* Tel.: +1 269 544 2716; fax: +1 269 387 3970. E-mail address:
[email protected]. Language & Communication 27 (2007) 245–257 www.elsevier.com/locate/langcom LANGUAGE & COMMUNICATION 0271-5309/$ - see front matter � 2007 Elsevier Ltd. All rights reserved. 1. Introduction Religious practitioners in Cuba’s popular religions deploy several highly marked regis- ters of ritual speech that index temporally-inflected African and Afro-Cuban ‘voices’ and therefore also sacred power. Two such registers, called ‘Lucumı´’ and ‘Bozal’, represent ideologically-charged voices of the past, African voices that sound like deities and slaves respectively. When ritual participants in Santerı´a ceremonies use Lucumı´ and Bozal in their speech, they bring the ‘past into the present’ in ritually powerful ways because each Abstract In this article I consider how registers of speech can index distinct chronotopes by indexing his- torical ‘voices’. Practitioners of Cuban Santerı´a can temporally inflect their ritual speech by deploy- ing two marked registers that contrast with standard Cuban Spanish and each other. These registers, called ‘Lucumı´’ and ‘Bozal’, are associated with particular historical or mythic character-types and with distinct but overlapping domains of ritual practice. I examine how contrasts in the ideological values of these registers bring three distinct chronotopes into play, thereby projecting the continuing power of ancestors and African deities, and serving as forms of ‘enregistered memory’. � 2007 Elsevier Ltd. All rights reserved. Keywords: Chronotope; Historical consciousness; Ritual speech; Religious language; Spirit possession; Santerı´a Enregistered memory and Afro-Cuban historicity in Santerı´a’s ritual speech Kristina Wirtz * Anthropology Department, Western Michigan University, Moore Hall, Kalamazoo, MI 49008, USA doi:10.1016/j.langcom.2007.01.009 246 K. Wirtz / Language & Communication 27 (2007) 245–257 of these registers invokes a particular ‘chronotope’, or notion of the nature of time, space, history, and (supernatural) agency. Bakhtin’s notion of chronotope or ‘time–space’ captures what he called ‘the intrinsic connectedness of temporal and spatial relationships’ in discourse and the ways in which these relationships shape character, agency, and historical consciousness in narrative emp- lotment (Bakhtin, 1981, p. 84). What is so useful about the concept of chronotope is that it suggests that space and time are mutually constituted in varying configurations, and always as part of broader cultural logics, which in the case discussed here include historical subjectivity, racial identity, and religious practice. Although Bakhtin applied his concept to understanding the construction of forms of time, space, and agency in literary genres like novels, the notion of chronotope also seems particularly useful for understanding the manipulation of time in magico-religious rituals, and especially how rituals not only create visions of past, mythic, or sacred space-times but vividly enact them in the here- and-now (see Hubert, 1999/1905; James and Mills, 2005, p. 10). As James and Mills argue, the interesting question is not so much how people represent time but rather what they do in time (James and Mills, 2005, p. 9; Parmentier, 1985). Widespread use of the concept of chronotope has generated as many questions as it has answered concerning how spatio-temporality is semiotically handled. Here, I address the issue of how chronotopes become manifest through the textual juxtaposition of speech registers, such that particular registers come to serve as forms of embodied memory (Connerton, 1989). I present an analysis of a genre of ritual performance common to San- terı´a and other related Cuban popular religions in which mythic-transcendent beings such as the orichas (the deities of Santerı´a) or muertos (spirits of the dead) manifest in the pres- ent, during possession trance. Register-shifts from unmarked everyday Cuban Spanish into two additional contrasting speech registers, Lucumı´ and Bozal, typify the voices of orichas and muertos respectively. I suggest that these contrastive registers take on histor- ical value through their association with these voices and are thus able to evoke two rit- ually important chronotopes that contrast with the everyday interactional chronotope of the ritual event. The juxtaposition of registers creates ritually meaningful relations among three distinct chronotopes: sacred, transcendent space-time (via Lucumı´), ances- tral/historical space-time (via Bozal), and the everyday plane of the here-and-now (via Cuban Spanish). The result is a temporal ‘telescoping’ through which transcendent and ancestral voices not only speak in the present ritual moment but temporarily inhabit it, conveying their historicity. 2. Registers as embodied memory To understand how chronotopic contrasts in Santerı´a ritual produce mnemonic effects on participants’ historical consciousness, it is important to attend to the language practices and ideologies that delimit speech registers and give them their indexical potential. My analysis will examine metadiscursive characterizations of two registers that forge indexical links with particular historical or mythic voices, such that shifts among these registers dur- ing rituals bring these distinct voices, and their associated chronotopes, together in perfor- mance. As Agha argues, the ideological work of what he calls ‘enregisterment’ uses contrasts across linguistic forms, such as accent, to index differentiated social types—‘per- sonae’—that become linked to those differences in form. He examines how a fundamen- tally bottom-up process accumulated across chains of interactions generates K. Wirtz / Language & Communication 27 (2007) 245–257 247 sociologically and characterologically distinct ‘voices’ that come to signify what they do because of how interlocutors align themselves toward each evocation of that voice (Agha, 2003, 2005). That is, durable social judgments about speech varieties and types of speakers arise out of instantaneous, real-time discourse processes. The salient ‘voices’ in the Cuban case I will describe index two contrasting, if related, ancestral sources of authority and authenticity: enslaved ancestors and the deities they brought from West Africa. The ritual ancestors that Santerı´a practitioners commemorate are contradictory figures who occupied the lowest rungs of Cuban society as African-born and/or slaves, even though some achieved ritual distinction as religious practitioners because of their association with the orichas (Brown, 2003). Their contradictory status is reflected in the contrasting social evaluations of the two ritual registers associated with Santerı´a. One register, Lucumı´, is seen by practitioners and scholars as a ‘high’ esoteric register, because practitioners identify it as an African language that is ‘the tongue of the deities’, but the other register, Bozal, is marked as a ‘lower’ or substandard sociolect of Cuban Spanish, because of its association with the speech of enslaved Africans. Note that the social evaluations of each register are based on the historical figures of enslaved Africans and African deities. I argue that the registers themselves, as well as the mythico- ancestral voices of muertos and orichas that speak through them, are examples what Conn- erton (1989) describes as embodied forms of collective memory, rather than ‘inscribed’ memory, where embodied memory seems particularly suited to transmitting subaltern his- tories because of its emphasis on pragmatic and performed, rather than strictly denota- tional meanings (Comaroff and Comaroff, 1992, 1993; Lambek, 1998). In short: Lucumı´, through its identification with its Yoruba origins, iconizes the tran- scendence of African traditions in Cuba even under the conditions of slavery. Bozal, which reproduces phenomena of linguistic interference, indexes the foreignness of its original speakers and their struggle to communicate in Spanish, and iconizes the transformations that produced creolized Cuban culture. The contrastively-defined values of these two reg- isters emerge out of their textual juxtaposition in real-time ritual events, coupled with met- adiscourses that characterize the divine or ancestral voices indexed by each register. In their collocation during rituals of spirit possession, these registers serve as embodied forms of memory as much as the bodily tropes that mark a person’s possession by an oricha or a muerto—such as the enactment of their behaviors, mannerisms, and tastes: enregistered memory accompanying corporeal habitus. 3. Register contrasts and ideologies of religious authority In Cuba, several distinct if overlapping religions comprise what Palmie´ has described as a spectrum of popular religious forms that mutually constitute each other as part of a gen- eral moral economy (Palmie´, 2002, pp. 159–168). One factor in this economy are judg- ments about how ‘African’, ‘European’, or ‘syncretized’ each tradition is (Argyriadis, 2000; Palmie´, 2002). Beyond Cuba, Santerı´a (also called Regla de Ocha) is the best known of these religions, but there are a number of others, including popular Catholic worship of the saints, the Reglas de Congo (sometimes called Palo Monte), Spiritism, and a less for- malized, more highly hybridized set of practices that Cuban scholars have called Muerterı´a (when it is recognized at all), which combines the others. I will focus here on Santerı´a and Muerterı´a. Santerı´a is closely identified with Yoruba traditions of West Africa that were recreated by enslaved ‘Lucumı´’ people in New World societies such as Cuba, Brazil, and Trinidad. It is considered to be faithful to its Yoruba origins in its focus on the orichas, who are transcendent African deities. Muerterı´a is understood to be a more local, Cuban phenomenon, and is scarcely recognized as a distinct ‘tradition’ at all but, rather, as 248 K. Wirtz / Language & Communication 27 (2007) 245–257 a more amorphous set of ‘syncretized’ practices combining European Spiritist and Congo- derived notions of working with the spirits of the dead (James Figarola, 1999). Muerterı´a focuses on muertos who are portrayed as historical types—especially slaves and subaltern migrant laborers such as Haitians. The practices comprising Muerterı´a are, for many Cubans, closely associated with brujerı´a (‘witchcraft’) or with attempts to protect oneself from brujerı´a: consulting spirits, asking for their protection, manipulating them to do trabajos (‘spells’), producing herbal medicines, purifications, and amulets, and feˆting the muertos. In the city of Santiago de Cuba, where I conducted ethnographic fieldwork among religious practitioners in 1998, 1999–2000, 2002, and 2006, there is a great deal of con- tact and mutual influence between Santerı´a and Muerterı´a, and many religious practitio- ners consider themselves to be both santeros and muerteros. One important type of worship for both kinds of practitioners is the festive ceremony of drumming and singing praises to the orichas or muertos that is generically known as a bembe´ or tambor.1 Depending on the type of ceremony, people may be possessed by either muertos or orichas. Orichas, generally speaking, comport themselves as powerful and even regal beings. They are widely understood to be transplanted African deities, so they have, in an important sense, transcended historical time and emerged from the Middle Passage and slavery essentially unchanged. Santeros do talk about how some orichas have disap- peared in Cuba while others have become more prominent, and how their domains may have shifted, but an underlying assumption in all of this discourse is that the orichas have an unchangeable essence that being transplanted to the New World did not and could not change. Part of this essence, indeed, is that they are multifaceted beings, who were already endowed with adaptability in changing circumstances so as to tran- scend history (Barnes, 1997). In contrast, the muertos usually represent set historical ‘types’ of persons—typically enslaved Africans or other highly marginalized groups in Cuba, including Haitian or Chi- nese laborers or, more rarely, (Native American) Indians. These historical types were all implicated in the colonization of the New World, the Atlantic slave trade, and its after- math of indentured servitude in Cuba—including Chinese and Mayan laborers in the late 19th century and Haitian and Jamaican laborers in the early 20th century (Perl and Valde´s Bernal, 1991). What these ‘types’ all have in common is their subaltern status because of histories of subjugation, a status reflected in their accented Spanish. During their lives, muertos were ordinary people who were profoundly changed by slavery. In Cuban histor- ical imagination, they were acculturated or at best transculturated but incompletely assim- ilated into Cuban life, speaking Spanish imperfectly, as a second language. In death their status has transformed to being powerful spirits who are willing to engage with humans, but they continue to personify ‘typical’ individuals caught up in the maelstrom of Cuba’s history. 1 According to local conventions in Santiago, a tambor or wemilere is done within Santerı´a (by recognized santeros, using bata´ drums, etc.) and a bembe´ or bembe´ de sau is not. 3.1. Lucumı´ According to local metalinguistic understandings, muertos or orichas who descend to possess someone during a ceremony do not speak in colloquial Cuban Spanish—‘Castel- lano’—but in distinctive registers. Whereas muertos are understood to speak in the register called Bozal, orichas speak in Lucumı´.2 Lucumı´ is closely associated with Santerı´a and car- ries two very different social valences. Practitioners sometimes call it the ‘tongue of the orichas’ because it is understood to be a divine language whose prototypical speakers are the orichas. At the same time, practitioners and scholars today sometimes refer to Lucumı´ as ‘Yoruba´’, emphasizing its historical connection to this modern African K. Wirtz / Language & Communication 27 (2007) 245–257 249 language. It is worth explaining my repeated collocation of Santerı´a practitioners and scholars because the dynamics between these two overlapping groups contribute to the historical valence of Lucumı´. First, among all of the popular religious practices of Cuba, Santerı´a has received a disproportionate amount of scholarly attention. Second, practitioners have been writing and circulating religious notebooks containing Lucumı´ vocabularies and other religious information for as long or longer than Cuban scholars have produced and circulated their own Lucumı´ glossaries and folkloric studies, and there has been con- siderable feedback, borrowing, and convergence between the two ostensibly distinct genres (Brown, 2003; Dianteill and Swearingen, 2003; Wirtz, 2004). It is therefore not surprising that both practitioners and scholars share an urge toward preservation and recovery of religious and linguistic knowledge that was lost in the translation from Africa to Cuba.3 In fact, some ritual experts do learn a lot of Lucumı´ vocabulary and gain considerable pragmatic expertise in performing Lucumı´ prayers, songs, and invocations. Some study religious and scholarly texts to improve their understanding, including materials on Yor- uba that are scarce in Cuba, but most practitioners regretfully admit to understanding very little of the semantic content of longer texts. The orichas, however, are regarded as fluent Lucumı´ speakers, and so proper communication between humans and orichas is predi- cated on santeros’ ability to fluently produce Lucumı´ utterances, even when their meaning is uncertain. Lucumı´, thus, has the double valence of being both a transcendent, divine language and an ancestral language that enslaved Lucumı´ ancestors once spoke but that their Spanish-speaking (ritual) descendents only imperfectly preserved. Although santeros and even some scholars metadiscursively characterize Lucumı´ as a ‘language’ and the orichas as its prototypically fluent speakers, those whom the orichas possess are not simply code-switching into Lucumı´. The speech of the orichas is in fact much more complex, as my next examples will show. 2 The situation reveals greater complexity, because orichas speak in heavily Bozal-accented Spanish and use occasional Lucumı´ words and phrases. Muertos I heard, too, used the same distinctive Bozal register and seemed not to use Lucumı´, although I have very few samples of muerto speech from which to generalize. From playback sessions with three santeros and one non-santero (including the transcriber who helped me prepare the excerpts I present) and other commentary on Lucumı´ I have heard, I do not believe that participants always distinguish Lucumı´ from Bozal when orichas speak. Perhaps not surprisingly, metalinguistic awareness seems to be shaped by genre expectations that orichas speak ‘Lucumı´’ and muertos speak ‘Bozal’ because Cuban religious practitioners associate Lucumı´ and Bozal with different spheres of religious practice (see Silverstein, 1981). 3 This concern with cultural decay from an African past exists in tension with scholarly and, to a lesser extent, practitioners’ focus on processes of cultural transformation—creolization and syncretization—in the creation of Cuban Santerı´a (Brown, 2003; Palmie´, 2002; Wirtz, 2007). The songs performed to attract the orichas to ceremonies are often entirely in Lucumı´, but the speech of orichas who possess someone tends to contain only occasional single Lucumı´ words or short phrases. That is, Lucumı´ words are substituted word-for-word into an otherwise Spanish matrix, and these Lucumı´ tokens tend to be names of reli- giously significant objects, persons, and states (of good or bad fortune). These tokens may have clear and concrete referents, such as the word orı´ or erı´ (‘head’) referring to someone’s actual head, but often their meanings are metaphorically extended, so, for example, orı´ might be used to refer to the relationship between a person and his principal oricha (who is said to reside ‘in the head’). Lucumı´ words often are combined with other 250 K. Wirtz / Language & Communication 27 (2007) 245–257 Pragmatic (I (Obatala) will not be here) In this example, three Lucumı´ word-tokens appear: the first-person pronoun emi, the noun ara´, which covers a wide semantic field including ‘land’, ‘people’, and ‘body’, and the emphatic particle o.5 The use of the Lucumı´ first-person pronoun ‘emi’ with the third-person inflection of the Spanish verb ‘to be’ (esta´ instead of standard estoy) help establish that the speaker is the deity Obatala´, not the female priest. Whereas the priest herself would presumably speak entirely in Spanish, the oricha’s voice is marked by both the esoteric lexical register of Lucumı´ and the seeming unfamiliarity with Spanish conju- gation. This pragmatic information shapes the interpretation of the oricha’s words, sug- gesting that he will not be present at the ritual for long. 4 Given the difficulties of transcribing/translating a register filled with as much indeterminacy and ambiguity as Lucumı´ is, my representations of this oricha’s speech are not necessarily the only possible ones. My transcriber and I engaged in repeated listening and long discussion to sort out the probable sounds, segmentation, and meaning of each utterance, a process of interpretation that the reader can partially reconstruct by working between the various ‘standardized Spanish’, literal, and pragmatic glosses provided for each line. 5 In form and meaning, these Lucumı´ tokens correspond closely to modern Yoruba words, although this fact is Standard Sp. Literal neither here nor the (Yo) no estoy (tierra) (emphatic) I am not land/body (emphatic) Emi no esta ara o ´ ´ Example 1 ambiguous, indirect, and even unintelligible forms to produce utterances with an aura of strangeness and mystery that require some effort (and a lot of pragmatic information) to interpret. Two examples, each followed by a short explanation, illustrate typical Lucumı´ usage, at least in my corpus of eight recorded (and an additional six witnessed but not recorded) instances of orichas speaking. The speaker is the male oricha Obatala´, who has possessed an elderly and highly respected female priest. The Lucumı´ tokens are in bold, and I pro- vide a comparison to more standard Spanish, as well as a relatively literal English trans- lation and a more pragmatically-informed translation provided to me by a priest who was in attendance and who also later helped me transcribe the recording.4 re as far as their local interpretations are concerned. Example 2 Pero dice Laroye´ no ta venı´ a su leri(to?) Standard Sp. Pero dice Laroye´ que no viene a su (cabeza)(cabecita?) gically ı´ from K. Wirtz / Language & Communication 27 (2007) 245–257 251 6 Linguists debate whether Bozal was a creolized form of Spanish, a Spanish- (or Portuguese-) substrate proto- creole, or simply non-fluent Spanish. (Castellanos, 1990; Lipski, 1986, 1998; Ortiz Lo´pez, 1998; Schwegler, 2006; Valde´s Bernal, 19 modified) infin itive to mark present, ongoing actions: in Example 2, ta venı´ (ven use of ta as a creole-like verbal aspect particle, which is combined with a (phonolo Literal But Laroye´ says that (he) does not come to his head (diminutive?) Pragmatic But (his principal oricha) Eleggua Laroye´ does not possess him The first Lucumı´ word in this utterance, Laroye´, is the proper name of an oricha. Orichas are understood to have multiple aspects or avatars, and Laroye´ is one of many avatars of the oricha Eleggua. The second Lucumı´ word, leri, was interpreted by my tran- scriber to be a version of orı´, meaning ‘head’. The head, in Santerı´a, is the physical and spiritual center of the body where one’s principal oricha is ‘seated’ during initiation, and where the possessing spirit or oricha takes control during possession trance. The expression ‘come to his head’ thus means ‘be possessed’. As I and a second listener heard it, the oricha did not simply say leri but lerito, adding the Spanish -ito and thereby creating a hybrid Lucumı´-Spanish neologism with a diminutive flavor: ‘his little head’. 3.2. Bozal Bozal is a sociolect of nonstandard Spanish that Cubans imagine to be the ‘broken’ Spanish once spoken by enslaved Africans (called ‘bozales’). The prototypical speakers of Bozal, thus, are no longer living, although Bozal, like Lucumı´, continues to circulate both as a metalinguistic label and as a ritual speech register.6 While there are no living bozals (African slaves), their ‘spirits’ or muertos continue to populate popular religious ceremonies through mediumship or possession trances in much the way the orichas des- cend in Santerı´a ceremonies. The muertos, Cubans say, speak in Bozal. (Orichas, too, speak in Bozal because they are not fluent in the Spanish they must sometimes use because of their devotees’ imperfect knowledge of Lucumı´.) In example 3, the powerful head of the orichas, Obatala´, sounds like a muerto as he expresses intimacy and affection for the elderly priest he has possessed and suggests that she may not live much longer. Obatala´’s use of Bozal in this case indexes the priest’s identity as a poor, elderly, and frail Afro-Cuban woman who is merely his vessel. Bozal as a register is marked by phonological, lexical, and morphosyntactic contrasts with unmarked Cuban Spanish that Cubans characterize as much like a foreign speaker’s errors of accent, pronunciation, and grammar and that some linguists characterize as pid- gin- or creole-like constructions. ‘Bozal’ pronunciation is on a continuum with colloquial Cuban Spanish (in which, for example /s/ is aspirated or dropped in some phonological contexts and there is a shift from /r/! /l/! /i/). In everyday contexts it is not uncommon to hear a word like puerta (‘door’) pronounced as /puel ta/ or /puei ta/ or esta´ (3rd-person singular ‘to be’) pronounced variously as /eh ta/, /e ta/ or /ta/. Such phonological varia- tions are emphasized in Bozal utterances, along with distinctive constructions, such as the 87). 252 K. Wirtz / Language & Communication 27 (2007) 245–257 The four lines in example 3 are full of phonological, lexical, and morphosyntactic alter- ations that distinguish the register from unmarked Spanish and therefore identify the speaker’s Bozal ‘voice’. I discuss only some of these alterations for reasons of space. There are numerous phonological characteristics—dropped or aspirated /s/’s and dropped /r/’s in particular (also /me jol/ for mejor)—that are characteristic of colloquial Cuban Spanish (as spoken by the santera when she is not possessed). My impression is that these phono- logical features seem to be heightened while the oricha is speaking, so as to give the speaker a heavier Bozal ‘accent’. These utterances also contain typical, awkward-sounding circumlocutions and ‘errors’, as if the oricha were groping for words in an unfamiliar lan- guage: For example, instead of saying Ustedes (plural 3rd-person) or even tu´ y tu´ (‘you and you’) while pointing at the author and her husband, the oricha says el suyo va el suyo (lit- erally: ‘yours goes yours’). In the second line presented earlier in example 1, there is a seeming lack of concord between the Lucumı´ first-person pronoun and the 3rd-person inflection of the verb. Spanish speakers characterize such lacks of concord as the kind Literal/Prag. Perhaps (she) will not be alive because (she) will be dead/is dying Standard Sp. A lo mejor no esta´ viva porque esta´ muerta (muriendo?) A lo mejol no ta vito po’que ta mo’ı´ Pragmatic So that the priest I am possessing then Literal So that the slaveship-mate (of him?) then Standard Sp. Ası´ (que) la carabela (de e´l?) entonces Ası´ el caramelo a e´l entonce Pragmatic (I (Obatala) will not be here) Literal I will not be land/body (emphatic) Standard Sp. (Yo) no estoy (tierra) (emphatic) Emi no esta´ ara o Pragmatic Later on, when you two (referring to KW and her husband) are well Literal You and you, when these (two people indicated) are well Standard Sp. Tu´ y tu´, es cuando e´stes (Uds) esta´n bien El suyo va e’ suyo e’ cuando e’ta´ e’te bien venir, ‘to come’) substitutes for viene (‘come(s)’) (Ortiz Lo´pez, 1998). Other widely remarked bozalisms, including lack of concordance between subjects and verbs or nouns and modifers, simplified verbal forms, or phonologically modified words, are generally interpreted by Cuban listeners as grammatical or pronunciation ‘errors’. This metaprag- matic characterization allows Bozal to index the disfluent Spanish voice of enslaved Afri- can ancestors. Several examples illustrate typical uses of bozal in utterances by the same oricha who spoke in the Lucumı´ examples above. In fact, example 3 is an expanded segment contain- ing the utterance from example 1. I have underlined segmentable bozalisms, keeping in mind the lack of a clear boundary between Bozal and the rest of Spanish and Schwegler (2006, p. 77) point that a single token of Bozal can cast a long metapragmatic shadow, such that an entire clause may sound ‘Bozal’ to a Cuban listener: Example 3 K. Wirtz / Language & Communication 27 (2007) 245–257 253 of mistake a foreigner learning Spanish might make. Another such ‘error’ appears in the speaker’s use of caramelo (‘candy’) for carabela, itself an interesting, chronotope-indexing word once used as a term of affection and fictive kinship among slaves who survived the Middle Passage aboard the same ship. The word choice marks the speaker’s familiarity with the special vocabulary of affection among Africans who survived the Middle Passage, helping produce a distinctively ancestral voice, as the oricha warns that he will not possess the santera for long, and that the santera herself will not live much longer. The Bozal-inflected speech of the oricha in example 3 illustrates the range of linguistic markers understood to be Bozal and how these work together with other linguistic and paralinguistic features to project a characteristically disfluent, African-sounding Bozal voice. The example also shows how markers of Lucumı´ and Bozal are deployed in a matrix of ordinary Cuban Spanish, such that the resulting sequence of register-contrasts is much more ambiguous and heteroglossic than a description in terms of code-switching would indicate. Such juxtapositions of register markers in real-time events can index chronotopes because they also cue interpretations in terms of religious and historical ideologies, in which particular registers index particular religiously and historically significant voices. 4. Chronotopic contrasts in real-time events I suggest that Lucumı´ and Bozal index contrasting chronotopes at least in part because they serve as metonyms of two contrasting religious practices and types of supernatural beings, each of which also indexes the same pattern of contrastive chronotopic values. Lucumı´ is tightly linked to Santerı´a and to the orichas, whereas Bozal is more loosely asso- ciated with other popular religious practices under the rubric of Muerterı´a that are under- stood to be more ‘syncretized’. Lucumı´ thus evokes for religious practitioners a transcendent, mythic chronotope as the divine language of the orichas and as the ritual register of Santerı´a, which is regarded as a proper religion that has carefully preserved African sacred practices and sensibilities. Bozal evokes a chronotope of a maelstrom of historical transformations centered on the experience of slavery, and the muertos personify ‘typical’ individuals caught up in this maelstrom in ways that shape them after death as well—namely, as speakers of ‘Bozal’ who can communicate only imperfectly despite their supernatural status. 4.1. The indexical creativity of register use: temporal ‘telescoping’ Note that a sort of temporal and spatial ‘telescoping’ has to happen in order for the orichas and muertos to reach out of the past that authenticates them and manifest them- selves as corporeal, powerful beings in the present. Their marked and accented speech helps accomplish this ritual telescoping of space-time. The marked speech of the orichas, in particular, is an important cue that other participants evaluate to validate and authen- ticate the presence and identity of the oricha. In possession trances, orichas and muertos take control of human bodies and in doing so make the transcendent and quotidian planes intersect. In the local idiom, orichas and muertos ‘descend’ or ‘rise’ to interact with other ritual participants, dance, do purifications and other procedures, and dispense advice, comfort, and warnings. They speak in ways that manifest the transcendent and historical chronotopes that define them, and they make those simultaneously divine and subaltern histories relevant to religious practitioners in the here-and-now. In doing so they address the p devot Lu non-s ples f 254 K. Wirtz / Language & Communication 27 (2007) 245–257 as much as I will if I am willing (estoy de acuerdo). When I come from my land yes because I am an African man. . . As this excerpt illustrates, the man obliges Ortiz-Lo´pez’s request for Bozal speech by taking on the persona of a ‘negro de nacio´n’, meaning an African-born man who would still have identified with his ethnic origin (nacio´n). He goes on in a similar register to describe life as a slave under lo branco (‘white people’), living in a bar- aco´(n) (sic, ‘slave quarters’) under the ‘whip’ (cuero), but still trying to maintain his religion despite being told ovvida a lo diose de lo negro (‘forget the blacks’ deities’). In short, the man performs a first-person account of colonial Cuban life from the perspective of, and in the voice of, a bozal slave who has possessed a female spir- itist medium. The man then, in Ortiz-Lo´pez’s words, ‘let himself get carried away until transforming into a true ‘negro congo’ . . . and transporting himself into that sociolinguistic past’. For 14 minutes, he astounded the researcher by producing ‘an unrecognizable discourse. . . that alternated between afrohispanic structures [like those above] and an unintelligible lan- guage’ that Ortiz-Lo´pez attempted to transcribe as follows (the extensive ellipses are his; I have underlined bozalisms and put possibly ‘Africanized’ words in bold, although the passage is almost glossolalic in places): Dicin gongo, dicin gongo ?...? y yo dicı´ ?...? y que ta´ bendijo cosa pa tu no cajo ni coja ni coja, pa logo ta´ yo ?. . .? alla´ nuyu tiera alla´ nuyu tiera y ta´ lejo de mı´o [asked for ´ tandard verb-aspect forms and lack of concordance that occurred in previous exam- rom my recordings. The translation is mine: Entonce ya ta´ decı´ pa’ ti. Nin˜o, tu´ ta´ quere´ que lo negro a´a decı´ cuanto yo va´ hace´, si me ta´ acoda´. Cuando yo ta´ venı´ de lo tierra mı´o sı´ poqque yo ta´ se´ negro de nac- io´n. . .So already (I? he?) speak(s) for you. Kid, you want the black man to speak topes, and it is this indexical power that allows them to serve as forms of enregistered memory, which I argue is a species of embodied memory because of its largely implicit pragmatic and performative function. That is, the register markers themselves can serve as creative indexes that can effectively transform ‘context’ through shifts in voice and chro- notope (Silverstein, 1976). One final example will illustrate how Bozal register use can trig- ger a creative chronotopic shift, even in the highly inappropriate and nonritual context of a sociolinguistic elicitation session. Ortiz Lo´pez (1998) was interviewing Cubans about Bozal as part of a study seeking traces of a missing Spanish Creole. He describes how one field consultant in Havana demonstrated for him how African slaves would have talked and slipped from what Ortiz-Lo´pez describes as conscious imitation to actually summoning his muerto by the very act of talking in Bozal like a muerto. Ortiz-Lo´pez explains: ‘he produced a striking and emotional monologue in which he abandons the present and unconsciously passes into that sociolinguistic epoch in which the black Afri- can lived on Cuban soil’ (p. 140) For several minutes, says Ortiz-Lo´pez, he ‘imitated a black woman during a spiritual session’. In the following sample from his transcription, I have underlined segmentable bozalisms, which in this sample mostly consist of the same ragmatic concerns of the moment, the hardships, heartbreaks, and woes of their ees. cumı´ words and bozalisms are at times almost electrically charged indices of chrono- rum] y tu narencio?. . . regist ized a tope of ordinary, everyday life, indexed by ordinary Cuban Spanish. My task has been K. Wirtz / Language & Communication 27 (2007) 245–257 255 to examine how these voices get enregistered in ritual and other metalinguistic practice, such that their utterance has the potential to telescope particular pasts into the present. Enregistered memory, I suggest, emerges out of contrastive relationships between two reg- isters of ritual speech linked to two ritually significant chronotopes. These contrasts artic- ulate with larger-scale and longer-term ideological processes that juxtapose religions, supernatural beings, and ways of speaking, and that, in turn, mediate the indexical rela- tions ers associated with their ‘voices’ index distinctive chronotopes that I have character- s mythic/transcendent and ancestral/historical, and that contrast with the chrono- Toward the end of the transcribed excerpt Ortiz-Lo´pez provides, the speaker seems to critique Ortiz-Lo´pez’s request for ‘palabras. . .como congo de Mandinguei’ (words. . .like a Congo of ‘Mandinguei’ (origins)), accusing him of ‘not believing in anything’: ‘Tu´ no ta´ cree´ na’. Upon again becoming conscious of his speech, the speaker started, began crying, and said, in regular colloquial Cuban Spanish: ¡Ay, cojone! Lui, Lui, ay ay ay cojone. . . para que me metı´ en e’to con˜o. . . Oh shit! Luis, Luis, oh, oh, oh shit . . . why did I get started in this, damn. . . He went on in considerable anguish to lament having done the imitation that appar- ently called down a real ‘muerto africano’ to possess him. It is significant that the man ini- tially chose to speak in Bozal by adopting the voice of what Cubans would agree is a ‘typical’ Bozal speaker: a muerto possessing an Afro-Cuban muertera. It is as if the histor- ical and religious subjectivity conveyed by Bozal speech is so tightly linked to this histor- icized figure that one presupposes the other. His performance in a muerto’s voice, something that usually only occurs in ritual settings, seems to have been sufficient in itself to creatively index the transformation of his consciousness into that of a man possessed by a muerto. 5. Discussion Speaking of his own muerto africano, identified for him by his Cuban consultants, Pal- mie´ (2002, p. 3) reflects on scholarly Black Atlantic historiography in light of the folk (Afro) Cuban historiography embodied by muertos. He comments: ‘In a very concrete sense, every form of historical knowledge involves propositions about the role of the dead in the world of the living, shaped as it invariably is by past human existence and agency’. Muertos and orichas are vivid and concrete presences that embody particular historicities through what Michael Lambek has described as a ‘synecdochic relationship to the past’ (Lambek, 1998, p. 119). He and others have described spirit possession in terms of the capacity of its poetic forms to mobilize contrasting chronotopes during ritual events and thereby offer what Lambek calls a ‘poiesis of history’, meaning its productive creation through forms of historicizing practice (Lambek, 1998; see also Stoller, 1995). I have given an account of how the distinct voices of the spirits and their associated chronotopes are creatively indexed by register contrasts. I have examined the linkages between the figures of muertos and orichas, the ritual prac- tices through which Cuban religious practitioners communicate with them, and the meta- pragmatic shadows these figures cast across registers of ritual speech they use. The hips between speech register, voice, and chronotope. The result is that concrete 256 K. Wirtz / Language & Communication 27 (2007) 245–257 speech forms—words, accents, morphosyntactic variations, and so forth—seem to cue chronotopes in a straightforward, first-order way, obscuring the much more complex, ideologically-mediated processes involved at two levels: during the real-time unfolding of rituals and in reflexive, historical processes through which registers cohere and gain durable social value (Inoue, 2004; Wilce, 2005). In the Cuban case described here, religious practitioners, muertos, and orichas are heard as speaking in distinct ways according to local religio-historical logic, regardless of their actual practices which combine registers in more complicated ways. Enregistered memory of the sort I have described for Lucumı´ and Bozal may be a rather common phenomenon, particularly in African diasporic settings. In an example with inter- esting parallels to my case, Armin Schwegler (1996) examines three chronotope-indexing speech varieties used in El Palenque de San Basilio, Colombia, where residents regard Spanish as ‘modern’ and a sign of their integration into wider Colombian society, where Palenquero indexes the local community united by the history that produced this creole that many still speak, and where a small number of elder women also employ an African- ized ritual register called ‘Lumbalu´’ in funerals that is explicitly marked as ancestral and African. Schwegler’s analysis suggests that the local identity of being Palenquero is consti- tuted through this complex of speech varieties. Such comparisons suggest new ground for exploring how the past is remembered and brought into the present, and what is accom- plished by its presence. Acknowledgments Thanks are due first and foremost to my Cuban field consultants. This article has tre- mendously benefited from the careful editorial attention of Michael Lempert and Sabina Perrino, as well as from thoughtful comments from panel discussant Richard Parmentier. I am also grateful to Stephan Palmie´, Armin Schwegler, and Bilinda Straight for conversa- tions that influenced my thinking on these matters. Fieldwork in Santiago de Cuba was funded by a Brody-Foley Grant, an International Pre-Dissertation Training Fellowship from the Social Science Research Council and the American Council of Learned Societies, with funds provided by the Ford Foundation, and a Foreign Language and Area Studies Fellowship. References Agha, A., 2003. The social life of cultural value. Language and Communication 23, 231–273. Agha, A., 2005. Voice, footing, enregisterment. Journal of Linguistic Anthropology 15 (2), 38–59. Argyriadis, K., 2000. Des Noirs sorciers aux babalaos: Analyse du paradoxe du rapport a l’Afrique a la Havane. Cahiers D’Etudes Africaines 160 (XL-4), 649–674. Bakhtin, M., 1981. The Dialogic Imagination: Four Essays by M.M. Bakhtin. University of Texas Press, Austin. Barnes, S.T., 1997. The many faces of Ogun: introduction to the first edition. In: Barnes, S.T. (Ed.), Africa’s Ogun: Old World and New, second ed. University of Indiana Press, Bloomington, pp. 1–26. Brown, D.H., 2003. Santerı´a Enthroned: Art, Ritual, and Innovation in an Afro-Cuban Religion. University of Chicago Press, Chicago. Castellanos, I., 1990. Grammatical structure, historical development, and religious usage of Afro-Cuban bozal speech. Folklore Forum 23 (1/2), 57–84. Comaroff, J., Comaroff, J., 1992. Ethnography and the Historical Imagination. Westview, Boulder. Comaroff, J., Comaroff, J. (Eds.), 1993. Modernity and its Malcontents: Ritual and Power in Postcolonial Africa. University of Chicago Press, Chicago. Connerton, P., 1989. How Societies Remember. Cambridge University Press, Cambridge. Dianteill, E., Swearingen, M., 2003. From hierography to ethnography and back: Lydia Cabrera’s texts and the written tradition in Afro-Cuban religions. Journal of American Folklore 116 (2), 273–292. Hubert, H., 1999/1905. Essay on Time: A Brief Study of the Representation of Time in Religion and Magic (R. K. Wirtz / Language & Communication 27 (2007) 245–257 257 Parkin and J. Redding, Trans.). Berghahn, Oxford. Inoue, M., 2004. What does language remember? Indexical inversion and the naturalized history of Japanese women. Journal of Linguistic Anthropology 14 (1), 39–56. James Figarola, J., 1999. Los Sistemas Ma´gico-religiosos Cubanos: Principios Rectores. UNESCO, Caracas. James, W., Mills, D., 2005. Introduction. In: James, W., Mills, D. (Eds.), Qualities of Time: Anthropological Approaches. Berg, Oxford and New York, pp. 1–15. Lambek, M., 1998. The Sakalava poiesis of history: Realizing the past through spirit possession in Madagascar. American Ethnologist 25 (2), 106–127. Lipski, J.M., 1986. Convergence and divergence in bozal Spanish: a comparative study. Journal of Pidgin and Creole Languages 1, 171–204. Lipski, J.M., 1998. Perspectivas sobre el espan˜ol bozal. In: Perl, M., Schwegler, A. (Eds.), Ame´rica negra: Panora´mica actual de los estudios linguı´sticos sobre variedades hispanas, portugueses, y criollas. Iberoamer- icana, Madrid, pp. 294–327. Ortiz Lo´pez, L.A., 1998. Huellas etnosociolinguı´sticas bozales y afrocubanas. Vervuet Verlag, Frankfurt y Madrid. Palmie´, S., 2002. Wizards and Scientists: Explorations in Afro-Cuban Modernity and Tradition. Duke University Press, Durham. Parmentier, R.J., 1985. Times of the signs. In: Mertz, E., Parmentier, R.J. (Eds.), Semiotic Mediation: Sociocultural and Psychological Perspectives. Academic Press, Orlando, pp. 131–154. Perl, M., Valde´s Bernal, S., 1991. Espan˜ol vestigial y minorı´as lingu¨ı´sticas en Cuba. In: Herna´ndez, C., Granda, C.P., Hoyos, C., et al. (Eds.), El Espan˜ol de Ame´rica. Junta de Castillo y Leo´n, Salamanca, vol. 3, pp. 1305– 1309. Schwegler, A., 1996. Chi ma nkongo: Lengua y rito ancestrales en El Palenque de San Basilio (Colombia). Vervuert Verlag, Frankfurt y Madrid. Schwegler, A., 2006. Bozal Spanish: Captivating new evidence from a contemporary source (Afro-Cuban ‘‘Palo Monte’’). In: Fuller, J., Thornburg, L.L. (Eds.), Studies in Contact Linguistics: Essays in Honor of Glenn G. Gilbert. Peter Lang, New York, pp. 71–100. Silverstein, M., 1976. Shifters, linguistic categories, and cultural description. In: Basso, K., Selby, H. (Eds.), Meaning in Anthropology. University of New Mexico Press, Albuquerque, pp. 11–55. Silverstein, M., 1981. The limits of awareness. Southwest Educational Development Laboratory, Austin. Stoller, P., 1995. Embodying Colonial Memories: Spirit Possession, Power, and the Hauka in West Africa. Routledge, New York. Valde´s Bernal, S., 1987. Las lenguas de African subsaharana y el espan˜ol de Cuba. Editorial Academı´a de Ciencias de Cuba, La Habana. Wilce, J.M., 2005. Traditional laments and postmodern regrets: The circulation of discourse in metacultural context. Journal of Linguistic Anthropology 15 (1), 60–71. Wirtz, K., 2004. Santerı´a in Cuban national consciousness: A religious case of the doble moral. Journal of Latin American Anthropology 9 (2), 409–438. Wirtz, K., 2007. How diasporic religious communities remember: learning to speak the ‘tongue of the oricha’ in Cuban Santerı´a. American Ethnologist 34 (1), 108–126. Kristina Wirtz (Ph.D University of Pennsylvania, 2003) is an assistant professor of anthropology at Western Michigan University. Her published and forthcoming articles and book (University Press of Florida, 2007) concern ritual performances, meaningful unintelligibility, modes of historical consciousness, and the making of religious community among Cuban religious practitioners. Enregistered memory and Afro-Cuban historicity in Santer iacute a " s ritual speech Introduction Registers as embodied memory Register contrasts and ideologies of religious authority Lucum iacute Bozal Chronotopic contrasts in real-time events The indexical creativity of register use: temporal ' telescoping ' Discussion Acknowledgments References