From Spell to Prayer. 135
as such, is that the epic vein decidedly predominates therein. The glowing periods in which the history of "the great transition" is recounted are not easily translated into the cold prose of science. Construed literally they appear liable to not a few serious strictures. For example, pure ratiocination seems to be credited with an effectiveness without a parallel in early culture. Almost as well say that when man found he could not make big enough bags with the throwing-stick, he sat down and excogitated the bow-and-arrow. Or again "unseen beings" seem to be introduced as " mysterious powers " sprung fully-armed from the brain of man, and otherwise without assigned pre-history.- Finally, magic and religion appear to be treated as in their inmost psychologic nature disparate and unsympathetic forces, oil and water, which even when brought into juxtaposition are so far from mixing that the observer has no difficulty in distinguishing what is due to the presence of each.^ One's first impression is that a purely analytic method has escaped its own notice in putting on a pseudo-genetic guise, that mere heads of classification have first been invested with an impermeable essence, and then identified with the phases of a historical development which is thereby robbed of all intrinsic con- tinuity. But on second thoughts one sees, I think, that to construe literally here is to construe illiberally. Dr. Frazer, in order to dispose summarily of an interminable question, may be supposed to have resorted to a kind of Platonic myth. A certain priority and a certain absoluteness within its own province had to be vindicated for magic as against religion, if the special problem of the Golden Bough was to be kept free of irrelevancies. This vindication the myth contrives, and the rest is, so to speak, literature. For the rest, in Dr. Frazer's promised work on the early history of religion, he doubtless intends to fill in what are manifest gaps in the present argument. Meanwhile, as ' G. />'.,• L, 7S. ' Cf. G. B.,' i., 33, 45> '-^c.
