Authors  |  About  |  Contact  |  Home
Essays Details
Published:  June 14, 2004
Print:  Printable PDF

Author Details
Author:  Brady Goodwin
At the time of publication, Brady Goodwin was a music major at the University of North Texas.
Linguistics As An Approach For Musical Analysis
by Brady Goodwin
Contrary to the first half of the twentieth century, today discussion of the relation between musical and linguistic analysis have become common among linguists, musicologists, and ethnomusicologists, with major review essays and articles appearing in journals specific to all three fields of study. This is a brief overview of several of those topics, ranging from Sausserian strategies of semantics to Lévi-Strauss’ study of mythological and musical structures and his critics, Powers’ observations between textual relations to music in a South Asian setting, and expanding to include several other anthropologists and ethnomusicologists. It is my intent to provide an explanation and introduction to the far reaching capabilities in the uses of linguistic models and methods and the relation to musical structure and syntax.


The idea is in the air: linguistics, that is, the methods used to describe a language, might be relevant to musical analysis (Nattiez 1973: 51).

The notion of using linguistics as a method to develop musical analysis is certainly not a new subject in the field of linguistics, musicology or ethnomusicology. This idea is traced back over seventy years to 1932, to a lecture given by Becking to the Prague Linguistic Circle. The first mention of using linguistics in a non-linguistic medium was first introduced by F. de Saussere in his Cours de linguistique générale (C.L.G.), published in 1916. The author proceeds to introduce the idea of semiology, or the existence of a series of signs outside of language that could be analyzed under the same forms of linguistic analysis that were already present.

It was after three decades that this idea actually became reputable and linguists began to pose the ideas that Saussere had daringly exposed in suggesting linguistic methods of analysis could be applied to non-linguistic forms. Semiology became defined by George Mounin as a discipline which undertakes as its first task to examine the validity of such a use: in order to do this he suggests that the structures of human language should be compared with those of the field in question (Nattiez 1973: 52-53). When applied to musical structures, this idea becomes known as musical semiology and must start with a basic comparison between the properties of language and with those of music. It can be roughly summarized as the similarities found within language semantics and syntax and music. To refer to music as a linear system is to suggest that music, like language, is a two dimensional medium that progresses in chronological order. Music arouses sensory activity that, in many cases, can not be achieved by the semantics of language; the idea is that music at its most basic is an audible form of communication that rarely will succumb to the same problems that are found in studying written language because it is a singularly audible form of communication. Nattiez attempts to enhance ethnomusicology’s scientific appeal by insisting that we must use the methodologies found in linguistic analysis and learn to apply them correspondingly to musical analysis (Feld 1974: 199). Of course, with the wealth in variety of language study, there is not one correct method to be used, and the ethnomusicologist inclined to research in semiology must be aware that the latest trends among anthropological linguists may not always be the best suited for their particular field.


What Claude Lévi-Strauss referred to as the capacite anagrammatique is the idea that while music and myth are both basically untranslatable when it comes to relating them to anything other than themselves, they are both inherently structural. He asserts that music can only be changed into music, and myth only changed into myth. However, both myth and music are seemingly related--each contains a basic dichotomy: theme and countertheme, both of which can be inverted, rhythmically distorted (through augmentation, diminution or otherwise), modally transformed, or presented in a new timbre (Hopkins 1977: 250). Beyond capacite anagrammatique, it is suggested that both mythological and musical passages are coded patterns that become selected by the culture in which they are surrounded. Referring to the codes implied divides music and myth into two separate levels: the external (physiological in the case of music and historical or supposed facts in the case of myth), and the cultural or internal, which is then split into two levels on its own, one being a pooling of possibilities exacted by culture (in the example of music, the pitches available and the variations possible within), the other the collection of schemes including all the possibilities (Hopkins 1977: 250; Lévi-Strauss 1964).

Lévi-Strauss’ studies in musical analysis through linguistic thinking yielded a good deal of criticism, some of it accusing Strauss of ethnocentricity in his research by overlooking study of West African drum music and focusing primarily in using Western European music as a model for analysis. S. Diamond voiced this opinion on this matter:

Among such peoples [West African] a drummer may, either alone or in combination with others, create incredibly complex contrapuntal rhythms, which disappear at the moment of invention; they are not fixed in any form of notation. Themes may be relatively limited, but the elaboration is rich and everyone seems capable of invention. Indeed the distinction between theme and elaboration becomes trivial, merely academic under such circumstances (1974: 298).

Scholars such as S. Diamond have held over time a view that written record is the only form of permanent documentation of the ways and habits of various cultures spread throughout the inhabitable continents. Of course, there is fallacy found in this notion, for what of mythical stories and musical traditions that are only communicated through oral history and audiation? Believing that through written records and formal structure permanency is found does a disservice to those cultures and therefore showcases a lack of knowledge on the part of these scholars.

The levels of code suggested by Lévi-Strauss have placed reservations in many modern musicologists who do not want to make the claim that music is possibly anything outside of what is presented (Powers 1980: 4). After all, it was Lévi-Strauss himself who first said that myth and music were untranslatable, so how could it be that music could be anything but just music? One will find that as the exploration into language and music continues, various textual commonalities appear that signal to those musicologists who feel apprehension towards the notion of “music only being music” that their viewpoint was similar to those who accused Lévi-Strauss of being ethnocentric.


Harold S. Powers describes three textual relations between music and language that have been found in South Asian and in both the South Indian and North Indian classical music tradition (Powers 1980: 2). Following the idea developed by Bright (1963), that the short and long syllabic phrases within the textual setting that correspond with shorter and longer musical duration, Powers agrees that the principle of this notion is exactly correct, and upon further discretion contains more complexity than was originally thought. Bright contrasted the musical phrases of the Indian raga with a Tegulu padam. The relationship between the textual Telugu-based construction of the raga is found to be both heterogeneous and complex, while retaining in its rhythms the Telugu prosody and implications of Sanskritic examples, from syllable and foot to stanza and beyond (Powers 1980: 3).

Does the textual relation found in these two examples of phonological structure of language and the basic assimilation of music constitute as an entire metaphor of language-as-music and music-as-language (Feld 1974:197)? Powers suggest that more is needed to make that assertion. In most instances the vocal music of a culture has a great deal to do with the influence of instrumental music in the same setting. Most North Indian (Hindustani) instrumental music, in addition to the classical music of South India has grown directly from the vocal repertoire established, and even the instrumental patterns (i.e. plucking of strings on a sitar). Every stylistic difference in plucking methods is a reflection of the long and short syllabic sounds describe in the above paragraph, except for the case of the jhala, which relates more to Middle Eastern methods of string plucking.

The third of these textual relations is based on the more abstract view of language forms and musical design in South Asian settings (Powers 1980: 3). South Asian musical performance in general is based largely upon the structures of later post-classical Sanskrit poetry, which unlike older poetry of the same language is more interconnected in terms of refrain and stanza connections, as opposed to each passage being its own separate entity. The order in which these refrains occur is a natural one, with each independent refrain leading into the next stanza, in some cases completing the last words of the stanza. These patterns usually end with stanzas and refrains containing matching rhyme scheme and sharing a word or two, ensuring that the end of the stanza or refrain flows seamlessly back to the beginning, benefiting repetition and the cycling of the refrains. In nearly all cases of classical ragas of melodic type are derived using these three characteristic commonalities between language and music.


These connections lead back into the desire of many ethnomusicologists to follow in Saussere’s footsteps and find a positive use of linguistic methods to analyze music. However, as Powers further notes, musicologists and ethnomusicologists often have the mistaken approach to music in only a bi-lateral sense, as opposed to a culture-spanning multi-lateral descent on linguistics-based analysis (Powers 1980: 9). We know of course that any attempt to define language in a simple bi-lateral method defeats the ways in which linguistic models were born. Powers points to the fact that linguistic models were devised from an enormous number of languages of all origin and that that were tested against each other, unlike musical methods of testing based strictly on a model derived from a variety of sources, such as field study or standard music theory (Powers 1980: 9). The methods so famously introduced in the aforementioned Cours de linguistique générale find their way back into this discussion of the semantically based relation of music and language. It is suggested by Powers that instead of relying on the new trends of musical analysis, musicologists and ethnomusicologists should reach back to reacquaint themselves with works of such influence such as Saussere (along the same desires that Nattiez proposes). It is noted that some still attempt to define music and thus define the methods (usually consisting of Western origin) of which they analyze it:

As is well known, there is not always agreement about what music is and about what is music, not just across cultures, but even within them. The very notion that music is something that can be segmented and analyzed, and the traditional terminology for doing so, have deep and particular roots in historical language for musical analysis are peculiar to Western European culture (Powers 1980: 9).


Before study into linguistic models as a means for musical analysis was a widely spread as seen today, much was focused on the overlapping nature of the music and language phenomena (Feld 1974: 197). Feld notes that the shifting popularity between the music and language phenomena and the study of linguistics in relation to musical study was in many ways initiated by Lévi-Strauss’ introductions to musical structure in relation to myth and Chomsky’s development of transformational linguistics combined with Sausserian semiotic study.

Looking into the overlap of music and language, two forms of study have been used primarily: that of music in language, referring to the correlation between stylistic and textual notions in basic song structure, and that of language in music, or the musical nature found in the spoken word. Beginning with studies concerning the coincidental nature of aligned musical and speech structures, continuing through Bright’s research pertaining to South Indian relations and following in several attempts by both linguists and ethnomusicologists to find an intermediate melody and the vocalization of language, these advances led to the rising interest in using these two forms of communication naturally intertwined with one another. Once these ideas had been presented, the uses for linguistic methodology in conjunction with music were thrust to the front of the interest of many of those who had begun in the former area.


In using linguistics model for musical analysis, it is emphasized how important these two forms of communication truly are, and the properties that both obtain in the emotional impact on a person. Perhaps this notion can best be summed up by Bright, who asserted that

It is widely felt that music, like language, conveys something--that a musical performance, like a linguistic message, contains something more than the physical properties of the individual sounds that make it up (1963: 28).

It is from here that the majority of linguists, musicologists, and ethnomusicologists find the inspiration to delve in such a far reaching subject that takes years to properly mediate upon. It is my hope that this has served as an introductory guideline to the sciences presented and sparked interest in often overlooked areas of anthropological study.
  • Bright, William 1963 “Language and Music: Areas for Cooperation”, Ethnomusicology 7 (1): 26-32.
  • Diamond, S. 1974 The myth of structuralism, in Rossi, ed., The Unconscious in Culture: the Structuralism of Claude Levi-Strauss in Perspective. New York.
  • Feld, Steven. 1974 “Linguistic Models in Ethnomusicology”, Ethnomusicology 18 (2): 197-217.
  • Hopkins, Pandora. 1977 “The Homology of Music and Myth: Views of Levi-Strauss on Musical Structure”, Ethnomusicology 21 (2): 247-261.
  • Lévi-Strauss, C. 1964 Mythologiques I. Le cru et le cruit. Paris. Eng. Trans., the raw and cooked. New York. 1969.
  • Nattiez, Jean-Jacques. 1973 “Linguistics: a New Approach for Musical Analysis”, International Review of the Aesthetics and Sociology of Music 4 (1): 51-68.
  • Powers, Harold S. 1980 “Language Models and Musical Analysis”, Ethnomusicology 24 (1): 1-60.
  • Saussere, F. de. 1916 Cours de linguistique générale. Paris.