Abstract Diessel

Causal and conditional constructions

Causality and conditionality are two basic concepts of human cognition that are commonly expressed in conjoined clauses. While it is often assumed that causal and conditional clauses are semantically closely related (cf. Dancygier 1998; Couper-Kuhlen and Kortmann 2000), they exhibit some striking cross-linguistic differences. Drawing on data from a representative sample of 50 languages, the paper shows that there are three general tendencies that distinguish conditional and causal clauses:

1. First, while conditional clauses usually precede the semantically associated clause, causal clauses tend to follow it. Since both causes and conditions are temporally (or logically) prior to the associated event, the positioning of conditional clauses is iconic, whereas the positioning of causal clauses violates the order-of-mention principle (Diessel 2001, 2005).

2. Second, while causal clauses are often only minimally different from main clauses, conditional clauses tend to be particular constructions that are formally distinct from other clause types (cf. Cristofaro 2003).

3. Third, while causal clauses are usually linked to an ordinary main clause, conditional clauses are often linked to clauses including a particular verb form or some other formal marker (cf. Comrie 1986).

Using corpus data from child and adult language, the paper argues that the formal differences between causal and conditional clauses are motivated by the way these constructions are used and processed. Both types of clauses occur in particular communicative situations. Conditional clauses are commonly used to express a contrast between two contradictory possibilities providing a conceptual framework for the interpretation of the main clause. This explains why conditional clauses tend to precede the main clause. In fact, if the conditional clause follows the main clause, the hearer might misinterpret important aspects of the associated main clause. Since reinterpreting linguistic material disturbs semantic processing, the occurrence of a final conditional clause is limited to constructions in which the conditional clause is either announced in the preceding main clause (e.g. by a particular verb form) or else functions as an afterthought or a metalinguistic speech act.

Causal clauses are commonly used to provide a reason for a controversial statement or action. This is perhaps most obvious in child language, in which causal clauses are embedded in particular discourse routines, consisting of a controversial statement, a causal question (e.g. Why?), and an isolated causal clause (cf. Diessel 2004). In adult language, causal clauses are also used to ‘describe’ a causal link between two propositions, but most languages employ different types of causal clauses to this end. For instance, English has a variety of causal connectives that mark the effect or consequence of a particular situation (e.g. so, thus), but it uses the causal conjunction because to ‘support’ a previous statement or action. This explains why causal (because) clauses, unlike conditional clauses, tend to follow the associated proposition violating the iconicity principle. Moreover, since causes and reasons tend to be communicatively important, causal clauses are usually not ‘deranked’ (Cristofaro 2003). That is, across languages, causal clauses tend to be expressed in non-embedded sentences (i.e. causal clauses are often only minimally different from main clauses), because causal clauses are commonly used as independent speech acts that syntactically are only loosely combined with the associated clause.