WMA Forum 2005

On Articulation and Truth in Music

Peter Dayan, (University of Edinburgh, Scotland)

The conference papers and publications I have produced over the past six years have been readings of a number of well-known French literary and theoretical texts, aimed at extracting their ideas concerning the relationship between music and literature; particularly, I have been trying to show how the concept of music is used to define literature, and vice versa. More than once, and quite justifiably, I have been taken to task for failing to distinguish rigorously between those ideas which (in traditional academic style) I was attempting to situate in their historical and ideological context, and general truths about the nature of music. I have not previously had the courage to speak directly about such general truths; how can one do so as an academic, given the enormous, indeed the insupportable weight of previous thought and commentary which bears at least indirectly on the subject which, so my professional conscience tells me, I would have to take into account, but which, even (or perhaps especially) if I were not ignorant of a large part of it, would, it seems to me, inevitably bury my own slight insights? At the WMA Berlin conference in 2003, I took advantage of what I saw as the particular character of the newly instituted WMA Forum: I allowed myself to speak, as it were, in the first person, in absolutely general terms, about music, language, and literature, without a single reference to any pre-texts or other authorities. I am well aware that one should not expect to get away with such things often, but I hope I may be forgiven just this once.

How do we ascribe meaning to words? I think it would be fairly uncontroversial, from the point of view of a linguist or of a literary theorist, to say that the process is necessarily a double articulation: in the realm of the signifier, and in the realm of the signified.

On the one hand, in the realm of the signifier, we distinguish a given word from other words in the language by its physical form, which is different from the physical form of other words. The word ‘song’ is physically different, in its phonemes or its letters, from all other words in English. In that sense, on the level of the signifier, each word acquires its individuality by being articulated against the other words in the signifying system. Each word is also articulated against other words in the syntactic system, which gives the word a grammatical status that helps us to identify it.

But on the other hand—the meaning of a word does not arise solely from its articulation against other words as signifiers, or from its syntax. The limits of a word’s meaning, indeed the limits of a word, are also defined by the articulation of its signified against other possible signifieds. The clearest evidence of this is contained in the existence of words like ‘bark’ which, to the user of the language, are actually two different words with exactly the same sound, spelling, and grammatical function, therefore indistinguishable as signifiers, but indisputably separated by their signifieds; and in the sheer possibility of translation. Whether I say ‘crow’, ‘Krähe’, or ‘corbeau’, I am referring to the same thing, a thing that we all believe exists outside language, and is of a different species from the blackbird. The meaning of the word ‘crow’ in English is limited in common parlance by the general size and shape of the bird, and in cultured circles also by distinctions from related species. The meaning, in other words, depends not just on the physical characteristics of the word, but on the physical characteristics of its referent—or rather, on our perception of its referent. And that is what is missing in music.

If music has meaning analogous to linguistic meaning, what might be the equivalent, in music, of a word? Could it be a note? or is a note more comparable with a letter, or a phoneme? But any such parallel quickly turns out to be untenable. There is no equivalent of a word in music because, as we have seen, the definition of a word is (partly but inescapably) dependent on its relation to a signified articulated against other signifieds, in a realm of signifieds which appears to us to exist independently of signifiers (as crows exist independently of the word that names them); and there is nothing in music which refers us to such a second articulation. Herein lies the essential difference between music and language.

I can translate the word ‘crow’ into various languages without difficulty. As I do so, the physical substance of the word, the letters of the word, may change entirely; but its signified meaning does not change with its physical substance because signifieds necessarily appear to us to exist as of right independently of the substance of the signifier. Crows would exist and would not interbreed with rooks whatever we called them; that which we call a rose by any other name would smell as sweet. The meaning of a note of music, on the contrary, is inseparable from its physical substance. It is not obviously or clearly defined by its articulation in any realm other than the realm of music; and if we change its physical substance so that it becomes unrecognizable, then we change its meaning. It is in this sense that music cannot be translated; one cannot transform the musical signifier without also transforming the meaning of the music. To give a gross illustration: let us say that an English composer wrote a piece of programme music about a bag, and introduced the motif ‘BAG’, as three notes of music, into the piece. If I were explaining this to a French person, I could translate the word ‘bag’ as ‘sac’. But how could one translate the motif? One could not change the notes to ‘SAC’ without completely changing the music—and the music’s meaning.

It is a commonplace in writing about music to talk of music as a translation of an impression, or a mood, or a poem, or a story into sound. The complications of this metaphor are confusing and interesting enough. But what I want to point out here is that among critical commonplaces about music one finds no real equivalent of the notion of translation between verbal languages, between English and French or Chinese and German. And that is simply because musical meaning does not appear articulated within an organised realm of signifieds, in the way that language is; in music, there is no arbitrary yet fixed conventional relationship between signifier and signified, as there is between the word ‘crow’ and what we believe to be the real bird. Because the musical signifier does not appear arbitrarily linked to a signified, we perceive the signifier itself as inseparable from its meaning. Musical meaning is generated by articulation within the realm of music, rather than in a realm outside music (as crows are outside language); therefore, the notes of music are irreplaceable.

For that reason, music does not have the same relationship to truth that language does. Truth in language involves correspondence between what is said in words and what appears to exist in the world outside words, and is mediated by our conviction that the link between the two realms, though arbitrary, is fixed by convention. If I see a blackbird and tell you it is a crow, I am not telling the truth. If I tell a German person that a crow, in German, is ‘Krähe’, I am telling the truth; if I say it is ‘Amsel’, I am not telling the truth. But this dynamic of truth and falsehood is missing from music because, once again, musical meaning is not articulated in the realm of the signified.

Since the idea of a certain divorce between music and signification began to be mooted in the nineteenth century, there has been a standard objection to it: music, say the objectors, certainly can refer; and the example normally given is that of the unmistakeable cuckoo in Beethoven’s Pastoral Symphony. But this objection completely misses the point. Beethoven’s cuckoo is not linked to the real cuckoo by an arbitrary convention. It actually sounds like a real cuckoo. Of course, the word ‘cuckoo’ also sounds like a real cuckoo. But in language, such onomatopoeic effects are not necessary to evoke a signified. In Scots, the same bird is called a ‘gowk’, which doesn’t sound at all the same, but refers to the same animal. If one changed the sound of Beethoven’s bird by as much as the word ‘cuckoo’ differs from the word ‘gowk’, one would simply lose the cuckoo entirely. Which confirms the general theoretical point. Certainly, it is possible for words, like music, to imitate that to which they refer; however, this tends, in practice, to be a marginal phenomenon, in music as in language. More often, language appears composed of arbitrary signifiers each of which is clearly linked to a signified, and is to that extent translatable; while music appears composed of signifiers each of which is not clearly linked to a signified, and to that extent is untranslatable.

So far, I have spoken in absolutely general terms about language and music. But I have to confess that to me, everything I’ve said is merely laying the groundwork for the really interesting questions which arise at the points where the distinctions I have made begin to break down; and those points, not surprisingly, are located precisely where the question of what constitutes literature emerges. One notion that poets and literary theorists seem to have found more and more difficult to get away from over the past couple of centuries is that poetry has a certain untranslatable element, because there is something in the words of poetry that is not to be understood solely by appreciating the articulation of their meaning in the realm of the signified. Poetry doesn’t just operate on the level of the kind of truth-telling I have been talking about. It also seems to operate within the level of the articulation between signifiers; just as music does. Hence the more ‘musical’ the poetry is, the harder it is to translate.

I have been talking about a kind of truth that exists in language by virtue of its double articulation, but that doesn’t exist in music, because of its single articulation. What I haven’t begun to talk about here is what kind of truth music might contain. I hope, however, that I have managed to suggest it might be useful to look at the parallel between the truth of poetry (as a discourse that refuses the dual articulation), and the truth of music. So the nature of that musical truth is, I think, best analysed through the way that poets, musicians, and critical theorists compare the distinctive nature of poetry with that of music. That enterprise is immense and fascinating. At its heart is the unavoidable question: are there, in fact, any analysable general truths, of any order, to be found about the nature of music or of poetry? I have to say that as far as I can tell, most academics today would answer no to that question, if asked. However, there remains an obstinate sense that perhaps we ought to be able to answer yes; yes, there is a truth about music in music, and about poetry in poetry. But, as far as I can tell, the truth of music, like the truth of poetry, can only be held to exist for as long as we can maintain the possibility that, at one level, it remains unarticulated.

Return to the top