Back to list

Detail of contribution

Auteur: Elitzur BAR-ASHER SIEGAL

The case of sentential negations in Jewish Babylonian Aramaic (=JBA)

Abstract/Résumé: While most logicians treat negation as an external truth-functional propositional connective, others (Katz 1977, Geach 1972, and especially Horn 2001) demonstrate that natural languages have either predicate denials or term negations. Accepting their conclusion in general, this paper makes the case for the following distribution for the two negators in JBA, la and law: the former functions as a “regular” non-connective negator, and the latter serves as a sentential connective. After demonstrating the problems with previous accounts for the distribution of these negators (Schlesinger 1928), I will support my hypothesis with three types of arguments. a. Syntactic differences: i. law is always the first element of the predication, but la can appear after the subject. ii. la tends to be next to the verb (a tendency noted by Jespersen 1917), while law always is in sentence-initial position and the verb situated in the sentence-final position. For Horn (2001), the fact that negation is not systematically assigned to sentence-initial position (or to other fixed positions) supports the idea that, unlike question markers, negation is not an operator applied to the fully formed proposition. Thus, the distribution in JBA suggests that law does function as a propositional operator. b. Sentential negation from the perspective of information structure: As noted in the literature (mentioned above), the pragmatic characteristics of sentential negations are the following: • They are less informative, not indicating why the “root sentence” is false. • Sentential negations as a connector affect the truth value of the sentence as a whole, without negating specific elements. Accordingly, sentential negations are expected either when the reasons for the denial of the proposition are insignificant in the context, or when the truth value of the sentence as a whole is important. law, as a sentential negator, meets these expectations, as it appears in the following contexts: i. To negate salient entailment (the reasons for the denial are unimportant). ii. In negative questions. iii. In conditional sentences, when only the truth value of the entire proposition matters. c. Diachronic considerations: The negator law derives from a cleft-construction [la+hu (=no+he) >law], which functioned as an explicit sentential negation, i.e. “it is not the case that...” In light of this, JBA is an example of a language in which the distinction between the two types of negation is formally expressed. At the same time, these data strengthen the hypothesis that natural languages, in general, do not demonstrate sentential negation.