- Conditionals and prediction: Time, knowledge, and causation in conditional constructions by Barbara Dancygier
This study describes conditionals as a set of related constructions defined by specific form-meaning correlations The author identifies several ‘basic parameters of conditionality’, which include (1) the contribution of the conditional marker if, (2) the verb forms in protasis and apodosis, (3) the different types of semantic relations between protasis and apodosis, and (4) the order of protasis and apodosis. These and other parameters are described one by one and progressively combined in the study, resulting in a coherent and well-argued typology of conditional constructions in English.
The most general contribution to the construction of the conditional marker if is described as ‘non-assertiveness’: if signals that for the clause in which it occurs ‘some of the felicity conditions for asserting do not hold’ (18). The author argues that this also applies to conditionals where the protasis is discursively ‘given’, such as If (as you say) your car is stolen, how will you get to work tomorrow?; in such cases the conditional marker signals mere repetition of a preceding assertion which need not correspond to the present speaker’s beliefs and therefore does not count as an assertion. Conditionals with if are the main focus of the study, but the last two chapters also include comparisons with related constructions like concessive conditionals, constructions with unless, or paratactic conditionals such as Say one word and I’ll kill you.
With the verb forms used in protasis and apodosis, the author makes a basic distinction between predictive and nonpredictive conditionals. In nonpredictive conditionals, like the discursively ‘given’ conditional cited above, the verb forms in protasis and apodosis are used according to the general principles which also apply in nonconditional contexts. In predictive conditionals, the verb forms do not follow the general rules but are characterized by two types of backshift. In if-backshift, the protasis uses the present tense to express an assumption about the future that is not itself a prediction but is used to make a prediction in the apodosis. In hypothetical backshift, the present tense in if-backshifted protases is shifted further to past or past perfect to signal the hypotheticality (‘distance’) of the assumption.
The different types of semantic relationship between protasis and apodosis are a third parameter in the analysis. The author makes a basic distinction between real-world, inferential/epistemic and speech act conditionals, following the framework of Eve Sweetser (From etymology to pragmatics, Cambridge: CUP, 1990), and links these semantic types to the predictive and nonpredictive categories: Epistemic and speech act relations typically involve nonpredictive constructions. In addition to the three basic types, she also discusses in detail the little-described metatextual conditional where the protasis comments on some aspect of the linguistic form or meaning of the apodosis. [End Page 218]
A final important parameter is the order of protasis and apodosis, which again correlates in different ways with the other parameters of predictive vs. nonpredictive and real-world vs. inferential vs. speech act vs. metatextual. Taken together, the combination of the different semantic and formal parameters outlined in this work results in a coherent and detailed typology of conditional constructions in English in which the prototypical member is the predictive real-world conditional with protasis-apodosis order.