The first important step in the vindication of Anquetil was made by his illustrious compatriot, Sylvestre de Sacy, De Sacy's Mémoires sur diverses An­tiquités de la Perse (1793). who, in 1793, published in the Journal des Savants his five celebrated Mémoires sur diverses Antiquités de la Perse, which dealt chiefly with the Pahlawí inscriptions of the Sásánian kings, for the decipher­ment of which he chiefly relied, apart from the Greek translations which accompany some of them, on the Pahlawí vocabulary given by Anquetil (vol. iii, pp. 432-526), “whose work,” as Darmesteter well says, “vindicated itself thus— better than by heaping up arguments—by promoting dis­coveries.” For the oldest extant manuscripts of the Avesta date only from the fourteenth century of our era, while the Sásánian inscriptions go back to the third, and could not, therefore, be set aside, even for a moment, as late forgeries; and if Anquetil's vocabulary furnished a key to these, it was manifest that the Pahlawí which he had learned from his dastúrs was the genuine language of Sásánian times; and that the occurrence in it of Semitic words, such as malká “king,” shanat “year,” ab “father,” shamsá “sun,” “not,” which Sir William Jones, regarding them as Arabic* (though he afterwards recognised them as Chaldæan),* cited as proof of the fictitious antiquity of the language in which they occurred, of Anquetil's credulity, and of his Pársí instructor's fraud, was an indisputable fact, whatever might be its true explanation. Tychsen insisted strongly on this point.

“This,” said he, “is a proof that the Pahlawí was used during the reign of the Sásánides, for it was from them that these inscriptions emanated, as it was by them—nay, by the first of them, Ardashír Bábagán—that the doctrine of Zoroaster was revived. One can now understand why the Zend books were translated into Pahlawí. Here, too, everything agrees, and speaks loudly for their antiquity and genuineness.”

*

The Pahlawí inscriptions thus deciphered by de Sacy had been known in Europe since Samuel Flower published in the Philosophical Transactions for June, 1693 (pp. 775-7) the Pahlawí inscrip­tions. copies of them which he had made in 1667, while further copies appeared in the works of Chardin (1711), Niebuhr (1778), and, at a later date, of other travellers;* but, though Hyde reproduced them in his book, de Sacy was the first to attempt with any success their interpretation.

Five years after the publication of de Sacy's Mémoires (1798), the Carmelite father, Paul de St. Barthélemy, published at St. Barthélemy. Rome his essay, De antiquitate et affinitate linguæ samscredamicæ et germanicæ, in which he defended the antiquity of the Avesta, and even uttered a conjecture as to the affinity of the language in which it is written with Sanskrit.

*

The first important step in the next, and perhaps the greatest, achievement of Persian scholarship—to wit, the Decipherment of the Old Persian Inscrip­tions.-Grotefend. decipherment of the Persian cuneiform inscriptions (writings of which the character and language were alike unknown)—was made early in the nineteenth century by Grotefend, whose papers on this subject—models of clear reasoning and acute insight— have only recently been unearthed from the Archives of the Göttingen Royal Society of Sciences and published in the Nachrichten of that Society (September 13, 1893, pp. 571-616) by W. Meyer. Of these papers the first was originally read on September 4, 1802, the second on October 2nd, the third on November 13th of the same year, and the fourth on May 20, 1803. Till this time, though Tychsen and Münter had made vain attempts at decipherment, it was, as we have seen when examining Hyde's work, very generally held, even by men of learning, that these characters were not writing at all, but were either architectural ornaments, the work of worms or insects, or mason's marks and numerical signs. Grotefend, primarily impelled to this inquiry by a dispute with his friend Fiorillo as to the possibility of arriving at the meaning of inscriptions whereof the script and language were alike unknown or buried in oblivion, arrived in his first communication at the following important general conclusions: Grotefend's general con­clusions. (1) That the figures constituting these inscriptions were graphic symbols; (2) that the inscriptions were trilingual, that is, that they consisted, as a rule, of three versions, each in a different language and script; (3) that the inscriptions which he proposed to explain, that is, those of the first class (the Old Persian) in particular, and also those of the second, consisted of actual letters, not of ideograms or logograms comparable to those employed in Assyrian and Chinese; (4) that all known cuneiform inscriptions were constant in direction, being in every case written horizontally from left to right.

From these general conclusions (all of which have since proved to be perfectly correct) Grotefend proceeded to Grotefend's method of procedure. examine more minutely two inscriptions of the first class, which he believed to be written in the so-called Zend (i.e., Avestic) language—a con­jecture which, though not the truth, was near the truth—and which he correctly referred to “some ancient king of the Persians between Cyrus and Alexander,” in other words, to the Achæmenians.* An examination of the Pahlawí inscriptions of the Sásánians, already deciphered by de Sacy, suggested to him the probability that the first word in the inscription was the name of a king of this dynasty, and the second his title. He then observed that that name which stood at the beginning of the second inscription was in the first placed after the title, which (again guided by the analogy of the Sásánian inscrip­tions) he rightly assumed to signify “King of Kings,” with a slight final modification, which he correctly conjectured to be the inflexion of the genitive case, from which he gathered that the two names in the first inscription were those of father and son. One of these names, which Tychsen had read Malkéusch, appeared to him to square best with Darius, whose name in the Books of Daniel, Ezra, and Nehemiah occurs in the form Dáriyávush (“Darjavesch”); another, read by Tychsen as Osch patscha, with Xerxes (“Khschhêrschê). For both these names consisted, in the Old Persian inscriptions, of seven separate characters (these being, as we now know, in the first, D. A. R. Y. V. U. SH, and in the second, K. SH. Y. A. R. SH. A), of which one (A) occurred three times, and three (R, Y, SH) twice, in the two names; and the assumption as to the reading of these names was confirmed by the order of the component letters of each. Now it was known from the accounts of the Greek historians that Darius was the son of Hystaspes, which name appeared in Anquetil's work in the native forms Gushtásp, Vishtásp, &c.; and, from the analogy of the inscription of Xerxes, it appeared probable that Darius also in his inscription would mention this, his father's name. And, in effect, there occurred in the proper place in this inscription of Darius a group of ten letters, of which the last three (now known to represent H. Y. A.) had already been recognised as the case-ending of the genitive. Of the remaining seven, two—the third (SH) and fifth (A)—were already known, while, from what was common to the Greek and Avestic forms of the name, the fourth, sixth, and seventh might fairly be assumed to represent T, S, and P respectively. There remained the two initial letters, of which it was pretty evident that the first was a consonant (G or V), and the second a vowel (not U, already known, and therefore presumably I); but Grotefend actually read them as G. O. instead of V. I.

Such were the great and definite results of Grotefend's discoveries. Further than this he endeavoured to go; but, Grotefend's results. on the one hand, he was misled by his belief that the language of the inscriptions was identical with that of the Avesta, and by the fact that Anquetil's account of the latter was imperfect and in many details erroneous; and, on the other hand, the materials at his disposal were inadequate and did not supply sufficient data for full decipherment and interpretation. Hence his scheme of the values of the letters was, as we now know, scarcely even half correct, while his interpretations and transcriptions of the texts which he attacked were but approximations. Thus one of the Persepolitan inscriptions with which he especially dealt (Niebuhr, Pl. xxiv; Spiegel's Keilinschriften, ed. 1862, p. 48, B), is now known to read as follows:—