'It was the myths, above all, that seemed to defy rational analysis and to give rise to the idea that their makers were rambling around in a kind of mystical fog. Yet closer observation, and the whole tendency of anthropologists to treat tribal peoples with increasing respect, have shown that most of the apparently illogical connections in 'primitive' myths are not really so. Rather, the logical systems involved are different from those standardized in western cultures. Levi-Strauss showed in La Pensee sauvage that many simple societies, far from having no category structure at all, have systems of immense range and complexity. That is fact; yet one has to guard against sentimentality at this point, for it is all too easy to add (as many now do) that these alternative logical structures are 'just as good' as the ones we happen to use. After all, they say, even Aristotelian logic has had to be replaced by different kinds in certain conditions, much as the Euclidean system, which once seemed the essence of logical geometry, is now recognized as too restricted for study of the world at large. But the truth is that for many purposes Aristotelian logic, which has established simple and consistent rules of cause and effect, is greatly superior to alternative systems depending on loose grades of symbolic association. Some aspects of myths can be appreciated more fully by these alternative systems, but there are also elements and qualities to which discursive analysis can properly be applied, at least as a preliminary stage.'
--G.S. Kirk, The Nature of Greek Myths, pp. 42-3