BEMIDBAR RABBAH:

By: Marcus Jastrow, J. Theodor

The Midrash commentary upon Numbers, called in the editio princeps of Constantinople (1512) "Bemidbar Sinai Rabbah," and so cited frequently by Naḥmanides and others. It is the latest component of the "Rabbot" collection upon the Pentateuch, and as such was unknown to 'Aruk, Rashi, and Yalḳuṭ. It consists of two parts, which are of different origin and extent. The first portion, sections i.-xiv. (ed. Venice, 1545, parashah "Bemidbar," fol. 135a to 145c; parashah "Naso," fol. 145c to 178b)—almost three-quarters of the whole work—contains a late haggadic commentary upon Num. i.-vii.; the second part, sections xv.-xxxiii. (fol. 178b to 194d, ed. Venice), reproduces the Midrash Tanḥuma from Num. viii. almost word for word. The consideration of the second portion will therefore be found in the article Tanḥuma. There also the form of the homilies of the Tanḥuma Midrash, their halakic introductions, their proems, their exposition covering in each case only a few verses of the text, their regular formulas of conclusion, are more appropriately considered. Suffice it to state here that the second portion of Bemidbar Rabbah follows closely those readings of the Tanḥuma which are familiar from the oldest edition (compare Buber's Introduction to his edition of the Tanḥuma, pp. 38a et seq.); and that M. Beneviste, in the preface to "Ot Emet" (Salonica, 1565), drew attention as early as 1565 to the fact that Tanḥuma and Bemidbar Rabbah are almost identical from the section "Beha'aloteka" onward. Buber gives a list of the variations between the two (ib. 39a et seq.). The passages drawn from the Pesiḳta Rabbati (Zunz, "G. V." p. 259, note) are to be found exclusively in the first or later part of this Midrash. This is true also, with the exception of the interpretation of the numerical value of the Hebrew word for "fringes," of the other passages pointed out by Zunz as originating with later, and notably French, rabbis. This numerical interpretation just mentioned forms a part of a passage, also otherwise remarkable, at the end of the section "Ḳoraḥ" (xviii. 21), which, taken from Bemidbar Rabbah, was interpolated in the editio princeps of the Tanḥuma as early as 1522 (Constantinople), but is absent from all the manuscripts. Another long passage, ib. 22, which belongs to the beginning of "Ḥuḳḳat," as in Tanḥuma, is erroneously appended in the editions to the same section, "Ḳoraḥ."

The halakic exordium at the beginning of the second part, on Num. viii. 1, is cut down to its concluding passage; the Paris MS., Cod. No. 150, and a MS. in the possession of Epstein, contain the exordiumcomplete with its customary formula, , as usual in Tanḥuma, which formula reappears throughout this portion of Bemidbar Rabbah, while in the editions, in section xv., Nos. 11, 17; sec. xvi., Nos. 1, 26; xvii. No. 1; sec. xx. Nos. 21, 22; sec. xxi. No. 2; xxii. No. 7; xxiii. Nos. 1, 7, the formula is changed to . In section xxi., beginning, and Nos. 16 and 23, the exordiums of Tanḥuma, Pinḥas, Nos. 12 and 15 (Nos. 11, and 13, Buber) have been omitted, as also in section xxii., beginning, Nos. 1 and 2 (compare Buber, pp. 47b, 51b).

The portions of Numbers to which there are Tanḥuma homilies in this portion of Bemidbar Rabbah are intended for public worship according to the divisions of the cycle of the sedarim and the Pesiḳta. The well-known variations existing in the division into sedarim probably explain why some of the old sedarim, as Num. xi. 23 (xvii. 16), xviii. 25, xxiii. 10, xxviii. 26, xxxi. 25, are here without these homilies, while such are appended—or at least fragments of such—to the passages, Num. viii. 5, xiv. 26, xv. 37, xx. 7-13, xxiv. 3. In an article in the "Monatsschrift," 1885, p. 351 et seq., upon "the Midrashim to the Pentateuch and the three-year cycle of Palestine" (to which reference may be had for many details omitted here), the undersigned has registered 32 or 33 sedarim in Numbers (see "Monatsschrift," 1886, p. 443), while the rabbinical Bible of Venice, 1617, contains a note stating that, according to some codices, Numbers contains 28 sedarim.

It is evident that in this portion of Bemidbar Rabbah, as in its source, the Tanḥuma, the collected homilies have been considerably metamorphosed and disjointed. Many are quite fragmentary, and others so discursive that they treat of the whole seder in extenso, contrary to the usual practise of this Midrash. Although the marking of the parashiyot at their beginnings and in marginal superscriptions is a departure in the Venice edition (in the editio princeps, the expression stands only at the end of section v.), the sections of the second part are indicated according to the usual notation of the parashiyot. With the exception of sections 16 and 17, which belong to "Shelaḥ Leka," each section contains a parashah of the one-year cycle, which was already recognized when Bemidbar Rabbah was compiled; there are even Tanḥuma Midrashim extant with divisions according to the parashiyot, while the Tanḥuma, in its earliest editions, is alone in using the original arrangement based on the sedarim-cycle. In Bemidbar Rabbah, even in the newest editions, the divisions according to separate homilies are no longer recognizable. The following conspectus of the contents of this second part may therefore be interesting: (1) section xv. 1-10, upon Num. viii. 1; (2) ib. 11, 12, upon Num. viii. 5; (3) ib. 13-16, upon Num. x. 1; (4) ib. 17-25, upon Num. xi. 16; (5) section xvi. 1-11 (= Tanḥ., "Shelaḥ," 1-7, Tanḥ. ed. Buber, 1-11) upon Num. xiii. 1; (6) ib. 12-23 (= Tanḥ. 8-13; Tanḥ. Buber, 12-25, is not in the Vatican MS.) comment upon Num. xiii. 17 to xiv. 23; (7) ib. 24, 25 (Tanḥ. Buber, addition, 1-6, Vatican MS.) homily upon Num. xiv. 11; (8) ib. 26-28 (compare Tanḥ. Buber, addition, 7-14, Vatican MS.) upon Num. xiv. 26; (9) section xvii. 1-4, upon Num. xv. 1; (10) ib. 5, 6, upon Num. xv. 37; (11) section xviii. 1-20, upon Num. xvi.; (12) ib. 21, an addition , contained in none of the Tanḥuma MS.; (13) ib. 23, a fragment of a homily on Num. xvii. 16; (14) ib. 22 and section xix. 1-8, on Num. xix. 1; (15) ib. 9-14, commentary on Num. xx. 7-13; (16) 15-33, continuous exposition of Num. xx. 14-xxi. 35; (17) sec. xx. 1-20, explanation of Num. xxii. 2-xxiii. 24; (18) ib. 21, upon Num. xxiv. 3; (19) ib. 22-25, upon Num. xxv. 1; (20) section xxi. 1 and 3-7, upon Num. xxv. 10; (21) ib. 8-13, upon Num. xxvi. 52; (22) ib. 2 and 14, 15, upon Num. xxvii. 15; (23) ib. 16-22, upon Num. xxviii. 1; (24) ib. 23-25, on Num. xxix. 35; (25) section xxii. 1, upon Num. xxx. 2; (26) ib. 2-6, upon Num. xxxi. 1; (27) ib. 7-9, upon Num. xxxii. 1; (28) section xxiii. 1-4, on Num. xxxiii. 1; (29) ib. 5-12, on Num. xxxiv. 1; (30) ib. 13-14, upon Num. xxxv. 9.

Since the second part of Bemidbar Rabbah, additions excepted, is derived from the Tanḥuma Midrashim, the question arises whether it and part 1 (sec. i.-xiv.) are to be ascribed to one author. That the author of the comparatively late commentary on the parashiyot "Bemidbar" and "Naso"—supposing that the Midrash upon these two is the work of a single author—should have deliberately rounded out his incomplete work with the Midrash Tanḥuma is certainly highly improbable. According to Epstein ("Beiträge zur Jüdischen Alterthumskunde," p. 70) some unknown author wrote the Midrash upon the parashah "Bemidbar" in order to complete the Sifre, which commences with Num. v. 1; another then continued it with the commentary on "Naso," and in order to complete the work for the remainder of Numbers, the commentary for the remaining parashiyot was drawn from Tanḥuma. It must also be mentioned that Cod. Hebr. 149 of the Paris National Library, dating from the year 1291, contains only the parashah "Bemidbar," while the Munich Cod. 97, 2 (Steinschneider), dated 1418, covers only this and "Naso."

Even the first part contains much that is taken from the Tanḥuma: "but a copious stream of new Haggadah swallows the Midrash drawn from this source and entirely obscures the arrangement of the Yelamdenu" (Zunz, "G. V." p. 260). In the parashah "Bemidbar," the outer framework of the original composition is still recognizable. There are five sections, containing five homilies or fragments of such, taken from the Tanḥuma upon Num. i. 1, ii. 1, iii. 14, iii. 40, and iv. 17, which are expanded by some very discursive additions. As Tanḥuma only treats of the first verses of each chapter, no doubt the author's intention was to supply haggadic commentary to the others. But in the section upon "Naso," which is more than three times the volume of that preceding, there are long passages which have no relation to the Tanḥuma homilies, based as they are upon the sedarim-cycle, and commencing in "Naso" with Num. v. 11. Sections vi., vii., viii., x., which, like the other lengthy sections in which the material derived from the Tanḥuma is overwhelmed in a flood of new Haggadot, show even more clearly the endeavor to supply homilies and continuous expositions for all sections of "Naso." Very truly hasZunz ("G. V." p. 261) said: "Instead of the brief explanations or allegories of the ancients, instead of their uniform citation of authorities, we have here compilations from halakic and haggadic works, intermingled with artificial and often trivial applications of Scripture, and for many pages continuously we find no citation of any source whatever." Nevertheless, most remarkable indeed was the industry of the unknown author of this imperfect work—a fragment, no doubt, of his original purpose. The skill calls for wonder and appreciation, which enabled him (sections xiii. and xiv. on Num. vii.) to give a different interpretation to each one of the twelve passages enumerating the offerings of princes of the tribes—identical in all but the name of the prince ("Monatsschrift," xxxv., p. 445).

This portion of the Bemidbar Rabbah shows all the marks of the late haggadic age; there is much which can be referred to R. Moses ha-Darshan, and which reveals a connection with Midrash Tadshe. The work is, according to Zunz, hardly older than the twelfth century.