Brandom on Representation and Inference

Abstract

It will take repeated engagements to come to terms with Brandom's huge, cohesive, quirky, and brilliant book. Here I can make only a first pass at one of its architectonic themes. The book is organized around an opposition between two explanatory strategies-representationalism, which Brandom says has been dominant since the Enlightenment (p. xvi), and inferentialism, the complementary approach Brandom adopts. This gives Brandom's story the attractions of bucking a trend. More seriously, he motivates his approach, to a large extent, by its competitor's flaws. The fundamental insight here is that semantics must answer to pragmatics (p. 83). Semantic terms are not intelligible apart from how they pull their weight in enabling us to make sense of what speakers do, in the way we make sense of rational behaviour in general. And inference is a central element in the framework of rational requirements in which we place linguistic behaviour when we make sense of it in that way. Now representationalism, as Brandom conceives it, holds that concepts of the representational directedness of language towards the extra-linguistic world can confer a self-standing intelligibility on the idea of, e.g., the content of an assertion, in advance of our needing to advert to inferential relations between such contents. Consider, e.g., a theorist who undertakes to capture contents of potential assertions by associating sentences with sets of possible worlds, and then to capture inference-warranting relations between such contents in terms of inclusion relations between sets of possible worlds. By using the concept of assertion as he does, this theorist conforms to the letter of the principle that semantics must answer to pragmatics. But suppose he does not mind its seeming an afterthought that the world-directed contents he supposedly brings into view as contents of assertions stand in inferential relations to one another. Since inference is central to the framework in which we make rational sense of linguistic behaviour, such a theorist violates the spirit of the principle. We

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 88 citations exist upstream.