This week's reading presents arguments for and against the existence of basic color categories (BCC). Hardin argues that such categories do in fact exist in human perception, citing numerous studies in psychology, neurophysiology, and anthropology. Van Brakel objects to Hardin's conclusion, claiming 1) that in many cases the studies Hardin cites are suspect, 2) the empirical data doesn't support the conclusion of BCCs, and 3) several studies suggest contrary evidence.

Hardin's book, COLOR FOR PHILOSOPHERS, presents a long argument drawing from the psychology of perception, physiology of the retina, and anthropological studies to show that we, qua humans, have a particular set of BCCs. He suggests that there are six primary colors: black, white, red, yellow, green, and blue. Additionally, humans classify colors as pink, orange, brown, purple, and grey. These colors, while they are a mixture of two or more primary colors, are recognized as distinct and thus deserve their own BCC, at least socially if not biologically.

Hardin spends the most time discussing the physiology of the retina, especially the pigment cells, rods and cones. He explains that there are three basic types of cones: short-, medium-, and longwave. He argues that each of these can be correlated physiologically with particular primary colors. Hardin infers that we have a biological primacy for perceiving certain colors, particularly red, yellow, green, and blue. This is not merely an accident of our language but a fact of our physiology.

Further, the primary colors are perceived according to an opponent system of three color pairs: red/green, blue/yellow, and white/black. This physiological fact is significant because of its correlation with information we have regarding the psychology of color perception. While we recognize that orange is a mixture of red and yellow or purple is a mixture of red and blue, we do not recognize any color that results from mixing red and green or blue and yellow. Hardin argues that these opponent colors CANNOT be mixed due to a physiological limitation. Red and green are perceived through opposing processes of the same retinal cells; thus seeing one necessarily prevents seeing the other (and similarly for blue and yellow). Hardin concludes that our scheme for BCCs is limited due to these opponent systems. Also, the six basic colors are primary and categorically separate because of our biology.

Hardin's main argument in the third chapter is based on a study by Berlin and Kay (1969) who tested different people across many different languages and cultures as to how they categorize colors. Berlin and Kay used 320 color chips of maximum saturation but varying hues and brightness from across the range of visible light. The subjects were asked to group the chips into different color classes. According to Hardin, the study indicates that humans universally classify all color variations into eleven groups, the six primary colors and pink, purple, orange, brown, and grey. Moreover, there was a natural progression to the order in which the colors are chosen. Hardin states that in all cases, black and white were considered by the subjects to be the most primary or first chosen colors. Next, red was chosen, followed by either yellow and green (in no particular order), then blue, and finally the five additional colors (in no particular order). Some cultures and languages distinguished fewer than eleven BCCs–some as little as two. In these cases, the subjects chose classes according to the same order. In other words, all those cultures with four BCCs had black, white, red, and either green or yellow, but not both green and yellow and not blue.

Hardin infers from this data that not only do we have a natural system of eleven BCCs but that there is a natural evolution of the colors, beginning with black and white, moving to red, then yellow or green, and so on. The evidence from the Berlin-Kay study shows that colors are not ingrained in a language or culture, but in some lower level, i.e. biological, state. As further support, Hardin cites studies where people whose languages do not make particular color distinctions more easily learn to classify colors from across BCCs than from within BCCs. So, If a person's language does not distinguish red and pink, she will initially not distinguish the colors (presumably, she will classify shades of pink as shades of red). If she is presented with several Mansell chips of shades of red and pink, she will quickly and easily learn to separate the two. But, if a similar experiment is performed where the chips are different shades of red, the subject is much slower and less successful in distinguishing maroon from scarlet.

Hardin cites several other studies that he believes suggest biologically extant BCCs. These include studies where infants and small children who have no or very limited use of language also make color distinctions. He is also unmoved by inverted and shifted spectrum arguments, claiming that the biological basis of color vision makes such cases unrealistic or outright impossible. Similarly, Hardin claims that the problem of multiple reliability in reducing psychological color categories to physiological color categories is not at issue. Generally, Hardin's position is that BCCs are a HUMAN phenomenon, so issues regarding misconnected or non-humans is irrelevant to his discussion.

Van Brakel responds to Hardin's arguments primarily by attacking the Berlin-Kay study and Hardin's interpretation of the results. He also discusses some other philosophical mistakes Hardin makes in his book. Van Brakel concludes that there is no neurophysiological data that supports a biologically ingrained system of color classification nor do the psychological and anthropological studies of Berlin and Kay suggest such a classificatory system.

Van Brakel's attack on the Berlin-Kay study comes from several fronts. First, he preludes his objections by pointing to earlier color distinctions based upon the spectrum. The original seven color system of Newton (Roy G. Biv) is completely arbitrary, with very particular and unnatural conditions required to see seven distinct colors. Discussing Berlin and Kay directly, van Brakel questions the methodology used by the researchers. He states that the study was biased from the starting block because of their use of the Mansell coloring system which is based on a western division of colors. Also, the study, which intended to determine BCCs based upon the basic color terms (BCTs) of a language presumed that BCTs serve only a functional role in color discernment, while in several cultures, color terms–both the names themselves and the number of names–have significant cultural and mythical implications that effected the results of the study.

Van Brakel also questioned the way in which the study defined color terms. Hue, saturation, and brightness are loaded components of color derived from our language (at least potentially). Many languages separate colors not only according to their appearance but according to the object to which they are attached. For example, we have a hair color category of BLONDE, yet we generally do not think of blonde as a color in any other circumstance. In this instance the Berlin and Kay study might say something like, "In English, BLONDE is a shade (subset) of YELLOW." The most important criticism of Berlin and Kay's methodology, according to van Brakel was their use of the evolutionary sequence of color categories. If a subject did not pick out black and white as their first color categories, the researchers assumed them to exist anyway. Also, if a subject picked as the fourth color a term that meant green, and blue (since some languages do not distinguish these colors), the researchers assigned it the value GREEN, since green comes before blue in their evolutionary schema. This arbitrary placement happened in several cases.

Van Brakel concludes that we have no reason to accept the eleven basic color categories defined by the Berlin-Kay study or any similar study. More importantly, he feels that no such anthropological evidence supports any universal color categories existing across human languages.

Next, van Brakel discusses Hardin's arguments regarding the physiological basis for color categorization. Van Brakel claims that Hardin makes several hasty inferences and fails to see the complexities of visual psychophysics. Hardin discusses in great detail the workings of the retinal photoreceptors, but he still glosses over key details. First, the information about the opponent systems involved in vision only discuss boundaries and achromatic distinctions without reference to color or hue. Second, Hardin's psychophysical account begs the very tricromatic color system it presumes to describe. Van Brakel says that we do not have enough knowledge of the workings of the brain to infer Hardin's model. Finally, van Brakel contends that the psychophysical information that Hardin presents does not support his reductive materialism regarding color perception. Hardin wants to reduce our trichromatic color system to the tripartite of cones in the eye, yet we not only have no good reason to do so, but the reduction does not yield the right connections between phenomena.

Van Brakel concludes by suggesting that we should view color naming and categorization in a functional manner. This explains both the cross linguistic similarities and differences and emphasizes the arbitrariness of color categories without suggesting categories do not exist.