Billeder på siden



1. Degli avvertimenti della lingua sopra'l Decamerone. Venice, 1584. Vol. i, p. 261.

2. “Usually” I say, because in Italian as well as in Sassarese, a single consonant is pronounced as if written double, as it falls under the following general rules :

a. If, being initial and not followed by a consonant, it stands at the beginning of a sentence, whether commencing a period or clause (long or short) or following a comma.

b. If the preceding word, though ending in a vowel, be an oxytone, or a monosyllable derived from a Latin word which has dropped its final consonant, or final syllable beginning with

sonant, in becoming Italian or Sassarese. Thus the preposition a', derived from the Latin ad, the conjunction e, corresponding to et, sì derived from sic, “ne” nec, and truncated words like "amò" amavit, “potè” potuit, have all the property of giving a strong sound to the initial consonant of the word following; and though one sees written a Pietro, e voi, grande, questo quello, amò molto potè poco, one always hears appietro, evvoi, siggrande, necquesto necquello, amommolto, poteppoco,

The weak sound of the consonants, on the other hand, will obtain in every case, other than those noted in the above rules, in which the preceding word ends in a vowel. Thus in each of the following examples:- di Maria, i doni, la mente, le donne, mi dice, ti lascia, si gode ama molto, pote' poco, molto largo, the initial consonant of the second word is pronounced as written, weak; for either the Latin form of the preceding word (de, illi, illa, illæ, me, te, se, potui) ends in a vowel, or else, as ama and molto in ama molto and molto largo, the preceding word has not the tonic accent on its last syllable.

The property which many oxytones and monosyllables possess of giving a strong sound to succeeding initial consonants, does not then depend, as Salviati would have it, on their oxytonic or monosyllabic nature, but as I think I have sufficiently shown, on the final consonant of the original Latin form. This Latin final consonant, though it has disappeared in the derived dialects, retains its effect through the process

1 In Sassarese this preposition gives the strong sound to the l of the article only when the following word commences with a vowel:-a lu babbu, all'anima, all'ilcribi, a la peddra.



called assimilation, by virtue of which it is transformed into an Italian or Sassarese initial.

This being admitted, the oxytones and monosyllables which do not enjoy the property of giving a strong sound to initial consonants, and which are cited by Salviati as exceptions, cease to be such, and fall in with my general rules.

3. The word “altru' altro, is an exception, being pronounced with ? Italian.


4. Except in "eternu” eterno,

"eterniddai" eternità, "ternu” terno, “incarnaddu” incarnato, “incarnazioni" incarnazione, “ turnu” (the “turning-box” of a monks' parlour) and some others.

5. Except in "forsi" forse (also pronounced vulgarly fossi), “cumparsu” comparso, and some others.

6. Except in Israeli.

7. Except in those words in which i receives the tonic accent. In these the trigram is resolved into the digrain sc, and the vowel i, which last is given its proper value; as in “pascia" pasceva. The same may be said of any other trigram into which i enters as its third element, the Sassarese sgi and the Sassarese and Italian gli for instance. And just as the trigrams are resolved into digrams and vowels by reason of the tonic accent falling on the latter, the digrams themselves, such as ci and gi, are, under similar circumstances, split up into simple characters followed by a fully sounded vowel i. Thus, while in the Italian bambagia, gi exists as a digram possessing altogether merely the sound of g sibilant; in albagia, the same purpose is served, not by the digram gi, but by the single letter g preceding the i, which latter is distinctly pronounced with its own proper sound.

8. In Cornish, gwr gives place to wur by mutation of g into w.

9. I will remark here that one would need to be, if not blind, at least deaf, to be able to deny the identity in some cases, and the strong analogy in some others, between the Sardinian and the Celtic initial mutations, as far as concerns material points; though one should not for all that assume with absolute certainty the identity of the causes which produced these changes.


By H. W. LLOYD, M.A.

PERSONS of a literary taste, who may have lived long enough to remember Paris as it was in the early part of the present century, will probably not have forgotten M. Marcel, a learned Orientalist, who was sometime Director of the Imperial Printing Office under the first Napoleon. M. Marcel was by profession a publisher, and to his other pursuits, added that of bibliographical research. It was he who first brought to the notice of Prince L. Lucien Bonaparte a curious volume, printed in the year 1568, in a language evidently Celtic, but in a type and orthography exhibiting remarkable peculiarities, unlike those pertaining to any one of the existing families of that class of languages, and supposed by that gentleman to bear the nearest resemblance to the Cornish. Of this book Prince Lucien became the fortunate purchaser, and thus found himself the possessor of an unique copy of the "Athravaeth Gristnogavl”, a work which has just been reprinted as nearly as possible in facsimile by the Cymmrodorion Society, and which has contributed largely to the settlement of a curious controversy, as well as to the elucidation of some material facts in connexion with the publication of a larger, and to scholars, and, indeed, to the lovers and students of Celtic literature generally, a more interesting and important work, the Welsh Grammar of Dr. Griffith Roberts. To the Welsh title of this latter book is appended no press-mark; but simply the date of the year, and,

1 Read before the Honourable Society of Cymmrodorion, in London, on the 30th June, 1880.

[ocr errors]

in Latin, the day of the month;—1567. 1° Martij. We are told, however, by Dr. John Dafydd Rhøs, in the preface to his Welsh Grammar, that Gruffydd Roberts wrote a Grammar, and “Mediolani excudit”. A late librarian of the British Museum, Sir Antonio Panizzi, was unable to bring himself to believe that by “Mediolani” the city of Milan, in Italy, could be intended, because, as he averred, the peculiarities characteristic of Italian printing were not to be detected in the book. He, therefore, started an extraordinary theory of his own, which was that by “Mediolani” must be understood, not Milan, in Italy, but the place called in the old British Itineraries "Mediolanum”, a Roman military station, the site of which antiquaries have been greatly puzzled to fix with certainty, and has been placed by some in Cheshire, by others in Flintshire, but by others again, and with far greater probability, in Montgomeryshire. The villages in that county which have contended for the glory of it, like the seven cities of old for being the birthplace of Homer, being Llan St. Ffraid, Llanfair, Llanfechan, Llanfyllin, (from Myllin), and Meifod, the two last of which, approach the most nearly in sound at least to Mediolanum. Some months ago an announcement was made by Prince L. L. Bonaparte in the Academy, that in the colophon at the end of the preface to the Athrawaeth, which had proved to be a small Welsh catechism, printed by Dr. Griffith Roberts, the author of the Welsh Grammar, the words were found, "O dref, Fylen nosuyl S. Nicolas.”, and in that at the end of the book, “Ymylen. 1568. dyguyl. S. Nicolas.", words which it was declared, on no less an authority than that of the Rev. D. Silvan Evans, could be referred to no other place in the world than Milan in Italy; and thus no further room was left for controversy on the question which had been raised by Sir Antonio Panizzi, as Dr. Roberts must naturally have printed the one work at the place where he printed the other. No serious doubt, indeed, could be entertained on such a matter in the mind of a Welshman, as otherwise none could have been raised as to the whereabouts of the Roman station of Mediolanum, which would in that case readily have been identified by its very name. But no such place is or has been known either to historians or to the inhabitants in modern times. But if there were still room left for such a doubt after this discovery, still further materials are to be found for its solution in the existence of another work, the Drych Cristionogawl, or Christian Mirror, in the British Museum, which had been strangely overlooked by Panizzi, of which Dr. Griffith Roberts is unmistakeably the author. This (which, however, was printed, not by the author himself at Milan, but by his friend and fellow-worker, Dr. Roger Smith, at Rouen), contains, in Dr. Smith's introduction to it, a direct reference to a Welsh work printed by Dr. Roberts at Milan. It is also of great value for the light which it throws upon some other very interesting questions which have grown out of the publication of these and other works of a somewhat similar character, printed to all appearance in a sort of series, originating for the most part in a single cause, and culminating in a single object. Why, for example, was the printing of this series commenced in Italy and continued afterwards in France ? And secondly, why was the peculiar orthography and punctuation found in them, and in them only, that has chiefly led to these perplexities, adopted in the first instance by Dr. Roberts, and continued, with some variations, by his successor in the work of printing them, Dr. Roger Smith ? The answer to these questions, interesting alike to the critic of language and to the bibliographer, is fully supplied in the preface and in some supplemental additions to others of the different works; and, therefore, though somewhat long, I have ventured here to reproduce them.

Of the Drych Cristionogawl I am unable of my own knowledge to give the full and precise title, as the title

« ForrigeFortsæt »