Davidson, Donald (1917–2003)

7. Adverbial modification

We began with Davidson on events. The strongest argument he advances for the existence of events and for his conception of their nature derives, surprisingly, from his views in the philosophy of language. Any theory of meaning for a language must embody a view of the relationship between language and reality. Davidson’s conviction is that a theory of meaning, by providing a view about this relationship, offers substantive answers to the various metaphysical questions about reality. In particular, it will require events in order to explain the semantic (logical) form of action, event and causal statements.

Consider the English sentence (1). One obvious candidate for its truth-condition is (2):

In (2) language is both mentioned and used, and in this sense (2) ‘hooks up’ language to reality. This hook-up remains silent on the nature of reality. It simply tells us that the English sentence (2) requires for its truth that John hit Bill. Since an adequate theory of meaning must be finite (1984: 4–15), if we try to construct a theory for, say, English, we are forced to read structure into English sentences (see §4). There does not seem to be another way to generate infinitely many sentences from a finite vocabulary. But now consider sentences like (3)–(5):

There are no specifiable limits upon the number of kinds of adverbial modifiers which can sensibly attach to these sorts of sentences. Therefore, treating each distinctively modified sentence as attributing between John and Bill a distinct primitive relation (such as hitting-at-six-in-the-bedroom) threatens to offend against the condition that the theory be finite. On the basis of considerations of this sort, Davidson proffers a proposal which reveals the common elements in these sentences, issues in the correct semantic truth-conditions, and validates the requisite implications – for instance, that (4) implies (3), and so forth. He takes sentences like (1) and (3)–(5) to harbour existential quantifiers ranging over events. The thesis that there are events is true because the best semantics for English requires quantification over them, as witnessed by their counterparts in first order logic ( $1^{\prime }$ ), and ( $3^{\prime }$ )–( $4^{\prime }$ ) respectively:

This technique for discerning ontological commitments extends to all cases where quantification and predication are required in order to construct a satisfactory semantics for natural language. The method is a general method for doing ontology (see Adverbs; Ontological commitment).