The relationship between music and language continues to be the source of considerable debate. On the scientific front, there have been studies that demonstrate that our brains process music in a manner similar to the way in which we process language, while other studies demonstrate that different parts of are brains are involved in the processing of music and language (thus allowing victims of brain damage to retain their ability with respect to one while losing their capacity for the other). Lucy Patston, a Ph.D. student at the University of Auckland, conducted research that suggests that musicians hear music as a language while non-musicians do not.
On an aesthetic level, critics have nearly always referred to music as a language in at least the qualified sense that it is an act of communication based to some degree upon a shared repertory of meaning-units (gestures, harmonies, tonal relationships). In this sense, painting is also a language. And yet music, unlike painting, requires the temporal unfolding that language demands in making sense. Think about the structure of a joke as an extreme example of language’s necessary temporal grounding. The joke is strongly oriented toward its conclusion (the punch line). Once the end is reached, our understanding of all of the previous statements leading up to the punch line is altered with respect to the end.
Indeed, that is the joy of a good joke; we experience a belated clarification of statements that we had processed without complete understanding. The joke illuminates in retrospect. Music functions in a similar manner. We may expect (indeed we must have some expectation if we are closely following the music) a musical phrase to “go” a certain way, but the music may either fulfill or contradict those expectations (and furthermore may fulfill or contradict those expectations in a variety of ways). The manner in which a phrase concludes alters or confirms our understanding of the opening of the phrase; we always listen in two directions: forward with respect to expectation but backward with respect to understanding.
However, the most contested site for examining the relationship between music and language is, not surprisingly, the realm of linguistics. Most linguists (or at least those espousing a sort of common-sense linguistics) attempt to quash the notion that music is significantly language-like by insisting that while music shares with language a strong reliance upon syntax, it lacks a prominent semantics. Indeed, some who oppose comparisons between language and music go so far as to insist dogmatically that music has no semantics. This claim warrants some investigation. Syntax accounts for the ordering of a statement (and here we shall use the term “statement” in a broad enough sense to encapsulate both linguistic and musical utterances); more precisely syntax concerns the patterned regularities that determine the combination of words into a proper sequence. We can easily abstract away from a statement in order to reveal its underlying syntax.
For instance, in the sentence, “A man is rational”, we can abstract away from the actual terms to reveal an underlying syntax of S is P (where “S” is a placeholder for the subject of the sentence and “P” is similarly a placeholder for the predicate). What items we insert as the subject and the predicate do not matter (although we may produce sentences that are either true or false depending on what we insert—but that is a matter of semantics); the structure itself is recognizable.
Certainly, music has syntax. Limiting our purview for now to tonal music (I shall address atonal music further on), we expect a certain sense of order to emerge from musical utterances. Cadences come at the end to close off musical phrases and larger sections; the stronger the cadence, the more conclusive it seems and therefore the more suitable in fulfilling its form-defining function. Indeed, just as we saw in verbal language, we can abstract away from the specifics of a phrase in order to reveal its underlying syntax. This is a common approach within so-called formal analysis. A sonata form has the arrangement of Exposition-Development-Recapitulation as its syntax while the phrase type known as a “musical sentence” has the arrangement Presentation-Continuation-Cadence as its syntax.
Semantics, on the other hand, accounts for meaning. In studying semantics, we are pursuing the manner in which linguistic statements point away from the structure of language itself in order to postulate something about the outside (extra-linguistic) world. For instance, the sentence “The moon is a smelly, drunk man” is syntactically sound but it is clearly false. Nothing in the structure of the sentence tells us that it is false. We merely know on account of our experience of the world that the moon is not drunk nor is it a man. Rather the moon is a satellite of the earth composed of rock.
Now, music seems to lack semantics, at least in this strong sense. The quip is often adduced that no one has ever composed music that says, “Go get me a soda.” Imagine if someone had. What would ensue? Either a room full of concertgoers would get up to get the music a soda or else they ought to feel guilty for disobeying a directive, right? Clearly not. If we were attending a play and a character said to the audience, “Get me a soda”, we would not understand the sentence as a directive but rather we would consider it within another register (metaphorical perhaps or spoken to an unseen character beyond the “fourth wall”). This is the first point to be made and it is hardly trivial.
Insofar as music may have semantic value, the register within which its utterances are formed are not the register of normal conversation but are rather within the register of oration. Music attempts to move you but it does not ask you to speak back. (Only the fool or the philistine in the audience responds to Hamlet’s “to be or not to be” by answering “go ahead and die already”, thinking that Hamlet actually wanted his opinion.)
But does music have a semantic aspect even if in some mitigated sense? Despite the vitriol of the naysayers, it seems rather misguided to maintain that music lacks all semantics. For example, we often hear people say “X’s piece is so expressive” and we believe that this statement is meaningful. And yet “expressive” is not a simple attribute in the way “white” is. We can say “my kitchen floor is white” by which we mean that my kitchen floor has the attribute of whiteness but nothing has the simple attribute “expressivity”; rather we say that something is expressive of something.
Hence we can say with equal validity that Chopin’s Funeral March from his Piano Sonata in B-flat minor is expressive and that the opening movement of Beethoven’s Pastoral Symphony is expressive but we would not say that they are expressive of the same thing or that they are expressive in the same manner. The Funeral March is expressive of a dark emotion while the Pastoral is expressive of the bucolic and the joyous. Now we could say that the Chopin movement is expressive of something similar to the second movement of Beethoven’s Third Symphony (also a Funeral March) but this points to the heart of the matter. Both movements do express similar things.
Now an objection arises. Our objector concedes that there is a similarity between the two funeral marches but finds this unsurprising and marvels at the naiveté with which we have treated the example. “They are similar,” he assures us, “insofar as they are both funeral marches and therefore they share a certain connotation.” We should, of course, respond: “Right you are; that is the very point.” Music is not simply the architecture of sound (despite the insistence of Stravinsky) but rather relies upon certain cultural tropes that are invested with meaning. I say “invested” inasmuch as all cultural and linguistic meanings are (to a large extent) arbitrary.
Over the course of time, music accretes meaning; meaning continually changes but as it changes, elements of it solidify. Funeral marches are subsumed beneath the genus of marches as a whole. But not all marches, of course, are gloomy. Rather the funeral march shares some characteristics with the wider category of the march (the constant motion forward, the steady inexorable pace) but it inflects those shared characteristics with certain individual traits (intimations of the macabre, the mournful quality). Here the march pushes steadily forward but it is imbued with a deep sense of loss. This social connotation affects the listener even in a piece like the Chopin where the putative deceased is left unspecified. Indeed, part of the effect here derives from the fact that the funeral march is functional music (that is, funeral marches, as such, can actually be performed as part of the funeral ritual) that is transposed in the Chopin sonata into a purely abstracted realm.
The point here is not that music is a language, but rather that it shares certain important characteristics with language that permits it to signify in the way that it does. This key thought has taken on different guises throughout the history of ideas. Jean-Jacques Rousseau (see his Essay on the Origins of Language) believed that language and music were coeval in development, that the earliest utterances were essentially musical. They were the highly inflected, direct expression of emotion between interlocutors. It was only with time and the need for creating communities for survival that language split from music and became hardened (in a sense) through its calcified application to concepts; it gained conceptual depth by losing its direct ties to the passionate soul. Although less concrete in meaning, music for Rousseau (particularly melody) retained its ability to signify by influencing the human emotions.
The 19th-century music critic Eduard Hanslick, a staunch adversary of the simplistic view that the representation of emotion was the content of music, still left a space for a musical semantics, writing in On the Musically Beautiful: “Every shaped tone carries—if only indeterminate—a semantic load…. Tones possess fundamentally and individually, like colors, symbolic meaning which produces an effect outside of and primary to any artistic intention.” Hanslick’s conception is far more complex and deserves far more consideration than it is often given (indeed I intend to treat his writing at greater length in a future column) but even in this brief excerpt we gain a sense of the way in which music, for Hanslick, resembles a language without ever truly being a language. A composer operates with musical tones already imbued with some vague meaning and must shape the tonal material in such a way as to clarify the specific meaning he brings to the material. The composer’s intentions, in other words, are always in response to the social (and, for Hanslick, to an extent natural) signifying properties of the tones themselves.
Theodor Adorno makes a similar point, in an article entitled “Music and Language: A Fragment”, as he attempts to articulate the manner in which music’s historicity. Musical material, according to Adorno, develops over the history of composition—that is to say, it is historically conditioned. A diminished seventh chord in Mahler does not mean the same thing that it did in Bach. This has less to do with the nature of music and more to do with its social development. Musical devices, just like words and turns of phrase in languages, alter over the course of their use. Therefore Adorno insisted that a composer who stood in an authentic relationship to his time (that is, a composer who unlike Stravinsky refused to capitulate to false consciousness) must treat his musical materials at their present stage of historical development because it is in this way that musical material can be said to have meaning. Thus, composition partakes of an evolving dialectical process in which the individual composer (with his personal desires for individual expression) confronts the socially conditioned material in order to create musical works that are themselves inherently social and therefore political entities. Adorno’s view of musical meaning (and its relation to language) relies rather strongly upon the historical (that is, diachronic) unfolding of its development.
However, we may choose to concentrate more on music at a particular moment—that is, we may wish to view it synchronically. This was one of the major achievements in linguistics of Ferdinand de Saussure. Taken synchronically, language manifests itself through what Saussure termed parole (speech) and langue (language). Speech is individual acts of expression while language represented the abstract logical structure of expression that preceded the birth of the speaker and all of the speaker’s individual acts of expression. Every linguistic element (every sign) comprises a signified and a signifier—simply put, a concept and the sound-image that represents it.
It is important to note that for Saussure the sound-image (most typically, an individual word) does not directly represent a thing but rather a concept. Thus, the word “tree” does not represent that thing growing outside of my window but rather the concept I have of “treeness”. Moreover, the signifier is not the actually uttered word but rather its position within the system of language. We recognize the word “tree” not because of what it is but because it is different from a host of other terms in the language. “Tree” signifies “treeness” because it differs from words such as “flea”, “train”, and so on. As Saussure put it: “In language there are only differences without positive terms.”
This would seem to be a fruitful approach to music’s relationship to language. Indeed, in some ways, music exemplifies Saussure’s notion of language as a closed-off system of structured differences. In tonal music, the pitch we identify as tonic holds a particular position within the tonal framework of a piece. It is the tonic (or home pitch) in part because all of the other pitches are not. They are all non-tonics and therefore do not offer us the sense of relative resolution that the tonic does. Moreover, all of the non-tonics are non-tonics in their own way. Not all non-tonics are alike. The note a fifth above the tonic is tonally “closer” to tonic (as a harmony-grounding bass note) than the other notes while the seventh degree (or leading tone) is close to tonic melodically—indeed, it gets its name “leading tone” by virtue of the fact that it “leads” into the tonic.
But we hardly require tonality to make this point. Even atonal music (music that eschews all notion of a grounding tonic that serves as the central foundation for the harmonic and melodic relationships) relies upon structuring differences. We understand the unfolding of an atonal composition insofar as it presents a tone in relation to other tones with which it is not identical. Indeed, atonal music, in some ways, models Saussure’s idea of language as a system of differences without positive terms more closely than does tonal music. In atonal music (ideally speaking) all differences are negative differences. A specific tone in the piece has no actual being or identity outside of its difference with the other tones of the piece.
Tonal music, on the other hand, adds positive difference to its basic level of negative difference. That is why some non-consecutive tones (like the tonic and the fifth above it) are conceptually closer than others. The fifth above the tonic is important not just in its role as non-tonic (and non-everything-else but the fifth above the tonic) but also because it has a specific positive role within the tonal system. It is important because of its relationship to the tonic as much as (or more than) its difference from the other tones. In this sense, we might say that tonal music (still the primary musical idiom and the one most familiar to the majority of listeners) is conceptually deeper than atonal music (but, it should be kept in mind, this in no way need be a value judgment).
Jacques Lacan, building on the ideas concerning language proffered by Saussure but with an eye toward their application in the field of psychoanalysis, posited a sharp divide between speech and language that accounted for a person’s (that is, a subject’s) radical alienation from the logical structure of language. We come to know ourselves only through our perceived difference from others. Moreover, we represent ourselves through signifiers. But we do not directly communicate with other subjects. This is impossible. Rather, Lacan insists, “The signifier represents the subject for another signifier.” This is an important aspect of his thought. For Lacan, our speech acts do not represent us to another person but rather to the system of language to which we are inextricably bound for symbolized communication.
We have no actual control over how our speech-acts are understood. Whatever we say, whatever we attempt to communicate through speech, finds its meaning within the system of language (the logical laws of language pervert meaning by making it conform to a system) and ultimately is (mis)interpreted by an other (it must be misinterpreted in that there is always an unbridgeable gap between our speech and the system of language through which it is necessarily filtered). Our statements find their meaning within another person.
Perhaps it is this Lacanian notion that brings me back to an inquiry into the relationship between music and language. Lacan once asked the auditors at one of his seminars a rather peculiar question: “Why don’t the planets speak?” He never quite answered the question but I would posit that they fail to speak because they are pure language, a pure and calcified system of relationships. Indeed, the notion of the planets as a form of (musical) language goes back to pre-Socratic thought and the idea of the “music of the spheres”. Supposedly, always out there in the cosmos, nature itself sounded out the basic structure of music and it was this structure that constituted the organization of the cosmos itself.
All human musical acts were imperfect attempts to speak within and against that system of language. Speaking involves an attempt (futile but necessary) to force language to speak against itself. Speech involves saying something individual in a rigid system of conformity. Music seems to attempt something similar—or, more appropriately, people attempt something similar through music. In order to communicate, most music relies upon some shared, socially conditioned notion of musical material (whether it be a set of tonal relationships, or familiar rhythmical gestures, or even simply a basic sense of significant contrast). And yet even in the most conventional of musical forms of discourse, composers often attempt to communicate something particular—a speech-act that may be partially perverted by the musical language but that impinges upon our sensibility. We then create the meaning of the utterance through our understanding of the music.
It is no more (and no less) accurate to say that music produces purely musical meaning than it is to say that language produces purely linguistic meaning. Such tautologies are beside the point. We search for meaning. We search for significance. We do so both in the act of creation and the act of reception. Language in the Lacanian sense is what (paradoxically) unites us in alienation and alienates us in our unity. But it is only through the act of symbolization (speech-acts) that sparks of meaning pass through to another. This notion of a nearly impossible but altogether necessary manner of communication strikes me as deeply musical.