Jump to ContentJump to Main Navigation
The Kantian SublimeFrom Morality to Art$

Paul Crowther

Print publication date: 1991

Print ISBN-13: 9780198239314

Published to Oxford Scholarship Online: November 2003

DOI: 10.1093/0198239319.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 17 January 2019

The Background to Kant's Mature Theory of the Sublime

The Background to Kant's Mature Theory of the Sublime

(p.7) 1 The Background to Kant's Mature Theory of the Sublime
The Kantian Sublime

Paul Crowther (Contributor Webpage)

Oxford University Press

Abstract and Keywords

Introduces major theories of the sublime preceding Kant's. These feature Longinus, Dennis, Addison, and Burke. Kant's pre‐critical theory of the sublime is outlined.

Keywords:   Addison, Burke, Dennis, Kant, Longinus, the sublime


The sublime is an aesthetic concept of ancient lineage. In the eighteenth century, however, it underwent a major revival, and received especially important reformulations in the works of Joseph Addison and Edmund Burke. I shall now very briefly outline the salient aspects of these two theories in turn. First, through his conception of the ‘great’ (set out in his famous Spectator articles on ‘The Pleasures of the Imagination’) Addison is able clearly to separate those experiences which we would now commonly regard as ‘sublime’ from those which we would describe as ‘beautiful’. He also gives an important description of what is positive in the experience of the sublime. Consider, for example, the following passage.

Our imagination loves to be filled by an object, or to grasp at anything that is too big for its capacity. We are flung into a pleasing astonishment at such unbounded views, and feel a delightful stillness and amazement in the soul at the apprehension of them. The mind of man naturally hates everything that looks like a restraint upon it, and is apt to fancy itself under a sort of confinement, when the sight is pent up in a narrow compass, and shortened on every side by the neighbourhood of walls or mountains. On the contrary, a spacious horizon is an image of liberty, where the eye has room to expatiate at large on the immensity of its views.1

On these terms, Addison's major claim is that our encounter with vast natural phenomena involves a sense of being liberated from perceptual confinement. We experience an exhilarating feeling of self‐transcendence.

Edmund Burke's theory of the sublime, however, takes us in a (p.8) somewhat different direction. In his A Philosophical Inquiry into the Origins of Our Ideas of the Sublime and the Beautiful (1757), his fundamental claim is that anything which can occasion pain or terror or some kindred passion is a potential source of the feeling of the sublime. This feeling can be occasioned in two different ways. First, when the perceptually overwhelming properties of objects test and strain our perceptual faculties so as to cause a weak state of pre‐conscious pain; and second, where dangerous objects are encountered from a position of safety, thus causing a weak or moderated state of terror. In such cases as these, Burke informs us that

if the pain or terror are so modified as not to be actually noxious; if the pain is not carried to violence, and the terror is not conversant about the present destruction of the person, as these emotions clear the parts of a troublesome encumberance, they are capable of producing delight; not pleasure, but a sort of delightful horror; a sort of tranquility tinged with terror.2

For Burke, our delight in the sublime is due to the fact that the weak or moderated states of pain or terror which sublime objects arouse are ones which cause a healthy invigoration of those finer bodily tissues upon which the mental powers act. The feeling of the sublime is a feeling which is deeply bound up with our instinct for self‐preservation.

With this background material in mind, we can now address Kant's early theory of the sublime.


Kant's first attempt to articulate a theory of the sublime is found in his pre‐Critical Observations on the Feeling of the Beautiful and the Sublime, published in 1764. In the years immediately preceding this work, Kant had been much concerned with the relationship between feeling and morality, and Werkmeister is probably right in asserting that it was his indecision on this score which led him ‘to take a closer look at the nature of feelings’.3

(p.9) Kant's method in the Observations is primarily descriptive and is conducted, as he puts it, ‘more with the eye of an observer than of a philosopher’.4 Nevertheless, the work does present some points of philosophical substance (especially in Chapter 1), and it is to an exposition of these that I now turn.

We are first informed that ‘The various feelings of enjoyment or of displeasure rest not so much upon the nature of the external things that arouse them as upon each person's own disposition to be moved by these to pleasure and pain.’5 It is this fact which explains why it is possible for one person to derive joy from, and another person to feel aversion towards, one and the same thing. Now it is important to note that while (in the above quotation) Kant uses the term ‘feeling’ to pick out particular instances of pleasure and displeasure, he subsequently—and much more frequently—uses it to denote the subjective disposition which enables us to find things pleasurable or displeasurable. Thus Kant considers (amongst others) the stout man who likes a coarse joke, the merchant who enjoys calculating his profits, and the man who finds the opposite sex a mere object of pleasure: ‘all these have a feeling that makes them capable [my emphasis] of enjoying pleasures after their own fashion’.6 Kant clearly regards this capacity for feeling as somewhat crude, inasmuch as it can be employed ‘without any thought whatsoever’. There is, however,

another feeling of a more delicate sort, so described either because one can enjoy it longer without satiation and exhaustion; or because it presupposes a sensitivity of the soul, so to speak, which makes the soul fitted for virtuous impulses; or because it indicates talents and intellectual excellences.7

It is as modes of such ‘finer feeling’ that we experience the sublime and the beautiful. Again Kant plies us with examples. Mountains with peaks above the clouds, descriptions of raging storms, Milton's portrayal of hell, all arouse ‘enjoyment but with horror’; whereas flower‐strewn meadows, valleys with winding brooks, or descriptions of Elysium occasion a ‘joyous and smiling’ (p.10) sensation. Now, ‘In order that the former impression [i.e. enjoyment with horror] could occur to us in due strength, we must have a feeling of the sublime, and, in order to enjoy the latter [i.e. the joyous, smiling, sensation] well, a feeling of the beautiful.’8

On these terms, in order to enjoy the appropriate states of pleasure, we must have a disposition for the sublime and the beautiful. In the case of the sublime, Kant goes on to suggest that our disposition towards it is made occurrent in three characteristic ways. As he puts it,

Its feeling is sometimes accompanied with a certain dread, or melancholy; in some cases merely with quiet wonder; and in still others with a beauty completely pervading a sublime plan. The first I shall call the terrifying sublime, the second the noble, and the third the splendid.9

As examples of these, Kant cites (amongst others) great depths as terrifying, great heights as noble, and great buildings (such as the Pyramids) as splendid.

Rather than consider Kant's many other examples of things which engage our feelings of the sublime and beautiful, I shall concentrate on just one of these—discussed in Chapter 2 of the Observations. Kant suggests that moral qualities such as sympathy, friendliness, and honour are ‘amiable and beautiful’ because they can ‘harmonise’ with (i.e. facilitate) virtue. But while such qualities may help dispose us towards virtue, they are by no means constitutive of it. For example, sympathy for a needy person's plight may lead us into an act of charity that makes us forgo the repayment of a debt. In this case, Kant sees our ‘higher obligation’ as ‘sacrificed’ to mere good‐natured concern. However,

when universal affection towards the human species has become a principle within you to which you always subordinate your actions, then love towards the needy one remains; but now from a higher standpoint, it has been placed in its true relation to your total duty. Universal affection is a ground of your interest in his plight, but also of the justice by whose rule you must forbear this action.10

(p.11) Thus for Kant, while moral virtue presupposes a feeling of affection for humanity, this feeling only takes on its distinctively moral character when it issues in impartial principles of conduct, rather than ad hoc sympathetic responses. It is, indeed, this very subduing of immediate impulse through principle which Kant finds sublime. As he puts it, ‘as soon as this feeling of affection for humanity has arisen to its proper universality it has become sublime, but also colder.’11 I shall consider the reason why Kant finds this universality ‘sublime’ a little further on.

This brings us to the end of Kant's main philosophical points. The remainder of the Observations is for the most part concerned with the application of these points to questions such as the nature of human character‐types, the relation between the sexes, and the characters of different nations. I shall not concern myself with these questions, but will instead go on to discuss some of the issues raised by Kant's philosophical points. First, it is possible that Kant was at least indirectly familiar with Burke's theory through an extended review of the Inquiry written by Mendelssohn in 1758.12 However, the differences between the two theories are quite striking. This is, I would suggest, not due simply to Burke taking a systematic, and Kant taking an informal, approach; rather it hinges upon important philosophical differences. Burke, for instance, holds that the sublime is essentially a passion of modified terror or pain and pertains, thereby, to the instinct for self‐preservation. While he takes the subjective aspect of this passion as his starting‐point, he ultimately construes it as the causal effect of quite specific properties in objects. Kant, in contrast, asserts his independence from this view in the very first sentence of the Observations. It is not the properties of objects so much as the subjective capacity for feeling which determines the nature of our pleasurable or displeasurable responses. This allows him to find a greater diversity both in the employment of our feeling of the sublime, and in the objects and situations which can occasion it. For example, our response can embody ‘quiet wonder’ and the exhilaration of the ‘splendid’ as well as delightful horror. Similarly, it can be occasioned by such things as Friendship, (p.12) Understanding, and Virtue, as well as the more obvious ‘sublime’ phenomena such as raging storms, and the like. Indeed, for Kant ‘sublime’ can be a predicate ascribed not simply to the effects akin to terror of other persons and things upon us, but to actions undertaken by ourselves, out of, say, friendship or virtue—and which involve no risk to our physical well‐being. This marks a crucial, and (from the point of view of Kant's mature theory) decisive, advance upon Burke.

In marking out these differences between Burke, and Kant's early theory, I am, in effect, denying Mary Mothersill's recent claim that the latter ‘draws very heavily’13 upon the former. However, John Boulton has suggested that ‘certain isolated observations and phrases, point unmistakably to Burke's influence’14 upon the Observations. This idea of more piecemeal influence is more difficult to deny. It is, nevertheless, interesting that most of the examples with which Boulton illustrates his claim (such as high mountains, raging storms, Milton's portrayal of Hell, and infinity) are also to be found in a modified form in Addison. The influence of Addison upon Kant15 has never been adequately studied, but, as I have indicated in some of the notes to this section, many of Kant's ideas in the Observations are also to be found in Addison's Spectator essays on the Pleasures of Imagination. Given also the fact that Kant mentions the Spectator by name, late on in the Observations, I would suggest that the piecemeal affinities between Kant and Burke's texts are probably due to common source material.

Let me now consider a second major issue raised by Kant's treatment of the sublime. For a long time in the English‐speaking world, our view of Kant's pre‐Critical work has been strongly influenced by P. A. Schillp's Kant's Pre‐Critical Ethics. In this work, Schillp holds fundamentally that Kant was not a moral sense advocate in the manner, say, of Hutcheson. The following passage referring to the Observations sums up Schillp's position well.

nowhere in all these references to the ‘moral feeling’ or ‘feeling for morality’ . . . does there seem to be any reason for assuming that Kant (p.13) meant by the term a definitely independent ‘sense’ or separate ‘instinct’ such as the British moralists had in mind . . . Indeed there is a remark in the Beobachtungen which furnishes no small piece of evidence to the contrary. ‘Feeling is in no sense all of a piece’ . . . is an assertion of the variety of feelings and, as such, is inconsistent with the British moralists' doctrine of a distinct and unique moral sense.16

The most, indeed, that Schillp is prepared to admit is that ‘Kant agrees with his British and French tutors upon the importance for the moral life of feeling and of the emotional elements in general’.17 In relation to Kant's remark about feeling not being ‘all of a piece’ (leaving aside the fact that Schillp quotes it out of context), there is nothing in it which would count against the possibility of there being a distinct moral sense. A theorist such as Hutcheson does not hold that all feeling is moral, but rather that there are several varieties of feeling (or internal ‘senses’) of which the moral is one. If Schillp had directly compared Kant and Hutcheson's views he would have actually found them to be almost identical. For example (as we have seen), when Kant talks about the feeling of the sublime, and feeling for morality, he uses the term ‘feeling’ to denote, not our particular state of quiet wonder, or affection for humanity (or whatever), but rather our disposition to experience such states. Now consider the following passage where Hutcheson sets out his notions of the ‘sense of beauty’ and the ‘moral sense’:

These determinations to be pleased with any forms, or ideas which occur to our observation, are what the author [i.e. Hutcheson himself] chooses to call senses, distinguishing them from the powers which commonly go by that name, by calling our power of perceiving the beauty of regularity, order, harmony, an internal sense. And the determination to be pleased with the contemplation of those affections, actions, or characters of rational agents which we call virtuous is what he marks by the name of a moral sense.18

Hutcheson's definition of an internal sense as ‘a determination to be pleased with’ is what in modern philosophical parlance would (p.14) be called a disposition, and could (with only slight grammatical modification) be fitted into almost all those contexts where Kant talks of ‘a feeling of’, without any change of meaning taking place.19 I am suggesting, therefore, that Kant's treatment of the sublime and the beautiful as feelings in the Observations parallels, and is probably influenced by, Hutcheson's notion of an ‘internal sense’. Surprisingly, the one point which Schillp does see Kant as having in common with Hutcheson et al. (i.e. ‘the importance for the moral life of feeling and of the emotional life in general’) is one which is in fact somewhat at odds with the new direction which Kant's thought is beginning to take in the Observations. For, as we have seen, while Kant sees a feeling of affection for humanity as a presupposition of virtue, he construes ‘true virtue’ as a special employment of this feeling—namely acting in accordance with universal principles irrespective of our spontaneous emotional impulses.

To conclude this section, I shall now briefly and critically relate Kant's theory to that twofold root of the sublime outlined in Section I of this chapter. First, it is clear that Kant cannot be straightforwardly termed a self‐preservation theorist. For while, with Burke, he does allow a state of modified terror to be one form of our feeling of the sublime, he does not advocate that causal theory which enables Burke to construe modified terror and pain and (thereby) a link with self‐preservation as the sublime's definitive feature. Whatever defects Burke's causal theory may have, it does at least offer an explanation of what things give rise to the sublime feeling and why they do so. Kant, however, offers no such explicit account, and thence leaves a number of important questions unanswered. Why does our feeling of the sublime dispose us to be moved by such disparate things as precipices, virtue, and pyramids? Why are such disparate emotional states as enjoyment with horror, quiet wonder, and a sense of the splendid to be taken as occurrent instances of the same feeling? These two questions are not (p.15) unrelated, in that (from a modern philosophical perspective) to answer the latter, we must, logically speaking, be able to answer the former. The beginnings of such an answer are present in Kant's text, in so far as all the objects and phenomena which he sees as giving rise to the sublime are ones which exercise powers of non‐coercive physical and/or intellectual authority over us. The deep precipice has the potential to destroy us; true virtue asserts itself as an obligation and duty transcending personal inclination; the pyramid manifests the superiority of collective physical endeavour, and creative genius, over the mundane individual's talents. Kant, in other words, implicitly construes the sublime as occasioned by powers which transcend the self, in some specifiable way. What unites these phenomenologically disparate states of enjoyment with horror, quiet wonder, and a sense of the splendid is that they constitute a mode of reverence. Yet this invites the question of why our reverence for that which has authority over us should be in any sense a source of pleasure—rather than of mere reverence as such. The case of virtue which transcends personal inclination provides an important clue. Here we have something which, in its orientation towards the universal, exceeds our normal sensuous mode of being, but which is not—as a precipice or pyramid is—something external to us. As I shall show (in Chapter 2) of this study), it is by grounding the sublime in such self‐transcendence from the sensuous level of our being to the universal that Kant arrives at the basis of his mature theory.20 Before we can address this question, however, we must finally note some of the important developments which characterize the ‘Critical’ phase of his philosophy.


Kant's mature ‘Critical’ philosophy is first articulated in the Critique of Pure Reason (1781, revised 1787). In this work, Kant (p.16) argues that space and time are not properties of things in themselves, but are ‘forms of intuition’, that is features inherent in the general constitution of sentient creatures such as ourselves. On these terms, therefore, in so far as space and time are contributed by ourselves, we know a priori that all ‘intuitions’ (i.e. particular sensible existents) will appear to us as given in space and time. But for Kant cognition is not simply a receptive process. Any manifold of sensible intuition given in space and time is also actively structured by the faculty of understanding with its concepts. Kant's major arguments to this effect centre on the Transcendental Deduction in the Critique of Pure Reason. The gist of Kant's position here is that, in the most general terms, an object is a manifold of sensible intuitions which has been combined and unified in terms of some specific ‘rule’ (i.e. concept). Now such combination or ‘synthesis’ can itself only be carried out by a unified consciousness, that is one which can be (at least sometimes) aware of itself as distinct from that which it is aware of. But, reciprocally, such a consciousness can only be unified in these terms in so far as it synthesizes the manifold of intuitions on the basis of those fundamental rules or ‘pure concepts of the understanding’ which Kant calls ‘categories’. The depth of this reciprocal dependence is best illustrated through the category of causality. In the Second Analogy in the Critique of Pure Reason, for example, Kant argues that, if we are to distinguish statements about our own ‘representations’ (i.e. our present awareness of sensible objects, or our acts of recollecting or imagining or thinking of such objects) from statements about the objects themselves, then we must say that (as A. C. Ewing succinctly puts it), ‘the latter sort of statements gives an account of what any other normal observer would see under given conditions, and this is to assert a causal law governing human perceptions.’21

On these terms, the category of causality and the laws of nature which embody it are fundamental in enabling us to draw a distinction between an objective order of events and the subjective experience of these.

It must be noted, however, that, while we can have objective knowledge of the phenomenal world, we cannot have such (p.17) knowledge of that ‘noumenal’ or ‘supersensible’ substratum which provides the material from which the understanding's categories and the forms of intuition constitute phenomenal experience. That there is such a realm follows from Kant's overall philosophical position, but since, by definition, it transcends the bounds of what can be experienced it cannot form an object of knowledge for theoretical reason. This division between the phenomenal and noumenal also extends to the human self. For, on the one hand, we are embodied creatures of feeling and sensibility, who think and act in time and space. This means that as phenomenal beings we are a part of nature and are subject to determination by nature's causal laws. On the other hand, in so far as it is the human subject which imposes this causal network through the categories of understanding and the forms of intuition, the ultimate ground of the self must in some sense be presumed to lie beyond the phenomenal world. It must, in other words, be a noumenal or supersensible self. Now this area of Kant's philosophy is, unfortunately, deeply problematic, inasmuch as the relations between the phenomenal self, the formal unity of consciousness, and the supersensible self are never adequately clarified by him. Indeed, in terms of consistency with his overall Critical epistemology Kant can only give the supersensible self a negative characterization—namely as that aspect of the self which is not in space and time, and not subject to the categories. However, in the realm of goal‐orientated human activity—‘practical reason’—Kant holds that there are grounds for a much more positive characterization. His justification for this view consists in the fact that in the practical sphere our actions involve the exercise of free will on the basis of rational principles, rather than purely mechanical determination by that natural causality which arises from the application of the categories. This exercise of free will is embodied in ‘pure practical reason’ (i.e. morality). In this case (according to Kant) not only are our decisions motivated by rational principles alone (rather than by sensuous inclinations such as the desire for happiness) but they often involve us acting contrarily to those sensuous impulses which most emphatically locate us in the mechanistic framework of nature. In fact the very existence of morality (a fact which Kant takes as given) presupposes a rational and autonomous dimension to the self. For if our decisions were (p.18) always merely mechanically determined, then it would make no sense—in the way that morality demands—for us to hold people responsible for their actions, and to assign praise and blame accordingly. On these terms, the supersensible self must be presumed to be both rational and free.

Having outlined two salient eighteenth‐century theories of the sublime and Kant's own theory, and having also noted some of the crucial claims which characterize Kant's Critical philosophy, we can now attend to his Critical theory of the sublime.


(1) Joseph Addison, Collected Works iii, ed. H. Bohn (Bell and Sons: London, 1890), 397–8.

(2) Edmund Burke, A Philosophical Inquiry into the Origin of Our Ideas of the Sublime and the Beautiful, ed. John Boulton (Routledge & Kegan Paul: London, 1958), 136.

(3) W. H. Werkmeister, Kant: The Architectonic and Development of his Philosophy (Open Court: London, 1980), 36.

(4) Kant, Observations, 45.

(5) Ibid. 45.

(6) Ibid. 46.

(7) Ibid. 46. Kant's notion of feeling ‘of a more delicate sort’ is strongly reminiscent of Addison's description of ‘innocent pleasures’ in no. 411 of the Spectator essays on the Pleasures of the Imagination. Indeed Kant's placing of finer feeling between the pleasures of sense and those of ‘high intellectual insights’ parallels a similar placement in the Addison essay just mentioned.

(8) Kant, Observations, 46.

(9) Ibid. 47–8. Again, the combination of the sublime and the beautiful in terms of the ‘splendid’ parallels a similar move in Addison's essay no. 412. Burke, it is interesting to note, was much more reticent about the possibilities of such a combination.

(10) Ibid. 58.

(11) Ibid. 58.

(12) For a discussion of this, see the Editor's Introduction to Edmund Burke, op. cit. (n. 2 above), p. cxxv.

(13) Beauty Restored (Clarendon Press: Oxford, 1984), 234

(14) The Editor's Introduction to Burke, op. cit. (n. 2 above), p. cxxvi.

(15) Addison's Spectator essays on the Pleasures of the Imagination were translated into German in 1745.

(16) P. A. Schillp, Kant's Pre‐Critical Ethics (Northwestern University Press: Evanston, Ill., 1960), 60.

(17) Ibid. 60–1.

(18) Francis Hutcheson, ‘An Initial Theory of Taste’, in George Dickie and Richard Sclafani (eds.), Aesthetics: A Critical Anthology (St Martin's Press: New York, 1977), 569–91. This reference p. 570.

(19) Indeed, it is interesting that, in his An Inquiry into the Original of Our Ideas of Beauty and Virtue, Hutcheson points out that ‘Grandeur’ is an idea which differs from Beauty, though he does not elaborate upon the nature of the difference. Given that the weight of Kant's exposition in the Observations falls upon the sublime, it may be that he was consciously trying to fill a gap which Hutcheson had left. This claim is given added plausibility by the fact that Hutcheson's Inquiry was translated into German in 1762.

(20) Kant's series of lectures on Anthropology contains a discussion of the sublime; they were given over a period dating from before the ‘Critical’ phase of his philosophy. The version of the lectures published posthumously, however, has clearly been revised on the basis of his Critical position, and their discussion of the sublime is derived substantially from the mature theory outlined in the Critique of Judgement.

(21) A. C. Ewing, A Short Commentary on Kant's Critique of Pure Reason (Methuen: London, 1938), 162.