The Vocabulary of the Septuagint and Literary Criticism 63 Here, too, textual witnesses to the reading hçm ta exist, a fact which makes it reasonable to assume that the Septuagint also reflects such a text (20). According to this reconstructed Vorlage, the phrase hwhy hwx rçak appeared twice in these verses. The Greek text has no such duplication, however. The first time this phrase appears, the Septuagintal text employs the verb ejntevllomai, while the second time it uses suntavssw. Both these verbs are standard equivalents for hwx and represent appropriate translations of the Hebrew root (21). Why, however, did the translator alternate the Greek verb he used to translate the Hebrew verb hwx? Such alternation between the use of ejntevllomai and suntavssw to render hwx is indeed not uncommon in the Septuagintal text of the book of Numbers. The verb ejntevllomai translates hwx 18 times, whereas suntavssw does so 28 times (22). These two verbs also occur in close proximity to one another in Num 2,33-34; 8,20.22; 34,13, and 36,2.5-6. This discrepancy must be considered in the light of two other differences regarding the description of the implementation process in vv. 22-23, however. While the word rçak in v. 22 is rendered by the term kaqav, in v. 23, kaqavper appears. John Wevers has noted that while this is an exact translation, this is the sole place in Numbers where the word rçak is rendered as kaqavper in the phrase hçm ta hwhy hwx rçak or in similar formulations of the implementation of a divine command (23). Furthermore, the verb hwx occurs once more in our passage, in the description of Moses’ implementation of one of the commands connected 22,9; Judg 3,4; 1 Kgs 8,53.56; Ps. 77,21; Neh 8,14; 9,14; 10,30; 2 Chr 33,8; 34,14; 35,6. With the exception of 12 instances (Exod 9,35; 34,29; 35,29; Lev 8,36; 10,11; Num 27,23; Josh 14,2; 20,2; 21,8; Neh 8,14; 2 Chr 34,14; 35,6), the expression is rendered by the phrase ejn ceiri; Mwush. Of these exceptions, in three cases (Lev 10,11; 2 Chr 34,14; 35,6) it is translated using dia; ceiro;" Moush. In Josh 14,2, the expression ejn ceiri; ΔIhsou' occurs. In Exod 34,29, the phrase is translated as ejpi; tw'n ceirw'n Mwush, and in Exod 35,29 and Josh 20,2 as dia; Mwush. In the five remaining cases (Exod 9,35; Lev 8,36; Num 27,23; Josh 21,8; Neh 8,14), we find the phrase tw'/ Mwush. It is clear that, with the sole exception of Num 27,23, wherever the phrase hçm dyb appears in the book of Numbers it is translated using the Greek ejn ceiri; Mwush. For the translation of hçm dyb using the regular dative, see J. JOOSTEN, “L’Excédent Massorétique du Livre de Jérémie et l’Hébreu Post-Classique”, Conservatism and Innovation in the Hebrew Language of the Hellenistic Period (eds. J. JOOSTEN – J.S. REY) (STDJ 73; Leiden 2008) 104, n. 45. (20) Several Hebrew mss, Targum Pseudo-Jonathan (see BHS) and the midrash cited above reflect the reading hçm ta. Also in Neh 8,14 several Hebrew mss and versions reflect the reading hçm ta: see BHS there. (21) According to DOS SANTOS, Expanded Hebrew Index, entellomai is employed to jv translate hwx 339 times. ( ) These data follow KIM, “Multiple Authorship of the Septuagint Pentateuch”, 22 126. Kim notes that, in contrast to the translator of Leviticus, the translator of Numbers prefers suntavssw when the person giving the command is God, whereas he prefers to translate the verb hwx by ejntevllomai when a human being is the initiator (ibid, 147). On the use of Greek verbs signifying “command” in the LXX, see J.A.L. LEE, “A Lexical Study Thirty Years On, With Observations on ‘Order’ Words in the LXX Pentateuch”, Emanuel. Studies in the Hebrew Bible, the Septuagint, and Dead Sea Scrolls in Honor of Emanuel Tov (eds. S.M. PAUL – R.A. KRAFT – L.H. SCHIFFMAN) (VTSup 94; Boston 2003) 513-524. (23) WEVERS, Notes on the Greek Text of Numbers, 469; J.W. WEVERS, Text History of the Greek Numbers (AAWG. MSU 16; Göttingen 1982) 127.