The meaning of color terms: semantics, culture, and cognition

This paper argues that color concepts are anchored in certain “universals of human experience”, and that these universals can be identified, roughly speaking, as day and night, fire, the sun, vegetation, the sky, and the ground. Although our color sensations occur in our brains, not in the world outside, and their nature is probably determined to a large extent by our human biology ( which links us, in some measure, with other primates), to be able to communicate about these sensations, we project them onto something in our shared environment. Kay and McDaniel (1978: 621) have claimed that the semantics of basic color terms in all languages directly reflects the existence of pan-human neural response categories. But how can language be directly” linked to neural responses? Language reflects conceptualizations, not the “neural representation of color … in the pathways between the eye and brain” (Kay and McDaniel 1978: 617). The link between the neural representation of color and the linguistic representation of color can only be indirect. The way leads via concepts. Sense data are “private” (even if they are rooted in pan-human neural responses), whereas concepts can be shared. To be able to talk with others about one's private sense data one must be able to translate them first into communicable concepts. This paper argues, then, against the current accounts of color semantics such as those proposed by Kay and McDaniel (1978) or by Rosch (1972); and it proposes a new interpretation of the evolutionary sequence discovered by Berlin and Kay (1969).