CENTRAL PLAZA BURIALS IN SALADOID PUERTO RICO: AN ALTERNATIVE PERSPECTIVE William F. Keegan Some Caribbean archaeologists have assumed that the individuals buried beneath the central plazas of Saladoid sites in Puerto Rico lived in those villages during their lives. They interpret these central place burials as providing immediate access to the ancestors during ceremonies performed in this public space. The central plaza is viewed as the axis mundi, and through ancestor veneration the dead were called upon to intercede with the gods on behalf of the living. However, cross-cultural studies indicate that burial practices often are determined by descent, and those clan members whose post- marital residence was in communities other than their clan villages often were returned to their clan village for burial. It is argued here that central place burials do not reflect ancestor veneration, but rather social solidarity among widely scat- tered villages. Algunos de los arqueólogos que trabajan en el Caribe han asumido que los individuos enterrados en las plazas centrales de los sitios Saladoides de Puerto Rico habitaron estas comunidades durante su vida. Dicen que los enterramientos en los lugares centrales proveían un acceso inmediato a los ancestros durante las ceremonias realizadas en este espacio público. Esta plaza central es vista como el axis mundi en donde, a través del culto al ancestro, el difunto era llamado para interceder con los dioses por los vivos. No obstante, los estudios en diferentes culturas indican que las prácticas funerarias eran comúnmente determinadas por descendencia. Entonces, aquellos miembros del clan cuyo patrón residencial post-marital era en aldeas dis- tintas a su aldea de origen comúnmente fueron retornados a la aldea de su clan natal para su entierro. En este trabajo se argu- menta que estos enterramientos en los lugares centrales no reflejan un culto al ancestro, sino la solidaridad social que existía entre aldeas dispersas. working in Puerto Rico have Archaeologists noted that early ceramic age, Saladoid {circa 400 B.C. to A.D. 500) settlements tend to have circular, oval, or horseshoe-shaped community plans, and that human remains often were buried in the central clearing or plaza (e.g., Maisabel, Tibes, Tutu, and Punta Candelero sites; Curet and Oliver 1998; Righter 2002; Siegel 1992, 1996, 1999). Beginning about A.D. 600 there were significant changes in cultural practices, which are used to define a new culture-historical period, called Ostionoid (Rouse 1992). These changes included substantial modifications in the produc- tion and decoration of pottery vessels, a shift from large extended-family domestic structures to smaller nuclear-family households, the construc- tion of stone-lined courts and plazas, and a shift in burials from the central plaza to the interiors of houses (Curet and Oliver 1998). The archaeologi- cal sites discussed in this article are indicated in Figure 1. Saladoid burials in the central plaza have been interpreted as evidence for ancestor veneration (Siegel 1992, 1996, 1997). It is implied that these individuals lived in the community in which they were buried, and that by burying them in a central place they were present in spirit, especially during the rituals and ceremonies that were held in this public space. Thus, the living members of the com- munity could look to the ancestors for supernatural intervention, and maintain spiritual ties to those who lived before them. The notion that central place burials reflect immediate access to revered ancestors as part of a system of ancestor veneration seems logical. The image of the central plaza as the axis mundi - William F. Keegan ■ Curator of Caribbean Archaeology, Florida Museum of Natural History, P.O. Box 1 17800, University of Florida, Gainesville, FL, 32611 (keegan@flmnh.ufl.edu) Latin American Antiquity 20(2), 2009, pp. 375-385 Copyright ©2009 by the Society for American Archaeology 375 This content downloaded from 129.252.86.83 on Thu, 26 Feb 2015 05:17:28 UTC All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 376 LATIN AMERICAN ANTIQUITY [Vol. 20, No. 2, 2009 Figure 1. Map of the Caribbean islands showing site locations mentioned in the article. represented as a tree that ties the underworld where the ancestors are buried to the land of the living and the sky world of the spirits - is a compelling image (Siegel 1992, 1996, 1997). Yet, the issue explored here is why the Puerto Rican Saladoids adopted the practice of central place burials, and what this prac- tice meant in terms of social organization. The issue here is not whether or not the Saladoid peoples practiced ancestor veneration. Siegel (1992) makes compelling arguments that they did. The issue is whether or not central plaza burials were exclusively an outcome of this practice. This commentary presents an alternative explanation for Saladoid central place burials, and their apparent disappearance at the beginning of the Ostionoid period. The alternative presented here is that indi- viduals who lived their adult lives in other com- munities were returned to their natal or clan community for burial. Yet I also recognize that this reason probably is not universal. The current trend in Caribbean archaeology is to move away from monolithic notions of singular cultural identities (e.g., Rouse 1992) to recognized cultural diversity throughout both time and space (e.g., Curet 2003; Keegan 2004). I begin by reviewing the conclusions presented for ancestor veneration because these are the con- texts in which central plaza burials have been inter- preted to date. I then turn to the alternative model of postmortem mobility, and argue that this prac- tice better explains central plaza burials in Sal- adoid Puerto Rico in particular contexts. It is hoped and expected that other alternatives will be forth- coming. Ancestor Veneration: The Model Peter Siegel (1992) was the first Caribbean archae- ologist to emphasize that Saladoid community plans reflect prevailing concepts of cosmology. He recognized that these settlements typically had cir- cular or semicircular arrangements that seemed to reflect a concentric worldview in which the central plaza represented a sacred space, that the circular This content downloaded from 129.252.86.83 on Thu, 26 Feb 2015 05:17:28 UTC All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Keegan] CENTRAL PLAZA BURIALS IN SALADOID PUERTO RICO 377 arrangement of domestic structures and mounded middens reflected secular (profane) space, and that the natural world (including various spirits) held sway in the surrounding forests. Siegel (1992, 1996, 1997, 1999) then used a number of ethnographic examples from lowland South America along with ethnohistoric reports for the contact-period Tainos to flesh out his model of Saladoid cosmology. In addition to the horizontal expression of concentric zones that extended out- ward from the village center, he also recognized a vertical expression in which the central sacred space was the axis mundi that tied the village to the celes- tial dome and underworld. Finally, he interpreted the discovery of human burials in the central plaza as one form of evidence for ancestor veneration (Siegel 1992). In other words, these burials were placed to provide imme- diate access to the ancestors during ceremonies conducted on the plaza. He went on to propose that ancestor veneration involved ceremonies in which individuals were buried in the central plaza and rit- ual objects were buried in the surrounding mounded middens. Siegel developed a well-conceived and logical model of Saladoid cosmology. However, the assumptions and speculations on which the model is based deserve more careful scrutiny, with spe- cial emphasis on the possible reasons for central plaza burials and the fact that these are found almost exclusively in Saladoid sites on Puerto Rico. Ancestor Veneration Reconsidered I agree with Siegel's conclusion that the layout of communities reflects cosmology (Keegan 2007). However, my first issue is that none of the sites he uses to support his concentric model have a perfect concentric plan and distribution of mounded mid- dens (Siegel 1996:Figure 3). Indian Creek is cir- cular but the middens are sequential and not simultaneous; Golden Rock (although not com- pletely defined) and Sorcé appear to be semicircu- lar; Punta Candelero is oval; Maisabel seems to have two open plazas; Monserrate has a mounded midden near the middle of the plaza; and Trants is circular, but lacks central plaza burials (Figure 2). In addition, central plaza burials are not limited to the Saladoid; El Chorro de Maíta site in eastern Cuba has central plaza burials in a proto-historic context (Valcárcel Rojas and Rodríguez Arce 2005). This diversity of community plans indicates that the concentric model may not adequately capture the cosmological intentions of the people who built and occupied these settlements. One possibility is that astronomical alignments also influenced the placement of the mounded middens when these communities were established (Alegría 1983; Castellanos 1981; Keegan 2007; Rodríguez Alvarez 2003; Sullivan 1981). In addition, Ensor (2003) has suggested that circular villages reflect patterns of kinship. A second issue is the assumption that the all of the mounded middens at a site are contemporane- ous. Yet at least some of the sites seem to have developed through the sequential construction of middens over a long period of time (e.g., Indian Creek, Antigua). Although the outcome may appear to represent a roughly concentric distribution, this distribution may be the final product of centuries of occupation and not reflect accurately the layout of the settlement at any particular time. In other words, the village itself was never a concentric arrangement of structures. One of the beauties of Siegel's model is the sub- stantial number of ethnographic cases from low- land South America that he cites (Siegel 1992, 1996). However, and despite an extended discus- sion of the proper use of ethnographic analogy in archaeology (Siegel 1992:376-381), it should be asked whether the ethnographic and ethnohistoric cases that he uses are directly applicable to the data (see Maclachlan and Keegan 1990). Circular and concentric settlements occur throughout the world. There is no reason to assume that this community plan only reflects native Amazonian notions of cos- mology, or that all lowland South American groups shared the same worldview.1 Siegel's model of Saladoid community plans and cosmology is based on the assumption that all lowland South American societies shared very sim- ilar conceptions of cosmology and space. This is a problematical assumption. Heckenberger (2002) provides some support for Siegel's model in his recognition that circular or concentric villages probably are characteristic of the Arawakan peo- ples who first colonized the Caribbean. In addition, he concludes that "Arawak peoples reproduce a habitas predisposed to perpetuate (a) an ethos of This content downloaded from 129.252.86.83 on Thu, 26 Feb 2015 05:17:28 UTC All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 378 LATIN AMERICAN ANTIQUITY [Vol. 20, No. 2, 2009 Figure 2. Settlement layouts for Saladoid sites (based on Siegel 1996). This content downloaded from 129.252.86.83 on Thu, 26 Feb 2015 05:17:28 UTC All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Keegan] CENTRAL PLAZA BURIALS IN SALADOID PUERTO RICO 379 settled village life, commonly coupled with large, fixed populations, fairly intensive subsistence economies, and landscape alteration (rather than mobility and low impact); (b) institutional social ranking based on bloodline and birth order; and (c) regional integration (particularly coupled with a social preoccupation with exchange and a cultural aesthetic which places great symbolic value upon foreign things) and a foreign policy commonly characterized by accommodation and accultura- tion of outsiders" (Heckenberger 2002). Heckenberger' s (2005) research among the modern Arawakan Xinguano suggests that Sal- adoid peoples were not egalitarian (contra Siegel 1996), and that their social organization can be characterized as cognatic and nonlineal. Moreover, ancestor veneration is most common in ranked soci- eties and rare in those described as egalitarian (Paula Rubel, personal communication 2006). Heckenberger also concludes that Gê-speaking groups, which Siegel (1992, 1996) uses to support his model, are not appropriate analogs for the Arawakan Saladoids (Heckenberger et al. 2005). Siegel's use of ethnohistoric accounts recorded primarily among the Tainos in Hispaniola also can be questioned. Curet (2003:19) recently noted, "Hispañolan and Puerto Rican polities used sig- nificantly different ideological foundations, a reflection of differences in the nature of the polit- ical structure and organizations." He continues, "Judging from the striking differences mentioned, they likely developed from distinct types of ances- tral societies, and/or through different and diver- gent historical processes" (Curet 2003:20). I have reached a similar conclusion and have suggested that the Tainos of Hispaniola are the product of Archaic cultural and social foundations (Keegan 2004, 2006a). Thus, Taino ethnohistory may not be directly applicable to cultural developments outside of Hispaniola. Furthermore, Siegel (1999:214) states, "Father Ramón Pané... made the most accurate and detailed account of Taino religion " In fact, Pané reported that the information he collected about religious beliefs was related to him in the Macorix tongue (a non-Taino language; see Granberry and Vescelius 2004). After making his initial report, Christopher Columbus then asked Pané "to go and live with another leading cacique, named Guari- onex, the lord of many people, since the language [Taino] of these folk was understood throughout the land." Pané countered: "Lord [Christopher Columbus], how is it that Your Lordship wishes that I go live with Guarionex without knowing any lan- guage other than that of Macorix?" (Stevens- Arroyo 1988:76). It impossible that the Macorix and Tainos shared similar religious beliefs, and it is possible that Pané added information that he col- lected in the Taino province of Maguana, but these assumptions have not been proven. Archaeologists have been too eager to accept Pané as an accurate accounting of Taino religion (Keegan 2007). We also need to keep in mind that "Las Casas also says that Ramon was ... a simple-minded man so that what he reported was sometimes confused and of little substance" (Bourne 1906:314). With regard to the notion of an axis mundi (e.g., Eliade 1963), Siegel specifically cites the case of the Makiritare (Yecuana), a Cariban-speaking group of the upper Orinoco of Venezuela. Yet the Makiritare (Yecuana) associate this concept with their house structures and not the central plaza. It is appropriate to ask whether this cosmological concept is only associated with man-made con- structions, and whether open plazas or central open spaces were similarly designated in Arawakan cos- mology. In sum, it is not enough to say that because some lowland South American societies had particular concepts of cosmology that these can be extended to the Salaçloid peoples of the Caribbean. There must be test implications and archaeological cor- relates that can be discerned in the Caribbean in order for the ethnographic analogies to be relevant. The final issue concerns burial practices. Archaeologists working in Puerto Rico have pro- posed what seems to be a straightforward dichotomy between central plaza burials during Saladoid times and a shift to burials beneath domes- tic structures during the subsequent Ostionoid (Curet and Oliver 1998). However, burial practices are far more complex. Saladoid burials also are common in middens (Siegel 1992), and post- Saladoid burials are found in caves, mounds, and beneath house floors (Curet and Oliver 1998; Siegel 1999:217). The Maisabel site presently is exceptional with regard to the continuation of central plaza burials into the Ostionoid, although the Tutu site on St. Thomas also has central plaza burials that continue This content downloaded from 129.252.86.83 on Thu, 26 Feb 2015 05:17:28 UTC All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 380 LATIN AMERICAN ANTIQUITY [Vol. 20, No. 2, 2009 into the Ostionoid (Righter 2002). Unlike other Saladoid sites that either were abandoned or had their central spaces converted to stone-lined courts at the beginning of the Ostionoid period (Curet 2005), Maisabel continued unchanged and burials continued to be placed in the central plaza during Ostionoid times. The fact that Maisabel seems to be unique raises questions concerning the degree to which this community plan, and conclusions based on ethnographic and ethnohistoric analogy, can be generalized to the rest of the Saladoid Caribbean. What concerns me here is why some Saladoid burials were located beneath the central plaza, and why central plaza burials are reported only for Puerto Rico. What follows is an alternative expla- nation for central plaza burials in Puerto Rico. Postmortem Mobility In an unpublished paper (cited with permission of the authors), Paula G. Rubel and Abraham Rosman (2009)2note that: "Archaeologists considering the question of the social structure ... have dealt with the relationship between mortuary practices and societal definitions of the social person, as well as the relationship between mortuary practices and the rank system (Binford 1971; Saxe 1970). Thus far they have not usually been able to deal with the question of descent" (cf. Hodder 1979). In societies that practice exogamy every com- munity is composed of natal residents and in- marrying spouses. One result is that individuals who trace descent to the same apical ancestor are widely scattered among different communities. Rubel and Rosman (2009; also see Rosman and Rubel 2009) note that individuals often are returned to their natal or clan village for burial in societies that practice patrilineal, matrilineal, or cognatic descent reckoning. Although the conditions under which an individual would be returned to their natal or clan community for burial have not been for- malized, one measure seems to be the degree to which an individual was incorporated into the clan of their postmarital residence. From the examples given by Rubel and Rosman (2009), it appears that men are more often returned to their natal village, and that women, especially those who contribute offspring to their new community, are less likely to be returned after death. An interesting case concerns avunculocal resi- dence, which Keegan and his colleagues have sug- gested for the prehispanic Caribbean (Keegan and Maclachlan 1989; Keegan et al. 1998). Rubel and Rosman (2009) note: "avunculocal post-marital residence results in a situation in which a woman is usually born in her father's village (among his clansmen), on marriage moves to the village of her husband and his matri-kin, and at death is carried to the cemetery of her own clan. Throughout her life, she never resides in the village of her own clan. Only in death is she surrounded by her own clans- men." In sum, the main issue is the connection between postmarital residence and where spouses are buried. It cannot be assumed that the individuals buried in a community were life-long residents ofthat com- munity. The possibility that individuals who lived elsewhere during their life were returned to their natal or clan village for burial must be considered. Saladoid Burials: An Alternative Perspective Saladoid culture was a long-lived and widespread phenomenon. Archaeological sites with Saladoid ceramics are distributed from northeastern South America, throughout the Lesser Antilles, the Vir- gin Islands, and Puerto Rico. If ancestor venera- tion was a significant component of cultural practice, then we would expect to see evidence for it throughout the Antilles. Continuing this logic, if central plaza burials were a component of ances- tor veneration, then such burials should be evident throughout the Saladoid region. Yet burial in the central plaza seems to be unique to Saladoid sites in Puerto Rico. Similar sites from the same time period on Vieques and in the Lesser Antilles (e.g., Sorcé, Trants, Hope Estate, Morel, and Golden Rock sites) and South America typically do not have burials in the central plaza (Rouse and Crux- ent 1963). However, Heckenberger (2005) notes that individuals were returned to their natal village for burial in the southern Amazon (Xingu area), and he believes that this practice was part of the cul- tural inventory during the Arawakan diaspora that led to the colonization of the Antilles (Hecken- berger 2002; Heckenberger et al. 2005). The ques- tion remains, why was the Puerto Rican Saladoid unique? One reason that Puerto Rico is unique is that it This content downloaded from 129.252.86.83 on Thu, 26 Feb 2015 05:17:28 UTC All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Keegan] CENTRAL PLAZA BURIALS IN SALADOID PUERTO RICO 381 was colonized very early in the ceramic age set- tlement of the Caribbean. The earliest dates for Sal- adoid sites come from Puerto Rico and the northern Lesser Antilles (ca. 400 B.C.). The available data suggest that the initial ceramic age colonization of the Caribbean occurred as either a direct jump from Central or South America to Puerto Rico and the neighboring Windward and Virgin Islands or as movement through the Lesser Antilles during which most of these islands were bypassed during the initial episode of colonization (reviewed in Kee- gan 2004; also see Rodriguez Ramos and Pagan Jiménez 2007). The only early Saladoid site identified in the Windward Islands of the southern Lesser Antilles is Fond Brûlé on Martinique.3 Radiocarbon dates of circa 480 B.C. and four other dates ranging up to A.D. 135 have been reported. Is it a coincidence that Fond Brûlé, the one early Saladoid site in the southern Lesser Antilles, is almost exactly halfway between Puerto Rico and the South American main- land (about 500 km from the northeast coast of Venezuela and 450 km from eastern Puerto Rico)? If the Saladoid migration had progressed in a stepping-stone fashion through the Lesser Antilles, then there should be similar early sites on other islands. To date, these have not been found. The need to maintain social ties over very long distances would have promoted an emphasis on clan solidarity - an emphasis that may not have been so critical once additional colonies were estab- lished throughout the Lesser Antilles after A.D. 200 (see Fitzpatrick et al. 2010; Haviser 1997). New settlements, especially when they are a sub- stantial distance from the parent community, face a variety of economic, demographic, and social risks (Keegan 2004). Catastrophic events can lead to subsistence failure, appropriate spouses must be found to maintain demographic viability, and the full range of statuses and positions must be filled to ensure social reproduction. These factors form the basis for the "lifeline" model in which a colony maintained close ties to its parent community (e.g., Kirch 2000). In addition, Moore (2001) has shown through simulation studies that a colony has a sig- nificantly greater chance for survival and success if it maintains ties to at least one other community. The social glue that binds one community to another is marriage. Alliances between families, clans, and villages are formalized through the exchange of spouses. Thus, the return of a deceased spouse to their clan village completes a structural- ist complementarity. Moreover, in some societies the death of a spouse must be resolved by provid- ing a new spouse to the widow/widower (see Schneider and Gough 1961). The death of a clan member provides an occasion for feasting in which clan members from different communities gather to send the individual to their afterlife, and it also might involve a marriage ceremony in which a new spouse was provided to reaffirm social alliances. In this regard, burial practices reflect regional social alliances, and not simply local connections to the ancestors and gods through deceased members of the community. Second, Saladoid material culture emphasized regional connections. Keegan (2004) has inter- preted the widespread sharing of iconography as a veneer that served to unite widely scattered com- munities.4 The notion of veneer comes from the low frequency of shared motifs in concert with their wide distribution across the Antilles from South America to Puerto Rico. In other words, while the majority of artifacts in Saladoid sites reflect local activities, there also are highly decorated ceramics and exotic lithic amulets that share virtually iden- tical iconographie messages and that served to rein- force ties among widely scattered communities (Watters and Scaglion 1994). The expectation is that other cultural practices, including burials, should reflect this regional focus. The shift to a more local focus is evident dur- ing post-Saladoid times (Curet and Oliver 1998; Keegan 2004). A wide variety of local pottery styles emerged, and the emphasis on long-distance con- nections in order to ensure social reproduction and procreation was no longer crucial. The abandon- ment of central place burials in favor of burials beneath domestic structures, and in other contexts, is a symptom of the emerging localization of social identity. Third, individuals who moved to the village of their spouse at marriage may lack ties to specific households in their natal or clan village. Although they were born in a particular structure, the move- ment of their siblings at marriage, death of their par- ents, and other factors can obliterate specific ties. Over time they are no longer associated with one household, but are instead members of a wider clan. This situation is especially true in the avunculocal This content downloaded from 129.252.86.83 on Thu, 26 Feb 2015 05:17:28 UTC All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 382 LATIN AMERICAN ANTIQUITY [Vol. 20, No. 2, 2009 case cited above. As a result they should be buried in the clan cemetery and not beneath the floor of a domestic structure. At least in Puerto Rico, Saladoid cemeteries seem to be located in the center of the village. Thus, it is reasonable to assume that these cemeteries contain clan members who resided in the village and those who lived elsewhere during their adult lives and were returned for burial. The centrality of burial would serve to reinforce ties among clan members who were dispersed among widely scat- tered villages. After death, the return of a clan mem- ber would be the focus of grieving by the descent group as expressed in a feast that brought together the widely scattered members of the clan. Such clan solidarity was essential for ensuring survival in the risky milieu of the colonial enterprise. Fourth, if the individuals who were buried in the central plaza were indeed revered ancestors, then we would expect to find some indicators for this status. Yet Saladoid burials contain few grave goods, and there is little to indicate that these indi- viduals were imbued with any special status. More- over, it is not clear whether the mounded middens represent a community dump, or whether they belonged to particular social groups. We do know that exotic artifacts were buried in these middens, perhaps during mortuary rituals associated with the deceased. If this was the case, then the ritual bur- ial of objects in the midden affirmed social alliances and group solidarity rather than the significance of particular individuals. This interpretation makes sense because less complex societies tend to trace descent to a mythical ancestor and not to living indi- viduals. In my opinion, the characteristics thus far used to define the Saladoid are emblems of regional identity that served to unite a diversity of cultural groups. The wide distribution of ceramics and lithics with virtually identical forms and modes of expression reflect the strong integration of widely scattered communities that needed inter- community cooperation and support to ensure their survival. In this situation the focus would not be on individuals (i.e., revered ancestors) but rather on the social relations that bound communities to each other. This is exactly the situation in which we would expect the dead to be returned to their natal or clan village for burial. In other words, the deceased individuals represent regional solidarity, and not individuals who intercede with the gods on behalf of the living. Fifth, it may be possible to test the notion that burials reflect the postmortem mobility of clan members by analyzing strontium isotopes in the bones. In a recent study, Booden et al. (2008) mea- sured strontium isotope ratios in 19 individuals buried in the cemetery at Anse à la Gourde, Guade- loupe. They concluded that a number of the indi- viduals had "nonlocal" Sr isotope signatures. What is intriguing is that these individuals often were buried on the periphery of the cemetery. Their study suggests that individuals who did not live at this site were buried in the cemetery. Finally, the issue of cultural diversity in the pre- Columbian Caribbean is only now being addressed.5 New evidence suggests that the Sal- adoid cultures may have had closer ties to the isthmo-Colombian area of South America than to the lowland Amazon and Orinoco River basins (Rodriguez Ramos 2005; Rodriguez Ramos and Pagan Jiménez 2007). Thus the notion that lowland South American ethnographies can be used to inter- pret Saladoid cultures is in question. Moreover, Taino societies exhibit substantial diversity (Hulme 1993), and recent interpretations suggest significant differences between the Hispaniolan societies that were described by the Spanish chroniclers and the Puerto Rican societies that have been interpreted using these records (Curet 2003; Keegan 2004, 2007). Thus, the use of ethnohistoric evidence recorded among the Taino of Hispaniola and Cuba should not be used to justify the notion of ances- tor veneration in Saladoid Puerto Rico (see Mach- lachlan and Keegan 1990). Conclusions To his credit, Siegel develops a logical argument for the role of cosmology in structuring the layout of Saladoid communities. However, in order to sup- port his model he homogenizes Saladoid culture, lowland South American societies, and Taino eth- nohistory. In fact, none of the settlement plans that he illustrates are exactly the same (Siegel 1996:Fig- ure 3), the ethnographic cases that he uses come from societies with diverse cultural practices (including significant language differences), and not all Caribbean archaeologists are willing to accept that accounts of the ethnohistoric Tainos in This content downloaded from 129.252.86.83 on Thu, 26 Feb 2015 05:17:28 UTC All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Keegan] CENTRAL PLAZA BURIALS IN SALADOID PUERTO RICO 383 Hispaniola and Cuba are applicable to Puerto Rico or the Saladoid period (Curet 2003; Keegan 2004, 2007). Finally, the fact that central plaza burials are found primarily in Puerto Rico indicates that the notion that these burials are evidence for ancestor veneration should not be extended to other Sal- adoid groups in the Caribbean. Siegel's(1992, 1996, 1997, 1999) model implies that Saladoid burials in the central plaza reflect the interment of individuals who lived in the commu- nity throughout their lives. Cross-cultural studies indicate that this may not be the case (Rubel and Rosman 2009). It is equally likely that clan mem- bers were returned to their clan village for burial, and that these burials were located in clan or com- munity cemeteries some of which were situated in the central plaza. Moreover, ancestor veneration, as defined by Siegel (1992, 1999), places primary emphasis on the local community and its deceased residents as providing access to the gods. In contrast, the recog- nition of postmortem mobility places emphasis on clan solidarity among widely scattered communi- ties. Because Saladoid material culture emphasizes a regional identity, it is here argued that Saladoid burials in the central plaza more likely represent regional integration and not homage to the people who lived and died in the village. Finally, the disappearance of central place buri- als is viewed as reflecting a shift to a more local identity (albeit on a smaller regional scale). The need to reinforce long-distance social ties disap- peared as the number of settlements and the den- sity of population increased. As a result, mortuary practices shifted to a more local focus by the begin- ning of the Ostionoid period. Had access to the ancestors been important - and Siegel argues that ancestor veneration was an important component of Taino ideology - then central place burials should have continued even if these were restricted to members of the emerging elite (see Curet and Oliver 1998; Siegel 1999). The selection of the central plaza as the loca- tion for the village cemetery may reflect a decision to inter the dead in a focal place. In this regard, the notion of ancestor veneration remains a viable working hypothesis, and there is no reason that burial practices based on clan affiliation could not also reflect the identification of these individuals as ancestors. However Siegel's speculations fail to consider the broader implications of social organi- zation, and he ignores the possibility that individ- uals who were not resident in the community during their adult lives were returned to their natal or clan village for burial. Given the exigencies of coloniz- ing groups, it is more likely that social solidarity was the primary emphasis of mortuary practices. My goal in writing this article is not to show that Siegel's ancestor veneration is wrong and my clan burial is right. Both hypotheses require further elab- oration and testing. The central issue is that the Saladoid is far more complex than previously acknowledged, and that our models of cultural behavior must embrace this complexity. Acknowledgments. The assistance of Michael Heckenberger, Corinne Hofman, Menno Hoogland, José Oliver, Reniel Rodriguez Ramos, and David Watters is appreciated. Extensive comments from Peter Siegel were very helpful in refining my arguments. Reniel Rodriguez Ramos translated the abstract into Spanish, and Josh Torres assisted with Figures 1 and 2. References Cited Alegría, Ricardo E. 1 983 Ball Courts and Ceremonial Plazas in the Caribbean. Yale University Publications in Anthropology No. 79, Yale University Press, New Haven. Binford, Lewis 1971 Mortuary Practices: Their Study and Potential. In Social Dimensions of Mortuary Practices, Memoir No. 25. American Antiquity 36:6-29. Booden, Mathias A., Raphaël G. A. M. Panhuysen, Menno L. P. Hoogland, H. N. de Jong, Grant R. Davies, and Corinne L. Hofman 2008 Tracing Human Mobility with 87Sr/86Sr at Anse à la Gourde, Guadeloupe. In Crossing the Borders: New Meth- ods and Techniques in the Study of Archaeological Mate- rials from the Caribbean, edited by Corinne L. Hofman, Menno L. P. Hoogland, and Annelou van Gijn, pp. 214-225. University of Alabama Press, Tuscaloosa. Castellanos, Ricardo 1981 La Plaza de Chacuey, un Instrumento Astronomico Megalitico. Boletín del Museo del Hombre Dominicano 16:31-40. Curet, L. Antonio 2003 Issues on the Diversity and Emergence of Middle- Range Societies of the Ancient Caribbean: A Critique. Journal of Archaeological Research 1 1 : 1-42. 2005 Caribbean Paleodemo graphy. University of Alabama Press, Tuscaloosa. Curet, L. Antonio, and José R. Oliver 1998 Mortuary Practices, Social Development, and Ideol- ogy in Precolumbian Puerto Rico. Latin American Antiq- uity 9:217 '-239. Eliade, Mircea 1 963 Myth and Reality, translated by W. R. Trask. Harper and Row, New York. Ensor, Bradley E. 2003 Crow-Omaha Marital Alliances and Social Transfor- This content downloaded from 129.252.86.83 on Thu, 26 Feb 2015 05:17:28 UTC All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp 384 LATIN AMERICAN ANTIQUITY [Vol. 20, No. 2, 2009 mations: Archaeological Case Studies on the Taino, Hohokam, and Archaic Lower Mississippi Valley. Ph.D. dissertation, University of Florida, Gainesville. Fitzpatrick, Scott M, Michiel Kappers, and Christina Giovas 2010 The Southern Route Hypothesis: Carriacou's Posi- tion in Southern Antillean Prehistory. In Island Shores, Dis- tant Pasts: Archaeological and Biological Perspectives on Pre-Columbian Settlement of the Caribbean, edited by S. M. Fitzpatrick and A. Ross. University Press of Florida, Gainesville, in press. Granberry, Julian, and Gary S. Vescelius 2004 Languages of the Pre-Columbian Antilles. Univer- sity Press of Alabama, Tuscaloosa. Haviser, Jay B. 1 997 Settlement Strategies in the Early Ceramic Age. In The Indigenous People of the Caribbean, edited by S. M. Wil- son, pp. 57-69. University Press of Florida, Gainesville. Heckenberger, Michael J. 2002 Rethinking the Arawakan Diaspora: Hierarchy, Regionality, and the Amazonian Formative. In Compara- tive Arawakan Histories: Rethinking Language Family and Culture Area in Amazonia, edited by Jon D. Hill and Fernando Santos-Granero, pp. 99-325. University of Illi- nois Press, Urbana. 2005 The Ecology of Power, Culture, Place and Person- hood in the Southern Amazon, A.D. 1000-2000. Rout- ledge, New York. Heckenberger, Michael J., Joshua Toney, Mark Donop and M. Schmidt 2005 Community Spatial Patterns and Socio-Political Orga- nization in the Early Arawak Diaspora: Reflections on Recent Research in the Southern Caribbean. Paper pre- sented at the 21st Congress of the International Associa- tion of Caribbean Archaeologists, University of the West Indies, St. Augustine, Trinidad. Hodder, Ian 1 979 Social and Economic Stress and Material Culture Pat- terning. American Antiauitv 14:446-454. Hulme, Peter 1993 Making Sense of the Native Caribbean. New West Indian Guide 67:189-220. Keegan, Wiliam F. 2000 West Indian Archaeology. 3. Ceramic Age. Journal of Archaeological Research 8: 135-167. 2004 Islands of Chaos. In The Late Ceramic Societies in the Eastern Caribbean, edited by Andre Delpuech and Corinne L. Hofman, pp. 33^4. BAR International Series 1273. British Archaeological Reports, Oxford. 2006 Archaic Influences in the Origins and Development of Taino Societies. Caribbean Journal of Science 42: 1-10. 2007 Taino Indian Myth and Practice: The Arrival of the Stranger King. University Press of Florida, Gainesville. Keegan, William F., and Morgan D. Maclachlan 1 989 The Evolution of Avunculocal Chiefdoms: A Recon- struction of Taino Kinship and Politics. American Anthro- pologist 91:613-630. Keegan, William F., Morgan D. Maclachlan, and Bryan Byrne 1 998 Social Foundations of Taino Caciques. In Chiefdoms and Chieftaincy in the Americas, edited by Elsa Redmond, pp. 217-244. University Press of Florida, Gainesville. Kirch, Patrick V. 2000 On the Road of the Winds. University of California Press, Berkeley. Lévi-Strauss, Claude 1 963 Structural Anthropology. Basic Books, New York. Maclachlan, Morgan D., and William F. Keegan 1990 Archaeology and the Ethno-tyrannies. American Anthropologist 92:101 1-1013. Moore, John 2001 Evaluating Five Models of Human Colonization. American Anthropologist 103:395^08. Righter, Elizabeth (editor) 2002 The Tutu Archaeological Village Site. Routledge, New York. Rodriguez Alvarez, Ángel 2003 Astronomía en la Prehistoria de Puerto Rico: Caguana y Tibes, Antiguos Observatorios Precolombinos. Doctoral dissertation, Universidad de Valladolid, Spain Rodríguez Ramos, Reniel 2005 The Crab-Shell Dichotomy Revisited: The Lithics Speak Out. In Ancient Borinquen: Archaeology and Eth- nohistory of Native Puerto Rico, edited by Peter E. Siegel, pp. 1-54. University of Alabama Press, Tuscaloosa. Rodríguez Ramos, Reniel and Jaime R. Pagan Jiménez 2007 Las Antillas en el Contexto del Circun-Caribe: Cin- cuenta Años Después. In Proceedings of the XXI Interna- tional Congress of the International Association for Caribbean Archaeology, edited by Basil Reid, Henri Petit- jean Roget and L. Antonio Curet, pp. 778-786, University of the West Indies, St. Augustine, Trinidad. Rosman, Abraham, and Paula G. Rubel 2009 The Material Basis of Dual Organization. Unpub- lished manuscript cited with permission of the authors. Rouse, Irving 1953 The Circum-Caribbean Theory: An Archaeological Test. American Anthropologist 55(2): 1 88-200. 1992 The Tainos. Yale University Press, New Haven. Rouse, Irving, and José M. Cruxent 1963 Venezuelan Archaeology. Caribbean Series, No. 6. Yale University Press, New Haven. Rubel, Paula G., and Abraham Rosman 2009 Solidarity of the Sibling Group: Cognatic Societies and Their Mortuary Practices. Unpublished manuscript cited with permission of the authors. Saxe, Arthur 1 970 Social Dimensions of Mortuary Practices. Ph.D. dis- sertation, University of Michigan, Ann Arbor. University Microfilms, Ann Arbor. Schneider, David M., and Kathleen Gough (editors) 1961 Matrilineal Kinship. University of California Press, Berkeley. Siegel, Peter E. 1992 Ideology, Power, and Social Complexity in Prehis- toric Puerto Rico. Ph.D. dissertation, State University of New York, Binghampton. University Microfilms, Ann Arbor. 1996 Ideology and Culture Change in Prehistoric Puerto Rico: A View from the Community. Journal of Field Archaeology 23:313-333. 1 997 Ancestor Veneration and Cosmology among the Taino. In Taino: Pre-Columbian Art and Culture from the Caribbean, edited by F. Bercht, E. Brodsky, J. A. Farmer, and D. Taylor, pp. 106-111. The Monacelli Press, New York. 1999 Contested Places and Places of Contest: The Evolu- tion of Social Power and Ceremonial Space in Prehistoric Puerto Rico. Latin American Antiquity 10(3): 209-23 8. Stevens- Arroyo, Antonio M. 1988 Cave of the Jagua: The Mythological World of the Tainos. University of New Mexico Press, Albuquerque. Steward, Julian H., and Louis C. Faron 1 959 Native Peoples of South America. McGraw-Hill, New York. Sullivan, Shaun D. This content downloaded from 129.252.86.83 on Thu, 26 Feb 2015 05:17:28 UTC All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Keegan] CENTRAL PLAZA BURIALS IN SALADOID PUERTO RICO 385 198 1 The Colonization and Exploitation of the Turks and Caicos Islands. Doctoral dissertation, University of Illi- nois at Urbana-Champaign. University Microfilms, Ann Arbor. Valcárcel Rojas, Roberto, and César A. Rodríguez Arce 2005 El Chorro de Maíta: Social Inequality and Mortuary Space. In Dialogues in Cuban Archaeology, edited by L. Antonio Curet, Shannon L. Dawdy, and Gabino La Rosa Corzo, pp. 125-146. University of Alabama Press, Tuscaloosa. Waiters, David R., and Richard Scaglion 1994 Beads and Pendants from Trants, Montserrat: Impli- cations for the Prehistoric Lapidary Industry of the Caribbean. Annals of the Carnegie Museum, Vol. 63, No. 4, pp. 215-237. Wilbert, Johannes 1 972 Survivors of Eldorado: Four Indian Cultures of South America. Praeger, New York. Notes 1. Siegel's model of ancestor veneration is based on selec- tive sampling from ethnographic reports regarding South American societies that existed two thousand years after the Saladoid peoples, and on ethnohistoric reports that the Tainos of Hispaniola practiced ancestor worship through the curation of human remains. Careful reading of Siegel's numerous pub- lications on this topic leads one to believe that ancestor ven- eration was an unquestionable component of Saladoid culture, and that archaeological evidence is not needed to demonstrate this point. 2. Citing unpublished papers as the basis for an argument is less than satisfying. However, Rubel, Rosman, and I are revising and combining their two papers for publication. In the interim, we are happy to provide the original versions of their papers to anyone who requests them. 3. Here I am faced with a dilemma. One reviewer, citing personal communication with Benoit Berard (an archaeolo- gist who has lived and worked on Martinique for years), claims that the early dates for the site are inaccurate, and that there is no evidence for the colonization of the southern Lesser Antilles before A.D. 100-200. A second reviewer claims that the early dates from this site point to the early set- tlement of these islands. In either case, most (if not all) of the southern Lesser Antilles appear to have been bypassed during the initial ceramic age colonization, and that the earliest sites are located on Puerto Rico and neighboring Windward islands. 4. One of the anonymous reviewers questioned my use of the term "veneer," which I use in conformity with the first dictionary definition - "a thin finishing or surface layer of wood ... bonded to an inferior substratum." In the same way that a wood veneer conveys a particular outward message, the use of identical symbols on pottery and lithics conveyed a sense of regional identity to the widely scattered Saladoid peoples of the Antilles. Although most of their artifacts were made to facilitate local/domestic activities, they also manu- factured a relatively small percentage of artifacts (circa 10 percent) that conveyed social and cultural meanings. In this regard we can identify similar symbolic messages that are consistent across the Antilles (called Saladoid), but this thin layer of shared meaning masks variability at the local level. Similar conclusions have been proposed for the Adena/Hopewell and Southeastern Ceremonial Cult in east- ern North America. 5. For more than fifty years Caribbean archaeology has been structured by the debate between those who believe that the Antilles were colonized by peoples from the lower Orinoco (Rouse 1953, 1992), and those who view Antillean societies as belonging to what Julian Steward defined as "Circum-Caribbean Chiefdoms" (Steward and Faron 1959). Those who take a traditional view of normative culture his- tory probably have not recognized this debate and the ways that it was contextualized in Caribbean archaeology (based on the fact that they never cite these publications). Peter Siegel represents the former, while I recently have become convinced that the latter is more accurate (although it is not possible that Andean cultural developments are the source of the Taino social formation). What is at issue is the degree to which Amazonian ethnographies and Taino ethnohistory can be applied generally to early Caribbean societies (see Maclachlan and Keegan 1990). Submitted April 24, 2005; Accepted March 12, 2007; Revised June 3, 2008 This content downloaded from 129.252.86.83 on Thu, 26 Feb 2015 05:17:28 UTC All use subject to JSTOR Terms and Conditions http://www.jstor.org/page/info/about/policies/terms.jsp Article Contents p. 375 p. 376 p. 377 p. 378 p. 379 p. 380 p. 381 p. 382 p. 383 p. 384 p. 385 Issue Table of Contents Latin American Antiquity, Vol. 20, No. 2 (June 2009) pp. 267-391 Front Matter POSTMARITAL RESIDENCE PRACTICE IN SOUTHERN BRAZILIAN COASTAL GROUPS: CONTINUITY AND CHANGE [pp. 267-278] THE OCCUPATIONAL HISTORY OF GALINDO, MOCHE VALLEY, PERU [pp. 279-302] THE ROLE OF PROSOPIS IN ECOLOGICAL AND LANDSCAPE CHANGE IN THE SAMACA BASIN, LOWER ICA VALLEY, SOUTH COAST PERU FROM THE EARLY HORIZON TO THE LATE INTERMEDIATE PERIOD [pp. 303-332] MICROEVOLUCIÓN EN SAN PEDRO DE ATACAMA (NORTE DE CHILE): EL CEMENTERIO DE QUITOR [pp. 333-342] STYLE AND SUBSTANCE: EVIDENCE FOR REGIONALISM WITHIN THE AZTEC MARKET SYSTEM [pp. 343-374] CENTRAL PLAZA BURIALS IN SALADOID PUERTO RICO: AN ALTERNATIVE PERSPECTIVE [pp. 375-385] REVIEWS AND BOOK NOTES Review: untitled [pp. 386-387] Review: untitled [pp. 387-388] Review: untitled [pp. 388-389] Review: untitled [pp. 389-390] Book Note [pp. 390-391] Back Matter