The question of the affinity of the Celtic languages to the other languages of the world is one that has given rise to considerable debate since the epoch that was made in comparative philology by the intro ductiiin of the study of Sanskrit, about the commencement of the 19th century. In preceding centuries Welsh had been often compared with Hebrew. "It is commonly observed," says Llewellyn in his His torical and Critical Reniarks on the British Tongue,' " that the British and the Hebrew are similar languages," on the ground of their being alike in many peculiarities of construction, especially in permutation or the change incident to several letters in the beginning of words, and also in the paucity and confusion of tenses in the conjugation of verbs, and in the binding together in one word of some prepositions and pronouns. This degree of resemblance is certainly not sufficient to place the Welsh and the other Celtic languages in the Semitic family, but in these respects it does resemble the Hebrew and differ from the Greek and Latin, which form a portion of the Indo-European family to which it is now adjudged•to belong. The Indo-European family, or rather tribe, is now so extended that it comprises many languages formerly regarded as entirely disjoined, such as Greek and Persian, Russian and English,—while the Semitic only comprises a very small circle of languages bearing a close resemblance to each other—a circle, in fact, no larger than that of the Celtic family alone.
The history of the controversy is an instructive one. For the first thirty years of the 19th century the Celtic languages were supposed, to quote the words of the Rev. Richard Garnett, "to form a class apart, and to have no connection whatever with the great Indo European stock. This was strongly asserted by Colonel Vans Kennedy, and also maintained, though in more guarded terms, by Bopp, Pott, and Schlegel. The researches of Dr. Prichard in his Eastern Origin of the Celtic Nations,' and of Professor Pictet of Geneva, in his truly able work,' Sur l'Affinitd des langues Celtiques avec he Sanscrit,' may be considered as having settled the question the other way. The demon stration of Pictet is so complete, that the German scholars who had previously denied the connection, now fully admit it, and several of them have written elaborate treatises, showing more affinities between Celtic and Sanscrit than perhaps really exist." The work of Dr. Prichard, himself a Welshman, is entitled The Eastern Origin of the Celtic Nations proved by a comparison of their dialects with the Sanskrit, Greek, Latin, and Teutonic Languages.' The first edition appeared in 1831 ; a second, with extensive and important additions by Dr. It. G. Latham, in 1857. On the subject of the Celtic languages in the British Islands, there are some valuable and really instructive papers by the Rev. It. Garnett, in the Quarterly Review,' and the Transactions of the Philological Society,' which are collected in the posthumous volume of his Philological Essays,' edited in 1859 by his son.
The Welsh language is one of the oldest in Europe : the Irish and Welsh are in fact among spoken languages the most ancient of which any written monuments are preserved, unless we regard the modern as identical with the ancient Greek. The Welsh has poems now in existence, the origin of which is believed by the best critics to date back to the 6th century, to a period little after the time when the Romans left the country, in which of course, while they held it, the dominant language was Latin. It is true that Zeuss has shown that the language of these poems, when originally composed, must have differed in a considerable degree from that of the form in which they are preserved to us, and that one of the most learned Welshmen of a century ago, the Rev. Evan Evans, says in his Speci mens of the Bards,' published iu 1763, that he had shown the most genuine remains of Taliesin to the best Welsh antiquaries and scholars then living, and that " they all confessed they did not understand one half of any of his pieces." It is true also that Price, in his Hanes Cymru,' a book intended for ordinary readers, found it necessary to give a modern Welsh version of an ancient Welsh poem, which ho quoted from Gwalchmai, a bard of the 12th century, who is said to have accompanied Cceur de Lion to the Crusades. Still from about the Norman conquest downwards, there exists a mass of literary matter in a language which is readily intelligible to those who are acquainted with the modern Welsh after a slight degree of study. A similar observation is perhaps applicable to no other living European language except the Icelandic. The language of the Saxon contemporaries of Taliesin has been a dead language for centuries, and even in the time of Gwalchmai the present English was as yet unborn.
The Welsh has long been an object of study to those who speak it. "There are," says Owen Pughe (' Archmologia,' xiv. 220), "about thirty different old treatises on Welsh grammar and prosody preserved. Of these, one is particularly deserving of notice as a curious relic : it was composed by Geraint about 880, revised by Einion about 1200, and again by Edeyrn about the year 1270, and regularly privileged by the different sovereigns who then exercised authority in Wales." This work was first printed by the Welsh Manuscript Society in 1856, under the editorship of the Rev. John Williams ab Ithel. A portion of a grammar which appeared in 1567, from the pen of Griffith Roberts, and of which mention will he made hereafter as remarkable on other accounts, is remarkable also as containing some proposals on the subject of Welsh orthography of an ingenious character, which were not however adopted.