Poetics and Performance: Fanfa Bands and the Semiotic Landscape in Northern Haiti

Abstract

This paper examines the poetics and cultural significance of fanfa youth band performances in the rural commune of Limonade in northern Haiti. Drawing on observations during fieldwork in 2010 and 2016, it analyzes how fanfa bands, directed by maestros, create complex sign systems through music, movement, and materialities. Utilizing Roman Jakobson’s semiotic theory and Linda Waugh’s expansion of poetic function, the study explores the interpretive relations between these components and their role in constituting a unique cultural soundscape. By examining the selection and combination of musical pieces, routes, and accompanying elements, the research highlights the dynamic interaction between fanfa bands and their social environment. This semiotic analysis offers insights into the broader implications of cultural landscapes and the poetics of performance in Haiti.

Introduction

Now and then, in rural communes in northern Haiti, bands of young people with musical instruments gather under the direction of a maestro as a hired unit for any one of a variety of types of events. Their performances usually involve a processional through town at the behest of a patron. Hiring a fanfa is an investment in an announcement that is not necessarily a direct proclamation in words, but rather a stream of cultural signs written onto the landscape by the fanfa for the patron, often with the patron and members of his or her kin network involved as participants. This paper provides an analysis of observations made of two Haitian fanfa bands in the relatively rural commune of Limonade, Haiti in 2010 and 2016. Performances of music, movement, and materialities accompanying the fanfa become complex, multivalent, and bound together as sign systems that work to recognize themselves as a larger sign-unit in the fanfa performance, which works to constitute a unique cultural soundscape.¹

Roman Jakobson ([1956a], 76–77) and Linda Waugh (1985, 146) use semiotic theory to posit a meaning for what Jakobson outlined as the poetic function of language, a function that promotes the “palpability of signs” as such and leads to a diversity of genres that depend on a message both as a system of signs and as a sign in itself. An axis of similarity and difference is used as a means of constructing a sequence, leading to hierarchical structures of symmetries or parallels (Waugh 1985, 150). Thus, according to Waugh (1985, 153), the key to analyzing poetic speech is to look at interpretive relations between linguistic signs. Fanfa bands provide an opportunity to work outside of poetic language, from the “sound-shape” of language (Jakobson and Waugh 2002), to analyze the poetics—the interpretive relations between signs—of fanfa in rural Haitian spaces.

To ground the interpretive relations between component signs of Haitian fanfa in the cultural landscape of northern Haiti, I rely on Jakobson’s ([1956b]) two axes, the paradigmatic (based on selection or association of items) and the syntagmatic (based on combination of items). The selection set, working from relations of similarity, and combination set, working from relations of contiguity, are primary units of analysis in this paper, as I look at the components I have analytically derived as main constitutive units of a fanfa performance: music, movement, and accompanying materialities.

I argue that selection and combination for music, movement, and materialities drive the placement of fanfa performances in a social set of generic fanfa forms, rendering the performances understandable and informative to observers. Indeed, in this case, just as in the case of language, observers are also participants constituting the performance itself. Relations of similarity and contiguity, in supplying constructive generic work to place the fanfa performance, thus supply the poetics that make the message understandable as a whole, a sum of an infinite number of sign-systems at work, as well as rendering the landscape as a dynamic, yet constant and consistent, cultural, and acoustic artifact.

First, I discuss my empirical evidence in order to describe the genres of fanfa available for patronage in Limonade. The signs that constitute the genre and the genre itself preexist any instantiation of either the signs without context (which are empirically impossible) or the genre as a singular event. I look here to the bottom line, the social fact of the messages as wholes, of which genres are available in Limonade. Second, I show combination and selection at play for music-making, movement-making, and materialities accompanying the fanfa. Next, I offer a discussion of the possibilities for the analysis of cultural landscapes using semiotics and, particularly, an approach that looks for poetics of performance. I consider Jakobson’s and Waugh’s poetics coupled with four other theories of communication: Silverstein’s (2003) indexical order, Woolard’s (1999) bivalency, Goffman’s ([1983]) felicity’s condition, and Hanks (1987) discourse genres. Finally, I offer conclusions about how poetics and genre processes facilitate the common understandings that both cause and affect the practices of fanfa bands. Thus, I hope this semiotic analysis can begin to link the theoretical to the empirical and the cultural, to offer insight into one example of a constructed Caribbean cultural soundscape.

Fanfa and soundscapes

The observations used in this paper were made during scoping research in 2010 and then, in much greater detail, in 2016, when I lived outside of Limonade working on a research project with a colleague. Limonade is a rural commune about 10 miles outside of the large northern city of Cap-Haitien. Although I witnessed fanfa processionals and interviewed fanfa participants in Cap-Haitien and Limonade, dating back to 2010, I limit my discussion here to only those two with whom I had a fairly close relationship in Limonade, Fanfa Evangelik and Fanfa Melomane.²

I first saw Fanfa Evangelik (Evangelical Fanfa) as its members marched slowly down the main road that runs through Limonade, leading the way for a hearse toward the town cemetery. Willis Samuel is the maestro, the self-described “direktè, proprietè, general” (director, owner, and leader) (Fig. 2).³ Willis, as townspeople know him, was studying at the time to become a pastor at an Evangelical Christian church. The other Limonadien fanfa is called Fanfa Melomane, which its maestro and members described to me as meaning the “music-lovers’ fanfa.” The maestro of this fanfa is sometimes just called Melomane himself, although in other contexts he goes by his given name, Jean-Marie Romilien (Fig. 1). Melomane has 30 musicians and another dozen female majorettes and dancers, a larger membership than Evangelik’s 22 players. All members of these two fanfa are between the ages of 16 and 20.⁴

Figure 1. Jean-Marie Romilien standing outside the building that serves as Melomane’s headquarters, which is also part of his carpentry shop. Photo by author.

Figure 2. Interviewing Willis Samuel under a tree near his house.

Both of the fanfa in Limonade play for patrons at events for any religious denomination desired, and at least annually for the mayor of Limonade, who commissions processions in the town square on the feast day of St. Anne, the town’s patron saint, in July.⁵ Unlike Willis, who is fairly young and as yet unestablished in his career, Jean-Marie has been the town carpenter for over 20 years. This means that when a person in Limonade dies, Jean-Marie may very well be called upon to build the coffin, order the flowers, arrange for a hearse, and compose all the elements of a respectful fanfa processional with religiously appropriate (i.e., denomination sensitive) and event appropriate (i.e., mood sensitive) musical selections and arrangements. It was clear from my very first interviews with them about their work that Jean-Marie and Willis are both, as fanfa maestros, fluent in diverse religious and social sign systems.

Rural Haiti comes into anthropological literature first with the classic study of rural lifeways in the Artibonite Valley, south of Cap-Haitian in the country’s interior, written by Melville Herskovits in the 1930s (Herskovits [1937]). In the following decades, anthropologists came to the Caribbean and were able to broadly theorize historical and cultural changes to the urban and the rural (Mintz 1953; Mintz and Price 1976; Wolf and Mintz 1957). In terms of soundscape, ethnomusicologist and student of Herskovits, Alan Lomax, a noted folksong record-maker in the United States, travelled to Haiti and collected recordings during the period of American occupation, the 1930s (Lomax 1938; see also Averill 2008a).

Music in Haiti is a fruitful topic of social scientific inquiry. The history of music in Haiti (Averill 2008b), the history of Haitian art music (Grenier 2001), and the history of Vodou in terms of Haitian nationalist music (Largey 2006) have all been written. In anthropology, studies have been made into Lenten rara musical practices (McAlister 2002), the chan pwen sung by work cooperatives in southern Haiti (Smith 2001, 2004), and songs sent transnationally by cassette tape (Richman 2005). None of these works addresses the fanfa activities that are so integral to everyday life in northern Haitian cities and towns.

Likewise, soundscapes have been studied in a number of disciplines, including anthropology. Social sounds began to be studied more frequently following the publication of The Tuning of the World by R. Murray Schafer in 1977. Since the 1970s, anthropologists, geographers, and others have paired landscape with language (Basso 1996), space with thought (Crang and Thrift 2000), sound with sentiment (Feld [1982]), social sounds with historical space (Picker 1999), and the use of architecture with cultural symbols and meanings and semiotic systems (Geertz 1989). Likewise, work has been done in semiotics to build a methodology for understanding soundscapes (Jaworski and Thurlow 2010; Lotman 1990).

Music, movement, and materialities

Maestro Melomane and Maestro Willis actively work to select and combine elements of music, movement, and materialities, in order to shape the poetics, and thus the meaning and efficacy of their fanfa performances in Limonade. Selection and combination axes are analytic categories here, used to facilitate analysis of poetics, while the component groups (music, movement, and materialities) are also analytic categories used to group units constitutive of fanfa performances. The component groupings are subject to the selection and combination processes but are notably not immediately necessary for a fanfa.⁶ Table 1 visualizes the component groupings and the questions raised when selection and combination are taken separately.

Table 1. Component groupings for fanfa by selection and combination

The first component group is the music, which in both selection and combination is integral to providing signs for the system that determines the message of the fanfa as a whole. In early October, Fanfa Melomane invited me to attend one of their practice sessions, during which they played pieces from their full repertoire (Fig. 3).

Figure 3. Fanfa Melomane’s October 6, 2016 practice session. On the right, sitting at the two drums, are Jean-Marie’s two assistants, who help him train the other members of the band. Photo by author.

Jean-Marie explained to me that he himself arranged the music for each piece, often writing musical bridges to tie together a variety of songs into medleys. The pieces they played for me included religious songs for various occasions and denominations, festival pieces that were taken from national anthems and national songs, local traditional folk songs (for example, from the Cap-Haitian bands, Septentrionale or Tropicana⁷), and arranged versions of pieces originally recorded in North America by pop singers (e.g., Whitney Houston, Celine Dion, Mariah Carey).

The sample that Table 2 provides makes it clear that Jean-Marie and his fanfa have a variety of types of pieces for a variety of types of purposes. Song selection and instrumentation are two of the most important modes of selection within which Jean-Marie works to define the poetics of the message his fanfa is required by the patron to produce. In terms of combination, then, it is clear that order of songs and order of instruments in creating those songs are key decisions, as well. A funeral procession featuring “Au Ciel” will not suddenly begin playing the national anthem—unless the funeral is that of the mayor, perhaps. The music as a system of signs put onto and constituting the local soundscape is a key component to the fanfa’s readable existence for Limonadiens who are confronted by it.

Table 2. Pieces played by Fanfa Melomane at the early October practice session. Those lacking information were unrecorded. There were others that went unrecorded as well, despite my attempts to collect titles, sources, and recordings

The maestro of the fanfa is often also its leader in terms of movement. During the funeral procession I witnessed in September at 1 p.m. in Limonade, and during others I saw after it for other types of occasions, the geographical points to which the fanfa processed (and processed by) are ritually enacted stages of the event (cf. Turner [1969]) (Fig. 4, Fig. 5). This means that each location visited by the fanfa has a purpose within the ritual that the fanfa is helping to perform. In the case of a funeral, it may be overlapping, polysemic values that announce, mourn, grieve, and otherwise come to grips with a loss for social, political, economic, and kin-based reasons.

Figure 4. A photo of Willis Samuel leading his Fanfa Evangelik through town on the main road of Limonade for a funeral on 27 September 2016. Photo by author.

Figure 5. A map of Fanfa Evangelik’s procession in Limonade, Haiti. The procession began at the funeral home, processed down the main road through Limonade, turned to make a loop around the town square, processed to the church (where it played at the service), went to the cemetery (where it played at the interment), and then processed to the patron’s house (a brother of the deceased).

As for movement, then, at the behest of the occasion (and patron), the maestro must direct the fanfa to be either stationary or mobile (almost always both at particular parts of the route), as well as deciding on the appropriate and required route of performance. Less obvious from an aerial photo is the maestro’s decision concerning the speed (slow, medium, fast) of both marching and playing. These are all elements of the selection of types of movement for the fanfa band.

In combination, the order of speeds and order of places can mean a great deal. Selective and combinatory decisions can index not only basic meanings of announcement or information (constituting genre) but can simultaneously index religious or social sincerity. For example, if following the burial at the cemetery Willis’s fanfa went to the house of the deceased’s mistress, or if the fanfa placed the procession around the town square following the burial—these decisions on axes of selection and combination would (literally) dramatically affect the outcome of the message (and genre) of that instantiation of the fanfa’s performance.

What I call the materialities that accompany the fanfa provide the final component category that contributes directly (in this analysis) to the construction of the fanfa performance as a unified sign. The maestro must select which, if any, material items will accompany the fanfa on its procession. In terms of combination, he must select the order in which these material elements appear within the fanfa’s procession-performance. The materialities seem internal to the event as a unit (which is an outcome of the successfully unified and “genred” message). The example I discuss here is a materiality that is not, technically, a materiality: a poem.

Every July 25 is the celebration of the commune of Limonade and its patron saint, Sainte Anne. The mayor hires multiple fanfa bands (although Jean-Marie told me that sometimes the mayor lacks funds and Melomane plays for free anyway) to perform in the town square, where the enormous trunks of trees are painted white and blue, Limonade’s colors. Jean-Marie told me that in recent years, the mayor paid a Limonadien poet, Michle Petegny, to write a poem honoring the commune. The mayor then asked Jean-Marie to put that poem to music, a song that Jean-Marie named Hymne Communale. In this case, the commissioned poem, whether sung as lyrics to Melomane’s Hymne or recited contiguously to it, serves as an accompanying materiality to the fanfa performance. As such, its selection and combination with the other elements of the fanfa provide important poetic meaning that contributes to the genre and the sincerity of the genre of a performance in the town square.

Possibilities

Semiotics, and particularly the idea of poetics, has great promise for fruitful analysis of landscapes. The work of Roman Jakobson and Linda Waugh on a theory of poetics can be brought to bear on four linguistic-anthropological approaches to understand meaning-making signs: indexical order (Silverstein 2003), bivalency (Woolard 1999), felicity’s condition (Goffman [1983]), and genre (Hanks 1987).

Jakobson ([1956a], 77) asks, “what is the empirical linguistic criterion of the poetic function?” and, to answer it, he offers a theory of two basic modes of verbal arrangement, selection and combination. Selection works on the basis of equivalence, selecting one element from a list of possibilities, while combination works on the basis of contiguity, highlighting placement of elements (of different kinds) in relation to each other. “The poetic function projects the principle of equivalence from the axis of selection into the axis of combination” (Jakobson [1956a], 78). This definition for poetics shifts our focus to the decisions being made primarily in the combinatory mode. In other words, to get to an analysis of poetics, we must look at the sequence as a unified combination and set of relationships. Furthermore, Jakobson did not limit this analysis to actual poetry but encouraged analysis “outside of poetry, when some other function is superimposed upon the poetic function” (Jakobson [1956a], 79).

Waugh (1985) provides a thoughtful overview of this kind of poetic theory in linguistic analysis, detailing the poetic function’s orientation toward the message itself. She defines the message as (a) a system of systems of signs and (b) a sign with signans and signatum (Waugh 1985, 146). She posits three types of linguistic sign: signs codified as such, which may be seen to be words; signs in or as messages without being fully prefabricated in the code, such as phrases governed by rules of combination of words; and signs codified only as optional patterns of combinations, such as discourses (Waugh 1985, 146–147). An analysis of poetics, then, projects the principle of equivalence from the category of selection into that of combination. As Waugh explains, following Jakobson, orientation to the message-sign by the message-sign along with the use of equivalence relations between constitutive signs leads to reevaluation of the internal nature of the sign as a sign and a reevaluation of the signs’ relations in the message (Waugh 1985, 153). This is an important statement because it posits the message’s recognition of itself and self-contained evaluation by way of relations of its constitutive units.

Jakobson’s and Waugh’s understandings of the poetic function of interaction and communication can be coupled with a number of theories in order to further clarify the semiotic processes involved in poetic meaning-making. Silverstein (2003) analyzes the semiotics of communication by, building on Charles Sanders Peirce (1958) and others, acknowledging the concepts of presupposition and entailment, which he writes are the two aspects that constitute indexical meaning. Presupposition is the aspect of “indexical ‘appropriateness-to’ at-that-point autonomously known or constituted contextual parameters: what is already established between interacting sign-users, at least implicitly, as ‘context,’” thus a sort of moveable a priori. Entailment, on the other hand, is the aspect of “indexical ‘effectiveness-in’ context: how contextual parameters seem to be brought into being—i.e., causally and hence existentially entailed—by the fact of usage of the indexical (Sin)sign [=token] itself” (Silverstein 2003, 195). We might shorten these explanations to presuppositions as prior sharedness and entailments to effectivity in interaction. Silverstein argues from presuppositions and entailments to suggest that there are numerous and simultaneous indices at work in any given discourse or communication—a sort of polysemy of higher orders.

Looking at Silverstein’s argument in terms of poetics, it seems there might be room for a fruitful combination of theoretical interests. Silverstein brings an ontology of signs during the passage of time, with which Peirce was also concerned, that might add to Jakobson’s poetic function a certain force of history and set-up for future sign work. In other words, internal relations of elements are the products of historical chains of social signs (presuppositions). The product of internal relations of elements—their interpretation—is a further force in the sign work to be done among interacting humans in a specific context (entailment). Looking at the poetics of indices (e.g., can indices be related to each other? What is the order of the “indexical order”?) may be a further avenue for productive theoretical exploration.

Woolard’s (1999) work on bilingual zones has led her to conclude that “translinguistic markers” include code-switching, a familiar term in linguistic anthropology, as well as bivalency. She defines bivalency as simultaneous membership of an element in more than one system. The relevant question is whether interactants distinguish between “sequential alternation—a speaker shifting from foot to foot—and virtual simultaneity—the speaker jumping up and down on a third foot” (Woolard 1999, 17). Rather than a sequence of forms of voices, roles, relations, or identities, Woolard posits a simultaneity of those types of things. This allows for expression that incorporates ambiguity or multiple functions at once. This is a similar conclusion to Silverstein’s, although drawing less on semiotics and more on sociolinguistics. While Woolard does question, if not attack, the dichotomy between sequence and simultaneity, and while this may seem anathema to Jakobson’s original two axes, it also appears that Woolard’s idea of bivalent fluency can be used to work on understanding the message as a semiotic whole. To what extent do interactants acknowledge a distinction between the axes that Jakobson defines? Perhaps Woolard’s bivalency offers a path toward analyzing poetics from the strategic position of an agent.

The third possibility for a fruitful use of semiotics of poetics comes from sociolinguistics. Similar to (and prior to) Silverstein, Goffman ([1983], 168) considers “social presuppositions in language use.” He borrows from J.L. Austin’s list of six “felicity conditions,” which he notes were later turned into “sincerity conditions” by John Searle and worked on to become four “maxims” of cooperative speakers by Paul Grice. Goffman suggests “the felicity condition behind all other felicity conditions, namely, Felicity’s Condition,” which he defines as “any arrangement which leads us to judge an individual’s verbal acts to be not a manifestation of strangeness. Behind Felicity’s Condition is our sense of what it is to be sane” (Goffman [1983], 170–171). From a “hearer’s-ear view of talk, namely, what it is a hearer will be able and ready to respond to,” with Felicity’s Condition, Goffman is attempting to describe the ultimate, underlying, inalienable presupposition (Goffman [1983], 188). For the same reasons that semiotic work on poetics may benefit from the presuppositions and entailments of Silverstein’s and Woolard’s theses, so too it may benefit from the overarching presupposition—really, constraint—that Goffman suggests. It may be interesting to ask what is the primary or original constraint on the meaning-making of poetics? In other words, what is the baseline constraint of expression that enables the poetic function? What is it that guarantees one the possibility to interpret a message as such, and how might a poetic message be missed? Is it, as Goffman might have it, the assumption of action from a reasonable position? Is there some sort of poetic faculty that must be satisfied—the interactant must know what they are confronted with is interpretable as a sign-message?

Finally, we come to genre. Hanks (1987, 668) borrows Mikhail Bakhtin’s “sociological poetics” and Pierre Bourdieu’s theory of practice to define genres as “elements of linguistic habitus, consisting of stylistic, thematic, and indexical schemata on which actors improvise in the course of linguistic production.”⁸ Interestingly, the Bakhtin that Hanks synthesizes is similar to Peirce, which provides a basis for semiotics, in that Bakhtin’s “ideological horizon” defines human consciousness as apperceiving reality through the mediation of ideology. Every genre, for Bakhtin, has its own value-laden orientation, and Hanks adds his own reckoning that genres gain thematic organization from an interplay between systems of social value, linguistic convention, and the portrayed world (Hanks 1987, 671).⁹ Hanks’ “threshold genres,” as doubly interpretable or ambivalently blended genres, and his view of genres as ways of naturalizing the world, may be particularly useful in combination with analyses of poetics (Hanks 1987, 688–689). As is evident in my description of fanfas in Limonade, the concept of genre can work productively alongside poetics. The self-consciousness of the message and the self-interpretation of the poetic function of the message are similar processes to the naturalization of genres that Hanks describes. How do manipulations of the poetic function, through manipulations of equivalences of relations in combination, inform the genre by which the message as a unified whole (the poem) is performed and understood by interactants? Indeed, is genre separable from the poetic function at all?

Conclusions

This paper provides an empirical case for analysis using a semiotic definition of the poetic function of expression. I present an explanation of music, movement, and materialities—performance—that I observed in fanfa band performances and practices in Limonade, Haiti in 2010 and 2016. In describing the selection processes for each component grouping—which songs, which instruments, stationary/mobile, which locations, which speeds, which materials—and in describing the combinatory decisions for these elements, I hope to have demonstrated the considerable effort it is to successfully and fluidly carry out a meaningful fanfa performance in rural northern Haiti.

The maestros I interviewed, Willis Samuel and Jean-Marie Romilien, from the two fanfa bands, Fanfa Evangelik and Fanfa Melomane, both instantiated a fluency and competency in every version of their identities, as Haitians, Limonadiens, businessmen, musicians, and band leaders. The fanfa performance itself indexes simultaneous and bivalent meanings depending on the conditions for each particular performance.