We may learn what qualities he considered necessary for an editor in this field, from the latter part of his _Remarks on Popular Poetry_, in which he discusses previous attempts to collect English and Scottish ballads. Of Percy he speaks in the highest terms, here and elsewhere. We have seen that he felt a strong sympathy with Percy's desire to dress up the ballads and make them as attractive to the public as their intrinsic charms render them to their friends. He did not of course realize the extent to which the Bishop reworked his materials, as the publication of the folio manuscript has since revealed it, and Ritson's captious remarks on the subject were naturally discounted on the score of their ill-temper. But it is not to be doubted that Ritson had an appreciable effect on Scott's attitude, by stirring him up to some comprehension of the things that might be said in favor even of dull accuracy. Ritson's collections are cited in their place, with a tribute to the extreme fidelity of their editor. It is a pity that this accurate scholar could not have had a sufficient amount of literary taste, to say nothing of good manners, to inspire others with a fuller trust in his method. Scott expresses impatience with him for seeming to prefer the less effective text in many instances, "as if a poem was not more likely to be deteriorated than improved by passing through the mouths of many reciters."[47] He admitted, however, that it was not in his own period necessary to rework the ballads as much as Bishop Percy had done, since the _Reliques_ had already created an audience for popular poetry. His purpose evidently was to steer a middle course between such graceful but sophisticated versions as were given in the _Reliques_, and the exact transcript of everything to be gathered from tradition, whether interesting or not, that was attempted by Ritson. In his later revisions he gave way more than at first to his natural impulse in favor of the added graces which he could supply.[48] It is easy to see how his own contributions of word and phrase might slip in, since his avowed method was to collate the different texts secured from manuscripts or recitation or both, and so to give what to his mind was the worthiest version. Believing that the ballads had been composed by men not unlike himself, he assumed, in the manner well known to classical text-critics, that his familiarity with the conditions of the ancient social order gave him some license for changing here and there a word or a line. In determining which stanzas or lines to choose, when choice was possible, he was guided by his antiquarian knowledge and by the general principle of selecting the most poetic rendering among those at his command. This was his way of showing his respect for the minstrel bards of whom he was fond of considering himself a successor.