Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
Can there be a Finite Interpretation of the Kantian Sublime?
Sacha Golob (King’s College London)
Abstract
Kant’s account of the sublime makes frequent appeals to infinity, appeals which have been
extensively criticised by commentators such as Budd and Crowther. This paper examines the
costs and benefits of reconstructing the account in finitist terms. On the one hand, drawing on
a detailed comparison of the first and third Critiques, I argue that the underlying logic of
Kant’s position is essentially finitist. I defend the approach against longstanding objections,
as well as addressing recent infinitist work by Moore and Smith. On the other hand, however,
I argue that finitism faces distinctive problems of its own: whilst the resultant theory is a
coherent and interesting one, it is unclear in what sense it remains an analysis of the sublime.
I illustrate the worry by juxtaposing the finitist reading with analytical cubism.
§1 – Introduction
Kant’s account of the sublime makes frequent reference to infinity. The “intuition” of the
sublime “carries with it the idea of...infinity”; apprehension “can progress to infinity”
[kann…ins Unendliche gehen]; imagination “strives to progress towards infinity” [ein
Bestreben zum Fortschritte ins Unendliche]; reason demands that we “think the infinite as a
whole” (KU 5:255, 252, 250, 254).1 It is obvious that the infinite played a central role in
Kant’s own presentation of the problem. It is less clear whether such references are
1
References are to the standard Akademie edition of Kant’s gesammelte Schriften (Berlin: de Gruyter, 1900–;
abbreviated as Ak.): Anth: Anthropologie in pragmatischer Hinsicht (Ak. 7); KrV: Kritik der reinen Vernunft
(Ak.3-4); KU: Kritik der Urteilskraft (Ak.5); Refl.: Reflexion (Ak.16); Relig.: Die Religion innerhalb der
Grenzen der blossen Vernunft (Ak.6); V-MP/Volckmann: Metaphysik Volckmann (Ak. 28); V-MP-L1/Pölitz:
Metaphysik L1 (Pölitz) (Ak.28). For KrV I use the standard A/B pagination. I follow Pluhar’s translation of KU
(Kant, 1987).
1
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
philosophically necessary, desirable or even coherent. This paper addresses that issue by
examining what one might call “finitism” about the Kantian sublime.
“Finitism” is the view that Kant’s theory can and should be interpreted so that it
makes no necessary reference to infinity in explicating the distinctive features of the sublime.
The key words here are “necessary” and “distinctive”. The finitist accepts that some
individuals may interpret sublime experience in such terms, as Kant himself obviously did.2
The finitist also accepts references to infinity insofar as they are justified independent of the
sublime. Suppose the claim that “the power of numbers progresses to infinity” is read as
referencing the infinite character of the natural numbers (KU 5:251). This could obviously be
defended quite independent of the sublime – and as such would be compatible with finitism.
Finitism is not intended as a straightforward piece of exegesis. Rather, the finitist
claims to rescue a defensible Kantian sublime from what he sees as poorly motivated appeals
to the infinite. For Crowther, for example, talk of infinity is a function of “the baroque
architectonics of Kant’s philosophical system…[which] actively weakens the philosophical
viability of his explanation of the sublime” (Crowther, 2010:181). The commentator’s task is
to prune away this excess, revealing a more persuasive “austere” theory (Crowther,
2010:180). Budd similarly finds it “hard to make plausible sense” of Kant’s remarks “even
within the terms of Kant’s own philosophy” (Budd, 1998: 238-9). Again, the commentator’s
task is to save Kant from himself (Budd, 1998:245). The finitist thus sees himself as engaged
in rational reconstruction along essentially Critical lines – but now shorn of the misguided
appeals to infinity.
In attempting to improve on Kant’s official doctrine, finitism has provoked some
strong reactions. Allison, a textbook infinitist, sees “the reference to infinity [as] clearly
essential to the Kantian account”, whilst Crowther’s view is “a complete failure” (Allison,
2001:397 n22). Crowther has responded by attacking an interpretative approach which
allegedly privileges “rigid internal consistency” in the face of “gross contradictions”
(Crowther, 2010:183, 194).
I have three aims in this paper. First, I argue that finitism is deeply rooted in Kant’s
position: when we look at the details of KU, particularly in the context of KrV, the logic of
Kant’s arguments are often finitist. Second, I argue that approaching the text in this way
illuminates key issues from the pyramids’ example to the role of mereology. Third, I show
how those very same commitments create a series of philosophical problems: I contrast my
2
For example, Crowther, 2010:173, 181n.
2
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
views here particularly with Crowther’s. In short, my goal is neither simply to defend finitism
nor to dismiss it. Rather, I want to show how integral it is to Kant’s approach and how
problematic the resultant theory of the sublime becomes.
Before getting underway, I must bracket off a handful of issues. First, sublime
artworks: Kant clearly recognises the possibility, but also sees it as secondary, not “fully
appropriate for the critique of aesthetic judgment” (KU 5:252). One problem is that the
artistic sublime might be subject to the requirements of both beauty and sublimity (KU
5:326). Abaci and Clewis have recently debated the textual data (Clewis, 2010; Abaci, 2010),
but from my point of view the real issue comes with later artistic developments. Consider, for
example, De Maria’s 1977 The Lightening Field. This stages a classic sublime experience,
using metal poles to channel lightning strikes in the New Mexico desert. De Maria insisted
that “the land is not a setting for the work but a part of the work” (De Maria, 1980:52).
Clearly, this problematises any simple divide between a ‘natural’ and an ‘artistic’ sublime.
Yet the conception of the art object is so different from Kant’s own that integrating it into the
discussion would require a separate paper. In what follows, therefore, I reference works of
art, but I remain neutral on whether the works themselves are sublime or whether the
experiences they generate simply provide a useful point of argumentative orientation.
Second, my main interests is in finitism, its roots and consequences. I will not attempt
to fully reconstruct the various chains of thought that led Kant to appeal to infinity. As one
would expect, some are relatively direct: KU 5:254 states that “reason makes us unavoidably
think of the infinite”. Others hang on what are disputed questions in their own right, such as
the moral role of the sublime. For Budd, for example, the “introduction of the idea of infinity
into Kant’s account…was designed to secure” this “moralized conception” (Budd, 1998:240).
For both he and Crowther, however, appeals to infinity are perhaps most closely bound up
with Kant’s account of measurement (Budd, 1998:240-1, 256; Crowther, 1989:97; Crowther,
2010:179). Whilst I look in detail at those sections, my primary aim is to articulate a version
of finitism, rather than to catalogue and explain each of Kant’s references to infinity.
Third, I focus on the mathematical sublime because it is there that finitism comes
centre stage: references to infinity in the dynamical sublime are rare, do little argumentative
work, and often simply point back to the mathematical case (KU 5:261).3 For stylistic
3
I agree with Merritt that there is overlap between the two classes: some instances of the mathematical sublime,
perhaps including the famous “starry heavens”, are also plausible instances of the dynamical sublime (Merritt,
2017:453).
3
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
reasons, I talk simply of the “sublime”, but unless explicitly stated otherwise I mean the
mathematical sublime.
§2 – The Existing Debate Between Finitism and Infinitism
I begin with a sketch of the existing debate. Let’s consider the common ground between
finitists and infinitists.
First, all parties agree that the experience of the sublime occurs insofar as the
imagination struggles to meet some demand of reason. Kant alternately presents this demand
as that of “comprehension in one intuition” [in eine Anschauung] and “comprehension of a
given object in a whole of intuition” [in ein Ganzes der Anschauung] (KU 5:254 – original
emphasis; KU 5:257).
Second, the imagination’s struggles occur because comprehension, specifically
“comprehensio aesthetica” (KU 5:251), is subject to a limit or “maximum” (KU 5:251-2).
Whilst “apprehension” occurs smoothly with even the largest objects, and can “proceed to
infinity”, comprehension encounters some kind of problem with “absolutely great” objects
(KU 5:251-2).
Third, the reasons for this are linked to the imagination’s role in estimating scale or
magnitude. There are two forms of such estimation: mathematically “by numbers” and
aesthetically “by eye” (KU 5:251). The latter has explanatory priority: the ability to count 10
feet is useless unless I have a demonstrative grip on the size of a foot (KU 5:251).4 Kant
refers to this demonstratively defined extent as a “basic measure” or “Grundmaß”: “hence
our estimation of the magnitude of the basic measure must consist merely in our being able to
take it in [fassen] directly in one intuition and to use it…for exhibiting numerical concepts
(KU 5:251).
Perhaps the simplest way to combine these three claims is something like this. Faced
with very large objects, such as the pyramids or St Peter’s, the imagination cannot properly
synthesize our experience of them. This is because we lose awareness of the parts first seen
or “apprehended” before we take in the rest:
[W]hen apprehension has reached the point where the partial presentations of sensible
intuition that were first apprehended are already beginning to be extinguished in the
imagination, as it proceeds to apprehend further ones, the imagination then loses as
4
As Allison neatly puts it, knowing that the Ark was so many cubits long is useless “unless one also happens to
know that a cubit is a unit of measure based roughly on the length of a human forearm” (Allison, 2001:317).
4
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
much on the one side as it gains on the other; and so, there is a maximum in
comprehension that it cannot exceed. (KU 5:252)
The attempt and subsequent failure to comprehend the object as a whole demonstrates both
imagination’s “inadequacy” and its “vocation” to obey reason’s drive to unity (KU 5:256). It
thus triggers the distinctive feeling of sensible shortfall and rational elevation which Kant
associates with the sublime (KU 5:256).
This is a crude, and as we will see problematic, summary. But before pressing further,
I need to introduce infinity: how does this enter the picture? One option is to appeal to
morality. But taken alone, this begs Budd’s complaint: the moral dimension so strongly
present for Kant is “highly likely to be absent from the experience of many, if not most, of
us” (Budd, 1998:246). Infinitism would thus become an artefact of Kant’s “tendency…to
moralize, in one way or another, any experience he valued” (Budd, 1998:246). If the infinitist
is to avoid this, she must find some bridging premise, something that links infinity more
directly to the sublime. There are four obvious options.
First, the reference to infinity might enter via the theory of measurement. This is how
finitists tend to read the text. Here is Crowther:
Kant is saying that in the case of vast formless objects reason demands that we
estimate their magnitude in relation to a unit of measure provided by a single
intuition. In order to satisfy this demand, the imagination will at first try out easily
comprehended measures such as a foot or a perch, but is then driven to find larger
units as a measure for them, and then still larger units as a measure for these, and so
on and so on, until it arrives at infinity itself as the only appropriate measure. But by
this time imagination is simply overwhelmed. (Crowther, 1989:97)
Here is Budd:
The larger the object, the larger the unit of measure required to estimate its magnitude
by sight. In other words, the aesthetic estimation of the magnitude of larger and larger
objects requires the comprehension of progressively larger units of measure. (Budd,
1998:238-9)
The suggestion is that larger objects lead to a ‘bidding up’ of the measure used to estimate
them – all the way to infinity. This is not a caricature of the infinitist position: for Allison too,
imagination “must assume increasingly larger measures ad infinitum in pursuit of its goal”
(Allison, 2001:315). But the view is implausible, as Budd and Crowther observe (Budd,
1998:256; Crowther, 1989:97, Crowther, 2010:179). Most obviously, why, when faced with a
very large but finite object, would we ever employ a basic measure larger than the object
5
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
itself? What would it even mean to use “infinity” as a “unit of measure”? How could this fit
with Kant’s insistence that we “take in [fassen]” the unit “directly in one intuition”?
Second, the reference to infinity might enter via apprehension. The difficulty is that,
as noted in §1, infinitism requires that infinity explicates the distinctive features of the
sublime. But the links between apprehension and infinity, which strongly suggest some form
of iteration or counting, are not of the right type. KU 5:254, for example, refers to a
“combination” [Zusammensetzung] “required to present a magnitude”, “guided by means of
numerical concepts”, and which “on its own progresses to infinity”. Kant equates this with
“arithmetic” and the natural gloss on it is in terms of the open-endedness of counting: his
point would simply be that the set of the natural numbers is infinite.5 But if this is correct, it
is entirely compatible with finitism since entirely independent of the sublime.
Third, it might be the unification demanded by reason that requires reference to
infinity: immediately after stressing the need for “comprehension in one intuition”, Kant
states that “reason makes us unavoidably think of the infinite…as given in its entirety” (KU
5:254). The problem, as Crowther notes, is that a vast finite totality could still easily defy
comprehension: by extension, there is no need to read reason’s demands as infinite to
demonstrate its superiority over the imagination. Hence his “austere” reading on which the
unity sought by reason is construed in “phenomenal” terms:
[W]hat reason demands is the comprehension of the phenomenal totality of any given
magnitude in a single whole of intuition…By ‘phenomenal totality’ here, I mean all
the different major parts or aspects which an object can present to perception –
including those which, due to its or our location, are not presently accessible.
(Crowther, 1989:101-2)
Sublime objects would be those where the quantity – and, as we will see, arrangement of
their parts – makes such synthesis impossible: as illustrated by the pyramids’ case. But the
totality would remain very much finite: the Great Wall of China is only finitely large and yet
well exceeds my capacity to grasp all its “major parts” in one intuition. One might, of course,
have separate grounds, independent of the sublime, for always thinking of reason as infinite.
But the outcome, precisely as the finitist charged, would be that talk of infinity was not
driven by analysis of the sublime itself, but by antecedent commitments, for example from
the moral sphere.
5
I refrain from introducing post-Cantorian complications surrounding types of infinity.
6
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
Fourth, it could nevertheless be that “from a subjective point of view, this object
seems infinite” even if it is, as on Crowther’s “phenomenal” account not actually so
(Matthews, 1996:173 – emphasis added). For Allison, similarly, we experience it “as if” it
were infinite (Allison, 2001:323). But this leaves Matthews and Allison vulnerable to a
collapse into finitism:
[A] phenomenon can surely be experienced as overwhelming per se without
appearing to be infinite. Indeed, if one person insists that the infinite must be
involved, then another, with equal validity, can claim that this is not the case, in so far
as they can find no introspective evidence that it figures in their own experience of
the sublime. (Crowther, 2010:181n.)
Once it is a question of subjective interpretation of an experience, the finitist is well set: she
does not deny that such thoughts may occur; she simply sees them as extraneous symptoms
of other Kantian commitments. Prima facie, the same applies to recent infinitist readings such
as Smith’s on which the link is even weaker: “it needn’t be maintained that the object
actually seems to be infinite (Matthews), but just that the object prompts or triggers an
experience of the imagination’s running on to infinity” (Smith, 2015:113).6
This survey is not, of course, exhaustive. But it provides a background against which I
can introduce my own account.
§3 – Apprehension: Consciousness and Mereology
I present my own account in three stages: in this section, I examine apprehension and in §§45 comprehension. I am ultimately going to argue that Kant offers a finitist theory which
tracks a very specific form of information overload. To understand that overload, we need to
see how KU’s account of apprehension and comprehension is rooted in KrV’s model of the
mind. The place to begin is therefore not with the third Critique, but the first.
In the A Deduction, Kant analyses the relationship between intuition and
apprehension:
Every intuition contains a manifold in itself, which however would not be represented
as such if the mind did not distinguish the time in the succession of impressions on
one another; for as contained in one moment no representation can ever be anything
other than absolute unity. (KrV, A99)
6
I return to Smith’s interesting reading in §4.
7
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
Every intuition “contains a manifold” because Kant holds that space and time are quanta
continua: i.e. “no part of them is the smallest” (KrV:A169/B211; A525/B533). In KU, he
makes the point via telescopes and microscopes: the scale of resolution can always be
magnified revealing new components, “all the way to the magnitude of a world” (KU 5:250).
These newly revealed parts are constitutive components of the original intuition: to see a man
at distance I must represent “his eyes, nose, mouth”, since the “representation of the whole
(of the head or of the human being) is composed of these partial ideas” (Anth. 7:135).
Nevertheless, such representations are “obscure”: I have no direct, non-inferential
consciousness of them (Anth. 7:135; 24:410).7
What marks the boundary between those parts of which I am directly conscious and
those of which I am not? As in the citation above, the answer is the synthesis of
apprehension: it is the mechanism by which the manifold becomes “represented as such”
(KrV:A99). Kant labels directly conscious states “perceptions” (KrV:A320/B376–7): thus
“by the synthesis of apprehension I understand the composition of the manifold of an
empirical intuition, through which perception, i.e. empirical consciousness of it (as
appearance), becomes possible” (KrV:B160). For Kant, this is necessarily a successive
process: “if the mind did not distinguish the time in the succession of impressions on one
another”, the result would be an “absolute unity” (KrV:A99). It occurs insofar as I divide the
manifold up according to some particular unit or measure. In the case of space and time, as
quanta continua, this measure is up to me:
Every quantum is either continuum, or discretum...A quantum continuum is not made
up of simple parts ...The multiplicity of parts of which it is made up is not determined
by its magnitude, and I can allow the multiplicity I wish [eine Menge, wie und soviel
ich will]. (V-MP/Volckmann 28:423)
For example, simply by focusing in or walking closer, I can bring more of the man’s face
over the threshold of consciousness. However, the scene, the object, or my other
commitments will often prompt or even demand the use of a particular unit: for example, if
something is conceptualized as an organism, I cannot assume its parts will be always be
further divisible (KrV:A526/B554). Concepts are thus “rules for apprehension” (KrV:A103–
4; Refl. 16:557), rules determining the conditions under which we become conscious of the
manifold.
7
A full analysis would require treatment of the multiple forms of consciousness and self-consciousness
recognised by Kant: for recent work on “obscurity” in particular, see Liang, 2017.
8
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
If we now return to KU, we can better understand the underlying framework Kant is
assuming. Imagine walking towards the pyramids from a distance. At the start, I can easily
see the whole structure, but the level of detail will be poor. In Kant’s terms, the basic
measure, what I can “take in directly in one intuition” or in “one glance” (KU 5:251, 254), is
very coarse-grained. This leaves the “apprehended parts (the stones on top of one another)
presented only obscurely” (KU 5:252), just as the distant man’s eyes and nose were
“obscure” in the Anthropology. My intuition of the pyramid at this point “contains a manifold
in itself which is not represented as such” (KrV:A99). As I come closer, I am able to
perceive, in both the modern and Kantian sense of that term, the parts, the stones, of this vast
object. In Kantian terms, I can apprehend them, as what I can “take in in one glance”,
becomes more fine-grained. Assuming I possess the relevant category of quantity, I can also
count these parts (KrV:B203).
In §1, I suggested that Kant’s references to apprehension “going on to infinity” were
unproblematic for the finitist. We can now better see why. They are motivated by two
assumptions: (i) that bodies are “divisible to infinity without, however, therefore consisting
of infinitely many parts” (KrV:A525/B553), and (ii) that the process of counting such parts
may go on indefinitely in line with the infinite nature of the natural numbers.8
Furthermore, two important, new points are also visible. First, Budd and Crowther
assume that infinity enters via what I called a “bidding up” process, in which I desperately
search for an ever-larger measure. They, rightfully, dismissed the resultant picture as absurd.
But we can now see the basic case, at least, for Kant runs in the opposite direction. We start
with an intuition, the man’s head, and progressively move to a more fine-grained measure,
such as his eyes. Second, we can see that whatever reason’s demand for “comprehension in
one intuition”, it must mean more than simply having the entire object in view: provided I
stand far enough back, I can easily do that. To establish what exactly Kant had in mind, it is
time to turn directly to his presentation of the pyramids’ case and of comprehension.
§4 – Comprehension I: The Pyramids, the Absolutely Great and the Absolute Measure
The heart of Kant’s account is comprehension, specifically “the comprehension of the many
[elements] in one intuition, to the limit of the imagination’s ability” (KU 5:254).
Comprehension is a form of imagistic synthesis: essentially, I fuse the successively
apprehended parts.
8
This analysis is not intended as an exhaustive treatment: I leave aside Kant’s brief remarks on the use of nonintuitive measures, for example “the degree of a certain virtue” (KU 5:249).
9
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
[C]omprehending a multiplicity in a unity (of intuition rather than of thought), and
hence comprehending in one instant what is apprehended successively, is a regression
that in turn cancels the condition of time in the imagination’s progression and makes
simultaneity intuitable. (KU 5:258–9)
Such comprehension “cancels the condition of time in the imagination’s progression and
makes simultaneity intuitable” because it generates a single, simultaneous image out of the
successively apprehended parts. This amounts to a “violence” which is nevertheless “still
judged purposive for the whole vocation of the mind” insofar as it generates an experience of
the sublime, making visible both sensibility’s inadequacy and reason’s demand for unity (KU
5:259). This section will address three connected aspects of Kant’s treatment of
comprehension: the pyramids/St Peter’s cases, the “absolutely great” and the “absolute
measure”.
First, we need to recognise that KU’s discussion of comprehension merges two very
different phenomena. One we might call the “extinguished parts case” or EPC. This occurs
when, in apprehending some object, the “partial presentations of sensible intuition that were
first apprehended are already beginning to be extinguished in the imagination as it proceeds
to apprehend further ones” (KU 5:252). In effect, the earlier parts lose their
phenomenological vivacity before I reach the object’s end. The other is what we might call
the “image limit case” or ILC. This occurs when, in trying to comprehend the object’s parts
in a single image, the complexity creates an awareness of the imagination as under intense
strain, operating at its upper limits
Now if a magnitude almost reaches the limit of our ability to comprehend [it] in one
intuition, but the imagination is still called upon to perform…an aesthetic
comprehension in a larger unity; then we feel in our mind that we are aesthetically
confined within bounds. (KU 5:259)
This confinement produces a “displeasure” which, in line with Kant’s overarching view of
the sublime, is nevertheless “subjectively purposive” (KU 5:260).
These two cases are mixed together in Kant’s presentation, but they are importantly
different. Smith comes the closest to recognizing this when he talks of a “failed
comprehension” and a “successful comprehension that nevertheless strains [imagination] to
the limit” (Smith, 2015:103). But we can press the point further. EPC is a problem of data
retention, whilst ILC is a problem of data combination. Suppose EPC were somehow fixed;
not only would that not fix ILC, it would likely make it worse since a greater quantity of
information would be available, increasingly the likelihood of overburdening the imagination
10
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
in synthesizing it. Phenomenologically, they differ too. The hallmark of EPC is a sense of
content slipping away; the hallmark of ILC is an excess of content that can barely be
mastered. Finally, they differ as to their triggers. Adding more content to an ILC will
eventually lead to the collapse of the phenomena by entirely overwhelming the imagination.
EPCs, in contrast, are more easily generated and have no upper bound: any number of parts
above a certain limit will do.
The distinction is vital if we are to understand the pyramids’ case. Allison has argued
that this is not a genuine instance of the sublime (Allison, 2001:317-8). This is echoed by
Doran for whom “Allison reads the Pyramids example, convincingly in my view, as an
instance of the ‘simply great’” (Doran, 2015:235). Allison’s worry is over Kant’s observation
that if one stands too close, the effect is lost (KU 5:252).
In virtue of what Kant had just said about comprehension, it might seem that he has
here gotten things precisely backward. For if the breakdown of comprehension is of
itself to be the source of the emotional effect of the perception of the pyramids, then it
would appear that the proper thing to do in order to attain such an effect is to stand as
closely as possible to them! (Allison, 2001:317-8)
What Allison misses is that the pyramids are intended to illustrate ILC: moving too close
might create a more overwhelming effect, but what Kant is trying to highlight here is a
specific experience in which “we feel in our mind that we are aesthetically confined within
bounds”. Merritt helpfully talks of a “sweet spot”: “we need to stand at just the point where
we will approach having a comprehensive view of the thing, but always just fail” (Merritt,
2012:39). Yet this is still not quite right: the sweet spot comes into play when we just barely
succeed as in the ILC. And when it does come to failure, as in the contrasting EPC, the
margin of it is unimportant: on Merritt’s reasoning, a truly gargantuan object that forces an
extreme EPC failure would not be sublime, whereas a far smaller one that only barely
exceeded our grasp would be.
If the pyramids’ case illustrates ILC, Kant’s other example, St Peter’s, is much less
fleshed out: it is natural, although not obligatory, to read it as an EPC. But at this point
Allison might well respond: “the fact that there are two such phenomena does not establish
that they are both sublime”. So why should we assume the pyramids are sublime for Kant?
One quick answer is that Kant’s description of the ILC includes the characteristic
reference to a painful struggle that is nevertheless purposive (KU 5:259). But it is important
to go deeper and doing so allows me to clarify the key concepts of the “absolute measure”
and the “absolutely great”.
11
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
I discussed the “basic measure” or “Grundmaß” in §3. It is vital to distinguish this
from the “absolute measure” or “absolutes Maß”. The absolute measure refers to the
underpinning of the ILC, what I called the “image limit”, i.e. the maximum amount of
content that the imagination can fuse into a single image. Thus an “aesthetic estimation of
magnitude” “exhibits absolute magnitude to the extent that the mind can take it in in one
intuition” (KU 5:251). Such formulations echo those of the basic measure, but the two
notions are very different. For example, the basic measure is, as noted in §3, highly variable:
as I move closer to the pyramids, the amount of conscious detail changes. In contrast, the
absolute measure is an upper limit, hard-wired by the mind’s structure. The following
illustrates their interaction:
[F]or the aesthetic estimation of magnitude there is indeed a maximum. And
regarding this latter maximum I say that when it is judged as [the] absolute measure
[absolutes Maß] beyond which no larger is subjectively possible (i.e. possible for the
judging subject), then it carries with it the idea of the sublime and gives rise to that
emotion which no mathematical estimation of magnitude by means of numbers can
produce (except to the extent that the basic aesthetic measure [Grundmaß] is at the
same time kept alive in the imagination). (KU 5:251)
It is only insofar as the “basic aesthetic measure” is “kept alive in the imagination”, i.e. not
extinguished as in EPC, that the imagination can test the boundaries of the “absolute
measure” by trying to fuse these “basic measures”. Kant’s tendency to blur the absolute and
basic measure has led to understandable confusion. For example, Budd writes that:
Kant’s idea of there being a maximum for the unit of measure for an aesthetic
estimation of magnitude is not the idea of there being, for each human being, some
size beyond which such a unit cannot be comprehended by the imagination in a single
intuition. Rather, it is the idea that, if, in the circumstances, a proposed unit of
measure takes too long to apprehend, it cannot on that occasion be used as a unit of
measure in an aesthetic estimation of magnitude. (Budd, 1998:237)
We can now see that both ideas are in fact in play at various points: the former is the
“absolute measure”, the latter is the “basic measure”, in this instance presented by Budd in an
EPC scenario.
This still leaves Allison’s question – why are cases, like the pyramids, that involve
pressure up to but not beyond the absolute measure genuinely sublime? The answer is that
Kant defines the sublime in terms of the “absolutely great” [schlechthin groß] (KU 5:248).
As has been noted, his presentation is a confusing one.
12
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
But suppose we call something not only large but large absolutely [schlechthin,
absolut], in every respect (beyond all comparison), i.e. sublime. Clearly, in that case,
we do not permit a standard adequate to it to be sought outside it, but only within it. It
is a magnitude that is equal only to itself. (KU 5:250)
Yet Kant immediately engages in just the sort of comparison he seemed here to forbid: “that
is sublime in comparison with which everything else is small” (KU 5:250). How should we
understand this?
The answer is that sublime objects are those the experience of which cannot serve as
the relata for an imagistic comparison. Hence, they are great “beyond all comparison”. Of
course, we can know that Everest has such and such a height, and compare that with the
height of other mountains. But we are unable to engage in “comprehensio aesthetica”, the
imagistic juxtaposition of the relevant objects. This is either because, in EPC, we were unable
to form an image of the sublime entity in the first place or, in ILC, because doing so takes us
right up to imagination’s boundaries, such that adding any further content, including the
comparative image, would exceed its capacities. The result is that sublime objects are those
which are beyond all comparison, and which “do not permit a standard adequate to it to be
sought outside it, but only within it” – i.e. terms of their distinctive effect on us. Such objects
are thus simultaneously beyond comparison and that “in comparison with which everything
else is small” insofar as only they trigger the requisite impact on the
apprehension/comprehension relationship and thus the distinctive feeling of scale (KU
5:250). In short, in response to Allison, the pyramids are sublime because the definition of
the “absolutely great” allows for both EPC and ILC instances.
Before proceeding, it is worthwhile contrasting my account with that of Smith, who is
acute in his criticisms of both Crowther and Matthews. Smith is an infinitist: “the experience
of infinity is necessarily connected with determining our response to the mathematical
sublime” (Smith, 2015:111). Two pages later, he takes himself to have substantiated that
claim (Smith, 2015:113). But the demonstration is unclear. He begins by noting that
imagination plays a central role for Kant in the generation of time (Smith, 2015:111). This is
absolutely correct: as I have argued elsewhere, the full story is that the act of iterating the
basic measure is successive because it is such action, or “motion of the subject”, that first
generates the pure intuition of time by affecting inner sense (Golob, 2016). The details of that
story are highly complex and not relevant here. What matters is the move Smith next makes,
in defending infinitism.
13
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
Because inner sense is the basis of intuition as such then it is not surprising that the
imagination in its synthesizing function would, if unchecked, proceed indefinitely.
(Smith, 2015:111)
But the existence of links between inner sense, imagination and the generation of intuition
does not show that imagination “would, if unchecked, proceed indefinitely”. Indeed, it is hard
to evaluate such a counterfactual since Kant places complex restrictions on imagination’s role
in time generation, subordinating it to apperception (KrV: B160-2, including the relevant
note). Smith next suggests that with the sublime, this counterfactual is realized: there the
imagination “has lost its restrictions in that it is no longer bound to a determinate magnitude”
(Smith, 2015:112). Suppose for the sake of argument, we grant this. After all, when an object
is “absolutely great”, it does lack “determinate magnitude” in a quite specific sense: we are
unable to achieve a stable image of it against which others might be compared. But that is not
the same as saying that imagination operates “unchecked” or that it is “indefinite in its
limitations” (Smith, 2015:111, 113). On the contrary, it is precisely because imagination is
nevertheless operating under limitations that the sublime experience is triggered. This is
easily seen with ILCs: as Kant himself notes, there we feel “aesthetically confined within
bounds” (KU 5:259). But it also applies to EPCs. We are continually achieving and losing
consciousness of the parts of objects around us: by the end of my walk to work, much of what
I saw at the start has simply faded away. What is distinctive about the sublime object is that
the quantity of content is such that it is nevertheless lost before I can complete the
apprehension of one and the same entity. This version of the sublime is thus produced not
because the imagination is “unchecked” or running onto infinity but because it is operating
within a very specific limit, albeit one it cannot meet: namely, to process the whole of some
specific object. There is an alternate way one might read Smith on which imagination is
“unchecked” just if the object it is processing cannot be taken up in a single intuition (Smith,
2015:113). This makes it true by definition to say that imagination has no such limits for
EPCs at least. But that simply moves the difficulty: his argument now fails when he states
that because it is unchecked the imagination “runs onwards towards infinity” (Smith,
2015:113). This merely asserts infinitism: what the finitism is arguing is precisely that an
experience may resist capture in a single intuition, and yet nevertheless not to require any
necessary reference to infinity.
It is striking that the cases recognised by Kant do not include the dominant postmodern one, what you might calls the “image breakdown case”, in which the imagination
retains such a richness of content, unlike an EPC, that it is not merely strained, unlike the
14
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
IPC, but actually defeated in trying to synthesise it. But the present task is to complete the
story by looking at the role of mereology in Kant’s model: I will then be able to explain the
precise form of overload or ‘being overwhelmed’ tracked by Kant’s theory.
§5 – Comprehension II: Mereology and Reason’s Demands
To do that we must understand exactly how “comprehension (comprehensio aesthetica)”
functions and how it underpins the two versions of the sublime discussed. I locate my reading
in relation to Crowther’s since it is the most influential version of finitism.
For Crowther, Kant’s “official” theory contains a vital lacuna which he aims to fill.
[I]s it that all vast phenomena are sublime? This would surely be counter-intuitive.
We encounter such phenomena in some (lack of) shape and form almost every day,
but it is, surely, only exceptionally that we find them sublime. Kant, in other words,
offers no criterion for distinguishing the merely overwhelming from the sublime.
(Crowther, 2010:179)
This is “a key difficulty which Kant does not address” (Crowther, 2010:184 – original
emphasis). Crowther therefore proposes to do so himself by expanding on his “austere”
reading, introduced in §2. In this the totality demanded by reason is a purely phenomenal
one, combining “all the different major parts or aspects which an object can present to
perception” (Crowther, 1989:101-2). Crowther’s early work presents matters as fairly
straightforward: the larger the object, the more parts, and so the harder it will be to achieve
the desired phenomenal totality (Crowther, 1989:102). But this would not separate the merely
vast from the sublime. His later work therefore posits a more sophisticated structure whereby
a sublime object’s form or situation prompts an awareness of its parts:
There must be some perceptual cue in the object which leads us to differentiate the
whole into its multitudinous parts rather than simply take them as given. It is
reasonable to suppose that when such differentiation occurs our awareness of the
phenomenon’s overwhelming size is energized, even though we may be perceiving it
from a great distance. (Crowther, 2010:185)
In the case of the pyramids, for example, the parts are made salient through their repetition:
The insistent repetitiveness of their configuration challenges us to comprehend this
accumulation in an imagined whole wherein all these elements are presented as
individual parts simultaneously. But, of course, this overwhelms our imaginative
capacity. (Crowther, 2010:185)
15
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
The same effect can be triggered when motion highlights an object’s components or when
“suggestive concealment” leads us “to imagine all the parts which might be hidden”
(Crowther, 2010:186). Either way, the result is the same: we are “gradually overwhelmed by
the sheer profusion of parts” (Crowther, 2010:180).
Whilst Crowther’s ingenious examples are not in Kant’s text, we can now see how
much of his basic framework is already there in KrV’s story about apprehension from §3.
This is a synthesis which precisely separates the manifold into parts. Similarly, as at the end
of §3, whatever reason’s demand for “comprehension in one intuition”, it must mean more
than simply having the entire object in view: provided I stand far enough back, I can easily
do that. Instead, as Crowther suggests, the key is mereology. Whether something is sublime
or not is a function of the apprehended parts which the imagination is seeking to comprehend,
and that in turn is a function of the “basic measure” used. St Peter’s seen as a dot on an image
of Rome is obviously not sublime, nor is it if I stand with my nose pressed against the front
wall. This is what I meant when I spoke of finitism as deeply rooted in Kant’s system. KrV’s
account of apprehension provides the perfect framework for a mereological story on which
the imagination is overwhelmed when the level of conscious detail, as determined by the
basic measure, and thus the number of parts becomes too great to reconcile into a single
image of the whole. This is entirely compatible with those mereologies being finite, just as a
finite weight can overwhelm our physical capacities.
We can now go further: such finitism can be defended against the classic objection to
it. As noted in §2, Crowther presents reason in purely phenomenological terms:
[W]hat reason demands is the comprehension of the phenomenal totality of any given
magnitude in a single whole of intuition…By ‘phenomenal totality’ here, I mean all
the different major parts or aspects which an object can present to perception including those which, due to its or our location, are not presently accessible.
(Crowther, 1989:101-2)
The standard objection is that this simply confuses reason and understanding.
Crowther’s version of reason’s demand suffers from a complete failure to distinguish
between the totality required by the understanding, that is ‘Allness’ [Allheit]…which
like all the categories, expresses a condition of the unity of apperception (or the
understanding), and the absolute totality or unconditioned required by reason.
(Allison, 2001:397)
Matthews makes a similar complaint:
16
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
I would argue that if the object has too many parts to hold in front of our mind, then
the way to see it as a whole is through an empirical concept of understanding. For
example, we might understand that the object is a house with fifteen rooms and that
they all go together. This does not appear to require an idea of reason. (Matthews,
1996:173)9
Allison adds that the type of totality demanded by Crowther is not only “not…a demand of
reason…it is also one that the imagination can easily meet” (Allison, 2001:397).
These responses are natural. The successive nature of apprehension means that Kant
needs an account of how we represent the simultaneous co-existence of an object’s parts.
What Allison and Matthews are picking up on is that, in KrV, this comes via the
understanding, most fundamentally in the 2nd and 3rd Analogies (for example, KrV:B257).
Yet this familiar line of attack misses the mark. Most obviously, imagination cannot
always “easily meet” the demand as we have seen in EPCs and ILCs. More fundamentally,
“comprehensio aesthetica”, the unity which imagination seeks and cannot deliver, is different
from that of the understanding. When I conceptualise something as a house, as in Matthews’
example, I subject my intuitions to a given “rule”: in modern terminology, I take on a set of
inferential commitments (KrV:A126; A106). But I do not necessarily have a phenomenal
awareness of that object’s parts: indeed, we typically employ concepts, in conversation for
example, without any distinctively imagistic experience at all. Kant’s account of the sublime
can therefore be construed on finitist terms provided we are willing to recognise a distinctive
drive towards imagistic comprehension, rather than merely inferential order: this separates
the relevant unity from that of the understanding. What the mathematical sublimes frustrates
is not our ability to get a rational grip on the world, but a sensible one.
What might this imagistic synthesis actually amount to? Here is Crowther’s
description of the imaginative unity supposedly characteristic of “comprehensio aesthetica”.
In the case of small objects, we can easily form an image which comprehends all their
various phenomenal aspects. For example, on seeing a tobacco tin, it is easy to
imagine all those parts of it which are hidden from our present viewpoint. (Crowther,
1989:101)
Leave aside the issue of size for the moment, and focus on “imagine”. It cannot mean
“imagine seeing all the parts from a single perspective” since, irrespective of the object’s
scale, this will typically be impossible. From what angle do I see all of the front, underside
9
See similarly Fenves, 1994:70.
17
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
and rear of a book? It is this impossibility which led Husserl to talk of a necessary and
“radical non-completion” [eine radikale Unvollständigkeit] inherent to perception (Husserl,
1973:51). So, we must read Crowther differently: when there are many parts, it is harder to
imagine them all in some more subtle sense. One might, for example, put the point in terms
of perceptual presence as Sellars did: I do not simply believe in the unseen parts of an object,
rather I imagine them.
Whilst [the interior whiteness and underside of an apple] are not seen, they are not
merely believed in. These features are present in the object of perception as
actualities. They are present by virtue of being imagined. (Sellars, 2007:458 – original
emphasis)
The more parts I am conscious of, the harder this imaginative presence is to sustain. This
identifies a genuinely imagistic unity, irreducible to the understanding and yet compatible
with Crowther’s finitist, phenomenal reading of “reason”. Hence finitism can be defended
against the classic Allison/Matthews objection.
§6 – From the Sublime to Cubism
I have argued that finitism is deeply embedded in the logic of Kant’s position, and that
positions such as Crowther’s can be defended from the classic objections to them. But I also
claimed that finitism was deeply problematic – and it is that issue to which I now want to
turn.
One set of problems is with the phenomenal unity just discussed: even if I am right
that it is not the unity of understanding, why should it have anything to do with reason? Why
can’t one talk solely of an imaginative unity?10 But I want to focus on another difficulty: it is
unclear that what finitism delivers is really a theory of the sublime. The term is, of course,
used very widely across many traditions. But confining ourselves to Kant’s framework, we
can identify several distinct components to his account: (A) a set of objects illustrating the
extension of the term and including the familiar tropes of the Longinian tradition – vast
edifices, high mountains, soaring eagles and so on (KU 5:252, 256, 315); (B) the account of
apprehension/comprehension; (C) the moral framework distinctive to Kant: talk of our
“sublime vocation” (KU 5:264) or the “sublimity of our moral vocation” (Relig. 6:50). As
has been widely discussed, Kant’s moralizing of the sublime creates a delicate balancing act:
10
See, for example, Kant’s account of Abbildung, Nachbildung, and Vorbildung in V-MP-L1/Pölitz 28:235.
18
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
the two must be intimately linked, without collapsing the former into the later. But my
present concern is not with the moral, but with the relationship between (A) and (B).
The root problem is this: what Kant’s account picks up on are two forms of
mereological information overload, triggered by the use of a basic measure that brings a large
number of the object’s parts over the threshold of consciousness (on the specific reading of
imaginative unity defended at the end of the last section, such consciousness would mix
distinctively perceptual and distinctively imaginary modalities). Such overload will at times
come from traditional cases of the sublime: standing at the mountain viewing platform, my
eyes scramble to take in the full scene. But the connection is very much a contingent one. The
(A) group is defined by a mix of scale and cultural associations of majesty and grandeur. Yet
what best predicts the Kantian sublime experience is the basic measure used by the viewer: if
this is sufficiently fine-grained then, given the quanta continua assumption, any object can be
parsed sufficiently to yield enough parts to trigger EPC or ILC. As Kant himself observes,
with the right equipment, a grain of rice can be “expanded for our imagination all the way to
the magnitude of a world (KU 5:251). Conversely, as we saw with the pyramids and St
Peter’s, vast objects will cease to have the requisite effect if I simply use an extremely
coarse-grained basic measure. Other cases, such as the soaring eagle, might be covered by
some other aspect of Kant’s theory, for example the dynamical sublime, but they have even
less connection with the apprehension/comprehension story.
The point can usefully be put in terms of resolution. Imagine a digital image file – it
may be a large file because it shows a wide expanse at very low resolution or a very small
expanse at a high resolution. What Kant’s theory picks up on effectively is the problem of
integrating large amounts of content, i.e. many large files. But this is compatible with their
being large in either sense just distinguished.
Crowther’s early work considers a version of this worry, where the viewer is shrunk
and finds microscopic structures sublime as they now tower over him (Crowther 1989:118).
But, as we saw, Crowther takes it for granted both that sublime objects must be larger than
the human body and that, barring such science fiction scenarios, comprehension is
unproblematic in the case of small objects.
In the case of small objects, we can easily form an image which comprehends all their
various phenomenal aspects….it is easy to imagine all those parts of it which are
hidden from our present viewpoint. (Crowther, 1989:112)
Yet this is not the case. What matters is the image resolution, not the object size: fusing a
huge number of incredibly fine-grained images into a single overview will test the
19
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
imagination even if the object is tiny. To insert a restriction such that sublime objects must be
over a certain size or larger than my own body would be purely ad hoc.
More broadly, there are plenty of putatively sublime experiences which do not suit
Kant’s model. On the one hand, one has ‘false negatives’. Consider a piece of light art in
which the viewer is suddenly immersed in an intense, all-encompassing, white light. This is
non-successive and there are no obvious ‘parts’ involved – but it is a natural candidate for the
sublime.11 A similar point could be made with sound art, which problematizes Kant’s focus
on images, and their successive apprehension and reassembly. On the other, one has ‘false
positives’. Take something like Grosz’s The Funeral. One could easily see how the viewer
might be “gradually overwhelmed by the sheer profusion of parts”, as Crowther put it
(Crowther, 2010:180). Grotesque, disordered, it also jibes with Kant’s talk of the horrible”,
even the “chaotic” (KU 5:245). Yet those features do not cohere well with the rest of Kant’s
theory: clearly, horrible chaos is neither necessary (the eagle, the moral law) nor sufficient
(the sewer pipe bursting on a crowded train) for sublimity, and it is hard to see Grosz as an
example of the sublime.12
These examples illustrate prima facie divergences between Kant’s theory and the
traditional set of sublime objects. Alone, however, this is relatively unthreatening – why must
Kant bow to vague feelings about the term’s extension? But what makes mereological
information overload important is that it is the central case tracked by Kant’s apparatus.
Crowther is correct that the merely overwhelming is obviously not sublime: the sound of a
traffic accident, a noxious smell, the stress of work, all of these might be overwhelming.
Kant’s theory tracks a particular form of the overwhelming, that triggered by a certain
relation between parts and whole. What we can now see is that this form has no direct link to
the sublime
These problems cannot be solved simply by appeal to the doctrine of subreption.
Under this Kant holds that sensible objects are never truly sublime: the predicate should
properly be applied only to our own states of mind (KU 5:256). But the issue is whether,
given that principle, any class of sensible entities has special status, in virtue of its ability to
trigger such experience. After all, there is clearly some sense in which Kant regards
11
I use art examples, subject to the disclaimer given at the end of §1. The work of James Turrell is the kind of
thing I have in mind, but I have simplified the case to avoid complexities created by the interaction of space,
particularly depth perception, and light in pieces such as Turrell’s Event Horizon.
12
Similarly, large scenes that are ‘verminous’, with an overwhelming number of crawling, shifting parts will
meet Crowther’s definition, even leaving aside the issue of scale. One response would be to argue that
something like this would necessarily constitute a threat and so trigger the dynamical sublime: in this way the
latter might backstop the mathematical sublime.
20
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
mountains as more closely connected to the sublime than pugs: thus, he talks of the “use
judgement makes naturally of certain objects so as to [arouse] this (feeling)” (KU 5:250 –
emphasis added). The problem is that when we press on that special status and on the
structure of Kant’s theory, we find that the “trigger entities” actually diverge sharply from
those Kant lists and from those on which the sublime tradition is standardly based.13
I presented finitism as a reconstructive project, and independent of this result, there
are undoubtedly passages in Kant’s text which call for a more inflationary approach: for
example, when he states that the “proper unchangeable basic measure of nature is the
absolute whole of nature” (KU 5:255). The finitist has never denied this; his view is that
these are architectonic flourishes, rather than genuine explanations of the sublime. The
problem we now see is that the apparatus underpinning finitism does not match up well with
that phenomenon either.
One final example makes the point vividly. Recall, this from Kant on the relationship
between apprehension and comprehension.
[C]omprehending a multiplicity in a unity (of intuition rather than of thought), and
hence comprehending in one instant what is apprehended successively, is a regression
that in turn cancels the condition of time in the imagination's progression and makes
simultaneity intuitable. (KU 5:258–9)
Such comprehension is “violent” because it makes present in a single image a series of
perceptions successively apprehended (KU 5:258–9). There is a style which sought precisely
to explore this phenomenon. But it is not the sublime. Rather, it is analytical cubism as
theorized both by artists such as Jean Metzinger and critics such as Simeon Giedion. Here is
the latter in 1938:
The cubists did not seek to reproduce the appearance of objects from one vantage
point; they went round them…Cubism breaks with Renaissance perspective…And in
so dissecting objects it sees them simultaneously from all sides – from above and
below, from inside and outside....The presentation of objects from several points of
view introduces a principle which is intimately bound up with modern life –
simultaneity. (Giedion, 1959: 432)
13
Moore’s theory comes closest to simply dispensing with the privileged class, treating all natural objects on a
par, and focusing entirely on the ideas of reason (Moore, 2018:365-7). Aside from the fact that Kant’s argument
surely does structurally privilege a given class of natural entities, albeit not the ones you might expect, it is hard
to see how a theory which made this move would remain a theory of the sublime in any substantial sense (as
opposed to an internal Kantian story about the significance of reason, say – Moore himself may be willing to
accept that as necessary for a genuine respect for the subreption doctrine).
21
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
We recognise this cubist ideal in Crowther’s phenomenal totality: all the major parts of the
object are successively seen and then simultaneously presented in one image, at the price of
classical perspective as noted by Husserl at the end of §5. In short, the best exemplar for the
underlying logic of Kant’s theory is not a vast mountain or a huge stone façade, but cubism’s
synthesis of the visual slices of everyday objects – a cup, a violin, a house.
At this juncture, one can imagine the infinitist re-entering the debate: doesn’t the very
fact that finitism fails to track the sublime show that we must appeal to infinity after all? But
I hope to have shown that the situation is much more complex than that. Kant’s apparatus,
from apprehension to comprehension, meshes perfectly with finitism and with what I called
“mereological information overload”. The root cause is that he has ported his existing theory
of perception onto the case of the mathematical sublime: this generates the characteristically
deep ties between that theory and his philosophy of mind. While it is undeniable that
something is missing from finitism, it is less clear where and how references to infinity
would enter the argument, or what explanatory role they would play. As seen in §2, one
cannot retain the same basic logic but simply append ad hoc references to the infinite, and
appeals to infinity are obviously not going to fix the problem just noted, a failure to
distinguish between finite but massive and finite but tiny entities. As stated in the
introduction, we need to recognise both how integral finitism is to Kant’s approach and how
problematic the resultant theory of the sublime becomes.14
Bibliography
Abaci, U. (2010): Artistic Sublime Revisited: Reply to Robert Clewis: Journal of Aesthetics
and Art Criticism, 68, 170-173.
Allison, H. (2001): Kant’s Theory of Taste. Cambridge: Cambridge University Press.
Budd, M. (1998): Delight in the Natural World: Kant on the Aesthetic Appreciation of Nature
Part Iii: The Sublime in Nature: British Journal of Aesthetics, 38, 233-250.
Clewis, R. (2010): A Case for Kantian Artistic Sublimity: Journal of Aesthetics and Art
Criticism, 68, 167-170.
Crowther, P. (1989): The Kantian Sublime: From Morality to Art. Oxford: Clarendon Press.
———. (2010): The Kantian Aesthetic: From Knowledge to the Avant-Garde. Oxford:
Oxford University Press.
De Maria, W. (1980): Some Facts, Notes, Data, Information, Statistics, Statements: Artforum,
18, 52-59.
14
My thanks to John Callanan, Jess Leech, and an anonymous referee for their extremely helpful comments on
an earlier draft – constraints of space unfortunately prevented me from taking up all of their suggestions, but I
hope to do so in subsequent work.
22
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
Doran, R. (2015): The Theory of the Sublime From Longinus to Kant. Cambridge: Cambridge
University Press.
Fenves, P. (1994): Taking Stock of the Kantian Sublime: Eighteenth-Century Studies, 28, 6582.
Giedion, S. (1959): Space, Time and Architecture. Cambridge, MA: Harvard University
Press.
Golob, S. (2016): Kant as Both Conceptualist and Nonconceptualist: Kantian Review, 21,
367-391.
Husserl, E. (1973): Ding Und Raum, Vorlesungen 1907. The Hague: Nijhoff.
Kant, I. (1987): Critique of Judgment (W. S. Pluhar, Trans.). Indianapolis: Hackett.
Liang, Y. (2017): Kant on Consciousness, Obscure Representations and Cognitive
Availability: Philosophical Forum, 48, 345-368.
Matthews, P. (1996): Kant’s Sublime: A Form of Pure Aesthetic Reflective Judgment:
Journal of Aesthetics and Art Criticism, 54, 165-180.
Merritt, M. (2012): The Moral Source of the Kantian Sublime. In T. Costelloe (Ed.), The
Sublime: From Antiquity to the Present (pp. 37-49). Cambridge: Cambridge University
Press.
———. (2017): Sublimity and Joy: Kant on the Aesthetic Constitution of Virtue. In M.
Altman (Ed.), The Palgrave Kant Handbook (pp. 447-470).
Moore, T. (2018): Kant’s Deduction of the Sublime: Kantian Review, 23, 349-372.
Sellars, W. (2007): The Role of Imagination in Kant’s Theory of Experience. In In the Space
of Reasons (pp. 454-469). London: Harvard University Press.
Smith, S. (2015): Kant’s Mathematical Sublime and the Role of the Infinite: Reply to
Crowther: Kantian Review, 20, 199-120.
23
Forthcoming in The Kant Yearbook, Vol. 11 (2019)
Final Draft – Please cite the published version for correct pagination
24