By explaining the term 'empirical evidence', Davidson states that it includes any contingent propositions that the radical interpreter may legitimately appeal to warrant his interpretation.6 Thus for him, T-theory is to be selected on the basis of 'evidence plausibly available to an interpreter' where an interpreter is 'someone who does not already know how to interpret utterances the theory is designed to cover' (Davidson, 1984, 128/ Fordor and Lepore, 1992, 72).
In this regard, Davidson states that a successful meaning theory is one where the meaning of an expression must provide an extensionally adequate truth theory and it meets further evidential constraints. In the context of radical interpretation, the primary question raised is: 'How is it the case that whatever the native said to the field linguist in the behavioral form, the field linguist would receive it in the same form or same spirit?' Further, it may be asked what characteristics the field linguist must have to understand the behavioral approach of the natives? Again, is there any guarantee that the field linguist possesses the required features what he needs to understand the natives' behavior?
To resolve all these problems Davidson introduces the notion of 'network of belief'. According to him, there are two sorts of belief, 'network of belief' and 'dense pattern of belief'. Meaning and belief form a nexus in case of 'network of belief'. But in case of 'dense pattern of belief' meaning and belief are conjoined and forms an agreement pattern, i.e., the maximum agreement between meaning of a proposition and the belief form. He further subsumes that beliefs form a 'belief system'. Ascribing the belief system Quine enunciates that any utterances can be considered as language if it arouses the possibility of radical interpretation. This asserts "nothing can be a language unless the correct T-theory for it could be selected by the sorts of observations plausibly available to the child or the linguist" (Fordor and Lepore, 1992, 73). Endorsing this view Davidson says, "a child learns his first words and sentences by hearing and using them in the presence of appropriate stimuli" (Epistemology Naturalised, p.18/ Fordor and Lepore, 1992, 72). Thus, it is stated that both a child and the linguist do succeed in choosing the right meaning theory on the basis of evidences available to them. Hence for both of them, meaning is to be identified as cognitive meaning.
| 6 | Ibid. |