Chronological Research

By the 1920s, professional archaeologists had accumulated enough assemblages to use as the basis for setting up sequences of local periods, and by the 1930s, they had formed sufficient sequences to begin synthesizing them into regional charts or chronologies. As in other parts of the world (see, e.g., Willey and Phillips 1958), they placed the local periods whose assemblages were most alike at the same heights in the charts on the assumption that the inhabitants who were able to mutually influence each other would have produced the most similar assemblages. They checked their results by hypothesizing the existence of horizons, that is, local complexes of cultural traits presumed to have been contemporaneous, and by determining whether these “time markers” actually did have a horizontal distribution on their charts. If they did not, the charts and hypotheses were reconciled in an effort to produce the most accurate and most practicable systems of local periods within which to organize the assemblages.

Gudmund Hatt, a professor at the University of Copenhagen, pioneered the new approach. In 1924, he undertook excavations in the formerly Danish part of the Virgin Islands, which the United States had just purchased from his country. He classified his finds into three successive groups and named each group after sites from which he had obtained typical assemblages: (1) Krum Bay, (2) Coral Bay–Longford, and (3) Magens Bay–Salt River. He placed (1) at the beginning of his sequence because it lacked ceramics and put (2) before (3) because, wherever the two occurred in the same site, the assemblages belonging to (2) underlay those in (3). He also noted that (2) was characterized by white-on-red-painted pottery, which had not previously been found in the Greater Antilles, whereas the modeled-incised pottery of (3) resembled that of the Taino Indians, who occupied the Greater Antilles in Columbus’s time (Hatt 1924; Birgit F. Morse, personal communication).

In 1935, Cornelius Osgood, professor and curator of anthropology at Yale University, established a Caribbean anthropological program with the intention of tracing the spread of cultural traits from South, Middle, and North America into and out of the West Indies. He sent Froelich G. Rainey and irving rouse, his first two graduate students, to the West Indies with this aim in mind. Like Hatt, they realized that they would first have to do chronological research. Rainey, assisted by Rouse, set up a sequence of local periods in Haiti, which, like Hatt’s, consisted of a preceramic period followed by two ceramic periods. The first of the two was marked by local pottery, and the second by pottery in the Taino style resembling that encountered by Hatt in the Virgin Islands. Rainey then went to Puerto Rico where he was able to distinguish two ceramic periods comparable to those Hatt had found in the Virgin Islands (Rainey 1940, 1941).

Rainey named the two ceramic periods in his Haitian sequence after typical sites, as Hatt had done in the Virgin Islands, but preferred to call the two Puerto Rican periods Crab and Shell in recognition of the fact that such food remains were dominant in their respective assemblages. He was ahead of his time in calling attention to this shift in the local diet, but Rouse found when he continued the Puerto Rican research that ecofacts such as food remains do not provide an adequate basis for chronological research because their presence in archaeological assemblages depends upon their availability and crabs and edible shellfish are limited to the coasts of large islands such as Puerto Rico. In his doctoral dissertation, therefore, Rouse (1939) substituted the names of typical sites for Crab and Shell, following the example of Hatt in the Virgin Islands and Rainey in Haiti.

Rouse also synthesized the three established sequences into a regional chronology and continues to refine and expand this chronology as new information becomes available, adding other local sequences to his charts as they have been formulated (Fig. 2).