. 4
( 10)


I notice that the willow and the linden differ in respect to their trunks, this
synthesis occurs with recognition, the recognition of the respect in which
the trees differ. I synthesize the two representations by intending that they
differ regarding their trunks: The recognition accomplishes the synthe-
sis. In both the A and B Transcendental Deductions, Kant argues that
such a recognizing consciousness is possible only insofar as it is possible
for the subject to be conscious of the act in which the representations are
synthesized.21 And in both editions, this ability in turn is seen to depend
upon the ability to re¬‚ectively relate one™s representations with the unity
of consciousness. Thus human cognition (Erkenntnis), even in its intuitive
form in which we intuit the differences among things in some regard or
other, requires acts of recognition (erkennen), and acts of recognition
always involve a relation to the unity of consciousness.
So for Kant, to carry out a “comparison of representations in relation
to the unity of consciousness” is to represent two representations together
so as to be able to note their similarities and differences in some respect
or other. In the note to Section 6 of the J¨ sche Logic, Kant tells us that
“to make concepts out of representations one must be able to compare.”
That is, in order to intend the concept ˜trunk™, for example, one must
be able to compare representations of trees with trunks so as to come to
note the respect in which they are both similar and different. Since this
comparison is necessary for generating or ˜making™ the concept ˜trunk™,
the representation in which the comparison is carried out cannot itself
be fully conceptual. The suggested origin of the concept demands that I
be able to compare the willow and the linden in respect of their trunks
even while I lack the concept ˜trunk™. Thus, what is required in order to gen-
erate the empirical concept ˜trunk™ is the intuitive ability to compare,

21 Cf., for example, A103 and B130.
Mark Okrent

in relation to the unity of consciousness, the intuitions of, say, a willow
and a linden. Such an act of comparison involves a relation to the unity
of consciousness insofar as, as a synthetic act, it essentially involves the
possibility that the agent of the act can become conscious of the activity
in which it generates the unity of the two intuitions in the comparison,
and thus can become conscious of the rule, or concept, it follows in per-
forming the act of comparison. Although humans need not be explicitly
conscious of the concepts that are implicitly operative in their intuitive
acts of comparison, the fact that those comparisons are carried out in
relation to the unity of consciousness implies that they are always capable
of forming the discursive judgments that are intuitively made present in
acts that compare intuitions.
Animals, according to Kant, lack the ability to generate concepts
or form discursive judgments. As such, they also lack the ability to be
acquainted with something with consciousness. They lack the ability to note,
or notice, the respects in which things differ. Kant thinks this because
he believes that it is a corollary of his observation that animals lack the
capacity to judge. For Kant, animals can intuitively grasp differences and
similarities among things, but they can never intend the respects in which
things are similar or different. Animals, for Kant, can note that two objects
differ, but they can never note or notice how they differ, or the way in
which those objects differ. That is, for Kant, animals can never become
conscious of the unity of the act of comparison in and through which the
synthetic representation (in which the representations of the willow and
the linden are compared) is produced. And this is precisely the respect
in which animal and human intuition differs for Kant. While animals do
compare intuitions, they lack the type of intuition necessary to form con-
cepts, and thus can never represent the respect in which they carry out
this comparison.
Unfortunately, this is no solution to the problem of animal sapience. It
only regenerates the same aporia we have been following right along. In
addition to representations of the items involved in the comparison, every
act of comparison, whether animal or human, as an act of comparison,
must involve an intention directed toward the relation between the items
intended, their similarity or difference. On Kant™s account, animals do
note similarities and differences between things. Mac does not only see
the house and the bus differently. He also sees their difference, even if
he can never intend any particular way in which they differ and can
never become conscious of the speci¬c differences between them. And
such an animal act of comparison necessarily involves a synthesis of the
Acquaintance and Cognition 105

representations that are compared in the act. Now either such a synthetic
act necessarily involves the possibility on the part of the agent of the act
to engage in the type of re¬‚ection that Kant calls consciousness, and thus
the ability to use concepts, or it does not. If it does, then animals can™t
be acquainted with objects, because they can™t perform comparisons,
and for Kant this is the minimal ability necessary for acquaintance with
objects. That is, they can™t have intuitions in any sense at all. If such
synthetic acts don™t require the ability on the part of the agent to re¬‚ect
and form concepts, then such abilities are not necessary for intentions
directed towards objects. So Kant™s “two kinds of intuition” solution to
his problem can™t possibly work.

4. conclusion
What has gone wrong? The short answer to this question is that the prob-
lem of animal sapience points to a deep problem in the way in which Kant
tends to report his own results. As I emphasized earlier, it is absolutely cen-
tral to the Kantian project that we see what we judge, that the objects we
perceptually encounter have, and must have, the same structural charac-
ter as the objects about which we form judgments. It is only for this reason
that Kant can conclude that we can have a priori knowledge of objects
of possible experience, that is, a priori knowledge of the objects that we
can intuit. We can know that the pure concepts of the understanding can
validly be applied to the objects of intuition only because the forms of
unity that make it possible to intuit an object are the very same forms
of unity that allow us to judge that object: “The same function which
gives unity to the various representations in a judgment also gives unity
to the mere synthesis of representations in an intuition. . . .” Kant assigns
to the faculty of imagination the ability to relate or synthesize sensible
presentations so as to present us with the intuition of an object. As we are
presented with intuitions of objects (and even more fundamentally with
the pure intuitions of space and time in which the intuitions of objects
are arranged), and we are not conscious of any activity on our part as
necessary for such presentations, Kant says that imagination is a blind,
unconscious faculty. The operation of this faculty thus does not require
that we be conscious of its operation.
At this point, however, we can ¬nally see what has gone wrong with
Kant™s attempts to solve the aporia. In the footnote to Section 6 of
the Logic, Kant says that by comparing the willow and the linden with
one another, I note that they are different from one another in regard
Mark Okrent

to . . . For such a noticing to be possible, it certainly must be possible for
me to re¬‚ect on the way in which I represent the trees together in the
comparison. And so for me to be able to generate and apply concepts in
judgments, I must be capable of becoming conscious of the respect in
which the comparison is carried out, that is, self-conscious of the unity of
the act of synthesis through which the act of comparison is performed.
Without the possibility of becoming conscious of the unity of the act in
which I generate the comparison with respect to sameness and differ-
ence, the synthesis of recognition in a concept is “altogether impossible.”
But, strictly, this fact provides us with no reason to think that the ability
to notice the respects in which I compare the representations of trees is
necessary for me to be able to compare the representations of trees. On Kant™s
own account, my dog Mac can and does compare his sensible represen-
tations of trees in regard to their sameness and difference, but lacks the
ability to ever intend how they are different. So the second, recognizing,
ability cannot be necessary for the ¬rst ability, the ability to be acquainted
with difference and similarity. On the other hand, it is quite implausible
to believe that a subject could have the ability to represent or notice the
respect in which trees are the same or different without also having the
ability to intend that they are the same or differ. If I were not intuitively
acquainted with the trees in such a way that I could in some manner
represent their being similar or different, if I lacked the representational
ability that Kant calls kennen, then I could not form concepts. But Kant
has given us no reason to believe that such kennen requires the possibility
of erkennen, the capacity to note the respect in which the comparison is
carried out. Before we can be rational creatures who possess the discur-
sive capacity to judge regarding the respects in which objects differ and
are the same as one another, we must be animals who possess the intu-
itive ability to represent the similarity and difference of objects. That is,
we must be able to see those similarities and differences.
In effect, the aporia arises out of Kant™s failure to distinguish rigorously
between two distinct capacities. The capacity to have a unitary represen-
tation of the outcome of an act of synthesis, by having a single represen-
tation of that which is represented through that act, is logically distinct
from the capacity to represent the act itself, which generates such a uni-
tary representation. Kant™s willingness to admit that animals can compare
things in respect to their similarity and difference, although they lack the
ability to form and apply concepts, displays the fact that having the ability
to intend the character of the act that generates a synthesis of represen-
tations “ the ability that is necessary for the use of concepts “ is in no
Acquaintance and Cognition 107

way necessary for the generation of such a synthesis. And for this rea-
son, the re¬‚ective capacity to attach the ˜I think™ to our representations,
while necessary for those representations to be anything for me, is not a
necessary condition for our being able to represent objects as objects.
Kant argues that it is only in the representation that is the concept,
the representation of the unity of the act of synthesis, that the act of
synthesizing various representations is in fact carried out.22 That is, the
unitary representation of the comparative relation between the two trees
is the conceptual, recognizing representation of the manner in which
they are the same and different. This is the cash value of the crucial claim
that without “such [re¬‚ective] consciousness . . . cognition of objects is
altogether impossible.” But insofar as it is the case that there can be
intuitions that represent objects, and it is possible for non-concept-using
animals to be acquainted with the similarities and differences of these
intuitively presented objects, this just can™t be right. And we are given no
independent reason for thinking that the ability to recognize the manner
in which things are similar and different is necessary for being acquainted
with their being similar and different.
Having said this, it does not follow that we have a good grasp on what is
involved in having a nonconceptual intuitive grasp of difference and sim-
ilarity. The fact that my dog is acquainted with objects, although he lacks
our facility with concepts, guarantees that such nonconceptual intuition
of similarity and difference occurs. Fully understanding what is entailed
by this fact regarding the nature of both conceptual and nonconceptual
cognition, however, is another matter entirely.
So we have come to a perhaps surprising conclusion. We started with
an inconsistent triad of Kantian views: 1. Intuitions involve a reference
to an object. 2. Animals, although they lack the ability to apply concepts,
have intuitions. 3. Cognition of objects requires a unitary consciousness
of the act through which a manifold is combined and the ability to apply
concepts. The surprising conclusion we have reached is that to resolve this
inconsistency Kant should give up (3), the claim that cognition of objects
requires the possibility of conceptual, judgmental cognition. Discursive
understanding is only possible on the basis of the intuitive presentation of
objects, the possibility of such an understanding is not necessary for the
possibility of such an intuition, and it is only possible to articulate what is

22 Kant, of course, explicitly adopts this position in the B edition Transcendental Deduction,
most notably in Section 26, where he asserts that the ¬gurative transcendental synthesis
of the imagination is the effect of the understanding on the imagination.
Mark Okrent

involved in discursive understanding by ¬rst articulating what is involved
in intuitively presenting objects. I take it that this is what Heidegger meant
when he said that “intuition is the original building site of all knowledge,
to which all thinking is directed as a means.”23
What does Mac see when he sees a bus? He sees an object, of course.
And the object that he sees is an object that he would describe as a big,
smelly, moving, noisy one; that is, he would describe it in that way if he
could describe it at all. Because he would be right to describe that object in
that way, if he were able to describe it, he sees that this object is different
from that other object that he would describe as a house if he could
describe it. And because he sees this difference, he responds to them in
quite different ways, even though he could never say, or judge, how these
objects are different. Mac also sees the difference between me and the
bus. But, this is quite a different difference from the difference between
the bus and the house, and Mac is aware of this as well, thank goodness.

23 Heidegger (1997), 58.
part two


Dialogue: Paul Guyer and Henry Allison on Allison™s
Kant™s Theory of Taste

remarks on henry allison™s kant™s theory of taste by
paul guyer
In Kant™s Theory of Taste: A Reading of the Critique of Aesthetic Judgment,1
Henry Allison has provided a detailed study of the two introductions and
the ¬rst half of Kant™s Critique of the Power of Judgment to stand along-
side his previous volumes on Kant™s theoretical and practical philosophy,
his Kant™s Transcendental Idealism of 1983 and Kant™s Theory of Freedom of
1990.2 Allison is as deeply committed to the unity and coherence of
Kant™s thought as any contemporary interpreter of his philosophy, and it
is a major aim of the present book to demonstrate the deep connections
between the two halves of Kant™s third Critique, that is, Kant™s aesthetics
and his teleology, as well as to demonstrate the profound connections
between Kant™s aesthetic theory and his theoretical philosophy on the
one hand as well as his practical philosophy on the other.
I focus here on Allison™s defense of Kant™s general conception of re¬‚ec-
tive judgment and the principle of purposiveness, his defense of Kant™s
deduction of pure judgments of taste, and his interpretation of Kant™s
connection between aesthetics and morality. In the following section I
recap those parts of Kant™s Theory of Taste that take up these three topics,
and in the section after that I discuss several points of difference between
his views and mine in each case. In order to defend the unity of the
third Critique, Allison defends Kant™s characterization of the judgment of

1 Allison (2001), xvi, 424. In this essay, page references to Kant™s Theory of Taste will be given
parenthetically, without further identi¬cation.
2 Allison (1983); Allison (1990).

Paul Guyer and Henry E. Allison

taste as a species of re¬‚ective judgment, and argues that our judgments
of beauty do indeed depend upon the principle of the purposiveness of
nature that is also the key to the judgments about the systematicity of
natural laws analyzed in the introductions to the work as well as to the
judgments about objective purposiveness analyzed in its second half. To
display the connections between Kant™s aesthetics and his theoretical phi-
losophy, Allison uses the ¬rst Critique™s distinction between the quid facti
and the quid juris, the analysis of the content or meaning of particular
forms of judgment and the justi¬cation of using such forms of judgment,
to explicate the relation between Kant™s Analytic of the Beautiful and
his Deduction of Pure Aesthetic Judgment, and then defends the success
of Kant™s strategy for the deduction of judgments of taste by grounding
them in the general conditions of the possibility of cognition. Finally, in
order to explain Kant™s connection between aesthetics and morality, Alli-
son interprets Kant™s theory of the “intellectual interest” in the beautiful
as evidence of nature™s amenability to our achievement of the ends of
morality, an assurance that we need not in order to be motivated to act as
morality requires but rather to prevent our motivation to be moral from
being undermined by our own predisposition to radical evil, and then
defends Kant™s thesis that all beauty, and thus natural as well as artistic
beauty, can be seen as an expression of aesthetic ideas, and speci¬cally
of the morally requisite idea of nature™s amenability to the achievement
of our moral objectives.

Allison explores the connection between “Re¬‚ective Judgment and the
Purposiveness of Nature” (13) in Chapter 1 of Kant™s Theory of Taste. He
bases his account on Kant™s de¬nition of re¬‚ection in the First Introduc-
tion to the third Critique as the activity of comparing and holding together
given representations “either with others or with one™s faculty of cogni-
tion” (20, citing FI 20:211).3 This de¬nition of re¬‚ection links aesthetic
judgment with other forms of re¬‚ective judgment: The comparison of
perceptual representations with each other is the basis for the formation

3 Citations from the Critique of the Power of Judgment (CPJ ) or its so-called First Introduction
(FI) will be located by volume and page number as in the Akademie edition. Allison
generally follows the translation by Werner Pluhar (1987). Unless quoting directly from
Allison, I will quote from Immanuel Kant, Critique of the Power of Judgment, edited by Paul
Guyer, translated by Paul Guyer and Eric Matthews (2000). Both of these translations
provide the Akademie edition pagination.
Dialogue 113

of empirical concepts of objects in nature, a process that is structured by
the a priori concepts of the pure understanding (the categories) but by
no means fully determined by those concepts, while the comparison of
particular perceptual representations with one™s own faculties of cogni-
tion is the basis for judgments of taste. The heart of Allison™s argument
in Chapter 1 is that Kant replies to Hume™s doubts about the uniformity
of nature over time, and thus about the very possibility of reliable empir-
ical concept formation, with a principle of purposiveness, which claims
“not that nature is purposive, that is, that we have some sort of a priori
guarantee that it is ordered in a manner commensurate with our cogni-
tive capacities and needs,” but rather that “we are rationally constrained
to approach nature as if it were so ordered”; “in Kant™s own terms, at
the basis of all re¬‚ection on nature (the search for empirical laws) lies
the a priori principle that ˜a cognizable order of nature in terms of these
[empirical] laws is possible™ ” (39, citing CPJ, 5:185). This principle is
thought of as having “normative or prescriptive force” (40), although it
prescribes not to nature but to the faculty of judgment itself, and is thus
a principle of heautonomy rather than autonomy (41). This principle
seems to be necessary simply because it is irrational to engage in an activ-
ity, such as scienti¬c inquiry, without some form of assurance that success
in that activity is at least possible: “It is right, that is, rationally justi¬ed, to
presuppose the principle of purposiveness because judgment legislates it
to itself as a condition of the possibility of its self-appointed task” (41).
Chapters 7 and 8 present Allison™s account of Kant™s deduction of
pure aesthetic judgments. In Chapter 7, Allison argues that the fourth
moment does not introduce an additional condition for the quid facti, but
rather “provides a unifying focus for the conditions that must be met, if a
judgment of taste is to be pure” (144). The heart of his argument in this
section, however, is his claim that “the argument of §21 for the necessity
of presupposing a common sense, though not itself part of the deduc-
tion of taste, nonetheless . . . provides grounds for postulating a cognitive
capacity that is a necessary (though not a suf¬cient) condition of the
possibility of taste” and “removes a worry generated by the account of the
conditions of a pure judgment of taste that the very idea of a common
sense . . . might be incoherent” (145). The gist of Allison™s interpretation
is that the “capacity to judge” ¬ts between a manifold of imagination and
the faculty of understanding by means of feeling is the condition of the
possibility of any cognition, for if we always needed a concept in order
to judge whether a concept ¬t a manifold, we would be saddled with an
in¬nite regress (154). The argument of §21 thus prepares the way for
Paul Guyer and Henry E. Allison

the argument of §38, discussed in Chapter 8, where Kant infers from the
premises that the suf¬cient conditions of aesthetic judgment are also the
necessary conditions for cognition in general, and that we are justi¬ed
in presupposing the latter in everyone the conclusion that if an object “is
subjectively purposive for me, then it must be subjectively purposive for
everyone” (176). This is the answer to the quid juris, although, of course,
invoking it in the case of any particular judgment presupposes that the
quid facti has been satisfactorily answered, that is, that the particular judg-
ment one has made is in fact a pure judgment of taste. In this chapter,
Allison rebuts my criticism that Kant™s deduction is a failure because it
does not show that the free harmony of the faculties must occur in every-
one in response to the same particular objects, or even that everyone
who is capable of achieving a harmony of the cognitive faculties under
the guidance of concepts must also be able to achieve it without such
guidance, on the ground that I expect the deduction to answer the quid
facti rather than the quid juris (180“2).
Allison presents his interpretation of Kant™s connection between aes-
thetics and morality in Part III of Kant™s Theory of Taste, arguing that the
premise for Kant™s connection is that it is “a central, though frequently
overlooked, feature of Kant™s moral theory . . . that the moral law dictates
the pursuit of certain ends” (203): The primary way in which aesthetic
experience is conducive to morality will then be that such experience,
as experience of the purposiveness of nature, gives us some general evi-
dence that nature is conducive to the realization of our ends, even our
moral ends, which conviction can then prevent our motivation to act as
morality requires from being undermined by fears of the futility of so
acting, fears that can all too easily be exploited by our own disposition
to radical evil. Allison argues that Kant™s antinomy of taste “ the appar-
ent con¬‚ict between the thesis that judgments of taste cannot be proven
from concepts but must yet in some way be based on concepts for it
to be rational for us to quarrel about them (239) “ cannot be resolved
simply by appeal to the indeterminate character of the concept of the
harmony of the cognitive faculties, which could explain how both the-
sis and antithesis can be true, but also requires the recognition that an
essentially indeterminable concept is in fact an idea of reason and thus
an idea of the supersensible (248). He claims that Kant is right to assert
in §51 that the experience of natural as well as artistic beauty points us
to this supersensible idea of reason as the basis of the essentially indeter-
minable concept of the harmony of the faculties, and that Kant™s thesis in
§59 that beauty is the symbol of the morally good depends “precisely” on
Dialogue 115

the premise that “aesthetic ideas . . . indirectly exhibit ideas of reason (in
virtue of their analogous ways of gesturing to the supersensible” (257).
The recognition that taste has a supersensible foundation thus facilitates
“the thought of oneself as a member of an ideal community subject to a
universally valid norm” (265) and is conducive to morality for that rea-
son. Since the experience of beauty is not the unique way in which one
could come to have this self-conception, however, this analogy cannot
ground a strict duty that everyone have or develop taste, but only “a duty,
as it were, or an indirect duty” (265).

(i) Reflective Judgment and the Principle of Purposiveness
Here I want to raise three issues.
First, I do not think that Allison has convincingly shown that we actu-
ally employ the principle of purposiveness, that is, the assumption that
nature is commensurate with our cognitive capacities, even in its merely
“heautonomous” form, in the process of making judgments of taste. In
fact, Allison has not actually speci¬ed what role this principle plays in
searching for a system of empirical concepts of nature except to suggest
that our conviction of this principle, by assuring us of the possibility of
success in scienti¬c inquiry, is necessary to motivate us in the conduct
of that inquiry. I have myself argued that the principle of purposiveness
plays a more complex role in the conduct of scienti¬c inquiry than that;
on my account, the transcendental assumption that nature itself is sys-
tematic and thus amenable to the logical systematization of our concepts
of it does not merely buck us up when our scienti¬c energies might falter,
but also gives us some speci¬c guidance in the construction and testing of
empirical hypotheses (where empirical data underdetermine empirical
hypotheses, at least begin by testing those that ¬t better into a system
of currently con¬rmed laws rather than those that ¬t less well), and the
membership of a particular lawlike generalization in a system of empirical
laws also gives some basis for the ascription of necessity to that general-
ization.4 But the principle of purposiveness appears to play none of these
roles, neither that to which Allison points nor those that I have analyzed
in the case of judgments of taste. Kant connects our pleasure in a beautiful
object with the fact that the harmony in our experience of its manifold
does not appear to follow from any concept or rule, and thus appears

4 See Guyer (1990a) and (1990b).
Paul Guyer and Henry E. Allison

to us entirely “contingent” and unintentional (CPJ, 5:188). This means
that an antecedent conviction that nature is commensurate with our cogni-
tive capacities, which is what the principle of purposiveness amounts to,
plays no role in our experience of beauty; if anything, such a conviction
would block the experience of beauty, making the harmony between the
manifold presented by the object and our own cognitive requirements
seem necessary rather than contingent. In con¬rmation of this point,
one might note the “direction of ¬t” in Kant™s explanation of the intellec-
tual interest in beauty: It must be the experience of natural beauty that
provides a hint that nature is amenable to our moral ends, rather than
an experience of nature™s amenability to our moral ends that provides a
hint of natural beauty, precisely because the existence of beauty must not
appear to derive from any more general rule if it is to strike us as thor-
oughly contingent and unexpected. Beyond this, since the judgment of
taste must be free from determination by concepts, the process of mak-
ing a judgment of taste cannot derive any guidance from the principle of
purposiveness in the way that the conduct of scienti¬c inquiry can derive
guidance from the conception of nature as systematic. That conception
leads us to formulate and prefer certain concepts of nature over others,
at least for the purposes of testing, on the ground that they ¬t better with
a larger system. But the assumption that beauty is purposiveness gives us
no particular guidance in trying to discern whether a particular object is
beautiful. If one likes, one might say that the conception of purposive-
ness of form tells us to focus on the form of an object in attempting to
judge whether it is beautiful, but the term ˜purposiveness™ does not really
give us any additional guidance; it does not tell us what to look for in
the form. The recommendation that we should look for purposiveness of
form does not tell us anything about how we should go about searching
for harmony between imagination and understanding. ˜Purposiveness™
is just a label that can be applied to that condition, which is detected
without any guidance from the principle of purposiveness at all. To be
sure, the label of subjective purposiveness offers us enlightenment about
his explanation of our pleasure in beauty: It connotes the fact that the har-
mony of the faculties pleases us because it is an unexpected satisfaction
of our general goal or purpose in cognition. So we can admit a principle
of purposiveness as part of Kant™s explanation of aesthetic response; but
no such principle plays any role in actually making particular judgments
of taste.
My second worry about Allison™s account of re¬‚ective judgment con-
cerns his suggestion that feeling can be equated with the capacity to judge.
Dialogue 117

I agree that the capacity to judge, speci¬cally to judge that empirical con-
cepts apply to perceptual objects, ultimately requires perceptual abilities
to apply predicates that cannot themselves derive any further guidance
from concepts: For example, to apply the concept ˜red book™ to volumes
in my library, I simply have to be able to see which ones are bound in
red, as opposed to blue, yellow, and green. No further rules can help
me here. But, of course, Kant is concerned to distinguish pleasure and
pain as subjective states that cannot become predicates of objects from
other subjective states, such as the sensations of colors, which can (CPJ, §1,
5:203“4), and it is not obvious that the patently discriminative function
that can be assigned to those subjective states that can become predicates
of objects should also be assigned to those that cannot. Perhaps feelings
of pleasure and pain must remain thoroughly subjective precisely because
they are not by themselves suf¬cient to ground judgments about objects,
at least beyond the simple judgments that those objects cause pleasure
or pain.
What would it mean for a feeling of pleasure or pain to be in and
of itself a capacity to judge or discriminate? One thing it might natu-
rally be thought to mean is that the feeling is a suf¬cient condition for
making a judgment, and thus that a feeling of pleasure is a suf¬cient
condition for judging that something is beautiful and a feeling of dis-
pleasure is a suf¬cient condition for judging that something is ugly. For
this to be the case, it would in turn be necessary that feelings of pleasure in
beauty are phenomenologically distinguishable from all other feelings of
pleasure, and feelings of displeasure at ugliness are phenomenologically
distinct from all other feelings of displeasure. But I at least have never
been able to ¬nd evidence that Kant believed this, although others of his
time, such as Hume, apparently did. Kant explicitly asserts that there are
“three different relations of representations to the feeling of pleasure in
displeasure, in relation to which we distinguish objects or kinds of rep-
resentations from each other,” namely, as agreeable, beautiful, or good
(CPJ, §5, 5:209“10), but he gives no hint that there is any phenomenolog-
ically evident or qualitative difference between the feelings of pleasure
in these different cases themselves. Instead, his view seems to be that in
order to make a judgment of taste we have to exclude other possible
causes, namely interests, for our feeling of pleasure in an object. But this
is precisely to say that the occurrence of a feeling of pleasure is not by
itself a suf¬cient condition for making a judgment of taste, and thus it
seems peculiar to equate a feeling of pleasure with a capacity to judge
beauty. The occurrence of a feeling of pleasure is a necessary condition
Paul Guyer and Henry E. Allison

for making a judgment of beauty, and that makes such a judgment an
aesthetic judgment (see 54), but it is not suf¬cient.
This is why my interpretation of Kant™s theory of aesthetic judgment is
aptly called a ˜causal™ interpretation: Judging that an object is beautiful
on the basis of a feeling of pleasure is analogous to judging that an effect
is caused by one sort of cause rather than another. I have never denied
that there are some elements of Kant™s view that speak against such an
interpretation; in particular, in §12 he seems to deny that the connection
between a particular object and our pleasure in it can be causal because a
causal relation must be a priori, while this connection is a posteriori (CPJ,
§12, 5:221“2). But this does not seem to be a serious problem: If we take
Kant™s appeal to the a priori here to be an expression of the necessary law-
likeness of causal connections, we could allow that the relation between
the harmony of the faculties and the feeling of pleasure is lawlike, even
though it follows from the very concept of an unintended and apparently
contingent harmony that there can be no lawlike relation between the
forms of objects that allow us to realize this harmony and the harmony
itself. In any case, the virtue of the causal interpretation, at least as I see
it, is that the fact that we have to judge that an object is beautiful indi-
rectly and by exclusion, that is, by excluding sensory or moral interests
as alternative causes, and that the latter cannot be done with certainty
is precisely what explains the fact, which Kant stresses, that particular
judgments of taste are themselves never certain. Allison accepts Kant™s
insistence on this point, but I am not sure what his own explanation of it
is given his apparent assumption that feelings of pleasure or displeasure
are suf¬cient conditions for making aesthetic judgments.
This is related to the last of my present points, which is simply that I
¬nd Allison™s claim that the feeling of pleasure in a judgment of taste pos-
sesses “intentionality” (53) undeveloped and unsustained. Allison does
not deny that the feeling of pleasure in a beautiful object is an effect of
the harmony of faculties that object allows, but he denies that it is “simply
the effect of such a harmony” because “it is also the very means through
which one becomes aware of this harmony” (54). But that by itself does
not imply that such a feeling has any internal content or structure that
determines its reference, which is how I would interpret the traditional
conception of intentionality. An effect is often precisely how we become
aware of a cause even when that effect clearly has no intentionality, at
least if we have enough other evidence to interpret that effect as a sign of
its cause. For example, a mushroom cloud might be the effect of either
a small nuclear device or a large volcanic eruption, but if we can exclude
Dialogue 119

one of these possible causes, say the ¬rst, by radiological data, then the
effect can also be taken as the sign of the other. Allison would have to pro-
vide an account of the internal structure of a feeling of pleasure, which
makes it by itself a sign of the harmony of the faculties, or of subjective
purposiveness in order to undermine the interpretation that it is a sign
of such a thing when but only when other possible causes of it can be

(ii) The Deduction of Pure Aesthetic Judgments
Here I want to make two points. First, I want to stress that although there
are differences in terminology and detail, there is no substantive disagree-
ment in our approaches to the strategy of the Analytic of the Beautiful
and its relation to the deduction of aesthetic judgments. But then I want
to explain why Allison has not convinced me that the deduction itself is
more successful than I have argued it is.
As I have noted, Allison organizes his account of the relation between
the Analytic and the Deduction around the contrast between the quid
facti and the quid juris: The moments of the Analytic answer the quid facti
by analyzing the meaning or content of the concept of a pure judgment
of taste, and thereby telling us what conditions any particular judgment
must satisfy in order to count as a pure judgment of taste, rather than,
say, a mere report of sensory agreeableness or a masked judgment of
instrumental or moral goodness; the quid juris, however, is answered by
the deduction, which, by arguing for the universality of the conditions
for the occurrence of the harmony of the faculties, justi¬es the demand
for universal agreement to any particular judgment that is in fact a pure
judgment of taste. Since I found both ˜de¬ning™ and ˜justi¬catory criteria™
among the four moments of the Analytic,5 it may seem as if I failed to
distinguish between the roles of the Analytic and the Deduction, or the
tasks of satisfying the quid facti on the one hand and the quid juris on
the other. And indeed, in his rebuttal of my critique of the deduction,
Allison accuses me of precisely this error, that is, of confusing the task
of justifying the claim that a particular judgment is a pure judgment of
taste with that of justifying the claim of universal validity for any pure
judgment of taste, and of mistakenly concluding that the deduction is
a failure because it does not accomplish the former as well as the latter

5 See Guyer (1997a), 108.
Paul Guyer and Henry E. Allison

But the differences in our terminology for describing the function
of the moments of the Analytic mask what I believe are only minor dif-
ferences in detail in our interpretations. What I had in mind by distin-
guishing between de¬ning and justi¬catory criteria was the following
model. The moments of subjective universality and exemplary necessity
de¬ne a judgment of taste insofar as what I mean by calling an object
beautiful is that I take it to have produced a feeling of pleasure in me,
independently of its being subsumed under any concept or rule, which
I can nevertheless suppose would necessarily occur in any other normal
human being, at least under normal circumstances.6 That the necessity
of agreement about the feeling of pleasure produced by an object inde-
pendently of its subsumption under a concept is what is meant by calling
an object beautiful, I argued, is suggested by the fact that it is in the
exposition of the second moment of the Analytic that Kant most explic-
itly appeals to our ordinary linguistic practices in support of his analysis.
By contrast, that a judgment of taste is disinterested and that it should
be a response to the purposiveness of form in an object are not (or, in
the case of disinterestedness, should not be) treated by Kant as parts of
the meaning of the term ˜beautiful™ or of the concept of a judgment of
taste, but are rather factors to which we can appeal in the justi¬cation
of a particular judgment of taste, that is, in the assessment of a particular
feeling of pleasure as pleasure in the beauty of an object. They are such
conditions because they are factors associated with the free play of the
cognitive faculties: The purposiveness of form in an object is what induces
such a condition, and the disinterestedness of the judgment of taste is a
consequence of it. This should make it clear that in calling disinterest-
edness and purposiveness of form ˜justi¬catory criteria,™ I did not mean
that they preempt the task of the deduction: They are used to justify the
assertion of any particular judgment of taste on the basis of the pleasure
actually felt in an object because of their association with the harmony
of the faculties, but the task of proving that the harmony of the faculties
must occur in the same way and under the same conditions for everyone,
and thus that of justifying the general practice of making judgments of
taste, awaits the deduction proper. But that means that, at least in general

6 Contrary to what Allison suggests, I did not claim that the second and fourth moments of
the Analytic make what is essentially the same claim because I identi¬ed the concepts of
(subjective) universality (the subject of the second moment) and (exemplary) necessity
(that of the fourth) (78“82), but rather because I believed both moments together to
state that a judgment of taste asserts the necessity of the subjective universality or universal
communicability asserted by a judgment of taste. See Guyer (1997a), 142“5.
Dialogue 121

terms, I distinguish between the tasks of discharging the quid facti and the
quid juris in the same way that Allison does.
Now, we might still argue over whether the disinterestedness of judg-
ments of taste and the formalism of beauty are actually included in the
meaning of the concept of beauty, as Allison contends, or not, as I do. I
might be persuaded that Kant at least intended that in the case of disin-
terestedness, although I continue to think there are good reasons why,
historically, he should not have; I think there is less room for debate in
the case of purposiveness of form. But space will not allow the pursuit of
this issue, so at this point I will turn instead to the question of the success
of the deduction itself. In Kant and the Claims of Taste, I argued that Kant
attempted to prove that the free play of the cognitive faculties would be
induced in ideal observers by the same particular objects twice, ¬rst in
§21, in the course of the exposition of the fourth moment of the Analytic,
and then in §38, in the Deduction proper, but that Kant™s argument was
in both cases inadequate. Allison argues that §21 and §38 have different
functions and that each argument is successful. The point of §21, he con-
tends, is to prove that all normal human beings have a general capacity for
common sense, which is just the ability to subsume intuitions under con-
cepts by means of feeling and without the guidance of further concepts,
which makes no direct claim about taste (153“4). Only in §38, he claims,
does Kant take the further step of arguing that if everyone must have the
same ability to subsume intuitions under concepts by means of feeling
rather than rules, then they must also have the same ability to respond
with pleasure to subjective purposiveness, or to detect purposiveness of
form by means of the feeling of pleasure without any subsumption of the
manifold of intuitions under concepts at all. It is only in §38, he con-
tends, that Kant asserts the principle that “If x is subjectively purposive
for me, then it must be subjectively purposive for everyone,” and this is “a
reasonable-enough claim,” he maintains, “given the connection between
subjective purposiveness and the conditions of judgment built into the
very de¬nition of such purposiveness” (176).
For present purposes, I will not dispute Allison™s division of labor
between §21 and §38. To the proposed argument of §21, however, I would
make two replies. First, while it is true that everyone who is capable of
applying concepts to intuitions at all must have a variety of perceptual
abilities to recognize the instantiation of the most elementary predicates
comprised in their concepts without further rules, it does not follow from
this that everyone must have the same set of such perceptual abilities. To
take the most obvious example, both sighted and blind persons will have
Paul Guyer and Henry E. Allison

a variety of perceptually based discriminatory abilities, but, of course,
blind persons will not have the ability to apply color terms via unaided
perception. But there may be subtler differences in the ability to apply
concepts without rules even among persons with the same basic percep-
tual capacities: For instance, some sighted persons might have the ability
to “see” the duck“rabbit ¬gure as both a duck and a rabbit, while others
with the same visual acuity might just not “get” it. Second, although per-
haps Allison will reply that this bears more on the argument of §38 than on
§21, I would reiterate my earlier suggestion that there is no obvious con-
nection between the ability to make basic perceptual discriminations by
“feel” and the alleged ability to judge purposiveness of form by “feeling”
More important, however, are my reservations about the success of
Kant™s argument in the deduction proper of §38. Allison™s suggestion that
I expected the deduction itself to provide suf¬cient conditions for the
justi¬cation of particular judgments of taste is not correct; while he cor-
rectly quotes me as saying, rather loosely, that “Kant™s own analysis of aes-
thetic judgment requires that his deduction come to particulars” (180),7
it should be clear that my general position was that the deduction is meant
to justify only the conditional that if one has correctly ascribed a feeling of
pleasure to the harmony of the faculties, then one may make a judgment
of taste because the harmony of faculties can be expected to occur under
the same circumstances for every ideally quali¬ed and situated observer.
As I put it at the beginning of my discussion of the deduction:

But attributing a feeling of pleasure to a source in the harmony of the higher
cognitive faculties can ground a rational claim to intersubjective validity only if
this harmony is itself subject to a valid imputation of intersubjectivity, or if the
occurrence of this harmony in the presence of a given object may rationally be
expected in anyone who does abstract from interest in its existence and con¬ne
his attention to its mere form of ¬nality. Thus Kant must provide an argument
that the harmony of the faculties occurs in different persons under the same
conditions. . . . 8

It is the task of the deduction to provide this argument, or to discharge
Allison™s quid juris; I do not think I confused the quid juris with the quid

7 The quotation is from 272 of the ¬rst edition of Kant and the Claims of Taste (Cambridge,
Mass.: Harvard University Press, 1979); it is on 242 in the second revised edition (Guyer
8 Guyer (1997a), 228.
Dialogue 123

My objections to Kant™s deduction, then, were not based on the com-
plaint that Kant fails to include the conditions for the justi¬cation of par-
ticular judgments of taste within that deduction itself, but rather that Kant
failed to prove that the harmony of the faculties must in fact occur under
the same circumstances even in ideally quali¬ed and situated observers,
that is, observers who have all satis¬ed the quid facti. I had two basic
objections to Kant™s argument: First, Kant did not prove that the gen-
eral ability to subsume intuitions under concepts without further rules
entails the ability to unify manifolds without concepts, or “that everyone
capable of cognition must respond in the same way to a given mani-
fold when his response is not guided by concepts at all.”9 Second, even
if we allow Kant the assumption that everyone capable of subsuming
intuitions under concepts without further rules must also be capable of
feeling the purposiveness of form in some manifolds without the guid-
ance of concepts at all, we cannot reasonably assume that everyone will
have this response to the same particular manifolds.10 Some might be
able to feel unity in very complexly patterned and richly variegated man-
ifolds, while others might respond this way only to less complex mani-
folds “ perhaps that™s why some people prefer the complex counterpoint
of Bach to the melodic invention of Bellini, and others the converse.
I do not think that Allison has addressed these objections. He simply
says that “cognition itself presupposes a common sense, understood as a
universally valid ˜feeling™ through which the conformity of universal and
particular is immediately apprehended in judgment,” but “if, as Kant now
argues, taste . . . is likewise a feeling directed to the conformity of given
representations with these same conditions, then it does seem reason-
able to assume the universal validity of this feeling as well” (177). But
I do not think that this suf¬ces to prove that everyone must be able to
feel the conformity of the same particular representations to the gen-
eral conditions of cognition, which is what Kant™s deduction requires.
Correctly discharging the quid facti, that is, attributing a particular feel-
ing of pleasure to the harmony of the faculties, will justify the asser-
tion of a judgment of taste only if the quid juris has been satisfactorily
answered, that is, if it has been proven that the harmony of faculties must
occur in response to the same objects and their manifolds of intuition
for everyone, and I do not believe that Allison has shown that Kant has
proven that.

9 Ibid., 284.
10 Ibid., 287“8.
Paul Guyer and Henry E. Allison

(iii) Aesthetics and Morality
Allison™s account of Kant™s connection between aesthetics and morality
focuses on two points. First, Allison argues that Kantian morality does
require us to attempt to realize certain ends, and that our experience of
beauty, above all of natural beauty, gives us some sort of evidence that it
will be possible for us to achieve these ends in nature, a hint that is the basis
for our intellectual interest in natural beauty. Second, Allison argues that
the essentially indeterminable nature of the concept of free play means
that this free play expresses a rational idea of the supersensible, and
thus that the experience of beauty, whether artistic or natural, really is
always an expression of an idea of reason and is linked to morality by that
I certainly agree with the ¬rst premise of Allison™s interpretation,
namely the supposition that morality imposes necessary ends upon us,
and that our attempts to act morally, even if motivated purely by respect
for the moral law and the ideal of duty rather than by any direct desire
for those ends, would nevertheless be rendered irrational if we did not
have some form of assurance that the realization of those ends is at least
possible.11 I likewise agree with his interpretation of Kant™s claim that we
can take an interest in the existence of beauty as evidence of this possibil-
ity without destroying the aesthetic character of the experience of beauty
itself. My only criticism of this part of Allison™s work is that it runs the risk
of simplifying Kant™s account of the connections between aesthetics and
morality by focusing exclusively on this aesthetic intimation of the possi-
bility of achieving the ends of morality. As I understand the program of
the third Critique, it is actually meant to bridge the gap between the realm
of nature and that of freedom or morality in at least three ways: A variety of
forms of aesthetic experience are meant to give us palpable evidence of
the possibility of acting disinterestedly, that is, the possibility of our being
motivated by the moral law and respect for duty; aesthetic experiences
are actually meant to teach us to love disinterestedly or even against our
sensible interests, that is, to assist in the production of feelings that are
actually helpful in our efforts to act in accordance with moral motivation;
and then, to be sure, there are experiences of re¬‚ective judgment that
are meant to assure us not only of the possibility of acting out of moral
motivation but also, as Allison argues, of the possibility of realizing the
ends that we would then be motivated to pursue.

11 I have been arguing for precisely this in much of my recent work on Kant™s moral
philosophy and teleology; see for example my 2001 and 2002.
Dialogue 125

Thus, for example, I interpret Kant™s thesis that the beautiful is the
symbol of the morally good in §59, where the analogy between beauty
and the morally good is based not on any connection between the content
of beautiful objects and the object or end of morality but on the character
of the experience of beauty, to imply that the experience of beauty gives
us a certain form of evidence of the possibility of our being motivated by
morality or acting autonomously rather than heteronomously: The heart
of this analogy is the claim that our experience of the freedom of the
imagination in our response to beauty suggests to us the possibility of
“the freedom of the will . . . conceived as the agreement of the latter with
itself in accordance with universal laws of reason” (CPJ, §59, 5:354), or,
more precisely, gives us some sensible con¬rmation of the more abstract
or intellectual acknowledgment of this possibility that is forced upon us by
our awareness of our obligation under the moral law in the so-called ˜fact
of reason™. Second, Kant™s famous comment in the General Remark on
the analytics of the beautiful and sublime that “The beautiful prepares us
to love something, even nature, without interest; the sublime, to esteem
it, even contrary to our (sensible) interest” (CPJ, §29 GR, 5:267) can be
taken to mean, again, that the experience of the beautiful teaches us
that it is possible for us to act disinterestedly, that is, out of disinterested
motivation, while the experience of the sublime reminds us that we must
often overcome sensible interests or inclinations in order to do this, but
also to suggest that the experiences of the beautiful and the sublime
actually prepare us to act morally by strengthening dispositions to feeling
that are, as Kant says in the Metaphysics of Morals, “serviceable to morality in
one™s relations with other people.”12 To make sense of this idea requires
that we think of the cultivation of feelings conducive to acting as morality
requires not as an alternative to acting out of the motive of respect for duty
but as part of the natural means by which we implement this motivation,
but I would argue that is precisely what Kant intends.13
Finally, while I agree that Kant™s account of our intellectual interest in
natural beauty is based on the premise that we need some form of pal-
pable assurance that we can actually realize the ends that we must form
when we are successfully motivated by the moral law, I would argue that
it is primarily in the Critique of the Teleological Power of Judgment that
Kant develops his idea that a form of the experience of re¬‚ective judg-
ment gives us this evidence. That is, it is our experience of the internal

12 See Metaphysics of Morals, Doctrine of Virtue, §17, 6:443, in Kant (1996), 564.
13 See my 2003a.
Paul Guyer and Henry E. Allison

purposiveness of organisms, not of their beauty or subjective purposive-
ness, that really forces upon us the thought that they must be designed,
and that if they are designed, then that must be for some end “ a thought
that we then naturally extend both to nature as a whole and to the exis-
tence of beautiful objects, whether organic or not, within nature (see
CPJ, §67, 5:379“80). The additional premise that only an unconditional
end could really provide the point for the existence both of the system
of nature as a whole as well as of natural beauty within it then leads to
the further recognition that only the development of human freedom
could be the ¬nal end of nature (CPJ, §84, 5:435 and §86, 5:443), and
that in turn leads to the thought that the realization of the highest good,
that is, the end that morality sets for us, must also be realizable within
nature (§87, 5:450).14 I would argue that the thesis that our intellectual
interest in beauty gives us a hint of nature™s amenability to our moral ends
is itself only a hint of this argument that is more fully developed in the
second half of the third Critique, and also observe that even there, Kant™s
position seems to be that our experience of organic nature ¬rst con¬rms
our recognition of the possibility of our acting freely before leading us
to further con¬rmation of the possibility of realizing the ends of our free
I now turn to Allison™s more controversial claim that the experience of
beauty is virtually identical to the presentation of an idea of reason, and
thus that all beauty expresses ideas and for that reason ultimately symbol-
izes the morally good. Allison develops this interpretation in response to
my criticism that Kant™s presentation of the Antinomy of Taste is redun-
dant and his solution to it ill-founded because his account of the expe-
rience of beauty as that of the indeterminate harmony of the cognitive
faculties already explains why we can rationally claim subjective univer-
sality for judgments of taste without being able to prove them from rules
and without requiring recourse to the idea of a supersensible substratum
of our cognitive faculties.15 Allison™s theory is that the concept of an essen-
tially indeterminable condition is already an idea of reason (247“8), and
that any beautiful object, natural as well as artistic, thus presents an idea
of reason that points us in the direction of the supersensible and there-
fore of morality: This is in turn why Kant™s claim that the beautiful is the
symbol of the morally good is appended to his solution to the antinomy
of taste (257).

14 I have expounded this argument in my 2000a and 2001b.
15 See Guyer (1997a), 299“307.
Dialogue 127

I am, to be sure, simplifying Allison™s complex and subtle account, but
let me make three objections that I think could be defended against even a
fuller presentation of his interpretation. First, I do not think that Allison™s
supposition that any “essentially indeterminable” concept is necessarily
an idea of reason can be sustained. While I am prepared to concede
Allison™s claim that the harmony of the faculties should be described as
“essentially indeterminable” rather than merely “indeterminate” (246),
I think there is an essential difference between the harmony of the fac-
ulties and an idea of reason: The harmony of the faculties is ˜essentially
indeterminable™ because it does not involve the subsumption of a mani-
fold of imagination under any determinate concept of the understanding
at all, even in the case of representational art, where concepts might be
regarded as part of the content of the manifold of imagination, whereas
ideas of reason are, by Kant™s account, always generated from concepts of
the understanding, although claiming for them an application beyond
what could be con¬rmed by sensibility, given its inherent limitations.
If this is correct, then the experience of the harmony of the faculties
could never be more than a symbol of any idea of reason, and it would
not seem to be automatic that every experience of the harmony of the
faculties should necessarily be such a symbol, at least without a further
framework for interpretation.
Second, while Allison™s account is meant to be a defense of Kant™s
notorious claim that all “[b]eauty (whether it be beauty of nature or of
art) can . . . be called the expression of aesthetic ideas” (CPJ, §51, 5:320), I
think that his approach simpli¬es Kant™s conception of an aesthetic idea.
His account supposes that we have an aesthetic idea whenever a mani-
fold of imagination, because of its essential indeterminability, connotes
a rational idea or, perhaps more precisely, the form of a rational idea.
But Kant™s theory is, I believe, more complex. As I understand Kant, an
aesthetic idea is the guiding and organizing theme of a work of art that
connotes a distinct idea of reason on the one hand, and is itself exempli-
¬ed and presented by the more concrete material or “attributes” of the
work of art on the other. Thus, for example, in a sculpture of Jupiter, the
personage of the god is the aesthetic idea that connotes more abstract
ideas such as omnipotence and justice on the one hand, and is realized
through such more concrete attributes as his eagle on the other.16 If this is
right, however, then an aesthetic idea is a distinct feature of a work of art,

16 For this interpretation, see “Kant™s Conception of Fine Art,” chapter 12 in Guyer (1997a)
and (1997b), 85“98.
Paul Guyer and Henry E. Allison

identical neither to the idea of reason that it suggests to us “ which is itself
a particular idea of reason, not the general idea of an essentially inde-
terminable idea that the experience of beauty would express on Allison™s
account “ nor to the harmony of the faculties that the work may induce
in us. The claim that the harmony of the faculties itself suggests by its
very form an idea of practical reason is, in my view, a separate and further
claim, which Kant does not suggest until he asserts the analogy between
the beautiful and the morally good in §59; it is not already entailed by or
necessary for the resolution of the antinomy of taste.
This leads to my third concern, which I might put by saying that even
if we accept the claim that the concept of the harmony of the faculties is
itself an idea of reason, it still would not follow that it is ipso facto a morally
signi¬cant idea of reason. All ideas of reason are not, of course, the same;
for example, the idea of a determinate cosmos is an idea of reason, but not
immediately of any moral signi¬cance. Some more complex argument
would be needed to identify the harmony of the faculties with a morally
signi¬cant idea of reason. To be fair, Allison does actually argue that the
harmony of the faculties is a morally signi¬cant idea of reason not just
because of its essential indeterminability, but because it is the idea of the
form of purposiveness (251), and moral ideas are obviously connected
to purposiveness as well. But here I would counter that the experience of
the harmony of the faculties is in the ¬rst instance an experience of the
form of cognitive purposiveness and not directly an experience of even the
form of practical purposiveness. If the experience of beauty is to retain its
disinterestedness, surely its connection to practical purposiveness can be
only analogical, as Kant indeed maintains in §59.
So I conclude that Allison™s attempt to rescue Kant™s antinomy of taste
is a failure. But I hardly conclude that Allison™s book is a failure. It is a
rich and provocative work that illuminates many aspects of Kant™s third
Critique, a book that is not merely about common sense but is also a
frustrating mixture of common sense and obscurity that has proven
surprisingly resistant to conclusive interpretation. Allison ably defends
many of Kant™s central theses and arguments, and even where he does
not, he presents cogent interpretations that will deservedly receive close
attention in the years to come.

response to paul guyer by henry e. allison
Paul Guyer has graciously acknowledged that we have wide areas of agree-
ment, indeed, even more than I had thought to be the case. Nevertheless,
Dialogue 129

as his critical comments make clear, there remains considerable disagree-
ment. Accordingly, I shall here focus on the disagreements that he empha-
sizes, most of which stem, I think, from our widely different attitudes
toward the systematic nature of Kant™s enterprise in the third Critique.
The initial set of objections concern issues involving the complex rela-
tionship between re¬‚ective judgment in general, the principle of pur-
posiveness, and aesthetic judgment. First, Guyer states that he does not
think that I have “convincingly shown that we actually employ the prin-
ciple of purposiveness . . . in the process of making judgments of taste”
(Guyer, this volume). This is quite true, since I endeavored to show no
such thing. On the contrary, I agreed with Guyer that Kant™s passing sug-
gestions that the principle of logical or formal purposiveness is also the
principle licensing judgments of taste are deeply misleading and do not
represent his considered view.17 Nevertheless, developing a suggestion of
Klaus D¨ sing, I also distinguished between a general principle (or con-
ception) of purposiveness and the speci¬c principle of logical or formal
purposiveness that is operative in the investigation of nature. And I fur-
ther indicated that what they have in common is their “heautonomy,”
that is, their grounding in and application to the power of judgment.18
Guyer™s second objection concerns my alleged suggestion “that feel-
ing can be equated with the capacity to judge” (Guyer, this exchange).
Although I hope that I didn™t suggest that, I did state that feeling for
Kant plays an essential judgmental role. Indeed, this is the only way in
which I can understand the Kantian conception of an ˜aesthetic power of
judgment.™ Thus, in the very ¬rst section of the third Critique, Kant states
explicitly that the feeling of pleasure or displeasure “grounds an entirely
special faculty for discriminating and judging” (5:204). In short, in agree-
ment with other eighteenth-century theorists of taste, Kant is committed
to the view that in a judgment of taste one judges through one™s feeling. By
contrast, on Guyer™s reading, the judgment is about the causal ancestry
of one™s feeling. But my problem with this is understanding how such a
judgment could be aesthetic, except in the Pickwickean sense that might
also apply to my judgment about the cause of my hangover. Certainly, it
does not require anything like a distinct power of judgment.
Guyer seems to have been led to his causal reading by the rejection
of the viability of something like the alternative I propose “ let us call it
the ˜intentional reading.™ The problem, as he sees it, lies in its apparent

17 Allison (2001), 59“62.
18 Ibid., 63“4.
Paul Guyer and Henry E. Allison

incompatibility with Kant™s insistence on the noncognitive nature of the
feeling of pleasure or displeasure. Since this feeling cannot be referred
to an object in a judgment, how can it function as a mode of judging in
anything like the manner I suggest?
Problematic as it may seem, however, the fact remains that Kant does
speak frequently of this feeling as one through which the mental state of
free harmony is represented.19 Moreover, to my mind, these texts, taken
together with the previously noted characterization of the power of aes-
thetic judgment, make the case for the intentional reading compelling.
Against this, Guyer insists that to attribute intentionality to a feeling
implies that the feeling must have “an internal content or structure that
determines its reference” (Guyer, this exchange), and he further claims
not to be able to ¬nd this in the pleasure of taste. Indeed, on his view, it
appears that all feelings for Kant are basically “raw feels,” which can differ
from one another in intensity and cause, but not in any phenomenologi-
cally discernible qualitative features. Although I cannot here enter into a
debate about the nature of intentionality, I believe it plausible to under-
stand intentionality as the directedness or aboutness characteristic of con-
sciousness. And since such consciousness need not amount to cognition
(˜recognition in a concept™ in Kant™s terminology), there is room within
a Kantian framework for viewing feeling as a mode of consciousness, and,
therefore, as having intentionality.
I likewise do not think it correct to deny that the pleasure of taste (or
any pleasure for Kant) has an “internal content or structure that deter-
mines its referent.” First, it has a distinct quality, namely, disinterestedness,
that characterizes the kind of liking it involves, one that Kant describes,
albeit brie¬‚y, in terms of a lingering (5:222). Second, if a feeling with a
richer phenomenology is required, one need look no further than Kant™s
account of the sublime. Granted, the latter is not the feeling for the beau-
tiful, but it is a feeling with a complex internal structure. Thus, it (along
with moral feeling) appears to be an important counterexample to what
Guyer takes to be the Kantian conception of feeling.
I believe that the true test of our respective accounts, however, is which
yields the more adequate rendering of Kant™s conception of a judgment
of taste. Guyer takes the great virtue of his interpretation to be its ability
to explain why the judgment of taste is always uncertain, something that I
also emphasize but allegedly fail to elucidate. On his view, he thinks that
this is readily explained by the fact that we have to judge that an object

19 See, for example, FI 20:223, 228, 230, KU 5:189, 190, 219, 220, and 222.
Dialogue 131

is beautiful “indirectly and by exclusion, that is, by excluding sensory or
moral interests as alternative causes” (Guyer, this exchange), a process
that can never yield certainty. He further questions whether this is possible
on my view, since I hold that “feelings of pleasure or displeasure are
suf¬cient conditions for making aesthetic judgments” (ibid.).
Guyer is correct that on my reading these feelings (as likings or dislik-
ings) are suf¬cient conditions for making aesthetic judgments or, more
precisely, judgments of taste. But he neglects to note the emphasis I place
on the distinction between a judgment of taste simpliciter and a pure judg-
ment of taste. Thus, I have argued that we can be certain about having
made a judgment of taste but not about having made a pure one. And I
develop this point by drawing the analogy with Kant™s notorious claim in
his moral theory that we can never be certain of having acted from duty
alone.20 No matter how sincere and conscientious one believes oneself
to have been in the performance of duty, it always remains possible that
the “dear self” lies behind the action. In other words, just as we can never
be certain of the purity of our motive, so, too, we cannot be certain about
the “purity” of our liking.
More importantly, on my view, this uncertainty cannot be understood
in the manner Guyer suggests. This becomes clear by considering his
example of the mushroom cloud. As he describes the situation, the prob-
lem is to determine which of two possible causes is responsible for the
phenomenon, and ruling out one on the basis of evidence establishes
the likelihood of the other. Similarly, Guyer reasons that, in the case of
the feeling for the beautiful, by ruling out possible interests as its cause,
we may judge (albeit not infallibly) that it was produced by the harmony
of the faculties.
I ¬nd this analogy revealing, since it points to a fundamental difference
in our readings of Kant. For I do not think it appropriate (particularly
from a Kantian perspective) to view interests as causes. Rather, they have
the status of reasons: either reasons to act or reasons to like (or dislike).21
As I see it, this follows from the very nature of an interest (whether sen-
suous or moral) as a product of practical reason.22 But if this is so, then
to exclude an interest as the ground of a liking is to exclude a certain
kind of reason for it (e.g., that one ¬nds the object agreeable or morally
uplifting) rather than certain possible causes. Thus, if, as Guyer suggests,

20 Allison (2001), 178.
21 At KU 5:221 Kant refers to interests as determining grounds of judgment.
22 See Gr 4:414n and 460n; KprV 5:79.
Paul Guyer and Henry E. Allison

determining whether a speci¬c liking is really for the beautiful is a mat-
ter of interpretation, then I think it is more like the interpretation of an
action in terms of an underlying intent than of an effect as the “sign” of
its underlying cause.
Since I believe that our differences regarding the deduction are largely
a consequence of those just discussed, I shall be somewhat more succinct
in my reply to Guyer™s second set of objections. To begin with, he seems
to have misunderstood the nature of my charge that he con¬‚ates the quid
facti with the quid juris. My quarrel with Guyer on this matter is not with
his inclusion of justi¬catory criteria within the Analytic of the Beautiful
(though I divide up the four moments in a quite different way than he
does). The problem, as I see it, concerns rather his view of the goal of
the deduction, which he claims at one point is to provide “an argument
suf¬cient to justify the imputation of speci¬c feelings to others on speci¬c
occasions.”23 In the present context, he takes this to require proving that
“the harmony of the faculties occurs in different persons under the same
conditions” (Guyer, this exchange). In either case, however, as a causal
claim about the conditions under which a certain mental state occurs,
it certainly concerns a matter of fact. And, as such, it might warrant an
expectation of agreement (or, under ideal conditions, a prediction); but, as
has often been noted, this is quite different from licensing a demand that
others acknowledge the appropriateness of one™s aesthetic response.24
Accordingly, Guyer is incorrect, or at least somewhat misleading, in
claiming that I fail to answer his objections to the deduction. For my
view is that these objections result from a misreading of Kant™s project.
Nevertheless, I think that Guyer is being perfectly consistent in viewing
the deduction in the way he does, since it is the logical consequence of
his causal interpretation of the judgment of taste. Thus, I am willing to
concede that if Guyer is right about that, then he is also correct in his
diagnosis of where the deduction fails. But since I do not think that he is
correct about the former, I have largely ignored his speci¬c criticisms of
the latter. Instead, I have endeavored to develop a different account of
the concern and structure of the deduction.

23 Guyer (1979), 272.
24 Interestingly, at one point Guyer himself distinguishes sharply between the notions of
expectation and demand, but connects the former with what he takes to be the epis-
temological reading of the deduction and the latter with an interpretation (such as
Crawford™s) that sees the deduction of taste as turning on its connection with morality
(op. cit., p. 261). Thus, he apparently fails to recognize any role for an ought (such as
the familiar ought of good reasons) in the analysis of taste independently of morality.
Dialogue 133

Fortunately, Kant himself is of considerable help in this regard, since
he is quite explicit about how he understands the task. On my reading,
the key passage occurs in §34, where Kant tells us that the project is to
“develop [entwickeln] and justify the subjective principle of taste as an a
priori principle of the power of judgment” (5:286). In other words, the
deduction consists of two steps: one in which the subjective principle
governing taste is developed or speci¬ed, and the other in which this
principle is justi¬ed. The former occurs in §35, where Kant identi¬es the
principle of taste with “the subjective principle of the power of judgment
in general” (5:286), and the latter in §38, which Kant entitles “Deduction
of Judgments of Taste” (5:289).
The argument of §35 is notoriously obscure, even by Kantian stan-
dards; for it is here that Kant speaks darkly of “schematizing without a
concept” and of subsuming, not intuitions under concepts, but the faculty
of intuitions (here identi¬ed with the imagination) under the faculty of
concepts (5:287). Nevertheless, at the risk of gross oversimpli¬cation, I
take Kant™s basic point to be that it is the power of judgment itself that
serves as the norm governing judgments of taste, which means that in
such a judgment the representation of an object is “subsumed” under
the conditions required by judgment to move from intuition to concept.
Thus, what is sensed in the judgment of taste is the conformity (or lack
thereof) of this representation with the norm (agreement of the imagi-
nation and understanding).25
If this is correct, then the task of §38, or the deduction proper, is simply
to show that the principle of judging aesthetically on the basis of confor-
mity to the subjective conditions of judgment is valid for everyone. But
this really requires little more than the reminder that these conditions, as
conditions of cognition, can be presupposed in all human beings. That
is why, as Kant notes in a remark attached to the section, the deduction
is so easy (5:290). For what it attempts to ground is simply the normative
principle that aesthetic judgments made in accordance with these con-
ditions of judgment, that is, pure judgments of taste, make a warranted
demand for agreement.26

25 See FI 20:220; 23.
26 I believe that this is supported by Kant™s claim that “it is not the pleasure but the uni-
versal validity of this pleasure perceived in the mind as connected with the mere judg-
ing of an object that is represented in a judgment of taste as a universal rule in the
judgment of taste, valid for everyone” (5:289). For Guyer™s quite different reading
of this passage, see Guyer (1997), 259. I discuss Guyer™s reading in Allison (2001),
Paul Guyer and Henry E. Allison

Otherwise expressed, Kant is claiming that the felt free conformity
of the representation of an object to the subjective conditions of cogni-
tion constitutes a compelling, universally valid “reason” to like an object
apart from any interest. And what makes this reason compelling is the
necessity of assuming (on pain of epistemological skepticism) that the
subjective conditions of cognition hold for the entire universe of cogniz-
ers. Accordingly, what you must agree with is not my ¬rst-order judgment
that x is beautiful (or nonbeautiful), though I would naturally welcome
such agreement. It is rather the normative principle that if my liking for
an object is, in fact, based on these grounds, you ought to like it also, and
that your failure to do so must be ascribed to a de¬ciency of taste. But,
of course, the problem is that since one can never be sure about the true
grounds of one™s liking, one is never in a position actually to demand the
agreement of others to a particular judgment.
Admittedly, this makes the conclusion of the deduction relatively weak,
certainly weaker than Guyer and others have taken it to be. Nevertheless,
the deduction, on my reading, is far from trivial. For, if successful, it
establishes that there is normativity in the domain of taste and that it
is based on a principle unique to judgment. Consequently, judgments
of taste are reducible neither to judgments of the agreeable, which are
altogether lacking in normativity, nor to judgments of perfection, whose
normativity is taken directly from cognition. Moreover, this is tantamount
to establishing the autonomy (or, better, heautonomy) of taste, which
both Guyer and I agree is a condition of its moral signi¬cance.
Since I cannot here attempt to provide anything like an adequate
response to Guyer™s critique of my account of the connections between
aesthetics and morality, I shall simply note a couple of places where I
believe he misreads my views and attempt to summarize these views in
a way that highlights what I take to be the main differences between us
on these complex topics. To begin with, Guyer suggests that I run ˜˜the
risk of simplifying Kant™s account of the connections between aesthetics
and morality by focusing exclusively on the aesthetic intimation of the
possibility of achieving the ends of morality” (Guyer, this exchange). But
though it is certainly true that I focus on this problem, particularly in
connection with natural beauty, I hardly think it fair to say that I do so
exclusively. On the contrary, I also insist that the experience of beauty
has an important propaedeutic function, helping us to break with our
sensuous interests and thereby providing a kind of bridge to morality.27

27 See Allison (2001), 264“6.
Dialogue 135

Up to this point, then, I do not think that there is a major disagreement
between us. Deep differences do emerge, however, when we turn to the
pivotal issues of how beauty supposedly symbolizes morality, whether this
applies to both natural and artistic beauty, and, most importantly, the link-
age between Kant™s accounts of symbolization and aesthetic ideas and the
resolution of the antinomy of taste in §57. In particular, Guyer dismisses
Kant™s move from the need for some (unspeci¬ed) indeterminable con-
cept to the concept of the supersensible as an unwarranted venture into
metaphysics, which is of no relevance to the account of taste. Instead, he
insists that all that is required to resolve the antinomy is the appeal to the
epistemological-psychological concept of the harmony of the faculties.28
By contrast, my project is once again to take the systematic dimension of
Kant™s account seriously and to provide a reading that makes some sense
out of its admittedly mysterious features.
In doing so, however, I do not, as Guyer suggests, attempt to elevate
the harmony of the faculties itself into an idea of reason. I argue instead
that the indeterminable concept required to resolve the antinomy is that
of the beautiful. Moreover, contra Guyer, I believe that Kant identi¬es
indeterminable concepts with ideas of reason.29
It follows from this that the concept of the beautiful is such an idea.
And I defend this claim by focusing on the explication of the beautiful
at the end of the third moment: “Beauty is the form of purposiveness of
an object, insofar as it is perceived in it without representation of an end”
(5:236). In brief, my claim is that the concept of the form of purposiveness
is an indeterminable concept and that purposiveness, though viewed in
the third Critique in connection with judgment, is with regard to its origin,
an idea of reason.30 Thus, in this respect I privilege the third moment,
whereas Guyer privileges the second; but I believe that my approach is
justi¬ed on the grounds that it is in the third moment that Kant focuses on
the aesthetic object. It also leads naturally to the talk of the supersensible,
which is the central feature of Kant™s resolution of the antinomy.
It is with this connection between taste and the supersensible in mind
that I attempt to explore how all beauty might be thought to symbolize

28 See Guyer (1997a), 340.
29 The key text here is KU 5:342, where Kant de¬nes such an idea as an “indemonstratable
concept of reason.” To say that it is indemonstrable is to say that it is indeterminable. And
I take Kant™s point to be that ideas of reason, and only such ideas, have this property. This
is what distinguishes them from aesthetic ideas, on the one hand, and from concepts of
the understanding, on the other.
30 Allison (2001), 249“50.
Paul Guyer and Henry E. Allison

the morally good. The starting point of my analysis is a formalistic under-
standing of the symbolization relation, which I explicate in terms of
a re¬‚ective isomorphism. Simply put, beauty symbolizes the morally
good because re¬‚ection on the former may be viewed as a sensuously
directed analogue of re¬‚ection on the latter.31 I believe that this accords
with Kant™s well-known illustrations of symbolization in terms of the
constitutional monarchy“living organism and the absolute monarchy“
machine analogies, where the analogies concern not the objects but
our manner of re¬‚ecting upon them (5:352). It also applies to both
artistic and natural beauty, since re¬‚ection on both assumes the same
I further argue, however, that there is a problem in moving from these
conceptually based examples of symbolization to one in which sensible
concepts are not involved (as is the case with the beautiful). Accordingly,
it is at this point that I appeal to aesthetic ideas and their relation to
ideas of reason. My claim is that since aesthetic ideas, though they are
themselves products of the imagination, nonetheless gesture toward the
supersensible, they provide the vehicle for understanding how something
sensible (a beautiful object of nature or art) could symbolize something
supersensible (the morally good). Moreover, since Kant maintains that
all beauty (natural as well as artistic) expresses aesthetic ideas (5:320), I
contend that this helps to explain how all beauty symbolizes the morally
good, which is just what Kant explicitly claims in §59.
Guyer objects to this on two grounds. First, he accuses me of oversim-
plifying Kant™s conception of an aesthetic idea, since I suppose that “we
have an aesthetic idea whenever a manifold of imagination, because of
its essential indeterminability, connotes . . . the form of a rational idea”
(Guyer, this exchange). Although I would not put it in quite these terms,
I do think that we have aesthetic ideas when something imaginatively
apprehended in mere re¬‚ection points to something supersensible. But
I fail to see why this involves an oversimpli¬cation. Of course, Kant™s ini-
tial discussion of such ideas is in connection with ¬ne art and the activity
of genius; but he further states that natural beauty expresses such ideas
as well. And I do not see how this can be understood in other than the
functionalist terms that I use, which, incidentally, also explains the role of

31 Ibid., 255.
32 Guyer apparently denies this, suggesting at one point that it applies only to natural
beauty. And he further suggests that what the latter symbolizes is our autonomy, which
I do not believe can be equated with the morally good. See Guyer (1996), 268.
Dialogue 137

such ideas in symbolizing ideas of reason. If Guyer has any better account
of this dif¬cult matter, I would be interested in seeing it.
Guyer™s ¬nal objection appears to concern my move from the expres-
sion of aesthetic ideas to the symbolization of morality.33 This points to
a real problem in Kant, since aesthetic ideas are said to express ideas of
reason and not all of the latter are moral ideas, much less ideas of the
morally good. But far from being unaware of the problem, I attempted to
address it by emphasizing the formal nature of the symbolic relation. On
my view, this makes it possible to conceive of the re¬‚ection on expres-
sions of aesthetic ideas as symbolizing the morally good, even though
these ideas themselves do not express the latter.34 Although this requires
distinguishing between what an aesthetic idea expresses and what it sym-
bolizes, on my understanding of symbolization this is not problematic.
Moreover, it does not preclude some beautiful objects from also being
connected with morality in a more substantive sense. This occurs when,
in addition to occasioning a form of re¬‚ection isomorphic with re¬‚ection
on the morally good, it also evokes speci¬c moral ideas.
Although Guyer claims that this account fails, it is not clear to me from
his criticisms that he has understood it correctly. Of course, the fault may
be my own for failing to express myself with suf¬cient clarity. And if this is,
in fact, the case, I hope that these remarks, sketchy as they undoubtedly
are, help to clarify my position and highlight the differences between our

33 Actually, he expresses his objection by saying that “even if we accept the claim that the
concept of the harmony of the faculties is itself an idea of reason, it still would not follow
that it is ipso facto a morally signi¬cant idea of reason” (26). The problem with that,
however, is that I do not af¬rm the antecedent.
34 Allison (2001), 261“3.

Intensive Magnitudes and the Normativity of Taste

Melissa Zinkin

What distinguishes a judgment of taste from a cognitive judgment?
According to Kant, both forms of judgment are normative and both rely
upon the transcendental faculties of imagination and understanding for
their normativity. Yet one refers to a subjective feeling of pleasure, while
the other refers to an object. In Kant™s account, it is the role of the imag-
ination that differs in aesthetic judgments and cognitive judgments. In
cognition, the imagination is subject to the rules of the understanding
with which its relation is “objective” (XX:223).1 In a judgment of taste,
the imagination™s relation to the understanding is “subjective” (XX:223)
because the imagination is not referred to a concept of the understand-
ing, but to the subject and his or her feeling of pleasure or displeasure
(V:204). It is also a relation of “free lawfulness” (V:240).
But why is the imagination is subject to the understanding in one
instance and not in the other? Many commentators have attempted to
explain the freedom of the imagination from concepts in judgments of
taste by ¬nding places in Kant™s theoretical philosophy that indicate that
there could be a form of synthesis without a concept.2 They point to
Kant™s discussion of the “threefold synthesis” in the ¬rst edition of the
Critique of Pure Reason, where it is only by the last of these syntheses, “the
synthesis of recognition in the concept,” that the manifold is uni¬ed by a
concept. This allows for the possibility that there can be a synthesis of the

1 All references to Kant™s works are given by volume and page number of the Akedemie
edition (see References), except for citations to the Critique of Pure Reason, which utilize
the customary format of “A” and/or “B” to refer to the ¬rst and/or second edition.
2 See, for example, Guyer (1997a), 75“6, and Zammito (1992), 65.

Intensive Magnitudes and the Normativity of Taste 139

imagination prior to that which is achieved by concepts and that judg-
ments of taste represent this arrested stage of a threefold process toward
cognition. Or commentators refer to the “judgments of perception” in
the Prolegomena, which “require no pure concept of the understanding,
but only the logical connection of perception in the thinking subject”
(IV:298).3 Still, merely to show that it is possible for there to be a form
of synthesis that is not determined by concepts is not to give a reason for
why such a synthesis occurs. Any account of Kant™s argument for a prin-
ciple of judgments of taste must be able to provide a reason why, in these
judgments, the imagination is not restricted by a concept, in order to
also be able to explain why the freedom of the imagination is an essential
component of aesthetic judgments.
In what follows, I argue that the imagination is not restricted to a
particular rule of cognition in judgments of taste because the form of
intuition by which it apprehends the manifold is different. In cognitive
judgments, the determination of the manifold by a concept requires that
this manifold be apprehended in an extensive form of intuition. However,
in judgments of taste, it is because the manifold is intuited in an intensive
form of intuition that it cannot be subsumed under any determinate
concept. Furthermore, I conclude, the sensus communis, which, according
to Kant, is what makes possible normative judgments of taste, is none
other than the a priori form of sense that makes possible the intuition of
intensive magnitudes constructed by the imagination. By understanding
judgments of taste as involving what is apprehended in an intensive form
of intuition, and by understanding the sensus communis as an a priori
form of sensibility, I believe it is possible to account for three of the
most important (and problematic) distinctive features of judgments of
taste: (1) They are not determined by a concept. (2) They give rise to
a distinctive kind of pleasure, the pleasure we take in a beautiful object.
(3) They make a subjective claim to universal validity.
My argument proceeds as follows: Section 1 explains the distinction
between intensive and extensive magnitudes in Kant. Section 2 argues
that, for Kant, concepts can only be applied to what is apprehended in
an extensive form of time. Section 3 focuses on the Critique of the Power
of Judgment and shows that there, the imagination “in its freedom” is
the imagination that apprehends the manifold intensively and is thus
“free” from determination by concepts. Section 4 argues that the sensus
communis is the a priori form of intensive magnitudes and is therefore

3 See Uehling (1971), 57.
Melissa Zinkin

the basis for the claim of a subjective form of universal agreement that is
distinctive of judgments of taste.

1. intensive and extensive magnitudes
In this section, after brie¬‚y discussing extensive magnitudes, I will explain
Kant™s view of intensive magnitudes. My account will focus primarily on
the Critique of Pure Reason, since my aim is ultimately to compare the
presentation of intensive magnitudes in the ¬rst Critique with that in
the third. For Kant, an intensive magnitude (intensive Gr¨ße) is a qual-
ity. Quantitative distinctions among qualities that are the same, such as
distinctions between different shades of the same color blue, are mea-
sured in terms of degrees of intensity.4 This is in contrast to an extensive
magnitude, which is measured in terms of the addition of homogeneous
units. Because intensive magnitudes are qualities, Kant refers to them
primarily in his discussion of the categories of quality, their schematiza-
tion, and the principle of the Anticipations of Perception. Although his
discussion of intensive magnitudes in the Critique of Pure Reason implies
that Kant assumes that they have an a priori ground, I argue that in this
work he does not provide an a priori form that would make intensive
magnitudes possible.

1.1. Extensive Magnitudes
An extensive magnitude is one “in which the representation of the parts
makes possible the representation of the whole (and therefore necessar-
ily precedes the latter)” (A162). In the Transcendental Aesthetic of the
Critique of Pure Reason, Kant describes both space and time as extensive
magnitudes and thus as having the essential characteristics of extensivity:
homogeneity and additivity. Homogeneity is the property of being of uni-
form quality throughout, and additivity is the property by which some-
thing can be measured in terms of the addition of units that are its parts.
The length of something, for example, is the sum of the length of its
parts. By contrast, an intensive magnitude such as density is not the sum
of the densities of its parts.5
Because Kant views space and time as extensive, he argues that the pure
schema of magnitude, or quantity, is number. The schema of magnitude

4 Maier (1930), 35.
5 Brittan (1986), 77.
Intensive Magnitudes and the Normativity of Taste 141

is that by which the categories of quantity can be applied to the intuition
of time and hence to objects of experience. Since time and space are
extensive, that is, homogeneous and additive, their quantity, and, indeed,
the quantity of anything that is spatial or temporal, thus ends up being
measured in terms of numerical units. Kant writes:
The pure image of all magnitudes (quantorum) for outer sense is space; for all the
objects of the senses in general, it is time. The pure schema of magnitude (quantitas),
however, as a concept of the understanding, is number that is a representation
that summarizes (zusammenfaßt) the successive addition of one (homogeneous)
unit to another. This number is nothing other than the unity of the synthesis of


. 4
( 10)