Before going further, it will be proper to say something more as to the essential peculiarity of Pahlawí to which we have already repeatedly alluded, namely, its Huzvárish or Zavárishn element of Aramaic words more or less defaced in many cases by the addition of Persian inflexional terminations and “phonetic complements.” When a Pahlawí text is read, a large proportion of the words composing it are found to be Semitic, not Íránian, and, to be more precise, to be drawn from an Aramaic dialect closely akin to Syriac and Chaldæan. Now since an ordinary modern Persian text also contains a large proportion of Semitic (in this case Arabic) words, which are actually read as they are written, and are, in fact, foreign words as completely incorporated in Persian as are the Greek, Latin, French, and other exotic words which together constitute so large a portion of the modern English vocabulary in our own language, it was at first thought that the Aramaic element in Pahlawí was wholly comparable to the Arabic element in modern Persian. But a more careful examination showed that there was an essential difference between the two cases. However extensively one language may borrow from another, there is a limit which cannot be exceeded. It would be easy to pick out sentences of modern Persian written in the high-flown style of certain ornate writers in which all the substantives, all the adjectives, and all the verbal nouns were Arabic, and in which, moreover, Arabic citations and phrases abounded; yet the general structure of the sentence, the pronouns, and the auxiliary verbs would and must continue to be Persian. Similarly in a sentence like “I regard this expression of opinion as dangerous,” only four of the eight words employed are really of English descent, yet the sentence is thoroughly English, and it is inconceivable that the pronouns “I” and “this,” or the particles “of” and “as,” should be replaced by equivalents of foreign extraction. In Pahlawí, however, the case is quite different. Haug goes, perhaps, a trifle too far when he says (Essay on Pahlawí, pp. 120-121) “all the case-signs and even the plural suffixes in the nouns; all the personal, demonstrative, interrogative and relative pronouns; all the numerals from one to ten; the most common verbs (including the auxiliaries) such as ‘to be, to go, to come, to wish, to eat, to sleep, to write, &c.’; almost all the prepositions, adverbs, and conjunctions, and several important terminations for the formation of nouns, as well as a large majority of the words in general (at all events in the Sásánian inscriptions), are of Semitic origin;” yet in the main such is the case, and “the verbal terminations, the suffixed pronouns and the construction of the sentence” are often the only Íránian part of the Pahlawí phrase, though they are its essential and characteristic part. But in addition to this we have a number of monstrous, hybrid words, half Aramaic, half Persian, which no rational being can imagine were ever really current in speech. Thus the Semitic root meaning “to write” consists of the three radicals K, T, B, and the third person plural of the imperfect is yektibún (Arabic, yaktubún), while the Persian verb is nabishtan, napishtan, or navishtan. The Pahlawí scribe, however, wrote yektibún-tan, but assuredly never so read it: to him yektibún, though a significant inflected word in Aramaic, was a mere logogram or ideogram standing for napish-, to which he then added the appropriate Persian termination. So likewise for the Persian word mard, “man,” he wrote the Semitic gabrá, but when he wished the alternative form mardum to be read, he indicated this by the addition of the “phonetic complement,” and wrote gabrá-um.
The analogies to this extraordinary procedure which exist in Assyrian have already been pointed out. In the older Túránian language of Akkad “father” was adda. “When the Assyrians,” says Haug, “wished to write ‘father,’ they used the first character, ad or at, of adda, but pronounced it ab, which was their own word for ‘father’; and to express ‘my father,’ they wrote atuya, but read it abuya; u being the Assyrian nominative termination, and ya the suffix meaning ‘my,’ which, in the writing, were added to the foreign word at.” So in like manner the Pahlawí scribe who wished to write “father” wrote abitar for pitar (pidar), the Assyrian ab being used as a mere ideogram, of which the Persian equivalent was indicated by the “phonetic complement” -tar.
Another curious (and, in this instance, valuable) feature of the Pahlawí script was that in the case of a Persian word recognised at that time as compound and capable of analysis, each separate element was represented by a Semitic or Huzvárish equivalent. Take, for instance, the common Persian verb pindáshtan, “to think, deem.” A modern Persian has no idea that it is capable of analysis, or is other than a simple verb; but the Pahlawí scribe was conscious of its compound character, and wrote accordingly pavan (= pa, “for”) háná (= in, “this”) yakhsanún-tan (= dáshtan, “to hold”); and Nöldeke has called attention to a similar analysis of the common word magar (“unless,” “if not”), which is represented by two Aramaic words, or Huzvárish elements, of which the first signifies “not” and the second “if.” And this principle has another curious and instructive application. The modern Persian pronoun of the first person singular is man, which is derived from the stem of the oblique cases of the corresponding Old Persian pronoun adam (= Avestic azem), whereof the genitive is maná. Of this fact the Pahlawí script takes cognisance in writing the Semitic lí, “to me” (or “of me”), as the Huzvárish equivalent of man.
These considerations, apart from external evidence, might have suggested to a very acute mind the belief that the peculiarities of Pahlawí lay almost entirely in the script, and that they disappeared when it was read aloud. Fortunately, however, there is sufficient external evidence to prove that this was actually the case.
First, we have the direct testimony of Ammianus Marcellinus, who says (xix, 2, 11): “Persis Saporem et Saansaan [i.e., Sháhán-sháh] adpellantibus et Pyrosen [i.e., Piruz or Pêrôz], quod rex regibus imperans et bellorum victor interpretatur.” This notice refers to Shápúr II (A.D. 309-379), whose title stands on his coins Malkán malká, but was in the actual speech of the people, then as now, Sháhán-sháh.
Secondly, we have the direct testimony of the learned author
of the Fihrist, Muḥammad b. Isḥáq (A.D. 987-8), who relies
here, as in other places where he speaks of matters appertaining
to Sásánian Persia, on the authority of that remarkable man
Ibnu'l Muqaffa'.
Ibnu'l-Muqaffa', a Persian Zoroastrian who
flourished about the middle of the eighth
century of our era, made a doubtfully sincere profession
of Islám, and was put to death about A.D. 760. He
was reckoned by Ibn Muqla, the wazír and calligraphist
(† A.H. 939), as one of the ten most eloquent speakers and
writers of Arabic, and Ibn Khaldún the Moor pays a similar
tribute to his command of that language; and with this he
combined a thorough knowledge of Pahlawí, and translated
several important works from that language into Arabic, of
which, unfortunately, only one of the least interesting (the Book
of Kalíla and Dimna) has survived to our time. Relying on
the authority of this learned man, the author of the Fihrist,
after describing seven different scripts (Kitába) used in pre-
“And they have likewise a syllabary [hijá, “a spelling,” not kitába, “a script”] called Zawárishn [or Huzvárish], wherein they write the letters either joined or separate, comprising about a thousand vocables, that thereby they may distinguish words otherwise ambiguous. For instance, when one desires to write gúsht, which means ‘meat,’ he writes bisrá like this [here follows the word written in the Pahlawí script], but reads it as gúsht; and similarly when one desires to write nán, which means bread, he writes lahmá like this [again follows the Pahlawí word], but reads it nán; and so whatever they wish to write, save such things as have no need of a like substitution, which you write as they are pronounced.”
*