Music Theory Online

The Online Journal of the Society for Music Theory


Volume 6, Number 2 May, 2000
Copyright � 2000 Society for Music Theory

Table of Contents

Raymond Monelle*

Martinez's Concept of "Intrinsic Semiosis"




[1] Most applications of Peircean theory to music have been sketchy, being usually confined to the trichotomy of icon, index, and symbol. The outstanding feature of Jose Luis Martinez's paper on Indian music to the Atlanta conference of SMT, reproduced in the January issue of MTO, was the introduction into music theory of the qualisign, sinsign and legisign, of the idea of emotional, energetic and logical interpretants, and of other aspects of Peirceanism that are normally the preserve of the philosophers. Here is a musician who is able to swim in the broader expanse of the Peircean sea.

[2] It is, of course, a commonplace that the Indian raga system acknowledges a level of referential meaning. Martinez cites raga malkauns, which is said to have a signification of "seriousness or peacefulness." In this respect, the raga becomes a legisign, a type of which the tokens are individual improvisations on the raga. And since the raga has qualities that cause it to resemble its object (it is performed slowly, and the notes mostly lie in the low register) it is an iconic legisign. This is a thoroughly convincing application of Peircean theory to music.

[3] This kind of semiosis is called "Musical Reference" by Martinez. He invokes two other kinds: one of these is Musical Interpretation, which is shown in musical treatises and critiques (it is "the study of the musical sign related to its interpretants"). In the Indian tradition, this may take the form of something that resembles our own music theory, but may also take narrative and mythical forms, which Martinez calls "analogic."

[4] The third kind of musical process is called "Intrinsic Semiosis." Kofi Agawu, in his response to this session of the conference (which included three other papers) relates Martinez's system of three semioses to Nattiez's "tripartition theory." Presumably, Intrinsic Semiosis made Agawu think of the "neutral level," and Musical Interpretation evoked the "esthesic level."

[5] Ideas like "intrinsic semiosis"--semiotic, but not related to an "outside world"--have appeared often in music theory. There may be a tenuous connection to the Molino/Nattiez theory, but I think there's a lot more to it than this. Let us, first of all, examine what this author means by his novel concept. At first sight, it seems to resemble Wilson Coker's "congeneric musical meaning" (Coker, 1972, pp. 60-88). There is, however, a fundamental difference. Coker describes congeneric meaning as intramusical, the signifying by one part of a composition of another part. For example, in the first five measures of Beethoven's Fifth, the first phrase (mm. 1-2) "predictively" signifies the second (mm. 3-5), and the second "retrodictively" signifies the first. The ground of signification is iconic, since they resemble each other. This writer is aware that the relation between phrase 1 and phrase 2 is most obviously syntactic, but he says that the syntactic coherence of a composition is established when "one part of a piece may iconically signify another" (Coker, 1972, p. 66). His subsequent descriptions of strophic form, sonata form, and so on, make it clear that he regards repetition and recurrence in music as a kind of iconic reference, that is, a type of signification. It is true that syntax can be made to signify; poets are fond of this device. But a syntactic relation is not in itself a semantic feature. The recurrence of a musical theme in a recapitulation is a means of confirming syntactic coherence. If, in addition, it is significant of something, then this is quite another matter. This kind of choice is stressed by Boris Asafiev, when he suggests that the full close in the tonic key, a syntactic feature in classical music, acquired a semantic weight in the works of Beethoven, coming to signify "unconditional affirmation, persuasiveness, confidence and rationality" (Asafiev, 1976/1942, pp. 763-764).

[6] When he insists that syntactic features are automatically aspects of signification, Coker seems to imply, for example, that a musical return signifies "return," which is of course illogical. More properly, the repetition or recurrence of a musical figure is somewhat like the repetition of a seme in a syntactic string; it inaugurates coherence and constructs redundancy. The coherent syntagm thus constructed may, admittedly, function as a sign.

[7] But Martinez's view is different. In Indian music, the performance is always an improvisation; traditional linguistic theories are thus easier to apply, for linguists usually discuss speech, while Western music is considered to be a repertoire of finished scores, something more like literature than language. The performance, then, is a sinsign of a legisign, the legisign being the raga. Since we recognize the raga in the performance by noticing common qualities, the process also involves qualisigns. "Legisigns," he writes, "signify by means of their actualization (replicas or sinsigns), which--in Indian music--are somehow flexible, because the performer improvises, choosing the sequence of a raga's motifs. . . . Further, each raga has its peculiar musical qualities, and this is a matter of qualisigns."

[8] Intrinsic semiosis is therefore not iconic--or at least, Martinez does not say it is iconic--but is based on the relation of qualisigns, sinsigns, and legisigns. But such a relation is not semiotic; it is logical. The sinsign is a real thing which is a token of a type. The token-type relation, though it is everywhere active in semiosis, is not itself a semiotic relation. To characterize the musical performance as a sinsign, is not to explain its function in the process of semiosis, but merely to classify it. "The token does not represent the type," writes David Lidov; "the vehicle does not represent the ground" (Lidov, 1999, p. 118).

[9] Martinez does not make the mistake of thinking that a syntactic relation is semantic, like Coker. Instead, he makes the quite different mistake of imagining that a sinsign signifies a legisign. If this were so, then a real Boeing 747 would signify the category "Boeing 747" or a real proper noun which I use in speech, like "Professor Agawu," would signify "proper noun," whereas it clearly signifies the person who responded to this session of the conference.

[10] As I have already mentioned, Agawu himself associates Martinez's "intrinsic semiosis" with Nattiez's "neutral level," itself derived from a famous article of Jean Molino (1975). Neither Molino nor Nattiez were writing in Peircean terms when they proposed this idea; they meant to frame it in the terms of structural linguistics. It corresponds, according to Nattiez, with the "immanent level where structural linguistics is situated" (Nattiez, 1975, pp. 50-51).

[11] This seemed a very promising line of thought when the Fondements appeared in 1975. Music seems abstract, yet it determines the sort of emotional and energetic interpretants that we associate with semiosis. Perhaps, then, by invoking the "immanent level," we can discuss music in semiotic terms without having to attribute meanings. This, at any rate, was what Nattiez set out to do in the second section of his book.

[12] But unfortunately, this proposal, like those of Coker and Martinez, is flawed by a systematic error. The idea of immanence makes us think of Saussure's famous statement that linguistics concerns itself with "differences, without positive terms" (Saussure, 1974, p. 120). For the great Swiss linguist, the phonological level and the semantic level of language were systematically related, although the meanings of words were simply arbitrary. We have moved a long way since then; Emile Benveniste showed that Saussure's logic was at fault in his idea of arbitrariness, because if we assume that phonological units are related to linguistic semantic units (not to real objects) then the relation is "necessary, not arbitrary" (Benveniste, 1971/1939, p. 45). By the time Benveniste was writing, the Prague school had shown that immanent linguistic theory required the concept of pertinence--the information that certain differences were operative within the system, others not. And this information can proceed only from a native speaker, who knows it because she thinks in terms of semantics and pragmatics; that is, because she understands language as a semiosis, not as an abstract system. This kind of immanence is magnificently summarized in Hjelmslev's Prolegomena, in which semantic distinctions, systematically considered, are an essential part of the immanence of language (Hjelmslev, 1961). It also rears its head in Chomsky's *Syntactic Structures* (Chomsky, 1957); deep consideration of the generative structuring of language leads the American to realize that grammar may mingle with semantics, and the the ill-formedness of certain hypothetical sentences may depend, not on grammar or syntax, but on meaning.

[13] The crisp rationalism of structural linguistics, then, is based on an admission that signification underpins every distinction. The bracketing of meaning by the linguist is entirely provisional. Semiotics, as opposed to semantics and pragmatics, is an inheritor of this kind of exclusive rationalism; it is a theory of how signification works, not a system of interpretation. But like structural linguistics, it either implies a foundation in semantics or it is not a theory of signs at all. Nattiez's "description of the neutral level" is, clearly, not a theory of signs, but a meta-rational system of music analysis. Perhaps the truest thing said about all sign theories is Greimas's principle that significations are "simply human" (Greimas, 1983/1966, p. 10). Other theories are echoed, negatively, in this statement; Hanslick's idea that musical content was "purely musical," for example, or the view of Clive Bell that the esthetic world is "a world with emotions of its own" (Bell, 1928, p. 27).

[14] Martinez is on safe ground with his musical reference and interpretation, especially since Indian music theory offers an elaborate account of the moods, pictures and traditions embodied in each raga. Western musicians find it more difficult to base their interpretations exclusively on such pictures; this kind of meaning is called "extramusical." In the West, we need some way of preserving the abstract level of music--the neutral level, the aspect of congeneric meaning, the concept of intrinsic semiosis--in order to avoid a trivial search for pictures and programs. If we deny such a level, then perhaps music is not a sign at all, and the whole enterprise (including Agawu's own writings, and, I confess, my own) is in vain and the semiologists must fall silent. We ought to consider this outcome with some degree of care, for nothing is to be gained from special pleading or the defense of academic empires. Reading Peirce, it is easy to assume that the whole world is composed of signs. Douglas Greenlee feels sympathy with "a suspicion that Peirce believed all events to be signs" (Greenlee, p. 45), though he thinks this view is ultimately false. A contemporary Peircean, David Lidov, lists "abstractive duality" (the distinction of token and type, for instance), "epistemic duality" (such as the relation of sensation and appearance) and "phenomenal duality" (the outward and inward experiences of a thing) as relations that are not semiotic (Lidov, 1999, pp. 118-119). But in a much simpler sense, ordinary things may not be considered signs. A pebble on the beach is not necessarily a sign, though we can consider it as an indexical sign of the eroding power of the sea, an iconic sign of a pigeon's egg, or a symbol of the number "one" if we are counting in pebbles. In spite of this, we do not normally think to classify a pebble as a sign.

[15] The music we play in order to accompany dancing may, indeed, be an indexical sign of the dance (so that we call the music "fandango" or "pavane," referring to the dance). But we do not need this dimension to hear it as music; we may not know that the piece is a fandango or that it is a dance at all. It is almost impossible to take such a view of language; the immanent level of language, as I have said, is defined by the level of signification which it brackets. But music can be understood as music without any idea of semiosis. "Language and music, in view of their immanent function . . . cannot be compared at all" (Harweg, 1968, p. 273). If language is opposed to the world, the one as signifying system, the other as merely an inventory of significations, then music is on the side of the world, according to Roland Harweg, like "sunsets and motor races."

[16] But if this is so, then why does music move us, and why does it suggest moods and pictures? I believe that we are bound to grapple with a view of musical signification as "simply human"; with an "extrageneric," "referential," semantic view. The difficulty of describing this signification in language must not send us running to abstract rationalisms.

[17] Again, Lidov offers a way forward (Lidov, 1999, pp. 181-184). Peirce's term "object" seems to suggest a concrete thing, event, or experience. But signs may be processes rather than things. If the representamen is a process, according to Lidov, the sign is a kind of ritual. "The lighting of the candles, the sweet food after fasting, or the confession" are signs of which the representamina are processes. The object, also, may be a process, and in this case the sign is a "symbol" (in the traditional sense, not in Peirce's sense). A tombstone refers not simply to a grave, but to a whole narrative of experience connected with our relation to the dead person. "The symbol actually provokes a sample of the experience it refers to." Finally, the interpretant may be a process. This is the case with art. "When engaged with art, we are cast into its perspective. The perspective sustained is an interpretant-process" (Lidov, 1999, p. 184). The feelings of excitement, expectation, suspense, delight which accompany our reading of a novel are parts of the interpretant, leading us to understand the text. Yet the object of this text may be obscure; in the case of a work with a program, like Hardy's plea for the non-judgmental acceptance of women in Tess of the
, or Eliot's praise of the gentle wisdom of the Jews in Daniel Deronda, it is possible to give some kind of account of the object, but in the best novels the object seems "suppressed" (Lidov's word). Speaking of the Five Orchestral Pieces, Schoenberg said that "an advantage of music was that one could confess all without actually revealing anything" (Lidov, 1999, p. 185). Therefore, in the case of a sign with a processual interpretant, the object will seem obscure, ambiguous, indescribable. Indeed, this is considered to be a mark of depth in an artistic sign.

[18] Oddly enough, this view would apply just as much to literary works as to music. The "object" of a novel or a poem is usually quite hard to discern. The difficulty of relating music semiotics to linguistics derives, not from the fact that language signifies a manifest object while music has no object at all, but from the mistake of taking language, instead of literature, as the material for comparison. The apparent "abstractness" of music is a general feature of artistic signification. In this sense, literature, painting, sculpture, and architecture are also abstract. This realization enables Lidov to present a semiotic account of the Funeral March from Beethoven's Eroica Symphony in which social and topical referents are woven into an elaborate system of signification (Lidov, 1999, pp. 231-239).

[19] Lidov has grasped the nettle of "extramusical" or "simply human" signification in music, and I have attempted to follow him in a work that will be published shortly (Monelle, forthcoming). Topic theory, proposed by Ratner (1980) and Agawu (1991), has been something of a Cinderella among music studies, but it can be broadened and deepened by a historical study of the cultural themes suggested by musical topics. Music may also signify cultural temporalities, and it may analyze levels of subjectivity. Indeed, the positivistic emphasis of structural linguistics, and even of Peircean semeiotic, may be inappropriate to musical semiosis; musical signification flows through the points of logical arrival, extending into an infinite chain of interpretants and suppressed objects--and this may be a mark of the strength of musical signification, not its weakness. The deconstructive implications of such a view are drawn out more fully in my new book.

[20] Martinez's extraordinary achievement in applying the breadth of Peircean sign theory to Indian music must not be diminished by a criticism of his concept of "intrinsic semiosis." Perhaps his doctoral thesis may appear in print so that we can examine his notions in more detail. It clearly has the philosophical strength to withstand our reservations, and indeed to inspire further thought, whether critical or sympathetic.

Raymond Monelle

Table of Contents

Return to beginning of article


Agawu, V. Kofi (1991). Playing with Signs: A Semiotic Interpretation of Classic Music. Princeton: Princeton University Press.

Asafiev, Boris V. (1976/1942) B.V. Asafiev's Musical Form as Process. Trans. by James Robert Tull. Ph.D Diss., Ohio State University.

Bell, Clive (1928). Art. London: Chatto.

Benveniste, Emile (1971/1939). "The Nature of the Linguistic Sign." In Mary Elizaeth Meek, ed., Problems in General Linguistics, pp. 43-48. Coral Gables: University of Miami Press.

Coker, Wilson (1972). Music and Meaning: A Theoretical Introduction to Musical Aesthetics. New York: The Free Press.

Greenlee, Douglas (1973). Peirce's Concept of Sign. The Hague: Mouton.

Greimas, Algirdas-Julien (1983/1966). Structural Semantics: An Attempt at a Method. Trans. by Daniele McDowell, Ronald Schleifer, and Alan Velie. Lincoln: University of Nebraska Press.

Hjelmslev, (1961). Prolegomena to a Theory of Language. Trans. by Francis J. Whitfield. Madison: University of Wisconsin Press.

Lidov, David (1999). Elements of Semiotics. New York: St. Martins Press.

Molino, Jean (1975). "Fait musical et semiologie de la musique." Musique en Jeu 17: 37-61.

Monelle, Raymond (forthcoming). The Sense of Music: Semiotic Essays. Princeton: Princeton University Press.

Nattiez, Jean-Jacques (1975). Fondements d'une semiologie de la musique. Paris: Union Generale d'Editions.

Ratner, Leonard G. (1980). Classic Music: Expression, Form, and Style. New York: Schirmer.

Saussure, Ferdinand de (1974). Course in General Linguistics. Ed. by Charles Bally and Albert Sechehaye, trans. by Wade Baskin. Glasgow: Collins.

* Return to Beginning

Table of Contents

Copyright Statement

Copyright � 2000 by the Society for Music Theory.
All rights reserved.

[1] Copyrights for individual items published in Music Theory Online (MTO) are held by their authors. Items appearing in MTO may be saved and stored in electronic or paper form, and may be shared among individuals for purposes of scholarly research or discussion, but may not be republished in any form, electronic or print, without prior, written permission from the author(s), and advance notification of the editors of MTO.

[2] Any redistributed form of items published in MTO must include the following information in a form appropriate to the medium in which the items are to appear:

This item appeared in Music Theory Online in [VOLUME #, ISSUE #] on [DAY/MONTH/YEAR]. It was authored by [FULL NAME, EMAIL ADDRESS], with whose written permission it is reprinted here.

[3] Libraries may archive issues of MTO in electronic or paper form for public access so long as each issue is stored in its entirety, and no access fee is charged. Exceptions to these requirements must be approved in writing by the editors of MTO, who will act in accordance with the decisions of the Society for Music Theory.

This document and all portions thereof are protected by U.S. and international copyright laws. Material contained herein may be copied and/or distributed for research purposes only.

prepared by
Brent Yorgason, Editorial Assistant
Updated 18 November 2002