Here the idea is not that silver is of the same colour as skin, nor gold as blood; but that the relation of colour between silver and gold may be compared with that between skin and blood: the skin throws the blood into relief, as a ground of silver would throw out a projection of gold. In license of this kind we can always trace both a rule and an aim. The rule is relaxed only for the particular occasion. The effect produced is that of tenderness, dignity, and purity. Had Shakespeare been describing the horrible carnage of a battlefield, he probably would have spoken of black or foul gore instead of using a brightening figure.
Now this purpose is not traceable in Homer’s use of certain words, if we are required to treat them as adjectives of colour. There is no Poet, whose rationale is commonly more accessible; but these cases, upon such a principle, do not admit of a rationale at all.
Take for instance his use of the rainbow. It is (1) πορφυρέη, and (2) like a δράκων, which is κυάνεος. Of these, the first may be construed dark with a hue of crimson; the second, dark with a hue of deep blue or indigo. Surely we have here, viewing it as a whole, a most inadequate treatment of the colours of the rainbow. Shakespeare indeed says[852],
His crest, that prouder than blue Iris bends;
and again, in the Tempest, Ceres addresses Iris thus[853];
And with each end of thy blue bow dost crown
My bosky acres....
But (1) blue differs from πορφύρεος, which is essentially dark, and is not blue. (2) Blue, taken largely, represents three of the seven prismatic colours: i. e. indigo and purple along with itself. (3) In the last quoted passage, Iris is also called ‘many-coloured messenger,’ and with ‘saffron wings.’ How different an effect do these words give, as they form a whole, from that of the simile in Il. xvii. In what manner then are we to understand Homer? I answer, in the way of metaphor; and with reference to light and dark, not to prismatic colour. The δράκοντες on the buckler and belt are dark and terrible: so is the storm of which Iris is the type, and it is in viewing the rainbow as a type of what is awful, that we are to find the reason of Homer’s simply treating it as dark, and not as a series and system of colours. Perhaps we ought not to overlook the possibility that Homer may also mean to compare the shifting hues of the serpent with the varied appearance of the rainbow.
Again, let us take his use of μελαγχροίης. Now the question is, did Homer mean by this simply to express darkness, that is to say was dark his idea of μέλας, or did he, with the specific idea of black in his mind, use the term which denoted it poetically for the olive complexion of Ulysses? Surely the former: for the latter use of it would have been bad. It would have been straining the figure in the wrong direction. For blackness would be a fitting trope only where the object was to describe something awful or repulsive.
But beauty of form in Homer always leans to light hues and not to dark ones, whence the Greeks are ξανθοὶ, and the Trojan Hector, though beautiful, is κυάνεος only. Therefore it was not Homer’s object to give an enhanced idea of darkness in the tints of Ulysses. And yet, if μέλας for him meant specifically black, then μελαγχροίης was the height of exaggeration in the wrong sense. But if by μέλας he only understood dark, that was a fair description of the olive tint, as compared with the withered and shrivelled skin of old age.