The lexicon of Jewish Babylonian Aramaic

The JBA lexicon has three layers: the common Aramaic core (shared with the targums), an eastern layer shared with Syriac and Mandaic (and partly with Akkadian, the old language of the land), and the loanwords of a Jewish community living in the Sasanian empire — Persian above all, with Greek and Latin words that came mostly through Palestinian rabbinic tradition. Hebrew is everywhere, as the language of the sources being discussed.

Sources: Jastrow 1903 (public domain; cited s.v.; 74 of its entries are marked "Pers." in the definition, our count 2026-10-17); Sokoloff 2002a, A Dictionary of Jewish Babylonian Aramaic (in copyright; wanted); Kaufman 1974, The Akkadian Influences on Aramaic; Shaked's papers on Iranian loanwords in JBA; Krauss 1898–1899, Griechische und lateinische Lehnwörter im Talmud, Midrasch und Targum (public domain; on disk, see Sources); Kohut 1878–1892, Aruch Completum (public domain; on disk). Counts: Wikisource Bavli (CC BY-SA, measurement only), unpointed tokens, 2026-10-17.

1. Core vocabulary

Frequent JBA content words and function words (beyond those in The technical language of the sugya):

2. Words that tell JBA from JPA and the targum

The pairs below are the most useful single-word cues for telling the dialects apart (Sokoloff 2002a, 2002b; Jastrow s.vv.; Dalman 1905):

3. Persian loanwords

JBA borrowed hundreds of Middle Persian (Pahlavi) words: administration, law, clothing, food, games, plants. A sample with Jastrow's etymologies and Bavli citations (Jastrow s.vv.; the identifications are Jastrow's and his sources', some revised since by Shaked and Sokoloff):

Checked 2026-10-18 against Kohut's Aruch Completum (cited by volume and PDF page; the Hebrew text of the OCR is readable): אַשְׁמְדַאי (vol. 1, p. 407: a compound of two Persian words, the first aēšma), בּוּסְתָּנָא (vol. 2, p. 143: Persian bustān), דַּנְקָא (vol. 3, p. 106: from Persian, "a sixth"), כּוּתָּח (vol. 4, p. 255: a Persian word), פּוּזְמְקֵי (vol. 6, p. 317: a Persian word), פַּרְדַּכְתְּ (vol. 6, p. 424: from Persian, "free from work"): all agree. For גּוּשְׁפַּנְקָא Kohut already gives a Persian etymology (vol. 2, p. 351: a ring for sealing, from Persian words, after Fleischer), against Jastrow's Aramaic guess. The other rows were not found or not legible in Kohut (unchecked).

Two remarks:

• Jastrow's etymologies are a century old. For גּוּשְׁפַּנְקָא and הַרְמָנָא Jastrow gives Aramaic etymologies, while later Iranists derive both from Persian (Shaked; Sokoloff 2002a s.vv.); for גּוּשְׁפַּנְקָא so did Kohut before Jastrow (above). Mova's lexicon should carry the etymology with its source and date, not as a fact.
• Persian words are a marker of tmr against jpa, where Greek and Latin loans dominate (The lexicon of Jewish Palestinian Aramaic).

4. Akkadian substrate

The Jews of Babylonia lived where Akkadian had been spoken; JBA (like Syriac and Mandaic) keeps Akkadian words that never reached western Aramaic (Kaufman 1974 is the standard study). Examples commonly cited:

(This table follows Kaufman 1974 in general; each item should be checked there before it is entered in a lexicon. Not every word that looks old is Akkadian: מִשְׁחָא "oil", for example, is native Aramaic (√משׁח).)

5. Greek and Latin

Greek and Latin words in the Bavli are fewer than in the Yerushalmi and the midrash, and many entered through the Mishnah and Palestinian traditions. Examples (Jastrow s.vv.):

Checked 2026-10-18 against Krauss 1899, vol. II (the dictionary; printed page = PDF page − 412 in the on-disk scan): אַכְסַנְיָא ξενία (p. 48), אֲוִירָא ἀήρ (p. 17), אַפִּיטְרוֹפּוֹס ἐπίτροπος (p. 104), דְּיָיתֵיקֵי διαθήκη (p. 197), דִּינָרָא δηνάριον / denarius (p. 207), טְרַקְלִין τρικλίνιον / triclinium (p. 274), לִיסְטִים ληστής (p. 315), סַנְהֶדְרִין συνέδριον (p. 401), סָנֵיגוֹר συνήγορος beside קָטֵיגוֹר (p. 404), קוּלְמוֹס calamus (p. 506): all agree. אוּשְׁפִּיזָא: Kohut (vol. 1, p. 410) takes it from Latin (the guest אושפיז, the inn אושפיזא), which supports hospitium / hospes. Disagreement on פְּרוֹזְבּוּל: Krauss (pp. 482–483) does not accept πρὸς βουλῇ βουλευτῶν; his entry and its bracketed additions weigh προβολή / παράβολον and προσβολή "handing over" (after Schürer) instead. The table keeps the traditional derivation; both are given here.

6. Hebrew in the Bavli, and the Bavli in Hebrew

• Hebrew inside JBA: whole sources (Mishnah, baraitot, verses), and also Hebrew words and phrases used as Aramaic vocabulary (מִצְוָה, תּוֹרָה, קַל וָחוֹמֶר, שְׁבוּעָה), often with Aramaic endings (מִצְוְתָא).
• JBA inside Hebrew: through the study of the Bavli, Aramaic words became part of rabbinic Hebrew and then of Modern Hebrew: אַגַּב "by the way", בְּדִיעֲבַד "after the fact", לְכַתְּחִילָּה "in the first place", אַדְּרַבָּה "on the contrary", מִסְתַּבֵּר "it stands to reason", תֵּיקוּ "a draw" (popular etymology as an acronym), דַּוְקָא "precisely; specifically", אַבָּא, סַבָּא, שְׁמַע מִינַּהּ … (Kutscher 1982, A History of the Hebrew Language, on Aramaic in Hebrew). A Modern Hebrew model meets these every day; a tmr lexicon links them back.

The Elamur team

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/languages/tmr/lexicon/