On the Conceptual-Procedural Distinction

Abstract

1. Introduction distinction between two types of semantic meaning--that which encodes representational (or conceptual) information and that which encodes a specification of how to take or manipulate a structure--is not a new topic in linguistics. Recently, however, the topic has been reintroduced within the relevance-theory framework, where the distinction is couched as versus Apparently, the first to suggest this distinction was Diane Blakemore, who in 1987, when discussing her analysis of discourse connectives such as but, after all, and so, wrote: On the one hand, there is the essentially conceptual theory that deals with the way in which elements of linguistic structure map onto concepts--that is, onto constituents of propositional representations that undergo computations. On the other hand, there is the essentially procedural theory that deals with the way in which elements of linguistic structure map directly onto computations themselves--that is, onto mental processes. (Semantic 144) Later, in 1992, she elaborated: [W]e have also seen that there are expressions whose meanings cannot be analyzed in representational [conceptual] terms at all. But, after all, moreover, and inferential so do not contribute to a propositional representation, but simply encode instructions for processing propositional representation.... [Linguistic meaning] may be either representational or procedural [bur not both]. (Understanding 150) (1) In The movie was over so we didn't bother going to the theater, for example, the presence of so signals that the discourse segment following the so should be processed as an effect or a conclusion following from the preceding segment and has no conceptual meaning at all. This line of argument has been pursued by Blakemore (Relevance) and many other adherents of relevance theory. Since the introduction of this schism in meaning, what falls under procedural meaning has been significantly broadened beyond the narrow confines of discourse markers (discourse connectives). distinction has been endorsed and discussed in relevance-theory literature in numerous papers, some researchers arguing for procedural status for one class of linguistic expressions (e.g., discourse markers, illocutionary markers, pronouns, tense, syntactic structures, and intonation contours), some arguing for conceptual status for another class (e.g., illocutionary adverbial, and parentheticals). In this article I will challenge the claim put forth by relevance theory, namely, that a linguistic form--a morpheme, a lexical item, a syntactic structure, or a stress or intonation contour--must be analyzed as having either conceptual meaning or procedural meaning but not both. To do this, I will first show that discourse markers/discourse connectives, some illocutionary markers, and pronouns that have been analyzed by relevance theorists as encoding only procedural meaning should be analyzed as having a conceptual component of meaning. I will then show that illocutionary adverbials, which have been analyzed by relevance theorists as encoding only conceptual meaning, should be analyzed as having procedural meaning as well. 2. So-called Procedural Meaning Only A. Discourse Markers/Discourse Connectives Discourse markers (DMs) are expressions such as but, and, however, so, and in addition which typically signal a relationship between the second segment of a sequence of messages and the preceding segment (see Fraser, What Are Discourse Markers?). Consider the sequence in example 1: (1) A: water won't boil. B: Thus, we can't have tea. Here, thus signals that the second segment functions as a conclusion inferred from the message conveyed by the first segment. In discussing DMs, proponents of procedural analysis have gone to considerable lengths to attempt to demonstrate that these expressions do not have conceptual meaning. …

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 — 33 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.