The Song of Moses has been analyzed with a theory of clause combining worked out for the verbal system of Biblical Hebrew (Isaksson 2015a). One of the main tenets in the theory is that clauses are combined in semantic relations, and that “switches” of grams and clause types supply the text with meaning and structure. Another tenet is that the partial loss of the morphological distinction between the VprefS and the VprefL grams was handled by a restriction of word order, in that VprefS verbs were placed in initial position in affirmative clauses, whereas VprefL verbs were used in clauses of the “X-VprefL” type (where X may also be a negation: lō or bal). A third tenet is that the conjunction we/wa basically marks a clause as an addition (related to a previous clause). VprefA (cohortative) clauses on the other hand exhibit a free word order, and the long -å̄ ending is facultative. These tenets have been applied to the archaic poetry of the Song of Moses, and the text has turned out to be communicatively successful.
The article takes a longer piece of direct speech (Gen. 45: 3-15) as a starting point for a discussion of clause linking in relation to the concepts of 'main line' and 'discourse type' as well as the verbal grammatical morphemes making up the Biblical Hebrew verbal system. It is suggested that the concept of 'discourse type' must be redefined in order to be productive: 'a type of text characterized by a standardized coding of the main line'. In this sense direct speech is not a discourse type, since its main line can be coded by many types of clauses. 'Direct speech' is a pragmatic concept referring to the utterances of actants in the text. With Joseph's speech as point of departure the narrative storyline and the main line in instructional/procedural discourse are discussed. It is shown that the wa-VprefS clause is used not only to code a storyline in SBH but also to elaborate a preceding finite or non-finite clause. It is also suggested that when clauses are co-ordinated they as a rule show the same syntactical structure: NCl + NCl, Vsuff+ Vsuff, VprefS+ VprefS, etc. (Isaksson 2013). If there are no discourse type constraints (as there are in narrative prose and instructional discourse) co-ordinated clauses may be juxtaposed with or without the conjunction w partial derivative/wa.
After 70 years of intensive research on the Dead Sea Scrolls it is conspicuous that scholarship has not arrived at even a mainstream view of the Hebrew language in the scrolls. The characteristic linguistic features of this written Hebrew point in several seemingly contradictory directions. There are both strong classicizing tendencies in the language and clear colloquial traits. This has led many scholars to regard QH as an intermediate state (in a diachronic sense) between Late Biblical Hebrew and Rabbinic Hebrew, in spite of the classicizing traits already in LBH and in spite of the fact that QH is contemporary with RH. The thesis in this article is that the scholarly discussions on the classicizing element in QH has failed to account for the nature of the biblical texts that the scribes in Qumran intended to imitate. An analysis of the biblical text tradition in Qumran must be able to account for the deviations from the Masoretic (Tiberian) text tradition. Among the features most difficult to explain in QH are the long forms of the personal pronouns (both independent and suffixed), as well as the retention of the stem vowel in verbs like yšqwdw (1QS 6:7). It is not enough to explain such features as dialectal traits (which they certainly are in the end). Because of the classicizing intention of the scribes at Qumran it is necessary to account for the deviating features in the light of the biblical reading tradition of the Qumran community. It is argued in the article that the reading tradition in Qumran shared many characteristics with the Babylonian and Samaritan traditions. Among the shared features are a strong tendency to long personal pronouns where Tiberian tradition (and our Hebrew Bibles) has short pronouns, and retention of the stem vowel in verbal forms of the type yiqtolū, where the Tiberian tradition reduce the vowel. The reading tradition in Qumran was heavily influenced by their colloquial language, and this oral reading tradition induced the colloquial features also in the indigenous scribal production of their biblical texts, as Kutscher has proved in his study of the Isaiah scroll. The article argues that the Qumran scribes when writing their sectarian texts intended to use the type of biblical language that they encountered in their biblical reading tradition. My conclusion is that the Qumran community followed an old biblical reading tradition when they used long forms of the personal pronouns such as hmh or -kmh or verb forms yiqtolū with retained stem vowel. The only central dialectal feature of QH that cannot be accounted for by comparison with other Jewish reading traditions are the long forms of the third person singular pronouns hwʾh and hyʾh. They might be analogical dialectal innovations by analogy, but comparative Semitic evidence makes a Protosemitic origin more probable. In such a case they indicate an old Hebrew dialectal variation between long and short third person singular pronouns, attested only in the reading tradition and the written sectarian literature of the Qumran community.
2nd rev. and expanded edition
The Biblical Hebrew wayyiqṭol clause-type is a primary constituent in the theory of ‘consecutive tenses’. This article uses recent advancements in the study of the masoretic text to clarify that such clauses were pronounced without gemination of the prefix consonant in SBH: wa-yiqṭōl (past perfective meaning). The gemination was an innovative feature of the reading tradition during the Second Temple Period. This opens for the question of the status of the conjunction wa before past perfective short yiqṭōl. It is shown that the traditional assumption of a special ‘consecutive waw’ before short yiqṭōl is unwarranted. The coding of pragmatic discourse continuity already has a signal: the clause-type with normal wa and initial verb (type wa-VX). The typical main-line clause in historical narration, wa(y)-yiqṭōl, signals discourse continuity because the verb directly follows the conjunction wa, and this conjunction was a normal ‘natural language connective’ wa ‘and’ in SBH.