Putting Brandom on His Feet: A Realist Interpretation of Inferentialism

Abstract

1. Inference and RepresentationThe view that the meaning of a word or sentence is identical its in language, and that language is primarily a social practice, is mainly associated with the name of Ludwig Wittgenstein (1958, §§ 30 and 43). It is opposed theories that construe meanings as abstract entities (for example, propositions) or as somehow determined by mental entities (for example, intentions). As is well known, however, Wittgenstein left it develop a systematic use of meaning. One of these others is Robert Brandom, whose thoughts about meaning, language, and social practices in general, have attracted much attention in the past two decades, more precisely, since the publication of his book Making It Explicit in 1994.Brandom's theory, which he also calls semantic pragmatism and pragmatic semantics, sticks out among other approaches of this sort because of its sheer size and richness of details. It is part of a comprehensive system that also deals with normativity, truth, and intentionality, mention just three important themes.1 The core of Brandom's semantics is the assumption that the meaning (or propositional content) of an assertion depends on its inferential role, that is, on its function as a premise or conclusion in inferences.2 One understands an assertion if one is able draw the relevant inferences. Brandom emphasizes that drawing inferences is a kind of knowing-how, a practical activity or capacity, rather than some kind of theoretical knowledge.Furthermore, Brandom also leaves no doubt that the main opponent of inferentialism is representationalism, which is his term for approaches that start with the concept of representation and this for defining the concept of inference. A representationalist theory would typically describe how words and sentences refer things and facts in the world, how the truth value of a sentence depends on the of the words appearing in the sentence, and finally, how true conclusions can be inferred from true premises.Brandom turns this explanatory strategy upside down. For him, the basic notion is that of inference.3 Inferences between sentences determine the meanings of these sentences and of the words contained in them. There are inferences and bad ones. Roughly speaking, neglecting a lot of details, we may say that the good inferences are those which are endorsed by the community. Truth is introduced into the theory only at a later stage, being defined as that which is preserved in the transition from premises conclusions in good inferences.As reference, this is not a relation between language and the world, such as between the word dog and a class of hairy animals. It is rather a relation between the word and another part of discourse. Imagine, for example, a dialogue about someone's pet. At some point in this dialogue, the would perhaps be specified as a dog, which would establish a relation between the word dog and previous occurrences of the word pet in the same dialogue. Brandom calls this an anaphoric account of refers. Anaphoric is intralinguistic and is not be confused with extralinguistic reference, which does not exist in Brandom's system. The purpose is to show how an analysis in terms of anaphoric mechanisms can provide the resources for a purely intralinguistic account of the of the English sentences by means of which philosophers make assertions about extralinguistic referential relations. (Brandom 1994, 306)In order better understand this anti-representationalist treatment of representational concepts, it may be useful distinguish three kinds of representationalism: In a first sense, a theory can be said be representationalist if it uses representational concepts such as reference at all, whether these figure as basic or as derived concepts and however they may be defined within the theory. …