Adina Moshavi, «Two Types of Argumentation Involving Rhetorical Questions in Biblical Hebrew Dialogue», Vol. 90 (2009) 32-46
Rhetorical questions (henceforth RQs) often express a premise in a logical argument. Although the use of RQs in arguments has been widely noted, the modes of reasoning underlying the arguments have not received sufficient attention. The present study investigates argumentative RQs in the prose dialogue in Genesis through Kings in the light of pragmatic argumentation theory. Two logical forms, modus tollens and denying the antecedent, are identified as accounting for the majority of arguments expressed by RQs. The first type is generally intended to deductively establish its conclusion, while the second, formally invalid form is used presumptively to challenge the addressee to justify his position. There is also a presumptive variety of the modus tollens argument, which is based on a subjective premise. Both modus tollens and denying the antecedent have similar linguistic representations and can be effective means of refusing directives.
Two Types of Argumentation Involving Rhetorical Questions 37
speakerâ€™s intentions in making the argument. Arguments may also be
judged from the perspective of effectiveness. An argument is effective, or
successful, if it achieves its goal.
An important point made by Walton is that arguments in natural
dialogue are often not intended to be logically valid. In a valid argument,
the truth of the conclusion follows from the truth of the premises. An
example is the syllogism â€œAll Aâ€™s are B. X is an A. Therefore, X is B.â€ If
the first two premises are true, then it is also true that X is B. Many
arguments in everyday conversation, in contrast, are â€œpresumptiveâ€, shifting
the burden of proof to the addressee to show that the conclusion is not
true(29). Walton argues that some logically invalid arguments can be
effectively used as presumptive rather than deductive arguments (30).
Waltonâ€™s views are directly relevant to understanding the Biblical use of
RQs in argumentation, as shown below.
3. The data for the study
The corpus for the study is the prose portions of Genesis through
Kings(31). The questions in the corpus were located by means of
computerized searches for the various interrogative particles (32).
Clauses with alh were omitted due to the difficulty in distinguishing
questions with the interrogative-negative combination alh from affirmative
assertions with the homonymous clausal adverb alh (33). Unmarked yes-no
questions were excluded as well, due to the impossibility of finding these by
(29) See, e.g., WALTON, Argument Structure, 240, 245; WALTON, The New Dialectic,
(30) D. WALTON, Informal Fallacies. Towards a Theory of Argument Criticisms
(Pragmatics & Beyond 4; Amsterdam 1987) 3-4; D. WALTON, Argumentation Schemes.
(31) The text for the study is the MT, as represented in BHS. I follow BHS in
categorizing passages as prose or poetry. Translations are my own, based primarily on
NJPS and NRSV.
(32) Clauses containing the formally identical exclamatory particles (i.e., hm, Ëšya)
were excluded from the results of the search. There are clear syntactic differences
between the homonymous interrogative and exclamatory particles, as pointed out by
STEINER, â€œMah Nishtannaâ€, 16-17; for example, hm in exclamations functions as an
adverb of degree (e.g., Num 24,5 wbf hm â€œhow beautifulâ€), while hm in questions is most
frequently a pronoun (e.g., 1 Sam 20,1 ytyÃ§[ hm â€œwhat did I doâ€) or an adverb of manner
(e.g., Gen 44,16 qdfxn hm â€œhow can we prove our innocence?â€). On the relation between
RQs and exclamations, see below.
(33) On the clausal adverb alh see J. BLAU, A Grammar of Biblical Hebrew (Porta
Linguarum Orientalium Neue Serie 12; Wiesbaden 21993) Â§103.3; M.L. BROWN, â€œâ€˜Is it
Not?â€™ or â€˜Indeed!â€™: HL in Northwest Semiticâ€, Maarav 4 (1987) 201-219; A. MOSHAVI,
â€œSyntactic Evidence for a Clausal Adverb alh in Biblical Hebrewâ€, JNSL 33 (2007) 51-
63; D. SIVAN â€“ W. SCHNIEDEWIND, â€œLetting Your â€˜Yesâ€™ Be â€˜Noâ€™ in Ancient Israel: A
Study of the Asseverative al and alÃ¸h'â€, JSS 38 (1993) 209-226; R.C. STEINER, review of J.
BLAU, An Adverbial Construction in Hebrew and Arabic. Sentence Adverbials in Frontal
Position Separated from the Rest of the Sentence, Afroasiatic Linguistics 6 (1979) 147-
149. Since any negative RQ can be reformulated as an assertion consisting of the
affirmative implication of the question, it is generally impossible to prove that a alh
clause is a negative RQ as opposed to an assertion with the clausal adverb; thus there is
no way to know whether rmal Ëšyla ytrbd alh (Num 23,26) is â€œDidnâ€™t I tell youâ€, or â€œI
told you!â€ Note that the opposite is not the case; there are many cases in which syntactic
and/or pragmatic criteria suffice to demonstrate that alh must represent the clausal