… and Why People Keep Saying It Is.
If it sounds profound, maybe that’s because we know it’s not true. Not literally anyhow. What may give it a ring of truth is hard to say exactly. It promises deeper understanding, but conceals its own obscurity.
Taken literally, music is (a) language is evidently nonsense. We don’t speak music. We don’t order a sandwich, arrange a date, or discuss the meaning of music in music. We may deliver a telegram singing, or decide to have a conversation as if we were performing an opera, but the message is in the words, not in the music.
A language is something like English, Chinese or Swahili: some locally grown variety of the communicative system that all humans share, language. In principle, anything said in one language is translatable into another. Try replacing any piece of music into English or Dutch. Or to replace it with another piece of music that sounds very different but means more or less the same. It’s a pointless assignment.
If the slogan nevertheless seems to make sense, it might be metaphorical. Music is similar to (a) language. Just like body language is a little bit like language, or according to another popular slogan, fashion is a language. In this case, clothes are used to make a statement, or rather, express an attitude (I’m cool. I’m macho. I’m retro. I don’t care about how I look). People emit signals by the way they’re dressed. In that sense, clothes take on a communicative function. But their primary function is to clothe us; communication is secondary, and anyhow, very limited. That people conduct a conversation by their attire is maybe imaginable, but certainly rare and above all highly unpractical.
Music is no doubt often (but not always) communicative and expressive, in a far richer sense than clothing. But to say that music is therefore (metaphorically) a language is missing an important point. An effective metaphor is a link between two things that are in all other respects strikingly dissimilar (language and fashion; a black sheep and a disliked person). Treated as a metaphor, music is a language obscures the fact that in interesting ways music is substantially similar to language.
“Language of Feeling”
Spoken language has rich expressive qualities. Tone of voice gives it a quasi-musical quality (prosody) that may come close to singing. It is this observation that originally motivated calling music the language of feeling or of the emotions; even the universal language of feeling. This was an eighteenth-century novelty that rapidly became a cliché. When Immanuel Kant proposed this he was not saying anything new:
Every expression of speech has as a coherent whole an intonation, appropriate to its meaning. This intonation indicates more or less an affect of the speaker, which it produces also in the hearer […]. Just as this modulation is like a universal language of feelings which every man may understand, music employs this independently and in its full force, viz. as a language of the affects […].
Immanuel Kant, Kritik der Urteilskraft §53 (1790) (my transl.)
Look again at the picture above (after Chodowiecki’s Natural and Affected Behaviours, 1779). That the two art connoisseurs talk in musical phrases may illustrate the absurdity of musical conversation. But we may also observe how easily these bits of melody may be associated with text. For instance: How beautiful! – How marvellous! It is the shape of the musical phrases that allows us to associate them with a specific text, because this shape itself is speech-like.
Grammatical and Ungrammatical Music
It is common in (classical-romantic) music theory to speak of musical phrases, sentences, and periods, even in instrumental music. That these terms have been borrowed from linguistic grammar implies an analogy. For eighteenth-century theorists this analogy was very much alive; their modern successors have managed to ignore it to an astonishing degree. Without being able to banish it entirely. This observation by Theodor W. Adorno could have been written in 1780:
Traditional theory of musical form distinguishes sentence, phrase, period, punctuation; question, exclamation, parenthesis; subordinate clauses are found everywhere, voices are raised and lowered, and in all of this music borrows its bearing from the voice that speaks.
Th. W. Adorno, Fragment über Musik und Sprache (1956) (my transl.)
The affinity between speech and music signalled by Kant and Adorno is fairly obvious. There is however another, more hidden aspect of the analogy between music and language.
What gives musical sentences their coherence is something similar to what in language is known as syntax: the right forms of the right kind of words, in the right order. John looks tired is a sentence, tired John looking is not.
The workings of syntax are most easily observed in sentences that are ill-formed. Here’s a rather silly but correct little piece of music in mid-eighteenth century style:
When in music the syntax goes wrong, it will sound “ungrammatical”. Here’s nearly the same piece, with the syntax mangled:
I must admit that inventing this example of broken musical syntax was not as easy as coming up with tired John looking. Not because the music is complex, but because it is hard to decide what exactly to do wrong, what properly belongs to musical syntax.
As you may hear, I left the melody more or less intact. Melody (including rhythm) most closely resembles the prosody of language. What gives music its formal coherence seems to be primarily the succession of chords (harmony). This is where the “ungrammatical” version breaks the rules.
In language we may distinguish syntax (rules) from the lexicon (words) and semantics (meaning). The same syntactic pattern allows us to produce sentences with very different meaning: John looks tired and Roses smell sweet have the same form. In music we may preserve the harmonic structure while altering everything else, drastically changing its character or expression. This is precisely what happens in classical variations and traditional jazz improvisation.
But because in music there is nothing like words, the distinction between syntax and all the rest is not that clear. The analogy between music and language is substantial and significant, but limited all the same.
Making and Breaking the Rules
This harmonic syntax has been developed within the tradition of European music since the sixteenth century. With the spread of Western culture it has become a worldwide standard. But its nearly global dominance doesn’t mean that it’s universal or natural. Presumably our syntax-parsing abilities are able to handle a broad variety of syntaxes on the basis of certain elementary principles. This suspicion is confirmed in increasing detail by neurological research. There is an intriguing conformity in patterns of brain activation when we’re occupied in musical and linguistic-syntactic tasks.
Some listeners may prefer the ungrammatical piece to the grammatical one – precisely because it breaks the rules in unpredictable ways! We might say that it obeys new rules that are invented ad hoc. This does not alter the fact that according to well-established standards, the piece is simply wrong. With sufficient exposure to “regular” music, we identify the mistakes, even if our aesthetic preference calls them “right”.