What is Communicative Success?

I Introduction

In most of our communicative transactions we are confident of success. We read the newspaper and are normally pretty sure that we have understood the text. We usually think we understand what is said in the TV broadcast and in casual conversation, and only rarely do subsequent events make us revise our judgment that we did.¹ But that we are confident of success, even if we are right, and even if we are both right and justified, does not mean that we have a clear idea of what success consists in.

Ideas about successful communication have had a prominent place in the philosophy of language in recent decades, mostly as a step on the way to meaning theoretical claims. This has been the case with philosophers like Wittgenstein, Quine, Davidson, Dummett, and Evans. Mostly, the discussion of communicative success has been of limited extent, serving as a subordinate part of some longer argument. I follow this tradition of giving communicative success a central role in theorizing about meaning, but I mostly disagree with the ideas that over the years have been offered in it.² In fact I am going to defend what I call the classical view of communicative success, a view that I attribute both to Locke and to Frege (see section II). I am going to defend it against modern views of a behaviorist/pragmatist character, views that impose a requirement of knowledge, or of reliability, and views by which public language semantics must be in place before communicative success can be accounted for.

Before proceeding, I will pause for a brief methodological interlude.³ As with many other basic notions in philosophy, when we are trying to come up with a general account, it is not so clear what the success conditions of the account itself is. Moreover, as with many other basic notions, it is not obvious that we can come up with neutral and generally agreed upon criteria of adequacy that are separate from the competing accounts themselves. On the contrary, often a dispute about what is the best account involves a dispute over criteria of assessment, and I don't expect things to be different in the present case. I shall simply state what I take the criteria to be.

First, I do not think that there is a well-defined pre-theoretic concept of communicative success that is available as object of conceptual analysis. What we have is a common sense practice of judging communicative success or failure in vernacular terms such as 'He did not understand,' 'She misinterpreted him,' 'He got the message,' etc. As speakers of English we do have intuitions about the correctness of such utterances with respect to various scenarios, even if we have much poorer intuitions about the exact content of the judgments expressed, about the evidence that is appropriate for making them, or about the extent to which the judgments are supported by the evidence used. I do take it, nevertheless, as a general criterion of adequacy, that the rate of success, as judged by the defined concept of the account should be roughly equal to the rate of success as judged by the standards of the pre-theoretic practice. We should, I think, by default attempt to provide a systematic reconstruction of the common sense standards by which communicative success is usually judged. If there is a significant difference between the two, then either there is a conceptual difference between the account and the common sense practice, or we have to ascribe a high rate of error to the practice from the theoretical point of view. The second alternative is not acceptable unless we have strong independent reasons for believing both that the proposed account is correct and that its concept of communicative success is in line with notions employed in common sense. Without such reasons, it is a criterion of adequacy that the rate of success as judged by the theory matches the rate of success as judged by the standards of common sense.

There is, however, a further question whether common sense standards are reasonably constant or...