It has been my intention to track the fluctuations of meaning and implication in the early Greeks' use of color terms, as a basis for analysis of their understanding of these terms specifically, and the visual world in general. The first part of this project has been carried out by researchers far more competent than myself, such that I can only reiterate and remark upon the patterns they chart. The second tier, that of analysis, has seemingly been tackled repeatedly. The only excuses for adding to this debate would be if some pertinent evidence had been overlooked, or some new research been done that could be relevant. While the canons of Greek literature have not made any drastic recent shifts of which I am aware, the question of human color perception has been more active in the last few decades, and could come into play.

Greek color terms have long confounded historians, spawning as many interpretations as there are interpreters. The difficulty is most extreme in it's earliest appearance. Homer used color words dramatically to paint the events of his epics, but limited these to four, roughly translated as black, white, greenish yellow and purply red. He also distinguishes the shades of metals, as we know from the Ekphrasis on the Shield of Achilles, in which he describes how the different metals are used for visual effect, but goes on to describe the actual sky as bronze colored. The confusion to modern readers is in defining the words and describing their implications. While as centuries progressed terms standardized, it is hard to equate the early usages to later definitions. Homer's limited range of color has been fodder for developmental theories, as well as suggestions such as color blindness (with the implication that all ancient Greeks were color blind). The problem is not only that he refers only to four colors, but the wide variety of things he describes as being those colors. For example, the sea is the color of wine, as are sheep. The same color, usually interpreted as yellow-green, applies to honey, sap and blood. Even as later writers expanded their descriptive palate confusing word choices persist. Discussion of this issue over the years has been influenced by various color theories as they have arisen; in the wake of Darwin, the theory was advanced that the early Greeks' retinas had not evolved the ability to perceive colors. Irwin seems conclusive in both her summary of the theorists, and her support of her own argument that the terms of contention refer not to hue precisely but to surface qualities, for example, sheen or dampness. What she does not address (it is not necessarily her intent) is what these observations may reveal about the nature of color perception in general. To the extent that we have preserved intact the works of Homer we have a representation of an immediate response to visual stimulus relatively uninfluenced by constructed ideas of color. Irwin does not question modern notions of color as fact¹, but provides some impetus for doing so.

The hypothesis of linguistic relativity states basically that a person can only perceive what they have words to describe². It would propose that the Greeks in Homers time merely had a limited number of categories into which to divide the colors they saw. A greater variety of terms would have been developed as finer distinctions were observed and described. It is the case that all languages of which we have documented the history at one point had words only for a limited spectrum. Japanese, for example, has in its vocabulary about four native color terms, others have been added from English and Chinese, and behave differently grammatically. Also significant is that the earliest theorist on color, Empedocles, ascribed all the colors to variations between light or white, dark or black, and red and yellow, which must exist as elements in the eye in order to be perceived. This selection being fairly consistent with Homer supports the premise that earlier Greeks thought of color in only these distinctions. Aristotle and others ascribed all color to gradations of light and shadow³. Xenophanes described the rainbow as having three colors; a purple, a yellow green, and a red.

The linguistic solution would explain the limited vocabulary, but not its odd application. Irwin adds to this an interesting delineation. Her work suggests that not only were Homer's colors less defined as to chroma, the concept of surface color as opposed to texture and shade was less distinct. She suggests that one term could have contained all these meanings. Thus a word that seems to indicate yellow or light green also meant fluid, fresh and living, and in one context was appropriately used describe blood, as the human sap. It is important to realize that the idea of surface color that we now take for granted was not formulated until fairly recently, and is still contested. Having not had refined synthetic colors as we do, the Greeks may not have had our concept of pure chroma. The colors they as craftsmen were able to isolate and reproduce may have some bearing on the chromatic scheme of the language, although the relationship is not as simple and direct as a one to one correlation between color terms and pigments. While black figure vases roughly contemporary to Homer reiterate his color palate, earlier frescos of the late bronze age include a cobalt blue color, which presumably must have had a name. Perhaps even given a complete literary record we would not find a continuously developing concept of color throughout the history of Greek civilization.

Maxwell-Stuart points out that it may be worthwhile to consider prose and poetry separately, and to question the factual reliability of prose when assessing what is meant by a color term, although Irwin would disagree that such a distinction even existed in ancient writings. When the eyes of a tiger are called by a word which usually seems to indicate blue, it may be worth noting wether it is probable that the author ever saw a tiger⁴. The difficulty may be in the information, not the lexicon. In poetry, the symbolic use of color can not be overlooked. Homer, Euripides, and other dramatist repeatedly use black and white to indicate gender distinctions, white to accentuate the feminine, particularly vulnerability and beauty, and darkness to indicate masculinity, as in the epigram "white armed". Homer calls the flesh exposed outside a warrior's armor white, to point out the possibility of injury. These same conventions can be seen in visual renderings of the time. An Attic black figure vase dated c. 540-530b.c., roughly contemporary with Homer shows the black/white gender symbology, but within two decades the distinction becomes less consistent with the appearance of light skinned male figures. Fifty years later we see a mural such as that in the Tomb of the Diver, c. 480 B.C., which retains a dark light distinction between the sexes, but includes blue. In Greek literature black and white are used in relation to a wide variety of sets of opposites, night and day, death and life, earth and sky, bravery and cowardice. And there are a myriad of shades of implication between the two. It would be a mistake in any case to dismiss "poetic" usages as falling outside the realm of meaning or significance. The fact that a description does not appear literal does not negate its usefullness in delineating the range of color experience.

All early color terms had some application of physical illness or emotional upset or imbalance. This points up the power and weight these terms carried, but it is also interesting how each can be related to the body, often as an effect carried out on it. A boy's cheeks become purple, or Medea becomes pale, as likewise do her children, and Achilles' lungs darken in his anger. This can be related to the Greek's concepts of health and physiognomy, having as its basis the principle of four bodily fluids, called bile. These are identified by their color, yellow, red, dark or black, and clear or white. These words have physical and symbolic manifestations that run parallel. The fearful person becomes pale, but a pale animal will symbolize cowardice. Even as the selection of colors expanded, in later years they continued to be not entirely literal in use. Maxwell -Stuart's study of terms for eye color demonstrates that color had emotional implications; pale (or blue) eyed meant not only that, but suggested personality traits, again cowardice but also dishonesty, homosexuality and insanity. The direct correlation of color terms to the body exemplifies their pervasiveness and importance. Irwin notes patterns of color use in Homer that fit the development of narrative. A climax passage will always be marked by increased use of color words, and deliberate contrast.

The question of synaesthesia also arises. It has often been noted that color terms were applied by Homer to sounds and voices, but similarly, he includes sounds in the supposedly pictorial Ekphrasis. These may be strictly metaphor, or they may indicate the expanded range of meaning that these words had. In the case of color that occurs within the body, ostensibly out of sight, it seems very possible that the phenomenon is not strictly visual.

The question of what color meant to the Greeks remains pertinent as a key to the basic nature of visual perception. Simply put, if color exists in an ordered spectrum of seven hues, why has this not been observed, by the majority of cultures, until relatively lately? There is the linguistic factor, which would hold that even perceiving a full spectrum the viewer may be limited in their terminology, hence deriving fewer categories. Conversely, the variance between the accepted model of color perception, and how color has actually been observed prior to this model may suggest the fallibility of the model.

In the past ten years, much study has been made of the mechanics of the visual apparatus. It is commonly known that the human eye receives light signals through receptors referred to as rods and cones. The rods are active in night vision, and the cones are of three varieties, responding to three ranges of wavelength. Thus the various wavelengths that reach the eye are interpreted as color within the brain. Each type of cone is loosely responsible for interpreting one axis of color vision; red/green, yellow/blue and black/white: therefor our vision is referred to as being trichromal. Something that is less common knowledge is that the cones of a human retina do not actually correspond to the full spectrum of wavelengths. The spectrum of colors we presumably all see is actually interpolated by the brain from an abbreviated selection of readings. "Colours are not intrinsic to objects in the physical world or to the neural processes in the visual system; rather, they are properties of the world taken in relation to the perceiver. Thus on the question of whether colours are intrinsic properties or relational properties, I side with the received view that they are relational."⁵ As it turns out, the number of cone types and their range is genetically mutatible; hence color-blindness (bichromy rather than trichromy) only occurs in males. Many animals are bichromal, but some are also tetrachromal. There may in fact be human females who are tetrachromal, these test subjects exhibit slightly higher levels of color differentiation. The evolution idea may not have been so farfetched.

The points that are brought to bear by this foray into cognitive science are as follows; (1) the human eye does not absorb the full spectrum of light to interpret as color⁶, (2) color is almost certainly constructed to some extent by the mind, (3) color vision in primates has evolved and is subject to fairly rapid mutation⁷. Consider the case were two color axis interpreted as conjoined by the brain, supposing the red/green axis were at some point aligned with the blue/yellow. The result would be a perceived range of colors going from a purply-red to a greenish-yellow. The result would be the now-familiar color scheme of the early Greeks; it is quite likely that persons seeing in this manner would be highly aware of surface qualities and refraction, as both Irwin and Maxwell-Stuart have shown that the Greeks were. This conjecture would be absurd were it not that the development suggested between the time of Homer and modern times, is not a physical evolution but of a refinement of decoding systems, quite similar to the linguistic development that is already assumed to have occurred.

I would propose that our predecessors cannot be dismissed as mistaken when they consistently describe their picture of the world, even if it does not match our own. Any of us seeking to interpret the words of writers dead for millennia can only guess, and bring the best of our knowledge to the effort. What may be too often forgotten is what these records can in turn contribute to deciphering the knowledge of our own age. Considered as a whole research into our past and present history creates a more complete picture of human perception. Based on the work of many others, it is clear that the early Greeks' senses were not faulty, but more unified than ours today, at once limited but somehow fuller.

Bibliography

Byrne, Alex and David Hilbert, Ed. Readings on Color
Cambridge, 1997: Massachusetts Institute of Technology

Edgeworth, Robert Joseph. The Colors of the Aeneid
New York 1992: Peter Lang

Fountain, Henry. Proof Positive That People See Color With the Tongue
New York times, March 30, 1995

Goldhill, Simon and Robin Osborn, Ed. Art and Text in ancient Greek Culture
Cambridge, 1994: Cambridge University Press

Irwin, Eleanor Color Terms in Greek Poetry
Toronto, 1974: A. M. Hakkert Limited

Lattimore, Richmond, trans. The Iliad of Homer
Chicago, 195 1: University of Chicago Press

Maxwell-Stuart, P. G. Studies in Greek Colour Terminology, Volumes one and two
Netherlands, 198 1: E. J. Brill

Thompson, Evan Color Vision: A Study in Cognitive Science and the Philosophy of Perception
London and New York, 1995: Routledge