the use of chronological charts as many Caribbeanists assumed that the newly developed techniques of measurement rendered the charts obsolete. Experience has shown, however, that neither is sufficient by itself. The results of the two procedures need to be checked against each other in order to identify and eliminate invalid dates and to improve the accuracy of the charts.

The charts and dates have been called measures of absolute time because their temporal values remain the same throughout the regions covered. Caribbeanists also work with measures of relative time known as ages, which vary from place to place. Five ages are distinguished: lithic, which is marked by stone chipping; archaic, by the addition of stone grinding; ceramic, by the first appearance of pottery; formative, by the development of ball and dance courts; and historic, by the introduction of writing. These ages have an irregular distribution when inserted in the chronological charts because the innovations that define them took place at different times in different places (Fig. 2).

Culture-Historical Research

Chronological charts, dates, and ages play a number of roles in research on the final two levels of interpretation. They may be used at the start of a project to identify sites, artifacts, and other information pertinent to the problem under study and to retrieve these data from the ground or from storage. To assist in doing the latter, the Yale Peabody Museum, which is now computerizing its Caribbean collections, is adding the name of the local period when each item was produced to the information already recorded about the nature of the item and the locality from which it came (Hill and Rouse 1994).

More important, the chronological systems have added a new dimension to the study of the history of cultural traits envisaged by Osgood when he founded the Yale Caribbean program. They have made it possible to reconstruct the trajectories of traits as they spread from period to period as well as from area to area and to study the changes that took place en route. For example, prior to the construction of chronological charts and the advent of radiometric dating, it was assumed that the custom of building ball and dance courts had diffused from Mexico to the Greater Antilles because there are superficial similarities between the courts in the two regions. Thanks to chronological research, local archaeologists have been able to show that the Antillean courts originated in Puerto Rico during the latter part of the first millennium a.d. and spread east and west from there without ever reaching Jamaica or central and western Cuba, the parts of the Antilles closest to Mexico. The courts became simpler as they spread (González Colón 1984; Rouse 1992, 112–116).

West Indian scholars originally assumed that all their finds had been produced by the ethnic groups who inhabited the area in the time of Columbus, that is, by the Guanahatabeys (also known as Ciboneys), Igneris, Tainos, and Island-Caribs (Map 3). For example, Heye attributed all of the Lesser Antillean artifacts in his Museum of the American Indian to the Island-Caribs. And when the Swedish archeologist Sven Lovén wrote his scholarly summary of West Indian archaeology and ethnology in the 1930s, he began it with a statement that all four ethnic groups had come full-blown from either North or South America (Lovén 1935, 2). That assumption relieved him of the necessity of studying the evolution of the groups’ cultures, since that would have taken place before they reached the islands. He only needed to concern himself with the histories of individual traits.

Lovén’s assumption soon began to be contradicted by the results of chronological research. Some sequences of local periods in the Greater Antilles have now been extended back to ca. 4000 b.c. (Rouse 1992, Fig. 14), and the ethnic groups encountered by Columbus could hardly have retained their separate identities over such a long period of time. Consequently, we must distinguish the historic ethnic groups from their prehistoric ancestors and predecessors.

Ethnic groups are defined in terms of sociocultural criteria, which are difficult to infer from prehistoric remains—hence the position of sociocultural research at the end of the sequence of stages under discussion here. In the absence of adequate knowledge of sociocultural criteria, Caribbeanists have formulated prehistoric