New Book: World Color Survey

Using data collected in 110 languages over 30 years, ICSI's Paul Kay and Richard Cook, along with other researchers, have supported and expanded Kay and Brent Berlin and Paul Kay's 1969 hypothesis that the world's languages conform to a universal pattern as they develop names for colors. In The World Color Survey, released in February by Stanford's CSLI Publications, the researchers argue on the basis of the new data that there are indeed universal constraints on the ways in which languages name colors and how color naming systems evolve.

When Berlin and Kay's Basic Color Terms was published in 1969, the prevailing doctrine held that local cultural factors determine how colors are named in different languages, specifically that "each culture has taken the spectral continuum and has divided it upon a basis which is quite arbitrary." Berlin and Kay found that color terms in different languages tend to be based on a limited number of colors, and that languages acquire new names for colors in a partially fixed order.

In 1976, Kay, Berlin, and the late William Merrifield established the World Color Survey (WCS) project to investigate these hypotheses more fully and to respond to (legitimate) criticisms. Critics argued that Basic Color Terms had been based on a small number of speakers, mostly of written languages spoken in developed societies, who were also speakers of European languages and who were currently living in the U.S. The WCS project gathered data on 110 unwritten languages in their own localities to test Berlin and Kay's hypotheses. While finding general confirmation of the hypotheses of universality and evolution of color naming systems, the project has also made several significant revisions.

The new book revises the model of how color–naming systems evolve. The current model admits the possibility of a language with only two color terms, as previously documented, but the survey did not document any such languages. In such a language, there is one term comprising white and all the "warm" colors and another comprising black and all the "cool" colors. A three–term language separates a "warm" term from the earlier warm–white term, producing a system of white, warm, and black–cool terms. Further stages of development successively divide these categories until a six–term language comprises separate terms for black, white, red, yellow, green and blue. Terms such as brown, purple, orange, pink, and gray tend to be added late in development, but not at precisely predictable points.

A separate chapter is devoted to the analysis of each of the 110 color terminologies. Each chapter contains several tabular and diagrammatic summaries of the data as well as an overall descriptive summary and evolutionary categorization of the color naming system. The 110 individual language chapters are preceded by chapters dealing with the history of the subject and the methods of the survey.

The book is authored by Paul Kay, Brent Berlin, Luisa Maffi, William Merrifield, and Richard Cook.