Archaeological Records from the Southern and Eastern Caribbean Area.

How Different and how Similar are they?

Aad H. Versteeg

© 1999  A.H. Versteeg, The Archaeological Museum Aruba.

This text can be used, if properly cited. The correct citation is:
Versteeg, A.H., 1997: Archaeological Records From the Southern and Eastern Caribbean Area. How Different and how Similar are they? Paper presented at the 18th International Congress for Caribbean Archaeology, Nassau, Bahamas, july 1997, (in press).

Keywords: archaeology, prehistory, Aruba, Dabajuroid, St. Eustatius, Saladoid,

Summary
Extensive contiguous areas have been opened up in five ceramic period sites in the eastern and southern Caribbean area, the largest ones being the Golden Rock site on St. Eustatius and the Tanki Flip site on Aruba. All five sites studied belong to the Saladoid (and subsequent Post-Saladoid) and Dabajuroid Traditions and date between ca 500 AD and ca 1300 AD. Striking corresponding aspects are noted in their archaeological record. For instance, in the settlement organization such as this is expressed in the most common house types, the caches, the plazas, ceremonial structures, but also in burial gifts and other aspects. The settlement organization of these communities is believed to be a reflection of their social organi-zation. The common elements (there are also differences) are suggestive of common roots of the Saladoid and Dabajuroid people. Such roots must be sought in a formative agro-ceramic phase in the Upper Orinoco, or in the Upper Amazon area. That so many aspects of the 'proto-culture' were kept through time by both groups, argues for strong and long-standing continuities. José Oliver already argued such common roots in 1989, mainly on a linguistic basis.

This paper was presented during the 18th International Congress for Caribbean Archaeology held in Nassau, Bahamas in july 1997. It was handed out in print to colleagues.
This paper will be published in the Proceedings of this congress. They have not yet been published up to now.

INTRODUCTION
Recent large-scale excavations of uninterrupted areas of Ceramic Period sites in the Caribbean region (Figure 1) have yielded relatively rich archaeological records. An uninterrupted area of ca 240 m2 (ca 12 x 20 m) or more has been excavated in five sites of which the most important data have been published. Two of these five sites are on St. Eustatius (Figure 2)(Golden Rock at 3300 m2 and Smoke Alley at 255 m2), one is on Saba (Kelbey's Ridge at 3822), and two are on Aruba (Figure 3) (Santa Cruz at 840 m2 and Tanki Flip at 2275 m2). The five sites yielded floor plans of several domestic structures: ten at Tanki Flip (NOTE 1), six at Golden Rock, six at Kelbey's Ridge, two at Smoke Alley and two at Santa Cruz. This is a total of 26 domestic structures. Two structures in plaza-areas are interpreted as buildings for ceremonies. Burials were found in all sites. Special finds that are interpreted as caches were encountered in some of the sites. The question this paper addresses is the following: how do these archaeological records compare, and what are the implications of the shared respectively not-shared aspects?

The most important data are to be found in five publications. Two of these are monographs: Versteeg & Schinkel, 1992 on the Golden Rock site, and Versteeg & Rostain, 1997 on the Tanki Flip site. Three other publications discuss the three other sites: Versteeg, Schinkel & Wilson, 1993 on Smoke Alley; Hoogland, 1996 on Kelbey's Ridge, and Versteeg, 1997 on the Santa Cruz site.

These archaeological records came into being through actions of three Indian groups: Saladoids (Golden Rock site) and Post-Saladoids (Smoke Alley and Kelbey's Ridge sites) in St. Eustatius and Saba and through actions of Dabajuroids (the historical Caquetío) in Aruba (Santa Cruz and Tanki Flip sites). These distinctions are primarily based on the different pottery of these three groups; at least their decorated pottery is classified into different groups. All sites are dated to the relatively late part of the pre-Columbian era, between ca 500 AD and 1300 AD.

DOMESTIC STRUCTURES AND SPECIAL STRUCTURES
At Golden Rock in St. Eustatius, large, circular pre-Columbian malocas were ex-cavated in which large and heavy posts were in the outer circle (Figure 4) and in the interior area. Sizes are up to ca 19 m diameter (Figure 5) and (Figure 6) (NOTE 2). Oval malocas of 9 x 14 m were excavated at Tanki Flip (Figure 7) and Santa Cruz (Figure 8) in Aruba, all oriented strictly E-W. The postholes were shallower and the posts were smaller than the Golden Rock ones.
There is an important difference in shape and size (the Statia houses are larger), but the other differences probably reflect more the availability of timber and the type of subsoil of the settlement than anything else. From these few examples (Santa Cruz yielded a similar maloca as the Tanki Flip ones [Versteeg, 1997]), and noting that José Oliver excavated a similar maloca (NOTE 3) in Falcón at the FAL-7 site (Figure 9) , a first glimpse appears that oval and round malocas of sizes between 12 and 22 m maximum length were a wide-spread phe-nomenon in the pre-Columbian Caribbean. They are present in five sites if we include those from the FAL-7 and Luján I sites (see note 1) to those from the St. Eustatius and Aruba sites. These large domestic structures were either extended-family-dwellings or multi-family-dwellings.
A smaller, circular house type of ca 8 m diameter was excavated at all five sites (Figure 10) primarily discussed in this paper (see note 2 for a possible sixth site with comparable structures). It seems to have been the other widely pre-vailing house type. In fact, judging on the basis of the limited archaeological record excavated up to now, this would have been the most common pre-Columbian domestic structure. It is striking that all structures of this size (between 6 and 9 m diameter) are circular. This type of structure typically provides housing for a nuclear family.
Structures interpreted as ceremonial structures are reported for Golden Rock and Tanki Flip. In both cases it concerns structures, smaller than 6 m, with corners at 900. These two are the only rec-tangular structures excavated and reported in the pre-Columbian Caribbean up to now. Both are situated in areas that were kept open through time and that are interpreted as plazas. Special objects and/or artefacts (caches) were found buried in these plazas. For an example from the ethnographical record of oval malocas around a plaza with a small rectangular ceremonial structure see Basso, 1973:45. (Figure 11).


CACHES
Caches (items buried on purpose not being (parts of) human bodies are reported from Tanki Flip, from Golden Rock and a single one from Smoke Alley. In these sites complete pottery items and parts of sea turtles were found in such a context that it is clear these vessels were not dumped but buried on purpose. At Tanki Flip, stone beads, complete pottery vessels and bones of non-endemic animals (deer and cat/ocelot) were exclusively found in features interpreted as caches (and burials).

At Golden Rock, the only stone beads found in the site were in one feature, the child burial discussed above (Figure 12). In this Saladoid site ánd in the Dabajuroid site at a 1000 km distance, stone beads held a very special place in society. The same holds true for complete pottery vessels (Figure 12) (Figure 13) (Figure 14). Items with a ceremonial background that are specific for the Dabajuroid sites is the non-endemic animal bone discussed above. Zemis are exclusively found in the (post-) Saladoid sites (Figure 15).


POTTERY
Saladoid (and the subsequent post-Saladoid) and Dabajuroid pottery is classified as belonging to three groups and to two Traditions (Saladoid and Dabajuroid). However, if the pottery collections of these Traditions are compared on a more basic level, both groups have bowls, vessels and griddles. The Dabajuroid sites have large burial urn vessels that are lacking in the Saladoid ones. Decoration elements of both groups comprise painting of a small part of their better finished pottery products in two or three basic colours, purple being the most frequently used Dabajuroid color and red and white the Saladoid one. Both Traditions have pottery adornos that represent stylized humans or animals. Naturalistic representations are not found among these adornos. For Tanki Flip this group was described as canonized products (Versteeg in Versteeg & Rostain, 1997:302) and that term applies equally well to the distinct, strictly limited patterns of the Saladoid Golden Rock pottery (Versteeg in Versteeg & Schinkel, 1992:71-2).
Incisions are not as well represented in Dabajuroid sites as they are in Saladoid ones. The execution of the pottery certainly is different between these groups. However, the basic concepts and the function of the pottery seem to be similar. Pottery that fulfilled utilitarian functions is found only in sherds. Complete pottery items are found either in burials or in caches to which a ceremonial function is ascribed as has been discussed above. And there is in both groups ser-ving ware sensu Oliver, at least in the Saladoid and Dabajuroid sites. That type of very well-made, thin pottery is only lacking in the post-Saladoid sites in the Lesser Antilles. Also, these sites show up a low percentage of decorated pottery (less than 2% for the post-Saladoid sites in islands such as Nevis, St. Eustatius, and St. Kitts, compared to ca 20 % in the preceding Saladoid period in those islands). Probably the function of pottery changed consider-ably in the transition Saladoid post-Saladoid. Note that this contrasts sharply with events in the Greater Antilles where the post-Saladoid pottery is rich in decorations.

ORIGIN
José Oliver states the following on the Dabajuroids:
''Speaking the Caquetío language, these peoples who produced the Dabajuro style, could only have come from the Llanos, and ultimately from a Proto-Arawakan hearth. No longer should we argue for western vs. eastern centers, and no longer does de "H" theory adequately portrays the cultural historical processes that characterize Venezuelan prehistory. Moreover, if anything, I have shown that the prehistory of Venezuela is intimately correlated with the riverine network of the Orinoco Basin (including the Llanos). It is this rather large lowland area, but in specific niches, where I find that the key to the understanding of the cultural history of the northern parts (Caribbean Coast) of South America (Oliver, 1989:505).

José's words imply that the Dabajuroids originate from the Upper Orinoco area (NOTE 4) or beyond (a Proto-Arawakan hearth); see also Wagner (in: Oyuela-Caycedo & Scott Raymond, 1998:147-8) for a discussion of the same view. The Saladoids also originate from Upper Orinoco areas or a Proto-Arawakan hearth. If they have a common origin, it comes as no surprise that they share organizational, cultural and social concepts such as: circular 8 m diameter houses but also malocas that have a surface area, on the average, of twice the basic house, stone beads and complete pottery products in caches and burials, sea turtles, and in Aruba, exotic animal bone (i.e. bone of mainland animals).

It is striking that the post-Saladoid archaeological records (of Smoke Alley and Kelbey's Ridge) do not have malocas. Also, houses at Tanki Flip, interpreted as the latest ones in time (Str-4 and -12), are smaller than the earlier ones (Versteeg in Versteeg & Rostain, 1997:455). Not only the percentage of decorated pottery seems to decrease considerably in the transi-tion from Saladoid to post-Saladoid (from ca 20 % to less than 2 % [Versteeg in Versteeg & Schinkel, 1992:72]), but the house size seems to be limited to relatively small structures, at least according to the information that is presently available.

CONCLUSIONS
The sites discussed share house concepts and house shapes, stone beads, complete pottery items and bones of specific animals in special features (caches and burials). Basic aspects of pottery decoration and the function of pottery seems to be similar for the Saladoid and Dabajuroid sites. In the larger sites (Tanki Flip and Golden Rock), some areas were kept free of building activities through time (interpreted as plazas) except for right-angled structures (interpreted as ceremonial struc-tures). The conclusion is that these peoples, previously ascribed to different cultural zones - i.e. de eastern and western Venezuelan cultural zones -, share many concepts that must have been important on an organizational (e.g. the houses and plazas) and symbolic (e.g. the caches, burials and rectangular plaza-structures) level of these societies.
These specific conclusions can be drawn because the used type of investigation provides context for the excavated features and for some clusters of excavated features. The function of some features such as the caches reported for Golden Rock and Tanki Flip could be inferred from their special position and their special content in these large-scale excavations. That is unusual in the Caribbean region and for that reason the number of sites studied here has been limited to five.
This does not mean that there are no differences to be noted. What is typical for the Aruba sites are secondary burials in urns, hearths with different types of stones associated with burials and caches. The St. Eustatius and Saba sites have zemis as their typical artefact with a ceremonial background.
Parallels such as documented in the houses, caches and the ceremonial use of stone beads and of canonized pottery products, the similar basic elements in this pottery suggest that both groups have a common origin. Long-standing traditions and a considerable amount of cultural continuity make the common origin of a proto-Arawakan group recognizable far from the origin in time and geography; the differences show that changes occurred after they left the Upper Orinoco region and after they changed their former riverine environment for a coastal/insular one.

A view of the pictures of this article

Literature cited

Ardila Calderón, G.I., 1996. Los tiempos de las conchas. Editorial Universidad Nacional. Bogotá. 249 pp.

Basso, E.B., 1973. The Kalapalo Indians of Central Brazil. Holt, Winehart & Winston, Inc., New York etc.. 157 pp.

Goldman, I., 1979. The Cubeo, Indians of the Northwest Amazon. Univer-sity of Illinois Press, Urbana (2e ed). 315 pp.

Hoogland, M.L.P., 1996. In Search of the Native Population of pre-Columbian Saba (400-1450 A.D.). Part two. Thesis Leiden University. 286 pp.

Oliver, J.R., 1989. The Archaeological, Linguistic and Ethnohistorical Evidence for the Expansion of Arawakan into Northwestern Venezuela and Northeastern Colombia. Thesis University of Illinois, Urbana-Champaign, Ill. 756 pp.

Oyuela-Caycedo, A & J. Scott Raymond (eds.), 1998. Recent Advances in the Archaeology of the Northern Andes. In Memory of Gerardo Reichel-Dolmatoff. Monograph 39 of the Institute of Archaeology of the University of California, Los Angeles. 173 pp.

Rivera Calderon, V., 1997. Estudio preliminar de la Distribucion espacial en la Comunidad aborigen de Luján I. Paper pre-sen-ted at the 17th International Congress for Caribbean Archaeology, Nassau, Bahamas. 9 pp.

Versteeg, A.H., 1993. Settlement Patterns within the Dabajuroid Santa Cruz site (Aruba). Paper pre-sen-ted at the 15th International Congress for Caribbean Archaeology, Puerto Rico. Internal Report Archaeological Museum Aruba 13:1-6.

Versteeg, A.H., 1997. Pre-Columbian Houses at the Santa Cruz site. In: Alofs, L., H.E. Coomans & W. Rutgers (eds.). Arubaans Akkoord, Stichting Libri Antilliani, Bloemendaal:89-101.

Versteeg, A.H. & C. Schinkel (eds.), 1992. The Archaeology of St. Eustatius: the Golden Rock site. Publication of the St. Eustatius Historical Foundation 2, Publication of the Foundation for Scientific Research in the Caribbean Region 131. 284 pp.

Versteeg, A.H., C. Schinkel & S.M. Wilson, 1993. Large-scale excavation versus surveys: examples from Nevis, St. Eustatius and St. Kitts in the Northern Caribbean. Analecta Praehistoria Leidensia 26:139-161.

Versteeg, A.H. & S. Rostain (eds.), 1997. The Archaeology of Aruba: the Tanki Flip site. Publications of the Archaeological Museum Aruba 8, Publications of theh Foundation for Scientific Research in the Caribbean Region 141. 519 pp.