. 7
( 10)


elicits pleasure is the free play and thus the mutual enhancement of the
cognitive capacities (imagination and understanding) in the apprehen-
sion of the object, together with the feeling that such a free play, and the
feeling it elicits, can be shared by all. The judging person™s state of mind is
therefore itself “purposive, without the representation of a purpose.” The
mental activity at work in apprehending the object judged to be beautiful
is accompanied by the feeling that a purpose is satis¬ed by it: The purpose
that the mind be precisely in the state it is in. And yet, here again we have
no concept of how such a purpose is satis¬ed. Like the form of the object,
the state of mind is purposive (it satis¬es a purpose, that of maintaining
the mind precisely in the state it is in) without the representation of a
purpose (i.e., without any determinate concept of this purpose).
This twofold purposiveness “ of the object, of the mental state itself “
explains, I think, the title of the third moment of the Analytic of the
Beautiful: “Third moment of judgments of taste, according to the rela-
tion of the purposes which in them are taken into consideration.” The
relation expressed in an aesthetic judgment is that of the purposiveness
expressed in the predicate to the purposiveness expressed in the sub-
ject. A purposiveness is expressed in the predicate because the pred-
icate ˜beautiful™ expresses the fact that a pleasure is elicited by the
universal communicability of the mutually enhancing play of the imagi-
nation and the understanding. This purposiveness has its ground in the
purposiveness of the subject of the judgment: the “purposiveness without
a purpose” of the apprehended (synthesized) form of the intuited object.
B´atrice Longuenesse

If this is correct, then the judgment of taste is the culminating point
of the Copernican revolution that began with the ¬rst Critique. For the
ground of the assertion of the predicate in the judgment of taste is the
intuited form of the object precisely insofar as it is synthesized by the subject.
So in the object, what grounds the assertion of the predicate ˜beautiful™
are just those features that depend on the synthesizing activity of the subject.
This point is con¬rmed if we now consider the implicit judgment
embedded in the predicate of the judgment of taste. I suggested earlier
that the predicate ˜beautiful™ might be explained in the following way:
˜Beautiful™ means “such that apprehending it elicits in me a pleasure
such that all judging subjects, in apprehending this same object, ought
to agree with my judgment.” The implicit judgment embedded in the
predicate (“all judging subjects, in apprehending this same object, ought
to agree with my judgment”) is a categorical judgment: The ground of
predication is to be found in the subject of the judgment, “all judging
subjects.” And yet that ground is not to be found in the concept of a judg-
ing subject: It is not by virtue of a character I know to belong universally
to all judging subjects that I claim that all of them ought to agree with my
judgment. Nor is the ground of the predication to be found in my empir-
ical knowledge of judging subjects. Rather, the ground for attributing the
predicate “ought to agree with my judgment” to all judging subjects (or,
in Kant™s terms, to “the whole sphere of those who judge”) is the capacity
I attribute to all of those who judge to experience the very same feeling
I presently experience. And my only ground for attributing to them this
capacity is the feeling itself as I experience it.
Let me recapitulate again. I have argued that according to the moment
of relation, the ground of the assertion of the predicate ˜beautiful™ is the
purposiveness without a purpose of the form of the apprehended object.
This purposiveness consists in the form™s capacity to elicit the mutually
enhancing play of imagination and understanding in the apprehending
subject. But the form of the object elicits such a mutually enhancing play
of cognitive capacities only because it is a synthesized form, a form that
is apprehended as the particular form it is only by virtue of the mental
activity of the apprehending subject. Thus what in the representation of the
object grounds the assertion of the predicate beautiful is its dependence on the
mental activity of the subject. I have also argued that the implicit judgment
embedded in the predicate of the aesthetic judgment (“all judging subjects,
upon apprehending this object, ought to experience the same feeling and
thus agree with my judgment”) is grounded on the capacity I postulate
in all judging subjects (and indeed, as we shall see, demand of them) to
Kant™s Leading Thread 213

experience the free play of their cognitive capacities I myself experience
in apprehending the object, and thus to share my feeling and agree with
my judgment.
We will have to keep these two features in mind to understand Kant™s
view of the modality of judgments of taste, to which I now turn.

4. the subjective necessity of judgments of taste
The modality of a judgment of taste, says Kant, is that of necessity. But what
is ˜necessary™? Is it the connection between the predicate and the subject
in the manifest judgment about the object (“this rose is beautiful”)? Or
is it, rather, the connection between the predicate and the subject in
the implicit judgment about the judging subjects (“all judging subjects,
upon apprehending this rose, ought to experience the same pleasure
and thus agree with my judgment”)? If the former, what is said to be
necessary is the connection between the object considered in its form
and the pleasure I feel in apprehending it. If the latter, what is said to be
necessary is the connection between the obligation implicitly assigned to
all judging subjects (they “ought to agree with my judgment”) and these
judging subjects considered simply as such.
I submit that Kant wants to assert the necessity of both connections. He
asserts at the outset that the relation between the object and the satisfaction it
elicits is necessary: “Of the beautiful, . . . one thinks that it has a necessary
relation to satisfaction” (§18, 237). But he then immediately goes on to
assert the necessity of the agreement of all judging subjects with my judgment,
taken as the example of a rule:

[The] necessity that is thought in an aesthetic judgment . . . can only be called
exemplary, i.e. a necessity of the assent of all to a judgment that is regarded as an
example of a universal rule that one cannot produce. (Ibid.)

Note that the situation here is not parallel to that of quantity. The
quantity of the manifest judgment about the object was different from
that of the implicit judgment about the judging subjects (the former was
singular, the latter universal). In contrast, here the necessity of the lat-
ter (the implicit judgment about the judging subjects) seems to ground
the necessity of the former (the manifest judgment about the object):
Because all judging subjects ought to judge as I do, the relation of the
predicate ˜beautiful™ to the subject of the manifest judgment can legiti-
mately be asserted as necessary. We can understand why this is so: What
is beautiful is the object as apprehended, and being beautiful is the same as
B´atrice Longuenesse

being judged to be beautiful. To say that all judging subjects ought necessar-
ily to agree with my judgment is to say that the object ought necessarily to be
judged beautiful or that the connection between the predicate ˜beautiful™
and the object is necessary.
This still does not tell us, however, how we should understand this
modality of necessity. Is the necessity of the connection between “all judg-
ing subjects” and “ought to agree with my judgment” to be understood on
the model of the subjective necessity of judgments of experience (because
I claim objective validity for my judgment, I claim that all judging sub-
jects ought to agree with my judgment)? Or is it to be understood on the
model of a moral imperative: “All rational beings ought to act in such
and such a way” (under the categorical imperative of morality)? Similarly
“all judging subjects ought to judge as I do”?
Kant™s response, I suggest, is that both models are relevant. Indeed,
both serve to clarify the crucial notion of a sensus communis on which Kant
will later base his deduction of judgments of taste, namely, his justi¬cation
of their claim to (subjective) universality and necessity.
Already in §20 of the fourth moment, Kant states that the subjective
necessity of the judgment of taste is af¬rmed only under the condition
that there be a common sense, Gemeinsinn. By ˜common sense™ he means
“not any external sense, but rather the effect of the free play of our cog-
nitive powers” (§20, 238), that is to say, the feeling that we have of this
free play and of its universal communicability. This is in direct continu-
ity with what was said in the ¬rst two moments of the Analytic of the
Beautiful. As we saw, according to the ¬rst moment, the aesthetic plea-
sure is a disinterested pleasure elicited in the mind by its own activity in
apprehending the object. According to the second moment, this activity
is one of free play of imagination and understanding, and the pleasure
expressed by the predicate beautiful is a both a ¬rst-order pleasure taken
in this free play and a second-order pleasure in the universal communi-
cability of the feeling thus elicited. The agreement of imagination and
understanding in cognition and the universal communicability of that
agreement provide an argument for at least supposing the possibility of a
similar universal communicability of the state of mind in the free play of
imagination and understanding, and thus a sensus communis aestheticus as
the ground for the aesthetic pleasure expressed in the predicate ˜beau-
tiful™. In this context, the obligation assigned to all judging subjects to
agree with my judgment is not analogous to a moral obligation. Rather, it
is analogous to the obligation to submit oneself to the norm of truth (the
rule-governed agreement between imagination and understanding) in
Kant™s Leading Thread 215

cognitive judgments. And indeed, it is by drawing on the a priori agree-
ment of imagination and understanding in cognition that Kant initially
justi¬es the supposition of a common sense as the ground of aesthetic

One will thus with good reason be able to assume a common sense [so wird dieser
mit Grunde angenommen werden k¨nnen], and without appealing to psychological
observations, but rather as the necessary condition of the universal communi-
cability of our cognition, which is assumed in every logic and every principle of
cognitions that is not sceptical. (§21, 239)

But there is something surprising about this justi¬cation. For as we saw
in discussing the second moment, what grounds the subjective universal-
ity and thus also the subjective necessity of cognitive judgments in the ¬rst
Critique is not the free agreement of imagination and understanding, but
their agreement for the production of concepts, that is to say, according to the
rules imposed by the understanding. The fact that there is such an agreement
(not free, but ruled by the understanding) may perhaps give us reason
to believe in the possibility of a similar agreement even without a con-
cept. But that does not give us suf¬cient grounds for af¬rming that such
an agreement exists, and still less for af¬rming that it necessarily exists.
Indeed Kant is more cautious when he writes:

This indeterminate norm of a common sense is really presupposed by us: our
presumption in making judgments of taste proves that. Whether there is in fact
such a common sense, as a constitutive principle of the possibility of experience,
or whether a yet higher principle of reason only makes it into a regulative principle
for us ¬rst to produce a common sense in ourselves for higher ends, thus whether
taste is an original and natural faculty, or only the idea of one that is yet to
be acquired and is arti¬cial, so that a judgment of taste, with its requirement
[Zumuthung] of a universal assent, is in fact only a demand of reason to produce
such unanimity in the manner of sensing, and whether the ought, i.e. the objective
necessity of the convergence of everyone™s feeling with that of each, signi¬es only
the possibility of such agreement, and the judgment of taste only provides an
example of the application of this principle “ this we neither want nor are able
yet to investigate here; for now we have only to resolve the faculty of taste into
its elements and to unite them ultimately in the idea of a common sense. (§22,

As we can see, here the model for the subjective necessity of the judg-
ment of taste is no longer the claim to necessary agreement proper to
a judgment of experience, but rather the demand of moral duty. The a
priori agreement of imagination and understanding in cognition allows
us only to accept as possible the ˜common sense™ that would ground
B´atrice Longuenesse

aesthetic judgment; but the request of a universal agreement of ratio-
nal agents under the moral law now appears to be a ground to demand
that we cultivate in ourselves the capacity to develop a common sense. As
we saw, already in the course of the second moment, Kant maintained that
we postulate the “universal voice” under which we formulate a judgment
of taste (cf. §8, 216).
Kant does not always clearly distinguish between the mere possibility
of an agreement of everyone with my own evaluation, based on the free
play of imagination and understanding, and the postulated existence of
this agreement, as a capacity that each judging subject has an obligation
to develop in himself and to demand of others. But it is important to keep
this distinction in mind in order to free Kant of the burden of an all too
evident objection, which we have already encountered in our examina-
tion of the second moment: If the sensus communis, understood generically
as the universally communicable agreement of imagination and under-
standing, is the common ground of cognitive judgments and aesthetic
judgments, why isn™t every cognitive judgment the occasion of aesthetic
pleasure? On the other hand, if there is merely a kinship, not a generic
identity, between the sensus communis that grounds judgments of taste (a
universally communicable free play and mutual enhancement of imagina-
tion and understanding in apprehending the object and re¬‚ecting upon
it, known by feeling) and the sensus communis that grounds judgments in
empirical cognition (a universally communicable agreement of imagina-
tion and understanding in apprehending the object and re¬‚ecting upon
it, known by virtue of the concepts that express it, and thus not free, but rule-
governed), why would the latter be a suf¬cient ground for admitting the
existence of the former? This objection falls if Kant™s argument for the
existence of a sensus communis grounding aesthetic judgments has the two
distinct steps mentioned earlier: (1) the universal communicability of the
state of mind in cognition shows that it is possible that the agreement of
the imagination and the understanding, even when it is not ruled and
re¬‚ected by concepts (when it is a free play eliciting a feeling of pleasure),
be universally communicable; (2) we demand that this agreement should
be universally communicable, and because we demand it, we make it “as
if a duty” to bring it about in ourselves and in others.
These two steps are expressed in the form of a question in the text
quoted earlier: Should we consider the sensus communis as a natural capac-
ity, or rather as the object of a higher demand of reason that we develop
this capacity in ourselves and in others? The two steps will be con¬rmed
Kant™s Leading Thread 217

and ampli¬ed in the deduction of the judgment of taste (although again
somewhat ambiguously). In the very short paragraph entitled “Deduction
of the judgment of taste” (§38), Kant asserts again that the claim to sub-
jective universality and necessity of our judgments of taste has the same
ground as the claim to subjective universality and necessity of judgments
of empirical cognition, justi¬ed in the ¬rst Critique. This is the ¬rst step in
the two-step argument summarized previously. In the next section (§40),
Kant adds:

If one was allowed to assume that the mere universal communicability of his
feeling must in itself already involve an interest for us (which, however, one is
not justi¬ed in inferring from the constitution of a merely re¬‚ective power of
judgment), then one would be able to explain how it is that the feeling in the
judgment of taste is required of everyone as if it were a duty (gleichsam als P¬‚icht
jedermann zugemutet werde). (§40, 296)

Here™s how I understand this passage: By itself, the “merely re¬‚ective”
use of the power of judgment, namely, the use in which the play of imag-
ination and understanding does not lead to a concept, would not suf¬ce
to explain why we demand of everyone, as if it were a duty, that they
share our pleasure in the object we judge to be beautiful. Something
else is needed in order to explain this demand, something that would
make the sensus communis not only a Gemeinsinn (a common sense) but
a gemeinschaftlicher Sinn: a sense by virtue of which we take ourselves to
belong to a community of judging subjects. This something else is an
interest that we take not in the object of the judgment (that possibility has
been excluded in the course of the ¬rst moment), but in the very fact of
the universal communicability of the judgment, that is to say, in the very fact
that through this shared judgment we progress toward a community of
judging subjects.
Indeed, in the next two sections Kant sets about explaining successively
(1) that there is an empirical interest attached to the judgment of taste,
that of developing sociability in ourselves, and (2) that there is an
intellectual interest (an interest we have insofar as we are rational) in
recognizing in nature and in ourselves the sensible sign of a common
supersensible ground. In recognizing this supersensible ground, it is our
own moral nature that we also recognize, and this makes the ought in
“All judging subjects ought to agree with my judgment” closer to a moral
“ought” than to the obligation assigned to cognitive subjects, to yield to
the norms of truth in empirical judgments.
B´atrice Longuenesse

There is a caveat here. Only the beautiful in nature can give rise to
such an intellectual interest. For only judgments about nature serve the
interest of morality by pointing to the supersensible ground common to
nature and to us. As for the beautiful in art, at most it serves the interest
we have in the development of our natural tendency toward sociability,
which is an empirical interest, grounded in the empirical characteristics
of humanity as a natural species (§41, 96“7). Does this mean that only
judgments of beauty in nature have the modality of necessity Kant tries to
justify in his deduction of judgments of taste? This would be surprising,
for all the examples Kant gives to illustrate the demand of a universal
agreement with our judgments of taste concern the beautiful in art (see
§§32“3). How are we to understand this apparent inconsistency? I think
there are two answers.
The ¬rst can be found in the relation between sensus communis and
Aufkl¨ rung. Kant emphatically endorses the three mottoes he attributes
to Aufkl¨ rung: to think for oneself, to think by putting oneself in the
position of all other human beings, to think always consistently (see §40,
294). Now, the universal communicability of judgments of taste, whether
they apply to nature or to art, makes them uniquely apt to satisfy the
¬rst two maxims of the Aufkl¨ rung. And in their case, the third maxim
is irrelevant: Any singular aesthetic judgment carries its own exemplary
norm and thus is in no need of consistency with other judgments. In
short, in the case of aesthetic judgments, the mere possibility of universal
communicability of a feeling becomes the normative necessity of a duty to
create the conditions of such universal communicability. And this applies
to our experience of beauty in art just as much as in nature.
The second answer lies in Kant™s conception of genius as a state of mind
in which “nature gives the rule to art” (§46, 307). Relating artistic creation
to genius de¬ned in this way means giving judgments of taste applied to
works of art their full share in the relation to the supersensible that is the
ground of the subjective universality and necessity of aesthetic judgments
applied to nature. This point is con¬rmed in the dialectic of the critique
of taste, where Kant describes genius as the “faculty of aesthetic ideas”
(§57, 344). An aesthetic idea, he says, is a sensible presentation of the
supersensible, of which we neither have nor can have any determinate
concept. Despite Kant™s very Rousseauian suspicion of art and its relation
to the ends of self-love, it remains that the beautiful in art, insofar as art is
the creation of genius, lends itself to the same demand for the universal
and necessary agreement of all judging subjects as does the beautiful in
Kant™s Leading Thread 219

Now we may well ¬nd that we are asked to accept too much here. To
have to suppose a consciousness of the supersensible ground common
to the object and to ourselves, as the ground of the subjective univer-
sality and necessity of the aesthetic judgment, is more than most of us
can swallow. However, Kant™s analysis of the two judgments present in the
judgment of taste “ the manifest judgment about the object, the implicit
judgment about the judging subjects “ may lend itself to a lighter read-
ing. One might accept the striking combination of a normative judgment
about the judging subjects (expressed in the predicate of the judgment of
taste as I have proposed to develop it) and a descriptive judgement about
the object considered in its form (expressed in the manifest judgment of
taste, “this x is beautiful”) while rejecting Kant™s appeal to the supersen-
sible as the ultimate ground of judgments of taste. One would then no
longer have any reason to grant any privileged status to the beautiful in
nature over the beautiful in art, since the main reason for that privilege
seems to be that nature, not human artifact, is a direct manifestation of
the supersensible that grounds aesthetic experience. In accounting for
the speci¬c features of judgment of taste, one may still maintain that the
mere possibility of universally sharing aesthetic pleasure becomes a norma-
tive necessity, an obligation made to all human beings to take their part
in the common effort to constitute humanity as a community of judging
subjects, beyond the particular limitations of each historically and bio-
graphically determined sensing, feeling, emotional access to the world of
sensory objects. This is, I think, the lasting legacy of Kant™s view.
part three


Re¬‚ection, Re¬‚ective Judgment, and
Aesthetic Exemplarity

Rudolf A. Makkreel

Although Kant considers a pure aesthetic judgment to be re¬‚ective in
nature, he is also able to account for a wider range of prejudgmental
and judgmental aesthetic responses. In the case of works of art, I will
show that he allows for both re¬‚ective judgments about their beauty and
determinant judgments about their meaning. But such an intersection of
re¬‚ective and determinant judgments should not be seen as supporting
the conclusion that their judgmental functions merge. Since determi-
nant judgment proceeds from a given universal to particulars, it clearly
involves a subordinating mode of thought. I will argue, however, that re¬‚ec-
tive judgment, which tends to begin with particulars, is a coordinating
mode of thought. Determinant judgment appeals to universals to either
describe the nature of particular objects or explain their behavior by
subsuming them under the laws of the understanding. Re¬‚ective judg-
ment, by contrast, is an expansive mode of thought that appeals not just to
the understanding, but to reason as a framework for interpreting particu-
lars. Because Kant calls re¬‚ection the power to compare a representation
either with other representations or with our own cognitive powers, I want
to underscore that re¬‚ective judgment is not so much about objects per
se as about their relations to us. I will also make a case for the thesis that
re¬‚ective judgment is orientational in that it enables the apprehending
subject to put things in context while discerning his or her own place in
the world.1

1 See Makkreel (1990); the relation between re¬‚ective judgment and reason is exam-
ined in Chapter 6 and the relation between re¬‚ective judgment and orientation in
Chapter 8.

Rudolf A. Makkreel

In the second half of this essay, I will claim that when the re¬‚ective
power of judgment (Urteilskraft) functions aesthetically in relation
to sense and feeling, it becomes a mode of evaluating or assessing
(Beurteilung) that has normative force. I will explore this normativity
of aesthetic judgment in relation to the idea of exemplarity. How is it
possible for certain particulars to become exemplars, and to what extent
can they orient us in making aesthetic assessments?

1. logical and transcendental re¬‚ection and their
relation to re¬‚ective judgment
Kant™s assertion in the First Introduction to the Critique of the Power of
Judgment that when judgment applies universal concepts of nature “its
re¬‚ection is at the same time determinative”2 is used by B´ atrice Longue-
nesse to align the functions of re¬‚ective judgment and determinant judg-
ment more closely than has been usual. Accordingly, she claims that there
are “judgments relating to the sensible given, which are not merely re¬‚ec-
tive, but determinative as well.”3 A successful re¬‚ective judgment “consists
not only in forming empirical concepts, but also applying them,”4 which
is a determinative activity. It follows from this interpretation that the
merely re¬‚ective judging involved in the claims of taste produces failed
judgments because they do not “determine” anything. Indeed, Longue-
nesse writes: “What makes judgments merely re¬‚ective is that in them, the
effort of the activity of judgment to form concepts fails.”5
It strikes me as hermeneutically implausible to claim that the most
explicit example of re¬‚ective judgment given by Kant, namely the aes-
thetic judgment, is a de¬cient version, whereas some other operations
discussed elsewhere are successful versions. It behooves us to see whether
there isn™t a better way to relate re¬‚ective and determinant judgment.
I will argue that it is re¬‚ection “ not re¬‚ective judgment, as Longue-
nesse claims “ that can lead to determinant judgment. Moreover, the
re¬‚ection that Kant relates to judgment in the Critique of the Power of
Judgment is not a “re¬‚ection on the sensible,”6 as suggested by Longue-
nesse, but a “re¬‚ection about the concept of nature in general.”7 Wanting

2 Ak 20:212.
3 Longuenesse (1998), 164.
4 Ibid., 165.
5 Ibid., 164. I am grateful to Eric Wilson for pointing me to this passage and its import.
6 Ibid., 165.
7 Kant, First Introduction to the Critique of the Power of Judgment, Ak 20:212.
Re¬‚ection, Re¬‚ective Judgment, and Aesthetic Exemplarity 225

to transport re¬‚ective judgment from the Critique of the Power of Judgment
back to the Critique of Pure Reason, she asserts that the application of the cate-
gories already requires re¬‚ective judgment. This application is said to pre-
suppose “a progress from sensible representations to discursive thought:
the formation of concepts through comparison/re¬‚ection/abstraction,
which is just what re¬‚ective judgment is.”8 I disagree: When Kant speaks of
re¬‚ection in relation to the formation of empirical concepts, this is not
yet re¬‚ective judgment; nor is this re¬‚ection a condition for the applica-
tion of the categories. Being a priori and formal, categories such as sub-
stance and causality are applicable to all possible phenomenal objects.
No special re¬‚ective or technical skill is necessary for their application.
Whereas general logic leaves us without any guidance for applying con-
cepts, transcendental logic is distinctive in supplying this for its concepts.
Kant writes in the Critique of Pure Reason: “in addition to the rule . . . which
is given in the pure concept of the understanding, it can at the same
time indicate a priori the case to which the rules ought to be applied.”9
This application is spelled out in the chapter on schematization, not in
the Amphiboly appendix on concepts of re¬‚ection, where Longuenesse
looks for it.10
By linking Kant™s discussions of re¬‚ection in the lectures on logic,
concepts of re¬‚ection in the Critique of Pure Reason, and re¬‚ective judg-
ment in the Critique of the Power of Judgment, Longuenesse attempts to
extract a common procedure whereby “sensible objects are re¬‚ected under
concepts.”11 In contrast, I will analyze these same three discussions in Kant
with the intent of showing what is distinctive about each. They show
some similarities, but they do not add up to a procedure of re¬‚ective
We can differentiate between re¬‚ection and re¬‚ective judgment by
examining Kant™s ¬nal J¨ sche Logic. Re¬‚ection as it relates to concept for-
mation is treated in Sections 5 and 6, whereas re¬‚ective judgment is not
examined until Sections 81 to 84. Re¬‚ection is related to the discursive

8 Longuenesse (1998), 164“5.
9 Critique of Pure Reason, A135/B174.
10 I refer to the twofold claims made in Chapter 6 of Kant and the Capacity to Judge that the
concepts of re¬‚ection of the Amphiboly appendix “express the rules for the re¬‚ective
genesis of all concepts or . . . judgments” (Longuenesse 1998, 131) and that these come
“prior to the subsumption of empirical objects under categories” (166).
11 Longuenesse (1998), 115.
12 Despite my criticisms of this, there is much to admire in the impressive ways Longuenesse
has been able to elaborate the relations between Kant™s categories and the traditional
table of logical or discursive judgments.
Rudolf A. Makkreel

tasks of the understanding, whereas re¬‚ective judgment is considered a
mode of inference geared to the more comprehensive aims of reason.
Re¬‚ection as a logical act of the understanding considers the form of
a concept “subjectively; not how it determines an object through a mark,
but only how it can be related to several objects.”13 This kind of logical
re¬‚ection, as de¬ned in Sections 5 and 6, is merely about subjective repre-
sentations not about how they might produce determinate cognition of
actual objects. Re¬‚ection as a purely discursive logical activity stays within
the realm of mental representations and does not yet broach the realm
of transcendental logic, which makes possible the cognition of objects.
In Section 6 of the J¨ sche Logic, re¬‚ection is described as one of three
logical acts of the understanding that can produce concepts in terms of
form. It is preceded by comparison and followed by abstraction. Compar-
ison relates the representations to the unity of consciousness, re¬‚ection
then considers “how various representations can be encompassed in one
consciousness,”14 and, ¬nally, abstraction separates out “everything else
in which the given representations differ.”15 Comparison and re¬‚ection
both involve a “zusammenhalten”16 “ they are ways of holding represen-
tations together without subordinating them to some de¬ning mark of
an object. Kant gives the example of generating the concept tree from
comparing and re¬‚ecting on representations of a spruce, a willow, and a
linden: “By ¬rst comparing these objects with one another, I note that they
are different from one another in regard to the trunk, the branches, the
leaves, etc.; but next I re¬‚ect on that which they have in common among
themselves, trunk, branches, and leaves themselves, and I abstract from
the quantity, the ¬gure, etc., of these; thus I acquire a concept of a tree.”17
This concept can be referred to many objects, but does not yet constitute
the determining mark under which objects can be subsumed to be cog-
nized. What this means is that the re¬‚ected concept simply analyzes what
several subjective representations have in common (gemein); it is not the
kind of synthetic concept that can determine an object in the universal
(allgemeine) way that is necessary for objective cognition. The resulting
analytical concept of tree is more general in form than the concept of
spruce, but it is still a weak representational form, not yet a rule-giving
content or de¬ning mark.

13 Kant (1992), 591.
14 Ibid., 592.
15 Ibid.
16 See Kant, Re¬‚exionen zur Logik, 2878; Ak 16:556.
17 Kant (1992), 592; Ak 9:94.
Re¬‚ection, Re¬‚ective Judgment, and Aesthetic Exemplarity 227

A constant theme of all of Kant™s claims about re¬‚ection is its subjec-
tive basis. This is also true for Kant™s ¬rst Critique discussion of concepts
of re¬‚ection, which “only serve to describe in all its manifoldnesss the
comparison of representations that is prior to the concept of things.”18
Kant™s concern with re¬‚ection here is not about arriving at knowledge of
objects, nor is it about the logical re¬‚ection that goes into the form of con-
cepts. It is about transcendental re¬‚ection, which considers representations
in relation to the cognitive faculty to which they belong. Whereas logi-
cal re¬‚ection relates representations to consciousness in general, tran-
scendental re¬‚ection relates them to the faculty that gave rise to them.
This new mode of re¬‚ection distinguishes representations according to
whether they belong to sensibility or to understanding. Once this subjec-
tive differentiation has been made about representations, we can then
discern whether their relation to each other is one of identity or differ-
ence, agreement or opposition, whether they are related internally or
externally or in terms of matter or form.19 Kant™s examination of tran-
scendental re¬‚ection uses these four pairs of concepts of re¬‚ection not
to give determinant knowledge of objects as categories or empirical con-
cepts do, but only to sort out what kind of object a representation can
be about. Re¬‚ection now becomes orientational in that as it compares
representations, it also assigns them their place in relation to possible
objects of sense or understanding.
We have indicated that transcendental re¬‚ection is about relational
concepts that come in coordinated pairs like inner and outer, matter
and form. Although concepts such as matter and form are compara-
tive in being reciprocally related, they are also contrastive and demand
a weighing of alternatives. Thus, according to whether representations
are located in sense or in the understanding, the priority of form or
matter will differ. For phenomenal objects of sense, form precedes
matter, but for intellectual objects of the understanding, matter pre-
cedes form. This re¬‚ective way of orienting ourselves to our represen-
tations is important, according to Kant, if we are to avoid the one-sided
metaphysical stances of Locke and Leibniz, respectively. Locke™s mis-
take was to “sensitivize” concepts, while Leibniz made the opposite error
of “intellectualizing” appearances.20 Concepts of re¬‚ection do not add
to our knowledge of objects themselves, but consider them in terms

18 Critique of Pure Reason (1997a), A269/B325.
19 Ibid., A263/B319“A266/B322.
20 Ibid., A271/B327.
Rudolf A. Makkreel

of how we should represent them. They establish what Kant calls a “tran-
scendental topic”21 that differentiates between representations accord-
ing to whether we should refer them to a phenomenal or noumenal
In Section 2 of the Introduction to the third Critique, Kant moves from
this abstract topic that guides re¬‚ection to a more concrete topology
for re¬‚ective judgment by distinguishing different regions in which we can
locate or frame objects. Thus, when we refer concepts to objects, we
delineate either a ¬eld (Feld), a territory (Boden), a domain (Gebiet), or
an abode (Aufenthalt).22 Kant writes that “insofar as we refer concepts to
objects without considering whether or not cognition of these objects is
possible, they have their ¬eld, and this ¬eld is determined merely by the
relation that the objects of these concepts have to our cognitive powers
in general.”23 Here again re¬‚ection relates objects to the subject. When
we think of objects without determining whether they can be actualized
in experience, we locate them as part of a ¬eld. They are conceived as
logically possible, but not yet as transcendentally possible or actualizable
for cognition. A ¬eld is the most neutral way of framing objects and
allows us to regard a ghost as belonging to the context of what can be
conceived, even though it is an illusory object. When a concept refers to
or means (bedeutet) an actual sensible object, then we can say that it has a
territory. Nature as experienced by us can be said to be our territory. This
territory of nature is a domain to the extent that the concepts legislate
to it. Categorial concepts such as causality have their domain in nature
because they necessarily apply to it. Empirical concepts simply have their
abode in nature because we have derived them from what we contingently
¬nd there.24
We may elaborate Kant™s orientational or regional distinctions between
¬eld, territory, domain, and abode by correlating them with the possible,
the actual, the necessary, and the contingent, respectively. This enables us
to speak of the ¬eld of the possible, the territory of the actual, the domain
of the necessary, and the abode of the contingent as four modal relations
that objects can have to us. When we locate an object in a territory or
a domain, we are in a position to make a determinant judgment about
it. By contrast, the ¬eld of the possible is merely the correlate of logical

21 Ibid., A268/B324.
22 Critique of the Power of Judgment, Ak 5:174.
23 Ibid.
24 Ibid.
Re¬‚ection, Re¬‚ective Judgment, and Aesthetic Exemplarity 229

re¬‚ection, and the abode of the contingent is the correlate of re¬‚ective
The latter relation becomes intelligible once we recognize that re¬‚ec-
tive judgment has the capacity to discern “lawfulness” in what is ordinarily
thought to be contingent.25 Whereas determinant judgment subsumes
particulars under already available universals, whether they be rules or
laws, re¬‚ective judgment seeks universality wherever we are still left with
a remainder of particularity. The abode of the contingent involves a col-
location of facts that we happen to come across and that demonstrate no
objectively necessary connection. What re¬‚ective judgment looks for then
is a subjective necessity. From the standpoint of the understanding, every
event is subsumable under some law, but the explanative laws of nature
could be so diverse that we could never grasp nature as a whole as anything
more than an aggregate. A systematic order of nature demands a ratio-
nal coherence that is intrinsically contingent from the standpoint of the
understanding. “The power of judgment presumes [it] of nature . . . only
for its own advantage” as a formal purposiveness of nature.26 Kant makes
it evident that this concept of a purposiveness of nature is nothing more
than a subjective mode of representing nature, or, to use more contem-
porary language, of interpreting it.
Being neither a concept of nature nor a concept of freedom, purposive-
ness is a mere re¬‚ective representational concept that has no rule-giving
content through which we could attain knowledge of any objects. “This
transcendental concept of a purposiveness of nature . . . attributes noth-
ing at all to the object (of nature), but rather only represents the unique
way in which we must proceed in re¬‚ecting on the objects of nature with
the aim of a thoroughly interconnected experience. Consequently, it is
a subjective principle (maxim) of the power of judgment.”27 The re¬‚ec-
tive principle of systematicity prescribes a purposiveness of nature that is
relative to the subject and its rational need for order. When we do ¬nd
such systematic unity we feel pleasure; here lies the connection between
re¬‚ective judgment and aesthetic judgment. Purposiveness as the lawful-
ness of the contingent is never predictable. It is something unexpected,
and the surprise in discerning it produces pleasure. Since there is always

25 This becomes evident in §§76 and 77 of the Critique of the Power of Judgment, where Kant
re¬‚ects on contingency at length and shows that what we humans call “purposiveness”
is the “lawfulness of the contingent” (§76, Ak 5:404).
26 Kant, First Introduction to the Critique of the Power of Judgment, Ak 20:204.
27 Kant, Critique of the Power of Judgment, 71; Ak 5:183.
Rudolf A. Makkreel

something presumptive about the principle of purposiveness, re¬‚ective
judgment can only prescribe it to itself, not to nature.28
When we return to Kant™s J¨ sche Logic and consider Sections 81“4, it
becomes even clearer that re¬‚ective judgment is more than the logical
re¬‚ection that goes into comparing and analyzing our representational
concepts. For re¬‚ective judgment is now explicated in terms of two modes
of inference. Whereas logical re¬‚ection in the service of the understand-
ing ¬nds what is common among given representations, re¬‚ective judg-
ment goes beyond such particular givens in search of a universal. Being
inferential, it is geared to reason. The search for universality involved
in re¬‚ective judgment proceeds “either 1) from many to all things of a
kind, or 2) from many determinations and properties, in which things
of one kind agree, to the remaining ones, insofar as they belong to the same
principle.”29 The ¬rst mode is an inductive inference, the second an infer-
ence from analogy in accordance with a “principle of speci¬cation.”30
Induction argues from the premise that many x™s are y to the conclu-
sion that all are y. Specifying analogies proceed from the partial simi-
larity of things to their overall similarity. Although both inferences are
oriented to reason, they are empirical and do not produce true univer-
sality. Induction really argues from particularity to “general rather than
universal propositions;”31 specifying analogies really proceed from parts
to wholes.
How then does this twofold characterization of re¬‚ective judgment in
the J¨ sche Logic relate to Kant™s discussions of re¬‚ective judgment in the
third Critique? Obviously, aesthetic judgments as functions of the power
of re¬‚ective judgment are not inductive inferences, because they remain
singular. When Kant claims the judgment “this rose is beautiful” to be uni-
versally valid, he is not generalizing that all roses are beautiful. Instead, an
aesthetic judgment about a particular rose indicates a subjective response
that is also claimed to be valid for others. The contingent pleasure that
I feel in contemplating the rose is judged to apply not just to me, but
to “us human beings.”32 The generalization made is not an inductive
prediction, but anticipates how we humans should respond.

28 Kant makes a stronger assertion in the case of teleological judgments. Here an organism
is described as functioning in such a way as to preserve itself (a regulative claim), and
we interpret this effect as its telos (a re¬‚ective claim valid for us humans).
29 Kant (1992), 626, Ak 9:132.
30 Ibid., 626, Ak 9:133.
31 Ibid., Ak 9:133.
32 Critique of the Power of Judgment, Ak 5:462.
Re¬‚ection, Re¬‚ective Judgment, and Aesthetic Exemplarity 231

If the aesthetic judgment is not inductive, is it then a specifying infer-
ence by analogy? This suggests a more promising alternative because
Kant does in fact appeal to re¬‚ective analogies when he explores the
contribution of aesthetic ideas to matters of taste. But the capacity to
appreciate beauty by no means requires us to specify the objects being
judged in such detail as to argue from partial similarity to an overall sim-
ilarity. Total similarity is more relevant to teleological judgment where
organisms are considered as integral systems in which the parts function
harmoniously and can maintain equilibrium. The re¬‚ective principle of
speci¬cation that guides questions of system “ whether they concern the
systematization of the laws of nature or some more limited teleological
questions about organisms “ seems to push the aesthetic apprehension
of re¬‚ective analogies a step further.
Re¬‚ective judgment in the Critique of the Power of Judgment, whether it is
used aesthetically or teleologically, is clearly comparative and coordina-
tive. It would be wrong, however, to assume that the coordination involved
in re¬‚ective speci¬cation means nothing more than placing things side
by side as a mere aggregate. According to Kant, coordination produces
an aesthetic-technical “order.”33 The re¬‚ective concern with coordina-
tion is to ¬nd “suf¬cient kinship”34 among empirical laws to allow them
to be part of a common system. To be sure, when relating laws within
a systematic framework, we must consider whether some empirical laws
may turn out to be species of higher generic laws. Kant™s full answer to the
problem of systematization requires us to approach it not merely as the
mechanical task of subordinating particular laws to already known higher
laws. Instead, the task of systematization is a technical one of adjusting
parts to wholes.35 The higher laws must themselves be revisable or speci-
¬able to make room for the lower laws. Thus Kant writes that we must
“make the universal concept more speci¬c by adducing or taking into
account (anf¨ hren) the manifold under it.”36 Kant here rejects the com-
mon assumption that speci¬cation applies merely to the particulars that
may be subordinated to a universal. Instead, the re¬‚ective speci¬cation
involved in systematization applies to the universal itself. It is the process
whereby a universal comes to re¬‚ect particulars by being coordinated
with them. In the end, coordination is less about lateral comparison than
about the mutual adjustment of the parts of a system.

33 See Kant, Re¬‚exionen zur Logik, Ak 16:121.
34 First Introduction to the Critique of the Power of Judgment, Ak 20:215.
35 See ibid., Ak 20:213f.
36 Ibid., Ak 20:215.
Rudolf A. Makkreel

Coordination as a feature of re¬‚ective judgment is to be distinguished
from mere juxtaposition on the one hand, and subordination on the
other. Re¬‚ective coordination can be characterized as a search for the
appropriate context in which it becomes possible to understand phenom-
ena in reciprocal terms, that is, not merely by means of relations of depen-
dence, but also as potentially interdependent. As part of a more general
process of orientation to the world at large, re¬‚ective coordination allows
for a differentiation of modes of intelligibility “ we have already seen this
in the regional distinctions between ¬eld, domain, territory, and abode.
These modalities cannot all be de¬nitively subordinated to each other.
In this section, we have seen that logical re¬‚ection holds representa-
tions together to discern an analytical unity, that transcendental re¬‚ec-
tion differentiates paired opposites, and that re¬‚ective judgment speci-
¬es the relation between universals and particulars. All these operations
function in a comparative and coordinative context and suspend the sub-
ordinative nature of synthetic cognition. Whereas subordinative deter-
mination establishes unidirectional, top-down relations of dependence,
re¬‚ective coordination is a process of reciprocal adjustment. For these
reasons, I ¬nd no basis in Kant for Longuenesse™s recurring language
of objects or sensible manifolds being “re¬‚ected under concepts.”37 We
should thus resist efforts to make re¬‚ective judgment part of the sub-
ordinative mode of thought that characterizes determinant judgment.
There is no doubt that re¬‚ective considerations go into empirical deter-
minations, but we should not reduce the systematic considerations that
led Kant to explore re¬‚ective judgment to something that, in the words
of Longuenesse, is “prior to the subsumption of empirical objects under
So far, I have argued that the re¬‚ective judgment appealed to in the Cri-
tique of the Power of Judgment is not the logical re¬‚ection that Longuenesse
sees at work in empirical concept formation and in the inductive proce-
dures of scienti¬c inquiry grounded by the Critique of Pure Reason. To more
fully explore the second question about the relation between aesthetic
judgment and the re¬‚ective appeal to analogies, I will provide a more
open-ended analysis of judgments of taste. Section 1 was an attempt to
¬nd the proper place of re¬‚ective judgment in Kant™s intellectual system.
Section 2 is an attempt to locate aesthetic judgments in their historical
and cultural contexts and to consider their normative status.

37 Longuenesse (1998), 64, 68, 122, 165, 179, 185, 298f.
38 Ibid., 166.
Re¬‚ection, Re¬‚ective Judgment, and Aesthetic Exemplarity 233

2. re¬‚ecting on prejudices of taste and
aesthetic schematization
Ordinary judgments of experience are determinant judgments about the
properties of objects that can be con¬rmed or discon¬rmed by observa-
tion. Aesthetic judgments also refer to objects, but disclose more about
our state of mind than about what is apprehended. Thus beauty, for Kant,
is not some experiential quality that adds to our knowledge of any object.
When we make a purely aesthetic judgment about x, we merely assert
that x has a form that puts our cognitive faculties in a state of harmony
that we feel to be pleasurable. It is a subjective judgment of taste, but it
is more than a report of the causal effect that the object has on us. If
it were merely that, it would also be a determinant judgment based on
self-observation or introspection. And because introspective claims are
subject to doubt, it would be a de¬cient mode of determinant judgment
about a particular state of mind.
A pure judgment of taste is not an empirical report about a subject™s
emotional state, but a normative re¬‚ective judgment that projects a felt
agreement with other subjects. It is not a descriptive Urteil, but a prescrip-
tive Beurteilung.39 What is the source of this normativity? For Kant, it is
transcendental. The aesthetic judgment transforms an empirical deter-
minant judgment about an object into a disinterested re¬‚ective judgment
that expresses a subjective assessment. A disinterested judgment is one
in which I suspend my normal theoretical and practical interests in the
existence of the object being apprehended. If I can so neutralize my
relations to an object, I should be able, according to Kant, to appre-
hend it purely. Thus he assumes that beauty involves the feeling of a
pure disinterested pleasure. This feeling may be a response to the object
apprehended but is not determined by it.40 Aesthetic appreciation is a
free evaluative response to the object™s formal purposiveness, which does
nothing more than allow the cognitive faculties to operate in harmony
with each other. Aesthetic pleasure is the feeling of this equilibrium. On
the assumption that all human beings share the same faculties, Kant con-
cludes that a disinterested judgment of taste is universally valid. Without
this Enlightenment tenet, all we can say is that the aesthetic judgment
makes a normative demand that it be a shared judgment. Whereas Kant
assumes that the sensus communis postulated by taste represents a universal

39 First Introduction to the Critique of the Power of Judgment, Ak 20:211.
40 See also the later discussion of the noncompelling in¬‚uence of precedents.
Rudolf A. Makkreel

community, we merely expect a general sharing that may in fact be less
inclusive. There is no need to equate the common, the general, and the
Kant ¬rst calls the aesthetic judgment gemeing¨ ltig, which means gen-
erally or commonly valid, in Section 8 of the Critique of the Power of Judg-
ment,41 but later substitutes the term allgemeing¨ ltig, which means univer-
sally valid. A similar shift is discernible in Section 22, where Kant writes:
“The necessity of the universal assent that we think in a judgment of taste
is a subjective necessity that we represent as objective by presupposing a
common sense (Gemeinsinn).”42 If we can presuppose that we all have a
common sense, then we can ground the judgment of taste in a constitu-
tive principle of experience and treat it as objective. However, this is not
really Kant™s standpoint. Section 34 makes it clear that an objective prin-
ciple of taste is impossible. Moreover, the sensus communis appealed to in
Section 40 is not a grounding principle, but what I would call an ˜orienta-
tional™ principle. This communal sense (gemeinschaftlichen Sinnes) is a pos-
sibility to be cultivated rather than something presupposed. Rather than
being an innately endowed common sense, the communal sense aims
at an ideal community. Thus the generality of the aesthetic judgment
is not rooted in some existing commonality, but projects a normative
universality to be arrived at by the human community.
Having introduced these distinctions, it is possible to generate the
following scale of aesthetic consensus: (1) a shared pleasure that is the
product of a common background, (2) a general pleasure that is re¬‚ective
or comparative in a purifying sense “ here the subject abstracts from its
private peculiarities “ and (3) Kant™s own ideal of a shared aesthetic
pleasure that is universal in claiming to be valid for all human beings.
The last kind of aesthetic consensus corresponds to Kant™s “power to
judge that in re¬‚ecting takes account (a priori) in thought of everyone
else™s way of representing.”43
The ¬rst kind of aesthetic consensus is a function of popular taste
and is therefore subject to fashion. The second kind involves re¬‚ective
judgment and is normative in a transcendental sense. I think that this
comes closest to what Kant actually justi¬es in his attempts to purify
human feeling and clarify taste. The third and ¬nal kind of aesthetic
consensus can be seen as aiming at a universal ideal. It takes what is

41 Critique of the Power of Judgment, Ak 5:214.
42 Ibid., Ak 5:239.
43 Ibid., Ak 5:293.
Re¬‚ection, Re¬‚ective Judgment, and Aesthetic Exemplarity 235

re¬‚ectively orientational and coordinated with a general sense of com-
munity (the second kind of consensus) and transforms it into an a priori
expectation of universal agreement that I can only regard as regulative
and hypothetical.44
What I will focus on as most relevant in Kant™s aesthetics is the relation
between (1) and (2). The ¬rst kind of consensus is due to a common
cultural background. This is what, in his lectures on logic, Kant speaks of
as being the product of a prejudice of taste. The second kind of aesthetic
consensus, by contrast, can be called the pure judgment of taste. What then
is the relation between a prejudice of taste and a judgment of taste?
In The Blomberg Logic, Kant distinguishes between logical and aesthet-
ical prejudices. The latter are also called prejudices of taste in the same
lectures. Kant sees aesthetical prejudices as imitating fashion and warns
that “Taste is quite ruined by imitation, a fertile source of all prejudices,
since one borrows everything, thinks nothing of a beauty that one might
be able to invent and come up with oneself.”45 One™s taste is a mere
prejudice if one reproduces what is common or customary in one™s back-
ground “ one merely imitates the examples of fashion. Criticisms like this
have led Gadamer to claim that Kant was too harsh on prejudices and
shared an Enlightenment prejudice against prejudices.46 Actually, Kant
warns that we should not immediately reject each and every prejudice!
Indeed, we should, in his words, “test them and investigate whether some-
thing good may yet be found in them.”47 He then goes on to say, “one
can actually ¬nd a kind of prejudice against prejudices, namely, when one
immediately rejects everything that has arisen through prejudices.”48
It is important to underscore that Kant allows our cultural tradition
to provide us with useful prejudices that can guide our taste. Tradition is
able to offer us possibilities from the past that are lacking in the present.

44 I ¬nd an interesting analogue to my earlier discussion of B´ atrice Longuenesse in the
fact that in this volume she argues that the merely re¬‚ective standpoint used by Kant to
explicate pure taste does “not suf¬ce to explain why we demand of everyone, as if it were
a duty, that they share our pleasure.” (See her essay in this volume.) If our overriding
concern becomes a deduction of taste, then the free coordinative assent involved in
pure taste can evolve into an actual duty to which we must subordinate ourselves. The
“demand” for agreement that Longuenesse attributes to Kant represents a rather forceful
translation of Zumutung, which Guyer and Matthews translate more appropriately as
“expectation” and which I would render even more modestly as “presumption.” We may
merely presume that others will agree with our judgments of taste.
45 Kant (1992), 136; Ak 24:173.
46 Gadamer (1992), 271“6.
47 Kant (1992), 133, Ak 24:169.
48 Ibid., 133, Ak 24:169.
Rudolf A. Makkreel

But not until we re¬‚ect on prejudices will they be properly appropriated.
The common prejudices with which we grow up will become useful only
if we use re¬‚ection to suspend them and transform them into something
provisional or preliminary. Both prejudices (Vorurteile) and preliminary
judgments (vorl¨ u¬ge Urteile) are modes of prejudgment (pr¨ judicium).49
a a
But whereas prejudices are unre¬‚ective and tend to rush to judgment,
preliminary judgments introduce a re¬‚ective moment that sets the stage
for further inquiry. They transform a common prejudice into a judgment
to be re¬‚ected on (judicium re¬‚ectens). The fact that preliminary judg-
ments appeal to re¬‚ection does not as such make them into re¬‚ective
judgments as they are de¬ned in the third Critique. A re¬‚ective judgment
is not merely subject to re¬‚ection, but is the outcome or product of actual
re¬‚ection and could be called a judicium re¬‚ectivum.50 Another difference
seems to be that a preliminary judgment re¬‚ects on the content of a prej-
udice, whereas a re¬‚ective judgment re¬‚ects on its form. A preliminary
judgment neutralizes the content of a prejudice to transform it into a
hypothetical claim that can then regulate further inquiry.
In his J¨ sche Logic, Kant characterizes the re¬‚ection involved here in
the way he had de¬ned transcendental re¬‚ection in the Critique of Pure
Reason, namely, seeing “to which cognitive power a cognition belongs.”51
Re¬‚ection allows us to pass from the persuasion that comes with hold-
ing a prejudice to the conviction that investigation can produce, but it
need not. The transition occurs only if re¬‚ection is used critically to
temporarily suspend judgment. It can also be used skeptically to per-
manently refrain from judgment.52 Accordingly, Kant says that “many
remain with the persuasion of prejudice, some come to re¬‚ection, few
to investigation.”53 Only investigation can produce proper determinant
judgments.54 It is clear once again that re¬‚ection is not a subsidiary com-
ponent of determinant judgment, for the deferral of judgment that is
involved in re¬‚ection “consists in the resolution not to let a mere prelim-
inary judgment become determining.”55 A preliminary judgment is thus
a product of re¬‚ection “in which I represent that while there are more

49 Ibid., Ak 24:161.
50 See Makkreel (1996), 70“1.
51 Kant (1992), 576, Ak 9:73.
52 Ibid., Ak 9:74.
53 Ibid., 576, Ak 9:73.
54 The quali¬cation “proper” was added because prejudices are already determinant “ to
be sure, pseudo-determinant “ judgments.
55 Ibid., Ak 9:74.
Re¬‚ection, Re¬‚ective Judgment, and Aesthetic Exemplarity 237

grounds for the truth of thing than against it, these grounds still do not
suf¬ce for a determinate or de¬nite judgment.”56
The aesthetic prejudice that considers Greek sculpture to be great can
be transformed by re¬‚ection either into a proper aesthetic judgment or
into a cognitive inquiry governed by the question, what has made Greek
sculpture so culturally signi¬cant to so many generations? The ¬rst route
is that of re¬‚ective judgment, the second that of a preliminary judgment,
which can generate hypothetical or regulative propositions that can be
tested by empirical investigation and thereby become a determinant judg-
ment. This empirical investigation into taste is not the route that Kant is
proposing. For him, a pure judgment of taste declaring a Greek statue
to be beautiful would use re¬‚ection to produce a normative re¬‚ective
judgment. It goes to the heart of the process of aesthetic apprehension
by focusing on its formal purposiveness in creating a harmony of the
faculties that enlivens the mind. It suspends the prejudice of taste, not
regulatively or hypothetically, but transcendentally. What this means is
that the example that is blindly imitated by aesthetic prejudice is trans-
formed into something properly exemplary.
The pure judgment of taste may re¬‚ectively orient itself to the judg-
ments of others, but it should not imitate them. Similarly, Kant suggests
that an artist can orient himself to the works of prior talents and be
“thereby awakened to the feeling of his own originality, to exercise his
freedom from coercion in his art in such a way that the latter thereby
acquires a new rule by which the talent shows itself as exemplary.”57
When we look at others as exemplary, we are inspired to draw on our
own resources as well. The examples passively stored in prejudices of
taste can be transformed into exemplary judgments of taste that we can
actively share. What we use as an example serves solely as an external
constraint. But what we take as exemplary functions as an external guide
that awakens an internal source as well. We ¬nd this same coordination of
external and internal references in Kant™s account of spatial orientation.
Thus I discern my place in the world by reference both to the external
position of the sun and to my internal capacity to distinguish left from
We might pause here to note that the capacity of re¬‚ective judgment
to make some particulars exemplary is the counterpart of its capacity to

56 Ibid., 577, Ak 9:74.
57 Critique of the Power of Judgment, Ak 5:318.
58 See Kant, “What Is Orientation in Thinking?”, Ak 8:134.
Rudolf A. Makkreel

specify a universal, discussed in Section 10. In both cases, particulars and
universals are brought in proximity by being coordinated.
Whereas examples provide determinate images with the power to bind
us, the exemplary can be said to be like an indeterminate schema that
leaves the imagination some ¬‚exibility in how to proceed. In Section 35
of the Critique of the Power of Judgment, Kant says that the aesthetic imagina-
tion schematizes without concepts, but he is not very helpful in spelling
out what this means. He contrasts this with schematization in the Critique
of Pure Reason by explaining that here we are not subsuming particular
intuitions under concepts, but rather coordinating the imagination and
the understanding in general as faculties that can reciprocally enliven
each other. But there is more to be said about schematization without
a concept. In the ¬rst Critique, schematization is the process whereby
concepts that are purely logical are explicated temporally to make them
experientially meaningful: The imagination serves the understanding by
making possible determinant judgments that apply universal concepts to
particular objects. In the third Critique, aesthetic schematization without
concepts must be conceived in terms of re¬‚ective judgment “ thus here
the movement is from particular to universal. An aesthetic or re¬‚ective
schema is a particular that is felt to be valuable. There being no directing
concept, aesthetic schematization is merely orientational. It allows a par-
ticular fact embedded in the “abode” of our prejudices to be freed from its
contingency by relocating it in a broader “territory” “ giving us more space
to gain our bearings. Aesthetic schematization may be seen as a kind of
epoch´ that suspends prejudices of taste to prepare us to judge things
from the general perspective of human sensibility. It is “merely re¬‚ec-
tive” in the positive sense of indicating something humanly valuable or
If schematization as such is the explication of meaning, then we can
distinguish two modes of such explication: the determinant explication
of the meaning of universal concepts as rules of application and the
re¬‚ective explication of the meaning of intuitive particulars that locates
something common or general in them. To re¬‚ectively schematize a par-
ticular would be to present it as normative “ as an exemplary model or
general type.
We need re¬‚ective schemata in judging beauty because we cannot rely
on self-evident determinant schemata or rules. Although judgments of

59 See also Gasch´ (2003), who argues that the “merely re¬‚ective” quality of aesthetic
judgment points to its ability to “shed light on an affective dimension” that is a “necessary
subjective component of all cognitive mental life” (26).
Re¬‚ection, Re¬‚ective Judgment, and Aesthetic Exemplarity 239

taste must be self-validated, it would be foolish not to consider the judg-
ments of others. Kant writes: “If each subject always had to start from
nothing but the crude predisposition given him by nature, [many] of
his attempts would fail, if other people before him had not failed in
theirs.”60 We can be autonomous even while we take into account what
others before us have done, if only to learn from their mistakes. Accord-
ingly, Kant de¬nes cultural progress in terms of the human species as a
whole. Whereas good taste must overcome the fashions of prejudice based
on unthinking imitation, it will always need to rely on exemplary models
or precedents. By relating taste to the exemplary and the in¬‚uence of
precedents, Kant places aesthetic norms in a public framework. It is not
enough to say that the individual subject adopts “a normative attitude
towards her mental activity” by taking it “to be appropriate” and then
expecting “that everyone ought to judge it in the same way.”61 Having
located the normative in the public domain from the start, Kant requires
us to regard the constraints of the normative as coming from both within
and without. He attempts to create this coordinative balance by consider-
ing a kind of historical in¬‚uence that does not deprive the individual of
his or her autonomy. To conceive this noncompelling kind of in¬‚uence,
he introduces a distinction between emulation (Nachfolge) and imitation
(Nachahmung). This is how Kant puts it: “Emulation of a precedent, rather
than imitation, is the right term for any in¬‚uence that products of an
exemplary author may have on others; and this means no more than
drawing on the same sources from which the predecessor himself drew,
and learning from him only how to go about doing so.”62 To consider a
model as a precedent is not to appeal to it as a determining ground, but
merely to orient oneself by it as potentially valuable or worth committing
to.63 The task of emulation is not merely to reproduce a standard, but to

60 Critique of the Power of Judgment, Ak 5:283.
61 This is how Hannah Ginsborg characterizes the normativity of taste in her essay in this
62 Critique of the Power of Judgment, Ak 5: 283.
63 The question is, what kind of constraint does a precedent put on the judging subject?
Although Kant considers the position of the other in re¬‚ective assessment, I do not think
he is ready to acknowledge Robert Brandom™s double scorekeeping, where “the commit-
ments a scorekeeper attributes to someone outrun those that the individual acknowledges”
(Brandom 1994, 646). Because Brandom sees us as embedded in social and linguistic
practices, many of the normative commitments that an individual acknowledges involve
inferential commitments that only another scorekeeper can recognize. There is thus
always a gap between the normative theoretical commitments that individuals acknowl-
edge and those that they have implicitly undertaken.
Brandom begins with something like Kant™s own distinction between what I hold to
be true (F¨ rwahrhalten) and what is true. But whereas for Kant this gap can be ¬lled by
Rudolf A. Makkreel

re¬ne it. This demands something like what is called ˜re¬‚ective endorse-
ment™ in the moral domain.64 The emulation of re¬‚ective precedents
allows us to relate the ˜territory™ of aesthetic judgment to a ˜domain™ of
mutual accountability, but in a way such that no one is compelled.
A re¬‚ective schema or precedent is not a determinate example, but
an indeterminate exemplar. In fact, Kant™s judgments of taste declaring
something to be beautiful are indeterminate too “ precisely to leave room
for human agreement. However, not all aesthetic judgments need be that
indeterminate. Especially when we judge a work of art, we want to be able
to specify some of the features that make it beautiful. Thus, in drawing
attention to aspects of its form, it is appropriate to describe how the
¬gures in a painting stand in relation to each other and how the distri-
bution of light leads the eye in certain directions. These are descriptive
determinant claims that can be used by a critic to try to convince others
that a painting deserves acclaim. Here we are making cognitive claims
about an aesthetic object, which means that we are moving beyond pure
aesthetic judgments. Kant himself acknowledges that in responding to
works of art, it is appropriate to recognize that the artist has a creative
purpose that he is communicating. Whereas beauty as such is purposive
without having a purpose, that is, merely playful and suggestive, artistic
beauty is purposive in a more determinate way. Artists are generally seri-
ous about creating an effect on their audience, although they must not be
too obvious about this. To the extent that literary artists have a purpose in
creating, concepts can be found to articulate their intent. But a great poet
will always express more than what is directly communicated by available
Such a surplus of meaning is illustrated by a line of poetry that Kant
quotes in Section 49. The line “The sun streamed forth as serenity streams
from virtue”65 provides more than a conceptual description of a beautiful
sunrise. According to Kant, the poet is presenting an aesthetic idea that
evokes more thought than can be “comprehended within a determinate

expanding my own perspective into that of a shared community or tradition (expanding
the I into a we), Brandom sees here a wider gap between an I and thou, which means
that “there is never any ¬nal answer as to what is correct” (ibid., 647). For Brandom, it
may be what I am unconsciously committed to that is ultimate, whereas for Kant, there
can be no ultimate commitment that has not been consciously endorsed. This becomes
especially obvious when we move from the domains of the theoretical and practical to
the territory of the aesthetic. For how am I to assert a judgment of taste without its
adequately re¬‚ecting me?
64 See Korsgaard (1996), 49“89.
65 Critique of the Power of Judgment, Ak 5:316 “ Kant here quotes a line from J. Withof.
Re¬‚ection, Re¬‚ective Judgment, and Aesthetic Exemplarity 241

concept.”66 The joy produced by the rising sun is suffused with the seren-
ity produced by virtue. A relation between the natural world and the
realm of freedom is suggested that enlivens our cognitive powers. It is
clear then that in judging art we can use concepts to make at least some
determinant cognitive judgments. Yet works of art are suggestive in ways
that may leave many of our judgments anticipatory and indeterminate
again. Aesthetic ideas are introduced to orient us to rational ideas, and
it is through aesthetic ideas that beauty can become a symbol for the
rational idea of moral virtue.
The symbol adds another mode of exhibiting meaning for Kant
and further extends his theory of schematization. In Section 59 of the
Critique of the Power of Judgment, he describes schemata and symbols as two
modes of presenting or explicating meaning.67 We have already spoken
of schemata as explicating the meaning of concepts of the understanding.
Then we suggested that schematizing without a concept when making a
judgment of taste allows us to consider particular examples of good taste
as orienting models. Symbols, in turn, explicate the meaning not of con-
cepts of the understanding, but of ideas of reason. Whereas the categories
of the understanding are directed at making sense of our experiences,
ideas of reason tend to transcend experience. How then can a symbol
exhibit the meaning of ideas? A symbol does not create a direct intuitive
counterpart for the abstract idea. It merely allows us to intuit something
analogous. There is no sensuous content that can exemplify the rational
idea of moral goodness. The sun by itself cannot be a symbol for the good.
Yet we may be said to stand in similar relations to the sun and to the good
to the extent that both uplift us. The sun becomes a symbol of the good
if in re¬‚ecting on our relation to both we ¬nd some formal analogies
there. Once we have found some analogue, we can use our re¬‚ection on
experiential relations to specify relations among more abstract ideas.
Symbolization can create re¬‚ective analogies between our experience
and ideas that surpass (¨ bertreffen) experience.68 Aesthetic judgments
that are symbolical could be said to approximate the re¬‚ective principle
of speci¬cation discussed earlier by presuming greater similarity on the
basis of partial similarity, but they will never claim overall or total simi-
larity. Analogies based on aesthetic ideas are used to ¬ll in or complete

66 Ibid., Ak 5:315.
67 For a more detailed analysis of schemata and symbols as functions of the imagination
see Makkreel (1990), especially Chapters 2 and 6.
68 Critique of the Power of Judgment, §49, Ak 5:314.
Rudolf A. Makkreel

experience rather than to totalize it in the manner of rational ideas.
Symbolical judgments are anticipatory, like preliminary judgments and
like re¬‚ective aesthetic judgments of taste that demand the agreement
of others. But what symbols anticipate cannot be con¬rmed objectively
by experience or subjectively by the agreement of others. Instead, sym-
bolical judgments can merely extrapolate from experience. Preliminary
aesthetic judgments anticipate content; re¬‚ective aesthetic judgments
anticipate formal agreement; symbolical aesthetic judgments anticipate
re¬‚ective analogies that allow us to coordinate the form and content of
experience. When aesthetic judgment goes over into symbolization, we
have a clear case of Uberlegung or re¬‚ection becoming Auslegung or inter-
pretation. Here re¬‚ective judgment coordinates representations to lay
out (aus-legen) a “¬eld” of possibilities and expands them by a process of
imaginative Ausbilding or completion. In this context, we can also make
sense of the fact that the German word used by Kant for symbolization is
Gegenbildung “ the process of creating coordinated counter images, that
is, images that reinforce or enliven each other.
Instead of regarding re¬‚ective judgment as proto-experiential or
subservient to the conceptual needs of the understanding, it should
be considered as meta-experiential.69 It attempts through a process of

69 Paul Guyer™s essay in this volume argues for a meta-cognitive interpretation of Kant™s
aesthetic harmony that has some parallels with my meta-experiential approach. Never-
theless, Guyer classi¬es the harmony disclosed by my conception of the aesthetic imagi-
nation as being merely precognitive. Actually, I wrote in Imagination and Interpretation in
Kant that “the aesthetic imagination is not limited to the preliminary precognitive func-
tions often assigned it, but plays a role in re¬‚ective judgment™s systematic concern with
knowledge in general” (1990, 66). Guyer rejects this link between the aesthetic imagina-
tion and systematic re¬‚ection by claiming that the aesthetic use of judgment is “entirely
distinct” from the systematic use of re¬‚ecting judgment in the Introductions to the
Critique of the Power of Judgment (Guyer, footnote 16). To this I respond by pointing to
Section VII of the Introduction, where Kant relates the aesthetic use of judgment to the
more general re¬‚ecting uses of judgment. Thus, in the judgment of taste, “the object
must be regarded as purposive for the re¬‚ecting power of judgment” (Critique of the
Power of Judgment, Ak 5:190; 77). Then in Section VIII we ¬nd an explicit link between
aesthetic judgment and systematic re¬‚ection about nature: “In a critique of the power of
judgment the part that contains the aesthetic power of judgment is essential, since this
alone contains a principle that the power of judgment lays at the basis of its re¬‚ection on
nature entirely a priori, namely, that of a formal purposiveness of nature in accordance
with its particular (empirical) laws for our faculty of cognition” (Ak 5:193; 79). This
purposiveness is what allows our understanding to systematize the laws of nature.
The fact that Kant places aesthetic pleasure “merely in the form of the object for
re¬‚ection in general” (Ak 5:190; 77) indicates that the accordance that exists between
the aesthetic imagination and “the way categories are generally schematized” (Makkreel
1990, 56) abstracts from reference to individual categories. It was never my view, pace
Re¬‚ection, Re¬‚ective Judgment, and Aesthetic Exemplarity 243

coordination to complete our experience and thus partly ¬ll in the total
system of experience that ideas of reason can only project abstractly.
Being meta-experiential, re¬‚ective judgments about art are often parasit-
ical on background determinant judgments. The most explicit example
of a claim by Kant in the Critique of the Power of Judgment that involves the
intersection of determinant and re¬‚ective judgment is not about art, but
about the cultural context in which art functions. It is his assertion that
man is the ultimate purpose (letzer Zweck) of nature because, as a cul-
tural being, he can set himself ¬nal purposes (Endzwecke). The claim that
human beings constitute an ultimate purpose is a teleological re¬‚ective
judgment, but it is based on a determinant judgment of practical reason
about their ability to establish themselves as ¬nal purposes.70 I conclude
from all this that whereas Kant de¬ned a pure aesthetic judgment to be
nonconceptual, most aesthetic judgments are at least in part conceptual.
They may presuppose already familiar empirical concepts, as when we
refer back to prejudices of taste, or more general concepts, as when we
orient our judgment to exemplary models. Moreover, they may project
aesthetic ideas that disclose af¬nities with rational ideas and can in turn
suggest new concepts. Since we grow up with logical as well as aesthetic
prejudices, it is unlikely that we ever confront the world without any
concepts. They may be inadequate concepts, or mere representational
concepts as found through the logical re¬‚ection discussed in Section 1 of
this essay. This means that the so-called nonconceptual judgment of taste,
“This rose is beautiful,” and the more generic judgment, “This ¬‚ower is
beautiful,” use vague representational concepts rather than determining
concepts with the de¬ning marks of things. And even when determin-
ing concepts do intrude into a judgment of taste, as in the more speci¬c
claim, “Hybrid Tea Roses are especially beautiful,”71 its cognitive content
can be abstracted from or re¬‚ectively neutralized. Even botanists are able
to suspend what they know about ¬‚owers in order to merely contemplate
their form.

Guyer, that we think of causation as such or of a particular kind of causation when appre-
ciating beauty. Yet aesthetic harmony can exhibit a felt analogue of causality, namely, the
power to prolong itself. Aesthetic pleasure “has a causality in it, namely that of maintain-
ing the state of the representation” (Ak 5:222, 107).
70 See Makkreel (1990), 137“8.
71 This is how a rose catalog elaborates on the beauty of Hybrid Tea Roses: “These are the
¬‚owers we envision when the word ˜Rose™ is mentioned, and the image conjured up “
that of large, elegant blooms with high centers and numerous, substantial petals “ is a
true one. Hybrid Teas usually produce one bloom per long stem. . . . they will astound
you with their beauty.” The Complete Rose Catalog (Hodges, SC: Wayside Gardens, 2003), 3.
Rudolf A. Makkreel

The fact that re¬‚ective and determinant judgments can at times inter-
sect does not mean that their functions merge. Indeed, I have argued
that it is re¬‚ection rather than re¬‚ective judgment that can lead up to,
and in that way merge with, determinant judgment. But even re¬‚ection
as such is not about making objective determinations. Thus, if its results
are to be used to contribute to our knowledge of the world, then the
impetus will have to come from without. Neither re¬‚ection nor re¬‚ec-
tive judgment inherently subordinates itself to determinant judgment.
On the contrary, re¬‚ective judgment, being orientational and interpre-
tive, provides a more general framework for the more delimited claims of
determinant judgment. It is because re¬‚ective judgment frames our expe-
rience that re¬‚ection by itself can precede as well as follow the making of
determinant judgments. We have seen Kant use re¬‚ection both to lead
up to concept formation and to suspend prejudices. Indeed, if re¬‚ection
can raise questions about the premature determinations of prejudices, it
should also be able to point to the limits of our capacity to attain mature

Understanding Aestheticized

Kirk Pillow

Many interpretations of Kant™s ¬rst and third Critiques tend to treat these
texts as at best tangentially related. Kant™s distinction between determi-
native and re¬‚ective judgment “ the former largely the purview of the
Critique of Pure Reason™s Transcendental Analytic, the latter the subject
of the Critique of the Power of Judgment “ invites a view of these mental
acts as wholly distinct. Determination, after all, involves the subsumption
of objects under concepts, while the re¬‚ective judgments of the third
Critique appear to involve no determining concepts. Pure aesthetic judg-
ments especially seem irrelevant to our conceptualizing efforts, for their
basis lies in a certain kind of pleasure rather than in cognition. A healthy
corrective to this tendency is provided by Beatrice Longuenesse in her
Kant and the Capacity to Judge (1998). Longuenesse makes a compelling
case that re¬‚ection plays a role in empirical concept formation in all
our acts of judging particulars determinatively. She argues that the “con-
cepts of re¬‚ection” of the ¬rst Critique Amphiboly guide the generation of
empirical concepts through the application of the logical forms of judg-
ment to particulars. Re¬‚ecting on the particular to produce a universal
under which to subsume it is then one aspect of determinative cognition.
Even the categorial structure of experience, Longuenesse argues, is the
outcome of acts of judgment with a fundamentally re¬‚ective origin.1
Longuenesse weds the ¬rst Critique to the third Critique theme of re¬‚ec-
tion, however, by sharply contrasting a re¬‚ective aspect of cognition with
the variety of “merely re¬‚ective” judgments that comprise the actual sub-
ject matter of the Critique of the Power of Judgment. On her reading of Kant,

1 See Longuenesse (1998), Part Two, especially Chapter 6.

Kirk Pillow

“every judgment of empirical objects as such is re¬‚ective,” but also “determina-
tive as well”; what distinguishes aesthetic and teleological judgments is
not that they are re¬‚ective but that “they are merely re¬‚ective judgments,
judgments in which re¬‚ection can never arrive at conceptual determi-
nation.”2 This means that Longuenesse secures a role for re¬‚ection in
Kant™s account of cognition only by setting aside aesthetic judgments as
irrelevant to cognition. Because for Kant aesthetic re¬‚ection does not
subsume particulars under determining concepts, the aesthetic again
seemingly contributes nothing to our cognitive efforts. Mere aesthetic
re¬‚ection has merely to do with pleasure, and so with subjective con-
ditions rather than with making cognitive claims about matters of the
object(ive). The result of Longuenesse™s interpretation is the alignment
of only a certain element of re¬‚ection with cognition such that the divide
remains between the cognitive work of determinative understanding and
the “merely” re¬‚ective play of aesthetic experience.
I wish to argue that this seeming irrelevance of the aesthetic to cog-
nition is the result of an impoverished conception of cognition. While
Kant places conceptualization and judgment at the heart of cognition
and locates this activity in the individual subject, the range of twentieth-
century epistemologies has provided us a richer picture of the opera-
tions of human cognition. Philosophers as varied as Kuhn and Davidson,
Heidegger and Sellars have led us to see the context dependence of epis-
temic practices, the interpretive dimension of knowing, and the social or
intersubjective nature of inquiry. Yet in most cases, this broadened con-
ception of understanding has been advanced while still ignoring aesthetic
experience and the insights of aesthetic theory, and largely maintaining
the gap between the cognitive and the aesthetic upon which Kant™s own
account of determinative judgment relies. A notable exception to this
tendency is the work of Nelson Goodman and Catherine Elgin, who
have made aesthetic concerns central to their account of cognition or
understanding. From Goodman™s Languages of Art (1968) and Ways of
Worldmaking (1978), to Elgin™s Considered Judgment (1996), they have pre-
sented an ecumenical conception of knowing to which traditionally cog-
nitive and aesthetic capacities and responses contribute equally. My aim
here is to follow and draw upon their example by expanding the Kantian
conception of ˜understanding™, to include in it what he excludes from
it through his sharp distinction between determinative judgment and
aesthetic re¬‚ective judgment.

2 Ibid., 164.
Understanding Aestheticized 247

I will contrast Kant™s conception of cognition as the recognition of
objects under concepts with Goodman and Elgin™s more nuanced con-
ception of understanding. Understanding in their sense accommodates
a variety of interpretive practices traditionally relegated to the sphere of
the mere aesthetic. This broadened (and more compelling) conception
of understanding will allow us to see that central features of Kant™s aes-
thetic theory, especially his theory of the aesthetic idea, contribute to the
satisfactory characterization of understanding thus broadened. Further-
more, reading Kant in this way will allow us to make use of his analysis of
the judgment of taste in an account of how we assess the cognitive claims
of an interpretive understanding. Conceiving understanding broadly as
inclusive of both cognitive and aesthetic dimensions will provide a critical
perspective on the divide between aesthetic and cognitive judgments in
Kant™s thought, but will also remind us that Kant opens the way to rec-
ognizing human understanding as an interpretive endeavor. As we will
see, however, a requirement of this reading of Kant will be the rejection
of one of the more implausible features of his aesthetic theory: the strict
requirement of aesthetic disinterestedness.
Concepts for Kant function as rules for determining which objects fall
under them, that is, which objects possess the properties marked by the
predicates of the concept. Understanding is the source of conceptual
rules (empirical and categorial), and judgment is the capacity for sub-
suming particulars under these rules.3 Determinative judgment, which,
following Longuenesse, I read as inclusive of re¬‚ective empirical concept
formation, is the “art” both of bringing particulars to appropriate, estab-
lished concepts and of inventing concepts that generalize the features
of the previously ill-conceived particular. Kant regards these capacities as
the core of cognition and practically identi¬es understanding with such
judgment: “We can . . . trace all actions of the understanding back to judg-
ments, so that the understanding in general can be represented as a fac-
ulty for judging” (A69/B94). Indeed, the view defended by Longuenesse
and others, that re¬‚ective judgment plays a role in empirical concept for-
mation, strengthens the identi¬cation of understanding with judgment,
for it makes judgment essential to the acts of understanding in which
conceptual rules are produced for use in cognizing particulars in deter-
minative judgments. All cognition of objects for Kant comes down to

3 Critique of Pure Reason, A132/B171. Subsequent references will provide A/B pagination
in the body of the text. See also the First Introduction to the Critique of the Power of
Judgment, in Kant (2000), 8 (Ak 20:201).
Kirk Pillow

acts of judgment in which we apply established concepts or re¬‚ectively
produce concepts that provide new ways of determining the objects.
Yet these capacities for conceptual subsumption are but one element
of the more sophisticated conception of understanding articulated by
Goodman and Elgin. They conceive understanding as the perpetual
effort to make coherent sense of things in ways that advance pursuit of
our goals and help us to imagine new ones. Understanding so construed,
as Elgin writes, “is more comprehensive than knowledge ever hoped to
be. We understand rules and reasons, actions and passions, objectives
and obstacles, techniques and tools, forms, functions, and ¬ctions, as
well as facts.”4 This broad understanding employs myriad epistemic tech-
niques, such as those described by Goodman as “ways of worldmaking,”
that include but go beyond the acts of conceptual subsumption to which
Kant reduced cognition. We not only categorize objects through our con-
cepts; we also advance understanding by weighing the relevance of var-
ious judgments for different explanatory purposes, or by ordering ¬nd-
ings variously for different practical or expressive aims.5 For Goodman
and Elgin, judgments of Kant™s determinative sort are not the only con-
stituents of cognition; they are but one of many devices through which
we seek understanding.
Understanding in their broader sense is both coherentist and inter-
pretive. As coherentist, understanding makes claims that depend for
their meaning and referential power on their position in a larger whole
of established claims. Understanding is guided by the regulative goal
of ordering hypotheses and ¬ndings into consistent systems of thought,
which means that one measure of the validity of new claims is how well
they cohere with what we already hold ¬rmly. Claims that cannot be
squared with our cognitive commitments are unlikely to take root unless
further inquiry reveals advantages to adopting the claim suf¬cient to
warrant the recon¬guration or rejection of other beliefs. Understanding
is thus constrained in its claims by a history of cognitive commitments,
many of which we would be unlikely to revise or reject. This means that
the mere coherence or internal consistency of a set of claims is insuf¬-
cient to assure their contribution to understanding, because the claims
must ¬t into the broader network of our established convictions if they

4 Elgin (1996), 123.
5 Goodman (1978), Chapter 1. In what follows, I will focus primarily on Catherine Elgin™s
more recent re¬nements of the position that Goodman and she have somewhat differently
Understanding Aestheticized 249

are to compel. Through the ordering of claims into coherent wholes,
and through the assessment of them in light of given cognitive commit-
ments, understanding moves in and out of what Elgin calls (developing
terminology of Rawls™s) “re¬‚ective equilibrium.” In a system of cognitive
commitments in re¬‚ective equilibrium, the various components are “rea-
sonable in light of one another,” but the whole is also “reasonable in light
of our initially tenable commitments,” those convictions that understand-
ing has already endorsed (CJ 107). Understanding is a perpetual process
of revising convictions to maximize tenability, where tenability is judged
not with reference to any absolute standard but with reference to a history
of advancements of understanding to which we are committed. These
commitments to the already understood are not absolute either, however,
because new claims can potentially unseat even deeply held positions. “A
cognitive system is not a static framework of ¬rmly established ¬ndings,”


. 7
( 10)