The Classical Journal

It may be proper to mention here our belief, that a very slight metrical pause was made after every foot in reciting Latin as well as Greek hexameters. See on this subject the first part of the "Critical Investigation of the Versification and Prosodial Usages of the Iliad and Odyssey."

2. When a syllable formed as above is so situated as not to have a metrical pause after it, it is to be considered long or short, according as it does or does not receive the metrical accent; the two consonants, of themselves, requiring little more exertion to be pronounced together than a single consonant requires. So, Ecl. v11. 36. Si fœtură gregem suppleverit, aureus esto. Geor. 1. 235. Quam circum extremæ dextra lævaquě trahuntur.

320. Sublime expulsam eruerent: ita turbine nigro. 11. 217. Quæ tenuem exhalat nebulam, fumosque volucres. We have to observe in the third place, relative to the power of the ictus metricus in Virgilian hexameters, that occasionally, in virtue of it, a final short syllable, formed by a short vowel followed by a consonant, occupies the place of a long syllable; as in the verses cited in the beginning of this treatise, and the following:

Ecl. 1. 39. Tityrus hinc aberat. Ipsæ te, Tityre, pinus. x. 69. Omnia vincit amōr: et nos cedamus amori. Georg. 1. 138. Pleïadās, Hyadas, claramque Lycaonis Arcton. IV. 463. Atque Getæ, atque Hebrūs, atque Actias Ori

thyia.

En. 1. 308. Qui teneant, (nam inculta vidēt,) hominesne, feræne.

XII. 883. Te sine, frater, erīt. O quæ satis ima dehiscat. The instances of this usage in Virgil's poems amount in number to more than fifty.

We remark, fourthly, that in three cases a short syllable, formed by a short final vowel before an initial consonant, is put for a long one, in consequence of the lengthening efficacy of the ictus metricus. The verses referred to are,

En. 111. 91. Liminaque, laurusque Dei: totusque moveri. 464. Dona dehinc auro gravia, sectoque elephanto. XII. 363. Chloreaque Sybarimque Daretaque Thersilochum

que.

It should appear that this usage is not only of very rare occurrence, but is also restricted to the case when the initial consonant is a liquid, or the letter s.

Lastly, the metrical emphasis occasionally enables the first syllable of religiō, reliquiæ, rèpĕrit, rětůlít, quătŭăr, and a few

others, to stand for the first of a dactyl, as in the subjoined lines,

Geor. 1. 270. Religio vetuit, segeti prætendere sepem.

Æn. 1. 30. Troas, reliquias Danaum atque immitis Achillei. Geor. 11. 22. Sunt alii, quos ipse via sibi reperit usus. En. v. 598. Retulit, et priscos docuit celebrare Latinos.. Geor. 1. 258. Temporibusque parem diversis quatuor annum.

It remains only to observe, that in the last three cases the accent probably rested particularly on the consonant. We have the authority of Mss. for considering that religio, the first syllable receiving the ictus, was uttered, in measure, relligio; and in reference to the third and fourth cases mentioned, the fact, that the writers of correct Latin hexameters studied to let the consonant be one of those which most forcibly reverberate in pronunciation when accented, as r, l, s, furnishes a strong presumptive evidence in favor of the position, we have assumed. And let it be remembered, that in respect of most points of metrical science, great probability is the utmost at which we can arrive by critical deduction: to positive and absolute certainty we are almost entire strangers.

ZEND AND PAHLAVI MANUSCRIPTS.

FRO

ROM some of those ingenious correspondents who occasionally diversify the classical pages of your Journal with articles illustrating Eastern literature, I would beg leave to solicit information respecting a large and valuable collection of manuscripts, brought from India above thirty years ago by Mr. Samuel Guise, and announced for sale by private contract. One of the printed catalogues which described this collection is now in my possession. The Arabic and Persian books were numerous, and some among them appear to have been rare and curious; but the Mss. that form the subject of my present inquiry, are those works composed in the ancient dialects of Persia called Zend and Pahlavi, many of which have been ascribed to authors of very remote antiquity, and some even to Zerátusht, or Zerdasht, the great Zoroaster himself. Mr. Guise procured those extraordinary writings at Surat, where be resided many years as a physician or surgeon: most of the Zend and Pahlavi manuscripts were purchased at a considerable price from the widow of Daráb, a learned destour or priest of the Parsee fire-worshippers, who had some time before instructed the accomplished Frenchman, M. Anquetil du Perron, in the different

dialects used by the ancient Persians, as far as their modern descendants could understand them through the assistance of books and of verbal traditions. Thus informed, M. Anquetil, on his arrival in France, published the "Zendavesta,” a work of considerable labor and ingenuity, in three quarto volumes; one of which comprises a list of the Mss. collected by himself in India. On a comparison, however, of this list with Mr. Guise's catalogue, it appears that our fellow-countryman was fortunate in obtaining some works which the French Orientalist had been unable to procure. Now the object of my inquiry is to ascertain whether Mr. Guise's collection has been purchased for one of our universities or other public institutions, or by some private individual of this country; whether it has found a place in any continental library, or whether it still continues unsold, and, in this case, how an application for the purchase of it should be made.

I am aware that many doubts have been entertained respecting the age and authenticity of those Zend and Pahlavi compositions; and a work lately printed at Bombay from a manuscript supposed by some to be a genuine specimen of the ancient Persic language, has been condemned as spurious by able critics, or regarded as a modern fabrication. That the learned Dr. Hyde of Oxford, who, about one hundred and thirty years ago, published his elaborate treatise "De Religione Veterum Persarum," expected some important results from a knowlege of the old Persian language, is evident, not only from the time and labor which he devoted to the study of it, but from the expense incurred in causing a font of metal types to be very handsomely and accurately cut in imitation of the Zend and Pahlavi characters. Whether these metal types are still preserved at Oxford, or were removed several years ago, as I have heard, to the British Museum, is another circumstance on which information would be highly acceptable.

But the exertions of Dr. Hyde, however laborious, have not in any considerable degree facilitated our acquiring a knowlege of the ancient Persic dialects: in his time England possessed but few Zend or Pahlavi Mss., and he wanted the assistance of vocabularies or dictionaries. It was reserved for M. Anquetil du Perron to furnish ample and curious specimens of those dialects, which have for some centuries been considered by the Parsee fire-worshippers as genuine remains of the language used in Persia, not only while the Sassanidan monarchs governed that empire, but during the age of Darius, and, as I before remarked, of Zoroaster himself. The Mss. brought to France by M. Anquetil, and now preserved in Paris, if they were collated with those brought to England by Mr. Guise, would furnish materials for some interesting works which might be printed in the types cut under Dr. Hyde's inspection. Our public libraries contain a few volumes of this rare class, and others are deposited in the private collections VOL. XXXIX. CI. JI. NO. LXXVII.

B

only takes place when one or both of the consonants are to be pronounced immediately after the vowel, and not when both are to be separated from it in pronunciation, by an intervening vocal pause. On this principle it is that the first two rules of the Eton Grammar are founded; and, if true, the sentiments of Dr. Carey on the subject of the third rule must also be founded on this principle. To this principle, however, the advocates of these sentiments do not appear to have been sufficiently attentive; and in endeavoring to establish their doctrine, they seem not to have regarded the difference in recitation between poetry and prose. We certainly think that there is no reason why, in pronouncing the first of the above lines, que in Eurique should be separated from Zephyrique; or why, in pronouncing the last, telu should not be connected with the initial s of scandite. That this latter procedure is not of the most elegant kind we readily admit; still it was certainly more allowable than arbitrarily to change the quantity of the last syllable of tela; and as to the former, it is an usage the most correct and proper, since according to no consistent theory of reciting hexameter verses can a pause be made in the middle of a foot after so unimportant a word as the particle que. The only regulation that can be made relative to the quantity of the syllables referred to must be of a general nature, as follows: "When the vowel is so situated, as to be separated in pronunciation from the succeeding consonants, the syllable is short; and when the vowel is so situated, as to be connected in pronunciation with one or both of the succeeding consonants, the syllable is long."

With regard to the fourth rule of the Eton Grammar, though it is correct as containing a general representation of the fact of the case, yet it is defective as furnishing no explanation of the principles of the usages in question. It may with propriety be subdivided into the two following regulations, the latter of which exhibits the influence of the ictus metricus on Latin prosody in a most important particular :

1. When a syllable formed by a short vowel, followed by a mute and one of the liquids, is so situated as to have a metrical pause made after it, its quantity depends on the vowel being joined to one of the consonants, or separated from both by that pause. So,

Ecl. 1. 53. Et fontes sac,ros frigus captabis opacum.
VI. 54. Ilice sub nīg,ra pallentes ruminat herbas.

1. 32. Pan primus calamos cera conjungere, plures.
vi. 42. Caucaseasque refert volu.cres furtumque Promethei.