M. Michela Sassi
We all live in a colorful world. Color is an essential feature of things as we perceive them with our most precious sense, vision, and thus it is fundamental not only to our knowledge of the outside world, but to our experience of it as well. In fact, amongst the sense qualities color is also the most effective in inducing responses in the pleasant–unpleasant, beautiful–ugly continuum. This makes the realm of color aesthetics a particularly complex one, and, in fact, an understanding of how colors affect people in their everyday experience (i.e. aesthetics in Baumgarten’s sense) is necessary at a lower, yet basic level, in order to investigate the level of the aesthetic discourse, where explicit aesthetic judgments on colors in both worlds, of nature and art, take place. If the field we intend to explore is that of Classical antiquity, such inquiry is even more complicated. I am not only thinking of the well-known fact that the Greeks and Romans did not produce any systematic aesthetic theory, so that we are allowed to reconstruct at best a few general trends through prudently reading a vast set of literary texts, philosophical writings, and art works. With color aesthetics we have to raise a further core issue, namely, whether the Greeks and Romans “lived” colors the way we do, with our “modern” sensibility.
A negative answer was indeed given to this question by several scholars and scientists during the nineteenth century, who were attempting to explain some “odd” features of the color vocabulary to be found in the texts of Greek literature (as well as in the Old Testament and the Veda). Goethe led the way in his Farbenlehre (1808–1810), where he observed, for example, that the term xanthos can cover most shades of yellow, from the shining blond hair of the Homeric heroes to the blaze of fire, and therefore he deduced that the Greeks were not used to distinctly defining the yellow area of the Newtonian spectrum with respect to red, on the one hand, and to blue, on the other. Later on, William Gladstone, the eminent politician and Hellenist, insisted in his Studies on Homer (1858) on the idea that Greek color vocabulary displays maximum sensibility to light impressions (think of the word leukos, whose root is the same as that of Latin lux), as opposed to a quite vague discrimination of the prismatic colors. In particular, Gladstone remarked on the lack of words specifically meaning “blue”: in fact, kyaneos denotes a dark shade of blue merging into black, whereas glaukos denotes a sort of “blue-gray,” in which the impression of brightness definitely prevails over the indication of hue – remember Athena’s epithet glaukopis, referring to the “gray-gleaming eyes” the goddess shares with her sacred bird, the glaux; or that untranslatable word, porphyreos, which covers a large range of shades in the red area, but can also trespass onto the blue one, when used for the shining reflections of the sea. All in all, there was enough for Gladstone to conclude that the visual organ of the ancients was still in its infancy, so paving the way for diagnosing the Greeks as blue- and yellow-blind (Schultz 1904). The diagnosis, soon extended to the Romans, although in a lesser degree given the greater precision of the Latin lexicon, would become a lasting topos, favored by the post-Darwinian climate of the late nineteenth century. In the following decades, such a conception prompted a host of investigations either on specific writers or on genres of Greek and Latin literature striving to prove that the ancient chromatic categories, “vague” as they are, do not fit in with modern taxonomies.
It may be worth remembering that, according to the well-established trichromatic theory of vision, which was first proposed by the English physician Thomas Young at the very beginning of the nineteenth century and elaborated by Hermann von Helmholtz half a century later, there are three types of cones (the photoreceptors, owing their name to their shape, that are specifically responsible for the vision of colors and details), which are preferentially sensitive to the blue, green, and red regions of the chromatic spectrum, that is, to the short, medium, and long wavelengths of light. This model can explain various kinds of color vision deficiency, or Daltonism, through dysfunctionality of one or more sets of cones, the most severe defectiveness of them causing total inability to see colors (people with complete achromatopsia must rely on their rods, the photoreceptors specialized for night vision). Today we know that cone disorders are caused either by accidental damage to the individual’s retina, optical nerve, or brain (as was the case for the painter who became color-blind, famously described by Oliver Sacks; Sacks and Wassermann 1987) or by inherited diseases, whose spread rates inside a given group are determined by genetic mechanisms (remember the color-blind population of the Micronesian atoll of Pingelap that has been studied, again, by Sacks 1997). These are, however, pathological variations of the color vision capability humans have had since they have evolved. It seems that only primates, and not all of them, have trichromatic vision, which must have obvious selective advantages in allowing better recognition of predators, food, and mates, and no one would think today that there has been a stage in the history of humanity when some colors were “not yet” perceived.
In any case, throughout the twentieth century the developmental account of color categorization gradually yielded to an approach more suitable to the anthropological paradigm of the time. To jump to the end of this story, which, while fascinating, is too complex to be included here (Sassi 1994, 2003), the common view held by the scholarship on color vision today is that every culture has its own way to name and categorize colors, which is not determined by the anatomical structure of the human eye (which is uniform), but by the fact that different ocular areas are stimulated, and do trigger different emotional responses, according to different cultural contexts. In short, humans cross-culturally tend to react differently to different wavelengths. There was, however, a grain of truth in the old view, inasmuch as the processing of sense data, if not the quality of the perception itself, seems indeed to have had a development through time in the various cultures, as I am going to argue by introducing a notion drawn from modern optical research. According to the most widely used color system, elaborated at the end of the nineteenth century by Albert Henry Munsell, an American artist, any color sensation is defined by three interacting aspects: the hue, determined by the position in the Newtonian spectrum, by which we discriminate one color from another; the value or lightness, ranging from white to black; and the chroma, corresponding to the purity or saturation of the color, depending on the distribution of the wavelengths (fire red and sky blue, for example, are highly saturated, whereas gray, white, and black are not at all so; artificial colors tend today to be more intense than natural ones, thanks to the modern chemical pigments and coloring techniques). This model can be extremely helpful for understanding, among other things, the specific ways in which every culture segments the huge range of possible combinations of the three dimensions (more than seven million have been calculated) by privileging one or another aspect. Interestingly, assuming the three-dimensionality of chromatic sensation seems to have allowed diachrony to be restored to the picture. In this connection, Marshall Sahlins made the important point that the linguistic definition of hue tends to proceed by a speed proportionate to the saliency of the color, namely, to its capacity to catch visual attention, so that red, the most salient color, is generally the first to be defined as hue, followed by yellow, green, and, at a greater distance, blue. On the contrary, green and blue, being less salient colors, are usually perceived as brightness before being (slowly) focused as hues (Sahlins 1976).
As far as antiquity is concerned, the Munsell model proved fruitful a few decades ago in the seminal work of Eleanor Irwin, which can be seen as a turning point in the scholarship on the subject. Irwin offered, also through analogies with other cultures, the first plausible explanation of the puzzling fluidity of Greek color vocabulary, tracing the remarkable predilection of Greek poetry for notations of brightness and emotional resonance of colors (think, again, of the warm reflections of the xanthos hair, or the alarming light of the glaukos eye) to the fact that the Greeks saw colors in degrees of light and darkness rather than in terms of hue (Irwin 1974). This claim, which I think may be extended with the proper qualifications to the whole body of Greek and Latin literature, is one of the two assumptions steering my inquiry. My second assumption is taken from Mark Bradley’s recent book Colour and Meaning in Ancient Rome (2009), whose central argument is that Roman perception of colors is not so much influenced by light and luminosity as by ontology, namely, by the notion of color as an inherent property of an object. In this view, color was to the Romans a basic unit of information and understanding of their social world, so the meaning of a color qualification is primarily conditioned by the ethical values assigned to the object it is most often linked to – hence, for instance, the versatility of purpureus, determined by the social and economic features of purple dye and its associations with luxury and decadence, the freshness connoted by viridis, thanks to its connection with vegetation, or the preference for the natural body color over cosmetic embellishments, which is central to the moral debate on distinguishing the true from the fake (Bradley 2009). In short, I assume the visual propensity to luminosity values and the social role of color in the communicative strategies as two parameters that, far from being mutually exclusive, have to be joined in understanding the issue of ancient color aesthetics.
Let us make a fresh start, observing that the notion of color as an inherent property of the object is indeed a universally widespread pre-theoretical notion (Sassi 2009 ), and, not surprisingly, it was shared by the ancient Greeks as well. In fact, the Greek word chroa/chroia means both the colored surface of a thing and the color itself, and is significantly related to chrōs, which means “skin” and “skin color,” with connotations of vigor and youth when applied to the bright Homeric heroes (Carastro 2009a). Chrōs notably occurs with the sense of “color” in Parmenides’ Fr. 8, where “changing place and altering in bright color [chroa phanon ameibein]” are mentioned among the characteristics mortals ascribe to reality, “trusting them to be true” (28 B 8.39–41 Diels-Kranz). One can find in this text, where color first appears in a philosophical writing, two features which would remain constant in ancient reflections, namely, the emphasis on brilliance with its power to catch the eyes, and the suspicion that color may be, or, worse, be exploited as, a deceitful covering of the truth.
The ambiguity of color is shown, for instance, in Empedocles’ Fr. 23, where the mixing by Love and Strife of the four roots that build the sensible world is famously compared to the work painters do when mixing a limited number of different pigments (polychroa pharmaka) in variable proportions, so as to be able to create likenesses of all things, be they trees, men and women, animals, or gods. It has been argued that the term pharmakon hints at a magical procedure, so implying that the human mind can be deceived (apatē, v. 9) as much by the appearance of natural bodies as by painted images, given that in both cases the basic elements of the composition are concealed (Skarsouli 2009). The splendor effect was likely to take part in this process, for the production of all the different colors is due, according to Empedocles’ account, to the mixture of two elements, fire and water, which correspond respectively to white (light) and black (darkness), considered as the two extremes in the chromatic continuum (Ierodiakonou 2005). Incidentally, it is also noteworthy that in all ancient theories white and black are considered to be colors, often primary or among the primary, while we have learnt from Newton that white light arises from the sum of all the hues of the spectrum, whereas black is its absence. In any case, the aspect of deception would emerge again in the parallel Gorgias establishes in the Encomium of Helen between the enchanting power of logos and the strong emotional impact that both the colored compositions of pictures and statues may have (82 B11.17–18 Diels-Kranz) – I wonder whether Gorgias may be thinking of statues too as colored (about which see below).
One can guess that Democritus, while appreciating the pleasure given by beautiful sights, gave instead a rather neutral judgment of it. While we cannot unfortunately figure out what Democritus had to say on painting and perspective in the writings he devoted to those subjects, we are well informed, thanks to Theophrastus’ De sensibus, about the content of his writing in On Colors. We are told that he made the nature of colors dependent on the interaction between daylight and the microphysical structure of objects, which means that he was considering brilliance a factor as important as hue for defining colors. Moreover, in explaining the various colors as mixtures of a basic set of four (white, black, red, and green), or as mixtures of the primary mixtures, he considered the mixture of red and white (corresponding to the golden and copper color) plus a small amount of green to give “the most beautiful color” (to kalliston chrōma). While it is difficult to identify what kinds of objects Democritus was thinking of, he is expressing here an unquestionable aesthetic preference for the attuning of black and white, corresponding to the minimum and the maximum degree of light, the little touch of green adding a sense of freshness and life. Also, significantly, the third primary mixture in the list, the color purple, is said to “appear delightful” (hēdu phainesthai), as it comes from white, black, and red, the presence of white being indicated by its brilliance and luminosity (to lampron kai diaughes; Theophr. De sens. 76–77, in 68 A 135 Diels-Kranz).
The same appreciation of brilliance is found in Plato, whose account of vision in the Timaeus (45b–e) is centered on the interacting of three factors, namely, the fire internal to the observer’s eye, the daylight, and the “flame” (that is, again, the light) transmitted by the colored objects (it is noteworthy that both notions of visual ray and sensible emanation are borrowed from Empedocles, as Meno, 76c–d shows). So color vision depends on the eyes (then the soul) being affected by a luminous complex, whose impression changes according to the quality of external light and the quantity of fire included in the composition of the natural bodies. It is no wonder that the fourth basic color in Plato’s list, which includes white, black, and red, is the “brilliant” and “shining” (lampron te kai stilbon, Tim. 68a), which is not a color at all to us – the people of Sacks’s island excepted.
Plato also speculates that the purer (more saturated, in our terms) and more brilliant the colors, the purer the pleasures they give to our senses (purity being determined in both cases by unmixing, respectively with other colors or with pains: Philebus 51a–53c, 59c), yet he assumes that perfect colors (like pleasures) are not within reach of ordinary experience, as he makes clear in the Phaedo, vividly describing the “real earth,” situated in the pure heaven, with such colors as purple, golden, white, which are there “brighter and purer” (lamproterōn kai katharōterōn), and “more beautiful” than those we see (and the painters employ) in the hollows where we live, amidst water, mist, and air (109b–110d, esp. 110c; the descriptions of heaven and the spindle of Necessity in the Republic, 616b–617b, and of the walls of ancient Atlantis in the Critias, 116a–c, should be mentioned as well in this connection). Plato’s distrust of worldly colors extends, of course, to those used in painting, which are “samples” or “illustrations” (deigmata) of the heavenly ones (Phaed. 110b) – the allusion to painters working here as a metaphor for the phenomenal world, as in Empedocles’ Fr. 23, yet with a more disparaging tone, due to Plato’s general emphasis on the dangerous power art has of stirring people’s emotions, and of seducing them into a false vision of reality (Keuls 1978; Baj 2009).
A direct thread joins Plotinus, “the most important thinker on light and colour in late Antiquity” (Gage 1993, 26), to Plato’s reflections on light and color. Plotinus identifies Beauty with the One, which is often described as Light (Enneads V 3.12, 17), and thinks that the One’s light is held within the Intellect, where it is fragmented into multiple unities (the Forms). Differently from Plato and most ancient theorists, Plotinus considers light to be an incorporeal power, and colors to be modes or reflections of the intelligible light fallen into the darkness of matter (Enn. II 4.5, IV 4.29, IV 5.7). On the other hand, he follows the Timaeus model in stressing the role of light within the eye in color vision (Enn. V 5.7), and the Philebus in expressing his preference for “simple” colors, where light shines most brightly. While Plotinus’ personal top list includes in fact lights rather than colors (namely, gold, lightning, fire, and the stars: Enn. I 6.3), his view clearly develops Plato’s emphasis on brilliance as an aspect of color. It may be more surprising to find a similar emphasis, in a quite different metaphysical frame, in Aristotle.
Aristotle shares indeed Plato’s predilection for “pure” and brilliant colors. In the chapter he devotes to color in the De sensu, he first argues that the various colors arise from different proportions in the mixtures of white and black, these corresponding to fire and water in the bodies and determining the transparent medium as light and darkness respectively; then he specifies that purple (halourgon) and red (phoinikoun), being produced by ratios which are “uncomplicated,” namely, easily translatable into numbers, are “the most attractive [ta hēdista]” (Arist. De sens. 3.439b 32–440a 6). It is of major interest to us that Aristotle attempts here to apply to colors (and next, ch. 4, to flavors) the mathematical treatment which had been successfully used for music: a path Newton too followed in adding indigo to the set of colors more visible in the spectrum, so that they would be seven just like the musical notes.
It is not too daring a guess that Aristotle (not unlike Newton perhaps) was driven to this approach, beyond his willingness to apply mathematics to natural science (Sorabji 1972), by some acquaintance with synesthesia, namely, the sensory (and particularly emotion-loaded) experience of two or more senses being simultaneously activated and intertwining, so that, for example, one may hear a sound and mentally see a color or a shape; or, most commonly, one sense is expressed in terms of another (e.g. by defining a sound “bright”). Aristotle was indeed aware that different senses cooperate in perceiving a given object as a whole complex of qualities, an essential cognitive activity he traced back to a specific psychological faculty he named a “sensation in common” (aisthēsis koinē, cf. Arist. De anima III 2, 426b 8–427a 14, De sens. 7, 449a 14–20). A more direct clue may be offered, however, by the fact that linguistic expressions of a sort of “synesthetic consciousness,” close as they are to metaphors, are quite frequent in Greek literature as in many others. For instance, the adjective leirioeis/leirios, “lily-like,” can describe both a visual quality (whiteness) and a tactile one (softness), recalling the flower, but also a clear voice, like that of the Muses in Hesiod’s Theogony 41, since “clearness” is in its turn a synesthetic notion appealing to the senses of sight and touch. Let us note here that “les fleurs sont le lieu par excellence de la synesthesia,” as Ribeyrol (2009, 57) puts it, and nicely emerges in the pleading we read in Plutarch (Quaestiones Convivales 646B–C) for the use of garlands in the symposia, because of the natural, and thus decent, pleasure colors and scents of flowers allow, unlike the artificial fragrances and dyes coming from the Orient. Incidentally, Plutarch’s interest in aesthetical psychology is in itself remarkable (Svoboda 1934). One should recall at least, regarding color, his observation (anticipated, however, in the pseudo-Aristotelian Problemata, XXXI 19) that eyes irritated by “painful” colors may take rest “by mixing shadow to them” or by looking at green and pleasant things (De exilio 599F–600A, De tranquillitate animi 469A) – remember also that in antiquity green used to be endowed with positive, even curative properties, not least owing to its association to thriving vegetation (Trinquier 2002).
A phenomenon I would like to mention next to synesthesia is the strict relation the sensations of brightness and of movement display in the visual process. A number of Greek words convey, indeed, a combined sense of brightness and movement: for instance, aiolos (which can describe both a “shining” surface and a “flashing” movement), argos (which can be used of bird feathers or of lightning), amarugma/amarugē (Irwin 1974, 205–216). This is also one among the reasons for the high appreciation of purple in antiquity. The very name of porphura is related to porphurō, meaning “swirling” besides “growing/dying purple,” and recalling the technology of the production of purple in the ancient world: more dyes from yellow to scarlet red to blue-violet were obtained from the juice secreted by many species of murex, by mixing them with each other and stopping the photochemical exposure (or the boiling) at certain points (this also explains why the dye was considered particularly valuable, especially in the varieties of red and purple, as they were obtained with undiluted juice, of which one mollusk provided just a few drops). What is more, the several hue changes occurring in the dyeing process would determine the final effect of purple, which was seen shimmering under the light rays, with a side effect of movement. This effect is well caught in the lively analogy with dyeing that Aristotle introduces in the scientific discussion on color contrast in the rainbow (Meteorologica III 4.375a 22–28; tr. H.D.P. Lee):
The same effect can also be seen in dyes: for there is an indescribable difference in the appearance of the colours in woven and embroidered materials when they are differently arranged; for instance, purple is quite different on a white or a black background, and variations of light can make a similar difference. So embroiderers say they often make mistakes in their colours when they work by lamplight, picking out one colour in mistake for another.
Yet the most brilliant literary report on the purple effect, which deserves to be quoted at some length, is likely found in Philostratus the Elder, describing a painting scene with hunters, whose leader catches the eye by his rushing in golden-and-purple guise (Imagines I 28.4 tr. A. Fairbanks, modified in a few points):
… and [his horse, white and black-headed] has golden trappings, and a bridle of Median scarlet; for this colour flashes on the gold as fired stones would. The youth’s garment is a chlamys bellying out in the wind; the colour is the sea-purple which the Phoenicians love, and it should be prized above other purple-dyes; for though it seems to be dark it gains a peculiar beauty from the sun and is infused with the brilliancy of the sun’s warmth. And from shame of exposing himself unclad to those about him he wears a sleeved chiton of purple […] He smiles, and his eye flashes, and he wears his hair long, but not long enough to shade his eyes when the wind shall throw it into disorder.
In ancient literary descriptions, some things stand out for being versicolored on their own. This is the case, for instance, for the sea, whose color is changing par excellence, which explains, among other things, why it is often strikingly described, starting with Homer, with such different terms as porphyreos, ioeides, glaukos, polios, oinops, melas, kuaneos (Ferrini 1999–2000; Christol 2002). The Greek adjectives glaukos and kuanous and the Latin caeruleus cannot be directly translated into our category of “blue,” not only because this color is slowly focused as hue because of its poor salience, as explained above, but also because of their prototypical association to the sea, whose unstable appearance and fearful resonances they tend to evoke even when qualifying other objects (Stewart 2006). Interestingly, this situation has such different outcomes as, in poetry, the glamorous description of the sea by that great admirer of the spectacula of nature, Lucretius (De rerum natura II 763–775), and, in the philosophical debate on the reliability of perception, the use of sea color as a proof of the fallacy of vision (cf. Cicero, Academica Priora II 105, commented on by Romano 2003). Another case is that of the colors of feathers in the neck of a dove or the tail of a peacock, whose changing according to the different angles from which light is reflected is a frequent spur to such vivid descriptions as that supporting Lucretius’ emphasis on the role of light rays in the perception of color, influenced by the atomistic paradigm (De rerum nat. II 795–809, see also IV 72–89), or that enhancing Lucian’s ekphrasis in the De domo 11.
Since human beings pay great attention to their own complexions, a key aspect of how they look to others, cosmetic treatments have always focused on the color of the face, and the very word chrōs, meaning both “color” and “skin,” can remind us that Greeks are no exception. In fact, the contrast between the pale complexion of women, due to their living in the darkness of the domestic sphere, and the highly colored skin typical of men, tanned and strengthened by physical exertion and outdoor sports, is an essential part of the ancient social hierarchy of values, which derives its force from being presented as a fact of nature (Sassi 2001, 1–11). This is also the reason why any artificial means aimed at correcting the natural/social code, being artificial, was to be rejected, and the application of cosmetics was equated with concealment and deceit, the exact opposite of the natural beauty associated with moral well-being. An honest bourgeois such as Xenophon makes very clear on two occasions that makeup, and also perfumes and luxury fabrics, are unworthy of a free woman. In the episode of Heracles at the crossroads, for instance, the woman who symbolizes virtue is “beautiful and of noble mien, her body adorned with purity, with downcast eyes, modest expression, dressed in white,” while her counterpart (vice) is “fat and soft, and so made up as to seem unnaturally red and with high-heeled shoes so as to appear taller than nature had made her” (Memorabilia II 1.22). Taking a more concrete stance, the rich landowner Ischomachus, who is Socrates’ interlocutor in Xenophon’s Oeconomicus, does not tolerate seeing his wife “much enamelled with white lead, no doubt to enhance the natural whiteness of her skin … rouged with alkanet profusely, doubtless to give more colour to her cheeks than truth would warrant.” He dissuades her from this practice by appealing to an ideal of authenticity: if she wants to be “truly beautiful,” not just “apparently” so, she should not sit about like a slave, but be active and oversee the running of the household. This exercise will whet her appetite, she will feel better and her complexion will be healthy, unlike those women who do nothing but sit around applying makeup and making trouble (Xenoph. Oeconomicus VII 21ff., 33). Even more famous is the passage in the Gorgias (465b) where Plato describes cosmetics as deceitful, wicked, despicable, “unworthy of a free man,” as they produce an “unnatural [or alien: allotrion] beauty,” the exact opposite of the natural beauty of a body shaped by gymnastic exercise. In remarks on cosmetics focused specifically on men, Plato also deplores the dissimulation that the effeminate man seeks to achieve. By neglecting physical effort and masculine exertion, he ends up with a pallid complexion, contrary to men’s “natural beauty,” which he attempts to cover up with “alien” hues (Phaedrus 239c–d). In short, whether they are women or effeminate men, those who titivate themselves do so in order to be seductive, and the more blatant their makeup, the more shameless their behavior.
Even outside the “anti-cosmetic” trend (whose interesting outcomes in Roman moral thought have been studied in depth by Bradley 2009 ), the view that colors found in nature are “more beautiful” than artificial ones is common, and it supports the notion that painters must aim at imitating the natural aspect of things (not, however, to the point that the divide between reality and representation be canceled, owing to the notorious ambivalence of the ancient concept of mimēsis). Hence the frequent insistence on painters’ need to apply the most appropriate colors, mainly when portraying human countenance, since one’s ēthos and emotions show through his or her complexion and the eyes (Plato, Cratylus 424d–e, 432b–c; Xenophon, Memorabilia III 10.1–8; [Aristoteles], De mundo 396b 11–15; Aristides, Quintilianus, De musica III, VIII, p. 15f.; Plutarchus, Cimon II 3, Alexander I 3, De Alexandri fortuna 335B, Quaest. conv. I 2. 618A–B; Philostratus, Imagines I 2). Plato’s remark in the Republic (420c–d) that a statue’s most beautiful parts (e.g. eyes), have to be painted not “the most beautiful colors” (e.g. purple), but “the most appropriate ones” (e.g. black) does confirm this realistic approach, in terms that Lucian would resume in one of his writings of art criticism (Imagines 6–8). The same assumption underlies a penetrating thought ascribed to Sophocles in a lively episode Athenaeus tells, whose source is claimed to be Ion of Chios (Deipnosophistae 603F–604A). Sophocles, excited during a symposium by the sight of a handsome cupbearer blushing from modesty, quotes a Phrynichus verse praising one’s “purple cheeks, where eros shines” (fr. 13 Nauck). As a pedantic schoolteacher objects that the cheeks of the boy would not appear that beautiful if painted purple in a portrait, Sophocles replies just reminding him of such further poetical expressions as the “purple lips” of a Simonides’ girl (fr. 585 Page), the epithet of Apollo “golden-haired” (a painter should paint the god’s hair black to get a better result), and finally the Homeric image of “rosy-fingered” Eos, which does certainly not imply that real people having such fingers would be appreciated, as these are more appropriate to a dyer than to a beautiful woman. We understand that the painters’ use of colors is bound by a principle of “literal” representation, while poets can use every available metaphor. Painters had, however, their proper means and skills. Noteworthy is the case of a bronze statue of Jocasta that Plutarch mentions, in discussing the issue of how we may take pleasure in painful representations: the artist added silver to her face to give the appearance of a dying person, giving pleasure to the observers and eliciting their admiration (Plut. Quaest. Conv. V 1, 674A; see also Philostr. Imag. I 2, giving credit to painters for the possibility they have, unlike sculptors, to render a person’s ēthos through coloring, and thus giving a particular light to, his or her eyes).
Color was furthermore perceived, both in the animated world and in painting, as an indicator of vitality and vigor. Let us consider Aristotle’s account of the embryo’s development in the De generatione animalium (II 6.743b 18–25, tr. A.L. Peck, with modifications):
The upper portion of the body is the first to be marked off in the course of the embryo’s formation; the lower portion receives its growth as time goes on. (This applies to the blooded animals.) In the early stages the parts are all traced out in outline [tais perigraphais]; later on they get their various colours [ta chrōmata] and softnesses and hardnesses, quite as if a painter were at work on them, the painter being nature. Painters, as we know, first of all outline [hupograpsantes tais grammais] the figure of the animal and after that go on to apply [enaleiphousi)] the colors.
In a few further passages of his biological writings, Aristotle interestingly compares the blood vessels, to which flesh is applied as to a framework (peri hupographēn), to the lines of the wooden skeletons (kanaboi) used in modeling (Gen. anim. II 6.743a 1–2, IV 1.764b 30–32; Hist. anim. III 5.515a34–b 7). These analogies are aimed at explaining how nature works, in accordance with Aristotle’s idea that the teleological processes of human technē imitate the profound teleology present in nature, at the same time making it more visible. What is more visible in painting, so as to be used for explaining the embryo’s growth, is how the pair line–color works: line provides first the essential features of image, then comes color to add “flesh” and the beauty of life. Nobody could deny that color has this advantage over line. Plato himself had to admit that pictures of living beings need colors to give them that clearness (enargeia) that the mere outline (perigraphē) is unable to do (Politicus 277c). In fact, as Plutarch would sum up, color owes both its major power of stirring emotions (kinetikōteron esti) and its dangerous deceptiveness (apatēlon) to its being lifelike (andreikelon: De audiendis poetis, 16B).
In fact, which element of the couple is to be judged “better” is an eternal issue in the history of art criticism. Aristotle himself, no less “formalist” than Plato in his aesthetic preferences (Porter 2010, 70–120), famously favors outline pictures in black and white (leukographia) for their giving more pleasure than images smeared with colors (Poetics 6, 1450a 38–B 2). Moreover, how drawing must have been central to ancient artistic practice is shown, in addition to a few scanty, yet significant, archaeological remains, by the focusing of Roman art discourse on the functional role of line and the plastic properties of color as two opposite virtues, respectively championed by Zeuxis (or Apollodoros of Athens) and Parrhasios (Quintilianus, Institutio oratoria XII 10, 4–5; Pliny the Elder, Naturalis Historia XXXV 60, 62, 65, 67–72; Plutarchus, De gloria Atheniensium 2, 346A; cf. Settis 2008). Incidentally, it is likely that within the tradition praising outline and drawing the view took form that earlier painters were to be praised for their using simple mixtures of just four colors (white, yellow, red, and black), after which the palette became richer (Cic. Brutus XVIII 70; Plin. Nat. Hist. XXXV 50). This must be a cultural construction, which does not correspond to what we know about the availability of pigments for the entire range of colors since early times: remember, among other things, the use of blue glass paste on the walls of the Mycenaean palaces (for example, in the famous frieze of Tiryns), or the blue and green traces left on the Korai of the Acropolis. An ideal development in the use of colors, going from Archaic simplicity to later luxuriousness, also underlies the distinction of colors austeri and floridi we find in Pliny, this reflecting, however, a real event: plenty of new pigments coming from the Orient invaded the Classical world toward the end of the fourth century, and, notorious as these were for their being both expensive and splendid, they became a favored target of the moralists. The discussion on colores floridi, however, cannot but be seen as a new expression of the old tradition of disparaging artificial colors in favor of the natural ones.
A different attitude emerges instead from a number of ancient descriptions of the aesthetic effect produced by coloration of statues, pervaded by unqualified emphasis on the brightening and enlivening properties of color. For instance, Euripides’ Helen, in blaming her life and beauty, focuses on the wish her colors were canceled as from a statue, so as to lose her charm (Helen 260–263).1 On this subject the literary evidence has recently received striking confirmation from important archaeological reconstructions of ancient sculpture polychromy, tangibly displaying at last that the effect sought, by applying the most brilliant colors, was exactly one of splendor, along with energy, movement, and life (Brinkmann and Wünsche 2004; Brinkmann, Brinkmann, Piening, and Primavesi 2011).
Let us pass now from living beings to the sky. When approaching atmospheric colors in the Meteorologica (III 2–4), Aristotle assigns to the intermediate air a coloring action, which causes the sun to be seen as red through a dark environment (374b 10–11, cf. 342b 4, 347a 4). This is not the only notion which, while being of minor interest in the psychological writings (De anima, De sensu), plays a primary role in the optical theory of the meteorological treatise. Aristotle even introduces here the notions of visual ray (opsis) and of light reflection (anaklasis), which are not at all present in the psychological writings, in order to explain such heavenly phenomena as rainbow, halos, mock suns, and rods appearing beside the sun. Appealing to the subject’s vision weakening as distance increases, for example, helps to explain that objects appear darker the farther away they are, because our sight (opsis) fails along the way (374b 10–35). Even the colors of the rainbow are determined by the diminishing of our vision according to the angles at which sunlight is reflected through in the atmospheric medium: “When the sight is strong enough the color changes to red [phoinikoun], next to green [prasinon], and when it is even weaker to purple [halourgon].” Aristotle adds that the three colors of the rainbow (notably overlapping the triad of porphureon phoinikeon chlōron found in Xenophanes’ Fr. 33) are “almost the only ones” painters cannot produce by a mixture (372a 6–11). One may remember that Aristotle claims red and purple to be the “most attractive” colors in the De sensu (3.439b 33–440a 1), yet it is even more interesting that in his view red, purple, and green (with dark blue, kuanoun) are primary mixtures of white and black, the remaining colors being mixtures of the primary ones (De sensu 4.442a 20–25). It seems that Aristotle does not know, or intends to ignore, what we have been used to consider the painters’ primary colors, namely the pigments yellow, blue, and red, from whose mixing on the palette all the other colors can be produced. This may be explained through the fact that Aristotle, like the Peripatetic author of the De coloribus (2.792b 16–32), is more interested in lights, and thus in an additive synthesis, rather than in hues, whose synthesis is subtractive (of light), and this prevents him from any experimental approach to the color problem (a similar motive may explain both Democritus’ and Plato’s approach: Struycken 2003 ). However, it is time to ask whether the palette was a regular tool of ancient painters after all.
In the De sensu (3.439b 18–440b 25) Aristotle sets out three “ways” (see tropos at 440a 6) in which colors other than white and black arise, namely, (i) through juxtaposition of small particles of white and black, (ii) through superposition upon each other (epipolasis), (iii) through chemical mixture. These situations should not be taken, pace Sorabji, as mutually exclusive “theories.” While arguing that only the third kind of combination produces a genuine mixture, Aristotle does not deny that the first and the second ones can actually occur and produce the impressions he actually describes. Painters, for instance, are accustomed to overlay (epaleiphousin) a given color over a clearer one (enargesteran) when they want a thing to appear through water or air, “just like” the sun, while being leukos, appears red through fog and smoke (440a 7–12). Aristotle does not specify how colors were actually mixed in the third kind of process (the fusion), and we are not committed to thinking that the mixing was done on a palette rather than directly on the painting support – there is no mention of something like a palette in either Greek or Latin sources. In any case, Aristotle’s dwelling over the superposition method may indicate that overlaying of paints was the favorite technique, as various tonalities could be produced and, what is more, light and shade effects could be achieved by superimposing a lightener (white) or a darkener paint (dark blue or black) over another (Bruno 1977; Kakoulli 2009). We can guess that the effect that was aimed at was a sort of chiaroscuro, a concept possibly corresponding to Greek skiagraphia. Skiagraphia is of course blamed by Plato for its producing a deceptive illusion of reality, and Plato says that it is more suitable for representing landscapes than for human figures, owing to its lesser precision (akribeia: Plat. Criti.107c–d; cf. Dionysius of Halicarnassus Isaeus 4). However, painters were not inhibited by this kind of criticism in their work: they knew, not unlike the philosophers, the power color has to give volume and life to the shapes in the painting, yet exploiting that power was their job. The Vergina frescoes on the one hand, and, on the other hand, the comments of several ancient art observers and critics (Plutarch, De Herodoti malignitate 863E, Tranquill. an. 473F; Lucianus Zeuxis 22.5; Longinus De sublime XVII 3) are there to prove that chiaroscuro prevailed.
Focusing at the start on ancient color experience, understood at the primary level of perception and emotional response, may indeed have been a rewarding option. Questions about the Greeks’ and Romans’ color preferences, the ancient scientific accounts of chromatic vision, and the explicit aesthetic evaluations of colored natural things or artifacts, cannot but be interrelated. After all, color is an essential stuff that life and art are made of.
While the most balanced account of the Greek color terminology remains Irwin (1974), Bradley (2009) provides an up-to-date discussion on the whole issue of ancient color categorization, as well as a comprehensive treatment, historically and anthropologically oriented, of the meaning of color in Roman culture. Since an investigation of color in antiquity requires the analysis of both texts and artifacts by ranging across various fields and methods (those of classical philology, archaeology, history of science, linguistics, and anthropology), the most stimulating approach to the topic is provided by a few collections of articles which make room for the contributions of diverse specialists: Beta and Sassi (2003), Carastro (2009b), and Villard (2002). As particularly regards the field of ancient art history, the discovery of the Vergina frescoes in the last decades of the past century, on the one hand, and, on the other hand, the increasing attention to the polychromy of ancient statues and architectural elements (along with some brilliant archaeological reconstructions) have triggered intensive study of Greek painting pigments and techniques, the most notable contributions being the volumes edited by, respectively, Rouveret, Dubel, and Naas (2006) and Brinkmann and Wünsche (2004), as well as Brinkmann, Primavesi, and Hollein (2010).