Rezetko, Robert, “Dating biblical Hebrew: Evidence from Samuel-Kings and Chronicles.” Pages 215-250 in Biblical Hebrew: Studies in Chronology and Typology. Edited by Ian Young. London: T&T Clark, 2003.
The primary distinction of Late Biblical Hebrew (LBH) features from Early Biblical Hebrew (EBH) is based on a comparison of the language of Samuel-Kings with that of Chronicles. This assumes that Samuel-Kings was composed in the pre-exilic period, while Chronicles is a product of the post-exilic period so that differences between the two reflect diachronic change. However, Rezetko points out that recent work by Auld has suggested that the Chronicler did not use Samuel-Kings as a source, but rather both works shared a common source. Thus their dates of composition may be closer than assumed. While Auld’s position is far from widely accepted, the disagreement should caution us about concluding automatically that the linguistic differences are diachronic. Rezetko suggests instead that they are explicable by non-chronological means such as stylistic idiosyncracy, dialectal difference, diglossia, or editorial and scribal activity.
He begins by treating 16 features of LBH for which he considers chronological explanations to be inadequate. For instance, the noun afformative וּת- (which I discussed a bit here) has generally been held to be a feature of LBH, the most common example being מַלְכוּת “kingdom”, used in contrast to ממלכה. In the Bible, Rezetko counts 71 unique common nouns formed with וּת- with a total of 380 occurrences. Two-thirds occur only once and very few can be contrasted with a non-וּת noun having a comparable meaning. Further, the number of unique nouns is similar in Samuel (5), Kings (4), Chronicles (4), Nehemiah (3), and Ezra (2). The misjudgment of the feature as late seems to be skewed by distribution of just one word, מַלְכוּת.
Another example is the supposed increase in the use of past tense weqatal forms and corresponding decrease in wayyiqtol forms in LBH, including the replacement of wayyiqtol in Samuel-Kings by simple weqatal in the parallel passages in Chronicles. However, Rezetko disputes these findings, counting 658 occasions in which Samuel-Kings and Chronicles both have wayyiqtol, 17 where Samuel-Kings has wayyiqtol but Chronicles has a qatal form, and 17 where Chronicles has a wayyiqtol and Samuel-Kings has a qatal form. There are only two occasions that Chronicles has a past weqatal that is parallel to a wayyiqtol from Samuel-Kings, but six times Samuel-Kings has a past tense weqatal that is parallel to a wayyiqtol from Chronicles.
Next, Rezetko turns to some issues of method. Here he calls for greater awareness of presuppositions, and more rigorous methodology. For instance, the study of the language “should be liberated from assumptions concerning the literary composition and development of these books.” Scholars have suggested dates for Chronicles from the early Persian period to the late Hellenistic, a span of 400 years. Further, Hurvitz’ method assumes that books like Samuel are not ‘chronologically problematic’ and are indisputably pre-exilic. He also argues that study of the language must take greater account of socio-linguistic factors, such as dialect variation and diglossia, as well as scribal factors such as editorial and scribal revision. Lastly, he argues that the diachronic study of language should be based largely on grammar rather than lexicography. Grammar is much more stable than lexicon. It is quite easy for a language to borrow or create new words and their distribution may have much more to do with regional or ideological differences as diachronic.