Among the Jews of Babylonia.

Language spoken by the Jews living in Persia. The earliest evidence of the entrance of Persian words into the language of the Israelites is found in the Bible. The post-exilic portions, Hebrew as well as Aramaic, contain besides many Persian proper names (especially in Esther; see I. Scheftelowitz, "Arisches im Alten Testament," part i., 1901) and titles (e.g., "satrap," "aḥashdarpenim"), a number of nouns (as "dat" = "law"; "genez" = "treasure"; "pardes" = "park") which came into permanent use at the time of the Achæmenidæ. More than five hundred years after the end of that dynasty the Jews of the Babylonian diaspora again came under the dominion of the Persians; and among such Jews the Persian language held a position similar to that held by the Greek language among the Jews of the West. Persian became to a great extent the language of everyday life among the Jews of Babylonia; and a hundred years after the conquest of that country by the Sassanids an amora of Pumbedita, Rab Joseph (d. 323), dared make the statement (Soṭah, end) that the Babylonian Jews had no right to speak Aramaic, but should speak either Hebrew or Persian. Aramaic, however, remained the language of the Jews in Palestine as well as of those in Babylonia, although in the latter country a large number of Persian words found their way into the language of daily intercourse and into that of the schools, a fact which is attested by the numerous Persian derivatives in the Babylonian Talmud. But in the Aramaic Targum there are very few Persian words (see "Z. D. M. G." xxviii. 56, 67), owing to the fact that after the middle of the third century the Targumim on the Pentateuch and the Prophets were accepted as authoritative and received a fixed textual form in the Babylonian schools. In this way they were protected from the introduction of Persian elements. On a possible early Judæo-Persian translation of the Bible see Judæo-Persian Literature, § 1.

Persian in the Talmud.

The explanation of the Persian derivatives in the Talmud (they are even more numerous in the geonic literature) is one of the most important tasks of Talmudic lexicography. R. Nathan explained about twenty words in the Talmud as being Persian (see Rapoport, "Biographie R. Nathans," note 6; Kohut, "Aruch Completum," Introduction, p. viii.). A comprehensive work on the subject is still a desideratum. Contributions to the subject have been made by Fleischer (addenda to Levy's "Wörterb."), Perles ("Etymologische Studien," 1871; "Zur Rabbinischen Sprach- und Sagenkunde," 1873; and in "Monatsschrift," 1893), and Kohut. The last-named, however, in his edition of the "'Aruk," has not always used the necessary discretion, and frequently declares what is of purely Semitic origin to be derived from the Persian ("Z. D. M. G." xlvii. 501-509).

Earliest Traces of Modern Dialect.

Nothing definite can be said in regard to the diffusion of the Persian language among the Jews during the long period that elapsed between the date of the completion of the Talmud and that of the earliest monuments of the Judæo-Persian literature. It is nevertheless beyond all question that a portion of the extant translations of the Bible originated in a much older period than did the Judæo-Persian writings whose dates may be determined. Even in these writings there are "so many ancient phonetic and lexical and at times even grammatical forms, such as are not found in the oldest Neo-Persian monuments, that this literature must be assigned to a comparatively early date" (Salemann, "Khudâidâd," p. ii.). At any rate the old forms in Judæo-Persian show that Persian had at a very early time become the mother tongue of the Jews that lived in those portions of the dominions of the califs where Persian was spoken. It is even probable that as early as the Sassanids there were Jewish communities which spoke Persian.

The earliest literary monument of Judæo-Persian is the curious document, dating from the eighth century, which Dr. M. Aurel Stein has found in the ruins of Khotan. Its language is almost free from Arabic admixture, and it contains no Hebrew words (D. S. Margoliouth, in "Journal of the Royal Asiatic Society," Oct., 1903, pp. 735-760). The second oldest document (the deposition of witnesses in a family lawsuit) dates from the year 1332 of the Seleucidan era (=1020 C.E.), and was written in Ahwaz (Persian,"Hormshir"), the capital of Khuzistan, in which city there was at an early date a large Jewish population (idem, in "J. Q. R." xi. 671 et seq.).

Contemporary Dialects.

The Neo-Persian language, which brought forth a flourishing national literature in the tenth century, probably became the mother tongue of the Jews who lived within the smaller territory where Persian was spoken. Persian is spoken to-day by the Jews of Persia proper, and, for the most part, by those under Russian dominion in Central Asia (in Bokhara, etc.). There is a colony of Bokharian Jews in Jerusalem. The total number of Jews who speak Persian is estimated at about 50,000. The so-called "Mountain Jews" of the Caucasus speak a dialect called "Tat," which varies considerably from the ordinary Neo-Persian. Their number is reckoned at 20,000 (see Jew. Encyc. iv. 558b, s.v. Dialects; W. Geiger, "Bemerkungen," in "Grundriss der Iranischen Philologie," i., section 2, p. 408; and Vsevolod Miller, in the third report of the Lazarevski Institute for Oriental Languages at Moscow). What Geiger says of the general characteristics of Judæo-Persian should especially be emphasized (l.c. p. 410): "It differs from the written language not nearly so much as the other dialects and patois of Persian. It is no more a historically independent dialect than is the Tadshiki (the Persian of the transoxanian lands), but is rather a corruption of the High Neo-Persian with local dialectic color." This is due to the fact that the Persian Jews busied themselves with Persian national literature, transcribing it in Hebrew characters.

Corruption from Literary Language.

In their own poetry also the Jews were influenced by this literature, which made itself felt even in their Bible translations. But in course of time corruptions crept in, especially in these Bible translations, as they were carried from place to place. Simeon Ḥakam in the introduction to his translation of the Pentateuch gives interesting examples of Persian words which are corrupted in vulgar speech and are used in this corrupted form by the Jews of Bokhara in their letters to one another. Thus, instead of ("âsmân" = "heaven"), they write ; instead of ("pâdishâh" = "ruler"), ; instead of (Arabic, "sajdah" = "bending"), ; instead of (Arabic, "kitf" = "shoulder"), ; instead of (Arabic, "jam'a" = "congregation"), ; instead of (Arabic, "'umr" = "life"), . Of these examples the last three show instances where the corruption consists in the transposition of the consonants. The same occurs in the poem of Chudâidâd (see Salemann, l.c. p. vi.). Simeon Ḥakam himself used the vulgar tongue ("lashôn hamôniyi") in his translations intended for his fellow countrymen of Bokhara; and his Persian ritualistic compendium may well be called a treasure-house for the Judæo-Persian idioms of Bokhara. On the other hand, Simeon asserts that the language of his translations is correct Persian, "leshôn pârsi ẓaḥ" (see the title-page of the Pentateuch translation in "Z. D. M. G." lvi. 730).

The following attempt to present an idea of the character of Judæo-Persian will be restricted to grouping the prominent peculiarities under several heads. A distinction is made between the language of the older literary productions (such as the dictionary of Moses Shirwani and the commentary on Samuel), and the most recent Judæo-Persian writings in Bokhara. Among the latter is the poem of Chudâidâd, although it was written at the beginning of the nineteenth century.

Transcription: Punctuation.

All of the Judæo-Persian literature, whether in manuscript or printed, presents the Persian text in Hebrew characters. In Bokhara at an early date the Hebrew vowel-signs were used to represent the Persian vowels in the copies of the poems of Yusuf Yehudi and his circle and of the Chudâidâd poem. In the commentary on Samuel the Persian is also partially punctuated. The Neo-Persian publications in Jerusalem intended for Bokhara are all punctuated. This has the advantage of showing the pronunciation of the Persian within the district in which it originated more clearly than is the case with the common Arabo-Persian alphabet with its scar-city of vowel-signs. Thus, for example, in the Hebrew transcription "k" is always distinguished from "g"; also, as regards vocalization, "î" from "ê" "û" from "ô." The Hebrew transcription shows also very clearly the changes which many sounds have undergone among the Persian Jews partly through dialectic variation of pronunciation, partly through carelessness, or as aids to pronunciation. An example of the transcription of the older period is furnished by the Samuel commentary (about the 14th cent.; see "Z. D. M. G." li. 398). In this work כ, when it designates "b," is frequently written with dagesh; where it stands for "w" it is usually written with rafeh. נ with or without dagesh corresponds to the Persian "g" ("gaf"); to Persian "gh," sometimes also to the ordinary "g"; is "j" ("jim"), sometimes also "tsh." ד represents also "dh"; כ with or without dagesh is "k"; כ is "kh" or "khw," although the latter is sometimes transcribed by ="ṣ" ("ṣad"). פ with rafeh is "f"; without it, "p." ="tsh." For the vowels, the following may be noted: Long "ā" is usually left undesignated, though it is often indicated by the vowel-letter א or by ḳameẓ; short "ă" ("ĕ") is often designated by shewa, sometimes by א; final "ăh" ("ĕh") here and there by א. Short "ǔ" is designated by the vowel-letter ו; short "ǐ" by the vowel-letter י (comp. "Z. D. M. G." liii. 412). As an example of the modern method of transcription may be mentioned the rule that Simeon Hakam lays down.

Various Forms of Gimel.

He uses four forms of gimel: (1) for the usual, hard "g"; (2) נ for "gh"; (3) for "j"; and (4) for "tsh." He has thus avoided entirely the use of for the last sound. In regard to פ, ף, and כ, ד, with or without dagesh, Simeon Ḥakam observes no rule: "We leave this to the reader, since there is a different pronunciation in every city. That of the people of Bokhara is not like that in the cities of Persia or in Balkh. "In Simeon's ritualistic compendium the transcription is less methodized than in the translation of the Bible (see "Keleti Szemle," iii. 15b; "Z. D. M. G."lvi. 759). In one of the Bokhara translations of Abot, printed in Jerusalem, "j" is rendered by ; "tsh," by .

Phonetics, Consonants: Consonantal Changes.

In the Samuel commentary and in Shirwani's dictionary the following important variations in sound from the written language are found, some of which occur also in the later literary productions: "k" is found for "g," and vice versa ( for "tuwānger," for "paigān," are, according to Nöldeke, "older forms which have preserved the original 'k'"); "ḳ" for "gh"; "g" for "ḳ"; "ṭ" for "t," and vice versa; final "t" instead of "d"; "d" for "th"; "d" for "ẓ"; "tsh" for "sh" ( = "shipish"); "j" for "z"; "h" for "ḥ"; "l" for "r" (in a modern publication stands for ); "w" (ו or ) for "b"; "b" for "f"; "m" (before "b") for "n"; "mb" (final) for "m." The changes which, especially among the Bokharian Jews, Arabic derivatives undergo, owing to the permutation of consonants, are such that the word-form often becomes almost unrecognizable (for examples see "Z. D. M. G." liii. 393, lv. 251 et seq., lvi. 746-753); changes due to the transposition of consonants, as the above-cited remark of Simeon Ḥakam shows, are also a peculiarity of the vulgar speech of the Bokharian Jews.


A marked characteristic of Judæo-Persian is the very frequent use of the vowel "u," it often being substituted for other vowels, for "a" ("e")or "i." The fact that in the Samuel commentary "u" (written ו) sometimes takes the place of "ǐ," may perhaps be explained by the supposition that in pronunciation the vowel "i" sounded like "ü," and that this was rendered by ו; hence, conversely, a י is sometimes found for "u" ( = "dushman" ="enemy"). It has already been noted that the Judæo-Persian texts carefully designate (by and ) the vowels "ē" and "ō," which in Persian writing are not distinguished from "î" and "û." Also the suffixes of the first person plural ("-îm," "-îd") are frequently written and (also ). Modern publications and manuscripts write "segol" instead of "ẓere" (see "Z. D. M. G." lii. 199). For short "ĭ," the pronunciation "ē" is also found (the Samuel commentary writes , also , for "sih" = "three"; , but also , for "gird" = "around," "about"). In the transliteration of Arabic words the Judæo-Persian texts of both ancient and modern times indicate the "imalah" of the "a" sound; the Samuel commentary also writes ("ishtew") for the Persian word "shitâb." For the shortening of vowels in Judæo-Persian see Nöldeke, "Litterarisches Centralblatt," 1889, p. 890. The above-mentioned transcription of short vowels by means of "shewa" points to a shortened pronunciation of the vowels.

Etymology: Sign of the Genitive.

The use of the particles "az," "ân" between two substantives to designate the genitive relation is found in the old Samuel commentary as well as in the most modern texts. The original sign of the genitive (the vowel "ĭ"), which was appended to the first substantive ("status constructus"), is attached in these texts to the substantive and to the genitive particle also; thus: = "king of the world." In the oblique case, besides the suffix "-ra," the particle "azmar" (instead of the "mar" used in the older language) is placed before the substantive both in the modern Bible translations ("Z. D. M. G." liv. 558) and in the old Samuel commentary (ib. li. 407). In the latter, "azmar" is also found alone, without the suffix "-ra." The use of the Arabic plural ending "-āt" in Persian words, e.g., "murghāt" ( = "birds"), is a peculiarity of the Tadshiki which has naturally affected the dialect of the Bokharian Jews ("Grundriss der Iranischen Philologie," i., section 2, p. 407). The preservation of the נ in the plurals ("arms"), ("knees") in the Samuel commentary is an ancient usage (Nöldeke, in "Z. D. M. G." li. 671). The form "dudum" (second), in the Book of Daniel, goes back to the Old Persian form of this ordinal ("Grundriss," i., section 2, p. 116).


As regards pronouns, the archaism "ēmā (we) is found. Noteworthy also are "māyān" and "shumāyān" for the first and second persons plural. The attachment of the enclitic pronoun "sh" by means of "i," in the Ezekiel commentary (Salemann), is important.

Conjugation of Verbs.

In the conjugation of the verb the following points are to be noted: The suffix of the first person singular is "-um" instead of "-am"; e.g., in the Samuel commentary; in that on Ezekiel. The suffix of the third person plural, "-and," throughout the Samuel commentary, is shortened to "-an"; Yusuf Yahudi uses both; while Simeon Ḥakam writes the ending sometimes ("-an"), sometimes ("-in"). The second person plural has at times the ending "-ētān." The imperfect plural in the Samuel commentary has the ending "-in" instead of "-īd" (e.g., = "ye wept," instead of "bigiryid"). The apocope of "-ast" to "-as" is frequent (e.g., for "kardast"). In the Samuel commentary and elsewhere the present participle is preferably formed with "-ā." The same commentary also furnishes many examples of the archaic formation of the passive, that is, with the employment of the verb "âmadan," instead of the usual "shudan." Another form of the passive, without an auxiliary verb, is to be found in the Ezekiel commentary: it is a form which had been known only in Middle Persian (Pahlavi), and can now serve as an important testimony to the survival of that form (Salemann, "Zum Mittelpersischen Passive," St. Petersburg, 1900).

Prepositions and Particles.

Of prepositions those should be noted in which the original initial "ă-" has been preserved. The Ezekiel commentary has for "bē"; for "bāz"; for "bar"; for "bē" (Salemann, ib. p. 270). The substantive "täh" ("under part," "ground") is used by the Jews of Bokhara as a preposition meaning "under." Especially remarkable are the particles which are used to-day by the Bokharian Jews, but have not yet been met with elsewhere: (1) The preposition "ḳaṭīr," also "ḳeṭī," meaning "with"; it is used also as a postpositive.Its origin may be traced to an Arabic substantive, "ḳiṭār," "ḳaṭār" (row), which is used also in Persian and Turkish. (2) The particle of comparison "warin," which is always placed after the substantive (probably from the particle "wār," "wārah," which is used only as a suffix to substantives, with the meaning of the Latin "instar"). (3) The adverbial particle "hamtōr" (meaning "at any rate," "certainly," "nevertheless"). (4) The interrogative particle "tshitō" (how?). In reference to these particles see "Z. D. M. G." lvi. pp. 730-739 (for No. 4, ib. li. 552). For the particle of negation with the imperative, Simeon Ḥakam uses "nă" instead of "mă" (e.g., "nakun" instead of "maknun").

Formation of Substantives.

Among the suffixes employed in the formation of substantives, "-ish" is often found in the older texts. In the written language it is used only in the shortened form "-ish," and is appended to the present stem. In Shirwani's dictionary nearly eighty such substantives are given (see Stade's "Zeitschrift," xvi. 231 et seq.; "Grundriss," i., section 1, p. 281). Another abstract ending, "-ih" (), of which many examples occur also in Shirwani, is added to adjectives to form substantives, and corresponds to the Pahlavi abstract ending of nouns "îh" (Stade's "Zeitschrift," xvii. 200, 202; "Z. D. M. G." li. 671). The adjectival suffix "-omand" (instead of "-mand") should also be noted in (in the translation of Isaiah, ed. Lagarde) and (in the Ezekiel commentary); both words mean "terrible."

Among the verbal formations are the numerous causatives ending in "-ânîdan" in the Samuel commentary ("Z. D. M. G." li. 672) and in the Ezekiel commentary (Salemann, l.c. p. 271).


As regards the syntax of Judæo-Persian the most noteworthy feature is the fact that the translations of the Bible follow exactly the syntactical construction of the Hebrew, in order not to lose any detail of the original text. The Hebrew participle is rendered by the participle without regard to tense; the Hebrew article, by the demonstrative "ān"; and the accusative particle , by "mar" (or "azmar"). The infinitive which stands before a finite verb in Hebrew is faithfully rendered by the Persian infinitive; and in the same way the infinitive with a pronominal suffix is literally translated. This tradition of the Persian translators of the Bible has been preserved by their most recent representative, Simeon Ḥakam. His Pentateuch in a way furnishes an ideal interlinear translation; and, in order to emphasize its merits as such, he has carried out the plan of using dots to separate single words or groups of words in the translation which correspond to single words in the text. This of course is characteristic only of translations of the Bible into Judæo-Persian; but it is possible that it may also have influenced other productions. However, Nöldeke has stated that the "Narration of Daniel" (a translation from the Aramaic) is free from the Hebraized syntax of the Bible translations ("Litterarisches Centralblatt," 1884, p. 889).

Pluralis Majestatis.

A peculiarity in the style of a modern literary production of Bokhara is the use of the "pluralis majestatis." In this work, a popular homily, the third person plural is used in speaking of the person represented as acting or speaking; similarly, a person is addressed in the plural of the second person. The singular is used, however, of wicked persons, and God also is always referred to in the singular. The reason for these two exceptions seems plainly to be the feeling that the polite form of the plural is not in place in speaking of God, whereas the respect implied by that form is not deserved by the wicked ("Z. D. M. G." lv. 250, lvi. 758).


The chief importance for Persian philology of the Judæo-Persian texts lies in the surprising wealth of additions to the vocabulary which all of them, without exception, offer. Lagarde has given a number of noteworthy lexical facts in his "Persische Studien aus der Propheten-Uebersetzung," and W. Bacher has also collected important expressions from Shirwani's dictionary and the Samuel commentary, as well as from the most recent Judæo-Persian writer, Simeon Ḥakam (Stade's "Zeitschrift," xvi.; "Z. D. M. G." li., lvi.). A number of words which can be traced back to Middle Persian, in part even to the Avesta language, have been brought to light and incorporated in the Persian dictionary; likewise interesting word-formations and meanings of well-known words which were not to be found elsewhere.

Further interest attaches to the Judæo-Persian texts on account of the large number of Arabic derivatives which they contain. These far exceed the number of Arabic words found in Persian dictionaries. Perhaps the influence of Arabic-speaking Jews, as well as familiarity with the Arabic Bible translations of Saadia, may have contributed to this.

Foreign Words Adopted.

A characteristic of the language of the Jews living in the northern lands where Persian is spoken is the intermixture of Turkish, especially East-Turkish, words. As early a writer as the lexicographer Solomon b. Samuel in the fourteenth century was influenced by Turkish ("Keleti Szemle," i. 27 et seq., 87 et seq.; "Ein Hebräisch-Persisches Wörterb." pp. 10, 27). He even explains a Biblical word (, Job xvi. 15) by a Turkish one having a similar sound (meaning "belly"). The above-mentioned homily contains a comparatively large number of Turkish words ("Z. D. M. G." lv. 255); but the greatest number occur in Simeon Ḥakam's ritualistic compendium (see "Keleti Szemle," lv. 157). The latter work is an interesting example of the fact that in modern times many words from European languages, especially from the Russian, have found their way into the language of the Persian-speaking Jews of Bokhara ("Z. D. M. G." lvi. 753 et seq.). The German word "Jahrzeit," in its ritualistic meaning, has been adopted by them (see "Zeit. für Hebr. Bibl." v. 154).

Admixture of Hebrew and Aramaic.

That which gives a Jewish character to Judæo-Persian is the use of mixed Hebrew and Persian forms, and the close union of the two elements. Combinations of a Hebrew noun with a Persian verb are frequentin the writings of Solomon b. Samuel ("Ein Hebräisch-Persisches Wörterb." pp. 20-22). By means of the Persian suffix "-î" he forms a word of Persian character from a Hebrew word; e.g., "she-lîḥî kunā" = "one who fills the office of prayer-leader" ("sheliaḥ ẓibbur"). He adds the Persian plural ending to a Hebrew word (e.g., ). The same Hebrew-Persian form of expression is found in the Samuel commentary. Simeon Ḥakam's ritualistic compendium offers the greatest number and most varied examples of the mixture of Hebrew with the Persian spoken by the Bokharian Jews of today ("Z. D. M. G." lvi. 755-758). In many of these a Persian (or Arabic) word is joined with its Hebrew equivalent in order to make the meaning more plain. Hebrew substantives are also joined by means of the Persian genitive suffix (; which is added to the first word); by the addition of the Persian ending "-nāk" an adjective is formed from a Hebrew substantive (e.g., = "dangerous"), etc. Aramaic words also are used in Persian, and are similarly combined with Persian ones. This appears to be due to the influence of the Targumim, many words from which were adopted into the common language of the Persian Jews at an early date. Especially is this noticeable in the writings of Solomon b. Samuel, who, for example, regularly expresses the idea of praying by "ẓelûtâ kardan." The Aramaic "shibṭâ" (= Hebr. "shebeṭ")is used throughout to designate a tribe. In the above-mentioned homily the tribe of Dan is called . Simeon Ḥakam in his translation of the Pentateuch everywhere renders or by the Aramaic word, writing it with ס instead of ש. He seems, however, to have kept the pronunciation with "s" for this Aramaic word, on account of the influence of the Arabic "sibṭ."

A curious fact, of interest in the history of language, is the custom of Solomon b. Samuel and of the writer of the Samuel commentary, when writing Hebrew, of using the Hebrew word exactly like the Persian "ăst," and of placing it at the end of the sentence, which is not done in Hebrew. Solomon b. Samuel furnishes another example of the influence of Persian syntax on Hebrew style (see "Z. D. M. G." li. 396; "Ein Hebräisch-Persisches Wörterb." p. 22).

For the importance of Judæo-Persian for Persian philology see, further, Lagarde, "Persische Studien," p. 68; Paul Horn, in Stade's "Zeitschrift," xvii. 203; Carl Salemann, "Khudâidâd," p. ix.; Wilhelm Geiger, in "Grundriss," i., section 2, p. 408.

  • Besides the Bemerkungen of W. Geiger, cited in the article, Salemann, on Middle Persian, in Grundriss der Iranischen Philologie, i., section 1, pp. 259, 269, 281, 291, 319, 332;
  • idem, in Litteraturblatt für Orientalische Philologie, ii. 74-86;
  • Horn, on the literary New Persian, in Grundriss der Iranischen Philologie, i., section 2, pp. 19 et passim, Index, pp. 526 et seq.;
  • Nöldeke, in Litterarisches Central-blatt, 1884, cols. 888-891;
  • idem, in Z. D. M. G. li. 669-676;
  • Ethé, in Litteraturblatt für Orientalische Philologie, i. 186-194.
G. W. B.