The plan of "The Grave of the last Saxon" is simple and coherent, the characters marked and contrasted, and the whole (as is the case with all the writings of the Rev. W. L. Bowles) conducive to the excitement of virtuous sympathy, and subservient to that which alone can give dignity to poetry-the cause of moral and religious truth.

The three poets, whose works we have just reviewed, however differing in many respects, agree in this, that they scorn the foreign aid of ornament, relying on that undefinable charm which every unvitiated taste must discover in just representations of natural emotion. Therefore we have thought fit, without regard to their relative poetical merit, to notice their respective productions in one and the same article. We discern, indeed, yet higher points of resemblance to justify us in so doing; Wordsworth, Southey, and Bowles, are alike deeply impressed with that salutary fear which is the beginning of wisdom; and they are all three warm lovers and staunch defenders of those holy and time-hallowed institutions under which they live-recognising in them the source of national safety, and the means of happiness hereafter.

ART. VIII.-A Manual of Comparative Philology, in which the Affinity of the Indo-European Languages is illustrated and applied to the primeval History of Europe, Italy, and Rome. By the Rev. W. B. WINNING, M.A. Bedford. London: Rivingtons. 1838.

ALTHOUGH the attention of many scholars has been directed to this subject, it is far from being exhausted. The fact is no less curious than true, that nearly all, if not all, the European languages bear distinct traces of an Asiatic origin; that words, on the etymology of which lexicographers have greatly erred, exist in the same state as with us in the Sanscrit and its dialects; and that where the difference may appear remarkable, a comparison of languages will suggest tables by which the variation and interchange of consonants become explicable. To the orientalist it is well known that a change of vowels is of little or no moment in a comparison of languages.

In examining this subject, we shall first bring the work of Mr. Winning under review, and afterwards introduce to the reader a few of our own sentiments.

Because German philologists, comparing their language with the Latin, have recourse to Gothic and Low German, which through the Sclavonian lead us back to the old Median, and because they compare High German with Greek, or directly with Persian, the author divides these idioms into Medo

German and Perso-German; we would rather have said PersicoGerman. In his observations on the Greek language, he makes a similar division into Medo-Grecian and Perso-Grecian; and as Latin is the least mixed, and entirely belongs to the Median division, he concludes the Latin to be of greater antiquity than the Greek. The Welsh maintaining a close affinity with Greek, and the Erse with Latin, Mr. Winning classes the Celtic family under Medo-Celtic and Perso-Celtic. From these facts he infers that the Sanscrit had no direct influence on the Western idioms, but that the Sanscrit words were brought to Europe through the Median and Persian, its kindred languages, in which we are very far from entirely agreeing with him. For though it be indisputable, that to the Median and Persian Europe is indebted for many terms, too little is known of the origin of the Sanscrit, and its earlier influence on other languages, to justify so hasty an opinion. If the theory of Colonel Vans Kennedy be correct, and we see no reason why it should not be correct, the Sanscrit must have had a most powerful influence on European tongues. We know too little of the Zend to authorize a general conclusion respecting the transit of words from it to us, at least respecting the proportion in which European vocables may be retraced to it as a source; but we have a clear evidence of the power of the Sanscrit in its dialects, which display a close, often an exact, similarity to our Western idioms, where the Sanscrit itself is not so striking in the analogy; consequently the dialects properly Indian must have influenced the languages of Europe, and it is inferible that many of the words may have existed, as in these dialects, in the primitive state of the Sanscrit. The Persian, indeed, exhibits numerous proofs of affinity; but were a balance drawn between the two, we, who have made an accurate comparison, and deeply investigated the question, think that it would be in the favour of the Sanscrit and its dialects.

The languages of which Mr. Winning treats are viewed under the following tabular arrangement:

IRANIAN.

Sanscrit, Zend, Persian.

Irano-Indian.

Sanscrit, Hindústáni, Bengáli.
Irano-European.

Zend, Persian, Sclavonian, Lithuanian, German, Celtic.

The subordinate languages of the four great European families :-
Sclavonian.

Russian, Servian, Croatian, Wendish, &c.
Lithuanian.

Lithuanian Proper, Lettish, Old Prussian, Latin.

GERMAN.

Lower German.

Gothic, Scandinavian, Dutch, English, &c.
Upper German.

Old, Middle, and New High German, Greek.

Celtic.

Erse, Gaelic, Welsh, Bas Breton, Basque.

After some remarks on the mutes in their divisions into labials, linguals, and gutturals, and the subdivision of each class into tenues, medials, and aspirates, with a short exemplification of prefixes, he proceeds to the consideration of Grimm's law on the regular interchange of certain letters in different languages, which Bopp has extended to the Zend and Lithuanian. These are exceedingly remarkable; but the interchange is greater than either Grimm or Bopp have noticed. Instances, too, are observ, able, in which there are exceptions to this law, other letters taking the place of those which are most commonly interchanged.

Mr. Winning, whose meritorious labours cannot be too highly extolled, has laid down as a general law, that where the MedoEuropean languages use medials, either tenues or aspirates will be found in the corresponding Perso-Grecian, and Old High or Perso-German; and the examples which are given remove the assertion from all doubt. The vocabularies also fully demonstrate the general arguments; and it is to be hoped that the striking analogies which they have brought to light, will induce our future lexicographers to correct the silly errors into which their predecessors have fallen.

The chapter on the Sclavonian languages, interspersed as it is with critical remarks on the Latin, will be found of singular value to the philologist. The principles of etymological research are lucidly detailed in it; and the matter is so substantiated by history, and so corroborated by internal proofs contained in languages, that it ceases to be merely hypothetical, but may be accepted as a series of established facts. The Lettish, or Old Prussian, being nearly related to the Sclavonian and German, resembling more the latter in grammatical forms, and the former in the mass of its words (a characteristic peculiar to the Old Prussian and Latin), it has been supposed that this circumstance may throw light on the origin of the Romans; the notion of the Goths having constituted a portion of the earliest inhabitants of Italy, which arose from the similarity between the Gothic and Latin inflexions, being devoid of every historical trace. But it is affirmed that a connexion between the Prussian Cures on the Curische Haf and the Romans can still be historically shown; and since the affinity between the Old Prussian and Latin is as striking on other points, Mr. Winning conjectures that the

Gothic portion of the Latin, also, existed originally without the intervention of any Gothic settlers into Italy:

"From the name of the two people, the affinity of their languages, and the perfect identity of many of their customs, I am led to the conviction that the Sabines were of Prussian origin, and formed that part of the Roman people which introduced the peculiarities of the Lettish language and customs. The Prussian Sabines might easily coalesce with the Sclavonians, Wends, and other Medo-Grecians; while it is almost inconceivable that the Goths, of a totally distinct race, should force the grammar of their language upon Latiǹ tribes, who retained their Sclavonian vocabulary."

These ideas are strongly supported in the following remarks on the traffic in amber, which is abundantly found in Prussia:"It is also surmised that the amber-bearing Eridanus transferred its name to the Po, for Herodotus mentions the legend of a river of this name flowing into the Northern Sea, whence amber came; and that the modern Radaune, which joins the Vistula near Dantzic, on the banks of which amber is still plentifully found, is this river." The chapters likewise on the German and Celtic language are filled with erudition, and deserve a most minute consideration-one which we unfortunately cannot bestow upon

them.

On the relationship between the proper Persians and the Pelasgi of Greece the author's observations have much in the form of authority; but we greatly doubt that the name of Cyrus had any reference to the Curetic worship. A word exactly answering to the Hebrew, from whence the idea seems to have proceeded in Median or Old Persian, would be required, ere any probability could be conceded to the conjecture: the modern Persian name is. We also doubt whether was the name in Pehlvi; because is retained as a name of the sun in modern Persian, and in no way answers to the interpretation given of in the Ferhangi Jehangiri and Berhani-Kattea. Yet were this the name of Cyrus in Pehloi, the difference between it and would strongly militate against the hypothesis that he was thus distinguished as the head of the Curetic worship.

خورشید

We likewise think that many parts of the author's Hamite classification are conjectural, and not to be established, and that he assumes too much when he declares that the , the Cherethites of our version, were certainly Cretans. The idea is not new, and has been examined in our review of Dr. Russell's Connexion of Sacred and Profane History. Mr. Winning next devotes his attentive inquiry to the Pelasgians of Asia Minor, Greece, and Italy. This is a most perplexed subject, as fertile in intricacies and hypotheses as any, perhaps, that has been agitated by scholars, and one which, we may be assured, will

never be historically determined. Every one who has investigated it proceeds to his task with a prepared system; and the origin of the Pelasgi has thus been accommodated to systems too many to be enumerated. We honestly profess our own ignorance; but were we to trouble our readers with our conjectures, we have no doubt that we could make them as plausible as others which have been received for facts; nay, we could etymologically support our conjecture; yet, after all, who would accept it as a historical verity?

What we should expect to receive from public judgment, Mr. Winning will not object to receive from us. He is very ingenious; nevertheless, we require more positive evidence than that which he has given to us. That the reader may judge how diversified have been the opinions on the subject, we will inform him, that Vans Kennedy called the Pelasgi the ancestors of the Thracians, in which he is supported by the affirmation of Strabo, that the Getæ and Thracians spoke the same tongue; and by Ovid's statement, that the Moesian was a dialect of the Thracian. But some derive this Pelasgic or Thracian language from Asia Minor, and others from Middle Asia. Baron Cuvier, on the other hand, traces the Pelasgi from India, supposing them in crossing the Persian mountains to have reached Kâf, or the Caucasus, and thence embarking on the Black Sea, to have reached the shores of Greece. Now, the Thracians certainly had an extraordinary influence on the mythology of Greece: Thamyris, Orpheus, and Musæus were Thracians, which is in favour of the identity of them and the Pelasgi: for Strabo calls the oracle of Dodona Пeλary@v ispvua, and the Dodonaan Jupiter Пeλaoyukós; and in support of Vans Kennedy's notion, that this Thracian or Pelasgic was allied to the Sanscrit, we would urge the names of the gods, for instance, Jupiter is a compound of the Sanscrit Ju, æther, heaven, (which gives in its inflexion the Latin oblique cases, Jovis, &c.,) and PITRI, father, which is corroborated by the well-known line of Ennius,

66

Aspice hoc sublime candens, quem invocant omnes, JOVEM."

For it will hereafter appear, that the Pelasgic influenced the Latin and Greek in different ways. The Father of Gods and men is also not unfrequently called in Sanscrit JIVAPITRI, the Father of Life; and either term is sufficient to account for the Latin name. The proof becomes still stronger from his Latin epithet Diespiter, which is Indra's title DIWASPATI, Lord of the Day. It has likewise been imagined, that Zeùç is a corruption of Indra's title DEWESHA, Lord of Gods, the genitive Alòs being recognisable in its genitive DEWASYA; this may or may not be the case, as the name may also be retraced to the root corresponding to Záw, and thus would imply the same, as JIVAPITRA,