Advantages of a constructional approach to verb argument structure
Goldberg argues that there are a number of advantages to adopting a constructional approach to verb argument structure.
Avoids implausible verb senses
Firstly, the constructional approach avoids the necessity of positing several dis tinct senses for one verb (which is necessary in a lexically driven model), in order to account for all the constructions it can appear in; some of these might be implausible senses. Consider the examples in (3).
The verb sneeze is a prototypical intransitive verb (3a). That is, it normally occurs with a single argument: the subject (Lily). Despite this fact, sneeze can occur in a syntactic construction like (3b), which can be represented as XCAUSES Y TO MOVE Z BY SNEEZING:[X Lily] causes [Y the birthday cards] to move [Z off the mantelpiece] by sneezing. As Goldberg points out, if we assume that this ‘cause to move by sneezing’ sense is a property of the verb itself, then we might expect to find a language (or languages) somewhere in the world with a lexical item specialised for this meaning, yet the existence of a verb sense of this kind is not attested.
Avoids circularity
Secondly, Goldberg argues that a constructional account has the advantage of avoiding circularity. If we assume that verbs are ‘in charge’ of everything that happens in a sentence – for example, how many participants are required and in what order – we are forced to posit as many senses for a verb as there are constructions in which that verb can occur:
Goldberg argues that if the properties of the constructions in which a verb can occur are seen as the properties of the construction itself rather than proper ties determined by the verb, this problem is avoided.
Semantic parsimony
The third advantage that Goldberg claims for a constructional approach is that it enables semantic parsimony. In other words, if the range of constructions in which a verb can occur – as well as the subtle differences in meaning associated with different possibilities – can be accounted for directly in relation to the construction itself rather than by positing long lists of senses for individual verbs, the resulting explanation is more economical. For example, because the verb kick can appear in the eight different verb argument constructions illustrated in (4), a lexically driven approach would be forced to posit eight different senses or lexical entries for this verb.
In contrast, a constructional approach places the burden of explanation on the syntactic construction itself rather than on the verb.
Compositionality
The fourth advantage claimed by Goldberg is that a constructional account preserves compositionality, albeit in a weakened form. In other words, while all linguists would agree that words contribute to the meaning of sentences, there is considerable disagreement about what and how much they contribute. As we have seen, in a lexically driven approach, words (particularly verbs) are assumed to contribute not only their content meaning, but also their ‘requirements’ concerning the syntactic structure of the sentence. In a constructional approach, Goldberg argues, the problems inherent in a lexically driven approach can be avoided while preserving the point of agreement: words do contribute meaning to sentences, but not all the meaning. Put another way, sentence-level constructions have their own conventional schematic meaning independent of the verbs and other lexical items that are embedded in them. These sentence-level constructions represent symbolic units in their own right, much like the formal idioms discussed in the previous chapter, which can be lexically filled in a number of ways. In the next section, we will set out in more detail how this set of claims is substantiated in Goldberg’s theory.
