Notes on Ballad Origins. 1 55
markedly distinct classes." In part I agree with Professor Child. " Ballads are the work of individual persons," but where collaboration occurs, consciously, or in course of oral tradition, the quota of each individual must now be matter of conjecture ; and reciters filled gaps in memory with fragments of other ballads, or improved to their taste out of their own invention, as Monsieur de Puymaigre remarked some forty years ago. Thus no traditional ver- sion, as it stands, is likely to be the work of one individual. Some might improve on, others might spoil, the original piece from which the original author himself might fre- quently vary, adding or abridging. As to " markedly distinct classes," these, of course, did exist in the Border) but not much among the " simple," the " folk," the general populace.
If I understand Mr. Henderson's own theory (compare his chapter on Ballads in his Vernacular Literature of Scotland) he thinks that most, or many of our traditional romantic narrative ballads are versions of mediaeval literary romances modified and degraded by vulgar versifiers and by oral tradition. That some ballads are in this case I entirely believe. For example (as I have elsewhere said) Sir Aldingar may be derived from a literary source in William of Malmesbury (ob. 1143), a version also found in a French metrical Life of Edward the Confessor. But, as to William of Malmesbury, Professor Child says " we cannot well doubt that he is citing a ballad — a ballad is known to have been made on a similar and equally fabulous adventure," and he traces such fables to the middle of the seventh century. Professor Child says again (i., 98) : '^ The idea of the love-animated plants has been supposed to be derived from the romance of Tristram^ agreeably to a general principle, somewhat hastily assumed, that when romances and popular ballads have anything in common, priority belongs to the romances."
Every folklorist knows that many such inventions are
