APPENDIX. THE popular legends of Germany, and some other countries, having been collected and published within the present century, a rich harvest of legendary lore lay ready. I had the good fortune to be the first to gather it1; and my Fairy Mythology will therefore probably remain for some time a kind of text-book on the subject. As the present may be said to be a companion to that Work, I venture here to make some additions to it; but I put them in such a form as to be independent. It was only by degrees that I arrived at what I believe to be the true origin of the word Fairy; and my notions of it are scattered through the Fairy Mythology. I will, therefore, now give my perfect theory. There can be no doubt that our word Fairy is the French féerie, which originally signified illusion, and is derived from fée. I therefore reject, 1 A selection of stories from the Kinder- und Hausmärchen of MM. Grimm, with illustrative, critical, and antiquarian notes, appeared in 1825, under the title of German Popular Stories. The translator was, I believe, Mr. Edgar Taylor. There is a wide difference, I must observe, between popular legends, and stories; the former are objects of actual belief, the latter are only regarded as sources of amusement. with full conviction, all the etymons (such as that from Peri) which go on the supposition of fairy being the original name. The Italian fata, Provençal fada, French faé, faée, fée, are, beyond question, the words first used to designate the being whom we call Fairy. Of these words, I regard the Latin fatum as the root. In a coin of Diocletian the Destinies are, I know, named Fata, and this might seem to give a ready origin of the Italian and Provençal names; but there is so little resemblance between the Parcæ and the Fairies of romance, that I cannot adopt it. My opinion is, that, as from the Latin gratus came the Italian verb aggradare, and the French agréer, so from fatum came affatare, fatare, (Ital.) and faer, féer, (Fr.), signifying to enchant; and that fato, fata, faé, faée, fée, are participles of these verbs. I believe there is not a single passage in the old French romances, in which these last words occur, where they may not be taken participially; such are, les chevaliers faés, les dames faées, and the continually recurring phrase elle sembloit (or ressembloit) fée. La fée is, therefore, la femme fée, and une fée is une femme fée. The Italian fata is, in the romantic poems, always employed as a substantive; but it is well known that a number of substantives in all languages are in reality adjectives or participles, and in the Pentamerone fata and fatata are evidently employed as equivalents. I therefore regard fata as nothing more than fatata, contracted after the usual rule of the Italian language1, and esteem una fata to signify merely una donna fatata. I will now show what was understood by une femme fée and una donna fatata. In the romance of Lancelot du Lac we are told, that "all those (women) were called Fays (fées) who had to do with enchantments and charms, and knew the power and the virtue of words, of stones, and of herbs; by which they were kept in youth and in beauty, and in great riches." This definition will, I think, include all the Fées, without exception, that we meet in the old French romances. It will also, I apprehend, apply to the Italian Fata. In the Pentamerone, Fata and Maga are synonymous, and fata is also equivalent to 1 I cannot help suspecting that this rule of the Italian language, (i. e. throwing the syllable at out of participles of verbs in -are; e. g. from adornato making adorno, from guastato, guasto) has been derived from the Latin. I think the following words have the appearance of having lost at or simple a. In the verb poto -are the participle is potus (potatus?); aptus is rather, I think, a contraction of aptatus than the participle of apo, obs. We have paratus and paṛtus, juvatus and jutus, fricatus and frictus, secatus and sectus, inseratus and insertus (Virg. Æn. iii. 152.), necatus and nectus, crematus and cremus, creatus and cretus, truncatus and truncus, pulsatus and pulsus, quassatus and quassus, lavatus lautus and lotus, viduatus and viduus, orbatus and orbus; to which perhaps may be added, cavatus and cavus, nudatus and nudus. Some of these may be explained on other principles, but I think my theory (and I have not met it anywhere else) is not altogether devoid of some show of vraisemblance, |