Arthur Schopenhauer (1788–1860) begins his main discussion of music, in Section 52 of The World as Will and Representation, by noting that “It stands quite apart” from all the other forms of art (1969: vol. 1, 256). (All further references to Schopenhauer are to this work.) “[M]usic is by no means like the other arts, namely a copy of the Ideas,” he writes, but is rather “a copy of the will itself . . . For this reason the effect of music is so very much more powerful and penetrating than is that of the other arts, for these others speak only of the shadow, but music of the essence” (vol. 1, 257). This is a claim that composers and musicians have, perhaps unsurprisingly, found extremely seductive: Richard Wagner, for example, held that Schopenhauer captures “the position of music among the fine arts with philosophic clearness,” and in doing so “recognises the true nature of music” (Goehr 1996: 201); other admirers included Liszt, Brahms, Rimsky-Korsakov, Mahler, Schoenberg and Prokofiev. But why did Schopenhauer hold that music “stands quite apart” from the other arts? Just what does he mean when he says that it, unlike them, speaks “of the essence” of things? These are the central questions facing anyone wanting to understand Schopenhauer’s theory of music, and in what follows I shall sketch (the beginnings, at least) of answers to them.
First, however, a reminder of the metaphysics and philosophy of art that underpins the theory will be useful. As an Idealist, Schopenhauer holds that the true nature of things is quite different from that which is presented to us in sense experience. But his take on the nature of reality is highly distinctive: reality, he argues, is will, a single arational, impersonal force that is constantly “striving” or in flux. Like the Kantian noumena, will “is that of which all representation, all object, is the phenomenon, the visibility . . . It is the innermost essence, the kernel, of every particular thing and also of the whole” (vol. 1, 110). Unlike the Kantian noumena, however, will is, if not directly knowable by human beings, to some extent accessible, for it “manifests” or “objectifies” itself, with varying degrees of clarity, in phenomena: in forces of nature, in organic and inorganic matter, in sentient creatures, and most clearly or perfectly of all in human beings.
Schopenhauer’s point is not that will objectifies itself to a greater or less extent in different phenomena, in such a way in which, for example, there might be more of it in a plant than there is in a rock; what varies in degree is rather the objectification itself – the will is more “visible,” or more clearly objectified, in the plant than it is in the rock, and more clearly in an animal than in a plant. “Indeed,” he writes, “the will’s passage into visibility, its objectification, has gradations as endless as those between the feeblest twilight and the brightest sunlight, the loudest tone and the softest echo” (vol. 1, 128).
It is at this point that what he calls the Platonic Ideas come into Schopenhauer’s picture. As he writes, “These different grades of the will’s objectification, expressed in innumerable individuals, exist as the unattained patterns of these, or as the eternal forms of things” (vol. 1, 129). It is to these patterns or what might be called templates that Schopenhauer is referring in his talk of Ideas: modes of objectification of will that are expressed in particular individuals in space and time. With the exception of music, the point of art, in Schopenhauer’s scheme, is to facilitate our recognition of these Ideas or templates of objectification of will, and hence, in effect, to give us access to reality. The different forms of art, he argues – again, with the exception of music – are suited to the presentation of different Ideas, and just as the Ideas can be ranked, so to speak, according to the grade of objectification of will of which they are the pattern or prototype, so the various forms of art can be ranked according to the Ideas which it is their particular function to present or express. In short, the higher the grade of objectification of will represented in an Idea, the more valuable, because the more revelatory of the nature of reality, is the art form which presents and expresses that Idea. Poetry (and in particular tragedy), whose subject is the Idea of “man in the connected series of his efforts and actions” (vol. 1, 244), the Idea in which will is manifests itself most clearly, is at the top of the hierarchy; architecture, the artistic purpose of which is to express “some of those Ideas that are the lowest grades of the will’s objectivity,” such as those of “rigidity” and “hardness” (vol. 1, 214), is at the bottom.
Why is it, then, that music “stands quite apart” from this hierarchy, as a “copy” not of the Ideas, but of the will itself”? In the opening pages of Section 52 of The World as Will and Representation, Schopenhauer offers what amounts to a simple – at least in terms of structure – argument in support of his claim. The first premise is that “In [music] we do not recognise the copy, the repetition, of any Idea” (vol. 1, 256). The second is that music is nonetheless “in some sense . . . related to the world as the depiction to the thing depicted, as the copy to the original” (vol. 1, 256). And his conclusion, as we have seen, is that music is “a copy of the will itself.” Understanding Schopenhauer’s view of music depends on understanding why he makes and what he means by each of these statements, and in what follows I attempt to elucidate all three.
What then of the first premise, the assertion that “In [music] we do not recognise the copy, the repetition, of any Idea?” Schopenhauer offers no explicit defence of this claim, but the standard assumption by commentators is that his underlying thought is that music does not “copy” any Idea simply by virtue of the fact that music is not representational: Jerrold Levinson, for example, presents Schopenhauer as holding that “being non-representational, [music] presents for contemplation no Ideas, no perceivable objectifications of willing” (Levinson 1998: 249). However, while at first glance this may appear obvious – for if music is not representational then surely it cannot represent Ideas – the suggestion is nonetheless too quick. The non-representational character of music, the fact that it does not depict phenomena or individual things, would be a sufficient reason for thinking that music’s function as a form of art has nothing to do with the Ideas only if the depiction of phenomena were the sole way of providing epistemic access to the latter. And that, on Schopenhauer’s own theory of art, is not the case.
Section 52 is misleading on this matter. As we have seen, Schopenhauer says there that music “stands quite apart” from all the other arts in not offering us a copy of any Idea, and he says too that the provision of knowledge of the Ideas “by depicting individual things is the aim of all the other arts” (vol. 1, 257). What he suggests, that is, is (a) that music is unique in not offering copies or representations of the Ideas, and (b) that all the forms of art that do trade in Ideas – that is, all the non-musical forms of art – do so by way of “copying” or depiction. But on Schopenhauer’s own account of the non-musical arts, neither of these suggestions is accurate. For the first two forms of art that he discusses, namely, architecture and “hydraulic” art – “the artistic arrangement of water” in landscape design (vol. 1, 217) – he also holds to be non-representational. “Architecture is distinguished from the plastic arts and poetry,” he writes, “by the fact that it gives us not a copy, but the thing itself. Unlike those other arts, it does not repeat the known Idea” (vol. 1, 216–17). Works of architectural (and indeed hydraulic) art present to us things that works of the other, non-musical, forms of art (to the extent that they were concerned with those things at all) would copy, or represent, by depiction. Nonetheless, Schopenhauer suggests, architectural and hydraulic art – despite being non-depictive, non-representational – are continuous with the other non-musical forms of art in being concerned with the Ideas. In architectural art, “the artist simply presents the object to the beholder, and makes the apprehension of the Idea easy for him” (vol. 1, 217), and works of hydraulic art “reveal the Ideas of fluid heavy matter in exactly the same way as the works of architecture unfold the Ideas of rigid matter” (vol. 1, 218).
In short, if architectural and hydraulic art, despite being non-representational, nonetheless function aesthetically in such a way as to facilitate “the apprehension of the Idea,” then an appeal to the fact that music is non-representational gives at best an incomplete explanation of why Schopenhauer holds that music “passes over the Ideas” and hence must function to give us access to “the essence” of things in a way fundamentally different from that in which the other arts do so.
A better explanation is this: the Ideas are essentially templates of the will’s objectification in phenomena, which is to say in configurations of matter. Some of them are expressed in particular configurations of matter (such as the Ideas of particular species of animal); some of them are expressed in any and every configuration of matter – such as those of “the most universal forces of nature” (vol. 1, 130), which it is the business of architectural and hydraulic art to make clear. To recognise an Idea, then, one has either to see it in an actual configuration of matter (as Schopenhauer holds that one does in successful architectural/hydraulic art), or in the representation of a particular configuration of matter (as he holds that one does in successful artistic paintings, for example). Non-representational music, however, can present us with neither representations of configurations of matter, nor with matter itself. Hence music cannot give us epistemic access to the Ideas.
But at this point, a question that has been lurking in the wings becomes unignorable: on what basis is Schopenhauer assuming that music is (at any rate paradigmatically) non-representational? Is it not quite obvious that at least some music just is representational? Schopenhauer – as of course he must – accepts that this is so. However, he suggests, such music
does not express the inner nature of the will itself, but merely imitates its phenomenon inadequately. All really imitative music does this; for example, The Seasons by Haydn, also many passages of his Creation, where phenomena of the world of perception are directly imitated; also in all battle pieces. All this is to be entirely rejected.
(vol. 1, 263–4)
When he speaks of “The inexpressible depth of all music” (vol. 1, 264), then, or makes statements such as “[M]usic . . . is also quite independent of the phenomenal world” (vol. 1, 257), it is clear that Schopenhauer does not mean all music, but good or successful music; it is clear that he is using the term in an evaluative sense rather than a strictly classificatory one. (Indeed, this is true of his talk of the arts in general. As Schopenhauer uses the term, “art” refers to works that are, in his terms, aesthetically successful.)
One worry about any such usage, of course, is that it may be in one way or another question-begging or merely stipulative with regard to what is good or aesthetically successful. However, it is hard to level this charge against Schopenhauer; if he is right that there is music that expresses the very essence of the world, that puts us more closely in touch with the nature of reality than can any other form of art, it is hardly unreasonable to regard this as being music of the very highest value. His objection to representational music requires rather more by way of explanation, however. Representational music is “to be entirely rejected,” he says, inasmuch as it “merely imitates” phenomena, and is “imitation brought about with conscious intention by means of concepts,” as opposed to an expression of the composer’s “immediate knowledge of the inner nature of the world” (vol. 1, 263–4). These remarks reflect Schopenhauer’s view that the primary value of good art is a function of its capacity to reveal something of the nature of reality. Given his idealist metaphysics, this means that the creator of successful art must have transcended the everyday phenomenal world, the structure of which is governed by space, time, and causality, and have glimpsed its underlying reality, the world as will. Schopenhauer holds that this process of transcendence is not something that an individual can deliberately bring about. Furthermore, inasmuch as the world as will cannot be conceived in spatial, temporal or causal terms, Schopenhauer holds that concepts are wholly inadequate to grasping or communicating the nature of the latter, and hence are “eternally barren and unproductive in art.” It follows, then, that music (or indeed painting or sculpture or poetry) that is initiated by “conscious intention by means of concepts” cannot point beyond the phenomenal world, cannot be genuinely revelatory of the nature of reality, and hence cannot be good art (vol. 2, 235).
But must all representational music be of this sort? What rules out the possibility of music that is genuinely revelatory by means of representation? Why should we think that representational music is necessarily “brought about with conscious intention by means of concepts,” so that musical representation may only be of “phenomena of the world of perception” (vol. 1, 264)? The answer, though Schopenhauer never states this explicitly, is that, as we saw earlier, the Ideas are templates of the will’s objectification in configurations of matter, and it is simply not possible to represent the template of a configuration of matter in sound. There is no difficulty in seeing how the sound made by a particular (sort of) configuration of matter may be represented in music – a bird call, say, or (in a more complex sense) the sound of a canon being fired. But whether what is represented is an individual sound event (the opening canon shot of a particular battle, say) or (as Schopenhauer would insist, through abstraction) a kind of sound event, it clearly belongs to the phenomenal world.
However, this is not to say that music that reaches beyond the phenomenal world – music that is genuinely revelatory – is necessarily non-representational. Schopenhauer allows that “we are able to set a poem to music as a song, or a perceptive presentation as a pantomime, or both as an opera” (vol. 1, 263) in such a way that the work is genuinely expressive of will. This will only be successful, however, when the representational aspect of the work remains firmly in a “subordinate position”: “if music tries to stick too closely to the words, and to mould itself according to the events, it is endeavouring to speak a language not its own” – “a great misconception and an utter absurdity” (vol. 1, 261–2). This is presumably what Schopenhauer thinks is wrong with the Haydn oratorios he refers to in the passage quoted earlier. By contrast, “No one has kept so free from this mistake as Rossini; hence his music speaks its own language so distinctly and purely that it requires no words at all, and therefore produces its full effect even when rendered by instruments alone” (vol. 1, 262).
There are at least two points worth remarking on here. First, the coherence of the thought that music that is composed to have a vocal component should, if it is really good music, “produce its full effect” when that component is left out is to say the least questionable. But it may be that this is simply an overstatement of the view that in genuinely revelatory music that is accompanied by words, the relation between the words in question and the music must be very loose. As he writes, the “individual pictures of human life” painted by the words and
set to the universal language of music, are [in good music] never bound to it or correspond to it with absolute necessity, but stand to it only in the relation of a particular example, chosen at random, to a universal concept. . . . Even other examples, just as arbitrarily chosen, of the universal expressed in a poem could correspond in the same degree to the general significance of the melody assigned to this poem; and so the same composition is suitable to many verses.
(vol. 1, 263)
This thought is certainly more intelligible than the view that in good music with a vocal element the latter is dispensable with altogether. But it is nonetheless a very long way short of being persuasive. Recall Bach’s setting of the aria “Mache dich, mein Herze, rein,” in the St. Matthew Passion, for example, and the way that Bach spreads these words – has worked to spread just these words – so sinuously over a melodic line that seems unending. The thought that the words here could be replaced by other (“arbitrarily chosen”!) words that express the same thoughts, leaving “the music” (as if the latter could be isolated from the sung words) somehow untouched, is little short of grotesque. Indeed, only the crudest of analysis could see words and music here as related to each other in such a way that notions of subservience/dominance could have any purchase at all. But if all this is right, Schopenhauer’s theory directs us to conclude, Bach’s setting of the aria must – precisely in virtue of these facts – be deeply flawed. It is hard not to see this as a reductio.
Second, and more positively, if there can be genuinely revelatory music that has representational elements in the form of words or “perceptive scenes” or both (as in opera), we might ask why Schopenhauer does not – or at any rate, does not explicitly – allow that there may also be genuinely revelatory music with representational elements that are musical. Fortunately, there appear to be no theoretical considerations that rule this possibility out, for unless it is a possibility, there would seem to be no way for Schopenhauer to acknowledge as valuable a great deal of music that one might expect him to value (Rossini’s William Tell overture, for example); and no grounds (other than what are – in Schopenhauerian terms – more or less superficial ones, such as technique, complexity and the amount of pleasure provided by each) for valuing Beethoven’s Pastoral Symphony, say, more highly than the music that accompanies Wile E. Coyote’s attempts to catch the Road Runner.
To conclude our consideration of what I identified earlier as the first premise of the argument that structures Schopenhauer’s main discussion of music, then: the fact that “we do not recognise the copy, the repetition, of any Idea” (vol. 1, 256) in music is due not to the fact that music is non-representational, but rather to the fact that the Ideas are simply not the sort of thing that can be represented in sound. Indeed, although Schopenhauer does not explicitly acknowledge the fact, there is no reason that good music – music that is revelatory of the nature of reality – cannot represent aspects of the phenomenal world. All that his theory demands in this regard is that such music not be purely representational in this way – that it should not be, as he characterizes the Haydn oratorios referred to earlier, “really imitative” (vol. 1, 263; my emphasis).
The difference between “representational” and “imitative” is suggestive here, however. For while Schopenhauer is clearly committed to the view that good music cannot be more than superficially imitative of the phenomenal world, he is also committed to the thought that good music is in a different sense representational through and through. The fact that music does not represent Ideas, he suggests, does not imply that it is no more than the purely formal arrangement of sounds, something that can be grasped entirely (as the Pythagorean tradition has it) in logical or mathematical terms: as he says, “we certainly have to look for more than that exercitium arithmeticae occultum nescientis se numerare animi which Leibniz considered it to be” (vol. 1, 256). This brings us to the second premise that I identified earlier: the claim that music is “in some sense . . . related to the world as the depiction to the thing depicted, as the copy to the original” (vol. 1, 256).
It has to be said that the argument that Schopenhauer offers in support of the second premise is not impressive, and Eric Payne’s translation renders it even less so:
That in some sense music must be related to the world as the depiction to the thing depicted, as the copy to the original, we can infer from the analogy with the remaining arts, to all of which this character is peculiar; from their effect on us, it can be inferred that that of music is on the whole of the same nature, only stronger, more rapid, more necessary and infallible.
(vol. 1, 256)
This is confused. The argument that Schopenhauer actually has in mind, I take it, is in essence as follows: (1) the non-musical arts are related to the world “as the copy to the original”; (2) music’s effect on us is “on the whole of the same nature” (albeit “stronger, more rapid, more necessary and infallible”) as the effect on us of the non-musical arts; therefore (3) music is related to the world “as the copy to the original.” However, while this is more coherent than Payne’s rendition of the argument, as it stands it is hardly compelling. To strengthen the argument, much more would have to be said about both the extent and the relevance of the analogy cited in (2); in particular, more would have to be said about the effect of music on us. And what Schopenhauer does say about the latter in effect renders the argument by analogy redundant. His thought is that we simply do not experience (good) music as merely a formal arrangement of sounds; or, rather, we experience it as something more than this, as something that is somehow getting at – putting us in touch with – something deep, something of profound significance:
[I]t is such a great and exceedingly fine art, its effect on man’s inner nature is so powerful, and it is so completely and profoundly understood by him in his innermost being as an entirely universal language, whose distinctness surpasses even that of the world of perception itself . . . [and in] which we see the deepest recesses of our nature find expression. Therefore . . . we must attribute to music a far more serious and profound significance that refers to the innermost being of the world and of our own self.
(vol. 1, 256)
It is clear – and this is no criticism – that (despite the “therefore”) there is no argument in all this; Schopenhauer is simply appealing to our experience of (in his terms, good) music. To a person who has not had (or at least been persuaded that others have had) the sort of experience that he refers to, the appeal will be unsuccessful, of course; but then so will be the argument by analogy, since fleshing out the second premise of the latter in Schopenhauerian terms will in effect involve appeal to just the sort of thing he appeals to in the passage quoted above. But for anyone whose experience of music is consonant with that described by Schopenhauer, the argument by analogy will be simply redundant; anyone in this position will need no further persuasion that music is not merely a formal arrangement of sounds, that it in some sense or other refers or points to something beyond itself.
But refer or point to what? Schopenhauer’s argument thus far has been that music cannot represent the Ideas, but that nonetheless we experience (good) music as somehow revelatory of the “innermost being” of things. Music must, therefore, he concludes, be a representation or “copy” of “the will itself” (vol. 1, 257).
As Schopenhauer acknowledges, this conclusion is deeply problematic, for his own metaphysics dictates that the will – noumenal reality – can “never be directly represented.” Hence, as he says, “it is essentially impossible to demonstrate” that music represents the will. Nonetheless, he suggests, his conclusion “is quite sufficient for me, . . . and will be just as illuminating also to the man who has followed me thus far, and has agreed with my view of the world” (vol. 1, 256). But this is wholly unconvincing. The fact of the matter is that either Schopenhauer’s metaphysics, holding as it does that the will cannot be object of representation, falsifies his conclusion regarding the nature of music, or that conclusion falsifies his metaphysics – and indeed his theory of music, since that is based on his metaphysics.
So what has gone wrong? Why does Schopenhauer insist that music is “a copy of the will itself,” despite being committed to the position that the will cannot be represented? At least part of the reason is that he is hugely over-impressed by a range of what he calls “analogies” between music and the Ideas, which is to say between music and the phenomenal world. With one exception – of which more in a moment – these range from the fanciful (e.g. “the definite intervals of the scale are parallel to the definite grades of the will’s objectification, the definite species in nature” (vol. 1, 258)) to the ludicrous (e.g. “impure discords, giving no definite interval, can be compared to the monstrous abortions between two species of animals, or between man and animal” (vol. 1, 259)). Only someone as immersed in Schopenhauerian metaphysics, as committed to the revelatory power of music, and as determined to show that each explain and confirm the other as Schopenhauer himself could find them compelling. And Schopenhauer clearly does find them compelling. Although he does not explicitly express it in this way, his thought seems to go something like this: “Given that the phenomenal world is the objectification or expression of the will, and that music in so many respects mirrors or parallels aspects of that world, surely music too must be an objectification or expression of the will.”
However, his final “analogy” between music and the Ideas suggests an alternative to this conclusion – and an indication that in the argument that he does explicitly offer for it (“music does not represent the Ideas; it is nonetheless revelatory of reality; music must therefore represent the will”) Schopenhauer simply commits the fallacy of false dilemma. The analogy in question is this: “Finally, in the melody . . . I recognise the highest grade of the will’s objectification, the intellectual life and endeavour of man.” In particular, he suggests, melody “relates the most secret history of the intellectually enlightened will, portrays every agitation, every effort, every movement of the will, everything which the faculty of reason summarizes under the wide and negative concept of feeling” (vol. 1, 259).
It is important to note that the will referred to in the latter quotation is not the will that we – and Schopenhauer – have been referring to thus far. Here, “will” refers not to noumenal reality, but to the human will. And a strong case can be made for the view that Schopenhauer ought to have restricted himself to the thought that music represents some of the most fundamental aspects of the human will, or of human willing, rather than making the much more ambitious claim that “music expresses . . . the inner being, the in-itself, of the world” (vol. 1, 264). For one thing, all that he has to offer by way of justification of the latter claim is the series of analogies referred to above, none of which are remotely compelling. For another, the less ambitious claim has the significant advantage of not rendering his theory incoherent, for there is nothing in his metaphysics that entails that the human will cannot be object of representation. Finally, the idea that melody “relates the most secret history of the intellectually enlightened will” promises to deliver as much as Schopenhauer could reasonably want or is entitled to by way of content for the idea that music can be a source of profound insight. Indeed, in some of his remarks about the cognitive significance of music, this idea seems to be all that he has in mind; for example, the suggestions that “The inexpressible depth of all music . . . is due to the fact that it reproduces all the emotions of our innermost being” (vol. 1, 264), and that the significance of music refers to the “innermost being” of “our own self,” that in it “the deepest recesses of our nature find expression” (vol. 1, 256). In the end, then, the most charitable way of understanding Schopenhauer’s theory of music is to discard what many have seen as its most distinctive feature – the thought that music is somehow a direct expression of the ultimate foundation of reality – and to regard it instead as a distinctive version of expression theory.
See also Expression theories (Chapter 19), Nietzsche (Chapter 32), and Wagner (Chapter 35).
Goehr, L. (1996) “Schopenhauer and the Musicians: An Inquiry into the Sounds of Silence and the Limits of Philosophizing about Music,” in D. Jacquette (ed.) Schopenhauer, Philosophy, and the Arts, Cambridge: Cambridge University Press, pp. 200–28.
Levinson, J. (1998) “Schopenhauer, Arthur,” in M. Kelly (ed.) Encyclopedia of Aesthetics, vol. 4, New York: Oxford University Press, pp. 245–50.
Schopenhauer, A. (1969 [1818, 1844]) The World as Will and Representation, 2 vols, trans. E.F.J. Payne, New York: Dover.
Budd, M. (1985) Music and the Emotions: The Philosophical Theories, London: Routledge & Kegan Paul. (Ch. 5 critiques Schopenhauer as an expression theorist.)
Kivy, P. (1997) Philosophies of Arts: An Essay in Differences, New York: Cambridge University Press. (Ch. 7 critiques Schopenhauer’s philosophy of music in the course of an argument to the effect that absolute music lacks content.)
Magee, B. (1997) The Philosophy of Schopenhauer, rev. and enlarged, Oxford: Oxford University Press. (Ch. 8 explores Schopenhauer’s philosophy of music within the context of his broader philosophy of art.)
Tanner, M. (1999) Schopenhauer, London: Routledge. (Explores Schopenhauer’s philosophy of music within the context of his broader philosophy of art.)
Young, J. (2005) Schopenhauer, New York: Routledge. (Chs. 5 and 6 explore Schopenhauer’s philosophy of music within the context of his broader philosophy of art.)