Abstract
Most world languages have basic color terms, but how these terms emerge over time is unknown. To address this question, we report color-naming data collected in the field on the Hadza, a population of nomadic hunter-gatherers in Tanzania. Their language, Hadzane, is a language isolate and probably represents an early stage of color term evolution. Fifty-five color-normal (HRR plates) informants provided single name or "Dont Know" (DK) responses for each of 23 color samples, including focal examples of the 11 English basic lexical color categories (BCCs). All informants provided names for the red, black and white (RBW) samples, with only slight variation across individuals in the names chosen. The other 20 samples frequently elicited DK (32±11 DKs/sample; 11±4 DKs/informant). When informants did name non-RBW samples, consensus was low (9±2 terms/sample), even for the focal samples. Yet individual informants named colors lawfully; samples given the same name by an informant were usually adjacent in color space. Sets of samples named with the same term by individuals often resembled English BCCs. These sets were similar across individuals despite their designation with different terms. Our results provide insight into how color terms are acquired. Berlin & Kay (1969) hypothesized that all languages have basic color terms that completely partition color space. Contrary to this view, and in line with previous findings on color naming in another language isolate (Levinson, 2000), the shared lexical representation of color among Hadzane informants is remarkably sparse. That said, there appears to be emerging consensus in Hadzane about which samples should be grouped together by name and this consensus is consistent with the structure of the universal BCCs. More broadly, our data suggest that while not all colors necessarily belong to a BCC, when BCCs do emerge their locations in color space are subject to cross-culturally universal constraints.
Meeting abstract presented at VSS 2014