Turino, Thomas_Signs, Imagination, And Experience. a Peircian Semiotic Theory for Music

Signs of Imagination, Identity, and Experience: A Peircian Semiotic Theory for Music Author(s): Thomas Turino Source: Ethnomusicology, Vol. 43, No. 2 (Spring - Summer, 1999), pp. 221-255 Published by: University of Illinois Press on behalf of Society for Ethnomusicology Stable URL: http://www.jstor.org/stable/852734 . Accessed: 04/07/2013 14:14 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

.

University of Illinois Press and Society for Ethnomusicology are collaborating with JSTOR to digitize, preserve and extend access to Ethnomusicology.

http://www.jstor.org

This content downloaded from 201.174.75.10 on Thu, 4 Jul 2013 14:14:48 PM All use subject to JSTOR Terms and Conditions

VOL.43, NO. 2

ETHNOMUSICOLOGY

1999 SPRING/SUMMER

Signs of Imagination, Identity,

and Experience: A Peircian Semiotic Theory for Music THOMASTURINO / University of Illinois at Urbana-Champaign in many societies intuitively recognize the emotional power of

People music in their personal, family, and community life. If ethnomusicolo-

gists have come to agree on anything over the last decade it is that music is a key resource for realizing personal and collective identities which, in turn, are crucial for social, political, and economic participation. These observations are integrally related, and they form the basis of the central question for musicology: "Why music?" Like the habitus, identities are at once individual and social; they are the affective intersection of life experiences variably salient in any given instance. Identity is comprised of what we know best about our relations to self, others, and the world, and yet is often constituted of the things we are least able to talk about. Identity is grounded in multiple ways of knowing with affective and direct experiential knowledge often being paramount. The crucial link between identity formation and arts like music lies in the specific semiotic character of these activities which make them particularly affective and direct ways of knowing. Recent scholars of ethnomusicology have succeeded in illustrating the intimate interfaces of sound structures, social structures, and identity (e.g. Seeger 1980, 1986; Pefia 1985; Feld 1988; Pacini Hernandez 1995; Sugarman 1997). It seems to me that the challenge for the next generation is to develop a theory of music in relation to what is usually called "emotion"our inadequate gloss for that mammoth realm of human experience that falls outside language-based thinking and communication. Such a theory is necessary if we are to move beyond mere description of the central roles music and dance play in collective events ranging from spirit possession ceremonies, mass nationalist rallies, and weddings, to the teen dances taking place on a Friday night. ? 1999 by the Board of Trustees of the University of Illinois

221


222

Ethnomusicology, Spring/Summer 1999

My purpose in this paper is to begin sketching a theory of music, emotion, and identity based on the semiotics of the American philosopher and scientist Charles Sanders Peirce (1839-1914). The semiotic tools discussed are applicable to all expressive practices; here I will focus on music, dance, and propositional language with the understanding that other expressive media have their own unique qualities and capacities and require some separate analysis and application. Initially,Peircian semiotic theory is daunting for the amount of new terminology it requires. Sometimes, however, new words are needed to think new thoughts and to approach old problems in radically new ways. Working intensively with only a small portion of Peircean theory for over fifteen years, I have become convinced that its potential is nothing short of revolutionary for understanding the social effects of music, art, expressive culture, and people's myriad ways of experiencing the world.' My purpose is not simply to present a system that allows us to categorize signs more precisely, although this is a necessary step, but to elaborate the distinct workings and potentials of different sign types in human life. Peircian

Semiotics

In dramatic contrast to the autonomous systems approach of Saussurian-based structural linguistics, Peirce developed a theory of signs to understand how people are connected to, and experience, the world. Whereas structuralismused language as the primarymodelling system, Peirce defined the concept of sign in the widest, most flexible way as something that stands for something else to someone in some way, thus allowing for many different types of signs outside propositional language (e.g. Peirce 1955:99). It has always been surprisingto me that in the great musical semiotics boom of the 1970s, scholars chose the Saussurianline and attempted to show the similaritiesbetween music and language. A basic premise in my work is that the most important theoretical insights about the social power of music will be derived from outlining the differences between propositional, semantico-referentiallanguage, and non-propositional sign modes such as music and dance.2

Sign, Object, Interpretant, and Semiotic Chaining For Peirce, semiotic processes (semiosis) have three basic elements: (1) the sign, something that stands for something else to someone in some way; (2) the object, which is the "something else," or entity, stood for by the sign, be it an abstract concept or a concrete object;3 and (3) the interpretant, which is the effect created by bringing the sign and object together in the mind of a perceiver (1955:99; see Fig. 1).


Turino: A Peircian Semiotic Theoryfor Music

223

Figure 1. Sign or Representamen

Interpretant = What the sign creates in the observer; the effect the sign has in/on the observer, including feeling and sensation, physical reaction, as well as ideas articulated and processed in language.

= Something actually functioning as a sign.

Object = What the sign stands for.

Some Basic Principles 1. There can be an infinite unfolding of signs in the mind, a kind of chaining process. 2. Thirds include Seconds and Firsts; Seconds include Firsts;Firsts can only determine a First (whatever is a Third determines a Third, or degenerately a Second or a First, etc.). 3. A fully developed general purpose language must have icons, indices, and symbols, according to Peirce.

Semiosis involves a type of chaining process through time in which the interpretant at one temporal stage becomes the sign for a new object at the next stage of semiosis, creating a new interpretantwhich becomes the next sign in the next instant, ad infinitum until that "trainof thought" is interrupted by another chain of thought, or by arrivingat a belief or conclusion. In each instant in the chain, the new sign stands for a new object creating a new interpretant-multiple examples of this process will be provided in what follows. Contrasting with general postmodernist views, in Peircian theory signs are neither unmoored from the objects they signify, nor are signs necessarily only linked to other signs.4 Both these ideas, derived from Saussure's problematic binary conception of linguistic signs, collapse the basic triadic character of semiosis and the different moments of semiotic chaining-that is, how sign-object relations at one stage create a distinct effect (interpretant) which becomes the sign at the next stage in the chain (Peirce 1991:239). Peirce emphasized that a sign is not a self-evident idea or entity but is the catalyst for an effect. As conceptualized within Peircian semiotics, "chains of semiosis" move between particularly sensory and direct types of signs and effects to those


224


that are mediated by language (Peirce 1991:70-75). It is my thesis that the

power of music to create emotionalresponsesand to realizepersonaland social identities is based in the fact that musical signs are typically of the direct, less-mediated type. Music involves signs of feeling and experience rather than the types of mediational signs that are about something else. Interpretants and the Mind-Body Dichotomy Moving past the inadequate Cartesianmind-body and emotions-thought dichotomies, there are three basic kinds of dynamic interpretants5-three general classes of effects created by sign-object relations (e.g. Peirce 1958:393; 1960:5.475; see Fitzgerald 1966:71-90). Peirce called the first type an emotional interpretant, a direct, unreflected-upon feeling caused by a sign. Since other types of interpretants can also involve emotion, this term is confusing; sense, feeling, or sentiment interpretant might be closer to Peirce's idea. The second type is an energetic interpretant, a physical reaction caused by a sign, be it unnoticed foot tapping to music, an accelerated heartbeat from a police siren, or unreflexively drawing a finger back from a hot stove. The third type is a sign-interpretant, that is, a linguistic-based concept. All three interpretanttypes involve signs and all three involve perception and mental activity. This framework thus gives us tools for describing different types of mental activity, or "thought,"be it languagebased or not, and hence eschews the strict mind-body dichotomy as it has typically been conceptualized.6 Moreover, for Peirce, the concept of meaning, a long-debated problem in regard to musical meaning, is pragmatically simplified by defining it as the actual effect of a sign, that is, the direct feeling, physical reaction, or language-based concept inspired in the perceiver by a musical sign (Peirce 1955:30-36). When a Tree Falls in the Forest The first step in semiotic analysis is to determine what is the sign, what is the object, what is the effect, and to whom, in any instance. While seemingly simple, this basic step is often overlooked leading to the postmodernist conflation mentioned earlier. A fundamental premise in the Peircian framework is that a sign has to create an effect, an interpretant, within a living being; this precludes abstract assigning of meanings, and in fact the hypothetical manufacturing of signs and objects in social analysis. When a tree falls in the forest it creates waves through the air, a potential sign, but the waves do not function as a sign unless there is someone there to be affected by them. Likewise, musical signs are sonic events that create an effect in a perceiver; not everything happening in music necessarily functions as signs all the time (something might not be apprehended, might not cause


Turino: A Peircian Semiotic Theory for Music

225

an effect). But within the Peircian framework if aspects of music create an effect, signs are necessarily involved. In this context, ethnography becomes crucial to social and musical semiotic analysis since it allows us to identify what the signs are, in relation to what object, for whom, and in which ways.

Peircian Categories for Signs and Sign-Object-Interpretant Relationships Peirce developed three trichotomies of concepts for analyzing different aspects of a sign and distinct types of relationships between the three basic components of semiosis: sign-object-interpretant. Combining one component from each of the three trichotomies to more fully comprehend the nature of a given sign, Peirce arrived at ten basic sign types (e.g. Peirce 1955:98-119; 1991:23-33;1958:390-393; 1960:2.43-2.308). These range from signs that produce particularly direct effects without need for the mediation of linguistically-based thought, to signs, objects, and interpretants grounded in language. I will first go through all the concepts briefly and then return to those that have the most potential for explaining music's power to create affect and forge social identities.

Trichotomy I: The Sign Itself The first trichotomy involves the nature of the sign itself (see Fig.2). Every chain of semiosis begins with the qualisign: a pure quality embedded in a sign such as redness, or the quality of a particular musical sound, or the quality of a harmonic or melodic relation. This aspect helps determine the identity and semiotic potential of the sign. The second concept in Trichotomy I is the sinsign which is the actual specific instance of a sign, e.g., each individual appearance of the word 'the' on this page or the redness of a particular rose. The third term is the legisign which is the sign as a general type, e.g., "The Star Spangled Banner"as a piece apart from any given performance of it, or the word 'the' apart from any instance of it, or the concept of "the color red." Both qualisigns and legisigns are dependent on actual realizations (the sinsign), just as any realization is dependent on the qualities of the sign (qualisigns) which allow us to apprehend it. Particularly important, the social meaning of a given instance of a sign is also informed by its belonging to general nested classes of phenomena (legisigns). Thus, the effects of a given performance of the "StarSpangled Banner"(sinsign) are informed by being related to the piece as a general class (legisign) so that we recognize it and relate it to former hearings. "The Star Spangled Banner"is also nested within other general classes of phenomena such as 'American nationalistic music,' and 'music'; these are other potential legisigns for a giv-


226


Figure 2.

Sign Trichotomy m: of the way a sign is interpreted as representing its object. 1. Rheme 2. Dicent 3. Argument

-

Interpretant

Trichotomy II: of the relation between sign and object. 1. Icon 2. Index 3. Symbol

Object

1. emotional

interpretant Dynamical Interpretant

Trichotomy I: of the sign itself. 1. Qualisign (tone) 2. Sinsign (token) 3. Legisign (type)

2. energetic

interpretant 3. a "sign"

1. Immediate object = "the object as the

signrepresentsit-contained within the sign."

2. Dynamicalobject= the objectoutside the sign; "the reality which by some means contrives to determine the sign." Note: The sign must indicate the dynamical object by a "hint,"and this hint (contained within the sign) is the immediate object.

en performance. As socially-relative categories by which phenomena are conceptually grouped, legisigns are a foundational aspect of culture. Trichotomy II: Sign-Object Relations Peirce's second trichotomy of concepts, involving the icon, index, and symbol, specifies three ways that the sign and object are related in a perceiver. This is the aspect of Peirce's work that has received the most attention. The term icon refers to a sign that is related to its object through some type of resemblance between them. The degree, basis, and even accuracy of resemblance is not so much at issue as the fact that resemblance calls forth the object when perceiving the sign. Thus, if a literal musical quotation or even the vaguest trace of another piece brings that piece to mind, iconicity is involved-the experienced quotation or trace is the sinsign, the



227

piece as general class (legisign) is the object. Motivic unity and most aspects of musical form operate iconically. This much is obvious. More importantly, common musical devices such as a rising melodic line, accelerando, and crescendo may create tension and excitement in a listener because they sound like so many human voices we have heard rising in pitch, speed, and volume when the speaker becomes excited. For most listeners, such signs are typically not processed in terms of language-based thought but are simply felt because of a direct identity established by resemblance between the musical signs and other expressions of excitement. Peirce suggests three types of icons: an image, a diagram, and a metaphor (1955:104-105). In an image, the sign-object relation is based in simple qualities shared; a musical "trace"or quote in one piece calling forth another piece would be of this type, as are most musical icons. A diagram involves analogous relations of the parts between sign and object as the basis of similarity between them; a map is of this type. In metaphors, juxtaposed linguistic signs, which are not iconically related to their objects or to each other, posit some parallelism or similarity between the objects of the signs-e.g., "A mountain of a man" suggests that 'the man' is 'large,' 'hard,' or 'durable.' The concept of metaphor has become popular in anthropology and ethnomusicology to denote iconicity in general and even other types of semiotic relations.7 Often lacking clear definition, the term has lost its usefulness for semiotic and cultural analysis whereas, as with Peirce's other formulations, his definition of metaphor more precisely pinpoints what is going on semiotically. The second concept in Trichotomy II is index which refers to a sign that is related to its object through co-occurrence in actual experience. Smoke can serve as an index of fire, a TV show's theme song can come to serve as an index for the program, a V7-I progression may index musical closure in European societies, the "StarSpangled Banner"may serve as an index for baseball games, Fourth of July parades, school assemblies, or imperialism depending on the experiences of the perceiver. The power of indices derives from the fact that the sign-object relations are based in cooccurrences within one's own life experiences, and thus become intimately bound as experience. Peirce uses the term symbol in a particular way that differs, and must be actively divorced, from standard usage.8 The Peircian symbol is a sign that is related to its object through the use of language rather than being fully dependent on iconicity or indexicality. Symbols are themselves of a general type (legisigns) whose objects are also general classes of phenomena (Peirce 1955:102). Most linguistic signs-words-are symbols,9 and language is the only semiotic mode that, in and of itself, has symbolic capability.10Language also uses iconic and indexical processes but it is par-


228


ticularly in propositional speech, and in the semantico-referentialfunctions of language (i.e., language used to refer to and define other parts of language) where its symbolic capacities differentiate it radically from other semiotic modes like music. After the early stages of language acquisition, we learn the objects of words through linguistic explanations, those objects being general concepts which are also articulated through symbols. For example when we explain that 'a cat is a furry animal,' both 'furry'and 'animal' are general languagebound concepts. We can experience what the feeling of furriness is by patting an actual cat, but we can not designate the general feeling without symbols anymore than we can reproduce the sensation through them. The symbolic function of language is what allows us to think in, and express, generalities. Yet because they are mediational signs which do not resemble, or can be removed from direct connections with their objects, symbols can not reproduce the feelings and experiences of those objects. Symbols are signs about other things, whereas icons and indices are signs of identity (resemblance, commonality) and direct connections. Whereas the meanings of indices are dependent on the experiences of the perceiver, and thus can be quite fluid and varied, the meanings of symbols are relatively fixed through social agreement. Dictionaries, math books, and Morse Code manuals document the conventional meanings of symbols. If symbols are to serve their special function of signification in general, relatively context-free, ways, their meanings must be basically fixed and agreed upon, or, as in this paper, (linguistic) arguments must be made for why their meanings should be altered or refined. Icons and indices have distinct semiotic functions and operate differently. For the most part, musical sounds that function as signs operate at the iconic and indexical levels. The sound of a particular Indian raga x may become a symbol for 'morning' (object) if the relationship is established in general terms through language as, for example, through verbal explanation in an American classroom, and if, upon hearing raga x subsequently, a student thinks the general concept 'morning.' But note that in the initial setting up of this relationship, the sound of the raga was the object of linguistic signs referring to the music and linking it to the general concept of a given time of day. More typically, musical sign-object relations are established without the mediation of symbols. When growing up in India if a young girl frequently heard a particular set of musical sounds (raga x) being played in the morning over the radio in her home, she might come to experience the sensation of 'morning' or 'home,' or myriad other things indexed by the sounds when hearing them later in life. The affective potential of signs is highly dependent on the manner in which the sign and object are linked. The wealth as well as depth of asso-



229

ciations with ragax are likely to be quite different for the Indian girl growing up with it because of the number and variety of indexical associations, as compared to the American student studying the raga largely through propositional and semantico-referential speech in a given class. Indices are experienced as "real"because they are rooted, often redundantly, in one's own life experiences and, as memory, become the actual mortarof personal and social identity. When given indices are tied to the affective foundations of ones personal or communal life-home, family, childhood, a lover, war experiences--they have special potential for creating direct emotional effects because they are often unreflexively apprehended as "real"or "true" parts of the experiences signified. By contrast, symbols are general, mediational signs about rather than of the experiences they express. Trichotomy III: How the Sign Is Interpreted Peirce's third trichotomy--rheme, dicent, and argument--involves the way a given sign is interpreted as representing its object. A rheme is a sign that is interpreted as representing its object as a qualitative possibility (Peirce 1955:103). A rheme is a sign that is not judged as true or false but as something that is simply possible. Peirce used the example that any single word, say common nouns like 'cat,' 'god' 'unicorn' or 'nation,' are rhemes because they suggest the possibility of these entities without (in themselves) asserting the truth or falsity of that possibility (1958:392). Likewise, a painting of an unknown or imaginary person or scene may be interpreted as a rheme. The second concept in Trichotomy IIIis the dicent. This is a sign which is understood to represent its object in respect to actual existence (Peirce 1955:103). The most important feature here is that a dicent is interpreted as really being affected by its object. A weathervane is a dicent-index for 'wind direction' (object) because the wind direction actually affects the position of the weathervane (it is indexical because of co-occurrence of wind and weathervane). A linguistic proposition is a dicent-symbol because the truth of the sign is interpreted as really being affected by the relations of the objects expressed through symbols." Dicent-indices are among the most direct and convincing sign types because typically they are interpreted as being real, true, or natural. They are often taken for granted and apprehended with a part of our awareness that does not involve linguistic-based signs (i.e., at the levels of feeling or energetic interpretants). The field of kinesics--"body language"--theorized by Gregory Bateson (1972), Ray Birdwhistell (e.g., 1960, 1970), and Edward Hall (1977) is largely the study of dicent-indexical signs.12 "Bodylanguage" is a dicent sign because it is interpreted as being the direct result of a person's actual attitude (object) and is thus apprehended as actually being


230


affected by that object. Facial expression, body position, and gesture typically create effects at the levels of emotional or energetic interpretants.13 It is true that signs that usually operate as dicent-indices such as tone of voice and "body language" can be manipulated, for example by actors, used-car salesmen, politicians, and false lovers. In daily interactions, someone who becomes known for being able to do this, however, is branded a phoney. Such people are particularly mistrusted because we are used to taking dicent-indices at face value and are especially offended when people manipulate these types of signs. The third concept in Trichotomy III is argument, involving both symbolic propositions as well as the language-based premises upon which the propositions can be interpreted and assessed. Argument is largely within the propositional, semantico-referential linguistic domain and is not particularly relevant to the analysis of musical signs. Rhemes and dicent signs, however, are key to artistic practice and meaning, and I will emphasize these two types later in the discussion. The Combination of Components

from the Three Trichotomies

Above I have already begun to illustrate how the components from the three trichotomies must be put together to better comprehend the full character of a given sign. Described in respect to the three trichotomies, a common noun is a rhematic-symbolic-legisign. It is symbolic because the sign-object relation is determined through language and because both sign and object are of a general type. The term legisign is redundant in this case because all symbols are legisigns. As explained above a noun is a rheme because it is interpreted as standing for a possible type of object ratherthan a specific existential object. As another example, a sudden, very loud sound in music might function as a rhematic-iconic-legisign with objects like 'thunder,' or 'explosion.' The rhematic aspect here is that these are possible objects rather than any specific instance of thunder or explosion. The iconic aspect is that the sign and object are related in the mind through resemblance. This is a legisign when the loud musical sound is a general type of icon for such objects. A weathervane is a dicent-indexical-legisign. It is a dicent because the 'direction of the wind,' which is the object of the sign actually affects the sign (the direction of the weathervane); it is indexical because the sign and object are related through co-occurrence, and it is a legisign because weathervanes are a general type of cultural phenomenon (we have seen them before). Peirce defined a weathercock as a dicent-indexical-sinsign (1955:115), and any given instance is, indeed, a sinsign. I believe, however, that our understanding of the significance of the direction of the weathervane depends on its status as a general type of sign that we have seen



231

before and hence know how to interpret, i.e., its character as a legisign. This is more than a matter of labelling. My emphasis on legisign in this case is key to my analysis of how I understand the sign to be functioning. When framed to be taken literally, facial expressions, vocal quality, the manner of articulation involved in plucking a guitar or blowing a sax can all function as dicent-indexical-legisigns. Any given instance of a sign involves quality (the qualisign) which allows us to recognize it, and is a sinsign. These features can usually simply be assumed. Most sinsigns in culture, which is to say most sinsigns, signify because they are immediately related to one or more general classes of phenomena-legisigns. The way sinsigns stand for their legisigns-i.e., the way they are categorized and grouped with other sinsigns to form a general type-is usually a culturally relative matter and is often key to cultural analysis. This is true for "body language" as well as weathervanes. The meanings of a smile are not self-evident cross-culturally or even across different social frames within the same society (Birdwhistell 1970). We learn to interpret smiles by linking them to general classes of dicent-indexical signs that we have experienced before in given contexts, that is, we understand them because they are legisigns. All signs can be analyzed in relation to aspects from the three trichotomies, producing ten basic sign types (see Figure 3). In discussion, however, signs are best identified by emphasizing the element(s) most prominent to their function in a given instance of semiosis or for a given purpose in analysis. The same sign, then, might be called simply icon, or rhematicicon, or rhematic-iconic-legisign depending on what the analysis or description requires. The Three Basic Categories: Firstness, Secondness,

Thirdness

We have now gone through the three basic trichotomies and suggested how ten possible sign types can be identified from the combination of their components. This entire semiotic framework is predicated on Peirce's three most basic categories for all phenomena (Peirce 1955:74-97). These are Firstness, something in and of itself without relation to any second entity; Secondness, relations between two entities without the mediation of a third; and Thirdness, involving the mediational capabilities of a person to bring a first and a second entity into synthetic or general relationships with each other. The initial term in each of Peirce's three trichotomies (qualisign, icon, rheme) and Trichotomy I (of the sign itself), pertain to Firstness which is the realm of oneness, quality and possibility. The second terms in the trichotomies (sinsign, index, dicent) and Trichotomy II (relations between sign and object) pertain to Secondness, and this is the realm of actual ex-


232


isting relations and reality connections.14 The third terms (legisign, symbol, argument) and Trichotomy III(how the sign is interpreted) are in the realm of Thirdness and are the most highly mediated, general signs appropriate for abstraction. The three types of interpretants outlined earlier also pertain to Firstness (emotional interpretant), Secondness (energetic interpretant), and Thirdness (language-based concepts). While all semiotic processes involve Thirdness (the sign and object brought together in the interpretant by a perceiver), Peirce's classifications of signs and of the trichotomies themselves move from relative Firstness to Thirdness. Qualisigns (the quality embedded in a sign regardless of whether it functions as one) pertain to Firstness while argument (symbolic propositions and premises) is largely Thirdness.15 Within the Peircian semiotic framework there are multiple combinations of relative Firstness, Secondness, and Thirdness. An indexical-legisign is a type of sign that combines the elements of Secondness (index, direct connection) and Thirdness (legisign, general type). Iconic and indexical legisigns are thus a kind of compromise solution falling mid-way between signs that function in the most direct unmediated way (iconic sinsign) and signs that function at the most general context-free level (argument) as shown in Figure 3. The vast majority of musical signs are of three compromise types: rhematic-iconic-legisigns; rhematic-indexical-legisigns; dicentindexical-legisigns. The aspect of generality provided by the legisign for each is, in fact, the cultural component, and a major defining facet of culture universally. The grouping of phenomena into general categories or types which, as we know, varies across cultural groups, is a primary foundation of culture just as token-type ("practice-structure,""parole-langue") dialectics are crucial to cultural transformation. Semiotic Hierarchies and a Theory of Musical Affect Within the Peircian framework, higher level signs and effects (Thirds, Seconds) contain the lower levels (Seconds, Firsts) (see Figure 3).16 In order to understand music's special potential for creating emotional effects, I am interested in probing the instances in which semiotic chaining is halted before reaching the level of Thirdness (symbol, argument, linguistic-based interpretants). Signs which are Firsts, Seconds, and Thirds, will be more likely to create effects at the same or lower levels of interpretant types. Thus, icons (Firsts) will most likely produce emotional, sensory interpretants (Firsts) at that point in the semiotic chain. Indices (Seconds) will produce energetic (Seconds) or alternatively feeling interpretants (Firsts). These types of signs, in and of themselves, will usually not produce higher language-



233

Figure 3. Class of Sign

Trichotomy 1

Trichotomy 2

Trichotomy 3

I

A

A

A

II III IV

B B

A A

V

C C C C C C

A B B A

VI VII VIII IX X

B

B B C C C

10 Classes of Signs Firstness Qualisign (iconic rheme) Rhematic iconic sinsign Rhematic indexical sinsign Dicent indexical sinsign

B A A B A B C

Rhematic iconic legisign Rhematic indexical legisign Dicent indexical legisign Rhematic symbolic legisign Dicent symbolic legisign Thirdness

Argument

A=Firstness; B=Secondness; C=Thirdness 1l=Firstness; 2=Secondness; 3=Thirdness

Firsts

I

HI II

Qualisign

Icon

Seconds

Sinsign

Thirds

Legisign

I

Index ~

Dicent Argument

Symbol

inclusion

Rheme

possibleinclusion

mediated interpretants (Thirds) at the point in the chain where they are being processed. This notion is key to my theory of music, emotion and direct experience. Since musical signs usually operate at the levels of Firstness and Secondness they will produce interpretants at these same levels in the chain where they occur. In contexts where these types of signs prevail and are the center of attention-for example in certain rituals, concerts, and dances-emotional and direct energetic effects can be prolonged, and movement to the level of Thirdness (language-mediated thought) postponed.17 Peirce shows that any general-purpose semiotic system must have icons, indices, and symbols, which is the case for language but not for semiotic modes like music and dance. Peirce was particularlyinterested in higher level signs, their operations, and effects. My emphasis diverges in that I am


234


interested in exploring the lower level signs of possibility and direct experience for the ways they create emotion and social identification. My theory of musical affectivity is based on the hypothesis that the affective potential of signs is inversely proportional to the degree of mediation, generality, and abstraction. To reiterate, lower level signs are more likely to create emotional and energetic interpretants, whereas signs involving symbols are more likely to generate language-based responses and reasoning--effects often described as "rational"or "conscious" responses. The point here is that different types of signs have different potentials. Musical Signs of Identity,

Emotion,

and Experience

Iconicity of Style: Signs of Identity Icons are, at root, signs of identity in that they rely on some type of resemblance between sign and object, as, in fact, do all relationships of identity. Steven Feld (1988) has discussed how iconicity functions to create social identity and aesthetic systems based on identification within the social and ecological environment. Musical forms that "sound like," that is resemble, in some way, other parts of social experience are received as true, good, and natural (Becker and Becker 1981).18 The dense "in sync but out of phase" quality of Kaluli or African Pygmy singing--individual variations and improvisations merging within the dense collective performance-'"sound like" the broader quality of social relations and are, in fact, based on the same ethics. I have made a similar case for Aymarapanpipe performance in Peru (1989, 1993). Feelings of iconicity or "naturalness" created through the correspondence of style across different practices are involved here. The subtle rhythmic patterns-basic to how we speak, how we walk, how we dance, how we play music-are unspoken signs of who we are, whom we resemble, and thus whom we are with. Conversely, divergences in kinesic and other features of social style directly identify outsiders, those who are not like us. Such signs are typically felt as relative comfort or discomfort with others in daily interaction. Sonic and kinesic iconicity, or lack thereof, however, comes to the fore in participatorymusical and dance occasions because in such occasions these signs are the focal point of attention. Indices: Signs of Experience and Emotion While some attention has been paid to the emotion-producing potential of iconicity in art, little theoretical work has been done in relation to indexicality. In fact, iconic and indexical signs typically operate together



235

in expressive cultural practices, and indices have their own special potentials for producing emotional response and social identification. One source for the affective power of musical indices is the fact that they are able to condense great quantities and varieties of meaning-even contradictory meanings-within a single sign. Indices signify through co-occurrence with their object in real-time situations. Once such indexical relations have been established, however, actual co-presence of sign and object is no longer required; the index may still call to mind objects previously experientially attached. But when former indexically related objects are not present, or even when they are, new elements in the situation may become linked to the same sign. Of key significance to a theory of musical affectivity, indices continually take on new layers of meaning while potentially also carrying along former associations-a kind of semantic snowballing. Hypothetically, the song that comes to index a romantic relationship, "oursong," may have a very positive emotional salience for the lovers when things are going well. This song initially may have been established as an indexical sign for the relationship (or the other) if the lovers heard it on their first date, their first dance, or when making love for the first time. Hearing it on subsequent occasions while the relationship was flowering it might have taken on additional objects in relation to those occasions, and continue to have a powerful positive emotional salience. It might carry both this salience and great sadness if the relationship ends in heartbreak. Hearing the song later in life, feelings of 'new love,' 'the many times together,' and 'heartbreak,' might be called up simultaneously creating a complex response. The multiple, sometimes conflicting, objects creating the interpretant by multivocal indices are not usually processed, at least initially, in terms of symbolic concepts. Rather we are moved to react in a visceral way because of the very complexity and incoherent form of the objects presented. Due to the very density of the objects called forth by the sign, we experience layers of feeling which will tend to remain undifferentiated and simply felt. The emotional power of such signs, of course, depends on the salience of the objects indexed. Indexical relations are grounded in personal experience; the members of social groups will share indices proportional to common experiences. Thus, indexical communication is most prominent in intimate groups such as married couples, families, close friends, and further down the continuum, in small close-knit communities or neighborhoods. Indices are grounded in one's personal and social life and thus are constitutive of identityboth in the sense of being part-and-parcelof ones personal past, as well as being signs of shared social experience. Moreover, the ability to communicate indexically within a family, a community, or a group of friends, what


236


Edward Hall calls "High-Context" communication, in-and-of-itself makes common experiences, and thus identity, patent.'9 The mass media and advertising redundantly create indexical signs, signifying [conjuring up] common experience and identity, beyond smallscale, face to face groups. This process underpins Benedict Anderson's idea of "imagined communities" in significant ways (1983). Nonetheless, the meanings attached to indices are not general or fixed. Unlike the meanings of symbols, which can be confirmed by consulting a dictionary or a math book, indices are fluid, multileveled, and highly context-dependent.20 The effects of indices can be guided by controlling the contexts of reception but they can not be guaranteed. This semantically ambiguous quality of indices is precisely the point of Louise Meintjes' article about the varied reception of Paul Simon's Graceland album among different groups inside and outside South Africa (1990). In spite of their rather unpredictable consequences, indices are frequently harnessed for the construction of social identities-in advertising, in mass political rallies and propaganda, and in ritual and ceremonies-because of their emotion-producing potentials and as pre-existing signs of identity. Like the intimate "our song" example, or the case of the Indian raga discussed earlier, indices often carry personal meanings, and thus our emotional investment in them tends to be higher than for general signs (symbols), especially when attached to significant aspects of our lives. They are "our signs," and they are the primary sign types that signify our personal and collective histories. As Frith observes, the music of adolescence and the teen years, when people are struggling with identity and other intense personal issues, tends to remain the most emotionally salient throughout an individual's life (1987); musical indices are at work here. Most importantly, as signs of Secondness, indices signify our personal and collective experiences in a particularly direct manner, they are "really"attached to events and aspects of our lives, and hence are experienced as real; they are signs of our lives, not signs about them. The Semiotic Potentials of Music Music integrates the affective and identity-forming potentials of both icons and indices in special ways, and is thus a central resource in events and propaganda aimed at creating social unity, participation, and purpose. In terms of the density of sign complexes music also has special potentials. Any musical unit is comprised of a number of components including: pitch, scale type, timbre, rhythmic motion, tempo, melodic shape, meter, dynamics, harmony (where applicable), specific melodies, quotes, genresall sounding simultaneously. Any of these parameterscan and often do func-



237

tion as discrete icons, indices, rhemes, and dicent signs which may be meaningfully combined to produce a macrolevel sign, although the significance of certain components may be foregrounded in the musical context. This multi-componential aspect of music can not be overemphasized as a basis of music's affective and semiotic potential. Within any given section of music the timbre may function as an icon or index with certain effects. The rhythm, meter, tempo, mode, melodic shape, and texture likewise may each function as discrete signs that compliment, chafe, or contradict the other signs sounding at the same time-contributing to the power of a particular meaning, to new insights, or to emotional tension, respectively. This aspect is in addition to other sequential juxtapositions of musical signs through time. I have discussed the semantic snowballing of musical indices, that is, one sign or sign complex gathering multiple objects to it simultaneously. The feature I am describing here is different. Music has the potential of comprising many signs simultaneously which, like other art forms, makes it a particularlyrich semiotic mode. The multicomponential nature of music functions in the same way, and can be a multiplication of "semanticsnowballing" in relation to the interpretant: the ambiguity or density of the sign complex discourages a response in Thirdness and encourages unanalyzed feeling. It is this multi-componential, and yet non-linear character of musical "sign bundles" that allow for a different type of flexibility in the creation of complex, densely meaningful musical signs that compound the condensation of meaning, the polysemy, and the affective potential. Social Frames and Interpretation: Rhematic and Dicent Signs in Art Because the concepts in Trichotomy III involve the manner in which a sign is interpreted, they depend on the social frame defining the type of interaction taking place. Gregory Bateson (1972), elaborated the concept of frame as metacommunicative conventions about how signs within a given interaction or context are to be interpreted. Erving Goffman (1974) and Richard Bauman (1977) extended this idea through the study of cues that signalled specific frames. Thus, a wink and other facial expressions might cue a "joking frame" indicating that a proposition spoken should not be taken literally as a dicent. Within a joking frame, linguistic propositions are interpreted as rhemes-signs of imaginative possibility. Similarly, we understand that the action on a theatrical stage should not be taken at face value whereas usually (i.e., without cues to the contrary) linguistic propositions, facial expressions, and body language in daily interaction are understood to be literal, and are interpreted as dicent signs of people's actu-


238


al attitudes. Even in theatrical contexts where we know "people are only acting," signs like facial expression and tone of voice that usually function as dicents may still be affective because of our habitual way of receiving them. Many signs in art are rhemes. A painting of an imaginary being or an imaginary or unknown person may be interpreted as a rhematic-icon because the painting only suggests the possibility of such objects without positing their actual existence. The use of musical icons representing 'bird calls,' 'bombs,' 'thunder,' or more abstract qualities such as a 'pastoral setting' are rhemes in that they signify these things as qualitative possibilities not specific existential instances. As signs interpreted as representing possible or purely qualitative objects, rhemes are crucial to the semiotic functions of art because they allow for the play of imagination and creativity. Rhemes can denote and represent what does not exist ('unicorn'), or what does not exist yet ('rocket ships' in early science fiction), but they are crucial to bringing new possibilities into existence by imagining and representing the possibility materially in art objects or performances. While the rheme allows for the concretization of imagined possibilities in art forms, dicent signs are particularlypowerful and convincing because they are interpreted as being really affected by the object they signify; they have a built-in "truth"value. A photograph or a realistic painting socially framed as a portrait of an actual person (e.g., with a linguistic title such as "KingHenry") are interpreted as dicent-indices. The social frame 'portrait' suggests that the camera or painter captured the image of the object ('the person') through co-occurrence with that (posing) person, and that the photo or painting was actually affected by the appearance of the object reproduced in the sign. Like a reflection in a mirror, portraits and especially photographs have a strong iconic component, but at the next stage in the semiotic chain it is their identity as dicent-indices that make us interpret them as real representations, as "true." Dicent-indices are central to the power of musical performance. Roland Barthes' influential concept of "the grain of the voice"-the direct connection of body to body through certain ineffable sonic qualities of performance-is grounded on dicent signs (1977). In musical performance we often interpret the volume, articulation, and quality of musical instruments or voices as signs of the "true"sincerity, emotional state, care, or training of the performer (possible objects). Facial expression, gesture, and physical attitudes are likewise important dicents for the "inner"attitudes of performers ranging from 'cool control' to 'deeply felt passion.' Directly parallel to "body language"and "tone of voice" in everyday interactions, we often interpret sonic signs of vocal and instrumental quality as actually being affected by the actual attitude of the performer (object) and thus under-



239

stand them as "true."Vocal quality is particularlyconvincing in this regard whereas instruments comprise a second layer of mediation between the performer and listener. This may be why vocal music is so predominant in popular music. In my classes, undergraduates are particularlyoffended if I suggest that their favorite popular singer may not have actually experienced what she is singing about or may not, at the time of performance, be in the emotional state signified to the listener through the sonic qualities that they interpret as dicent-indices. They often do not accept my analogy that professional singers can operate like actors who train themselves to reproduce given emotional cues for the effectiveness of their art. When my students take such cues literally, these signs are operating as dicents rather than rhemes of possible emotional experiences; they are thus more affectively powerful because they are interpreted as real. For many music genres in our society, especially in the popular music field, a common assumption is that musicians really mean and are experiencing what they express through "the grain of the voice" and through physical cues.21 That is, unlike acting, musical performance in many popular genres isframed to be taken literally as emotional expression. In short, these types of sonic and physical dicent signs are powerful for us because they are interpreted as being the direct result of the feelings they express, and because they operate below the level of propositional speech which is more likely to invite us to assess truth or falsity. We know words can lie or be mistaken. Within many social frames, popular musical performance often being one of them, we habitually take dicent signs at face value and we believe them. Like paintings and photos, however, the social framing of different types of musical performances, recordings, genres, and artists may cause them to be interpreted as rhemes, or conversely dicent signs, in relation to the artists' attitudes during "the performance." In genres such as blues, "folk,"soul, and "roots rock," performance is typically framed in relation to authenticity of feeling (dicents). In other genres where artifice is more pronounced, or where aesthetic ideologies emphasize the separation of art and life (see Bourdieu 1984), the interpretation of emotional cues as rhemes (i.e., signs of possible emotions) may be more likely. Whether a sign functions as a rheme or dicent in relation to emotional cues, however, depends on the experience, knowledge, attitude, and even desire to suspend disbelief, on the part of the perceiver as well as on the skill of the artist to communicate with emotional cues. When my students take their favorite star's signs of emotion literally they are operating as dicents, whereas when I question their literalness, the signs are operating for me as rhemes and thus are less convincing.


240


Musical recordings are likewise framed differently in relation to their character as dicent signs or rhemes in relation to "live performance."When framed as "field recordings" or "live concert recordings" we are more likely to interpret the sounds as dicent-indices with the microphone(s) having a similar relation to the object as a camera lens. Earlierin the history of recording, the label "high fidelity" was meant to frame a given recording as a dicent-index in relation to "live" performance. The interpretation of recordings as rhemes or dicents of live performance, however, depends on sonic cues (e.g., audible double tracking, electronic effects, "naturalness") as to the degree of studio manipulation. I have often heard people assess a recording in regard to whether the band could actually reproduce the same sound "live"on stage; within certain styles this remains an important criteria of authenticity. In music framed as studio art (e.g., the late work of the Beatles or computer music) however, there is no pretension of representing a "live"performance and we respond to the music rhematically in relation to various possible objects. "Studio music" is based in a different aesthetic and is assessed according to different criteria than participatory music, live stage performance, or "high-fidelity"recordings framed as dicent-indices of live performance. The point here is that the power of musical dicent signs will depend on the framing of the genre, style of representation, and perceptible levels of mediation. Thus, Kalulifield recordings, and Bruce Springsteen or Aretha Franklin recordings are constructed to be interpreted as dicents of actual performance whereas much of David Bowie's work, disco of the 1970s, and Cage's electronic music are not (Frith 1987). Whether musical signs will be interpreted as dicents or rhemes of the performers' actual feelings will often be influenced by whether the form of presentation as a whole is interpreted (framed) as rheme or dicent. Nonetheless, as in theatrical presentations, we may still respond to certain musical signs (e.g., an emotive cry) as dicents regardless of the frame and level of mediation because of our habitual way of interpreting such signs. Dicent Signs and Social/Musical Synchrony Much of what Feld (1988) has discussed as iconicity, such as in-syncbut-out-of-phase singing among the Kaluli, does indeed have a strong iconic component; we feel a likeness with others and with our social environment. A more important source of social relations and identity generated through musical performance, however, are dicent-indices. It is important to remember that the same signs can function at different levels, and may have different semiotic potentials, within the semiotic chain. Hence, looking in a mirror has a strong iconic component, we "look like" the image of


Turino: A Peircian Semiotic Theory for Music

241

ourselves that we have seen before in mirrors. When I look in a mirror and see a smudge of dirt on my cheek and automatically move to wipe it off, however, I am reacting to the mirrorimage as a dicent-index, a sign of actual relations and fact; I automatically believe what I see. It is this "truthvalue" of dicent-indices that provides particular power for musical signs of social identity. Edward Hall's discussion of rhythmic synchrony-moving and/or sounding together-is key to the way dicent-indices create actual experiences of social identity, unity, and participation (1977; see also McNeill 1995). In music and dance when things are clicking, timing, attack, articulation, and body motion are guided in relation to other performers in the situation. Culturally specific styles of fitting in with others during performance, be it strict rhythmic unison, Keil's rhythmic "participatorydiscrepancies," or interlocking, likewise comprise experiential signs of the quality of the relations involved. The signs that emerge from each performer's manner of interacting sonically and kinesically affect and are directly affected by the kinesic and sonic signs of others. These signs are dicent-indices to the extent that they signify levels of stylistic competence and the nature of the social relationships within performance (object) and are simultaneously affected by those relationships and degrees of shared cultural knowledge (competence). When music makers and dancers are in sync, such signs move beyond felt resemblances to experienced fact of social connections and unity. While dicent-indices function like this in everyday interactions through "body language," tone of voice, etc., these signs have particular power in participatory music and dance performance because non-propositional sonic and kinesic signs are the focal point of attention. Thus direct kinesic and sonic response to others may well be experienced as a deep type of communion, although one can rarely fully express the feeling in words. Here we come to a prime difference between emotional and energetic interpretants, on the one hand, and the effects of symbols on the other. What Feld, Keil, Hall, and others point to is the deeply felt, yet often unspoken, experiences of being of a group through the "naturalness"of iconic signs and the direct experience of dicent-indices. Propositions and linguistic arguments about identity may even become emotionally heated, but because they call for mediated, word-based evaluations, they do not provide the feeling or direct experience of belonging; rather they are claims and arguments about belonging. The other arts involve iconicity and indexicality and have their distinct potentials for creating emotion and identity. But they typically do not engage large groups of people collectively in the actual doing of the activity that results in the experience of social synchrony. This is one of the special potentials of participatory music and dance.


242


Some Musical Examples Jimi Hendrix's Woodstock performance of "The StarSpangled Banner" provides an excellent example of how musical meaning and affect are created through icons and indices. In what I (following Greg Urban) will call creative indexing, Hendrix musically shifts the meaning of the national anthem much as was accomplished by patching the seat of blue jeans with the American flag during the Vietnam era. Creative indexing involves the juxtaposition of two or more indices in novel ways that play off of the original meanings of the signs. Another example would be the wearing of a tuxedo (indexing formality) with red tennis sneakers (informality). The original meanings of the indices creatively combined is further dependent on their identity as legisigns, that is, signs of a general type as defined within specific cultural contexts. Here, the legisign can only operate within groups who share the same experiences with the indices, hence, the meaning of the legisign may be quite varied across specific groups. In Hendrix's musical performance, the very use of the loud electric guitar with feedback and distortion (an indexical legisign of the rock counter-culture and all that it, in turn, signified in the late 1960s) for the performance of the anthem (indexically associated with nationalistic contexts) creates a new meaning specifically within the broader social context of the time. For me as a listener, "text-painting"such as the use of bomb sounds for the line "bombs bursting in air" (the text was not sung but is generally known), and sounds iconic for planes, sirens, and destruction added to the significance of the anti-war protest. At one point "Taps"was quoted, and in that context functioned as an icon (the quote) that carried indexical meaning through its usual association with funerals. Shifts of accent, rhythm, and phrasing, and the use of rock-riffconventions expressed sarcasm, again as creative indices at the microlevel in support of the macrolevel sign (the piece as a whole). My guess is that these aspects of articulation that expressed sarcasmwill operate for many listeners as dicent signs (i.e. signs directly affected by Hendrix's inner feelings), but this would have to be verified through discussions with listeners. Upon hearing a recording of this performance in 1989, the indexical meaning of the performance and its various microcomponents would be compounded in multileveled and conflicting ways ("snowballing" effect) if you had been at Woodstock, met a girl, and fallen in love; listened to the piece on records with her remembering Woodstock; and later lost her after being drafted and going to Vietnam. It might likewise be compounded in a starkly contrasting manner if you had lost your son in Vietnam and had thus developed a hatred of hippies. Using this example in university music classes, I found that the younger students, not yet born during the late sixties, reacted in a variety of ways.



243

For some, the performance indexed the World Series! For others it was deeply offensive because of the disrespect communicated by the creative juxtaposition of distortion (dicent signs probably operating here) and the national anthem; still others found the creative index funny or somehow disturbing (direct energetic interpretant = laughter). For an older man in the class, whose son was hurt by using drugs during that time, the performance indexed, and actually was iconic for, being under the influence of drugs (he said: "it sounds like being high on drugs"). This is what I meant earlier by the "looseness of reference" of high context (iconic and indexical) signs. The same sign may have radically distinctive meanings depending on personal experiences and situational context. What was striking in the class, however, was the depth of emotional response that the recording elicited. While the Hendrix performance itself is a particularly "programmatic" one that might be dismissed as atypical, the affective power of meaningful moments in music is due to the same process of combining multiple iconic and indexical legisigns, as well as dicents, in artful ways. The passage from the third to the fourth movements in Beethoven's Fifth Symphony, which for years frequently produced a physical response in me (chills in the spine), is a particularly clear example. In this passage a swirling or staggering sensation was created by the juxtaposition of ascending and descending melodic motion. The placement of rests and then their absence gave me the sense that the motion was accelerating. In this fashion, the melody rose out of the soft, low, dark timpani and strings, shifting from minor to major. At the end of the transition, a crescendo was added to the static and yet agitated bowing of strings, and arrival to the fourth movement was marked by the entrance of loud brass. Here, creative indexing (e.g., the major/minor and timbre contrasts) along with the use of mutually reinforcing conventions (e.g., crescendo with or following ascent) were involved in creating music that had a particularly powerful, emotional and energetic effect on me as a listener. The same types of processes and conventions (legisigns) are used in music generally to produce their effects. In the Hendrix and Beethoven examples, it was the juxtapositions of signs within the piece that produced the overall effects. In other cases a mere moment of music, apart from internal structural relations, may have a powerful effect. For example, several years ago upon hearing only the opening chords of Crosby, Stills, Nash, and Young's song "Ohio" on the radio I had a strong emotional reaction as well as a physical response (tingling of my spine), that was totally incommensurate with my feelings about the song when it was popular, or with feelings that I reflexively recognize about the deaths at Kent State. The fact that I responded upon hearing only the opening chords indicates that Meyer's (1956) psychological theory


244


regarding the inhibition of expectations to create musical affect does not apply (although, conversely, his theory may be understood in regard to indexical processes). I was driving my car and not paying much attention to the radio at the moment the piece came on, but it created strong emotional and energetic interpretants almost instantly. I am still not sure of all that was being indexed by the opening chords of "Ohio,"but this is precisely why the music was able to produce such a strong affective response: perhaps an important period of my life, political feelings of that time, or some other events were being recalled by an extremely condensed sign. The quality of the opening chords as iconic for a particular feeling may also have been involved. Since I can not link the objects of this sign with symbolic concepts, I am unable to say what was indexed. The objects were, nonetheless, lodged in me since a strong reaction was produced. The objects indexed by "Ohio"were of a different part of my memory and experience not approachable through symbols, only through the more basic iconic and indexical signs. The fact that certain parts of ourselves are only available through pre-symbolic signs is precisely why we need art and music, media that operate largely at the iconic and indexical levels. It is also why humans need distinct realms of practice which foreground the different semiotic levels of iconicity, indexicality, and symbolism to achieve subjective integration of the "whole" person. The distinct functions of different types of signs to link varied parts of ourselves and our experiences explains why pre-symbolic systems like art, music, and dance have been maintained in societies universally, even when the "more complete" semiotic mode of language (having icons, indices, and symbols) is available. My experience with the song "Ohio"was real and, I would guess, not atypical. For music scholars this type of phenomenon is only banal if we choose to ignore how music is really operating for most people in the social world. The import increases when we consider that the affective potential of music is constantly utilized, and in some cases manipulated, for a variety of highly significant social ends including the mobilization of collectivities to create or defend a nation. Rhemes as Signs of Possible Identities: The Case of Zimbabwean Nationalism Perhaps Benedict Anderson's (1983) celebrated idea of "imaginedcommunities" has been so widely applied to other types of identity formations because it pinpoints the fluid, constructed, semiotic nature of social identity in units that can not rely on face-to-face interaction. The general idea is relevant in some ways to the dispersed transnational immigrant groups



245

and diasporic identities that have become increasingly common in the contemporary world. As signs of possibility and imagination, rhemes are key to the construction of new social formations-they make the imagined possibility materially patent and public, thus helping to bring the possibility to fruition. Taken as a whole, a painting of a winged horse is an icon which might create the sense of an imagined ideal (object) such as 'flight/freedom' plus 'strength/ nobility.' The icon at the macrolevel is actually constructed through creative indexing: indices of birds22fused with the image of a powerful horse. Musical styles can function very much like the picture of a winged horse. In fusing the indices of various pre-existing social groups into a single rhematic-icon, the resulting sonic image projects the imagined possibility of forging a new synthetic social group and identity. Basic to nationalist movements generally are the processes of modernist cultural reformism-the blending of 'the best' of local culture and foreign 'modern' culture to forge images of a new 'modern' yet locally distinct "nation"(Gellner 1983; Chatterjee 1986; Buchanan 1995). This process is based on the manipulation and new combinations of indices to create a rhematic-icon as the macrolevel sign for a new social identity. Beginning around 1960, Robert Mugabe and the young middle-class nationalist leadership in Zimbabwe consciously set about to construct a unified African nation and party out of the disparate localist-indigenous groups in order to challenge white domination.23 Initially they did this through mass rallies, and observers of the time did not miss the fact that Mugabe's emphasis on indigenous music, dance, and ritual was specifically directed towards creating an emotional tie between the masses and the movement (e.g., Shamuyarira 1965:67-68; Bhebe 1989); the emotional potency of these types of semiotic practices has been theorized above. In the rallies, music and dance traditions that indexed specific indigenous social groups ('tribes') and regions were repeatedly juxtaposed on stage, and framed by symbolic discourse about the party and nation. Through repeated juxtaposition of these localist signs in a new context, their objects (specific regions and social groups) may become indexically linked to each other with the intention of creating a concrete image of a new composite social unit. When framed by symbolic discourse about party and nation within the rallies, they may become indexically linked to this new imagined entity: "the Zimbabwean nation.'"24Although not theorized in Peircian terms, this was the explicit intention of the nationalist leadership; they consciously tried to control the new linking of pre-existing localist indices of identity to the new objects of party and "nation." In the context of these rallies, in 1963 a new musical genre called the gallop was created by the cultural arm of the nationalist party. They were


246


also involved with creating a new national costume that would blend 'old' indigenous elements with 'modern' style. Like the costume, the gallop is described as incorporating indigenous rhythms and musical features into the context of a 'modem' electric guitar band. The creative juxtaposition of localist and modernist indices, typical of modernist reformism, was intended to serve as an icon for the new locally unique, yet 'modern' nation. Although undertaken for professional as well as political reasons later, Thomas Mapfumo's incorporation of mbira and other indigenous Shona genres into his guitar band repertory repeated this pattern. By realizing a visual and/or sonic image in the world through an artistic style (rhematic-icon), the new imagined entity becomes real in-so-faras it now has a public, concrete representation. Whereas the winged horse will probably remain in storybooks and people's dreams, the rhematic-iconic image of new imagined identities, when perceived as a unified image, and experienced as a unified quality, can influence how people perceive or construct their own identities dialectically. In short, rhematic-icons can be used to realize imagined realities, and through their actual representation can influence social reality by showing the way to what might be possible. But because they operate initially at the level of Firstness, icons tend to create emotional interpretants. The indexical components of such icons (wings, horses, musical indices for certain existing social groups and places) add the level of Secondness, that is, of existential phenomenon, a reality component. When properly constructed within the proper social context, such complex rhematic-icons can be felt (affect as interpretant) to represent reality: possibility can be felt (Firstness) or experienced (Secondness) as reality. At the right time and place, the image of a winged horse, the devil, a new social identity, or even a radically different political order might be made convincing through iconic and indexical means. Church leaders, politicians, and revolutionaries have long taken advantage of this. Immigrantand diasporic communities similarlyblend indices from the original home, the new home, and from members of the same diaspora elsewhere to create new composite signs to articulate who they are. For the people involved, the perception of theses signs may differ, however, when artistic styles emerge from grassroots processes rather than being consciously constructed by political vanguards. Dicent-Indices As Signs of Actual Identity: The Case of Peruvian Chicha The Peruvian cumbia andina, or chicha, genre provides a typical example of how creative indexing operated in the creation of a new musical



247

genre that stood for an emerging social identity. In contrast to the Zimbabwean case, the creation of chicha was not consciously constructed by a political elite, but rather came into existence more organically out of the experiences of highland Andean migrants, and especially their children, in the "foreign"locale of Lima. Chicha involved the juxtaposition of musical components such as scale types, phrase structure, and melodic shape from the highland wayno (a song genre indexing Andean social identity) with the rhythm of the urban cumbia (indexing urbanity among highland migrants), and finally electric instrumentation (guitars, bass, synthesizers) as an index for youth and modernity. The meaning of each of these specific indices depended on the fact that they were also legisigns. It is the particular significance of "cumbianess," "waynoness" and electric instruments as types specifically among the migrants that allowed them to signify what they did.25 Chicha largely pertained to the teenage children of Andean migrants in Lima. Although using other terminology, Peruvians themselves recognized that chicha serves as a logical sign for the identity of second-generation Andean migrantsin Lima.It organically fused indices that mirroredtheir ambiguous social position. Chicha fans were of highland parentage but not highlanders themselves; they were born in Lima but were not accepted as true Limefios. Out of this position a new sense of identity emerged for the children of Andean migrants that coalesced around chicha. The rock influences in the instrumentation indexed the age set of cumbia andina fans, and the vocal style functioned as a dicent sign for the youth and class of the star singers (Turino 1990). The same process of creative indexing was involved in constructing what, at one level, may be understood as an rhematic-iconfor the new social group. The organic emergence of chicha, however, has led some Peruvian migrants to actually interpret the style as a dicent-indexical sign. When signs, in this case a musical style, are interpreted as organically emerging out of a particularsocial position, they function as dicent-indices which are really affected by their object-that social position. In this case the music is experienced directly as real signs of an existing identity. This may be an important distinction for analyzing the effects of processes used in the purposeful construction of new possible identities in contrast to cases where styles emerged as a result of pre-existing identities and sensibilities perceived to have organically given birth to those practices and styles. The notion of "authenticity" has been highly contested recently in popular music studies and ethnomusicology as the basis for evaluating styles. Authenticity relates directly to how signs are interpreted within given social contexts (i.e., their character as rhemes or dicents). Dicents are interpreted as authentic and "true"because, like the weathervane, they are


248


understood as being directly affected and often the result of what they signify. While this is a subjectively and culturally specific matter, the rheme/ dicent contrast gives us a basis to think through claims about authenticity and the effects of "authentic" expressive cultural practices in relation to identity. Thirdness and Musical Experience By stressing the importance of Firstness (icons, rhemes) and Secondness (indices and dicents) in the preceding sections I do not mean to denigrate symbolic processes and Thirdness (legisigns, symbols, argument). Obviously symbols-language-are profoundly important for human understanding and social life. In relation to music, symbols are necessary to conceptualize and discuss aesthetics and affectivity in general or comparative ways, as I am bound to do in this paper. Through good discussion, symbols may open new possibilities for understanding broader human experience and the other sign modes themselves. After being trained about music through symbols, some individuals and social groups take particular pleasure in experiencing music at the level of symbolic thinking and discourse. Leonard Meyer offers some interesting observations in this regard: Whethera piece of music gives rise to affectiveexperienceor to intellectual experiencedependsupon the dispositionandtrainingof the listener.... Belief [framing]alsoprobablyplaysan importantrolein determiningthe characterof the response.Thosewho have been taughtto believe that musicalexperience emotionalandwho arethereforedisposedto respondaffectivelywill is primarily probablydo so. Thoselistenerswho havelearnedto understandmusicin technicaltermswill tend to makemusicalprocessesan object of conscious [read: symbolic]consideration.This probablyaccountsfor the fact thatmost trained critics and aestheticiansfavorthe formalistposition. Thus while the trained musicianconsciouslywaits for the expected resolutionof a dominantseventh chordthe untrained,but practiced,listenerfeels the delayas affect(1956:40). It is not surprising that technical musical training that relies heavily on symbols creates listeners who sometimes respond at the level of Thirdness. Meyer's clear division between an "intellectual"and an "affective"response precisely underlines what I have been suggesting about the different nature and effects of signs in Peirce's three categories, although I would like to avoid his thought-emotion dichotomy.26 Leonard Meyer's comments also speak to the semiotic ambiguity and the almost mystical quality that music may have for average "untrained" listeners. This is precisely because such listeners do not have symbols to rationalize or "domesticate" musical events at the level of Thirdness, that is, to understand them as a part of generalizable, predictable, mediated



249

categories and processes. Even people who do have a vocabulary to talk about music can not always domesticate musical experience with symbols, as in my case with the song "Ohio," or they may choose to compartmentalize different types of experiences. Ultimately, it is the more ambiguous nature of indices and the fact that icons and indices are not organized in a distributional-grammaticalsystem that allows them to be juxtaposed within sign complexes that compound the polysemy.27 Experiencing music through symbolic discourse is certainly valid and it can be useful. My point is simply that it is extremely different than experiences of Firstness and Secondness and it is a minority musical experience in our society and for most societies in the world. Thus, if we want to understand what music is doing, can do, for most people we will attend to these other types of semiotic processes and categories of experience.

Conclusions Peirce's semiotic theory helps in the analysis of signifying musical forms. But its even greater significance is as an avenue for understanding musical affectivity, different parts of ourselves and experiences, and the special potentials of music for the construction of personal and social identities. The frequently mentioned "mysterious," "untalkable"quality of music (its lack of domestication by symbols), allows for a heightened looseness of reference, and personal and group appropriation. Music has a great multiplicity of potentially meaningful parameters sounding simultaneously, and its status as a potential collective activity helps explain its particular power to create affect and group identities. Each sign component discussed has specific potentials. Symbols and argument are necessary for general and theoretical discussion and to expand synthetic understanding. The special power of symbols and argument in relation to these ends is specifically based on the precision with which the symbols and premises are defined (i.e., rooting out polysemy). This paper is entirely an exercise in symbolic discourse and argument. Lower level signs have other potentials which, conversely, are importantly based in their openness and fluidity for interpretation. Icons are signs of feeling and, in combination with rhemes, allow for the concretization of imagination which is key for making new realities possible. Writers who have used Peircian theory to think about music and art have emphasized iconicity and rhemes, signs of Firstness (see Kaelin 1983). In this paper I have emphasized signs of Secondness, indices and dicents as signs of reality connections, social relations, and direct experience. Although thinking in terms of Peirce's three basic categories of phenomena, pinpointing the "natural"quality of icons, and the snowballing,


250


multileveled, polysemic, ambiguous, condensed, and personalized nature of indices helps me understand music's affective potential, speaking in such terms can ultimately only point to the general ways such experiences happen. Even when Peircian analysis does shed new light, it does not satisfy, and it cannot demystify our most profound musical experiences. When people shift to symbolic thinking and discourse to communicate about deep feelings and experiences, the feeling and reality of those experiences disappear and we are not satisfied. This is because we have moved to a more highly mediated, generalized mode of discourse, away from signs of direct feeling and experience. This is Charles Seeger's dilemma about the "untalkables" of music, and the very point of my paper. Symbols do not pertain to all parts of ourselves, and they fall short in the realm of feeling and direct experience. This is why we need music.

Acknowledgements I was first introduced to Peirce and the potential of his theories for thinking about music by Professor Greg Urban during graduate school at the University of Texas, a gift I gratefully acknowledge. An earlier version of this paper was given at a special colloquium at the University Chicago in 1989. Two expanded versions were written for specific seminars at the University of Illinois in 1995 and 1996. This version was written for a colloquium at the University of Texas at Austin in February 1998; I would like to thank my host, Veit Erlmann. Too numerous to name, many students and colleagues have offered criticism and helpful comments on the earlier versions which gave rise to this one. I would like to especially thank Joanna Bosse, James Lea, Fernando Rios, and Chris Scales for the ongoing dialogue, as well as Charles Capwell, Steve Hill, Chris Morosin, Jon Sterne, and Rob Templeman for central insights offered and inspired.

Notes 1. I have concentrated on Peirce's work about signs and his basic philosophical categories, which comprise only part of his abundant philosophical and scientific writings. In these realms, at least, I have tried to be true to his mode of thinking. At times I extend his concepts to suit my purposes of developing a theory of music, emotion, and experience. I continue to elaborate the workings of iconic, indexical, rhematic, and dicent signs. 2. "Semantico-referential"denotes the capacity of using language to refer to, define the meaning of, and actually create, other signs in language. 3. Peirce suggests two types of objects. One is the existential object out in the world which he called the dynamical object (an actual tree stood for by the word tree). The second type is the immediate object, the object as the sign represents it and as contained within the mind. Thus, the general mental concept 'tree' is the immediate object for the word tree. The immediate object more or less corresponds with Saussure's "signified"within his binary concept of the sign. 4. There seems to be a general notion that the relationship between "signifiers"and "signifieds"has changed due to historical conditions in the so-called postmodern period. For example, Grossberg writes that "not only can we no longer confidently read the meaning or ideology of a text off its surfaces, but even the notion of a single identifiable fixed text is prob-



251

lematic-it is also a matter of historically different conditions, of the changing spatial and temporal complexity of the cultural terrain itself" (1988:19). While I agree with ideas about the fluidity of certain types of signs, I do not think that the character of semiotic processes has changed historically. Within the Peircian framework, certain types of signs must have relatively fixed meanings to function (e.g., symbols), whereas others are by nature fluid and contextually contingent (e.g., indices), and this is so regardlessof historical period. In my view, what has happened is a recent emphasis on, and attention to, fluid sign types whereas previously writers (e.g., Saussure and the structuralism that followed) overemphasized relatively fixed types. It was an analytical error to assume that all sign types functioned like Peircian symbols, which are akin to Saussure's idea of the sign comprising a unified "signifier"and "signified." 5. Peirce defines the dynamical interpretant as "the direct affect actually produced by a sign upon an interpreter of it" (Peirce 1960:4.536). 6. Emotional and energetic interpretants can become signs inspiring language-based interpretants at later stages in the chain of semiosis. For example, a loud explosion nearby (sign) might cause an immediate sensation (emotional interpretant--sign) setting off an adrenalin flow (energetic interpretant--sign) making one jump (energetic interpretant--sign) generating another unreflexive action such as spinning around to look for the source of the noise (energetic interpretant---sign) which might lead to blurting out, 'What the hell was that?!' (energetic interpretant--sign) generating language-based attempts at an answer (languageinterpretant--sign .. .). All of this can happen in an instant and involves different aspects of the nervous system. Even when reaching the linguistic-based signs and interpretants, earlier interpretants and signs, such as lingering sensations of surprise or fear, fast heartbeat, etc. can reinsert themselves in the chain (be noticed and cause a new effect) if they are strong enough. Thus a linear progression from emotional to energetic to language-basedinterpretants can not be assumed, and it becomes pointless to differentiate mind and body since the brain and nervous system are part of the body. 7. For example, in ethnomusicology the concept of metaphor has sometimes been used in cases where the hierarchical organization of an ensemble is seen as metaphoric for the broader social structure (e.g., Waterman 1990). In Peircian terms this would simply be an image-icon if the hierarchy in the ensemble called forth the object of 'general social hierarchy' for a perceiver. This case does not involve the creative juxtaposition of signs which posit a similarity between their distinct objects, but rather a perceived similarity between sign and object. 8. In fact, the word symbol is used in so many ways and its meaning has become so vague that it no longer serves well for semiotic and cultural analysis. I prefer and strictly use Peirce's specific concept. I also understand the reluctance to accept new technical meanings for such a widespread, general term. To understand the Peircian framework, however, such translation is necessary and, I have found, ultimately beneficial. 9. Some parts of speech, however, are not symbols. For example, pronouns in general, and especially demonstrative pronouns function as indices to other words in the utterance, or to elements in the environment: 'that chair.' Demonstrative pronouns are not symbols because they have no general object. Likewise, articles are not symbols. 10. This is why we must use language to communicate in general, relatively contextfree, terms about everything else-Charles Seeger's (1977:16-30) linguocentric predicament for musicology. 11. Peirce noted that linguistic propositions are all dicent signs, in this case, dicent symbols. Peirce goes on to discuss the dicent sign as a proposition that "professes to be really affected by the actual existent or real law to which it refers" (1955:104). Thus, the proposition, 'the rose is red' is affected by the "laws"or conventional meanings behind the concepts of 'rose' and 'red' and to the existence of those qualities in an actual flower. 12. A "wince" initially functions iconically to express pain or displeasure because it "looks like" other expressions of displeasure. Primarilyit is indexically related to emotional


252


states through co-occurrence. All indices depend on an initial iconic moment of recognition linking token and type to form the legisign. EarlierI gave the example that a rising melodic line and crescendo might function iconically in relation to excited speaking voices. This is true at an early part of the semiotic chain, but the real impact of these signs is based on the fact that rising pitch and volume when speaking co-occurs with excited states and we interpret these signs as being the result of excitement (the object). Thus at a later point in the chain the legisigns 'rising pitch' and 'crescendo' function as dicent-indices, and it is this character that explains their affective potential. 13. Often the effects of such signs are described as "unconscious" or "subconscious"; "consciousness" being delimited by linguistic thought and communication (e.g., Bateson 1972:141). From a Peircian perspective, for signs to function they have to be apprehended, although not necessarily through linguistic-based interpretants. The concepts emotional and energetic interpretant are useful in that they allow us to talk about different types of awareness and semiotic effect, avoiding the typical mind-body, emotions-thought, and consciousunconscious dichotomies. All three types of interpretants involve mental (which is also a physical) activity. 14. It is no accident that among all Peirce's categories the signs in Trichotomy II have received the most attention in social and ethnomusicological theory, and that the second terms in the latter two trichotomies are particularlyimportant for sociomusical analysis. Secondness is the realm of connections to the world, and the "reality"functions of signs. 15. The actual semiotic status of the qualisign is a matter for further reflection and debate. From one point of view it is pure Firstness and does not function as a sign at all inand-of itself until it is embedded in an actual sign, sinsign, at the level of Secondness. Nonetheless, it represents semiotic potential of pure quality which may be analyzed apart from its actual embodiment. 16. Logically it can be seen that Firsts stand alone; Seconds involve a direct unmediated relation between a first and a second; and Thirds bring a first and a second into a mediated or synthetic relationship. Thus higher level entities logically contain the lower. 17. Peirce has convincingly demonstrated that his three basic categories could be applied to all phenomena (e.g. 1992:145-284). It is certainly possible to extend them to experiential states in individuals.A state of Firstnessis when there is no consciousness of self, other, or world-when "trainsof thought," or "chains of semiosis" have not begun or are kept from beginning. Such states are achievable, for example, through certain meditation practices as well as other activities including dance and musical performance. This state can not be apprehended until after the fact, in memory; recognize it and you are already outside of it, and it is gone. States of Secondness involve experiences of intense concentration on one thing, person, or activity (your instrument, your dance partner, your lover, a tennis ball) such that no mediating language-based thoughts about self or the entity of attention occur. Intense experiences of Secondness are what MihalyCsikszentmihalyicalls "flow"(1988). He has elaborated a theory of "flow," particularly useful for musicians, involving an escalating balance of challenge and competence which he finds essential to intense, sustained experiences of concentration (Secondness). States of Firstness and Secondness, are often deeply satisfying and relaxing because the self and mundane matters of the world are temporarily transcended (not "thought" about). States of Thirdness involve language-based propositional and semanticoreferential thought and social interchanges. 18. More accurately, the signs of highly iconic musical forms are often simply taken for granted because they match one's broader experience and sense of reality when properly performed according to the cultural norms of the style ("good music"). This taken-for-granted quality is underlined in "poor performance," that is, when the iconicity is disrupted, thus often generating criticism. 19. Hall proposes a continuum between what he calls "High-Context"and "Low-Context" communication. "High-Context"communication is when much of the information is contained within the context, including the context of individual memory, rather than in the



253

code. Without using Peircian terminology, what he is referring to is indexical process. In "LowContext" communication most of the information is carried by the code, which presupposes its basis in Peircian symbols. 20. From the Peircian perspective, postmodernist assertions about the new fluidity of signs-the unmooring of signifier from signified-are non-sensical. Regardless of the historical period, indices have always been rather fluid, polysemic, multileveled, and non-guaranteed. Symbols, if they are to function well as symbols, by contrast, must be relatively tightly bound to their objects ("look it up in the dictionary"). The problem is that postmodernist semiotic discussions do not distinguish between the distinct nature of different sign types. 21. In Distinction (1984), Pierre Bourdieu theorizes this strong connection between art and life in popular aesthetics, and contrasts it specifically with elite Europeanaesthetics which are based on the notion of the separateness of art and life. 22. Here the wing of a bird is an index for the whole animal because of co-occurrence. 23. Robert Mugabe, who became the leader of the victorious ZANU(PF)party and president of Zimbabwe after independence in 1980, served as Publicity Secretary of the early nationalist parties. He played a central role in initiating cultural nationalism. 24. As in this case, the analysis of how different types of signs function in relation to each other for given purposes is crucial since each (e.g., rhematic icons, indices, and symbols) have different potentials. In the 1960s in Zimbabwe, the noun "nation"was certainly a rhematic-symbolic-legisign; not only was it general but at the time it was a yet unrealized possibility. 25. Among people in the highlands, 'wayno' does not signify 'highlands' but more specific things. Likewise, in Colombia, its place of origin, 'cumbia' does not necessarily index urbanity any more than electric instruments signify 'modernity' in the United States. 26. In many places, musical training largely consists of iconic-indexical processes: direct imitation of a teacher or model with few words spoken. In some situations, Indian classical music for example, learning takes place through direct imitation, yet there is also a (somewhat separate) tradition of theorizing about music through symbolic discourse. Certain social groups, the Aymaraof southern Peru for instance, do not have a highly developed symbolic system for talking about music and do not seem interested in doing so. They also do not have formal music training; people primarily learn by doing within actual festival performances. It is also interesting that, unlike that of their Quechua neighbors, Aymara music is primarily instrumental. It would be interesting to do cross-cultural and cross-group studies on the relative use of symbols in musical life (training, discourses about, song texts) and preferences for the different sign modes in musical experience. 27. The distributional or grammatical meaning of words reduces polysemy. For example, the meaning of a linguistic form that may either serve as a verb or a noun will be distinguished by grammatical marking, word order, suffixes, etc., to indicate its identity: "The light light lights lightly."

References Anderson, Benedict. 1983. Imagined Communities. London: Verso. Barthes, Roland. 1957. Mythologies. New York: Hill and Wang. . 1977. "Grainof the Voice." In Image Music Text, 179-89. New York: Noonday Press. Bateson, Gregory. 1972. Steps to an Ecology of Mind. New York: Ballantine Books. Bauman, Richard. 1977. Verbal Art as Performance. Rowley, Mass: Newbury House Publishers. Becker, Judith, and Alton Becker. 1981. "A Musical Icon: Power and Meaning in Javanese Gamelan Music." In The Sign in Music and Literature, edited by Wendy Steiner, 20315. Austin: University of Texas Press. Bhebe, Ngwabi. 1989. "The Nationalist Struggle, 1957-1962." In Rumoil and Tenacity: Zimbabwe 1890-1990, edited by Canaan S. Banana, 50-114. Harare:The College Press.


254


Birdwhistell, Ray L. 1960. "Implications of recent Developments in Communication Research for EvolutionaryTheory." In Report on the Ninth Annual Round Table Meeting on Linguistics and Language Studies, edited by William M. Austin, 149-55. Washington: Georgetown University Press. . 1970. Kinesics and Context: Essays on Body Motion Communication, Philadelphia: University of Pennsylvania Press. Blacking, John. 1977. "Toward an Anthropology of the Body." In The Anthropology of the Body, edited by John Blacking, 1-28. London: Academic Press. Bourdieu, Pierre. 1984. Distinction: A Social Critique of the Judgement of Taste. Cambridge: HarvardUniversity Press. Buchanan, Donna. 1995. "Metaphors of Power, Mataphors of Truth: The Politics of Music Professionalism in Bulgarian Folk Orchestras." Ethnomusicology 39(3):381-416. Chatterjee, Partha. 1986. Nationalist Thought and the Colonial World. Minneapolis: University of Minnesota Press. Cooke, Deryck. 1959. The Language of Music. London: Oxford University Press. Csikszentmihalyi, Mihaly, and Isabella Selega Csikszentmihalyi. 1988. Optimal Experience: Psychological Studies of Flow in Consciousness. Cambridge:CambridgeUniversity Press. Deutsch, Felix. 1954. "AnalyticSynesthesiology: Analytic Interpretation of Intersensory Perception." The International Journal of Psycho-analysis 35(3):293-301. Feld, Steven. 1974. "LinguisticModels in Ethnomusicology." Ethnomusicology 15:353-62. -. 1988. "Aestheticsas Iconicity of Style, or 'Lift-Up-OverSounding':Getting into the Kaluli Groove." Yearbook for Traditional Music 20:74-114. -. 1994. "Communication,Music, and Speech about Music."In Music Grooves by Charles Keil and Steven Feld, 77-95. Chicago: University of Chicago Press. Fernandez, James W. 1986. Persuasions and Performances: The Play of Tropes in Culture. Bloomington: Indiana University Press. Frith, Simon. 1987. "Towards an Aesthetic of Popular Music." In Music and Society: the Politics of Composition, Performance and Reception, edited by Richard Leppert and Susan McClary, 133-50. Cambridge: Cambridge University Press. Gellner, Ernest. 1983. Nations and Nationalism. Oxford: Basil Blackwell. Giglioli, Pier Paolo (ed.). 1972. Language and Social Context. Harmondsworth, Middlesex, England:Penguin Books. Goffman, Erving. 1974. Frame Analysis: An Essay on the Organization of Experience. New York: Harper Colophon. Grossberg, Lawrence. 1988. It's a Sin: Essays on Postmodernism, Politics and Culture. Sydney: Power Publications. Hall, Edward. 1977. Beyond Culture. Garden City, N.J.: Anchor Press. Hymes, Dell. 1972. Directions in Sociolinguistics: The Ethnography of Communication. New York: Holt, Rinehart, and Winston. Kaelin, E. F. 1983. "Reflections on Peirce's Aesthetics." In The Relevance of Charles Peirce, edited by Eugene Freeman, 223-37. La Salle, Illinois: The Hegeler Institute. Keil, Charles and Steven Feld. 1994. Music Grooves. Chicago: University of Chicago Press. Keiler, Allan R. 1981. "Two Views of Musical Semiotics." In The Sign in Music and Literature, edited by Wendy Steiner, 138-68. Austin: University of Texas Press. Kertzer, David I. 1988. Ritual, Politics and Power. New Haven: Yale University Press. Lyons,John. 1969. Introduction to Theoretical Linguistics. Cambridge:Cambridge University Press. . 1977. Semantics. 2 vols. Cambridge: Cambridge University Press. McNeill, William H. 1995. Keeping Together in Time: Dance and Drill in Human History. Cambridge: HarvardUniversity Press. Meintjes, Louise. 1990. "PaulSimon's Graceland, South Africa, and the Mediation of Musical Meaning."Ethnomusicology 34(1):37-73. Meyer, Leonard B. 1956. Emotion and Meaning in Music. Chicago: University of Chicago Press.



255

Nattiez, Jean-Jacques. 1990. Music and Discourse: Toward a Semiology of Music. Princeton: Princeton University Press. Pacini Hernandez, Deborah. 1995. Bachata: A Social History of a Dominican Popular Music. Philadelphia: Temple University Press. Peirce, Charles Sanders. 1955. Philosophical Writings of Peirce. Edited byJustus Buchler. New York: Dover. . 1958. Charles S. Peirce: Selected Writings. Edited by Philip P. Wiener. New York: Dover. . 1960. The Collected Papers of Charles Sanders Peirce. Edited by Charles Harshorne and Paul Weiss, Cambridge, Mass.: The Belknap Press of HarvardUniversity Press. . 1991. Peirce on Signs. Edited by James Hoopes, Chapel Hill: University of North Carolina Press. . 1992. The Essential Peirce: Selected Philosophical Writings. Vol. 1. Edited by Nathan Houser and Christian Kloesel. Bloomington: University of Indiana Press. Pefia, Manuel. 1985. The Texas-Mexican Conjunto: History of a Working-Class Music. Austin: University of Texas Press. Powers, Harold S. 1977. "The Structure of Musical Meaning: A View from Banaras."Perspectives of New Music 14(2):308-334. Saussure, Ferdinand de. 1966. Course in General Linguistics. Edited by Charles Bally, Albert Sechehaye, and Albert Riedlinger. New York: McGraw-HillBook Company. Seeger, Charles. 1977. Studies in Musicology, 1935-1975. Berkeley: University of California Press. Shamuyarira,Nathan. 1965. Crisis in Rhodesia. London: Andre Deutsch. Sugarman,Jane C. 1997. Engendering Song: Singing and Subjectivity at Prespa Albanian Weddings. Chicago: University of Chicago Press. Trudgill, Peter. 1974. Sociolinguistics: An Introduction. Harmondsworth, Middlesex, England: Penguin Books. Turino, Thomas. 1989. "The Coherence of Social Style and Musical Creation among the Aymara in Southern Peru." Ethnomusicology 33:1-30. . 1990. "Somos el Peru: 'Cumbia Andina' and the Children of Andean Migrantsin Lima." Studies in Latin American Popular Culture 9. . 1993 Moving Away from Silence: Music of the Peruvian Altiplano and the Experience of Urban Migration. Chicago: University of Chicago Press. . 1995. "ParticipatoryMusic as a Separate Art: Music Making in Peru, Zimbabwe, and the United States, Thoughts About Alternative Futures,"Colloquium paper presented at the University of Wisconsin at Madison. Turner, Victor. 1967. The Forest of Symbols: Aspects of Ndembu Ritual. Ithaca: Cornell University Press. Waterman, Christopher A. 1990. Juju: A Social History and Ethnography of an African Popular Music. Chicago: University of Chicago Press. West, Cornel. 1989. The American Evasion of Philosophy: A Genealogy of Pragmatism. Madison: University of Wisconsin Press.


Turino, Thomas_Signs, Imagination, And Experience. a Peircian Semiotic Theory for Music

Recommend Documents