In this way each Japanese word came to have two pronunciations: first, its own original sound for colloquial purposes; and second, its borrowed sound for purposes of writing. At the outset the spoken and the written languages were doubtless kept tolerably distinct. But by degrees, as respect for Chinese literature developed, it became a learned accomplishment to pronounce Japanese words after the Chinese manner, and the habit ultimately acquired such a vogue that the language of men--who wrote and spoke ideographically--grew to be different from the language of women--who wrote and spoke phonetically. When Hiyeda no Are was required to memorize the annals and traditions collected and revised at the Imperial Court, the language in which he committed them to heart was pure Japanese, and in that language he dictated them, twenty-nine years later, to the scribe Yasumaro. The latter, in setting down the products of Are's memory, wrote for the most part phonetically; but sometimes, finding that method too cumbersome, he had recourse to the ideographic language, with which he was familiar. At all events, adding nothing nor taking away anything, he produced a truthful record of the myths, traditions, and salient historical incidents credited by the Japanese of the seventh century. It may well be supposed, nevertheless, that Are's memory, however tenacious, failed in many respects, and that his historical details were comparatively meagre. An altogether different spirit presided at the work subsequently undertaken by this same Yasumaro, when, in conjunction with other scholars, he was required to collate the historical materials obtained abundantly from various sources since the vandalism of the Soga nobles. The prime object of these collaborators was to produce a Japanese history worthy to stand side by side with the classic models of China. Therefore, they used the Chinese language almost entirely, the chief exception being in the case of the old poems, a great number of which appear in the Records and the Chronicles alike. The actual words of these poems had to be preserved as well as the metre, and therefore it was necessary to indite them phonetically. For the rest, the Nihon Shoki, which resulted from the labours of these annalists and literati, was so Chinese that its authors did not hesitate to draw largely upon the cosmogonic myths of the Middle Kingdom, and to put into the mouths of Japanese monarchs, or into their decrees, quotations from Chinese literature. "As a repertory of ancient Japanese myth and legend there is little to choose between the Records and the Chronicles. The former is, on the whole, the fuller of the two, and contains legends which the latter passes over in silence; but the Chronicles, as we now have them, are enriched by variants of the early myths, the value of which, for purposes of comparison, is recognized by scientific inquirers. But there can be no comparison between the two works when viewed as history.