I. Introduction
Immanuel Kant’s analysis of poetry and its cognitive role revolves around the clear demarcation of the activity of the poetic imagination from the influence of either emotions or concepts. My paper takes this as a starting point in order to show that these demarcations are not based on ‘external constraints’ (I use ‘external’ here in the sense of using poetry ‘as the means to an end’, for example cognition) applied to poetry but are, according to Kant, part of the essence of poetry as such. Seen in this way, poetry can make a significant contribution to our knowledge and can express a particular form of truth that is specific to poetry – namely, aesthetic truth. Historically, my discussion will revolve around Kant’s Lectures on Logic (LL, dated ca. 1770–1800) and on his Lectures on Anthropology (dated ca. 1772–1789), engaging with the Critique of the Power of Judgement (1790) only to show if and how some of his earlier ideas found resonance in his later work.
My main motivation for engaging with this topic can be encapsulated as follows: the concept of perfection that Kant uses in his earlier analysis of poetry in the LL (where he is influenced by Georg Friedrich Meier and Alexander Gottlieb Baumgarten, while adding his own reflections to refine this concept) is central to understanding the cognitive role of poetry but also its freedom from concepts. As the following sections will show, Kant insists on a certain freedom or independence of the poetic image from the distinctness involved in conceptual determinations. I will attempt to show that, in a similar vein to existing analyses that focus on aesthetic ideas,1 the specific concept of ‘perfection’ discussed in the LL can play a key role in improving our appreciation of the cognitive role of poetry, and its freedom, while serving as a background for a deeper understanding of the notion of ‘aesthetic truth’ in poetry.
A further goal of this piece is to expose and challenge a possible misconception about the value of Kant’s aesthetic reflections in the LL (including his analysis of poetry), as part of his rationalist aesthetics. I believe that the reluctance to bring them into closer dialogue with the CJ stems, first, from the idea that Kant’s earlier analysis of perfection can be dismissed once he rejects this concept in the CJ (§ 15) and, second, from the idea that the shift taking place in Kant’s aesthetics over time (his increasing preference for ‘free beauty’) renders his earlier analysis of perfection in poetry superfluous. In my view, however, the situation is much more complex2 because the two definitions of perfections in the LL and the CJ do not overlap and because the ‘free beauty’ of poetry in the CJ is not something self-evident. I will come back to this at the beginning of Section III. For the moment I can say that, if one wants to have a full account of the freedom that Kant grants to poetry, as a liberation from the excessive influence of either sensibility or concepts, resulting in a pure poetic image untainted by external influences (I use the term ‘poetic image’ in the sense of spontaneous associations of different representations that each reader makes in their own individual ways),3 one should not ignore these lectures and jump straight to the analysis of poetry in the CJ. For this reason, my paper will focus less on the CJ and more on the LL.
Before moving on to the main part, it should be mentioned that Kant’s analysis of poetry, although a marginal topic of research, has not been completely ignored. Of the few articles that have dealt with this topic,4 I am here most concerned with Gabriele Tomasi’s valuable article ‘Kant and the Cognitive Value of Poetry’ and with the comprehensive anthology Kant on Poetry, both of which have already covered a large area of the research. I consider my article to be a necessary contribution to these works, because it brings the concept of ‘perfection’ (not mentioned in the anthology) into closer dialogue with the question of the freedom of poetic beauty (which Kant seems to classify under ‘dependent’ beauty) and of the aesthetic truth (opening up another area of research than the one limited to the cognitive role of poetry through aesthetic ideas). While part of my argument overlaps with Tomasi’s extensive work on the topic (for example, his question of how cognitive value interacts with aesthetic value in poetry), and with the anthology (more specifically, the importance of reducing emotional intensity), the focus here is a different one.
Regarding the question posed in the anthology – ‘Did Kant’s conception of poetry have any real influence on this system of thought?’ – my article offers an additional answer to the one given there, namely that Kant’s anthropology, which includes poetry, is central to his system of thought, and therefore his analysis of poetry must also be central. My alternative answer is made possible by shifting the focus to the LL and showing how his earlier analysis of ‘perfection’ (and his engagement with rationalist aesthetics) contributed to the further development of his understanding of poetry. How this enhanced perception of poetry led him to reconsider his aesthetic principles remains a task for future research.
I will now give a brief overview of the main sections of this paper and how they support my claim. In Section II, I emphasize the great importance that Kant attaches to the reduction of emotional intensity in poetic writings as a necessary condition for originality. Without denying the essential role of emotions in the aesthetic experience of poetry,5 this is a first step towards showing that poetry should not aim to passively arouse sensations in the reader or seek to gain their attention and approval by lesser means than intellectual and moral improvement (as bad rhetoric does,6 for example). While Section II is concerned with what poetry should not aim to achieve, Section III takes a closer look at the real purpose of poetry, through an analysis of the struggle between logical and aesthetic perfection present in the LL. The aim of these two sections is to show that what appears to be a conceptual constraint of poetry in Kant’s analysis, in order to highlight its cognitive purpose, that is, the further cultivation of our intellect through the enrichment of our concepts, can actually contribute to a better understanding of the necessary independence of poetic imagery from the ‘distinctness’ of logical thought. In Section IV, I relate the previous discussion of perfection to the problem of aesthetic truth, showing that, through this interplay of aesthetic and logical perfection, poetry becomes capable of expressing truth without deceiving us, and that such a subjective truth can legitimately be called truth.
II. Emotional Intensity and Originality: Avoiding the Pitfalls of Poetry
In this section, I focus on Kant’s discussion of the key role of tradition in toning down the emotional intensity of poetry. The relevance of this section to the overall aim of the paper is to show how, according to Kant, any excess of emotionality in poetry can be a hindrance to the spontaneity of productive imagination. To show that a minimum condition for poetic originality and for the freedom of poetic imagination is to aim higher than an immediate form of empathy, I will consider two examples of Kant contrasting poets.
The first one is Kant’s comparison of Homer and Virgil, in the light of which his preference for moral and intellectual content in poetry over emotionality becomes clear. Kant regards Homer as an ‘unpolished’ genius in comparison to his successor Virgil, as he states in his Anthropology Mrongovius (1784–1785): ‘Genius is for the most part coarse like Shakespeare – he lacks polish; Homer, too. The virtuoso polishes it away; Virgil is like this.’7 As Nuria Sánchez Madrid has already pointed out, the fact that Virgil could draw on Homer’s example, thus ‘reducing the intensity of the passions in his poetry and hone the more intellectual features of his verses’,8 led to Kant’s placement of Virgil above Homer:
An example of this can be seen in Kant’s remark in Observations on the Feeling of the Beautiful and the Sublime (1764) that Homer’s characters in Iliad display a sublime, terrifying wrath, while the heroes of Virgil’s work appear sublime in a noble sense: […] The wrath of someone fearsome is sublime, like the wrath of Achilles in the Iliad.9
Later in the CJ Kant will quote the verses of Frederick the Great, describing how they evoke sublime and calming feelings that facilitate the cultivation of virtue in the reader (CJ, 5:316). We can now turn to the second comparison, that of John Milton and Friedrich Gottlieb Klopstock.
Klopstock’s aim was to create a new poetry on a par with Homer’s Iliad or Milton’s Paradise Lost. However, the infusion of emotion into his style of writing was not to Kant’s liking. In the Anthropology Mrongovius, Kant writes: ‘We must not judge beauty according to charm, [and] especially not according to emotions. For they mislead very much. Poems that surely are emotional do not have true beauty. Klopstock is like this.’10 Poetic language must not be imbued with emotion, aiming at the sympathy of its readers, because this kind of sympathy can only be of a momentary nature and cannot have a lasting effect on the reader:
Klopstock here makes the error of powerfully contorting the language, but he did not have a literary gift, for he mostly speaks the language of one who is astounded or terrified, and he arouses admiration through sympathy rather than arousing admiration through the depiction of objects, as Milton does.11
Kant’s verdict is that Klopstock’s poems lack originality, because they arouse sensations that everyone can already find in themselves. In the Anthropology Friedländer (1775–1776) he makes a similar statement about love poems: ‘Many poems are mere plays of sensation, for example, love poems. A talented poet must not bother with this, because it is very easy to arouse such sensations, since everyone already has such sensations on their own.’12 Throughout the LL, Kant repeatedly insists that the beauty of poetic language should not aim to arouse sensations in the viewer or listener. For example, in the Vienna Logic (ca. 1780) he states: ‘He who wants to convince his listeners must not waken their sensations but must instead present the matter to their intuition.’13 What is only stimulating or exciting, what only pleases the senses, does not require any spontaneity on the part of the reader. While Kant does not deny the role of emotion in poetry, he believes that original poetry should not aim to have a passive effect on the reader (by triggering already familiar sensations and images) but that it should challenge the reader to recreate the original poetic images through her own imaginative power. This seems to be the reason why, in a passage dated around 1776–1778 and included in his Reflexionen zur Anthropologie, he states that Milton’s poetic images are inimitable, unlike those of Klopstock: ‘Klopstock can be imitated very well, but Milton is difficult to imitate because his images are original’ (AA, 15:400, § 914).14 The term ‘original’ does not simply mean ‘new’ or ‘never encountered before’; it also implies that it cannot easily be reproduced.
Original poetic imagery, like Milton’s, creates original associations in the reader’s mind, as opposed to poems, which only overstimulate the sensations we already carry within us. While some authors question Kant’s familiarity with Milton’s texts,15 I will simply state here that the parallels he draws are in line with his philosophical theories about the productive imagination and the misleading effect of sensations. This will become clearer later in the paper.
My aim in this section has been to introduce Kant’s rejection of the overriding influence of passive emotion on spontaneous imaginative or intellectual activity in poetry. In the next section I will focus on how Kant described the freedom of poetic language from the constraints of determinate concepts.
Before concluding this section, I would like to address some potential criticisms of the concept of the poetic image as outlined thus far, which are also connected to the historical significance of Kant’s theory of poetic imagery. His analysis could offer some solution to the concerns of those aesthetic theorists, such as Edmund Burke, who denied that abstract words like ‘virtue’ or ‘honour’ could evoke any kind of representation.16 We have seen above that, according to Kant, poetic images serve to make abstract concepts intuitive, or ‘present the matter to our intuition’. While language as such may contain abstract concepts, the main task of poetic language is to provide a way out of the abstractness of language, on the condition that the poetic image prevails over the abstract general meaning of words. An example can be found in the CJ, where Kant states that even intellectual concepts like ‘virtue’ can animate our senses by aesthetic means:
Conversely, even an intellectual concept can serve as the attribute of a representation of sense, and so animate the latter by means of the idea of the supersensible; but only insofar as the aesthetic, which is subjectively attached to the consciousness of the latter, is used to this end. Thus, e.g., a certain poet says in the description of a beautiful morning: ‘The sun streamed forth, as tranquillity streams from virtue.’ (CJ, 5:316)
This point holds not only for poetry but also for prose, as an example from the LL shows:
The completion of the perfections of all our cognitions is finally to give them sensibility, so that one thinks of the abstractum in concreto in a single, individual sensible case. E.g., if I think of friendship, true love, and the mutual helpfulness that flows from these, in the case of Damon and Pythias. Here, then, I think the universal in individual cases.17
Kant could also provide an answer to Gotthold Ephraim Lessing’s claim in Laocoon against the poet as ‘a painter of images’ and his complaint that a poet cannot reconstruct a ‘harmonious picture’ of physical beauty, because ‘what painters can best express by lines and colors is least capable of expression in words’.18 Therefore, the poet’s task cannot be ‘to make us see and feel […] the whole beauty of the beloved object. He is conscious of the inadequacy of all verbal expression.’19 Again, this would not be a major concern for Kant. One does not need to imagine Dante’s Beatrice in all her details, let alone a recognizable individual face, but the poetic image can still be striking and make an abstract concept intuitive, in the case of Beatrice, the abstract concept of ‘beauty’: ‘With eyes of light more bright than any star / in low, soft tones she started to address me / in her own language, with an angel’s voice.’20
III. Aesthetic and Logical Perfection: The Inner Aims of Poetry
So far, we have seen how an excessive outpouring of emotion can stand in the way of the free activity of the productive imagination, an obstacle that the best poets should strive to avoid. My focus now turns to the freedom of poetic imagery from the constraints of determinate concepts, a freedom that the best poets should aim to attain. To this end, I will examine Kant’s analysis of aesthetic and logical perfection in the LL and his rejection of ‘distinctness’ in poetry. But, first of all, some clarifications are needed.
In analysing the freedom of poetic images starting from the discussion of ‘perfection’ in the LL, one must proceed with caution, for the following reason. In § 15 of the CJ Kant rejects the concept of ‘perfection’ from the domain of aesthetics because perfection is always connected to a concept or an end (‘what the thing is supposed to be’) and not to a feeling, to an objective, not a subjective purposiveness (CJ, 5:227). Later, in § 16, Kant writes that, while beauty does not benefit from its combination with perfection, ‘the entire faculty of the powers of representation gains if both states of mind are in agreement’ (CJ, 5:231). He later writes in § 48 that the concept of ‘perfection’ must be taken into account in judging the beauty of any work of art, because it refers to ‘the agreement of the manifold in a thing with its inner determination as an end’ (CJ, 5:312).
Leaving aside these internal contradictions in the CJ, for my present interest I need to point out that the two definitions of perfection in the LL and in the CJ do not overlap (as I will show later, this opens the possibility that the concept of perfection discussed in LL could still be implicitly involved in his discussion of poetry in CJ). Therefore, when Kant in the CJ contrasts the ‘free’ beauty with ‘adherent’ beauty – the ‘adherent’ beauty involving ‘perfection’ (CJ, 5:229) – we need to be aware that Kant has in mind the understanding of perfection as described in CJ, not in the LL. If we bear this in mind, it will be easier to understand why in the following I will allow myself to speak of the ‘freedom’ of poetic beauty in connection with ‘perfection’.
Having clarified this, we can now proceed to the analysis of logical and aesthetic perfection in the LL, which predates the third Critique. In these lectures, Kant uses the philosophical terminology of Meier and Baumgarten and, by using their terminology, also aims to shift its meaning.21 His analysis of poetry in the LL is valuable for the following reasons: the tension between aesthetic and logical perfection and Kant’s criticism of the concept of distinctness employed by the rationalist aesthetics in the analysis of poetic language gradually pave the way for treating poetry as an original way of thinking with poetic means. This ‘rationalist’ phase of Kant’s aesthetics does not aim to reduce the aesthetic experience of poetry to its cognitive purpose (as an enrichment or illustration of abstract concepts) but to help understand how concepts can guide the free play of the imagination. I will now explain this idea more clearly.
Because the medium of poetry is language, we are often led to focus on the conceptual meaning, neglecting the fact that poetic meaning can sometimes consist of images, as spontaneous associations of different representations (according to the definition of the poetic image that has been provided in Section I of this paper). If we experience a poem that aims at this production of images in us, as delivering nonsense, we are denying the freedom of the imagination from the constraints of the understanding. This is my reading of Kant’s statement from the anthropology lectures Menschenkunde (1781–1782): ‘[W]hen the poet adorns a whole succession of thoughts with images, the beautiful must immediately shine, but the understanding must only come later and the thought must not immediately shine through, but only in the aftertaste’ (AA, 25:987–88).22 The more effort we put into recognizing and experiencing the peculiarity and novelty of the poetic act, leaving the understanding of its conceptual content to the ‘aftertaste’, the more valuable its contribution to thought will be. However, a certain tension arises here, because the freedom of poetry can never be understood in isolation from concepts. After all, one cannot take pleasure in reading poetry if it is divorced from meaning.23 In the Anthropology Pillau (1777–1778), Kant also insists that poetic images are always accompanied by corresponding concepts: ‘In poems, sensibility sets the end and the understanding must only give it unity. […] A poet takes a theme and depicts it through sheer images, which, however, have to be so constituted that they are accompanied by conforming concepts.’24
In what follows, I will show how the LL can help us to better understand Kant’s connection between the freedom of poetic images and concepts. Before we get there, however, a few clarifications are necessary at the outset. In the Vienna Logic, Kant links ‘aesthetic perfection’ with the cognition of the subject and its feelings and clearly distinguishes it from ‘logical perfection’, which is connected to the cognition of the object:
Logical perfection is concerned with the understanding and is the cognition of objects through it. Aesthetic perfection is concerned with feeling and with the condition of our subject, namely, how we are affected by the object. For through the beauty of cognition I do not seek cognition of the object but of the subject. (VL, 24:809)
According to J. Colin McQuillan, ‘the idea that logically perfect cognition agrees with the constitution of the thing, while aesthetically perfect cognition agrees with our feeling and tastes is not found in Baumgarten or Meier, so it would also seem to be original to Kant’.25
Although a clear definition of the concept of perfection as such seems to be missing from the lectures, which in turn makes the concepts of ‘aesthetic’ and ‘logical’ perfection rather vague, I believe that one way out is to break these two concepts down into their main features and talk about aesthetic and logical perfections: truth, distinctness, certainty, extensiveness (BL, 24:58), liveliness (BL, 24:126), and so on. Here I will only focus on two of them: distinctness (required for logical perfection) and liveliness (required for aesthetic perfection).
The (intensive) distinctness of philosophical language is attained by means of abstraction, while the (extensive) liveliness of poetic language is achieved through the combination of a multitude of marks in one representation. According to the Vienna Logic, ‘Poetry is reckoned first of all under aesthetic perfection’ (VL, 24:808) because it describes things through a multitude of marks, thereby creating the impression of liveliness. First, Kant attributes an ‘extensive’ distinctness to poetic language, as opposed to the ‘intensive’ – or ‘deep’– distinctness of philosophical language:
Distinctness, furthermore, is a perfection of cognition, and it has the peculiarity of belonging to logical as well as to aesthetic perfection as a necessary property. An extensive distinctness, which one encounters in the beautiful understanding, is an aesthetically pure one[;] intensive distinctness, however, which one finds in the deep understanding, belongs to logical perfection. (BL, 24:57)
Later, he attributes distinctness only to logical perfection, distinguishing it from the liveliness characteristic of aesthetic perfection: ‘Liveliness is an aesthetic perfection. Distinctness, however, […] is a logical perfection. […] One attains distinctness in a cognition, however, when one seeks to produce it by means of much abstraction’ (BL, 24:126). Later in the Vienna Logic Kant will also sharply distinguish between the two: ‘The essential aesthetic perfection is the opposite of the logical[;] distinctness of intuition is nothing more than liveliness in aesthetic truth’ (VL, 24:807). This contrast between liveliness and distinctness will be fundamental for Kant’s understanding of the freedom of poetic imagination.
Referring to Milton’s Paradise Lost, Kant states in the Blomberg Logic that ‘the beauty of all good poetic representations, e.g., of paradise lost, [has] an aesthetic perfection. It is sensed only in confused concepts, and it loses its value just as soon as the concept is made distinct’ (BL, 24:51). Why are indistinctness and confusion considered aesthetic ‘perfections’ and how are they related to liveliness? The explanation lies in the fact that liveliness ‘is attained with the help of much combination’, more precisely:
[I]n a description of spring I represent it in a lively way through a multitude of marks coordinate with one another. The poet does it thus. He shows, e.g., the budding flowers, the new green of the forests, the cavorting herds, the renewed rays of the sun, the lovely, charming air[,] the revival of the whole of nature. (BL, 24:126)
But how does this combination lead to ‘confusion’ and therefore to indistinctness? One explanation, coming from Gabriele Tomasi’s analysis, could be that in the poetic images these combined marks are melted or blended together (con-fused) and cannot be (logically) distinguished afterwards:
In fact, for a representation to be confused, it is necessary that its marks are not distinct from each other. Baumgarten considered these representations (the confused clarity of which he also labelled ‘extensive clarity’) highly poetic, and he therefore recommend their use in poetry. […] Confused/confusus derives from the Latin verb confundere, which is a compound of the prefix ‘con’ (‘together’ or ‘with’) and fundere. Interestingly, the latter word, in addition to meaning ‘to fuse’, also means ‘to spread out’ and ‘to extend’ – whence, as Frederick Beiser notes, ‘Baumgarten’s choice of the term extensive’.26
This also has a positive effect on cognition, because it allows the poetic image to represent ‘more parts of the sensed object’ than usually represented in the abstract language (of philosophy, for example). This leads Tomasi to draw an interesting parallel between this role of ‘con-fusion’ in cognition and the CJ’s analysis of aesthetic attributes that constitute an aesthetic idea that expands our concepts:
Now, extensive clarity is a function of confusion, namely of the number of marks (notae) of the thing represented together in a single representation. This means that in a sense representation that is extensively clearer than others, more parts of the sensed object will be represented. We can read the aesthetic enlargement of a concept through an aesthetic idea, of which Kant speaks, as a transformation of Baumgarten’s extensive clarity.27
A similar parallel is drawn by Claudio La Rocca:
In the aesthetic expansion (Erweiterung) of the concept, a transformation of Baumgarten’s concept of extensive clarity (extensiven Klarheit) can be recognised within the framework of critical theory, which contains a wealth of characteristics that cannot be ‘made clear’ (‘deutlich gemacht’).28
Their parallel is supported by § 49 of the CJ, where Kant describes the representation of imagination as that which ‘by itself stimulates so much thinking that it can never be grasped in a determinate concept [and] gives more to think about than can be grasped and made distinct in it (although it does, to be sure, belong to the concept of the object).’ We see how poetic language and the representations evoked by it ‘belong to the concept’ without being constrained by it. This is possible because the aesthetic attributes ‘go alongside the logical ones, and give the imagination an impetus to think more, although in an undeveloped way, than can be comprehended in a concept, and hence in a determinate linguistic expression’ (CJ, 5:315). The poetic imagination itself is capable of thinking, but the images it produces cannot be made ‘determinate’, like the philosophical concepts, for example. The freedom of poetic images from concepts should therefore be sought in the indistinctness and con-fusion, which make possible the liveliness of poetic images in contrast to the dryness, which emerges by means of abstracting from the marks of objects, as Kant emphasizes in the Blomberg Logic:
Liveliness in a cognition is attained with the help of much combination. One attains distinctness in a cognition, however, when one seeks to produce it by means of much abstraction. […] All sensibility brings about liveliness. Suitability to reason, however, produces dryness. (BL, 24:126–27)
He does so too in the Vienna Logic: ‘For a cognition can be thorough and logically perfect, but at the same time dry’ (VL, 24:809).
From the above quotations, it can be concluded that the LL and the CJ agree on the fact that liveliness plays an important role in cognition: the con-fusion of aesthetic attributes makes possible the aesthetic ideas, which in turn (climbing on the aesthetic attributes) refer to ‘invisible beings, the kingdom of the blessed, the kingdom of hell, eternity, creation’ or abstract concepts such as ‘death, envy, and all sorts of vices, as well as love, fame’ (CJ, 5:314).
Returning to the earlier idea that imagination itself is capable of thought, one possible explanation could be found in the Vienna Logic, where Kant repeatedly emphasizes that logical perfection, as the skeleton and basis of all cognition, is a prerequisite for aesthetic perfection. In his words:
Logical perfection is the conditio sine qua non and the basis of all thought. Aesthetic perfection cannot exist by itself, but is only the ornamentation of already correct logical perfection. Now one can, to be sure, leave something out of logical perfection. But one must not mutilate it. One can also […] ignore some of the truth […] yet without in the least causing any lack in logical perfection. (VL, 24:808)
The logical perfection that Kant has in mind here is truth, as he already announced in the Blomberg Logic: ‘truth is a perfection that is included in the logical as well as in the aesthetic perfection of cognition. It is a principal logical perfection, on which the others all depend. […] For aesthetic perfection, truth is also required’ (BL, 24:56).
If we are looking for a concrete example of the importance of truth in poetry, we could consider the creation of metaphors. Drawing on Kant’s analysis of imagination,29 Berys Gaut discusses the paradigm of metaphor-making:
[M]etaphors can also be extravagant and unconvincing and can misfire in various ways; a good metaphor […] must seem appropriate to its object. In this respect the cognitive content of the metaphor is important: if there are properties literally possessed in common between the two items linked by metaphor, then the metaphor will prove apt. It is because men really do have some salient attributes in common with (the ordinary conception of) wolves that the wolfish metaphor is an apt one. […] the value of aptness, […] rests on a cognitive insight. […] For in a good metaphor, concepts and domains of thought otherwise far removed from each other are brought into intimate contact, reconfiguring the familiar conceptual terrain into a place both hauntingly strange yet oddly right.30
Kant may as well have put it like this: the forging of an ‘apt metaphor’ is only possible if the logical perfection serves as basis. But is it only logical truth that we are dealing with here? In the Vienna Logic Kant states that in poetry ‘logical perfection is somewhat less in degree than aesthetic perfection, and aesthetic perfection always struggles with logical perfection for a place’ (VL, 24:808). The question that arises here is: why does Kant speak of a struggle if the logical perfection is a stable foundation and aesthetic perfection just its ‘ornamentation’? A possible answer to this question can be found in Kant’s analysis of the aesthetic truth of poetry, which stands in a certain tension with the logical truth, because poetry aims to express truth with poetic means. This will be the topic of the next section.
IV. The Aesthetic Truth of Poetry
In Section II we saw how the emotional excess in poetic language can be detrimental to the freedom of poetic imagination. According to the Jäsche Logic (ca. 1800), stimulation and excitement can also have a ‘disadvantageous influence on the logical perfection’.31 In Section III I analysed how Kant’s discussion of logical and aesthetic perfections in his LL pursues two aims: (1) to establish a certain freedom of poetic images from conceptual constraints (through the rejection of distinctness) and (2) to show that logical perfection must be a basis for aesthetic perfection, but that aesthetic perfection must have a certain priority.
So far, the focus has therefore been on the way in which the activity of poetic imagination is distinguished by Kant from the activity of the senses (Section II) and the activity of concepts (Section III). These boundaries are not external to poetry (turning it into a means to another, that is, a non-poetic end, such as cognition) but regulate its unconstrained play organically, from within the process of creation. In this section, I show how these demarcations make possible what Kant calls aesthetic truth.
As we saw in the previous section, effective metaphors and poetic images gain their striking character from their connection to truth. However, if poetry aims to express truths, this may set certain limits to it, so the poet chooses to mix ‘a lot of untruth with it’, as mentioned in the Anthropology Pillau:
Why is the poet happier in fable than in truth? Because his aim is not to promote the understanding, but he takes merely the imagination as his chief end. Truth sets limits for him, and he does not love that at all. The orator, however, is happier in truth than in fable; e.g. poets are not happy in describing life, or if that is suitable, they mix a lot of untruth up with it.32
Referring to Milton’s depiction of death and sin ‘as persons’ in Paradise Lost, Kant states in the Blomberg Logic:
Milton is reproached for representing death and sin as persons, as it were, because this is not appropriate to their constitution. But with the aesthetically perfect we do not require as much truth as with the logically perfect. With the aesthetic, something may be true only tolerabiliter. In this way it is aesthetically true that Milton represents the angels in the paradise lost as quarreling, and caught up in battle, for who knows whether this cannot occur. (BL, 24:56)
This mixture of truth and untruth should not make the reader feel ‘deceived’ or lead to a certain ‘uncertainty’ on her part (on the contrary, according to the Blomberg Logic, certainty must be an essential part of the aesthetic experience): ‘Certainty belongs to logical and to aesthetic perfection. The uncertain, the wavering, always displeases’ (BL, 24:57).
For instance, the poet does not describe the state of the things, but only how something ‘appears to our senses and seems to be’, as mentioned in the Vienna Logic: ‘The sun sinks into the water, says the poet. If he were to say that the earth turns on its axis, then he would assimilate to logical truth and not be a poet’ (VL, 24:810). While reading ‘the sun sinks into the water’, one is simultaneously aware of its logical untruth and its aesthetical truth. The fact that one can be aware of logical untruth shows how logical perfection still serves (in the background) as a basis for the aesthetic perfection of poetry, while allowing aesthetic perfection to assert its dominance through the momentary substitution of logical truth for aesthetic truth. The open question is why Kant is of the opinion that the aesthetical truth is worthy of being called such. I believe an answer to this question can be found in the Jäsche Logic, where Kant sets out the last two of the four ‘moments’ of aesthetic perfection:
aesthetic truth. A merely subjective truth, which consists only in the agreement of cognition with the subject and the laws of sensory illusion, and which is consequently nothing more than a universal semblance. aesthetic certainty. This rests on what is necessary in consequence of the testimony of the senses, i.e., what is confirmed through sensation and experience.33
Aesthetic truth deserves to be called such because it can be universally confirmed by the ‘testimony of the senses’ and is in agreement with our natural way of experiencing the world (the ‘rising’ and ‘setting’ of the sun): ‘That which agrees with universal illusion is aesthetically true, e.g., the rising and setting of the sun – thus it is quite distinct from logical truth.’34
Consider the following example given by Kant in the CJ:
[W]e must not take the sight of the ocean as we think it, enriched with all sorts of knowledge (which are not, however, contained in the immediate intuition), for example as a wide realm of water creatures, as the great storehouse of water for the evaporation […], or as an element that separates parts of the world from one another […]; rather, one must consider the ocean merely as the poets do, in accordance with what its appearance shows, for instance, when it is considered in periods of calm, as a clear watery mirror bounded only by the heavens, but also when it is turbulent, an abyss threatening to devour everything. (CJ, 5:270)
What we can glean from this passage is, that for aesthetic truth, we must operate an abstraction from our knowledge and rely instead on our ‘immediate intuition’. Aesthetic truth, therefore, requires a certain amount of effort on the part of the reader, and without this effort many poetic images can be disregarded as superfluous. It is useful to remember that Section II dealt with the lack of originality in the poetry that contains ‘emotions’ that the reader can immediately find within herself, therefore subjecting the readers to passivity. In this case, the immediacy in question cannot be attained by passive readers. As we will see below, the poetic ‘illusions’ that Kant is discussing here have the capacity to discipline our senses and therefore they involve a certain spontaneity on the part of the reader.
If one were to ask the question ‘Where exactly does the “poetic image” fit into Kant’s aesthetic theory as a whole?’ the answer could also be found here. In opposition to representations evoked automatically by everyday language35 (or affects),36 a poetic image is always produced spontaneously and voluntarily (compared to other arts, which, in Kant’s view, aim at the passivity of the audience, such as music):37 ‘Poesis [Dichten] is as much as the faculty of producing voluntary representations in the imagination.’38
The representations we produce when reading and writing poetry require our power of imagination and are therefore capable of freeing us from the limitations of the ‘law of association’:
The imagination (as a productive cognitive faculty) is, namely, very powerful in creating […] another nature, out of the material which the real one gives it. We entertain ourselves with it when experience seems too mundane to us; we transform the latter […] in this we feel our freedom from the law of association. (CJ, 5:314)
I now turn to a less-known piece presented by Kant to the University of Königsberg in 1777 as part of Johann Gottlieb Kreutzfeld’s appointment to the chair of poetics. In light of the previous discussion, the following passage is particularly relevant, which is why I am quoting it in full:
There are semblances through which the mind is not deceived but with which it plays. Whoever invokes it does not want to bring about error in unsuspecting minds, but truth in the form of appearance, which form does not obscure the inner being of truth but only displays it decorated for view. The appearance does not deceive the inexperienced and gullible through decoration and delusion. Rather, it utilizes the capacity for illumination of the senses and in this manner allows the meager, sapless picture of truth to come forth dipped in sensuous color. If there is contained in such semblance something by which it deceives, as it is commonly put, I would rather call it illusion. Semblance which deceives disappears as soon as its deceitfulness and (empty) play are perceived. But playful illusion, since it is nothing but truth as appearance, remains even when seen for what it is. The latter keeps the mind in pleasant motion, fluctuating as it were in the border region between error and truth, and greatly flatters the mind which is aware of its sagacity with respect to the temptations of semblances.39
As we saw in Section III, the ‘dryness’ – here called the ‘meager, sapless’ nature – of logical truth or logical perfection is opposed to the liveliness involved in aesthetic perfection. The depiction of abstract concepts through concrete sensible cases (as previously mentioned in Section II) makes possible a truth ‘dipped in sensuous color’, which makes our cognition lively.
Kant’s distinction between ‘illusion’ and ‘deception’ can help to understand better why the aesthetic truth of poetry involves spontaneity on the part of the reader. First of all, the poetic art has the power to refine our senses, making them subject to our increased control:
For this reason this art belongs to the noble and free arts, that is, to the arts which promote the freedom of the mind, since it soothes the senses and thereby ridicules their greedy longing and since it brings it about that those who have been attracted by its beauties and who have shed their crudity will all the more obey the lessons of wisdom.40
Second, while deception displeases and disappears as soon as it is discovered (see Kant’s example of the nightingale in § 42 of the CJ), illusion persists and pleases us. As the following passages from his Anthropology Friedländer and Vienna Logic show, in contrast to deception, illusion can be combined with truth: ‘All semblance is first an illusion, if it can harmonize with the cognition of truth. However, all semblance is deception as soon as it does not agree with the cognition of truth.’41 ‘The poetically true is not what is logically true, but instead what accords with illusion’ (VL, 24:807). One can be aware of the truth and still take pleasure in the illusion. Further, the poet has no intention to deceive us, because in his poem he presents an illusion that is in agreement with ‘the laws of sensory illusion’ (as quoted previously in Section II) and therefore does not falsify our ‘natural’ way of experiencing the world.42 The parallel between poetic illusions and optical illusions that Kant draws is helpful in this context:
On the other hand, in the case of playful optical illusions I do see through the semblance quite well and am protected against error and nonetheless continue to feel joy. Unquestionably the semblance here triggers joy because it does not deceive but strongly though unsuccessfully entices into error. Hence insofar as semblances deceive, they induce displeasure, insofar as they only play with us, they induce pleasure. In something like this consists the difference between the common sensory deceptions and illusions employed by poets. (AA, 15:908)
We experience pleasure in illusion because our mind becomes ‘aware of its sagacity with respect to the temptations of semblances’, of its rational nature, in not being deceived by this play. The mind’s fluctuation ‘in the border region between error and truth’ is possible due to the limitations imposed by our logical nature (showing once more why Kant thinks that in poetry logical perfection should be a basis).
The aim of this section has been to show how the freedom of poetic imagination can facilitate the experience of aesthetic truth in poetry. One recurring theme in this paper was the idea that freedom is established by setting limits. While Section II focused on the limitation of emotions, Section III focused on the mutual limitation of concepts and sensibility. While these limitations enable poetry to make an original contribution to knowledge, they should not be viewed as external constraints on poetic imagination, turning it into a means to a non-poetic end. Rather, they should be seen as safeguarding its freedom from excessive influence, whether from the senses or from the intellect.
V. Conclusion
In order to show how Kant can allow poetry to express a certain truth – aesthetic truth – without imposing any external constraints on the poetic imagination, I proceeded as follows.
In Section II I analysed how according to Kant, the spontaneity of the productive imagination can be hindered by any excess of emotionality in poetry. A limitation of sensations and emotions is therefore a prerequisite for the freedom and originality of the poetic imagination. In Section III I showed how Kant aimed to liberate poetic images from ‘distinctness’ as a logical perfection, while maintaining the tension between logical and aesthetic perfection, which is essential to poetic experience. In Section IV I linked this tension between the two perfections to the problem of aesthetic truth and its role in enlivening our cognitive powers and allowing poetic images to overcome certain conceptual and sensory barriers.
In conclusion, if we are to take poetry’s role in cognition seriously, then poetry should be approached by first paying attention to its original poetic way of expressing a particular content, and not through the distorting lens of conceptual thought. The freedom of the poetic imagination is not a given but the result of a laborious process.
Notes
[1] See Immanuel Kant, Critique of the Power of Judgment, trans. Paul Guyer and Eric Matthews (Cambridge: Cambridge University Press, 2000), § 49. Hereafter: CJ.
[2] Rachel Zuckert also points this out in her paper ‘Kant’s Rationalist Aesthetics’, Kant-Studien 98 (2007): 443–63.
[3] Küplen analyses how each of the complex mental representations evoked by poetic language ‘is itself construed of many different mental representations’. She offers the example of the word ‘sun’, which, when encountered in a poem, can evoke various other mental representations such as ‘light, round, warm and so on’. See Mojca Küplen, Kant’s Aesthetic Cognitivism (London: Bloomsbury, 2023), 111. Further discussion will be found in Section III of the present article.
[4] See Laura Penny, ‘The Highest of All the Arts: Kant and Poetry’, Philosophy and Literature 32 (2008): 373–84; C. E. Emmer, ‘Kiesewetter, Kant, and the Problem of Poetic Beauty’, in Natur und Freiheit: Akten des 12. Internationalen Kant-Kongresses, ed. Violetta L. Waibel, Margit Ruffing, and David Wagner (Berlin: De Gruyter, 2018), 2979–86; Gabriele Tomasi, ‘Kant and the Cognitive Value of Poetry’, Aesthetica Preprint, no. 116 (2021): 17–49; Iris Vidmar Jovanović, ‘Kant on Poetry and Cognition’, Journal of Aesthetic Education 54 (2020): 1–17; Fernando M. F. Silva and Bernd Dörflinger, eds., Kant on Poetry / Kant über Poesie (Baden-Baden: Olms, 2023).
[5] As Guyer notes, Kant could only have meant that ‘a taste that is pleased solely by emotion is defective, not that art that communicates emotion among its other merits is for that reason alone defective’. See Paul Guyer, ‘Form, Content, Emotion: The Place of Poetry in Kant’s Aesthetics’, in Silva and Dörflinger, Kant on Poetry, 36.
[6] According to § 53 of the CJ, if we understand ‘rhetoric’ to be ‘the art of persuasion, i.e., of deceiving by means of beautiful illusion’, then we must bear in mind that this type of rhetoric borrows elements from poetry in order to ‘win minds over to the advantage of the speaker before they can judge and to rob them of their freedom’. See CJ, 5:327.
[7] Immanuel Kant, Anthropology Mrongovius, trans. Robert R. Clewis, in Lectures on Anthropology, ed. Allen W. Wood and Robert B. Louden (Cambridge: Cambridge University Press, 2012), 25:1313.
[8] See Nuria Sánchez Madrid, ‘Kant’s Virgil: Poetical Authority as a Normative Source of Reason’, in Silva and Dörflinger, Kant on Poetry, 176.
[12] Immanuel Kant, Anthropology Friedländer, trans. G. Felicitas Munzel, in Lectures on Anthropology, 25:527.
[13] Immanuel Kant, The Vienna Logic, in Lectures on Logic, ed. and trans. J. Michael Young (Cambridge: Cambridge University Press, 1992), 24:807. Hereafter: VL.
[14] References to Kant’s untranslated works are to the standard Akademie Ausgabe (AA): Immanuel Kant Gesammelte Schriften (Berlin: Reimer/De Gruyter, 1900–). Translations are my own, unless stated otherwise.
[15] See Daniel Shore, ‘Milton and Kant?’, Milton Quarterly 48 (2014): 27–30.
[16] Edmund Burke, A Philosophical Enquiry into the Origin of Our Ideas of the Sublime and Beautiful (1757/59; Oxford: Oxford University Press, 1998), 150.
[17] Immanuel Kant, The Blomberg Logic, in Lectures on Logic, 24:128. Hereafter: BL.
[18] See Gotthold Ephraim Lessing, Laocoon: An Essay upon the Limits of Painting and Poetry, trans. Ellen Forthingham (Boston: Roberts, 1874), 131.
[20] See Dante Alighieri, The Portable Dante, ed. and trans. Mark Musa (New York: Penguin, 2003), 10.
[21] Clewis writes that ‘Kant uses Baumgarten’s ethics textbooks, adopting his predecessor’s language of perfection – though perhaps without endorsing perfectionism completely. Indeed, he grows increasingly dissatisfied with aesthetic perfectionism.’ See Robert Clewis, The Origins of Kant’s Aesthetics (Cambridge: Cambridge University Press, 2023), 176.
[22] As translated by Tomasi in ‘Kant and the Cognitive Value’, 42 (translation not provided in the Lectures on Anthropology).
[23] This point is made by Emmer: ‘Even though much or even most poetry is expressly composed with an attention to its acoustic materiality, practically no one enjoys listening to such poetry if they cannot understand the words’. See his ‘Kiesewetter’, 2983.
[24] Immanuel Kant, Anthropology Pillau, trans. Allen W, Wood, in Lectures on Anthropology, 25:760.
[25] J. Colin McQuillan, ‘Baumgarten, Meier and Kant on Aesthetic Perfection’, in Kant and His German Contemporaries, vol. 2, Aesthetics, History, Politics, and Religion, ed. Daniel O. Dahlstrom (Cambridge: Cambridge University Press, 2018), 18.
[28] Claudio La Rocca, ‘Das Schöne und der Schatten’, in Im Schatten des Schönen, ed. Heiner F. Klemme, Michael Pauen, and Marie-Luise Raters (Bielefeld: Aisthesis, 2006), 38, my translation.
[29] Gaut emphasizes that ‘Kant’s connection of creativity with imagination in its employment of metaphor-making is intriguing, and captures an important insight’. See Berys Gaut ‘Creativity and Imagination’, in The Creation of Art, ed. Berys Gaut and Paisley Livingston (New York: Cambridge University Press, 2003), 164.
[31] According to the Jäsche Logic, ‘one must be very cautious with stimulation and excitement, whereby a cognition affects sensation and acquires an interest for it, because attention can thereby so easily be drawn from the object to the subject, whence a very disadvantageous influence on the logical perfection of cognition must evidently arise’. Immanuel Kant, Jäsche Logic, in Lectures on Logic, 9:38.
[34] Immanuel Kant, Dohna-Wundlacken Logic, in Lectures on Logic, 24:723.
[35] According to the Metaphysik Mrongovius, language, designated as ‘facultas signandi’, is ‘mechanical, i.e. without any exertion of power, and also involuntary, e.g. if I say Rome, then the representation of this city immediately springs forth. I may want it or not. This comes from habit.’ Immanuel Kant, Metaphysik Mrongovius, in Lectures on Metaphysics, ed. and trans. Karl Ameriks and Steve Naragon (Cambridge: Cambridge University Press, 1997), 29:887.
[36] For example, in his Anthropology Mrongovius (25:1341) Kant writes: ‘He who can vividly imagine an affect can do this better than he who himself has such an affect. Great poets often were not full of affect, e.g., Young.’
[37] About music, Kant writes in the CJ (5:206) that it ‘occupies the lowest place among the beautiful arts […] because it merely plays with sensations’.
[39] Immanuel Kant, ‘Concerning Sensory Illusion and Poetic Fiction’, in Kant’s Latin Writings: Translations, Commentaries and Notes, ed. Lewis White Beck et al., trans. Ralf Meerbote (New York: Peter Lang, 1986), 15:907.
[42] This is also emphasized by Lemos: ‘By contrast with the sea appearing higher in the middle than at the shores, the delusions produced by natural-colour painted statues, makeup, or tricks of conjury are artificial (künstlich). As such, they may presuppose an intention of deceiving.’ See João Lemos, ‘Kant on the Deceitful Art of Rhetoric’, British Journal of Aesthetics 65 (2024): 5.
Acknowledgements
I presented this paper in various forms at the Lisbon Annual Conference on Kant’s Aesthetics in March 2024, as well as at the PhD group meetings in Tübingen. I would like to thank the participants and organisers in Lisbon as well as my colleagues and my supervisor in Tübingen for their helpful comments and questions. My article also benefited greatly from valuable feedback from the guest editor João Lemos, and two anonymous referees, Iris Vidmar Jovanović and C.E. Emmer, whose identities were disclosed by mutual agreement after acceptance.
Competing Interests
The author has no competing interests to declare.