11. Mishnaic Hebrew (hbo-x-mishnaic)

Mishnaic Hebrew (MH; in Hebrew ləšôn ḥăḵāmîm, "the language of the Sages") is the Hebrew of the Mishnah, the Tosefta, the halakhic midrashim and the Hebrew passages of the two Talmuds and the classical midrashim, roughly the first centuries of the common era. It is the same language as Biblical Hebrew in its sounds, its roots and most of its morphology, but its verb system, its pronouns, its relative and possessive constructions and much of its vocabulary are different, and in several of those points it is the direct ancestor of Medieval and Modern Hebrew. This chapter describes it as a layer of its own, with the code hbo-x-mishnaic that the research corpora use (resources.tsv, texts.tsv). Mova has no trees and no tools for it yet; section 12 says what would be needed.

Contents

1. What Mishnaic Hebrew is
2. Periods and branches
3. Spoken or learned? The old debate
4. Texts, manuscripts and reading traditions
5. Phonology and orthography
6. Pronouns
7. The relative שֶׁ־ and the possessive שֶׁל
8. The noun
9. The verb: stems
10. The verb: tenses and the participle
11. The infinitive, the imperative and modality
12. Syntax
13. Particles, adverbs, prepositions and conjunctions
14. Vocabulary: Hebrew, Aramaic, Greek, Latin, Persian
15. Worked passages
16. In Mova
17. Sources

1. What Mishnaic Hebrew is

Segal opens his grammar with four periods of Hebrew: Biblical, Mishnaic, Medieval (Rabbinic) and Modern (Segal 1927 §1). The second is the language in which the Tannaim (the teachers up to the compilation of the Mishnah, about 200 CE) taught and in which their teaching was written down; it continues, more and more mixed with Aramaic, in the Hebrew of the Amoraim (3rd–5th centuries) and of the later midrash. Gesenius–Kautzsch call it "New Hebrew" and place it in the Canaanite branch with Biblical Hebrew (GKC §1 and §3 a).

What makes it a layer of its own, and not merely late Biblical Hebrew, is a set of systematic differences (Segal 1927 §5, §§103–104; Fernández 1997 in "wanted"):

2. Periods and branches

Tannaitic and Amoraic. Research distinguishes MH1, the Hebrew of the Tannaim (Mishnah, Tosefta, halakhic midrashim, the tannaitic quotations in the Talmuds), from MH2, the Hebrew of the Amoraim, which is no longer a spoken language and is more strongly influenced by Aramaic (Segal 1927 §§16–21 on the rise and decline of the spoken language; Bar-Asher 1987, cited in Henshke 2020: 25 n. 1). Segal places the end of MH as a spoken idiom at the end of the Tannaitic period (Segal 1927 §§19–21).

Palestinian and Babylonian. Within Rabbinic Hebrew, manuscripts transmitted in the Land of Israel differ from those transmitted in Babylonia in spelling and in forms. A well-known marker is the spelling of אֶלָּא "but rather": written אילא with yod in Babylonian texts, without it in Palestinian ones, reflecting different vowels in the first syllable (Wormser 2020: 157–158, after Yeivin 1976). Wormser tests this criterion on the late midrash of the Tanḥuma–Yelammedenu type and finds that it does not always hold (Wormser 2020, ch. 6 in Heijmans 2020).

Western and eastern within Palestine. Of the three complete manuscripts of the Mishnah that carry the western (Palestinian) tradition, Kaufmann and Parma A are Italian, of about the 11th–12th century, and Cambridge Add.470 is Byzantine, mid-15th century (Henshke 2020: 25–26).

3. Spoken or learned? The old debate

In the 19th century Abraham Geiger described MH as an artificial creation of the schools, comparable to medieval Latin, and Siegfried and Strack held a similar view; others (Graetz, Luzzatto, Jacob Levy, Israel Lévi, Grünbaum, Bacher) held that it was the living speech of Jews in the Second Temple period, and Nöldeke took a middle position: a genuine survival kept up artificially by scholars whose daily language was Aramaic. Segal summarises the debate and argues at length that MH was a vernacular descended from the colloquial Hebrew of the biblical period, with its own development, and that its grammar and vocabulary show the marks of everyday speech: words for trades, tools, food and daily life (Segal 1909: 3–5; Segal 1927 §§6, 13–15).

This position is the standard one today. Heijmans names Segal's article of 1908 and his grammar of 1927 as the beginning of the modern academic study of Rabbinic Hebrew, after which the field moved largely to Israel and into Modern Hebrew (Heijmans 2020: 2). The Bar Kokhba letters (2nd century, written partly in a Hebrew close to MH) are usually cited as direct evidence of its use outside the schools (wanted: an open edition).

4. Texts, manuscripts and reading traditions

Why printed editions mislead. Printed Mishnah editions (Venice 1522 onwards, Vilna/Romm) went through copyists and printers who knew Biblical Hebrew best and often "corrected" MH forms towards biblical ones, or adopted Babylonian spellings. The study of MH since Kutscher therefore works from the good manuscripts: Kaufmann A 50 (Budapest), Parma De Rossi 138 (Parma A), Cambridge Add.470 (Lowe's edition of 1883), and Genizah fragments (Heijmans 2020: 1–2; Henshke 2020: 25–29).

Vocalisation. The Kaufmann and Parma A manuscripts are wholly or largely vocalised, Cambridge mostly not; where Cambridge's scribe did point words, the pointing mixes the Tiberian system with local traditions and is a witness to Byzantine Hebrew (Henshke 2020: 27–56, on phonology, morphology, homographs, ketiv/qere and foreign words). The vocalisation of manuscripts is later than the text and reflects the reading tradition of the community that pointed them.

Oral traditions. Living reading traditions of the Mishnah (Yemenite, Italian, Sephardi, Ashkenazi) keep MH forms that print has lost; Yemenite tradition in particular is used as evidence (Wormser 2020: 157 n. 14; Kantor 2023 and Hornkohl & Khan 2021 for the method of comparing reading traditions; Kantor 2023 is CC BY-NC, cited only).

What we hold. The research corpus has the Mishnah in Hebrew in several editions from Sefaria's export (Mishnah): "Mishnah, ed. Romm, Vilna 1913" and "Torat Emet 357" (both public domain, scope open), and an edition "based on the Kaufmann manuscript, edited by Dan Beʾeri" (scope private); English translations under CC0 (Sefaria Community Translation), CC BY (Kulp's Mishnah Yomit; Silverstein's Bartenura), CC BY-SA (Open Mishnah) and CC BY-NC (William Davidson). The examples in this chapter are quoted from the pointed Torat Emet text; the translations are ours.

5. Phonology and orthography

Gutturals. In MH, and especially in Galilean and later sources, the gutturals are weakened: ʾ, h, ḥ and ʿ are confused in writing in places, and the Talmud reports that some communities did not distinguish them (Segal 1927 §§41–43). The Cambridge manuscript shows guttural features of its own (Henshke 2011, cited in Henshke 2020: 25 n. 2).

Final m > n. Final mem becomes nun in many words and endings: the masculine plural ־ִין beside ־ִים (חַיָּבִין "liable", קוֹרִין "they read", m. Berakhot 1:1), אָדָן for אָדָם in some manuscripts, and the 3rd person plural suffix often ־ָן (מִצְוָתָן "their commandment", m. Berakhot 1:1). Segal treats the plural in ־ִין as old Hebrew colloquial and not only an Aramaism, since Biblical Hebrew already has some ־ִין plurals (Segal 1927 §281; GKC §87 e lists the biblical ones, e.g. חִטִּין Ezek 4:9, מִלִּין Job).

ה and א at word end. Final ה and א interchange in the writing of feminine nouns and Aramaic loans (קְתִדְרָא, אַכְסַנְיָא; Jastrow 1903 s.vv.), reflecting the same final vowel.

Plene spelling. MH writes vowels with matres lectionis much more freely than the Masoretic Bible: ו for o and u even in closed syllables, י for i and often for e (כּוֹתֵב, שִׁילֵּם), and doubled וו / יי for consonantal w and y (Segal 1927 §§38–39). This "full spelling" is the ancestor of the spelling of unpointed Modern Hebrew.

Vowels in loanwords. Greek loanwords in the Mishnah show vowel reduction in open pretonic syllables, as native words do, which shows that they were integrated into the Hebrew sound system (Heijmans 2020, ch. 8: 193–199).

6. Pronouns

6.1 Independent personal pronouns

(Segal 1927, the personal pronoun, §§ under "1. The Personal Pronoun"; GKC §32.)

The loss of אָנֹכִי is one of the sharpest markers: in the Torah אָנֹכִי and אֲנִי both occur often (chapter 5), while MH has only אֲנִי. In Late Biblical Hebrew (Chronicles, Ezra–Nehemiah, Esther, Qoheleth) the proportion already moves towards אֲנִי (chapter 12).

m. Bava Metzia 1:1 זֶה אוֹמֵר אֲנִי מְצָאתִיהָ zeh ʾômēr ʾănî məṣāʾṯîhā — this-one saying I found-it — "One says: I found it."

m. Avot 1:14 (Hillel) אִם אֵין אֲנִי לִי, מִי לִי. וּכְשֶׁאֲנִי לְעַצְמִי, מָה אֲנִי. וְאִם לֹא עַכְשָׁיו, אֵימָתָי ʾim ʾên ʾănî lî, mî lî? ûḵəšeʾănî ləʿaṣmî, māh ʾănî? wəʾim lōʾ ʿaḵšāw, ʾêmāṯāy? — "If I am not for myself, who is for me? And when I am for myself alone, what am I? And if not now, when?"

The short passage has five MH markers: אֵין negating a nominal clause with a pronoun subject, the reflexive with עֶצֶם "self" (לְעַצְמִי "for myself"), the compound conjunction כְּשֶׁ־ "when", the adverb עַכְשָׁיו "now" (BH עַתָּה), and the interrogative אֵימָתַי "when" (BH מָתַי).

6.2 Demonstratives

(Segal 1927, "2. The Demonstrative Pronoun".)

The pronoun זֶה … זֶה "this one … that one" for "one … the other" is typical of legal style: זֶה אוֹמֵר … וְזֶה אוֹמֵר (m. Bava Metzia 1:1). The demonstrative use of אוֹתוֹ, the object marker with a suffix, as "that (same)" (אוֹתוֹ הַיּוֹם "that day") is an MH innovation that survives in Modern Hebrew.

m. Berakhot 1:1 וְלֹא זוֹ בִּלְבַד wəlōʾ zô biləḇaḏ — and-not this only — "and not this alone"

m. Shabbat 7:2 הֲרֵי אֵלּוּ אֲבוֹת מְלָאכוֹת אַרְבָּעִים חָסֵר אֶחָת hărê ʾēllû ʾăḇôṯ məlāʾḵôṯ ʾarbāʿîm ḥāsēr ʾeḥāṯ — behold these fathers-of labours forty lacking one — "These are the main classes of work: forty less one."

6.3 Interrogatives

MH keeps מִי and מָה (מַה before a consonant cluster, often written joined: מַה שֶּׁ־ "what that, whatever"), and adds אֵיזֶה / אֵיזוֹ / אֵילוּ "which", כֵּיצַד "how", מִפְּנֵי מָה "why" (beside לָמָּה), and אֵימָתַי "when" (m. Avot 1:14; m. Berakhot 1:1 מֵאֵימָתַי "from when") (Segal 1927, "5. The Interrogative Pronoun"). The Biblical interrogative הֲ is rare; questions are marked by intonation or by a question word.

m. Berakhot 1:1 מֵאֵימָתַי קוֹרִין אֶת שְׁמַע בְּעַרְבִית mēʾêmāṯay qôrîn ʾeṯ šəmaʿ bəʿarḇîṯ — from-when reading(m.pl) OBJ Shema in-the-evening — "From what time do they recite the Shema in the evening?"

6.4 Pronominal suffixes

The suffixes are the Biblical ones, but the 2nd and 3rd plural feminine forms tend to merge into the masculine, and ־ָן / ־ֵן for "their" spreads (מִצְוָתָן "their obligation", m. Berakhot 1:1). On verbs, object suffixes are much less used than in BH; MH prefers the object marker with a suffix (אוֹתוֹ "him", אוֹתָהּ "her") or ל + suffix (Segal 1927 §§213–216).

7. The relative שֶׁ־ and the possessive שֶׁל

שֶׁ־. The relative and complementiser of MH is the proclitic שֶׁ־ (with dagesh in the following letter), which replaces אֲשֶׁר almost entirely. It is old: Biblical Hebrew has it in the Song of Songs, Qoheleth, Lamentations, some Psalms, and in a few early northern passages (Judg 5:7 שַׁקַּמְתִּי, 6:17, 7:12, 8:26), which is one of the arguments that MH continues a colloquial or regional Hebrew rather than inventing (GKC §36; Segal 1927, "3. Relative Pronoun", §§14–15). In the Torah, שֶׁ־ is virtually absent: the OSHB morphology analyses only one form as containing it, the disputed בְּשַׁגָּם of GEN 6:3 (counted 2026-10-17; chapter 12 gives the counts book by book); the Torah's relative is אֲשֶׁר (1 911 relatives in MACULA's Type=relative, 2026-10-17; chapter 5).

שֶׁ־ builds a family of compound conjunctions: כְּשֶׁ־ "when", מִשֶּׁ־ "from the time that", עַד שֶׁ־ "until", לִפְנֵי שֶׁ־ "before", מִפְּנֵי שֶׁ־ "because", בִּשְׁבִיל שֶׁ־ "so that", אַף עַל פִּי שֶׁ־ "although", כְּדֵי שֶׁ־ "in order that".

m. Berakhot 1:1 מִשָּׁעָה שֶׁהַכֹּהֲנִים נִכְנָסִים לֶאֱכֹל בִּתְרוּמָתָן miššāʿâ šehakkōhănîm niḵnāsîm leʾĕḵōl biṯrûmāṯān — from-the-hour that-the-priests entering to-eat in-their-terumah — "from the hour when the priests go in to eat their terumah"

m. Berakhot 1:1 עַד שֶׁיַּעֲלֶה עַמּוּד הַשָּׁחַר ʿaḏ šeyyaʿăleh ʿammûḏ haššaḥar — until that-rises pillar-of the-dawn — "until the dawn comes up"

שֶׁל. Possession is expressed analytically by שֶׁל (שֶׁ + לְ, "which is to"), as a word of its own or with suffixes: שֶׁלִּי "mine", שֶׁלְּךָ "yours", שֶׁלּוֹ "his". The construct chain remains, but mostly in fixed phrases and compounds (בֵּית הַמִּשְׁתֶּה "banquet house", עַמּוּד הַשַּׁחַר), while free possession uses שֶׁל, or the anticipatory construction "his-X of-the-man" (בֵּיתוֹ שֶׁל אָדָם), which follows Aramaic (Segal 1927, "4. The Possessive Pronoun").

m. Bava Metzia 1:1 זֶה אוֹמֵר כֻּלָּהּ שֶׁלִּי zeh ʾômēr kullāh šellî — this-one saying all-of-it of-mine — "One says: It is all mine."

In Biblical Hebrew the same thought would be a verbless clause with לְ: לִי כֻלָּהּ, or a construct with a suffix.

8. The noun

Plural endings. Besides ־ִים and ־וֹת, the masculine plural ־ִין is frequent (חַיָּבִין, קוֹרִין, חוּטִין "threads", m. Shabbat 7:2). Many nouns change the gender of their plural ending relative to BH (BH שָׁנִים "years", MH שָׁנִים and later שָׁנוֹת; BH מָקוֹם → מְקוֹמוֹת both), and some nouns have a plural in MH that they lack in BH, or the reverse (Segal 1927 §§281–292). Foreign nouns form plurals in ־ִין, ־וֹת, ־יוֹת (Segal 1927 §286).

The dual is restricted to paired body parts, numerals and time words, much as in BH; it is not productive (Segal 1927 §293).

New noun patterns (mishqalim) (Segal 1927 §§220–293):

The qəṭîlâ verbal noun is one of MH's chief resources: it gives every Qal verb a regular action noun, a function that BH spread over many irregular patterns and the infinitive.

m. Avot 1:2 (Simeon the Just) עַל שְׁלשָׁה דְבָרִים הָעוֹלָם עוֹמֵד, עַל הַתּוֹרָה וְעַל הָעֲבוֹדָה וְעַל גְּמִילוּת חֲסָדִים ʿal šəlōšâ ḏəḇārîm hāʿôlām ʿômēḏ: ʿal hattôrâ wəʿal hāʿăḇôḏâ wəʿal gəmîlûṯ ḥăsāḏîm — on three things the-world standing: on the-Torah and-on the-service and-on bestowing-of kindnesses — "The world stands on three things: on the Torah, on the (Temple) service, and on deeds of loving-kindness."

Note עוֹלָם "world": in the Torah עוֹלָם is "long time, eternity" (לְעוֹלָם "for ever"), and the spatial sense "world" is post-biblical (BDB s.v. עוֹלָם; Jastrow s.v.).

Article and state. The article is ה as in BH; it is kept after prepositions as in BH (בַּתּוֹרָה). MH increasingly uses the article on the first member of a construct-like phrase in compounds that have become single words (הַבֵּית הַכְּנֶסֶת is late; MH has בֵּית הַכְּנֶסֶת).

9. The verb: stems

The seven Biblical stems remain, with these shifts (Segal 1927 §§103–212):

• Nithpaʿal (נִתְקַטֵּל) replaces the Hithpael in the perfect: נִתְקַבֵּל "was received", נִתְכַּוֵּן "intended". It combines the נ of the Niphal with the ת of the Hithpael and usually has a passive or reflexive sense (Segal 1927 §§132–134). In the Torah, MACULA marks one form as HebBinyan=NITHPAEL (2026-10-17), against 179 Hithpael.
• Pual finite forms are rare; the Pual survives mainly as a participle with adjectival meaning (מְתֻקָּן "prepared", m. Avot 3:16; מְיֻחָד "special") (Segal 1927 §§126–131).
• Hophal likewise survives mostly in the participle (מוּתָּר "permitted", מוּקְצֶה "set aside").
• Piel for Qal: some verbs move from Qal to Piel (BH שָׁלַם → MH שִׁלֵּם "pay"; BH בָּקַשׁ is already Piel), and the Piel takes over causative functions in some roots (Segal 1927 §124).
• New intensive stems by reduplication: Pilpel-like forms (גִּלְגֵּל "roll", טִלְטֵל "move about") (Segal 1927 §§108–109, §195).
• III-ʾ and III-y verbs merge: III-aleph verbs inflect like III-he (קָרָא → קָרִינוּ "we read" in m. Berakhot 1:1, BH קָרָאנוּ), and the infinitive לִקְרוֹת follows the III-he pattern (Segal 1927 §204).
• Niphal infinitive loses its ה: לִכָּנֵס for BH לְהִכָּנֵס (Segal 1927 §115).

m. Berakhot 1:1 לֹא קָרִינוּ אֶת שְׁמַע lōʾ qārînû ʾeṯ šəmaʿ — not we-read OBJ Shema — "We have not recited the Shema."

The form קָרִינוּ, with î before the suffix, is a III-he form of a III-aleph root (BH would be קָרָאנוּ).

m. Berakhot 2:1 אִם כִּוֵּן לִבּוֹ, יָצָא ʾim kiwwēn libbô, yāṣāʾ — if he-directed his-heart, he-went-out — "If he directed his heart, he has fulfilled (the obligation)."

The idiom יָצָא "went out (of the obligation)" = "fulfilled his duty" is a technical sense of a common verb, typical of MH legal vocabulary.

10. The verb: tenses and the participle

This is where MH differs most from the Biblical prose of the Torah. The consecutive forms (wayyiqtol and weqatal) are gone (Segal 1927 §157 and §§306–323); the language has a three-way system much closer to a tense system:

The participle as present. The participle has become the ordinary present tense and the tense of the halakhic rule: קוֹרִין "they recite" = "one recites", אוֹמֵר "he says", נוֹטֵל "he takes" (m. Bava Metzia 1:1). It keeps its nominal origin: it is negated with אֵין, it agrees in gender and number but has no person, and a subject pronoun is needed (Segal 1927 §§324–343).

הָיָה + participle. The periphrastic past habitual: הוּא הָיָה אוֹמֵר "he used to say", the formula that introduces many sayings in Avot (m. Avot 1:2, 1:14, 2:5, 3:16) (Segal 1927 §§324, 343).

m. Avot 1:2 שִׁמְעוֹן הַצַּדִּיק הָיָה מִשְּׁיָרֵי כְנֶסֶת הַגְּדוֹלָה. הוּא הָיָה אוֹמֵר šimʿôn haṣṣaddîq hāyâ miššəyārê ḵəneseṯ haggəḏôlâ. hûʾ hāyâ ʾômēr — Simeon the-righteous was of-remnants-of assembly the-great. he was saying — "Simeon the Just was one of the last of the Great Assembly. He used to say: …"

qatal as narrative past. Where the Torah strings wayyiqtol forms, MH strings qatal forms joined by וְ:

m. Avot 1:1 משֶׁה קִבֵּל תּוֹרָה מִסִּינַי, וּמְסָרָהּ לִיהוֹשֻׁעַ mōšeh qibbēl tôrâ missînay, ûməsārāh lîhôšuaʿ — Moses received Torah from-Sinai, and-handed-it to-Joshua — "Moses received the Torah from Sinai and handed it on to Joshua."

Compare the Torah's narrative of the same kind of event, a wayyiqtol chain: וַיִּכְתֹּב מֹשֶׁה אֵת כָּל־דִּבְרֵי יְהוָה וַיַּשְׁכֵּם בַּבֹּקֶר וַיִּבֶן מִזְבֵּחַ "And Moses wrote all the words of the LORD, and rose up early in the morning, and builded an altar" (EXO.24.4, wlc:EXO:648; JPS 1917). In MH the וּ before מְסָרָהּ is a plain conjunction with a qatal, not the consecutive. The verb קִבֵּל "receive" is itself typical of MH and of late BH (Esther, Chronicles, Ezra), absent from the Torah, which uses לָקַח.

The yiqtol. It expresses the future, the modal (should, may), the jussive and purpose after שֶׁ־: עַד שֶׁיַּעֲלֶה "until (the dawn) rises", יִשָּׁבַע "he shall swear" and וְיַחֲלֹקוּ "and they shall divide" (m. Bava Metzia 1:1) (Segal 1927 §§306–323).

11. The infinitive, the imperative and modality

The infinitive construct is used almost only with ל: לֶאֱכֹל "to eat", לִקְרוֹת "to read", לִתְפֹּר "to sew" (m. Shabbat 7:2 עַל מְנָת לִתְפֹּר "in order to sew"). The Biblical temporal constructions with בְּ and כְּ + infinitive (בְּבֹאוֹ "when he came") give way to כְּשֶׁ־ + finite verb (Segal 1927 §§344–350). The infinitive with ל also expresses obligation: חַיָּבִין אַתֶּם לִקְרוֹת "you are obliged to read" (m. Berakhot 1:1), and with עָתִיד "about to, destined to" the future (Segal 1927 §348).

The infinitive absolute has almost disappeared as a verb form. Its BH functions (emphasis, command) are not continued; a few forms survive as nouns or adverbs (Segal 1927 §104).

The verbal noun (qəṭîlâ, section 8) takes over many functions of the BH infinitive as a noun: הֶקְטֵר חֲלָבִים וְאֵבָרִים "the burning of fats and limbs" (m. Berakhot 1:1).

Imperatives and exhortations. The imperative is as in BH, and the Hithpael imperative הִשְׁתַּדֵּל "strive" is common in sayings (m. Avot 2:5 "where there are no men, strive to be a man"). The cohortative and jussive forms have merged with the yiqtol; נָא is very rare.

m. Avot 1:1 הֱווּ מְתוּנִים בַּדִּין, וְהַעֲמִידוּ תַלְמִידִים הַרְבֵּה, וַעֲשׂוּ סְיָג לַתּוֹרָה hĕwû məṯûnîm baddîn, wəhaʿămîḏû ṯalmîḏîm harbēh, waʿăśû səyāḡ lattôrâ — be deliberate in-the-judgment, and-raise-up disciples many, and-make fence for-the-Torah — "Be deliberate in judgment, raise up many disciples, and make a fence around the Torah."

הֱווּ "be!" is the imperative of הָוָה, the Aramaic-type form of "to be" that MH uses in the imperative beside הָיָה. The adverb הַרְבֵּה "much, many" after its noun is an old Hiphil infinitive absolute fossilised as an adverb.

12. Syntax

Word order. With the loss of wayyiqtol, which forced the verb to the front, MH clauses lean to subject–verb order, especially with participles (הָעוֹלָם עוֹמֵד "the world stands", m. Avot 1:2; הַיָּד כּוֹתֶבֶת "the hand writes", m. Avot 3:16). Verb-first order remains common with qatal and in fixed formulas.

Negation. לֹא negates finite verbs; אֵין negates the participle and nominal clauses (אֵין בּוּר יְרֵא חֵטְא "an ignorant man does not fear sin", m. Avot 2:5; אֵין לוֹ "he has not"). The prohibitive אַל + yiqtol survives in sayings (Segal 1927 §§339–340).

Conditionals. אִם with qatal or participle in the protasis (אִם כִּוֵּן לִבּוֹ יָצָא, m. Berakhot 2:1); the negative condition וְאִם לָאו "and if not" (לָאו = לֹא + Aramaic-type emphatic form). The unreal condition uses אִלּוּ.

Casuistic law. MH legal style opens with a participle with the article as a case heading: הַקּוֹרֵא מִכָּאן וְאֵילָךְ לֹא הִפְסִיד "one who reads from then on has not lost" (m. Berakhot 1:2); הַמּוֹצִיא מֵרְשׁוּת לִרְשׁוּת "the one who carries out from domain to domain" (m. Shabbat 7:2). Compare the Torah's כִּי + yiqtol / אִישׁ כִּי case headings (chapter 8).

The dispute formula. דִּבְרֵי רַבִּי X. וַחֲכָמִים אוֹמְרִים … "the words of Rabbi X. But the Sages say …" — a recurring discourse frame that marks a minority opinion and the majority's. Conversation in the Mishnah has its own pattern of adjacency pairs (question–answer, assertion–counter-assertion) and argumentative steps (Shemesh-Raiskin 2020, ch. 3 in Heijmans 2020: 57–96).

Agreement and the "his-X of Y" construction. Possession with anticipatory suffix, בְּנוֹ שֶׁל רַבִּי "his son of Rabbi = Rabbi's son", is frequent (Segal 1927 §§ on the possessive).

13. Particles, adverbs, prepositions and conjunctions

MH loses many BH particles and gains new ones, several from Aramaic (Segal 1927 §§294–305):

m. Berakhot 1:2 שֶׁכֵּן דֶּרֶךְ בְּנֵי מְלָכִים לַעֲמֹד בְּשָׁלֹשׁ שָׁעוֹת šekkēn dereḵ bənê məlāḵîm laʿămōḏ bəšālōš šāʿôṯ — for-so way-of sons-of kings to-rise at-three hours — "for such is the way of princes, to rise at the third hour"

Note also שָׁעָה "hour", rare in BH (only in the Aramaic of Daniel) and the ordinary MH word.

14. Vocabulary: Hebrew, Aramaic, Greek, Latin, Persian

Segal argues that the bulk of MH vocabulary is Hebrew: much of it biblical, some old Hebrew not attested in the Bible but with cognates elsewhere in Semitic, and some biblical words with new meanings (Segal 1927 §§9–12, §§93–102). On top of that layer come loans:

Aramaic. The heaviest layer, in vocabulary and in grammar (the -āʾ ending, the possessive with anticipatory suffix, words like אַבָּא "father", אִמָּא "mother", צָרִיךְ "must, needs", לְהַלָּן "below, further"). Breuer shows how the forms רַבָּה / רָבָא and אַבָּא / אָבָא differ in spelling, pronunciation and meaning between the Aramaic dialects and MH, and argues that MH stands as a bridge between Western and Eastern Aramaic usage (Breuer 2020, ch. 1 in Heijmans 2020: 5–24). Stadel discusses how to identify Aramaic used by the Tannaim themselves (Stadel 2020, ch. 4: 97–118).

Greek. Hundreds of words, especially in law, trade, administration and daily life (Jastrow 1903 s.vv.):

Latin. Usually through Greek or Aramaic: טַבְלָא "tablet" (tabula; Jastrow s.v. טַבְלָא II), לִבְלָר "scribe" (libellarius), קוּלְמוֹס "pen" (calamus, Greek κάλαμος).

Persian. Already in late BH and continued: פַּרְדֵּס "orchard" (BH Neh 2:8, Song 4:13, Qoh 2:5; Jastrow s.v.; in later mystical tradition "Paradise"), גִּזְבָּר "treasurer", דָּת "law" (BH Esther).

New meanings of old words: עוֹלָם "eternity" → "world"; יָצָא "go out" → "fulfil (a duty)"; מְלָאכָה "work" → the technical category of work forbidden on the Sabbath; קָרָא "call" → "read, recite" (already BH in part).

15. Worked passages

15.1 m. Avot 3:16 (Rabbi Akiva), the image of the shop

הַכֹּל נָתוּן בְּעֵרָבוֹן, וּמְצוּדָה פְרוּסָה עַל כָּל הַחַיִּים. הַחֲנוּת פְּתוּחָה, וְהַחֶנְוָנִי מֵקִיף, וְהַפִּנְקָס פָּתוּחַ, וְהַיָּד כּוֹתֶבֶת, וְכָל הָרוֹצֶה לִלְווֹת יָבֹא וְיִלְוֶה

"Everything is given on pledge, and a net is spread over all the living. The shop is open, the shopkeeper gives credit, the ledger is open and the hand writes, and whoever wishes to borrow may come and borrow." Every main clause is a participle predicate after its subject: the MH present-tense system in one breath. There is not one wayyiqtol or weqatal.

15.2 m. Bava Metzia 1:1, a legal case

שְׁנַיִם אוֹחֲזִין בְּטַלִּית, זֶה אוֹמֵר אֲנִי מְצָאתִיהָ וְזֶה אוֹמֵר אֲנִי מְצָאתִיהָ … זֶה יִשָּׁבַע שֶׁאֵין לוֹ בָהּ פָּחוֹת מֵחֶצְיָהּ … וְיַחֲלֹקוּ

"Two are holding a garment. One says, I found it, and the other says, I found it … Each shall swear that he has no less than half of it in it … and they shall divide it."

• שְׁנַיִם אוֹחֲזִין: numeral as subject + participle in ־ִין: the case is posed in the present.
• זֶה … וְזֶה: "the one … the other".
• מְצָאתִיהָ: qatal 1 sg + object suffix 3 f.sg (BH-type suffix on a verb, still found).
• יִשָּׁבַע: Niphal yiqtol, modal "shall swear".
• שֶׁאֵין לוֹ: complementiser שֶׁ־ + negative existential אֵין + ל + suffix: "that he does not have".
• פָּחוֹת מֵ־: "less than" — comparison with מִן as in BH, with the MH adjective-adverb פָּחוֹת.

15.3 Sefer Yetzirah 1:1–1:2

Sefer Yetzirah (chapter 1 and docs/sefer-yetzirah/) is written in a terse Hebrew that shares many MH features: participles, number words used as nouns, verbal nouns, and almost no consecutive tenses. Its opening, in the standard recension held in docs/sefer-yetzirah/text/he_standard_warsaw1884_sefaria.txt (unpointed there; the points below are ours):

בשלשים ושתים נתיבות פליאות חכמה חקק יה … בִּשְׁלֹשִׁים וּשְׁתַּיִם נְתִיבוֹת פְּלִיאוֹת חָכְמָה חָקַק יָהּ …

bišlōšîm ûštayim nəṯîḇôṯ pəliʾôṯ ḥoḵmâ ḥāqaq yāh — "With thirty-two wondrous paths of wisdom Yah engraved …". The verb chain that follows uses qatal (חָקַק, בָּרָא) and lists, not wayyiqtol; the next mishnah counts the "ten sefirot without anything (בְּלִימָה)" and the "twenty-two letters of foundation" with the MH-like construction of a number followed by a noun in apposition. The yetzirah crate reads the book's letter tables, not its grammar (chapter 13).

16. In Mova

What exists (2026-10-17):

• No Mishnaic trees. tools/macula reads only MACULA Hebrew, i.e. the Hebrew Bible. There is no UD treebank of Mishnaic Hebrew in UD 2.18 (wanted).
• Texts in the research corpus: the Mishnah in public-domain Hebrew editions (Vilna 1913; Torat Emet) and in CC0 / CC BY English translations, under sefaria; the Kaufmann-based edition is scope private. The Hebrew text has vowel points (Torat Emet) but no morphology.
• Lexicons: Jastrow 1903 (public domain; Sefaria digital text Jastrow_Dictionary.jsonl and the Internet Archive scans) and Levy 1876–1889 (German, public domain) in dictionaries. Jastrow's entries carry references to Mishnah, Tosefta, Talmud and midrash, which could seed a Mishnaic dictionary in the shape of torah-en.tsv.
• Sefer Yetzirah: the text, its two recensions and the yetzirah automata (chapter 13).
• Rabbinic Hebrew NLP: BEREL, a BERT model of Rabbinic Hebrew (Shmidman et al. 2022, hbo-2022-shmidman-introducing-berel-*), shows that the domain is distinct from Modern Hebrew for models; Mova would use such a model at most as a teacher, never at answer time.

Gaps (not done; each a new direction that needs the operator's word):

1. A Mishnaic morphological analyser in Rust, starting from the Biblical one proposed in chapter 13 and adding the MH rules of this chapter (participle-as-present, Nithpaʿal, III-ʾ = III-y, ־ִין plurals, שֶׁ־ and שֶׁל as separate tokens).
2. A small gold set of Mishnah trees (e.g. Avot 1, Berakhot 1, Bava Metzia 1) written by hand or by a teacher and checked, under our licence, to measure against.
3. The macula prag rules transferred: MH has no wayyiqtol, so "narrate" must be read from qatal chains; the halakhic participle with the article is a rule, not a narration; הָיָה אוֹמֵר is a habitual frame of speech.
4. A Jastrow-based dictionary of the Mishnah's words in the shape of torah-en.tsv.

17. Sources

• Segal, M. H. 1909. Mišnaic Hebrew and its Relation to Biblical Hebrew and to Aramaic. Oxford (from JQR 20, 1908). PD-USA only; cited, not copied. segal-misnaic-hebrew-1909
• Segal, M. H. 1927. A Grammar of Mishnaic Hebrew. Oxford: Clarendon. PD-USA only; cited by §. segal-grammar-mishnaic-1927
• Heijmans, S. (ed.) 2020. Studies in Rabbinic Hebrew. Cambridge: Open Book Publishers. CC BY 4.0. doi.org/10.11647/OBP.0164 our card ; heijmans-studies-rabbinic-hebrew-2020. Chapters cited: Heijmans, Introduction (1–3); Breuer, "Rabba and Rava, ʾAbba and ʾAva" (5–24); Henshke, "The Vocalisation of MS Cambridge of the Mishnah" (25–56); Shemesh-Raiskin, "Adjacency Pairs and Argumentative Steps …" (57–96); Stadel, "Tannaitic Aramaic" (97–118); Wormser, "The Distinction between Branches of Rabbinic Hebrew …" (155–168); Heijmans, "Vowel Reduction in Greek Loanwords in the Mishnah" (193–202).
• Hornkohl, A. D., Khan, G. (eds.) 2021. New Perspectives in Biblical and Rabbinic Hebrew. Cambridge: OBP. CC BY 4.0. hornkohl-khan-new-perspectives-2021
• Gesenius, W., Kautzsch, E., Cowley, A. E. 1910. Gesenius' Hebrew Grammar. Oxford. Public domain. §§1–3 (history), §32 (pronouns), §36 (שֶׁ־), §87 (plural endings). Wikisource text gesenius-kautzsch-cowley-1910-wikisource
• Jastrow, M. 1903. A Dictionary of the Targumim, the Talmud Babli and Yerushalmi, and the Midrashic Literature. Public domain. Jastrow_Dictionary.jsonl; scans archive.org/details/dictionaryoftarg01jastuoft
• Levy, J. 1876–1889. Neuhebräisches und chaldäisches Wörterbuch über die Talmudim und Midraschim. Public domain. levy-neuhebr-1876-1889
• Brown, Driver, Briggs 1906 (BDB), s.v. עוֹלָם. Public domain.
• The Mishnah: Torat Emet 357 and Romm (Vilna 1913) Hebrew texts, public domain, via Sefaria's export, Mishnah. Passages cited: Berakhot 1:1, 1:2, 2:1; Shabbat 7:2; Bava Metzia 1:1; Avot 1:1, 1:2, 1:14, 2:5, 3:16. Translations ours.
• Kantor, B. 2023. A Linguistic Classification of the Reading Traditions of Biblical Hebrew. OBP; CC BY-NC (cited only). kantor-classification-reading-traditions-2023
• Shmidman, A. et al. 2022. "Introducing BEREL: BERT Embeddings for Rabbinic-Encoded Language." hbo-2022-shmidman-introducing-berel-*
• Cook, J. A. 2017. "Language contact and the genesis of Mishnaic Hebrew." hbo-2017-cook-language-contact-genesis-mishnaic-hebrew
• Mova: docs/sefer-yetzirah/, yetzirah, macula.

Wanted (not held, or held under terms that forbid use): Pérez Fernández, An Introductory Grammar of Rabbinic Hebrew (1997); Bar-Asher, Studies in Mishnaic Hebrew (2009, Hebrew); Haneman, A Morphology of Mishnaic Hebrew according to the Parma Manuscript (1980, Hebrew); Kutscher, A History of the Hebrew Language (1982); an open edition of the Bar Kokhba letters; an open, morphologically tagged Mishnah (the Kaufmann manuscript with morphology); a UD treebank of Mishnaic Hebrew.

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/languages/hbo/11-mishnaic/