. 7
( 10)


under concept A) is part of the whole of my present experience, then X
(recognized under concept B) is part of the same whole.™™ What we
represent to ourselves as the simultaneity of things in space is then
nothing other than the sensible (temporal) form, that is, the mode of
ordering individuals in time, resulting from a synthesis guided by the
capacity to analyze according to the discursive form of a hypothetical
judgment whose reciprocal converse is also believed to be true. In
accordance with this discursive form, asserting the presence (existence,
Dasein) of one of the objects perceived is represented as a sufficient
condition for asserting the presence of the other, and conversely the
presence of the latter is reflected as a sufficient condition for asserting
the presence of the former. Which specific determinations condition one
another (i.e. specifically what conditions what), we do not know. We shall
acquire such determinate cognition only by means of the indefinite,
never completed process of corrections and specifications of our discur-
sive judgments in actual experience. Nevertheless, Kant™s point is that

the relation, in Kant™s argument, between the respective roles of the logical form of
disjunctive judgment and that of hypothetical judgment, both of which, I maintain, play
a role (the latter more directly than the former) in our ordering appearances in such a way
that we experience their objective simultaneity.

this process finds its initial impulse in the mere consciousness of the
simultaneous existence of things in space, because such consciousness
itself already depends on a synthesis of sensible manifolds guided by our
capacity to judge, namely, a synthesis oriented toward reflection accord-
ing to the form of hypothetical judgments.25
If this is correct, objects are thus individuated in space and time by
their reciprocal interaction, and concepts of objects thus individuated
are concepts of relational properties. But this means that the empirical-
cognitive use of the form of disjunctive judgment, by means of which
we think of objects in nature as falling under a unified scale of genera
and species, is mediated by that of the form of hypothetical judgment,
by means of which we individuate objects by determining their uni-
versal interaction in one space and one time. This is why I said earlier
that the category of community is the most complex of all. It cannot be
understood except under the presupposition of the other relational
categories, and thus under the presupposition of the empirical use of
the logical forms they depend upon. I submit that this is why the third
category of relation has two names: Wechselwirkung (reciprocal action,
where the emphasis is on the relation of causal interaction) and
Gemeinschaft (community, where the emphasis is on objects™ belonging

Note that Kant™s reasoning here, just like his argument in the Second Analogy, displays
a complex web of interdependence between subjective and objective temporality. On the
one hand, awareness of the irreversibility or reversibility (order-determinateness or order-
indifference) of the subjective succession of representations is all that perceiving (experien-
cing) the objective temporal order of appearances amounts to. So, the perception of
objective temporal order depends on a specific feature of the subjective succession of
representations. But on the other hand, what generates our consciousness of such
a feature of the subjective succession just is our act of relating our representations to an
intentional object (an object they are the representation of). This is because relating our
representations to objects is attempting to reflect objects under concepts according to the
logical forms of categorical, hypothetical, and disjunctive judgment, and this in turn is what
generates “ depending on what is given to our senses “ our awareness of the irreversibility of
the subjective succession in case the pattern that emerges is that of a permanent object whose
states change, or the reversibility of the subjective succession in case the pattern that emerges
is that of several coexisting permanent objects whose states are interrelated. So, striving to
relate representations to objects is what generates the awareness of the reversibility or
irreversibility of the subjective succession, and this in turn just is what our awareness of
the objective temporal order (succession or simultaneity of states of things) amounts to. Thus
Kant™s Analogies of Experience should be understood as being essentially an explanation of
how we relate representations to objects in general: an explanation of intentionality (the
directedness of representations, their property of being representations of something), and
as a result, a theory of what makes it possible to apply concepts such as those of causal
connection and causal interaction to the objects of an empirical science of nature.

to one space, thus to one world-whole, and under one logical space
of concepts).26

Concluding remarks
Kant™s logic typically comes under heavy attack, on two main grounds.
First, it is suspiciously psychologistic. Second, it is caught within the
narrow bounds of an Aristotelian model of predication and syllogistic
inference, a model relegated to irrelevance by Fregean/Russellian exten-
sional logic. However, in the light of the use Kant makes of his ˜˜logical
functions of judgment™™ for solving the problems he addresses in the
critical system, I would like to suggest that the charges of psychologism
and archaism perhaps cancel each other. Because what Kant calls ˜˜pure
general™™ or ˜˜formal™™ logic is exclusively concerned with the ˜˜universal
rules of the understanding,™™ and understanding is the faculty of concepts
(defined as ˜˜universal and reflected representations™™), Kant™s logical
forms of judgment are nothing but forms of concept subordination, and
the forms of inference he is concerned with are merely the various ways
in which concept subordination (inclusion or exclusion of the extensions
of concepts, under an internal or external condition) allows for truth-
preserving inference. And because his ˜˜pure general logic™™ is so narrowly
defined, it can make a claim to being a description of the forms according
to which minds such as ours are capable of universalization of their
representations “ capable of combining their representations in such a
way that they are susceptible to being reflected under concepts and thus
related to objects, defined both logically as instances of concepts, and
intentionally as what our representations are representations of (the
intentional correlate of our representations). None of this makes Kant™s
˜˜general pure logic™™ a part of psychology, for logic, as Kant puts it, is
concerned not with the way we think, but with the way we should think:
the normative rules of concept combination according to which our
judgments are testable as to their truth and falsity (cf. A54/B78).27

It is striking that in the appendix to the Transcendental Dialectic, when Kant talks of
the demand of reason to unify all concepts of natural science under one highest genus,
the genus he cites is that of the concept of force (A649/B677), namely precisely that concept
he takes to justify, in the Amphiboly of Concepts of Reflection, the anti-Leibnizian
point that (empirical) substances are individuated only by their relational properties (see
Cf. also Jasche Logic, Einl. i , AAix , p. 14. And in this volume, ch. 4, pp. 89“91.

I pointed out earlier that in his explanation of logical forms of judg-
ment “ especially the form of disjunctive judgment “ Kant gives pride
of place to an extensional consideration of concepts and concept
subordinations, that is, to the consideration of the classes or multiplicities
(Mengen) of objects thought under concepts. This is because his main
concern is to elucidate the ways in which forms of concept subordination
are also forms according to which individual objects are subsumed
under concepts, and thus extensions of concepts are constituted in the
first place. And this in turn is related to the role Kant assigns to forms of
intuition (space and time) as the forms according to which objects are
individuated, distinguished from one another and brought together,
˜˜synthesized™™ so that they become susceptible to being reflected under
concepts. Examining and evaluating Kant™s notion of a form of intuition
is beyond the limits of this chapter, as is examining Kant™s account of the
synthetic a priori character of mathematics and its role in empirical
science. Nevertheless, in light of my examination of Kant™s logical form
of disjunctive judgment and its relation to the category of community,
I suggest that we should be attentive to the ways in which the notion of an
a priori form of intuition is meant to account for an original capacity to
represent (anticipate, generate) homogeneous multiplicities (multipli-
cities of objects thought under the same concept) just as Kant™s table of
logical functions is meant to account for an original capacity to form
universal concepts. Kant did not anticipate logical or scientific revolu-
tions to come, and certainly we have reason to wish he had been more
circumspect in his remarks on Aristotelian logic, Euclidean geometry, or
Newtonian science. But what he did provide was a striking model of how
elementary functions of minds such as ours “ functions of concept
formation and functions of object-individuation “ might account for
the unity of our unsophisticated, everyday perceptual world, and our
sophisticated, scientific worldview.
He argued, moreover, that these same elementary functions, when
related not to sensations, but to impulses and desires, are capacities to
develop a moral standpoint (Critique of Practical Reason); and that both
moral and theoretical standpoint are ultimately rooted in the peculiar
nature of the living, pleasure-seeking, purposeful beings we are (Critique
of Judgment).28 All three Critiques thus give us a view of human beings as
having a peculiar capacity to develop what we might call a standpoint on

See chs. 9 and 10 in this volume.

the whole: a standpoint whose elementary discursive form is the form of
disjunctive judgment and the grounding concept, that of community.
Just a few more words, before I close, about this concept of ˜˜community™™
and its further destiny in the critical system. In the first version of the Third
Analogy, after developing his argument to the effect that substances are
perceived as simultaneously existing only if they are in relations of universal
interaction, Kant notes that our own body is the mediator for our percep-
tion of the simultaneous existence of other bodies, or physical substances:
From our experience it is easy to notice that only continuous influence in
all places in space can lead our sense from one object to another, that the
light that plays between our eyes and the heavenly bodies effects a
mediate community between us and the latter and thereby proves the
simultaneity of the latter, and that we cannot empirically alter any place
(perceive this alteration) without matter everywhere making the percep-
tion of our position possible; and only by means of its reciprocal influ-
ence can it establish their simultaneity and thereby the coexistence
of even the most distant objects (though only mediately). (A213/B260)29

Because of this mediating role of our sensing body in our perception of the
community of material substances, the community of material substances is
also a community of our respective standpoints (the respective standpoints
of empirically given human beings located in space) on material substances,
and on the world as a whole. Now, in the third Critique “ the Critique of the
Power of Judgment “ Kant makes it one of the grounding maxims of
Enlightenment, that we should strive to think ˜˜from the standpoint of
everybody else.™™ And he grounds our capacity so to think in what he calls
a gemeinschaftlicher Sinn, a common sense or sense of community, namely
the capacity to develop a common standpoint on the whole (whether a
common epistemic standpoint on the whole of objectively existing things,
or a common normative/moral standpoint on the whole of interacting
human beings). This gemeinschaftlicher Sinn, or common sense, consists in
our capacity to use imagination and understanding in such a way that each
enhances the other in striving for a universal standpoint, albeit one pre-
mised on each of the particular standpoints we initially hold.30

On Kant™s view of the relation between self-consciousness, our consciousness of our own
body and our consciousness of a world of material objects in general, see my ˜˜Self-
consciousness and consciousness of one™s own body: variations on a Kantian theme.™™
Forthcoming in Philosophical Topics.
See Critique of the Power of Judgment, AAv, p. 293. Guyer and Mathews translate gemeinschaf-
tlicher Sinn as ˜˜communal sense.™™

We are more used to reading the critical system under the dominance
of the concept of cause: from Kant™s response to Hume™s skeptical doubt
in the first Critique, to his elaboration of the concept of free agency in the
second Critique. And certainly, there is a lot to say for this line of reading.
But I would like to suggest that from the community of substances in the
first Critique, to the community of standpoints on substances, also in the
first Critique, to the community of rational agents in the second Critique,
to the gemeinschaftlicher Sinn of the third Critique, there is another line
of reading, one that does not contradict the previous one but integrates
it into a more complete view of Kant™s philosophical project: relating, as
he says, all cognition to ˜˜the essential purposes of human reason™™
Finally, I submit that it is also from the standpoint of this concept and
its development throughout the critical system that we can best evaluate
Kant™s relation to his German Idealist successors. It is quite striking, for
instance, that in Hegel™s Phenomenology of Spirit the progress from
˜˜Sense-Certainty™™ to ˜˜Perception,™™ to ˜˜Force and Understanding™™ (the
first three chapters of the Phenomenology) is one where we gradually
become aware that only under a representation of universal interaction
is the identification of any individual object of sense-perception possible
for a consciousness such as ours. Hegel thus appears to espouse just the
kind of reasoning I have argued is Kant™s own in the Third Analogy. And
like Kant, he goes on to examine what relation between the conscious
subjects themselves is involved in the cognitive process just described
(fourth chapter of the Phenomenology, ˜˜Self-Consciousness,™™ and the
dialectic of desire and recognition). This being said, there are of course
major differences between the ways each of them proceeds from there
(not to mention the differences in the ways they arrive there). Where
Kant thinks that the same discursive (intellectual) functions by means of
which we represent the community of spatiotemporal substances can
also serve to think a purely noumenal (a-temporal and non-spatial)
region of being to which we belong as moral agents, Hegel, reasonably
enough, denounces the hypostatization of an ˜˜inverted world™™ (end of
the chapter on ˜˜Force and Understanding™™).31 On the other hand,
where Kant insists that our epistemic standpoint on the whole is irre-
trievably limited by the given spatiotemporal conditions of our human

Hegel™s Phenomenology of Spirit, trans. A. V. Miller (Oxford: Oxford University Press, 1977),
pp. 79“105; G. W. F. Hegel, Phanomenologie des Geistes, ed. Wolfgang Bonsiepen und
Reinhard Heede, in Gesammelte Werke, vol. ix (1980), pp. 82“102.

sensory knowledge, Hegel, unreasonably enough, strives to achieve a
standpoint that would amount to ˜˜the presentation of God, as he is in his
eternal essence before the creation of nature and of a finite spirit.™™32 It is
perhaps possible to interpret Hegel™s grandiose statement as gesturing
toward nothing more than some universal underlying logic of all concept
formation and correction.33 Just as it is perhaps possible to interpret
Kant™s talk of a ˜˜noumenal realm™™ as gesturing toward nothing more
than our moral use of reason in achieving a fully autonomous determin-
ation of action. Perhaps we can come to this kind of reasonable recon-
struction in both cases. Even so, I would suggest that the resistance
Hegel opposes to Kant™s ˜˜noumenal realm,™™ on the one hand, and the
resistance Kant opposes, preemptively as it were, to any ambition remotely
resembling Hegel™s logic of ˜˜absolute knowledge,™™ are, from each of
them respectively, a lasting legacy.34

Hegel™s Science of Logic, trans A. V. Miller (New York: Humanities Press, 1976), p. 50;
G. W. F. Hegel, Wissenschaft der Logik. vol. i: Die objective Logik, ed. Friedrich Hogemann
and Walter Jaeschke, in Gesammelte Werke, xi (1978), p. 21.
This kind of reading is defended by Robert B. Brandom, ˜˜Some pragmatic themes in
Hegel™s idealism,™™ European Journal of Philosophy (1999), pp. 164“89; repr. as ch. 7 in Tales
of the Mighty Dead: Historical Essays in the Metaphysics of Intentionality (Cambridge, Mass.:
Harvard University Press, 2002).
On this point, see my ˜˜Point of view of man or knowledge of God: Kant and Hegel on
concept, judgment and reason.™™



Kant starts the exposition of the Transcendental Ideal, in the Critique of
Pure Reason, by stating what he calls the ˜˜principle of complete determin-
ation™™ in the following terms: ˜˜Every thing . . . as to its possibility . . . stands
under the principle of complete determination [durchgangigen ¨
Bestimmung], according to which, among all possible predicates of things,
insofar as they are compared with their opposites, one must apply to it™™
(A572/B600). This principle is susceptible to different interpretations.
I suggest it has, according to Kant, a legitimate, critical interpretation,
which emerges from the Transcendental Analytic as a whole.1 I shall
consider that interpretation in a moment. But it also has an interpret-
ation in the context of rational metaphysics, from which Kant inherits
the principle in the first place.2 In this context, ˜˜complete determination™™
means complete determination by the intellect alone. As it gradually

Thus at the end of section two of the Transcendental Ideal, to the question: ˜˜How does
reason come to regard all the possibility of things as derived from a single possibility,
namely that of the highest reality, and even to presuppose these possibilities as contained
in a particular original being? ™™ Kant answers by sending us back to ˜˜the discussions of the
Transcendental Analytic themselves™™ (A581/B609). In what follows I shall attempt to out-
line what I take to be the most important features of those discussions for understanding
what the critical version of the ˜˜principle of complete determination™™ might be.
See Alexander Gottlieb Baumgarten, Metaphysica (Halle, 1739, repr. in Kant, AAXvii),
x148; Christian Wolff, Philosophia Prima sive Ontologia (Frankfurt-am-Main and Leipzig,
1736); repr. in Christian Wolff, Gesammelte Werke, ii-3, pp. 187“9.


appears while we progress through section two of the Transcendental
Ideal, this interpretation is one to which reason, according to Kant, is
inevitably drawn, and which leads to the dialectical reasoning that Kant
calls the ˜˜Transcendental Ideal,™™3 in accordance with the illusory principle
stated at the beginning of the Transcendental Dialectic: ˜˜If the condi-
tioned is given, then the totality of its conditions is also given.™™ In this case:
if limited realities are given, then the absolutely unlimited totum realitatis is
also given. This totum realitatis is then posited as a distinct being, the
ground of all finite reality: the ens realissimum of rational theology.4
The Transcendental Ideal is not the first instance in the Critique of Pure
Reason where Kant criticizes the rationalist notion of a totum realitatis, an
unlimited whole of reality. One memorable previous occasion for such
criticism was the appendix to the Transcendental Analytic, On the
Amphiboly of Concepts of Reflection, and more specifically, the analysis
of the concepts of reflection: ˜˜matter, form.™™ The rationalist concept of a
whole of reality, or unbounded reality, was then cited as a typical
instance of the error of intellectualist philosophers, according to which
the matter of thought (positive determinations or realities, thought by
concepts) is prior to its form (relations of these determinations according
to the principle of contradiction). ˜˜In respect to things in general,
unbounded reality was viewed as the matter of all possibility, but its
limitation (negation) as that form through which one thing is distin-
guished from another in accordance with transcendental concepts™™

Kant sometimes calls ˜˜Transcendental Ideal™™ the reasoning that leads to the representation
and hypostatization (positing as an existing object) of an ens realissimum (see A340/B398).
But he more often calls that representation itself, as an archetype and source of all reality,
˜˜the Ideal of pure reason™™: see A568/B596, A569/B597, A574/B602. Here I am referring to
the Transcendental Ideal in the first sense. Later in this chapter the expression
˜˜Transcendental Ideal™™ will mostly be used in the second sense.
On the steps of the illusion, cf. A582“3/B610“11. On the characterization of the dialectical
reasoning called ˜˜Ideal of pure reason,™™ cf. A340/B398. For a careful analysis of the steps of
Kant™s argument in section two of the Transcendental Ideal, see Michelle Grier, Kant™s
Doctrine of Transcendental Illusion (Cambridge: Cambridge University Press, 2001),
pp. 230“51. In the course of her analysis, Michelle Grier subjects to acute scrutiny what I
call, in the paper that was the original version of this chapter, the ˜˜critical version™™ of the
principle of complete determination and the ˜˜critical reduction™™ of the rationalist notion of
an ens realissimum (see Beatrice Longuenesse, ˜˜The Transcendental Ideal, and the unity of
the critical system,™™ Proceedings of the Eighth International Kant Congress, Milwaukee 1995
[Milwaukee: Marquette University Press, 1995], i , pp. 521“37. And Grier, Transcendental
Illusion, pp. 237“48). I think her criticisms are often well taken, and I have tried accordingly
to clarify my view in revising the paper for this chapter, which on several points extensively
revises the earlier paper. See also my review of her book in Mind, vol. 112, no. 448 (2003),
pp. 718“24, esp. p. 723.

(A267/B323, translation modified).5 To this conception, Kant then opposed
his own conception of the primacy of form over matter: the forms of
sensibility being a priori and making possible the consciousness of sensa-
tions, and therefore the matter of appearances as that which corresponds to
sensation (cf. A20/B34); the forms of thought being a priori and making
possible concepts and objects recognized or thought under these concepts.6
Now, relating Kant™s criticism of the Transcendental Ideal in the
Transcendental Dialectic to his criticism of the concept of the unbounded
whole of reality in the Amphiboly of Concepts of Reflection is interesting
for at least two reasons. First, it gives us a route to the critical reduction of
the rationalist notion of a whole of reality, by sending us back to
the exposition of the forms of sensibility and understanding and their
respective relation to their matter, according to the Transcendental
Analytic. But second, and less generally acknowledged, it also sends us
forward, to the analysis of matter and form of thought in the First
Introduction to the third Critique.7 Indeed, I want to suggest that
the critical version of the concepts of reflection ˜˜matter, form™™ expounded
in the Amphiboly chapter of the first Critique finds its ultimate develop-
ment in the concepts of matter and form (matter as ˜˜logical genus™™ and
its complete specification in the form of a system) which guide reflective
judgment according to the First Introduction to the third Critique.
My goal in this chapter is therefore threefold.
First, I propose to sort out the legitimate (critical) and illegitimate
(intellectualist) uses of the ˜˜principle of complete determination™™ in
section two of the Transcendental Ideal. In doing so, I shall be primarily

Guyer and Wood translate: ˜˜In respect to things in general, unbounded reality is regarded
as the matter of all possibility . . . ™™ Kant uses the past tense: ˜˜Auch wurde in Ansehung der
Dinge ¨ berhaupt . . . ™™ It is important to translate this past tense to make it clear that Kant is
describing a view made irrelevant by the critical standpoint he advocates. I should add that
in the original version of this paper, I said that in the Amphiboly chapter, Kant denounces
the illusion of rationalist metaphysicians (or what he calls ˜˜the intellectualist philosopher™™ “
see A367/B323). As Michelle Grier has pointed out to me, the doctrine of illusion belongs in
the Transcendental Dialectic, not the Transcendental Analytic. Here we can talk only of an
error. It will turn out, from the argument of the Transcendental Dialectic, that this error is
itself kept alive by an unavoidable illusion of reason. On this point, see below, esp.
pp. 233“4.
In the Amphiboly, Kant mentions only the primacy of form over matter in the sensible
given: the primacy of forms of intuition over sensations and thus appearances (cf. A267“8/
B323“4). But I think the point can be extended to the relation between matter and form of
thought: when thought is sensibly conditioned, its form is prior to its matter. I shall argue
for this in the second part of this chapter. See also KCJ, pp. 147“63.
Cf. Critique of the Power of Judgment, First Introduction, AAxx, pp. 211“17. And see below in
this chapter, pp. 230“2.

concerned, not so much with Kant™s account of the illusion of reason
which he calls the ˜˜transcendental ideal,™™ as with what I have called its
˜˜critical reduction™™: what the principle of complete determination and
the related notion of a whole of reality amount to, once they are disen-
tangled from the rationalist illusion.
Second, I intend to compare the concepts of totum realitatis and ens
realissimum expounded in section two of the Transcendental Dialectic,
with the criticism of those same concepts expounded in the Amphiboly
of Concepts of Reflection.
Third, and most importantly, I shall suggest that the analysis
and critical reduction of the transcendental ideal opens the way to an
articulation of reflection and determination in cognition which puts
the first Critique in closer connection to the third than is generally
recognized, and therefore puts both first and third Critiques beyond
the commonly assumed strict dichotomy between what Kant calls, in
the third Critique, ˜˜determinative™™ and ˜˜reflective™™ uses of judgment.
This has important consequences, which I shall briefly address at the
end of this chapter.
There is on my part an underlying conviction guiding the path I
propose to take, from Kant™s criticism of the transcendental ideal, back
to his criticism of the intellectualist conception of the unbounded whole
of reality, in the Amphiboly of Concepts of Reflection, and forward again
to the Introduction to the Third Critique. My conviction is that the
transcendental ideal proper (the pure concept of an ens realissimum
whose origin Kant traces back to an unavoidable and, once properly
recognized, ultimately beneficial illusion of reason) plays a less indispens-
able role than Kant claims it does, by the terms of his own analyses, if we
follow these analyses through each of the steps I just outlined. I leave it to
the reader to judge if my conviction is adequately supported.

Kant™s criticism of the Transcendental Ideal
At the beginning of section two of the Transcendental Ideal, Kant con-
trasts the ˜˜determinability™™ of a concept and the ˜˜complete determin-
ation™™ of an individual thing. The notion of determination here at work
is explained in the Jasche Logic, x15, where determination is opposed to
Through continued logical abstraction, higher and higher concepts
arise, just as through logical determination, on the other hand, lower

and lower concepts arise. The greatest possible abstraction yields the
highest or most abstract concept “ that from which no determination can
be further thought away. The most fully achieved determination would
yield a thoroughly determinate concept [conceptum omnimodo determinatum]
i.e., one to which no further determination might be added in thought.
Note. Since only individual things, or individuals, are thoroughly
determinate, there can be thoroughly determinate cognition only as
intuitions, but not as concepts; with respect to the latter, logical determin-
ation can never be regarded as completed.8

Determination here clearly means: specification. To determine a con-
cept is to produce a specification of it by adding to the initial concept a
mark that is not analytically contained in it. With respect to such a mark
our initial concept is indeterminate, it can be determined (specified) by
predicating of it either the affirmation, or the negation of an additional
mark: animals are either rational or non-rational, human beings are
either Athenians or Barbarians (non-Athenians), and so on. In this
sense, it is quite clear that only representations of individual things are
fully determinate, namely not further determinable or specifiable. For a
Leibnizian, such a thoroughly determined representation is an ultima
species, an ultimately specified concept. For Kant, it can only be an
intuition. The only fully determinate (not further determinable, i.e.
specifiable) representation is an intuition. Correspondingly, objects are
fully determinate, i.e. singular objects, only insofar as they are objects of
It may seem strange to say that only intuitions are fully determinate,
since, as is well known, for Kant an object which would be ˜˜merely™™ an
object of intuition would remain ˜˜indeterminate™™ (appearances are
˜˜indeterminate™™ objects of empirical intuition, they are determined as
objects, or phenomena, only by being thought under concepts: cf. A20/
B34; A249). This ambiguity is due to Kant™s ambivalent relation to the
rationalist tradition: on the one hand he maintains, against the rational-
ists, that only sensible intuitions, not concepts, are singular. Therefore, if
determination is specification, only sensible intuitions are fully determin-
ate. But on the other hand, it remains true that specification is a
conceptual operation. We determinately know an object only by

Jasche Logic, AAix, p. 99. What Guyer and Wood, for the Critique of Pure Reason, and
J. Michael Young, for the Jasche Logic, translate as ˜˜thoroughgoing determination™™ (omni-
modo determinatio, durchgangige Bestimmung), is what I also call, in the main text, ˜˜complete
determination.™™ Both translations are correct, but the latter is philosophically more

concepts: to determine an object for the intuition is to know it under
concepts, and we know it as determinately as our concepts are specified.
Now, this twofold meaning of ˜˜determinate™™ (singular, therefore intui-
tive; but determined by concepts) accounts for Kant™s adoption of the
˜˜principle of complete determination™™ which he inherits from the
Leibnizian rationalists, and at the same time accounts for the peculiar
meaning he assigns to this principle in the context of the Transcendental
Kant formulates this principle in the terms I quoted above: it says, of
every singular thing, that ˜˜among all possible predicates of things, insofar as
they are compared with their opposites, one must apply to it™™ (A572/B600).
Now, such a principle seems to make no sense at all unless one supposes
that one can indeed think, i.e. presuppose as given, ˜˜all possible pre-
dicates [and] their opposites.™™ Without such a presupposition, one is
simply left with the logical principle of contradiction on the one hand
(it is not possible to attribute to one and the same thing, considered
under the same respect, a predicate and the negation of that predicate);
and with the principle of excluded middle on the other hand (given a
pair of contradictory predicates, one or the other must be predicated of a
thing, there is no third alternative). What the principle of complete
determination adds to these two logical principles is precisely the refer-
ence to the totality of possible predicates. Kant indicates quite clearly this
difference between the merely logical principles of contradiction and
excluded middle, and the principle of complete determination:
˜˜through this proposition predicates are not merely compared logically
with one another, but the thing itself is compared transcendentally with
the sum total of all possible predicates™™ (A573/B601, emphasis mine).
But why should one admit such a principle, if logic does not demand
it? Why should we not be content with admitting the principles of
contradiction and excluded middle as rules for relating concepts and
thus for further and further determining our concepts of objects? For a
rationalist of the Leibnizian-Wolffian school, the answer is that the
principle of complete determination adds to these logical principles
the metaphysical principle that states how objects are individuated.
Each is a unique combination of affirmations and negations of essential
determinations or perfections in the divine understanding. Moreover,
this is how they are determined to exist, or on the contrary, to remain
mere possible components in unactualized possible worlds, according to
the principle of fitness, i.e. the wisdom of God™s choice. But Kant does
not consider that objects are individuated by complete determination

accessible to pure intellect. He expressly denies this. Objects are given in
space and time and individuated as objects of sensible intuition. So what
is his reason for affirming a ˜˜principle of complete determination™™ such
as this? Kant™s answer to this question in the Transcendental Analytic has
two components. The first is the role assigned to infinite judgment in the
table of logical functions of judgment. The second is the role of the unity
of apperception, and ultimately, of the unity of experience, in the
Transcendental Deduction of the Categories
So, very briefly on each of these two points.
(1) An infinite judgment, for Kant, is a judgment in which I affirm of a
subject-concept a predicate that is itself the negation of a predicate: ˜˜A is
not-B.™™ In doing so, I locate the subject-concept in the unlimited sphere
of all possible beings, to the exclusion of the sphere of the negated
predicate. Such a judgment, says Kant, does not have to be considered
in general logic, which ˜˜abstracts from all content of the predicate (even
if it is negative)™™ (A72/B97). In transcendental logic, on the contrary, it is
important to consider those judgments which take into consideration an
infinite logical extension (they locate the subject-concept in the ˜˜infinite
sphere of all possible beings™™), while being ˜˜limiting with regard to the
content of cognition™™ (A73/B98): the only determinate information pro-
vided by the predicate is the exclusion of the subject-concept from the
determinate sphere of a specific concept. The exclusion of infinite judg-
ment from general logic and the claim of its usefulness ˜˜in the field of
pure a priori cognition™™ exactly parallels the restriction of the principle of
complete determination to the field of transcendental philosophy.
Indeed, some Reflexionen call infinite judgment ˜˜judgment of complete
In the Transcendental Ideal, however, Kant associates complete
determination not with the form of infinite judgment, but with the
form of disjunctive judgment as the potential major premise of a dis-
junctive syllogism. And this certainly makes sense: left to itself, the
infinite judgment would leave almost entirely indeterminate, unspeci-
fied, the infinite sphere to which the sphere of the subject-concept is said
to belong. But on the other hand, disjunctive syllogism can function as
the ground for complete determination only if its disjunctive major
premise states the complete division of the infinite sphere of a concept
whose division would yield all concepts of possible beings: the logical

Cf. Reflexion 3063, AAxvi, p. 636.

form of complete determination has to be jointly grounded in the forms
of infinite and disjunctive judgments.
(2) Now, from the standpoint of the Transcendental Analytic, what
makes possible the use of infinite-cum-disjunctive judgment, i.e. the
indefinitely repeated endeavor to determine any subject-concept by its
inclusion in, or exclusion from, the sphere of all other known concepts of
things, is the unity of apperception, as described in the Transcendental
Deduction of the Categories: only if one and the same act of comparison
and reflection and before this, one and the same act of synthesis achieved
in order to compare and reflect, organizes our perceptions, can all
predicates be compared to all other predicates, and therefore can con-
cepts of objects be ever further specified. This is how the unity of
apperception gives rise to the unity of experience: the unified act of
synthesis and analysis (comparison and reflection) is what connects
objects in one space and one time, and reflects them under concepts.
The infinite sphere whose division would yield all concepts of possible
entities, in which infinite judgment thinks the object thought under its
subject-concept is then the infinite sphere of the concept: ˜˜object given
in space and time,™™ that is to say ˜˜object of experience.™™ The form of
disjunctive judgment is the logical form according to which this infinite
sphere is determined.
So this is how Kant can affirm on his own, critical grounds a ˜˜principle
of complete determination™™: any singular object of experience is fully
determinate by virtue of its being comparable to every other possible
object, i.e. by virtue of its belonging in the infinite sphere of the concept:
˜˜object of experience,™™ in which its concept can be related to all other
concepts either positively or negatively. Contrary to what was the case
for rationalist metaphysics, it is not necessary to suppose that the totality
of possible predicates be actually given (in God™s infinite understanding)
to assert that every thing is either positively or negatively determined in
relation to every possible predicate. It is sufficient to have shown that the
form of our understanding is such that necessarily, any determination of
an individual thing (namely, any mark of the concept under which we
cognize it) determines it positively or negatively relative to all the con-
cepts defining the possible subspheres of the one infinite sphere of the
concept: ˜˜object of possible experience,™™ or ˜˜object given in space and
If this is so, the principle of complete determination Kant formulates
at the beginning of section two of the Transcendental Ideal (A571“2/
B600“1) is not a new principle, in the context of the first Critique. It is a

principle that Kant could have given as a corollary of the principle of all
synthetic judgments: ˜˜the conditions of the possibility of experience are
the conditions of the possibility of the objects of experience™™ (cf. A111,
A158/B197).10 By defining complete determination in terms of concepts
alone, rationalist metaphysicians have run away with an illusory version
of a perfectly sound principle of cognition.
The same can be said of the idea of the sum total of all possibility,
which is presupposed in the statement of the principle; and also of the
idea of the sum total of all reality, omnitudo realitatis, which depends on
the first. This is how.
We already saw how the idea of a sum total of all possibilities (the
totality of all possible predicates) is contained in the very statement of the
principle of complete determination, and is precisely what makes it
different from the logical principles of contradiction and excluded mid-
dle. But what can we understand by ˜˜possible predicate™™? According to
the Transcendental Analytic, a possible predicate is a predicate that
˜˜agrees with the formal conditions of experience (in accordance with
intuition and concepts)™™ (from the Postulates of Empirical Thought in
General, A218/B265). If this is so, comparing the predicates of an indi-
vidual thing with the sum total of possible predicates is comparing them
with all the predicates which agree (1) with the forms of intuition,
(2) with the universal relations made possible in these forms by the cate-
gories and their schemata, and (3) with the present state of our empirical

What I mean by this is that if the principle (˜˜the conditions of the possibility of experience
are the conditions of the possibility of the objects of experience™™) is true, and if making use
of the forms of infinite and disjunctive judgment is among the conditions of possibility of
experience (as I recalled earlier in this chapter, see pp. 217“18), then it follows that, by
virtue of these forms, every object falls under, or is excluded from, the sphere of every
possible predicate, and thus the principle of complete determination as defined by Kant in
section two of the Transcendental Ideal is true of all objects of experience (i.e. all things as
appearances). In her discussion of the original version of this chapter, Michelle Grier
criticizes me for saying (according to her) that ˜˜the principle of complete determination is
not a ˜new™ principle at all, but essentially reiterates the already established doctrine that
the ˜conditions of the possibility of experience are the conditions of the possibility of the
objects of experience™™™ (see Grier, Transcendental Illusion, p. 239). But I do not take the two
principles to be identical, I only take the one (the principle of complete determination) to
follow from the other (the principle of the possibility of experience) once it is understood
that the latter includes the role of infinite and disjunctive judgment in reflecting objects
under concepts and thus coming up with representations of individuated objects for our
intuitions. Moreover, it remains of course true that this version of the principle is different
from the illusory, purely intellectual interpretation of it. Michelle Grier™s concern, in ch. 7
of her book, is mainly with the latter; my concern is mainly with clarifying what a critical
version of the principle of complete determination might be.

concepts. Now, we also know from the Transcendental Analytic that
among these empirical concepts, some are ˜˜positive determinations™™ or
realities, some are negative determinations, or negations. Realities are
˜˜what corresponds to sensation,™™ negations are what corresponds to the
absence of a sensation, or ˜˜a concept of the absence of an object™™ (see the
schemata of the categories of quality, A143/B182; also the table of nothing
at the end of the Analytic, A291/B347). Because of his relating reality to
sensation, and negation to the absence of sensation, Kant considers that
positive determinations, or realities, are prior to negative determina-
tions, or negations, which in fact are meaningless if one does not have a
prior concept of the corresponding positive determination. This being
so, saying that an individual thing is fully determined if it is compared to
the sum total of possible predicates can be reduced to saying that it is
fully determined if it is compared to the sum total of possible positive
predicates, or realities. From this, the comparison with negative predi-
cates analytically follows. Therefore, there is again a perfectly legitimate,
critical reading for the move from the principle of complete determin-
ation to the supposition of a sum total of all possibilities, and from there
to the supposition of a sum total of all realities, or totum realitatis.
Except, of course, in the critical context this totum realitatis remains a mere
idea: there is no given totality of positive predicates, the mere limitation of
which would give us the complete determination of each singular thing.
Predicates are not given once and for all in God™s infinite understanding,
but generated by the ˜˜logical use of the (human) understanding™™ reflecting
upon the sensible given. In other words, they are generated by what Kant
calls, at the end of section two of the Transcendental Ideal, ˜˜the distributive
use of the understanding in empirical knowledge™™ (A582/B610). So, from
the standpoint of the Transcendental Analytic, the representation of a totum
realitatis as the complete whole of positive determinations of things can only
be a goal which reason sets to the understanding for the improvement
of its knowledge, not an actually given whole. The illusion of rational
metaphysics is precisely to think that such a whole is actually given in
pure intellect alone, rather than having to be generated by the sensibly
conditioned understanding.
On the other hand, even from the critical standpoint, reality, as ˜˜that
which corresponds to sensation,™™ does indeed have to be presupposed as
given as a whole in space and time. In other words, the distributive use of
the understanding in experience does presuppose some collective whole
of experience and, corresponding to it, the unanalyzed whole of what is
given in space and time. Just as concepts of spatial and temporal

properties of objects presuppose space and time as formal intuitions,
˜˜infinite given magnitudes,™™ it seems that realities as positive determin-
ations of things which are objects of empirical concepts presuppose the
whole of reality as that which fills space and time. Kant says precisely this,
it seems to me, when he explains why reason not only forms the idea of a
totum realitatis, but moreover forms the erroneous belief that this totum
actually exists.11 The legitimate ground for this belief, he says, is that in
every one of our efforts to cognize empirical realities or empirical posi-
tive predicates of things, some totum realitatis must indeed be presup-
posed as existing (although Kant does not mention this particular point,
I suggest he may have in mind the fact that the principle of the perma-
nence of substance, for instance, would make no sense without such a
presupposition). But one should not confuse this experientially presup-
posed whole of reality with a discursively thought whole of realities or
positive determinations.
This distinction is clearly made in the following passage from the end
of section two of the Transcendental Ideal:
an object of sense can be completely determined only if it is compared
with all the predicates of appearance and is represented through them
either affirmatively or negatively. (A581/B609)

This, I take it, relates every positive predicate of empirical things to the
distributive use of the understanding in experience, and therefore, the
merely distributive, not collective, totality of discursively reflected posi-
tive determinations. Then Kant goes on:
But because that which constitutes the thing itself (in appearance),
namely the real, has to be given, without which it could not be conceived
at all, but that in which the real in all appearances is given is the one all-
encompassing experience, the material for the possibility of all objects of
sense has to be presupposed as given in one sum total [als in einem
Inbegriffe]; and all possibility of empirical objects, their difference from
one another and their thoroughgoing determination, can rest only on
the limitation [Einschrankung] of this sum total. Now in fact no other
objects except those of sense can be given to us, and they can be given
nowhere except in the context of a possible experience; consequently,
nothing is an object for us unless it presupposes the sum total of all
empirical reality [den Inbegriff aller empirischen Realitat] as condition of
its possibility. (A582/B610)

Cf. A581/B609.

This time, Kant states that every empirical thing, as given in intuition,
is related in experience to a presupposed whole of reality. It is just a few
lines after this passage that Kant goes on to say that we form the illusory
representation of an existing whole of positive determinations or reali-
ties because ˜˜we dialectically transform the distributive unity of the
empirical employment of our understanding into the collective unity of
a whole of experience.™™ Such a transformation, it seems to me, is the
transformation of the never-ending progress of the discursive use of the
understanding into the (illusory representation of) a given totality of
conceptual determinations of objects of experience. This illusory repre-
sentation of a ˜˜collective whole of realities™™ is ultimately hypostatized
(posited as a distinct being) into the representation of an ens realissimum,
as the single ground of all reality.12
Kant seems to waver between different formulations when he endea-
vors to lay out the relation between this purely intelligible being and its
limitations. On the one hand, he suggests that the relation of the ˜˜man-
ifoldness of things™™ and the ˜˜concept of the highest reality™™ is analogous
to that of figures and infinite space (A578/B606). But immediately after
that, he corrects the formulation and says that the highest reality is
related to the possibility of all things rather as their ground than as
their whole (Inbegriff ) (A579/B607). However one takes it, the relation
of the highest being to limited realities is merely the relation of an idea to
concepts: the relation of the purely intellectual idea of an omnitudo
realitatis, i.e. totality of positive predicates, to concepts of things in gen-
eral and their determinations, as limitations (cf. A579/B607).

Kant™s claim, at the end of section two of the Transcendental Ideal, is thus that the only
totum realitatis whose existence we can meaningfully assert is the whole of reality given to
the senses, which we presuppose as a condition for the unity of experience and for the
distributive use of the understanding by which realities (particular positive determinations
of things) are thought under concepts. This is the only refutation Kant ever gives (without
saying that he is giving it) of his own pre-critical proof of the existence of God. That proof is
none of the three proofs Kant goes on to criticize in the next sections of the Transcendental
Ideal. It rests on the principle that the matter of all possibility has to depend on one single
totum realitatis, an individual being or ens realissimum (see The Only Possible Argument in
Support of a Demonstration of the Existence of God, in Theoretical Philosophy, AAii, pp. 77“81).
Kant now says that the relation between a totum realitatis and limited realities, if thought by
pure concepts, is just this: a relation between an idea and concepts. As a relation between
existences, it is nothing over and above the relation between (1) the indeterminate whole of
reality presupposed for the distributive use of the understanding in experience, and (2)
the determinate limitations of that whole, reflected under concepts of realities or positive
determinations of things, negations (the absence of positive determinations) and limit-
ations (realities limited in relation to other realities).

This reduction of the purely intellectual relation between ens realissimum
and limited realities to a relation between an idea and concepts makes it a
mere form. Here Kant™s analysis complements and develops a theme that
was announced as early as the appendix to the Transcendental Analytic:
the Amphiboly of Concepts of Reflection, to which I shall now turn.
Let me first recapitulate. I have suggested, following an indication by
Kant himself (cf. A581/B609), that we can find in the Transcendental
Analytic the resources for a critical interpretation of the ˜˜principle
of complete determination™™ formulated at the beginning of section
two of the Transcendental Ideal. We can also find the resources for a
representation of the totum realitatis conditioning the application of the
principle that would conform to the restrictions of the use of the under-
standing in cognition established in the Transcendental Analytic. I have
suggested that according to this critical interpretation, the whole of
reality that grounds the representation of the complete determination
of things is the indeterminate whole of reality given in space and time,
presupposed in any empirical use of the understanding giving rise to
discursively represented realities or positive determinations of things (as
appearances). However, recognizing that the only existence of such a
whole is that of the whole of reality presupposed for the use of the
understanding in experience does not do away with the purely intel-
lectual representation of the relation between a discursively thought
(conceptual, intellectual) totum realitatis and the particular realities or
determinations of things. I now propose to compare what Kant says, in
the Transcendental Ideal, about this intellectual representation, to
Kant™s criticism of the rationalist totum realitatis and ens realissimum, in
the Amphiboly of Concepts of Reflection.

The Amphiboly of Concepts of Reflection
In the Transcendental Analytic, Kant™s criticism of the intellectualist
philosopher™s notion of a whole of reality occurs in the context of
Kant™s examination of the fourth pair of concepts of reflection: matter,
form. To understand the point Kant wants to make, it will be useful to
recall what he generally means by ˜˜concepts of reflection,™™ and the
specific role assigned to the last pair of these concepts.
Kant distinguishes two types of reflection: ˜˜logical reflection,™™ which
he also calls logical comparison; and ˜˜transcendental reflection.™™ Logical
reflection is ˜˜a mere comparison,™™ where ˜˜one completely abstracts
from the cognitive power to which the given representations belong™™

(A262/B318). Or again: ˜˜The concepts can be compared logically with-
out worrying about where their objects belong, whether as noumena
to the understanding, or as phenomena to the sensibility™™ (A269/B325).
This logical reflection, I suggest, is the same as the ˜˜logical use of the
understanding™™ which according to the 1770 Inaugural Dissertation was
common to all cognition and by which
when a cognition has been given, no matter how, it is regarded either as
contained under or as opposed to a characteristic mark common to
several cognitions, and that either immediately and directly, as is the
case in judgments, which lead to a distinct cognition, or mediately, as is the
case in inferences, which lead to a complete cognition.13

It is also the same as the ˜˜logical use of the understanding™™ described in
x10 of the Transcendental Analytic, where Kant characterized the
understanding as a ˜˜capacity to judge™™ (Vermogen zu urteilen) after saying
that we form concepts only in order to judge by their means (A68/B93).
In the Amphiboly, Kant indicates that this logical reflection or compar-
ison, namely (if I am right in the identifications I just suggested), the
logical use of the understanding, is guided by ˜˜concepts of reflection™™ or
˜˜concepts of comparison™™ which correspond respectively to the four
headings of the table of judgments: identity and diversity (Einerleiheit
und Verschiedenheit) for universal and particular judgments; agreement
and conflict (Einstimmung und Widerstreit) for affirmative and negative
judgments; internal and external (Innere und Außere) for categorical and
hypothetical judgments; matter and form (Materie und Form) for mod-
ality of judgments (A262/B318“A268/B324).
I understand these correspondences in the following way. First, iden-
tity and difference: we compare objects, or perhaps lower (more specific)
concepts, thought under a concept A, to find out whether they are
identical or different (einerlei oder verschieden) with regard to their being
also thought under a concept B; we thus form universal judgments (all
As are B) or particular judgments (some As are B, some As are not B).
Second, agreement and conflict (Einstimmung, Widerstreit): we compare
concepts, as regards their comprehension (the marks which belong to
them), to find out whether they are in agreement (As are B) or conflict
(As are not B). Third, inner and outer (Innere, Außere): we compare
concepts in order to find out whether one of them (say, A) contains in
itself (˜˜internally™™) the sufficient condition or ground to assert the other

Inaugural Dissertation, AAii, p. 393.

(categorical judgment: A is B) or whether an additional, ˜˜external™™
condition or ground should be added, in order to ground the attribution
of B to A (hypothetical judgment: if C is D, then A is B). Of course in this
third case, the situation, from a Kantian point of view, is complicated by
the fact that this ˜˜internal™™ or ˜˜external™™ condition for predication may
or may not have to take into account the intuition thought under the
subject-concept: the categorical, just as the hypothetical judgment, may
be analytic or synthetic. But as Kant says repeatedly, logic does not
take into account this difference, which concerns the origin and content
of the concepts, not the mere form of thought. Similarly, the description
of logical reflection in the Amphiboly chapter merely considers the form
of judgment and the concepts of reflection, or concepts of comparison,
guiding the act of judgment according to each aspect of its form.
Now the point of the Amphiboly chapter is to show that Leibniz
confused logical reflection or comparison, as I just characterized it
(comparison of concepts to form judgments, whatever the origin of
those concepts), with a comparison of objects. Leibniz thought that at
least for an infinite understanding, things could be known by concepts
alone, and therefore, the concepts or rules for comparison of concepts
could be understood as concepts or rules for comparison of things.
Things that are identical with respect to all possible predicates were
therefore numerically identical. Because no logical conflict, or contra-
diction, can be thought between two positive determinations or realities
thought by concepts alone, no conflict could be thought between two
positive determinations or realities in things. By pure concepts, a thing
could be known exhaustively through its intrinsic properties: predica-
tion under an external condition could be reduced to predication under
internal conditions; indeed these internal conditions were marks analy-
tically contained in the subject-concept. In opposition to all of this, Kant
maintains that things cannot be cognized by concepts alone. They
are given in space and time, the forms of our sensibility. Their indivi-
dual representation or intuition is radically distinct from any concept,
although concepts are of course formed by ˜˜comparison, reflection, and
abstraction™™ from what is given to sensible intuition. Therefore, logical
reflection must be complemented by transcendental reflection or
transcendental topic, which distinguishes between the comparison of
concepts and the comparison of objects given in space and time.
Notice, though, that transcendental reflection inherits its concepts of
comparison, or rules for the comparison of things in space and time,
from logical reflection. In transcendental reflection, one wonders what it

means for things, as opposed to mere concepts, to be identical or differ-
ent, in agreement or conflict, internally or externally determined. In
other words, objects are still compared through the grid of our discur-
sive understanding and its concepts of comparison or rules for compar-
ison. But the purpose of transcendental reflection is precisely to show
that these rules have to acquire a different use when they are applied to
objects given in space and time.
This is where matter and form come into the picture. These two
concepts, says Kant, ˜˜ground all other reflection, so inseparably are
they bound up with every use of the understanding™™ (A266/B322).
This gives them a status different from that of other concepts of reflec-
tion. They are second-order concepts, concepts by means of which
we are asked to reflect upon the act of comparison itself: every act
of comparison has a matter (the determinable, what is ˜˜given™™ in
thought) and a form (the determination, the processing of what is
given in thought). Kant thus makes a very un-Aristotelian use of
these Aristotelian concepts. Matter and form are matter and form not
of things, but of thought. Indeed, this is how Kant uses these concepts
consistently in the Jasche Logic:

The matter of concepts is the object, their form universality.14

The matter of the judgment consists in the given representations that are
combined in the unity of consciousness in the judgment, the form in the
determination of the way that the various representations belong, as
such, to one consciousness.15

The matter of inferences of reason consists in the antecedent propositions
or premises, the form in the conclusion insofar as it contains the

Jasche Logic, x2, AAix, p. 91.
Ibid., x18, AAx, p. 101.
Ibid., x59, AAix, p. 121. We already saw that by consequentia (or Konsequenz in German)
Kant means the relation between subject and predicate, antecedent and consequent,
concept and its divisions, in a categorical, hypothetical, or disjunctive judgment. We also
saw that it is the obtaining of the consequentia, not the independent truth-value of the
components of a proposition or an inference, that make the proposition true or the
inference valid (see above, ch. 4, pp. 97“9; ch. 6, pp. 150“5.; ch. 7, pp. 188“90). I believe
what Kant means, in the present case, is that the form of the inference is the consequentia
(the relation between subject and predicate) expressed in the conclusion, which itself
analytically results from the relations or consequentiis expressed in the premises of the

And finally:
The universal doctrine of method . . . has to deal with the form of science
in general, or with the ways of acting so as to connect the manifold of
cognition in a science.17

Now, the intellectualist philosopher™s mistake is to think that this
relation between matter and form of thought is sufficient to define the
relation between matter and form of things, which can thus be cognized
as noumena, objects of pure thought. In fact, retorts Kant, things as we
know them are not noumena, but phenomena. The matter and form of
phenomena are not matter and form of pure thought, but matter and
form of sensibility: matter as that which ˜˜corresponds to™™ sensation, form
as space and time, forms of intuition. Kant™s main point is that this being
so, the relation between matter and form of possibility is the reverse of
what the rationalists (as he understands them) made it to be: from the
rationalist standpoint, the matter of possibility is prior to its form, and
this is why the rationalist supposes an unbounded reality (the intelligible
˜˜matter™™ of all determinations of things) by limitation of which ( ¼ the
˜˜form™™ of all possible things) every particular thing is thought. But from
the critical standpoint, the form of possibility is prior to its matter. There
is of course more here than a mere reversal of priority: the very notion of
possibility is then completely redefined. The possible has no ontological,
but merely a transcendental status: what is possible is what agrees with
the formal conditions of our knowledge (intuition and concepts). In
other words, the ˜˜possible™™ has no existence of its own, be it as a pure
essence in God™s understanding, alternative possible worlds, or what-
ever else. Unbounded reality as the ground of all possibility is replaced
by something which has, left to itself, no reality (namely no positive
determination) at all: space and time, as mere forms of intuition:
[I]n respect to things in general, unbounded reality was viewed as the
matter of all possibility, but its limitation (negation) as that form through
which one thing is distinguished from another in accordance with trans-
cendental concepts. The intellectualist philosopher could not bear it that
form should precede the things and determine their possibility; a quite
appropriate criticism, if he assumed that we intuit things as they are
(though with confused representation). But since sensible intuition is
an entirely peculiar subjective condition, which grounds all perception

Ibid., x96, AAix, p. 139.

a priori, and the form of which is original, thus the form is given for itself
alone, and so far is it from being the case that the matter (or the things
themselves, which appear) should be the ground (as we would have to
judge according to mere concepts), that rather its very possibility pre-
supposes a formal intuition (time and space) as given. (A267“8/B323“4)

But given the argument of the whole chapter on concepts of reflec-
tion, which is itself a mere appendix to the argument of the Analytic as a
whole, Kant™s point can be extended: the primacy of form over matter
does not concern merely sensibility, but also discursive thought. The
whole array of forms of discursivity (the form of universality of concepts,
forms of judgment as forms of the capacity to generate concepts to be
combined in judgments, forms of syllogisms as imbedded in forms of
judgment, and finally the form of a system as the form of the unity of
empirical cognitions related to the unity of space and time) has to be
presupposed for any empirical object to be cognized under concepts. In
this developed assertion of the primacy of form over matter, the ens
realissimum of rational metaphysics and of Kant™s own pre-critical sys-
tem18 finds its overthrow. As a ground of all possibility, it is reduced to a
mere form, with no ontological status.
This point is vividly expressed in a remarkable Reflexion dated by
Adickes in 1783“4, where the ens realissimum is presented as a discursive
form corresponding to the intuitive forms of space and time:

That something be actual [wirklich] because it is possible according to a
universal concept, does not follow. But that something be actual because
it is completely determined through its concept among everything pos-
sible, and distinguished as singular [als eines] from everything possible,
means the same as: it is not a universal concept any more, but the
representation of a singular thing completely determined by concepts
in relation to everything possible. This relation to everything possible by
the principle of complete determination is the same, by concepts of
reason, as is the somewhere and some time [irgendwo und irgendwenn]
by conditions of sensible intuition. For space and time determine not
only the intuition of a thing, but at the same time its individuality by the
relation of place and instant . . .
From this it follows that the ens realissimum must be given prior to the
real concept of all possibility [zu dem realen Begriffe aller Moglichkeit vorher
gegeben sein musse]. And just as space cannot be first thought as possible,

Cf. above, n. 12.

but must be given, not as an object actual in itself, but as a mere sensible
form in which alone objects can be intuited, in the same way the ens
realissimum must be given not as an object, but as the mere form of
reason, in order to think the difference of every possible entity in its
complete determination; it must be given as an idea which is subjectively
actual, before anything is thought as possible; but from this it does not
follow that the object of this idea is actual. One sees nevertheless that in
relation to the human understanding (and its concepts) the idea of a
highest being is just as necessary as is space and time in relation to the
nature of our sensibility and its intuition.19

If we compare this to what is said of the ens realissimum in the
Transcendental Ideal, we can say the following. In the Transcendental
Ideal, Kant argued that reason, by an unavoidable illusion, forms the idea
of a totum realitatis, individuated as an ens realissimum, as the condition of
the complete determination of individual things in general. But as given
to us, things are completely determined, i.e. individuated, only insofar as
they are empirically given in space and time. The totum realitatis we do
have to presuppose as the given condition of their complete determina-
tion is thus the (indeterminate, collective) whole of reality given in space
and time. However, making this critical point was not putting an end to
the purely rational idea of a whole of reality. Rather, it was constraining us
to take it for what it is: a mere thought, without an object.
What we now see is that the Amphiboly of Concepts of Reflection had
already given a status to this thought: it is a mere form, or mode of
ordering our representation, as the highest condition of the systematic
division by virtue of which individual things are conceptually deter-
mined, i.e. located in a universal scale of genera and species. As a
discursive form, it does have a strictly intellectual status. What it does
not have is any relation to objectivity independently of the matter that it
determines (inferences, judgments, empirical concepts, and thus ulti-
mately and mediately, the matter of those empirical concepts them-
selves, i.e. the matter of appearances, ˜˜that which corresponds to
Now, this notion of form, culminating in the form of complete deter-
mination supposed to guide all reflection, brings us very close indeed to
the exposition of reflective judgment in the Introduction to the third
Critique. I now turn to this last text.

Reflexion 6289, AAxviii, pp. 558“9.

Reflective judgment and the affinity of appearances
In the First Introduction to the third Critique, Kant raises the following
question: how can we assume that what is empirically given to our effort
of cognition has such homogeneity as to allow for cognition under
empirical concepts and empirical laws? He answers this question by
stating that our power of judgment assumes, as a principle for its reflec-
tive use, that there is in fact in nature no ˜˜disturbingly unbounded
diversity of empirical laws and heterogeneity of natural forms.™™
Rather, ˜˜through the affinity of particular laws under more general
ones, nature qualifies for an experience as an empirical system.™™20
Now, such an affinity of appearances is precisely what the principle of
complete determination and its presupposition of a sum total of all
possibilities, and ultimately an ens realissimum, was meant to ground, in
the Transcendental Ideal:
Through this principle, every thing is related to a common correlate,
namely the collective possibility [die gesammte Moglichkeit], which, if it (i.e.,
the matter for all possible predicates) were present in the idea of an
individual thing, would prove an affinity of everything possible through
the identity of the ground of its complete determination. (A572/B600n)

But the critical analysis of the Ideal concluded that the ˜˜one thing™™ could
be asserted to exist only as the whole of reality presupposed for the unity
of experience, from which positive determinations or realities were
generated distributively through the empirical use of the understand-
ing. And the criticism of the Amphiboly of Concepts of Reflection had
previously given the warning that the form of affinity, or homogeneity of
the empirical given, was precisely this: a mere form, imbedded in the
concepts of reflection describing ˜˜in all its manifoldness the comparison
of representations which is prior to the concepts of things™™ (A269/B325).
The Introduction to the third Critique, it seems to me, makes the same
point. But it makes it more clearly than before, because it makes it
without going once more through the criticism of the illusory version
of complete determination in rational metaphysics. For the same reason,
the articulation of the unity conferred to nature by the universal princi-
ples of the understanding, and its unity all the way down to the sub-
sumption of individual objects under particular empirical laws thanks to
the principle of reflective judgment stated above, is fully elaborated. It is

Critique of the Power of Judgment, First Introduction, AAxx, p. 209.

often thought that this articulation consists in an opposition, or at least a
strict separation, between the determinative use and the reflective use of
the power of judgment.21 And true enough, Kant writes:
The power of judgment can be considered either as a mere power to
reflect according to a certain principle, upon a given representation, in
order to form a possible concept [zum behuf eines dadurch moglichen
Begriffe]; or as a power to determine a concept available as a ground,
by means of a given empirical representation.22

In the second aspect, one may recognize the use of the power of judgment
by means of which categories are applied to phenomena. The first aspect,
on the other hand, is the use of the power of judgment by means of which
empirical concepts and empirical laws are formed. But in fact these two
uses are, in Kant™s presentation, not opposed, but complementary, and
indeed, inseparable. There is no determination without reflection, deter-
mination by the pure concepts of understanding is indeed nothing but an
˜˜instruction for reflection.™™ What is also the case, though, is that reflection
(by which empirical concepts and laws are found) needs more than the
˜˜instruction™™ by the pure concepts of the understanding: it needs the
specific principle which presupposes, for the benefit of the power of
judgment, the affinity or homogeneity of phenomena. These two aspects:
(1) determination for reflection, (2) reflection under a principle of affinity,
are clearly stated, and stated together, in the following passage, which is
very famous but nevertheless insufficiently heeded:
With regard to the universal concepts of nature, under which a concept
of experience [ein Erfahrungsbegriff] (without any particular empirical
determination) is first possible at all, reflection already has its direction in
the concept of a nature in general, i.e. in the understanding, and the
power of judgment requires no special principle of reflection, but rather
schematizes the latter a priori [sondern schematisiert dieselbe a priori] and

Michael Friedman, for instance, suggests that an important aspect of the ˜˜transition project™™
of the Opus Postumum was to overcome the discontinuity between the formerly ˜˜entirely
independent domains™™ of reflective and determinative judgments. See his Kant and the
Exact Sciences, p. 262, and in general pp. 242“64. Friedman does acknowledge, however,
that reflection should be seen as playing a role in the Metaphysical Foundations. But he seems to
think that Kant became aware of this fact only at the time of the ˜˜transition project™™ and under
the spur of the problems raised in connection with the Aether Deduction (ibid., p. 320). If
I am right in the analyses I propose, the reflective aspect of judgment played an essential role
in the argument of the first Critique itself, and therefore also in Kant™s conception of the role of
the categories in the Metaphysical Foundations. On this point, see my concluding remarks.
AAxx, p. 211; see also AAv, p. 179.

applies these schemata to every empirical synthesis, without which no
judgment of experience would be possible at all. The power of judgment
is here in its reflection at the same time determining [emphasis mine] and its
transcendental schematism serves it at the same time as a rule under
which given empirical intuitions are subsumed.
But for those concepts which must first of all be found for given
empirical intuitions, and which presuppose a particular law of nature,
in accordance with which alone particular experience is possible, the
power of judgment requires a special and at the same time transcenden-
tal principle for its reflection . . . All comparison of empirical representa-
tions in order to cognize empirical laws . . . presupposes that even with
regard to its empirical laws nature has observed a certain economy
suitable to our power of judgment and a uniformity [Gleichformigkeit]¨
that we can grasp, and this presupposition, as an a priori principle of the
power of judgment, must precede all comparison.23

A lot should be said, which I cannot say here, about the role of schema-
tism in the ˜˜instruction for reflection.™™ I at least want to note this:
the relation between schematism and reflection was already present in
the first Critique, if one takes seriously what was said there about ˜˜logical
reflection™™ as guided by concepts of comparison, or concepts of reflection,
corresponding to each of the logical functions of judgments. Given that
these logical functions are also the origin of the categories, I suggest that the
picture that emerged from the first Critique was the following: understand-
ing, or the power of judgment, guides the syntheses in imagination of
what is given in space and time to make it analyzable according to
logical forms of judgment. This is how it produces schemata, or universal
forms of synthesis, by means of which appearances will ultimately be
recognized as subsumable under the categories. Categories, or pure con-
cepts of the understanding, are the ˜˜universal representations™™ of the
syntheses implemented by the understanding in order to make the sensible
given ˜˜reflectable™™ under empirical concepts combined in judgments.
Forms of judgment are forms of reflection, guided by ˜˜concepts of reflec-
tion™™ as described in the Amphiboly chapter. These forms culminate in the
form of systematicity, announced as early as the metaphysical deduction of
the categories in the joint forms of infinite and disjunctive judgment which
remain the constant horizon of the ˜˜distributive use of the understanding™™
as described in the critical reduction of the ens realissimum at the end of
section two of the Transcendental Ideal.

AAxx, pp. 212“13.

It may be objected that such an account of systematicity gives
short shrift to the regulative role of the ideas of pure reason, and
in particular the Ideal, expounded most notably in the appendix
to the Transcendental Dialectic. Well, actually, I do think that the
Transcendental Analytic, together with its appendix, was sufficient to
offer an account of systematicity which does away with the ontological
illusion carried by the Ideal of pure reason. And the conclusion of
section two of the Transcendental Ideal seems to endorse precisely
such an account. In the reading I have suggested, Kant criticizes the
hypostatization of the idea of a totum realitatis as an ens realissimum. And he
endorses the necessary supposition of a totum realitatis as the necessary
condition of the unity of experience and of the distributive use of the
understanding in generating empirical concepts of specific realities or
positive determinations of things.
Why then does Kant nevertheless affirm the necessity of the illusion,
and argue for a regulative use, not merely of the idea of a totum realitatis,
but even of the corresponding Ideal (the ens realissimum)? The answer to
this question, I believe, lies in the relation Kant establishes between the
theoretical and the practical use of reason. In this regard, there is an
interesting symmetry between the appendices to the two parts of the
Transcendental Logic. The appendix to the Transcendental Analytic, on
the Amphiboly of Concepts of Reflection, attributes the erroneous,
intellectualist conception of a whole of reality to a kind of inertia of the
understanding pursuing its course beyond its legitimate use, in which it
should be bound by the senses; overcoming this inertia and waking up
the understanding to the bounds of the senses leads to rejecting the idea
of the primacy of the matter of thought over its form (i.e. the primacy of
the unbounded whole of reality over its limitations) in favor of the
primacy of form over matter in experience. But the appendix to the
Transcendental Dialectic, on the regulative use of the ideas of pure
reason, explains the necessity of the idea and of the corresponding
Ideal not by the inertia of the understanding, but rather, by the dynamics
of reason,24 which demands both the idea (the totum realitatis) and the
supposition of its object (the ens realissimum), for its own practical
purposes. Therefore the idea and its object are also called upon to
play a role, but merely regulative, in cognition, and the appendix
to the Transcendental Dialectic balances out the appendix to the

I borrow this expression from Gerd Buchdahl: cf. Kant and the Dynamics of Reason: Essays on
the Structure of Kant™s Philosophy (Oxford: Blackwell, 1992).

Transcendental Analytic. The latter sends us backwards, to the whole
development of the Analytic, for an account of the form of systematicity
in the theoretical use of the understanding. The former sends us for-
ward, to practical reason and its postulates, meanwhile restricting all of
the ideas to a regulative role in the theoretical realm.25 In this way, it also
points to a unity between the realm of nature and the realm of freedom
which the third Critique will further elaborate as the articulation of two
legislations, that of understanding and that of reason.
What exactly, then, does the third Critique add to this picture? First of
all, it is no small achievement of its Introduction, in both editions, that it
makes explicit the cooperation between determination and reflection in
judgment. But this, if I am right, is not so much an innovation as just this:
a clarification. The true novelty of the third Critique consists in adding to
the picture I just drew, and relating to the cognitive use of judgment,
those merely reflective judgments, judgments where reflection is with-
out determination: aesthetic and teleological judgments. And finally, the
novelty of the third Critique consists in making the ˜˜merely reflective™™
judgments the locus of the articulation between the legislation of reason
through the concept of freedom, and the legislation of the understand-
ing through concepts of nature. So I am not trying to say that nothing
new could be added after the first Critique. What I am claiming is that, so
far as the problem of complete determination is concerned, the terms of
the problem, and the manner of its solution, present a fundamental
continuity when we move from the first to the third Critique. I think
that this continuity, if properly understood, gives added strength to both
Critiques, and makes both of them more interesting and convincing than
they would otherwise be. It allows for a better understanding of the
relation between the cognitive and the non-cognitive use of judgment,
and of the relation of both kinds of judgment to practical reason.
In sum: I have argued that the ˜˜principle of complete determination™™
formulated at the beginning of the Transcendental Ideal is not only a
principle that pure reason, when abstracted from all relation to the
senses, holds to be true of things in general. It is also a principle Kant
holds to be true of things as appearances. I have attempted to show what
meaning the principle has in relation to things as appearances, in light of

It would be desirable to specify in the case of each idea (the soul, the world, God, or the ens
realissimum) what its specific regulative role is, and how it relates to the form of systematicity
expounded in the first section of the appendix to the Transcendental Dialectic (A642“68/
B670“96). Within the limits of this chapter, my goal was only to clarify the role of the idea
of a totum realitatis and the related ideal (hypostatized singular object, ens realissimum).

the teachings of the Transcendental Analytic. I have argued that clarify-
ing the meaning of the principle in this context also gives us a better
understanding of what remains of the purely rational principle once
confronted with the severe restrictions the Critique of Pure Reason places
on any attempt to claim determinate cognition of things independently
of the conditions of sensibility. In comparing the criticism of the
Transcendental Ideal in section two of the Transcendental Ideal to
what was said of the rationalist notion of ens realissimum in the
Amphiboly of Concepts of Reflection, I have concluded that from the
standpoint of the Transcendental Analytic, the rationalist (and Kant™s
own pre-critical) notion of an ens realissimum became a mere form of
thought, or a component in the form of systematicity that determines its
own matter (inferences, judgments, concepts, ultimately the matter of
appearances). And finally, I have argued that relating this last form to
the notions of matter and form at work in Kant™s account of reflective
judgment in the First Introduction to the third Critique goes yet one
more step toward depriving the idea of the ens realissimum of any kind of
ontological status.
There remained the question: why is Kant so intent on asserting,
again and again, the necessity of the idea, the unavoidable illusion it
carries, and even the positive, regulative role it plays in cognition? My
suggestion is that none of this would be necessary unless Kant was intent
on maintaining its role for practical reason. The unity of theoretical and
practical reason is what drives the admissions of theoretical reason itself.
Whether the practical grounds for endorsing the idea of ens realissimum
are any stronger than the theoretical grounds, is a question I had no
ambition to answer in this chapter.


Kant says relatively little about moral judgment. He spends much more
time and care explaining and justifying the content of the moral law,
expounded in the different formulations of the categorical imperative:
˜˜Act only in accordance with that maxim through which you can at the
same time will that it become a universal law™™; ˜˜So act that you use
humanity, whether in your own person or in the person of any other,
always at the same time as an end, never merely as a means™™; ˜˜Act only so
that the will could regard itself as at the same time giving universal law
through its maxim.™™1
To be sure, these formulations of the categorical imperative are sup-
posed to function as principles or premises for inferences determining a
system of duties. In other words, they must serve as principles for two
kinds of moral judgments: (1) those by which we determine what we are
supposed to do or refrain from doing in virtue of a specifically moral
commitment, that is, a commitment independent of any consideration of
utility or of personal happiness; (2) those by which we subject to a moral
evaluation the actions already performed by ourselves or by others, and

Cf. Groundwork of the Metaphysics of Morals, trans. Mary Gregor (Cambridge: Cambridge
University Press, 1998), AAiv, pp. 421, 429, 434. As I did for all other works of Kant (except
for the Critique of Pure Reason), I give page references to Groundwork in the volumes of the
German Akademie Ausgabe (AA). Henceforth I will give those references directly in the
main text, citing volume and page numbers, e.g. (iv, 429).


the characters of those who performed them. So there is of course a
relation between moral law and moral judgment, the originality of
Kant™s position being precisely that he makes the former the uncon-
ditioned principle of the latter.
Nevertheless, it remains true that Kant is more informative about the
first than he is about the second. For instance, ch. 1 of the Critique of
Practical Reason is entirely devoted to the moral law and its formulation
in the categorical imperative. But ˜˜pure practical judgment,™™ that is,
moral judgment, is granted only two or three pages relegated to the
end of ch. 2, whose main topic is the ˜˜Concept of an Object of Pure
Practical Reason.™™2 And even there, Kant is mostly concerned with
explaining the fundamental difficulty we encounter in attempting to
think the relation between the moral law (which depends on the faculty
of reason alone, and thus on our belonging to a purely intelligible world)
and actions that unfold in the sensible world and are thus causally
necessitated. This metaphysical difficulty is according to Kant the root
of the difficulty of moral judgment, evaluating an action or the will of the
subject that performs that action (is it a good will or not?). For an
external event, given in space and time, does not by itself give us any
access to the internal motivation of the agent (did she act from respect
for the moral law, or on the contrary from egoistic interest?).
If this is so, should we not say that the question of moral judgment is,
by Kant™s own admission, the weak link in his moral philosophy? In
other words, even supposing Kant succeeded in his ambition to formu-
late ˜˜the supreme principle of morality™™ (iv, 392), did he not remain helpless
when it came to grounding on this principle the indisputable validity of
any moral judgment at all, whether determining (answering the ques-
tion: what should I do?) or reflecting (answering the question: ˜˜is this
action, and the will of this agent, morally good or evil?™™).3

Cf. Critique of Practical Reason, ch. 1, AAv, pp. 19“57; ch. 2, section ˜˜Of the Typic of Practical
Judgment,™™ AAv, pp. 67“71.
Here I am applying to moral judgment the distinction made by Kant in the Critique of the
Power of Judgment, between determining use of the power of judgment (where ˜˜the uni-
versal (the rule, the principle, the law) is given,™™ and ˜˜the power of judgment subsumes the
particular under it™™) and reflecting use (where ˜˜only the particular is given, for which the
universal is to be found™™): cf. AAv, p. 179. Kant does not make use of this distinction when
he speaks of moral judgment, but it seems illuminating to me in respectively characterizing
the (determining) application of the moral law in deciding to act, and the (reflecting)
evaluation of a given action, that is, the search for the rule under which it was performed.
Of course, ˜˜determining™™ and ˜˜reflecting™™ have a distinctively practical meaning here. For
˜˜determining™™ an action under the moral law is literally making it come about, not (as in the

Now to the suggestion that evaluating moral worth is, according to
Kant, a task that remains fundamentally opaque for us, one might
oppose Kant™s adamant insistence on the fact that moral judgment, in
contrast with theoretical judgment (which requires complex training
and is susceptible to error), is accessible to all. Its verdict is infallible, at
least for anyone who remains attentive to the voice of her conscience,
that is to say (as we shall see shortly) to the demands of pure practical
reason. Indeed Kant goes as far as to maintain that the verdicts of
common moral conscience are more worthy of trust than the subtle
distinctions and specious ratiocinations of academic moralists (iv, 404).
This being so, the sole merit of elucidating the fundamental principle of
morality (the categorical imperative in its various formulations) is to
reinforce common conscience in the moral judgments it was perfectly
capable of passing by itself, before their principle was made explicit. But
that merit itself, modest as it may be, shows that it is possible to elucidate


. 7
( 10)