A Unified Approach to Weqatal in Biblical Hebrew: A Cosubordination Analysis

This paper argues that the Biblical Hebrew weqatal Verb grammaticalises cosubordination, that is, that a weqatal clause is syntactically independent of but semantically dependent on the illocutionary and semantic features of its Donor Clause. The Donor Clause which can be any main or non-main Tiberian Hebrew Clause type. Such a cosubordination analysis is shown to have greater explanatory power and fewer (and less) problematic exceptions to the traditional account of weqatal. There are three traditional approaches to the problem of weqatal, and particularly its relation to the qatal. The first and oldest solution is that the apparent tense-/aspect-switching effect is due to the waw, claiming that a so-called Conversive waw in weqatal is a different morpheme from the regular Coordinator waw (e.g. Sasson 1997, 2001). This solution is often bolstered by the fallacious argument of the supposed prosodic difference between qatal and weqatal, whereby weqatal undergoes an accent shift is restricted environments (Revell 1984, 1985), but Garr (1998) dismantles such an argument thoroughly. The second, widely held solution is that weqatal grammaticalises Nonpast/Nonperfective, having gained this value from qatal’s common use in the apodoses of Conditional Clauses [(waw-)X + qatal]. The third and rarely held view is that weqatal and qatal are more similar than not, both encoding Perfectivity (Berry 1903; Cook 2002; cf. Baayen 1997). The Nonperfective analysis is most common, but it has two major weaknesses. Firstly, although in most cases it correctly predicts that weqatal expresses Nonperfectivity, it fails to predict that weqatal would inherit the interpersonal (e.g. illocutionary force) and other semantic values (e.g. frequentative, negation) of the DC. That is, if weqatal is a regular Finite Nonperfective Verb, we would expect in the majority of cases for it to introduce its own illocution, but it does not. Secondly, there are over one hundred systematic cases of weqatal following wayyiqtol or qatal, in which weqatal is past. Hence this paper offers a fourth solution, namely, that weqatal is distinct from qatal, but not in grammaticalising a tense/aspect distinction, but cosubordination (Van Valin & LaPolla 1997). In cosubordination, the Cosubordinate Clause inherits semantics (and other) features from its DC even though the DC and the Cosubordinate Clause are syntactically independent. I call this the Cosubordination Analysis. The proposed Cosubordination Analysis, in contrast, predicts both the inheritance of interpersonal/semantic values from the DC, and the inheritance specifically of past tense after past tense DCs. The exceptions to the Cosubordination Analysis are fewer and more amenable to explanation than other analyses.