our disposal. For that reason alone we can never attain to all predicates, yet we 
must always try. The insistence of reason on that task, although the conditions 
that are required for such determination are not to be found in experience, 
could be described as some sort of demand, perhaps - why not? - in its 
strongest, Lacanian sense of the term, demand which we can never fulfill. 
Thus, we are in an untenable position - although we cannot know all what is 
possible, reason demands that we must always try to attain it. This position 
could be described as some kind of ethical position, as far as we can talk about 
position at all, since each position is already always a non-position. And its 
fundamental claim: you cannot, but you must, is even more demanding and 
frantic than notorious claim of Kant's categorial imperative. 
If a certain impossibility is inherent in Kant's conception of thorough determi-
nation, is therefore a thoroughly determined concept possible at all? But - the 
thorough determination itself is already a concept, an ideal, which as every 
ideal presupposes a certain idea, in our case the idea of the sum-total of all 
possibilities. This idea has been, strictly speaking, undetermined until now. It 
was thought merely as the sum of all possible predicates and if we would 
succeed to determine it, we could, perhaps, have a thoroughly determined 
concept. Now - how do we proceed from this undetermined idea of the sum of 
all possible predicates to the thoroughly determined concept a priori? Under 
closer scrutiny, claims Kant, »we yet find, that this idea, as a primordial 
concept, excludes [ausstoße] a multitude [Menge] of predicates which as 
derivative are already given through other predicates or which are incompat-
13 Rohs, op. cit., p. 172. 
106 Peter Klepec 
ible with others; and that it does, indeed, define itself as a concept that is 
thoroughly determinate a priori. It thus becomes the concept of an individual 
object which is thoroughly determined through the mere idea, and must there-
fore be entitled an ideal of pure reason«. (B 601-602) 
This is the decisive point. »The idea of an omnitudo realitatis« (B 604) is not 
just an idea, but a very special idea. It is this idea that serves as some kind of 
basis for the transcendental ideal. This ideal is also not just an ideal, but a very 
special ideal. It »serves as basis for the thorough determination that necessar-
ily belong to all that exists. This ideal is the supreme and complete material 
condition of the possibility of all that exists - the condition to which all 
thought of objects, so far as their content is concerned, has to be traced back. It 
is also the only true [eigentliche] ideal of which the human reason is capable. 
For only in this one case is a concept of a thing - a concept which is in itself 
universal - thoroughly determined in and through itself, and known as the 
representation of an individual.« (B 604) 
At this point, everything seems to be clear for Kant. However, some of the 
problems still remain unsolved. But before we would draw any jumpy conclu-
sions, a few misunderstandings should be cleared up. First of all, Kant's 
language and conceptual apparatus which he uses repeatedly should not de-
ceive us. Kant describes transcendental ideal as »transcendental presupposi-
tion, namely, that of the material for all possibility, which in turn is regarded as 
containing a priori the data for the particular possibility of each and every-
thing.« (B 600-601) Moreover, though this ideal should not be conceived as 
given, for Kant to represent means »to represent everything as deriving its own 
possibility from the share which it possesses in this sum of all possibilities.« 
(B 600) Consequently, ideal is described as Urbild, prototypon, original of 
everything, on the other hand, things are described as mere ectypa, copies, 
which are derived from it.14 But all these repeatedly used terms are not, as Kant 
himself warns, »to be taken as signifying the objective relation of an actual 
object to other things, but of an idea to the concepts. We are left entirely 
without knowledge as to the existence of a being of such outstanding pre-
eminence.« (B 607)15 Although Kant repeatedly uses terms such as matter, 
14 Although Kant constantly uses the term derivation, there is, strictly speaking, no derivation at 
all. This applies, on the one hand, to two kinds of predicates - the first expresses a content, a 
reality without restraints, the second a certain degree of this content, a limited reality; although 
the second kind of predicates can be thought under the presupposition of the first, there is no 
derivation in the literal sense of a word, because both kinds of predicates are the result of the 
same synthesis and are produced simultaneously - and on the other hand, to the two kinds of 
negation, i.e. to the logical and the transcendental negation and their relation to the purified 
concept of transcendental ideal. 
15 There is another Kant's notice: »although in our first rough statements we have used such 
Kant's Transcendental Ideal as Fiction 107 
substratum, material etc., the transcendental ideal remains what it is, the idea 
of totality, the idea of reality, i.e. merely an idea: »It is obvious that reason, in 
achieving its purpose, that, namely, of representing the necessary thorough 
determination of things, does not presuppose the existence of a being that 
corresponds to this ideal, but only the idea of such a being.« (B 605-606)16 
However, if the transcendental ideal is to be understood strictly as an idea 
which is in the ultimate analysis nothing but a concept - how then to think it, 
since Kant obviously left us more or less in the dark? Perhaps the parallel 
between the transcendental ideal and the Kantian concept of space, the paral-
lel, which is proposed by Kant himself,17 would be of some help at this point. 
The transcendental ideal is, i.e. as space, not the concept which contains all 
predicates under itself, unter sich, but the concept which contains them within 
itself, in sich.n What is the difference? The concept, for Kant, contains infinite 
number of possible objects, infinite multitude of representations as their com-
mon mark. It contains them therefore under itself, unter sich. »But no concept, 
as such, can be thought as containing an infinite number of representations 
within itself. However, space is thought in this latter way...« (B 40) Space is 
single, infinite and divisible - we can divide only one single space, because 
there exists only one such space. But the trouble with the parallel between the 
transcendental ideal and space lies in the fact that, for Kant, space is an 
intuition, while the transcendental ideal is just a concept. As is well known, 
concept and intuition are strictly separated for Kant, which brings up the 
following question: is there any common ground for the parallel between 
language«. (B 607) In German: ob wir es gleich anfänglich im ersten rohen Schattenrise so 
vorstelleten. 
16 We have to admit that we are on very slippery grounds, at least as far as the level of language 
is concerned. It would suffice to understand the derivation as »a limitation of its supreme 
reality« (B 607), i.e. to conceive the Einschränkung as Teilung, and the manifold of things in 
the world would be understood as the effect of the supreme being. We could then easily proceed 
with the determination of such being in its unconditioned completeness and also succeed in 
determining it through all predicates. But such use of transcendental ideal would trespass the 
bounds of its legal employment »for reason, in employing it as a basis for the thorough 
determination of things, has used it as the concept of all reality without requiring that all this 
reality be objectively given and be itself a thing.« (B 608) But, »we have no right to do this, nor 
even to assume the possibility of such an hypothesis« (B 608). However, Kant knows very well 
that it »does not suffice merely to describe the procedure of our reason and its dialectic; we must 
also endeavor to discover the sources of this dialectic, that we may be able to explain, as a 
phenomenon of the understanding, the illusion to which it has given rise. For the ideal, of which 
we are speaking, is based on a natural, not on a merely arbitrary idea.« (B 609) 
17 Kant's example is: all figures presuppose different kinds and ways of limiting infinite space. 
(See: B 606). 
18 For Kant's distinction between unter sich/in sich see also, Logik (Jäsche) Werkausgabe, Bd. 
VI., p. 526 (§ 7), 529 (§ 13). 
108 Peter Klepec 
concept and intuition? Has not Kant himself already given an answer to this 
question claiming just few lines above that no concept as such can be thought 
as containing an infinite number of representations within itself? However, 
there may be a way out of this impasse if we do not search for their common 
ground, but only their common features. As Philonenko emphasizes19, space 
and the transcendental ideal are both forms, nothing but empty forms, and they 
are both something, not real, but ideal.20 This may be the reason why Kant 
characterizes the transcendental ideal as »simple [einfach]«. (B 607) And the 
simplicity of form could be - why not? - understood, as Hogrebe proposes, as 
»universal register, shrinked on a single one concept«21. This register, how-
ever, would be a paradoxical one: in fact empty, but always already fulfilled 
with content; nothing but empty form, and at the same time apparent substra-
tum; and finally, transcendental condition, yet not as real, as given. This 
paradoxical status of the transcendental ideal can also be formulated in the 
following way: although it seems that the transcendental ideal already con-
tains all possible predicates, although it appears to be some sort of an All or a 
Whole, it is actually not so - it can never be such a whole, it can never be 
Whole, for it can never be completed or accomplished. The reason for that is 
that the transcendental ideal does not have analytical, but synthetical character 
and it is the synthesis, the experience which confers upon the ideal its charac-
ter of syntheticity. Not surprisingly, this synthesis is also paradoxical - we are 
actually never ever in the position to add a missing, lacking, new or uncontained 
predicate to the sum-total of all predicates. It is due to the fact that it is 
impossible to find out whether a certain predicate is or is not contained in the 
transcendental ideal. Why? We would need to compare this predicate with all 
possible predicates and since there is an infinite number of possible predicates, 
there is no guaranty or Guarantor, which could ensure that the comparison 
would not last - infinitely long. In other words, such comparison would very 
probably last endlessly. 
If this is true, the transcendental ideal would represent a very inconvenient and 
thereby unneeded support. However, the time in question, i.e. the time needed 
for finding out the desirable predicate, is not such time to be characterized as 
logical and not as real time, since it is only with logical time that we can 
explain the immeasurable moment which passes between empty form as not-
yet fulfilled with content and always already fulfilled transcendental ideal as 
thoroughly determined? But as soon as we accept that the time which passes 
"Alexandre Philonenko, L'Oeuvre de Kant, 1. part, Paris 1989, Vrin, p. 316. 
20 That the form is a common feature of both space and transcendental ideal was already pointed 
out by Kant himself. See, for instance, R 6290, Kant's gesammelte Schriften, Bd. 18., p. 559. 
21 Wolfram Hogrebe, Pradikation und Genesis. Metaphysik als Fundamentalheuristik im 
Ausgang von Schellings »Die Weltalter«, Frankfurt/M 1989, Suhrkamp, p. 62. 
Kant's Transcendental Ideal as Fiction 109 
between not-yet and always already is the logical time, such time can also be, 
as immeasurable, characterized as infinite, for who can tell how long it takes to 
come to the conclusion from, for instance, the first and the second premise? 
How much time does it elapse during the logical operation, for example: if p 
then q? Therefore, the time which is needed for the thorough determination of 
transcendental ideal can be nothing but logical. Another reason for that lies in 
the fact that we, as human beings, are simultaneously on both sides. It is our 
experience which is identical with the transcendental ideal and it is merely an 
illusion that ideal itself »excludes [ausstoße]22 a multitude of predicates » (B 
601). It only appears that the transcendental ideal produces this multitude at 
the same time as it excludes the predicates. It is the same illusion which 
produces a semblance of the transcendental ideal as thoroughly determined. 
In spite of its paradoxical and illusionary status, the transcendental ideal is 
nevertheless thoroughly detennined. How? We have to determine repeatedly, 
whether a certain predicate belongs to it or not. However, it does not happen in 
real, but in logical time. Logic, i.e. the operation of disjunction, is crucial here. 
If a certain predicate is added to a particular thing, its opposite must be 
excluded. For example, if a thing is white, it is not red, blue, black etc., in 
short, non-white. In accordance with »either-or« of the upper premise of 
disjunctive syllogism, one predicate of each pair of contradictory opposites 
must belong to everything what exists. The upper premise in disjunctive 
syllogism contains logical division of concept or the division of sphere of a 
universal concept, the lower premise limits this sphere to the part, with which 
conclusion determines the universal concept. But the universal concept of 
reality cannot be divided a priori, for without experience we cannot know any 
kind of reality. The first or the upper premise is therefore nothing but represen-
tation of the sum-total of all predicates, i.e. the transcendental ideal. 
Where does the illusion that the transcendental ideal is identical to all reality 
have its origin? The answer is self-evident for Kant, and it lies in Transcenden-
tal Analytic. The possibility of objects of our senses is based in their relation to 
our thinking. We can think space and time a priori, but what makes out the 
content, the matter, the reality in the appearances (what corresponds to feel-
ing), must be given as real. If that condition is not fulfilled, if there's no 
content, no reality, we cannot think at all, and we cannot represent anything, 
because there is nothing to be thought or represented. Without the real there is 
also no thorough determination. »The material for the possibility of all objects 
of the senses must be presupposed as given in a sum-total [Inbegriff]; and it is 
22 German verb ausstoßen, which Kemp-Smith translates with the verb »to exclude«, can also 
mean: to extrude, to force out, to launch, to emit, to produce, to articulate, to express, to outlaw, 
to discriminate, to ostracize, to segregate, to separate, to except. 
110 Peter Klepec 
upon the limitation of this sum-total that all possibility of empirical objects, 
their distinction from each other and their thorough determination, can alone 
be based« (B 610), although we know that »as a matter of fact, no other 
objects, besides those of the senses, can be given to us, and nowhere save in 
the context of a possible experience; and consequently nothing is an object for 
us, unless it presupposes the sum [Inbegriff] of all empirical reality as the 
condition of its possibility« (Ibid.) But this possibility should be understood as 
a purely empirical principle, which is valid for appearances only, not for things 
in themselves. If we would demand of this principle to be valid for things in 
themselves too, we would sooner or later return to the old theological path, but 
that does not concern us here. 
Let us resume the result of our examination of Kant's concept of the transcen-
dental ideal. Its most notable feature seems to be its paradoxical status: in fact 
empty form, but never presented as such, i.e. pure and empty, for it is always 
already filled with content; the apparent substratum and transcendental condi-
tion, but not in the sense of something real. Though it is thoroughly deter-
mined, it is never a Whole, it is never completed or accomplished. This 
thorough determination of the transcendental ideal is paradoxical and illusion-
ary, which is due to the fact that it does not have analytical, but synthetical 
character, that it is the synthesis, the experience which confers upon it its 
character of syntheticity. This syntheticity is paradoxical as well - we are in 
fact never in the position to add a missing, lacking, new or uncontained 
predicate to the sum-total of all predicates since this predicate is always 
already contained in the transcendental ideal. It could be said, therefore, that 
the transcendental ideal is some kind of non-existing surplus, fiction, and yet 
as such the basis for every determination of things. As thoroughly determined 
concept, as a register of all possible predicates and as fiction which is the basis 
for our thinking and knowing, it resembles, with certain reservation, the 
conception of big Other of Jacques Lacan: that, too, is a universal register 
which is always already here, but not as existent. This Other is the Other of 
universal discourse, the Other which contains all what was said and what can 
be represented, it is the Other of Borges' total library, the treasure of signifiers. 
This Other has fictional status, too, it is pure symbolic order, but without it, in 
a very certain sense of the word, we could not even disagree... Of course, 
Lacan's concept of the Other has a broader meaning than the one presented 
here, but as a mere hint, it suffices at least to illustrate the fact that, although 
not fully developed by Kant himself, the concept of the transcendental ideal as 
presented here allows us to claim at least that, for Kant, fiction is certainly not 
just illusion.