Taxonomies take on a number of forms and vary from domain to domain and culture to culture. One way of analysing and comparing semantic structures is to record how rich they are in terms of the number of lexemes applied to a given range of phenomena. A large number of terms applied to a domain marks it as being of particular importance for the lifeways of the people who developed the semantic system. The usual example is the Inuit classification of snow into dozens of forms because of the vital importance of this material to their survival. This assertion has been called into question because Inuit word structures are quite different from English, and each "word" for snow is in effect a short descriptive phrase.
For further details see:
Spradley's taxonomy of "flops" (partial listing)
The domain of colour terms has been of particular interest to anthropologists, who have discovered intriguing differences in the number of special words for colours used from culture to culture (Berlin and Kay, 1969). Some cultures have only two terms, and numbers range from this minimum to over twenty. Furthermore, there is a surprising regularity in the actual selection of the specific terms. When two terms are used, they are invariably "black" and "white." A third term will always be "red." The fourth and fifth colours recognized will be "yellow" and "green," the sixth, "blue." Additional colours will be added in a regular order as well. This pattern is clearly evident in English, which has only five of its own terms: black, white, red, yellow, and green. All the other special words we use for colour are either loans from French, e.g., "bleu," or terms that refer to objects, such as pink, a flower.
Cross-cultural differences in the number of colour terms and the regularities
in their variation defy adequate explanation. It is clear that different
colour taxonomies are not correlated with differences in perception, i.e.,
people who classify everything as black and white have quite adequate colour
vision. There is a correlation between the number of terms and technological
development, but it is too weak to suggest a definite regularity.