Esquematismo Trascendental

Transcendental Schematism and The Problem of the Synthetic "A Priori"
Author(s): Henry E. Allison

Source: Dialectica, Vol. 35, No. 1/2 (1981), pp. 57-83
Published by: Wiley
Stable URL: http://www.jstor.org/stable/42968792
Accessed: 20-03-2016 06:54 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.
Wiley is collaborating with JSTOR to digitize, preserve and extend access to Dialectica.
http://www.jstor.org
This content downloaded from 128.111.121.42 on Sun, 20 Mar 2016 06:54:03 UTC
All use subject to JSTOR Terms and Conditions
Transcendental Schematism and The Problem of the Synthetic
A Priori
by Henry E. Allison*
Summary
The paper is concerned with the connection between Kant's conception of transcendental
schematism and his analysis of the conditions of the possibility of synthetic a priori judgments.
After dealing with some of the standard objections to Kant's theory, I argue that transcendental
schemata must be construed as pure intuitions. I then point out that the Principies of Pure
Understanding are a set of synthetic a priori judgments which assert the function of the various
schemata as necessary conditions of the possibility of experience, and that each of these
judgments involves the relation of a pure concept to pure intuition (as is required by Kant's theory
of synthetic a priori). Finally, I argue that the connection between a category and its schema must
itself be both synthetic and a priori. It therefore requires a "deduction", which Kant himself
does not provide, but which I endeavor, in part, to provide for him.
Rsum
Cet article tudie la connexion entre la conception kantienne du schmatisme transcendental

et son analyse des conditions de possibilit de jugements synthtiques a priori. Aprs avoir
rpondu certaines des objections habituellement faites la thorie de Kant, je montre que les
schmas transcendentaux doivent tre interprts comme des intuitions pures. Je relve ensuite
que les principes de l'entendement pur sont un ensemble de jugements synthtiques a priori qui
assignent aux divers schmas la fonction de conditions ncessaires la possibilit de l'exprience,
et que chacun de ces jugements implique la relation d'un concept pur un intuition pure (comme
le rclame la thorie kantienne du synthtique a priori ). Finalement, je montre que la connexion
entre une catgorie et son schma doit tre elle-mme la fois synthtique et a priori. Elle exige
donc une dduction , que Kant lui-mme n'a pas fournie, mais que je me propose d'esquisser
sa place.
Zusammenfassung
Diese Arbeit betrifft die Verbindung zwischen Kants Auffassung des Schematismus' und sei-
ner Analyse der Bedingungen fr die Mglichkeit von synthetischen Urteilen a priori. Nach der
Behandlung einiger bekannter Einwnde gegen Kants Theorie argumentiere ich, dass transzen-
dentale Schemata als reine Anschauungen konstruiert werden mssen. Ich zeige dann, dass die
Prinzipien des reinen Verstandes eine Menge von synthetischen Urteilen a priori darstellen, die die
Funktion der verschiedenen Schemata als notwendige Bedingungen fr die Mglichkeit von Er-
fahrung festlegen und dass jedes einzelne dieser Urteile die Relation eines reinen Begriffs zur rei-
nen Anschauung betrifft (wie Kants Theorie des Synthetischen a priori das verlangt). Schliesslich
* University of California, San Diego.
Dialctica Vol. 35, N<> 1-2 (1981)
58 Henry E. Allison
argumentiere ich, das s die Verknpfung zwischen einer Kategorie und deren Schema selbst so-
wohl synthetisch als auch a priori sein muss. Sie macht deshalb eine Deduktion erforderlich,
die Kant selbst zwar nicht liefert, die ich aber an seiner Stelle zu geben versuche.
In a letter to Reinhold, wherein he complains of Eberhars attack on the

Critique , Kant reflects:
Mr. Eberhard says: "One seeks in vain for Kant's principle of syn-
thetic judgment." But this principle is completely unambiguously
presented in the whole Critique , from the chapter on the schematism
on, though not in a specific formula. It is: All synthetic judgments of
theoretical knowledge are only possible through the relation of a given
concept to an intuition . If the synthetic judgment is experiential, the
underlying intuition must be empirical; if it is a judgment a priori, the
intuition must be pure.1
By suggesting that it is in the chapter on the schematism and not, as one
might expect, in the Introduction or the Transcendental Deduction that we
find the beginning of an account of synthetic judgments in general and syn-
thetic a priori judgments in particular, Kant attributes to this murky text a
systematic and philosophical significance which it is not usually considered to
possess. Behind this suggestion lies the appealing picture of the Transcenden-
tal Analytic as containing a unified line of argument, the goal of which is to
establish a set of synthetic a priori judgments with respect to objects of possi-
ble experience. This is equivalent to establishing what Kant calls
"metaphysics in its first part" (BXVIII), and which Paton has judiciously
termed a "metaphysic of experience".2 Within this project, the chapter on
the schematism is presumed to play a crucial role.3
The purpose of this paper is to examine this claim and to delineate the
precise contribution which the schematism makes to the Transcendental
Analytic. After a consideration of some of the standard criticisms of Kant's
account, I shall argue that transcendental schemata are to be construed as
pure intuitions which, as the letter to Reinhold indicates, are necessary for
synthetic a priori judgments. I shall further argue that, despite initial ap-
pearances, the synthetic a priori judgments afffirmed in the Principles of Pure
1 Kant's letter to Reinhold, May 12, 1789, Kants gesammelte Schriften , Edited by the
Kniglich preussischen Akademie, Berlin, 1902 ff. Vol. XI, p. 30.
2 H. J. Paton, Kant's Metaphysic of Experience, London, 1936, Vol. I, p. 72.
3 The role of the schematism in "laying the foundations of metaphysics" is, of course,
asserted most emphatically by Martin Heidegger in Kant und das Problem der Metaphysik, Bonn,
1929. In what follows I shall be arguing for a similar role for the schematism, but I shall be basing
my argument on a very different understanding of the nature of metaphysics than Heidegger's.
Transcendental Schematism and The Problem of the Synthetic A Priori 59
Understanding predicate pure concepts of pure intuitions. Finally, I shall sug-

gest that this interpretation of transcendental schemata and of the function of
this chapter brings to the fore a problem which Kant himself never clearly ad-
dressed and which has received scanty treatment in the literature. This is the
problem of establishing the connection between the particular categories and
their respective schemata. I shall show that a judgment affirming such a con-
nection between a category and its schema must be construed as both synthetic
and a priori . Kant's failure to provide a clear account of how such judgments
are to be justified thus suggests that the problem of the synthetic a priori reap-
pears in a particularly perplexing form in the very chapter which is assigned a
crucial role in its resolution. I shall also contend, however, that Kant has pro-
vided us with the necessary materials for dealing with this problem.
In addition the the charges of obscurity and internal incoherence, the most
common objection to the schematism chapter is that of superfluity. As is so
often the case, the classical formulation of this line of objection belongs to
Prichard. Moreover, since his charge is directed not only against the
schematism chapter, but the whole second book of the Transcendental
Analytic , it can be construed as a direct challenge to Kant's own evaluation of
the situation as it is reflected in his comments to Reinhold. In Prichard's
words:
We naturally feel a preliminary difficulty with respect to the existence

of this second part of the Analytic at all. It seems clear that if the first
part is successful, the second must be unnecessary. For if Kant is in a
position to lay down that the categories must apply to objects, no
special conditions of their application need be subsequently determin-
ed. If, for instance, it can be laid down that the category of quantity
must apply to objects, it is implied either that there are no special con-
ditions of its application, or that they have already been discovered
and shown to exist. Again, to assert the applicability of the categories
is really to assert the existence of principles, and in fact of just those
principles which it is the aim of the System of Principles to prove.
Thus to assert the applicability of the categories of quantity and of
cause and effect is to assert respectively the principles that all objects
of perception are extensive quantities, and that all changes take place
according to the law of cause and effect.4
4 H. A. Prichard, Kant's Theory of Knowledge, Oxford, 1909, pp. 77-82.
60 Henry E. Allison
A more recent version of the same basic objection, albeit with specific
reference to the schematism, has been provided by J. G. Warnock. He main-
tains that the presence of this chapter in the Critique is due to Kant's illicit
separation of the possession of a concept from the ability to use it. The pur-
ported goal of the Transcendental Deduction is to show that we possess a cer-
tain set of concepts. If this goal were obtained, there would be no remaining
question about the applicability of these concepts. But it is just this pseudo-
question with which Kant, according to Warnock, is concerned in the
schematism chapter.5
The standard, indeed, the mandatory strategy for the defense of the
schematism against this line of objection is first to stipulate what the
Transcendental Deduction establishes, assuming its soundness, and then to
suggest what tasks remain.6 The former, of course, is itself a highly controver-
sial matter; but it does seem possible here to ignore most of the interpretative
issues. The essential, and presumably uncontroversial point is that, in the
Transcendental Deduction, Kant claims to have established the objective reali-
ty or the "right" to use a set of a priori or "pure" concepts with respect to
the objects of human experience. He also claims that this is the very same set
of concepts which he has identified in the Metaphysical Deduction by means
of an analysis of the nature of the understanding and its characteristic activi-
ty, judgment. Kant's procedure, which becomes clearer in the Second Edition,
wherein the argument is explicitly divided into two stages, involves linking
these concepts with the two essential components of human experience: con-
ception and perception.7 The former is accomplished by connecting these con-
cepts with the transcendental or objective unity of apperception; the latter by
connecting them with the transcendental synthesis of the imagination, and
thereby with the spatio-temporal structure of human experience. As we shall
see shortly, the latter linkage is crucial for the interpretation of the
schematism, because it establishes that these concepts, which have their origin
in the very nature of the understanding, stand in a determinate relationship to
time.
5 G. J. Warnock, "Concepts and Schematism", Analysis IX, 1949, pp. 77-82.

6 Cf. Eva Schapere, "Kant's Schematism Reconsidered", Review of Metaphysics, XVIII,
1964, pp. 267-292; Moltke S. Gram, Kant , Ontology & the A Priori , Evanston, 1968; and
Lauchlan Chipman, "Kant's Categories and their Schematism", Kant-Studien, Vol. 63, 1972,
pp. 36-49.
7 Cf. Critique of Pure Reason , 21, B144-145 and 26, B159-160. The crucial point here is the
two step proof structure of the B Deduction. An interesting and influential interpretation of the
proof structure has been provided by Dieter Henrich, "The Proof-Structure of Kant's
Transcendental Deduction, Review of Metaphysics , XXII, 1968-69, pp. 640-659. My own inter-
pretation of this proof structure differs substantially from Henrich's, but these differences are
not directly germane to the present issue.
Admittedly, such a cursory review does not shed much light on the in-
tricacies of the Transcendental Deduction , but it should suffice to discredit
Prichard's charge. To cite his own example, it simply is not true that 4 4 to
assert the applicability of the categories of quantity and cause and effect is to
assert the principle that all objects of perception are extensive quantities, and
that all changes take place according to the law of cause and effect ".To know
that the concept of quantity is applicable to the objects of human experience is
not yet to know either that it applies universally, or that the objects to which it
applies are extensive quantities. The latter point, in particular, requires the ad-
ditional knowledge of how this concept is expressed in sensible terms, that is,
how it is schematized. Again, to know that the category of causality is ap-
plicable to objects of human experience is not to know either that it relates
specifically to alterations of the state of substances, or that it relates to all
such alterations.8
Similar considerations apply to Warnock's version of the objection. War-
nock, it will be recalled, suggested that the very presence of a theory of
schematism in the Critique is due mainly to Kant's mistaken separation of the
possession of a concept from its use. In reality, however, the Transcendental
Deduction presupposes that the pure concepts have a "logical use", that is, a
use with respect to the various forms of judgment. Presumably, this was
established in the Metaphysical Deduction , wherein Kant tried to argue that
judging under each of the designated "forms" presupposes the correspon-
ding concept (intellectual function).9 The question with which the
Transcendental Deduction deals is whether these same concepts also have an
extra-logical or "real use" with respect to a sphere of objects (objects of
possible experience). It can be read as an attempt to provide a general argu-
ment to the effect that these concepts must be assumed to have such a use, i.e.,
that they are applicable to objects of possible experience through their connec-
tion with the transcendental synthesis of the imagination. It does not,
however, even attempt to show how, or under what specific conditions, the
particular concepts are to be applied. Correlatively, it does not tell us what
characteristics of these objects count as the sensible expressions of the concep-
tual relations thought in pure concepts. For example, it does not inform us
what property or relation of appearances in time counts as the sensible expres-
sion of the logical relation of ground and consequent. It is the task of the
schematism to determine this. As Kant put the matter in an important Reflex-
ion: "The schematism shows the conditions under which an appearance is
8 Cf. Patn, op. cit. Vol. II, p. 67.

9 A similar point is made by Gram in his discussion of Warnock 's objection, albeit without
specific mention of a "logical use", op. cit. pp. 89-91.
62 Henry E. Allison
determined in respect to a logical function and therefore stands under a

category." 10
Further difficulties arise from the fact that Kant casts the whole problem
of finding sensible conditions which license the application of specific
categories to appearances in terms of subsumption. This occurs twice, and
' subsumption ' seems to be used in a different sense in each case.11 The first is
in the introduction to the Transcendental Doctrine of Judgment, where Kant
defines judgment as "the faculty of subsuming under rules, that is, of
distinguishing whether something does or does not stand under a given rule
( casus datae legis)99. (A132/B177) Here the rules are the pure concepts, and
the schemata function as the means for determining whether a given ap-
pearance stands under one or another of these rules. The second, and really
notorious passage, occurs at the opening of the schematism chapter itself.
Kant begins by noting that all cases of subsumption of an object under a con-
cept require some homogeneity between the two. This is presumably il-
lustrated by the relation between the geometrical concept of a circle and the
empirical concept of a plate (not, as one would expect, the plate itself). Here
the homogeneity is said to consist in the fact that what is intuited in the former
(the geometrical concept) is thought in the latter.12 Then, by way of contrast
to this relatively unproblematic situation, comes the decisive passage:
But pure concepts of understanding being quite heterogeneous from
empirical intuitions, and indeed from all sensible intuitions, can never
be met with in any intuition. For no one will say that a category, such
as that of causality, can be intuited through sense and is itself contain-
ed in appearance. How, then, is the subsumption of intuitions under
pure concepts, the application of a category to appearances, possible?
A transcendental doctrine of judgment is necessary just because of this
natural and important question. (A136-137/B176-177)
Unfortunately, most of Kant's commentators have found this question, at
least in the way in which he here formulates it, to be neither natural nor im-
portant. According to the standard line of objection, 'subsumption' is here
used in the sense in which it is found in the traditional theory of judgment,
that is, to designate the relation between a class concept and the particulars
10 Kants gesammelte Schriften , Vol. XVIII, p. 392, Reflexion , no. 5133.

11 This is pointed out by Ernst Robert Curtius, "Das Schematismuskapitel in der Kritik der
reinen Vernunft* ', Kant-Studien , Vol. XIX, 1916, pp. 338-366, esp. p. 348, and by Norman Kemp
Smith, A Commentary to Kans Critique of Pure Reason, Second Edition, New York, 1962,
p. 336.
12 Critique of Pure Reason , A137/B176. I am here following the original version of the text
rather than Vaihingens emendation which was adopted by Kemp Smith in his translation. For a
discussion of this point see Paton, op. cit., Vol. II, p. 26, note 1.
falling under it. To judge on this theory is just to subsume particulars under
such a concept. Evidence for this interpretation is presumably supplied by
Kant's example of the subsumption relation between the geometrical concept
of the circle and the empirical concept of a plate. Given this reading, it is then
easy to argue not only that this notion of subsumption is hopelessly inade-
quate to capture the nature of judgment, but also that such a formulation
mistakes the relation which, according to Kant's own doctrine, holds between
the categories and the sensibly given. As Kemp Smith puts it, this relation is
properly one of form and matter, structure and content, not universal and
particular.13
Admittedly, the opening paragraphs of the schematism chapter cannot be
characterized as a model of philosophical lucidity. Still, it would be surprising
if Kant were in fact as confused as his commentators assume him to be. First
of all, one can question whether Kant really intended his example of the rela-
tion between the geometrical concept of a circle and the empirical concept of a
plate to be taken as an instance of the subsumption of a particular under a
class concept. This suggests that the concept of a plate (or rather the plate
itself) can be taken as a member of the class of circles; but this can hardly be
what Kant meant. Rather, following Paton, we can see the key to Kant's in-
tent to lie in the initially puzzling reference to "the roundness which is in-
tuited in the former" (the pure geometrical concept of a circle).14 Assuming
this, we can then read Kant to be maintaining that it is the possibility of ex-
hibiting the geometrical concept in pure intuition (constructing a circle) that
explains the homogeneity between it and objects, such as plates, which are
given in empirical intuition, as well as the empirical concepts which are form-
ed by abstraction from the content of such intuition ("thought in the latter ").
The homogeneity, in short, is between pure and empirical intuition, not bet-
ween a class concept and a member of that class. Moreover, only such a
reading enables us to make any sense of the contrast which Kant wishes to
draw between the geometrical concept and the pure concepts of the understan-
ding, which "can never be met with in any intuition".
More significantly, such an interpretation of the circle-plate relation
relieves us of any necessity of construing the still problematic relation between
the pure concepts and appearances in terms of the notion of subsumption as it
is operative in the traditional theory of judgment. In fact, as the text already
13 Norman Kemp Smith, op. cit., pp. 335-336. He is here essentially following Curtius. The
line of objection which I have here tried to sketch is intended to capture the main thrust of the
criticisms of both Curtius and Kemp Smith.
14 Paton, op. cit., Vol. II, p. 26, note 1.
64 Henry E. Allison
indicates, Kant is using 'subsumption ' as a synonym for 'application '. 15

Thus, rather than making the former into her "trouser word", which would
force us to conclude that Kant has mistakenly construed the problem of the
application of the categories in terms of the judgmental notion of subsump-
tion, it seems to be both reasonable and in accordance with the text to read
'subsumption 9 here as simply meaning 'application
If we were to stop here, however, we would not so much have interpreted
'subsumption 9 as interpreted it away. Certainly, an adequate exegesis must
show that the term is not completely idle. Fortunately, this is quite easy to do.
The key point is that it is the syllogistic, not the judgmental, conception of
subsumption that is at work here, and that is supposed to provide an analogy
for understanding the problem of the application of the pure concepts of the
understanding to appearances.16 Kant sketched his views on this topic in the
Critique and in a parallel passage in the Lectures on Logic ( 58). As he puts it
in the Critique , inferring or "judging mediately" takes place "by the sub-
sumption of the condition of a possible judgment under the condition of a
given judgment". The "given judgment" is the universal rule which func-
tions as the major premise ("everything composite is alterable"). The minor
premise is characterized as "the subsumption of the condition of another
possible judgment under the condition of the rule" ("bodies are
composite"). The conclusion is the "mediate judgment", which results from
the application of the rule to the subsumed case ("bodies are alterable").
(A330-331/B386-387) The crucial term here is 'condition'. Recall that in the
minor premise, it is the "condition of another possible judgment" (bodies)
that is subsumed under "the condition of the rule " (not the rule itself). In ad-
dition, Kant tells us that this rule "states something universal, subject to a
certain condition". In the categorical syllogism in the text, the condition of
the rule is the quality of being composite. The minor premise then asserts that
this condition is met in the case of bodies. This licenses the conclusion that all
bodies are alterable. The condition of the rule is thus the middle term in the
syllogism, the "third thing", which connects the universal rule with the par-
ticulars to which it is applied in the conclusion.
Kant hardly wished to construe the application of the categories to ap-
pearances as a bit of syllogistic reasoning. Nevertheless, the analogy with such
reasoning does serve to underscore the particular problem which arises in the
15 Cf. Gerold Prauss, Erscheinung bei Kant , ein Problem der " Kritik der reinen Vernunft
Berlin 1971, p. 103.
16 Both Curtius, op. cit., p. 348 ff. and Kemp Smith, op. cit., p. 336, acknowledge the
suitability of this model for the presentation of the problematics of the schematism, but then pro-
ceed to criticize Kant for not adhering to it!
attempt to understand how such application is possible. As already noted, the

problem is the heterogeneity of the two elements to be brought into connec-
tion. This heterogeneity is due to the fact that the pure concepts of the
understanding, in contrast even to "pure sensible", i.e., mathematical, con-
cepts, which are likewise a priori , are derived from the very nature of the
understanding. As such, they have no direct relation to intuition. Yet, as the
Transcendental Deduction demonstrates (again assuming its soundness), they
do relate to intuition and therefore to appearances. Given all of this, it is easy
to see that the whole purpose of the analogy is to suggest that, just as the case
of syllogistic reasoning, the connection between the rule expressed in the ma-
jor premise and the instance to which it is applied in the conclusion is only
established by means of the subsumption of the instance under the condition
of the rule; so in the present case, where the pure concepts of the understan-
ding are the universal rules, there is a need for some analogue of the condition
of the rule, or the middle term of the syllogism, under which appearances can
be "subsumed". This analogue will, of course, turn out to be the
transcendental schema, the infamous "third thing", which makes possible
the mediation between category and appearance.
Finally, this analogy enables us to see the importance, if not the
naturalness, of the question which Kant raises in terms of the notion of sub-
sumption. Since concepts for Kant are "predicates of possible judgments",
(A69/B94) and since judgments which apply concepts to appearances are syn-
thetic, while those which apply a priori concepts are likewise a priori, it
follows that the question with which Kant is concerned at the beginning of the
schematism chapter is really the question of how synthetic judgments are
possible a priori. 17 When he first posed this question in general terms in the
Introduction to the Critique , without any specific reference to the pure con-
cepts of the understanding, Kant alluded mysteriously to an "unknown = x"
(B13) which is needed to ground the connection between the concepts that is
asserted in the judgment. When he returns to this question in the
Transcendental Analytic, this "unknown = x" is more precisely characteriz-
ed as a transcendental schema.
II
What, then, is a transcendental schema, such that it is capable of playing

this significant role? This would seem to be a straightforward question, for
which there must be a direct, unambiguous answer. When we turn to the text,
17 See my The Kant-Eberhard Controversy, esp. p. 60.
66 Henry E. Allison
however, we find a bewildering variety of answers, not all of which are ob-
viously compatible with one another. Limiting ourselves to the schematism
chapter, and even there ignoring minor variations as well as the important ac-
count of the schemata of mathematical and empirical concepts, we find the
notion of a transcendental schema characterized in the following ways:
1) As a "third thing" or "mediating representation", which is
"homogeneous on the one hand with the category, and on the other hand
with the appearance, and which thus makes the application of the former to
the latter possible". Kant also claims that such a representation must be
"pure" (non-empirical), and that "while it must in one respect be intellec-
tual, it must in another be sensible (A138/B177)
2) As a "transcendental determination of time", which, as suggested in (1), is
homogeneous with both category and appearance, and which therefore
"mediates the subsumption of the appearances under the category".
(A139/B178).
3) As the "formal and pure condition of sensibility to which the concept of
understanding is restricted". (A140/B179)
4) As the "representation of a universal procedure of imagination in pro-
viding an image for a concept ". (A140/B179-180) This is intended as a general
characterization of a schema. It leads directly to the brief account of the
schemata of mathematical and empirical concepts as rules for the construction
of images.
5) As "simply the pure synthesis, determined by a rule of that unity, in accor-
dance with concepts, to which the category gives expression". This formula-
tion underscores the point that, unlike those of other concepts, the schemata
of the pure concepts of the understanding cannot be construed as rules for the
construction of images. In this context, it is further described as a
"transcendental product of imagination, a product which concerns the deter-
mination of inner sense in general according to conditions of its form (time) in
respect of all representations, so far as these representations are to be con-
nected a priori in one concept in conformity with the unity of apperception ".
(A142/B181)
6) As "the true and sole conditions under which these concepts obtain relation
to objects and so possess significance" . (A146/B185)
7) As "nothing but a priori determinations of time in accordance with rules ".
(A145/B184)
8) As "only the phenomenon, or sensible concept of an object in agreement
with the category". (A146/B186)
In the face of this plethora of formulations, the thesis for which I wish to
argue is that a transcendental schema is to be construed as a pure intuition,
and that this is compatible with all of these formulations, with the possible ex-
ception of (4). Although none of these formulations explicitly identify a
transcendental schema with a pure intuition, such an identification is obvious-
ly compatible with (3) and (6). More germane, however, is the fact that Kant
does explicitly characterize a transcendental schema as a pure intuition in both
the Critique of Practical Reason and the Critique of Judgment . In the former
work he writes, in connection with the discussion of the Typic of Pure Prac-
tical Judgment:
The judgment of pure practical reason, therefore, is subject to the
same difficulties as that of the pure theoretical, though the latter had a
means of escape. It could escape because in its theoretical use
everything depended upon intuitions to which pure concepts of the
understanding could be applied, and such intuitions (though only of
objects of the senses), as a priori and hence concerning the connection
of the manifold in intuitions, could be given a priori in conformity to
the concepts of the understanding, i.e., as schemata.18
In the latter he notes:
Intuitions are always required to verify (< darzuthun ) the reality of our
concepts. If the concepts are empirical the intuitions are called ex-
amples: if they are pure concepts of the understanding the intuitions
go by the name of schemata.19
The claim that transcendental schemata are to be construed as pure intui-
tions is not new. Moltke Gram has insisted upon it in his own account of the
Transcendental Analytic . Gram, however, grants to this interpretation a
somewhat subterranean status, since he connects it with what he calls Kant's
" hidden" or "implicit" theory of syntheticity. According to this theory, syn-
thetic judgments predicate concepts of intuitions, and synthetic a priori
judgments predicate pure concepts of pure intuitions. Transcendental
schemata, construed as pure intuitions, are on this view the referents of the
schematized and not the pure concepts.20 Nevertheless, Gram explicity denies
that this interpretation of a transcendental schema is compatible with all of
Kant's accounts. In particular, he denies its compatibility with the 'third
thing' account. The heart of the problem, according to Gram, lies in what he
takes to be the incoherence of the notion of a "third thing", which is both
universal and particular, intellectual and sensible. Since these constitute two
sets of contradictory properties, nothing can possess both members of either
18 Kants gesammelte Schriften , Vol. V, p. 68.

19 ibid., Vol. VI, p. 251.
20 Gram, op. cit., esp. pp. 128-129.
68 Henry E. Allison
set. Moreover, even if something could, it would not be pure intuition, which
is by definition entirely sensible and particular.21
I have dealt in some detail with this objection in the Introduction to The
Kant-Eberhard Controversy . Here I can only repeat my central contention;
viz., that it rests upon a failure to distinguish between two senses in which
Kant uses the notion of pure intuition.22 Each of these senses involves a
reference to the notion of form; the distinction between them corresponds to
the distinction which Kant first clearly draws with regard to space in a foot-
note in the Transcendental Deduction of the Second Edition.23 This is between
space as a mere form of intuition or sensibility and the actual representation
of space as an object (as in geometry), which is a formal intuition . Both can be
called 'pure intuitions', and in the Transcendental Aesthetic Kant did not
distinguish between them. Nevertheless, the former is a completely indeter-
minate, unconceptualized pure intuition which presents a manifold as poten-
tial material for thought, while the latter is a pure intuition with a deter-
minate, conceptualized content.24 Obviously, of these two, only the first can
be said to be purely sensible; so that Gram's whole objection to the third thing
formulation rests upon the assumption that the notion of a pure intuition can
be taken only in the first sense.
Nor can this second (determinate) sense of pure intuition be taken as an
afterthought to which Kant merely alludes in an obscure footnote in the Se-
cond Edition. On the contrary, it is a central thesis of the Transcendental
Deduction, even in the First Edition, that it is only insofar as the "pure
manifold" of intuition (indeterminate pure intuition) is synthesized in accor-
dance with the categories that it can be brought to the unity of consciousness
and thus yield an actual content for cognition. As Kant clearly states, apart
21 Ibid., pp. 91-94.

22 The Kant-Eberhard Controversy , pp. 86-88.
23 The note reads as follows: "Space, represented as object (as we are required to do in
geometry) contains more than mere form of intuition; it also contains combination of the
manifold, given according to the form of sensibility, in an intuitive representation, so that the
form of intuition gives only a manifold, the formal intuition gives unity of representation. In the
Aesthetic I have treated this unity as belonging merely to sensibility, simply in order to emphasize
that it precedes any concept, although, as a matter of fact, it presupposes a synthesis which does
not belong to the senses but through which all concepts of space and time first become possible.
For since by its means (in that the understanding determines the sensibility) space and time are
first given as intuitions, the unity of this a priori intuition belongs to space and time, and not to
the concept of the understanding (Cf. 24). (B160-161, note a)
24 The distinction is also implicit in the response to Eberhard, wherein Kant denies the in-
nateness of the actual representations of space and time, The Kant-Eberhard Controversy , p. 136,
Kants gesammelte Schriften , Vol. IX, pp. 222-237. Perhaps even more germane, in a Reflexion
which deals specifically with the schematism, Kant distinguishes between the intuition and the
determination of time. (Zeitanschauung and Zeitbestimmung ), Reflexionen 6359, Kants gesam-
melte Schriften , Vol. XVIII, p. 686.
from such a synthesis, "not even the purest and most elementary representa-
tions of space and time, could arise". (A102) Surely, such representations
count as pure intuitions in the Kantian sense; so that we are led inevitably to
conclude that the conception of a determinate pure intuition is as central to
Kant's thought as the doctrine of the transcendental synthesis of the imagina-
tion, from which it is in fact inseparable. Moreover, it is just such a (deter-
minate) pure intuition to which the mathematician appeals when he exhibits or
constructs his concepts.25 The pure (formal) intuition that is produced by such
an activity is both sensible and intellectual (it is a sensible presentation of a
concept), and it is also both universal and particular. Indeed, Kant himself
claims as much in his account of geometrical construction when he notes:
To construct a concept means to exhibit a priori the intuition which
corresponds to the concept. For the construction of a concept we
therefore need a non-empirical intuition. The latter must, as intuition,
be a single object, and yet none the less, as the construction of a con-
cept (a universal representation), it must in its representation express
universal validity for all possible intuitions which fall under the same
concept. (A713/B741)
If, therefore, we take the term 'pure intuition' in the second, determinate
sense, there is no longer any difficulty in reconciling the claim that a
transcendental schema is a pure intuition with the initial characterization of it
as a "third thing". We must now consider whether it is compatible with the
second and decisive characterization of such a schema as a "transcendental
determination of time". Since most of the other characterizations are clearly
equivalent to or derivative from this one, showing the compatibility in this
case should suffice to justify the interpretation.
Some apparent plausibility is certainly derived from the fact that time, for
Kant, is a pure intuition. Thus far, however, the plausibility is merely ap-
parent; for from the fact that time is a pure intuition, it by no means follows
that a transcendental determination of time is one also. Moreover, the text is
not of very much direct help in this matter. Rather than providing us with an
account of just what he means by a transcendental determination of time,
Kant simply affirms its homogeneity with the category "which constitutes its
unity, in that it is universal and rests upon an a priori rule", and with ap-
pearance "in that time is contained in every empirical representation of the
manifold". (A138-139/B177-178) Nevertheless, by paying attention to some
25 In this regard, it is worthy of note that, in his polemic with Eberhard, Kant explicitly
characterizes mathematical construction as schematic rather than mechanical, The Kant-
Eberhard Controversy , p. 1 1 1 ; Kants gesammelte Schriften , Vol. VIII, p. 192. The point of this is
that what is constructed is the schema of the concept.
70 Henry E. Allison
of the hints contained here and, more particularly, by a consideration of what

Kant means by a ''transcendental determination ", it should be possible to ar-
rive at some understanding of this difficult notion.
Let us begin with a consideration of what, in general, it means for Kant to
"determine an intuition" (keeping in mind that time is an intuition). This ex-
pression is not very frequent, but where we do find it, it is clear that is means
to synthesize, conceptualize or subsume the given intuition under a concept
(all of these here being used synonymously) in such a way that the intuition is
related to, or represents, an object.26 Moreover, only by this means can a con-
cept itself come into relation to an object and thereby gain objective reality.
Now 'transcendental' is here equivalent to ' a priori ' i.e., universal and
necessary (hence the equivalence of (2) and (7)). Thus, a transcendental deter-
mination of an intuition would simply be one that is governed by an a priori
concept, which is precisely what Kant indicates in the passage cited above.
Putting this together and applying it to time, we arrive at the result that a
transcendental determination of time must be a conceptualization of time in
accordance with an a priori concept, which refers time to an object or objec-
tifies it, while also providing objective reality for the concept involved. To ob-
jectify time means to represent a temporal order as an intersubjectively valid
order of events in the phenomenal world, in contrast to a merely subjective or
"subjectively valid" order of representations in an individual consciousness.
But this is just what Kant claims in the Transcendental Deduction is effected
by the transcendental synthesis of the imagination. Consequently, a
transcendental determination of time can also be characterized as a product of
this synthesis or, as Kant puts it, a "transcendental product of the imagina-
tion". (5)27
These considerations all seem to lead to the conclusion that transcendental
determinations of time, as products of the transcendental synthesis of the
imagination, are universal and necessary (a priori) characteristics of objective
time or an objective temporal order. This would explain their homogeneity
with appearances, since all appearances are contained in such an order, and
with the pure concepts, since these concepts provide the rules whereby this
order is determined. The situation, however, is considerably complicated by
the fact that time (as well as space) is not itself an object and cannot be
26 Cf. Critique of Pure Reason , B 153-156, where Kant talks specifically about the determina-
tion of inner sense and his letter to Beck, July 3, 1792, Kants gesammelte Schriften , Vol. XI,
p. 348.
27 So construed, a "transcendental determination of time" is also equivalent to "the syn-
thesized in general" (das Zusammengesetzten berhaupt) to which Kant refers in his cor-
respondence with Beck and Tieftrunk. See especially Kant's letter to the latter, Dec. 11, 1797,
Kants gesammelte Schriften, Vol. XII, pp. 222-225.
Transcendental Schematism and The Problem of the Synthetic A Priori 7 1
perceived. Consequently, a transcendental determination of time must be

regarded as a universal and necessary characteristic of things in time
(phenomena) in virtue of which time can be represented in a determinate man-
ner (objective time relations can be expressed), rather than as a characteristic
of time per se.2% To cite a single example, the schema of the concept of
substance is said by Kant to be the "permanence of the real in time".
(A143/B183) Kant will argue in the First Analogy that it is only by reference to
something permanent that change, and with it the passage of time, can be in-
tuited. Our concern here is not with the cogency of this argument, but simply
with the claim itself, viz., that it is only by reference to the permanent that the
passage of time can be determined and, therefore, that the presence of
something permanent is a necessary feature of any world that can be ex-
perienced as temporal. Given this, together with the above considerations, it
follows that permanence, although a property of things in time, rather than of
time itself, can nonetheless be regarded as a transcendental determination of
time.
But does not this account of a transcendental determination of time com-
pletely undermine the claim that a transcendental schema is a pure intuition?
How can such a necessary characteristic of things in time, or even the
representation of such a characteristic, be termed a pure intuition? To con-
tinue with our example: permanence is a concept which refers to any number
of possible objects; and the same can be said, mutatis mutandis of any of the
schemata. This is undoubtedly why transcendental schemata are frequently
construed as concepts and equated with the schematized categories.
Moreover, such an interpretation is not only suggested by the above account,
but it also seems to be confirmed by Kant's own characterization of the
schema as "only the phenomenon, or sensible concept, of an object in agree-
ment with the category". (8)
Since a transcendental schema has already been defined as a determinate
pure intuition, and since this has been seen to mean one that is conceptualized,
it should come as no great surprise to find a conceptual component in its
characterization. Here the emphasis must fall on the fact that it is a sensible
concept. Although Kant, of course, begins with the radical separation of sen-
sibility and understanding, intuition and concept, the very heart of his ac-
count of knowledge consists in the claim that any cognition of an object (syn-
thetic judgment) involves both elements. With regard to permanence, as well
as with all of the transcendental schemata, the intuitive element must be
located in the irreducibly sensible component of the representation. To think
28 Paton, op. cit., Vol. II, pp. 28-30.
72 Henry E. Allison
of something permanent is just to think of it as lasting through time. The con-

cept thus "rests upon" or presupposes the intuition of time, which it deter-
mines. In this respect it is analogous to a mathematical concept, such as that
of a triangle, which has an essential reference to the pure intuition of space,
wherein it is exhibited or constructed, that is, realized.
It will here be objected that the comparison ignores the crucial difference
between mathematical concepts and pure concepts of the understanding. This
difference consists in the fact that the former can be constructed, i.e.,
presented in pure (formal) intuition, while the latter cannot. Indeed, it was
this very heterogeneity between pure concept and intuition which generated
the problem of the schematism in the first place. Thus, the basic justification
for characterizing the realized (constructed) mathematical concept as a pure
(formal) intuition seems to be totally lacking with regard to the schemata of
the pure concepts of the understanding.
In order to deal with objection, it is necessary to consider the account of
pure intuition which Kant provides in the Transcendental Aesthetic . As
already indicated, it is consistently correlated with the notion of form. The
differences between the various characterizations turn on the different ways in
which this latter notion is construed. Thus, in the preliminary definitions,
"pure intuition" is initially equated with "form of sensibility".29 But since
the defining characteristic of empirical intuition is the presence of sensation, a
form of sensibility is likewise termed a "form of empirical intuition". Cor-
rectively, since "appearance" is defined as the "undetermined object of an
empirical intuition ", a pure intuition can also be designated as a "form of ap-
pearance".
In the Transcendental Aesthetic and elsewhere Kant seems to treat these
expressions as equivalent, albeit without providing anything resembling a
satisfactory explanation. Fortunately, he does at least provide some hint as to
the meaning of 'form'. This hint is to be found in the "conclusions" which
Kant draws from the "Metaphysical" and the "Transcendental Exposition"
of space and time. Kant there maintains that space and time are to be assigned
the metaphysical status of forms of appearance (and its equivalents) on the
grounds that they have been shown in these expositions to function as condi-
tions of sensible intuition.30 As the context indicates, 'condition' has a strictly
epistemic sense. It refers to that which makes possible or is presupposed by a
certain kind of representation. For example, space is held to be a condition of
outer appearances because it is only by reference to the representation of
29 Critique of Pure Reason , A20/B34-45.

30 Ibid., A26/B42, A34/B50.
space that objects can be experienced as external to the self and to each
other.31 Since we are now concerned only with Kant's terminology, we need
not consider the cogency of this argument. The crucial point is simply that
'form', at least in this context, has essentially the same meaning as 'condi-
tion' (taken in the epistemic sense). Moreover, it follows from this that
transcendental schemata can be construed as pure intuitions if they can be
shown to function as forms or conditions of sensible intuitions.
Kant certainly seems to affirm such a function for transcendental
schemata when he characterizes them as "formal conditions of sensibility".
(A140/B179) Even apart from this, however, it should by now be clear that
this is precisely the function that is assigned to transcendental determinations
of time. As conditions of empirical time determination, they are certainly con-
ditions in a different sense than space and time themselves. The latter are
general forms or conditions of sensibility, that is, conditions under and with
reference to which the data of empirical intuition are given to the mind; while
transcendental determinations of time are specific temporal conditions of ac-
tual empirical intuitions (empirical time determination). Nevertheless, they
are still conditions of empirical intuition and, therefore, pure intuitions in the
Kantian sense.
Ill
This brings us to the question of how the interpretation of transcendental

schemata as pure intuitions helps us to understand their role in the synthetic a
priori judgments of a "metaphysic of experience", and thus the function of
the schematism chapter in the Transcendental Analytic. As already suggested,
the gist of the answer lies in the recognition that synthetic a priori judgments
involve the predication of pure concepts of pure intuitions. Construed as pure
intuitions, transcendental schemata would thus seem to be attractive can-
didates for the role of subjects of such judgments.
As we have also noted, in his own account of the nature of transcendental
schemata, Gram construes the notion of pure intuition in the first of the two
senses which have been distinguished. This leads directly to his rejection of the
"third thing" characterization of the nature of a transcendental schema.
Although I cannot pursue the matter here, it can easily be shown that it also
31 Ibid., A23/B38. Strictly speaking Kant's argument here is designed to show that the
representation of space is a priori. He endeavors to do this, however, by showing that it is a condi-
tion of the possibility of outer experience.
74 Henry E. Allison
underlies his central criticism of the actual arguments of the Analogies.32 In

contrast to Gram, I have argued not only that we must distinguish between the
two senses of pure intuition, but also that transcendental schemata must be
construed as pure intuitions in the second sense. A natural corollary of this is
that pure intuitions must likewise be taken in this sense, if they are to function
as the subject of synthetic a priori judgments involving the categories.
The specific judgments at issue here are contained in the Principles of Pure
Understanding . An adequate treatment of this topic would, therefore, require
a detailed consideration of these principles, a task that is obviously beyond the
scope of this paper. Nevertheless, even a cursory glance at this section of the
Critique suffices to indicate that the basic difficulty confronting such an inter-
pretation. The problem is posed by Kant's contention that, in these principles,
appearances are " subsumed, not simply (schlechthin) under the categories,
but under their schemata". (A181/B223). As an illustration of this, we need
look no further than the First Analogy in the First Edition which asserts: "All
appearances contain the permanent (substance) as the object itself, and the
transitory as its mere determination, that is, the way in which the object
exists." (A182) Since that which is subsumed under a concept in a judgment is
the subject, while the concept under which it is subsumed is the predicate, it
seems that the real subjects of such judgments are appearances (not
transcendental schemata). Moreover, since the schema is here said to function
as a predicate, Kant must be equating it with the schematized category. Put-
ting these factors together, we arrive at a more or less standard reading,
wherein the principles apply the schematized categories to appearances.
How, in the face of this, can the original claim be maintained? Before we
are in a position to deal with this question, two points must be noted. The first
is that the predication of a concept of an intuition, or, to invoke Kans own
manner of expression, the relation of the former to the latter in a judgment, is
really equivalent to the determination of the latter by the former.33 The se-
32 Cf. Gram, op. cit., p. 139-140 where he writes with respect to the First Analogy: "Kant set
himself the task of showing that a category has something corresponding to it in pure intuition.
And that pure intuition is too impoverished to instantiate the concept Kant wants to instantiate
does not show the absence of the covert theory of syntheticity in the argument. It shows only that
the covert theory makes demands on pure intuition which it cannot fulfill." By the "covert
theory" Gram means the theory that synthetic judgments predicate pure concepts of pure intui-
tions. The reason why pure intuition cannot fulfill the demand which, according to Gram, this
theory places on it (providing a referent for pure concepts) is that he construes pure intuition in
the first, indeterminate sense.
33 Admittedly, Kant frequently refers to acts of determination and seldom, if ever, to acts of
predication. Nevertheless, I feel that this interpretation is justified by the fact that such deter-
mination always involves the application of concepts, and that Kant explicitly identifies real (as
opposed to logical) predicates with determinations. Passages wherein this occurs include: Critique
of Pure Reason A578/B626 ff., Lectures on Logic , 36, note 1 , Kants gesammelte Schriften , Vol.
IX, p. Ill; Reflexionen 4634, Vol. XVII, pp. 616-617, and 4795, Vol. XVII, p. 731.
cond is that the determination or "objectification" of time by means of the

transcendental synthesis of the imagination does not involve making time
itself into an object, but rather the determination of the objective relations of
appearances in time. In other words, one determines time transcendentally by
relating appearances to one another in certain determinate ways.
Given this, one would expect that a judgment which asserts such a deter-
mination of time, and which thus predicates a pure concept of a pure intui-
tion, would consist in a claim about certain universal and necessary
characteristics of appearances qua temporal. More specifically, it would not
assert that time itself is necessarily x (where V stands for any categorial
feature, e.g., permanence), but that things in time are necessarily x, or involve
some element of xness. Now this is exactly what one finds in the Principles of
Pure Understanding, particularly in the Analogies. Each of these Analogies
makes a claim to the effect that some element of xness (a transcendental time
determination) functions as a condition (form) of empirical time determina-
tion. For example, permanence, the schema of the concept of substance, is
claimed to be a necessary condition for the experience of duration (the
magnitude of time); while rule governed succession, the schema of the concept
of causality, is claimed to be a necessary condition for the experience of an ob-
jective sequence of events.
Thus, despite their references to appearances, these judgments really refer
to the transcendental schemata; for they assert that the schemata function as a
priori conditions of time consciousness and, therefore, of experience.
Moreover, since, as we have already seen, a judgment which makes such an
assertion, can be said to predicate a pure concept of a pure intuition, this can
also be said of these judgments. In order to realize this, however, it is
necessary to distinguish between what can appropriately be called the "sur-
face" and the "deep" structure of these judgments. Their surface structure,
as well as some of Kant's (unfortunately) most influential accounts of such
judgments, suggest that they involve the relation of one concept (pure) to
another concept (empirical), e.g., every alteration has a cause. A cursory
glance at the Principles of Pure Understanding indicates that they involve the
application of the schematized categories to appearances. A somewhat deeper
analysis, however, reveals that what is actually being asserted in such
judgments is the determining function of the schema of the concept. Finally,
carrying this analysis one step further, we note that this involves the deter-
mination of time by the corresponding concept, and that this is equivalent to
the predication of a pure concept of a pure intuition.
76 Henry E. Allison
IV
The preceding account of the "deep structure" of synthetic a priori

judgments suggests a problem which Kant himself seems hardly to have con-
sidered, but which must be resolved, if the program of the Transcendental
Analytic is to be defended. The specific problem which I have in mind is the
explanation and justification of the connection, which Kant simply
dogmatically asserts, between the various categories and their corresponding
schemata.34 This is, of course, a question that must be faced on any inter-
pretation of transcendental schemata. It becomes particularly pressing,
however, given the systematic function which has here been assigned to the
schematism.
There are two passages in which Kant can be taken as addressing himself
to this problem. The first is from the general introduction to the Analytic of
Principles . In commenting on the "peculiarity" of transcendental
philosophy, he there notes that "besides the rule (or rather the universal con-
dition of rules), which is given in the pure concept of understanding, it can
also specify a priori the instance to which the rule is to be applied ". In light of
this, he further reflects: "It (transcendental philosophy) must formulate by
means of universal but sufficient marks the conditions under which objects
can be given in harmony with these concepts." (A134-136/B174-175) The se-
cond passage is from the schematism chapter itself. Here Kant notes curtly
"that pure a priori concepts, in addition to the function of understanding ex-
pressed in the category, must contain a priori certain formal conditions of sen-
sibility, namely those of inner sense". (A140/B179) As already noted in a
previous reference to this passage, these "formal conditions of sensibility"
are transcendental schemata.
Unfortunately, neither of these passages helps very much in the resolution
of the problem. Both presuppose the general result of the Transcendental
Deduction, viz., that the pure concepts stand in relation to time by means of
their connection with the transcendental synthesis of the imagination. Given
this result, Kant seems to be claiming that it must be possible to specify the
schema corresponding to the category or, equivalently, that the category
somehow "contains" its schema (the formal condition of sensibility). But he
does not offer a trace of justification for this claim, and he provides no ac-
count of how one can ever establish the connection between a given category
34 In fact, Kant only lists eight schemata for the twelve categories, providing a separate
schema for each of the categories of relation and modality but only one schema each for the three
categories of quality and of quantity. A plausible interpretation of this is given by Paton, op. cit.,
Vol. II, pp. 63-64.
and its schema. In order to pose the problem in its sharpest form, we need on-
ly ask about the nature of a judgment which asserts such a connection. The
fact that Kant merely lists the schemata in connection with the various
categories might suggest that it must be analytic. But the utter heterogeneity
between the intellectual and the sensible, which is precisely what generates the
need for a schematism in the first place, makes it clear that this cannot be the
case. Equally clearly, it cannot be synthetic a posteriori . This would make the
connection between category and schema a contingent matter. It must,
therefore, be synthetic a priori . This, however, leads to the paradoxical result
that the whole problem of the synthetic a priori breaks out within the doctrine
of the schematism, even though this doctrine is intended as an essential step in
the resolution of this problem.
An initially plausible way of dealing with this problem is in terms of the
familiar distinction between pure and schematized categories. Thus, one
might argue that, since the pure concepts stand in no connection with time, no
schemata can be supplied for them; but that, since the schematized categories
already stand in connection with time (as rules for the transcendental synthesis
of the imagination), the connection between these categories and their
schemata can be determined analytically. For instance, if, with Paton, we
define the pure category of substance as 4 'the concept of the synthesis of sub-
ject and predicate", and the schematized category as the concept of the syn-
thesis of the permanent and the changing in time"35 then, while it would be
impossible to provide a schema for the former, it would be a trivial matter to
do so for the latter.36
As tempting as such a line of argument might at first appear, it simply will
not do. Even assuming the distinction between pure and schematized
categories, which Kant himself never explicitly draws,37 the obvious difficulty
with this move is that it only serves to push the problem back one step. For if
the connection between schematized category and schema is viewed as un-
problematic, then the locus of the problem shifts to the connection between
the pure and the schematized categories. Thus, to continue with Paton's
35 Paton, op. cit., Vol. II, pp. 52-53.

36 According to Paton, the schema is simply the product of the synthesis according to the
schematized category, op. cit., p. 53 and more generally pp. 42-43.
37 This distinction is almost a commonplace in the literature and they are frequently treated
as if they were two distinct sets of concepts. This is certainly true of Paton's account and even
more so of Gram's. For the latter, see especially op. cit., pp. 126-127. I do not believe, however,
that this has any warrant in the text. If one wishes to distinguish between pure and schematized
categories, then the distinction should not be construed as between two sets of concepts, but bet-
ween two functions (judgmental and perceptual or experiential) that are exercised by the pure
concepts. Although I cannot pursue the matter here, these two functions correspond to the two
parts of the Transcendental Deduction in the Second Edition.
7 8 Henry E . Allison
distinction between the pure and the schematized categories of substance, we

are led to ask about the connection between the concept of the synthesis of
subject and predicate (which concerns the relation between representations in
a judgment) and the concept of the synthesis of the permanent and the chang-
ing in time (which concerns the relation between entities). It is easy to show
that this connection must be both synthetic and a priori; so that we are back to
our original question.38
We must, therefore, find the "ground " of the synthetic a priori judgment
connecting category and schema. In order to achieve this goal, we must go
beyond what Kant explicitly tells us, but it may still be possible to do so on the
basis of materials which the Transcendental Analytic provides. Two points are
here of particular relevance: 1) that the categories, as rules for the
transcendental synthesis of the imagination, serve to determine time (the form
of inner sense), and 2) that the schema is in each the product of such deter-
mination. Assuming these premises, I believe that it is possible to construct a
plausible case for the schemata which Kant assigns to the categories. A really
adequate treatment of this topic is here impossible; and in what follows I do
not pretend to be doing more than providing the outlines or strategy for such a
treatment. To this end I shall limit my consideration to the modal categories,
where the connection between category and schema is easiest to show, and to
the categories of substance and causality, where it is most important.
The modal categories and their respective schemata are: possibility ("the
agreement of the synthesis of different representations with the conditions of
time in general"); actuality ("existence in some determinate time "); necessity
("existence at all times"). In his recapitualtion of these schemata, Kant adds
by way of a general comment: "/T/he schema of modality and of its
categories is time itself as the correlate of the determination whether and how
it belongs to time." (A145/B185) Both the specific schemata and the general
comment strongly suggest that what Kant has provided us with here is a set of
translations of logical into real modalities. To claim that something is really ,
not merely logically, possible, is to claim that it is subject to the conditions of
time (which are equivalent to the conditions of possible experience). Similarly,
38 Jonathan Bennett, Kans Analytic, Cambridge, 1966, p. 151 suggests that "The schema
of any category, then, is just the category itself with the condition of temporality added". But
then our question becomes: how, in each case do we determine the "sum " of category and condi-
tion of temporality?
to claim that it is actual is to claim that it exists at some determinate time.39

Finally, to claim that it is necessary is to claim that it exists at all times.
Such translations, and the same can be said for all of the schemata, pro-
vide grounds for determining whether appearances are in accord with the
categories.40 In this sense they can be construed as rules for the application of
rules (the categories), and this constitutes the "moment of truth" in the
prevalent "rule theory" of the schematism.41 The key point, however, is that
they can only function as such grounds or rules, if it is already assumed that
they express the result of the translation of the categories into temporal terms.
For example, only if I already know that the temporal and hence experiential
translation of 'actuality' must be "existence at a determinate time", can I use
this as a basis for determining whether or not something is actual. In each case
it is the initial claim that is crucial. Thus, rather than simply describing
transcendental schemata as rules, it is less misleading to say that they can serve
or function as rules.42
Assuming that the translation must be into temporal terms, the connection
between the modal categories and their schemata seems to be intuitively ob-
vious. Nevertheless, the necessity for a translation into an exclusively tem-
poral language is itself controversial. Has not Kant here, as well as with the
other schemata, unjustifiably ignored space? And is it not necessary that the
categories have spatial as well as temporal schemata? The response to this fre-
quently expressed objection is a simple no. It rests upon the failure to
distinguish between the sphere of objects to which and the necessary and suffi-
cient conditions under which the categories are to be applied. The sphere in-
cludes spatial as well as temporal entities, and thus all objects of possible ex-
perience. Nevertheless, the categories relate to these objects in virtue of their
temporality. This follows directly from the role given to time as the universal
form of all appearances (since all appearances are "modifications of inner
sense"). Nor can we say that Kant moved away from this position in the Se-
39 One might object to this analysis on the grounds that the notion of logical actuality
(logische Wirklichkeit) is vacuous, and thus that in the case of actuality there can be no such con-
trast. Nevertheless, Kant makes use of this notion to express the groundedness or objective validi-
ty (transcendental truth) of a judgment. Cf. Kant's letter to Reinhold, May 19, 1789, Kants
gesammelte Schriften , Vol. XI, p. 47, and Reflexion 2181, Vol. XVI, p. 261 . The notion is discuss-
ed by Klaus Reich, Die Vollstndigkeit der kantischen Urteilstafel (Berlin 1932), p. 44, 59-60, and
by Rainer Stuhlmann-Laeisz, Kants Logik (Berlin, 1976), p. 63.
40 Cf. Schapere, op. cit., p. 272.
41 The ' 4 rule theory" is already presented by Curtius. In the recent literature its most
rigorous advocate is Robert Paul Wolff, Kant's Theory of Mental Activity, Cambridge, Mass.
1963, pp. 206-223.
42 Gram provides a convincing critique of the "rule theory" along similar lines, op. cit.,
pp. 95-100.
80 Henry E. Allison
cond Edition, with its increased emphasis on space.43 To be sure, Kant does
there assert that in order to establish the objective reality of the categories we
need not merely intuition, but outer intuition. (B291) The point, however, is
that outer intuition is seen to be necessary to express the temporal determina-
tions contained in the schemata. Consequently, the fact that we need outer in-
tuition in order to experience something permanent (the schema of substance)
does not at all affect the status of this schema as a transcendental determina-
tion of time.
This brings us to the more difficult task of "deducing" the schemata of
the categories of substance and causality. Taking our clue from the brief
discussion of the modal categories and their schemata, we can assume that this
"deduction" must consist in the determination of the temporal expression or
translation for the intellectual function which is thought in the pure concept.
However, before we can consider what can count as the temporal expression
or translation of a particular intellectual function, we must first determine
what in general is meant by an intellectual function. This, in turn, requires a
step back into the morass of the Metaphysical Deduction.
Although Kant tends to speak of the intellectual function thought in the
category in connection with the "manifold of an intuition in general", what
he really means by this is its function with respect to judgment. In determining
this function, we abstract from any consideration of the content and reflect
merely on the form of judgment. Kant, as we all know, thought that he had
provided an exhaustive list of these forms, and with it of the "functions of
understanding". Fortunately, we need not deal with this issue here. The im-
portant point is only that he attempted in the Metaphysical Deduction to
establish a direct correspondence between certain forms of judgment and cer-
tain concepts (intellectual functions). Thus, judging under the categorical
form entails determining the content (the given manifold) in terms of the
logical relation of inherence and subsistence (the pure concept of substance);
while judging under the hypothetical form entails determining the content in
43 This is argued for by Gregg E. Franzia, "Space and the Schematism", Kant-Studien, 69,
pp. 149-159.
Transcendental Schematism and The Problem of the Synthetic A Priori 8 1
terms of the logical relation of ground and consequent.44 Consequently, these

categories can be characterized as the conceptual conditions for the deter-
mination of the manifold of any intuition, if it is to be subject to the
categorical and hypothetical forms of judgment respectively. Assuming this, it
follows that in order to provide the schemata of these categories, it is
necessary to establish the conditions under which a temporal manifold must
be determined, if it is to be subject to the categorical and the hypothetical
forms of judgment.
Let us begin with the pure concept of substance. As indicated above, what
we must do is to consider how a specifically temporal manifold is to be deter-
mined, i.e., conceptualized, if it is to be subject to the categorical form of
judgment. This is equivalent to specifying the conditions under which one can
say of anything temporal that it is a real subject of accidents (rather than a
merely logical subject of predicates). Perhaps the most obvious candidate for
such a necessary condition is reidentifiability. Only something that is reiden-
tifiable throughout a change of states can be distinguished from one or more
of these states and considered as their "real subject", that is, as something to
which these states pertain as modifications or in which they "inhere" as ac-
cidents. But in order for something temporal to be reidentifiable, it is
necessary that it endure or be permanent. Strictly speaking, of course, this on-
ly requires a relative permanence and this generates a problem that is crucial
for the First Analogy. For our present purpose, however, we can rest content
with the result that at least a relative permanence is a necessary characteristic
of any phenomenal object that can be conceived of as the "real subject" of
change or accidents. As such, it is the sensible condition under which the
44 What follows rests largely on an interpretation of the Metaphysical Deduction which, un-
fortunately, I cannot even attempt to justify here. This interpretation essentially involves two
points: 1) The claim for which I argue in The Kant-Eberhard Controversy , pp. 63-67, that Kant's
account of judgment shows the necessity for recognizing a set of "pure" concepts which are
presupposed rather than produced by the activity of judging. 2) Kant's own claim that the
categories are just the logical functions of judgment insofar as they are employed in the deter-
mination of the manifold of a given intuition, (B143), Prolegomena , 39, and Metaphysische An-
fangsgrnde der Naturwissenschaften , Kants gesammelte Schriften , Vol. IV, pp. 474-475, note.
The former establishes that there must be pure, i.e., a priori concepts; the latter is the basis for
determining the identity of these concepts. I construe Kant's identification of the categories with
the logical functions of judgment to be equivalent to the claim that to judge under a certain form,
e.g., hypothetical^, is just to conceptualize the given data in terms of the relation of ground and
consequent. This interpretation has some affinities with the brief yet suggestive account of Arthur
Melnick, Kant's Analogies of Experience, Chicago and London, 1973, pp. 30-42. My differences
with Melnick largely concern the details of his account of an "epistemic concept" and the ques-
tion of the relationship between the Metaphysical and the Transcendental Deduction. These are
not, however, directly relevant to the present concern. Unfortunately Melnick fails to deal at all
with the schematism.
82 Henry E. Allison
category of substance can be applied in a judgment to objects of possible ex-

perience and thus the schema of the concept. As Kant himself puts the matter:
If I leave out permanence (which is existence in all time), nothing re-
mains in the concept of substance save only the logical representation
of a subject - a representation which I endeavor to realise by
representing to myself something which can exist only as subject and
never as predicate. But not only am I ignorant of any conditions under
which this logical pre-eminence may belong to anything: I can neither
put such a concept to any use, nor draw the least inference from it. For
no object is thereby determined for its employment, and consequently
we do not know whether it signifies anything whatsoever.
(242-243 /B300-301 ) .
Similar considerations apply to the pure concept of causality (ground and
consequent), which is at work in the hypothetical form of judgment.45 As
before, if the judgment is to be applicable to a temporal manifold, it must be
possible to specify the conditions under which successively given items in this
manifold can be connected in judgments of this form. Clearly, their mere suc-
cessiveness will not suffice; for as Kant well knew, Hume had shown that the
mere fact that A precedes B in time (even invariably) does not justify the claim
that A is the cause or ground of B. What is needed in order to make judgments
of this form about temporally given items is some notion of necessary or rule
governed succession. Consequently, Kant is essentially following Hume when
he tells us that the schema of causality "consists in the succession of the
manifold, insofar as that succession is subject to a rule". (A144/B187)
Admittedly, this "deduction" seems to be subject to an obvious objec-
tion. For the claim that rule governed succession is the schema of the concept
does appear to entail the claim that the cause must invariably precede the ef-
fect in time. This is readily explicable in terms of Kant's reliance on Hume,
but it does not help us with the nagging question of what to do in those cases
where the cause and effect are simultaneous. Kant was, of course, well aware
of this problem, and he attempted to deal with it in the Second Analogy by
means of the distinction between the order and the lapse of time. An examina-
tion of Kant's argument would be necessary in order to provide an adequate
treatment of his identification of the schema of the concept of causality with
rule governed succession. Such an examination cannot be undertaken here.
Even apart from it, however, we can see that Kant's claim is at least plausible;
and we know what would be required to make it more than that.
45 Following Melnick, op. cit., p. 39, the hypothetical form of judgment for Kant is here
construed as the non-material hypothetical.
I have tried in this section both to indicate and to make some suggestions
as to how one might fill a significant lacuna in the argument of the
Transcendental Analytic . Kant claims that, given the category and given time
as the form of inner sense, it is possible to provide the schema or the sensible
condition for the realization of the category. But he does not give us any clue
as to just how this is to be done. Instead, he contents himself with hinting
darkly about 4 4 an art concealed in the depths of the human soul". I hope that
the preceding account has shed a bit of light on this 44 art In any event, I do
believe that I have pointed to a line of argument that is required by the overall
program of the Transcendental Analytic and that is distinct from the
arguments of both the Transcendental Deduction and the Principles of Pure
Understanding . The former merely argues in general terms that the concepts
identified in the Metaphysical Deduction have objective reality in virtue of
their connection with the transcendental synthesis of the imagination. The lat-
ter attempts to prove that certain sensible conditions (the schemata of these
concepts) constitute the universal and necessary (formal) characteristics of a
temporally ordered world. In order to link these two, it is necessary to link the
categories with their schemata. This is, or ought to be, the systematic function
of the schematism chapter in the Critique of Pure Reason .
Dialctica Vol. 35, N 1-2 (1981)

Esquematismo Trascendental

Uploaded by

Document Information

Original Description:

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

Esquematismo Trascendental

Uploaded by

Copyright:

Available Formats

Transcendental Schematism and The Problem of the Synthetic "A Priori"

Author(s): Henry E. Allison

Cet article tudie la connexion entre la conception kantienne du schmatisme transcendental

* University of California, San Diego.

Dialctica Vol. 35, N<> 1-2 (1981)

In a letter to Reinhold, wherein he complains of Eberhars attack on the

Understanding predicate pure concepts of pure intuitions. Finally, I shall sug-

We naturally feel a preliminary difficulty with respect to the existence

4 H. A. Prichard, Kant's Theory of Knowledge, Oxford, 1909, pp. 77-82.

5 G. J. Warnock, "Concepts and Schematism", Analysis IX, 1949, pp. 77-82.

8 Cf. Patn, op. cit. Vol. II, p. 67.

determined in respect to a logical function and therefore stands under a

10 Kants gesammelte Schriften , Vol. XVIII, p. 392, Reflexion , no. 5133.

indicates, Kant is using 'subsumption ' as a synonym for 'application '. 15

attempt to understand how such application is possible. As already noted, the

What, then, is a transcendental schema, such that it is capable of playing

17 See my The Kant-Eberhard Controversy, esp. p. 60.

18 Kants gesammelte Schriften , Vol. V, p. 68.

21 Ibid., pp. 91-94.

of the hints contained here and, more particularly, by a consideration of what

perceived. Consequently, a transcendental determination of time must be

28 Paton, op. cit., Vol. II, pp. 28-30.

of something permanent is just to think of it as lasting through time. The con-

29 Critique of Pure Reason , A20/B34-45.

This brings us to the question of how the interpretation of transcendental

underlies his central criticism of the actual arguments of the Analogies.32 In

cond is that the determination or "objectification" of time by means of the

The preceding account of the "deep structure" of synthetic a priori

35 Paton, op. cit., Vol. II, pp. 52-53.

distinction between the pure and the schematized categories of substance, we

to claim that it is actual is to claim that it exists at some determinate time.39

terms of the logical relation of ground and consequent.44 Consequently, these

category of substance can be applied in a judgment to objects of possible ex-

Dialctica Vol. 35, N 1-2 (1981)

You might also like