The Vocabulary of the Septuagint and Literary Criticism 65 that the Septuagintal text of vv. 22b-23 diverges from the verses which precede it. In light of these facts, we can suggest that the translator of vv. 22b-23 was not the same as the translator of the rest of the section on Joshua’s investiture. This is probably due to the fact that the Hebrew text that lay before the initial translator did not include vv. 22b-23, making it impossible for him to have translated these verses. Only at a later stage, when this additional section became an accepted part of the Hebrew text, did another translator find the need to translate it and add it to the Septuagintal text. He focused on his mission to translate the additional Hebrew verses ignoring the meaningful differences between his translation and the previous translation. The differences in translation vocabulary between vv. 19-22a and vv. 22b-23, combined with the arguments regarding the textual status of vv. 22b-23, thus witness to a two-stage process of translation as a result of two stages in the composition of the Hebrew text. The attempt to harmonize the Septuagintal text with that of the MT is a well-attested phenomenon, known both from the Septuagintal revisions and from a comparison of the Greek biblical scrolls from Qumran with the later Septuagint version (29). It would appear that in our present case the addition of the translation of vv. 22b-23 came about through similar reasons. The recognition that the authoritative text included a passage not reflected in the Greek translation led to its addition in the Septuagint. The later translator did not adapt his new translation to the translation before him, however, but worked it anew. It is therefore possible to identify traces of his handiwork through his choice of vocabulary. As stated above, on the basis of various factors I concluded that v. 19 was added to the pericope of Joshua’s investiture in Numbers 27 at a later stage and that the description of the implementation in vv. 22b-23 was similarly absent from the original text of the passage. Our identification of two stages in the Greek translation enables us to gain a more precise insight into the development of the passage. According to the Septuagintal evidence, the description of the implementation was incorporated at a later stage than the addition of v. 19. The different translations of the verb hwx demonstrate that the addition of the description of the implementation was made by a translator other than that of v. 19. In light of these findings and conclusions, we are led to assume that v. 19 lay before the initial translator, whereas vv. 22b-23 did not exist in his version. In other words, the development of the text included a stage which contained v. 19 but not vv. 22b-23. We can thus conclude that the additions were made in two phases: (29) See SEELIGMANN, “Problems and Perspectives in Modern Septuagint Research”, 185; E.C. ULRICH, “The Septuagint Manuscripts from Qumran: A Reappraisal of Their Value”, Septuagint, Scrolls and Cognate Writings (eds. G.J. BROOKE – B. LINDARS) (Septuagint and Cognate Studies 33; Atlanta, GA 1992) 49-80; E. TOV, Textual Criticism of the Hebrew Bible (Minneapolis, PA 22001) 143-148; E. TOV, “The Greek Biblical Texts from the Judean desert”, The Bible as Book. The Transmission of the Greek Text (eds. S. MCKENDRICK – O.A. O’SULLIVAN) (London 2003) 97-122; S. KREUZER, “From ‘Old Greek’ to the Recensions: Who and What Caused the Change of the Hebrew Reference Text of the Septuagint?”, Septuagint Research. Issues and Challenges in the Study of the Greek Jewish Scriptures (eds. W. KRAUS – R.G. WOODEN) (Septuagint and Cognate Studies 53; Atlanta, GA 2006) 225-237.