Welsh Journals

CAMB10 = BR1T0N, March, 1820. NULLI QUIDEM MIHI SATIS ERUDITI VIDENTUR, QUIBUS NOSTRA IGNOTA SUNT. Cicero de Lembns. WELSH LANGUAGE. THE BARDIC LETTERS: JF ROM the elementary analysis of the Welsh Language, to which the last Essay under this head was devoted, a transition to those symbols, which are employed to represent its articulations to the eye, seems easy and natural. " Words/' says Aristotle, " are the marks of thoughts, and letters of words:" and St. Au- gustin, pursuing the same idea, observes, " verba sunt signa au- " dibilia, signa verba visibilia." And nothing can be more certain than that this was the natural progress of language: the oral or audible first, and then that, which was written or visible. The elementary sounds, of which mmh account has already been given, were supplied by Nature herself as the representatives of ideas in a primitive state of society; but, had human intercourse pro¬ ceeded no farther than this, it must of necessity have remained extremely rude and imperfect. The same impulse, therefore, that taught man to convey his thoughts to the ear, instructed him also to embody the representations of those thoughts to the eye *. And this he did by resolving his oral language into its original principles, and by assigning to those principles such distinctive characters, whether imitative or symbolical, as natural attributes or arbitrary accidents presented for his adoption f. " The ana¬ lysis of language into its elementary sounds," says Mr. Astle in * M. De Gebelin has upon this subject the following very accurate re¬ mark : " L'invention de Pecriture, ainsi que celle de tous les arts, fut de la " plus grande simplicite. On vouloit peindre vine idee; mais cette idee " peignoit un objet. Un n'eut done qu'a peindre eetobjet, qu'a en tracer " la figure, et l'idee fut peinte : ainsi on ecrivoit par la ineme moyen qu'on " parloit. La nature en fit tous les frais."—Monde Primitif, torn, iii^ p. 379. t See Harris's Hermes, p. 331. VOL. r. 2 I .