Language and Music


Language and Music



What kind of phenomenon is language? I shall consider this question by comparing language with music—though I shall also be concerned with what kind of phenomenon music is, especially toward the end. Language occupies a curious no-man’s land, being neither purely external not purely internal.[1] We can talk to others out loud or we can talk to ourselves silently; and we can talk in a semi-whisper to no one in particular. It is sometimes hard to tell whether an act of speech is mental or physical: is your larynx moving slightly as you talk to yourself? You can make what feels like a smooth transition from inner to outer speech with no abrupt shift from inner to outer. You can also write words down as well as speak them, and this writing can take several forms. Audition need not be involved. It is hard to tell whether one’s inner speech is auditory or modality-neutral; for deaf people it is not likely to consist of auditory imagery. So language can manifest itself in many forms; it is not tied to a single medium or sensorimotor capacity. One might be tempted to call it a psychophysical phenomenon in an effort to capture its protean character, but that presupposes a kind of dualism better avoided (we don’t want to be procrustean about the protean). We could say multiform or variegated, though that doesn’t offer much in the way of illuminating description. What we can assert is that it would be quite wrong to assimilate language to a type of physical behavior, as it might be the sounds produced by the vocal apparatus; there is much more to language than that, and also much less.

Here music affords a helpful analogy: where does music exist? We can play an instrument or sing out loud, or we can quietly hum a tune to ourselves, or we can silently rehearse a melody in the mind. It may not be clear to us whether the music is purely within us or engages our motor faculties: were you minutely flexing your fingers to tap out that imagined rhythm, did your vocal cord contract slightly when you inwardly reached for a high note? The inner-outer dichotomy seems insufficient to capture the full range of musical expression: some music is more outer than other music, some more inner. And which is basic? Did music exist in our minds before we ever gave public expression to it? Music, like language, occurs in many forms, which we clumsily describe with words like “psychological” and “physical”; in reality it is fluid and various, seemingly indifferent to its contingent vehicle. It is possible to practice one’s singing without making a sound, just by inner note production aided by an active vocal cord. To identify music with external musical behavior would be hopelessly reductionist. Where does music exist? Everywhere and nowhere: it slips and slides from one domain to another, varying within the sensorimotor system and sometimes detaching itself from that system altogether. I would be inclined to say that it is primarily mental, as I would be inclined to say the same of language—though that too risks the procrustean. Music and language are creatures of many guises, many exemplifications.

This is partly because they are both essentially abstract, mathematical even. Language consists of a finite array of digital elements that can combine by rules to produce an infinite number of complex structures—a kind of computational system. As such it can take many forms without losing its identity: wherever the abstract system exists there is language–in the head, in the air, on paper, in neon. It is a form beneath a substance—a pattern not a stuff. Someone could be operating with language and be completely different from us mentally and physically, so long as they instantiate the abstract pattern of language: a combinatorial system equipped with syntax and semantics (a bit like a computer program, as is often said). But music is very similar, only it uses a different kind of abstract mathematical structure. Music is analogue not digital: it uses a continuous magnitude (pitch) and divides it into discrete units. Thus we obtain scales and keys by segmenting the sound medium in various ways: tones, semitones, named notes, octaves, etc. The major scale, say, is just a mathematical construct defined by using intervallic structure: it can be reproduced on any instrument permitting variations of pitch (which itself is just frequency of vibration). Rhythm is the same: intervals of time (not pitch intervals) punctuated by discrete sounds. Musical notation represents melody and rhythm by means of geometry: the placing of a mark in a higher or lower position on the stave, and the use of differently shaped marks for note lengths as well as distance along the stave. The abstract structure can exist in an instrument or in a voice or purely in the head. Without such a structure no music can exist (just noise), and where the structure exists so too does music.

Accordingly, when we hear language spoken we are hearing an abstract structure with certain formal properties (discrete infinity being central), and when we hear music being played we also hear an abstract structure with certain formal properties (a continuous infinity segmented according to recognizable rules). Breaking the rules destroys language and destroys music (unless another set of rules is set up): random order is the enemy of both language and music. Linguistic theory studies the form of these rules, as music theory studies the form of musical rules. In both cases it is an abstract structure that constitutes the phenomenon not its contingent vehicle: the same tune can be played on a piano or a trumpet or a human voice (outer or inner), as the same sentence can be spoken or written down or rehearsed inwardly. We hear sequences of elements in time that have been constructed according to certain rules and which admit of multiple realizations. We should not confuse the thing itself with its passing manifestations. The processing mind must be attuned to a formal structure as well as to a type of physical stimulus. The underlying psychology is mathematical.

We also have a competence-performance distinction in both cases: the competence can exist when performance is lacking through injury or fatigue. The competence has a heavy cognitive component, though it also contains motor instructions; by no means is it merely a matter of stimulus and response. Learning is involved in both cases, easier in childhood than adulthood, and no doubt there is an innate component (a gene for language, a gene for music). The two can sometimes converge as in song: the language faculty must mesh with the musical faculty to produce words with melody. Discrete digital infinity meets continuous analogue infinity. Speech rhythms interact with musical rhythms. Is this why song is so universally appreciated? It combines two of our most impressive attributes (sadly lacking in other animals that have no appetite for song). When you listen to a song your language faculty is fully engaged, with all its productive power, and in addition your musical faculty is fully engaged, with its capacity to resonate to melody and rhythm. Even a mediocre singer is operating at a very high cognitive and motor level; aliens might marvel at our routine ability to sing a song, thinking what superb mathematicians we must be in order to wield the abstract structures involved simultaneously. Also, language and music are regarded as expressive of thought and emotion. Music can use language this way, as in song, but language can use the resources of music too, as in variations of pitch and volume, or staccato utterance. The two faculties are not completely disjoined in practice, despite the distinction of modules. It would be wrong to say that language is a type of music or that music is a type of language—they embody quite different constitutive principles—but at a more general level the affinities are clear. Abstract mathematical structure with multiple realizations—a kind of fluid fixity, disciplined variety, creatively rule-governed. A twenty-six-letter alphabet, a twelve-tone scale—but all that coiled creativity! This is the power of combination applied to an economical base. In both cases, too, a lot happens behind the scenes: we don’t consciously apprehend what music theory and theoretical linguistics reveal—these belong to the unconscious mechanisms that enable us to hear the sound stimulus as exemplifying music and language. The brain computes and the mind enjoys, or at least understands. Teachers instruct you in the rules of grammar and musical composition, often bafflingly, but your unconscious was onto to this stuff long ago.

So we see an affinity between language and music, but there is one difference that stands out and threatens to undermine the analogy, viz. that there is no such thing as sign music. The deaf can use sign language, thus demonstrating the independence of language and the auditory medium, but no one “listens” to sign music, thus demonstrating the independence of music and the auditory medium.[2] Music seems more closely tied to sound than language is: sound strikes us as essential to music. How then can music consist in an abstract structure that is only contingently exemplified in an audible medium? So is music not like language after all? Is it a modality-specific phenomenon? These are good questions, but I think they have an answer—namely, that it is only contingent that non-auditory music doesn’t exist. Consider rhythm first: we usually register rhythm through our ears, but it is possible to register it through touch. We can feel pulses of rhythm in the air or by direct physical contact with the body, say by tapping; sometimes we feel our heart beating. It is imaginable that someone might develop this type of perception further and even come to enjoy the pleasures of touch that are involved. In the case of melody we need an analogue of pitch; here colors might do the trick, with blues for low notes and reds for high notes, along with degrees of saturation. We could map pitch differences onto color differences and produce sequences of colors with the same intervallic structure as pitches. Then an identical abstract structure would exist in the visual sense as now exists in the auditory sense. A composition could combine color presentations for melody with tactile stimuli for rhythm. Wouldn’t that be music? Suppose we came across an alien species that was constitutionally deaf yet musically inclined. They might have an elaborate musical culture built around vision and touch. We can suppose that the same emotional associations exist for them with respect to sights and feels as exist for us with respect to sounds. They have an advanced technology that enables them to combine visual and tactile stimuli into musical compositions that resemble our compositions—perhaps a body suit with a visual headset. They have never experienced or even heard of audible music and just assume that everyone consumes music as they do. They resonate to an abstract musical structure defined by an analogue of our pitch relations. A song is a series of visual and tactile stimuli that combine language and melody (defined by visual variations). Nearer to home, imagine that synesthesia is common, so that colors regularly evoke sound images in the mind: a musical performance consists of a sequence of visual stimuli that are known to evoke specific auditory images. The spectator gazes at the array of shapes and colors and experiences sounds inwardly, so that vision and sound blend together. Wouldn’t that be music? And not just because an auditory component is involved—the visual component would also be musical. Suppose the synesthesia faded away so that only the visual element was left: would that mean that these people no longer possess a musical sensibility? Music is certainly not tied to external auditory perception, since we can hear music inwardly, so why should it be tied to auditory phenomenology at all?[3] Poetry isn’t necessarily tied to hearing, given that there could be poetry in the form of sign language, so why should music be? It is true that we find it hard to imagine non-auditory music, but that tells us little about what is really possible; maybe we can’t fully grasp the musical phenomenology of the deaf aliens, but that doesn’t mean there isn’t one. Isn’t it just parochialism to suppose that music must exist for others in the form it exists for us? Maybe there is a Beethoven on another planet who composes purely in colors and whose symphonies are much prized; maybe even tastes and smells come into it! After all, music is based on intervallic structure, but that is not peculiar to sound. Or there might be music lovers possessed of another sense entirely, say one responsive to electrical fields: music is composed by means of varying electrical currents. Deep feelings might be aroused thereby. Just as language is a matter of patterns, so music is: and the patterns can exist in unlimited forms, i.e. types of sensory phenomenology. Imagine if there were a kind of inverted spectrum for sounds whereby a low frequency sound is heard by some people as a high note and by other people as a low note. Does that mean that music would be different for the different kinds of hearer? It’s the structure that counts not the absolute quality of the sound. Generalizing, we can move from this case to the synesthesia case to the alien case—all bona fide consumers of music. So it turns out that music and language are both inherently modality-neutral, contrary to initial expectations (it was difficult for hearing speakers to understand at first that deaf people speak a language just like theirs). For us music is inextricable from sound, but not so for all possible musical beings. On some planets they sing the blues using the color blue.[4]


[1] I could as well say speech, focusing on the performance aspect of language; in a deeper sense language might be identified with an underlying competence or cognitive structure. But speech itself exhibits the kind of variety I am alluding to before we even get round to examining the cognitive foundations.

[2] I am told that this is not entirely true—that there are musical compositions for the deaf that employ visual and tactile stimuli. Great: but they don’t seem to have much traction or scope, and are not appreciated by the hearing community. Still, this fact is grist to my mill, as will become apparent.

[3] Another kind of case to consider would be blindsight music, i.e. sound waves reach the ears but no conscious auditory response is produced yet there is a psychological effect on the hearer. We could describe these people as hearing music but not consciously. Evidence for this would be the occurrence of effects in these hearers like those experienced by ordinary hearers, e.g. improved mood or knowledge of the tune that was presented. People with this condition might well listen to music though their conscious mind shows no awareness of it: the musical response would occur in the non-conscious part of their mind.

[4] We can enquire whether other art forms are musical, at least partially. Dance is a good candidate: there is surely something musical about dance, which is why music and dance go so regularly together, and we might conjecture that it is because dance embodies in another medium some of the structure of music—rhythm certainly, but also something analogous to melody (segmented continuous movement in space). A machine might be hooked up to a dancer that converts movement into melodic sound, and musical performances produced thereby. Even pictorial art might be seen as having a musical aspect in the form of harmony and a fixed palette of “notes”. The mathematical forms of music can in principle be transposed to other artistic mediums. If so, music has already exceeded its conventional sonic bounds.


One response to “Language and Music”

  1. jeffrey g kessen says:

    That’s a good piece. Dug it….(it would be a very ingenious—and charitable—machine that could convert my dance movements into melodic sound).

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.