The conceptual/procedural distinction was introduced in Relevance Theory by Blakemore (1987) to explain the differences between content words and discourse connectives. The main arguments were that content words encode concepts that contribute to the proposition expressed by an utterance while the meaning of a discourse connective is better described in terms of constraints on the inferential phase of interpretation, as for instance within the Gricean notion of conventional implicature (Wilson and Sperber 1993). In a different semantic framework, procedural information has been explored as argumentative instructions for discourse connectives, however without a clear connection to a theory of pragmatic inferences (Ducrot et al. 1980). In another framework, discourse markers signal cognitive discourse relations (Sanders et al. 1992, 1993) and guide the hearer in the interpretation process, therefore facilitating the processing of the text following them (Sanders and Noordman 2000; Sanders 2005).
Numerous studies aim at defining conceptual and procedural distinction and at proposing qualitative discriminating features for the two types of content. The first attempts to define and characterize conceptual vs. procedural information include qualitative features such as representational vs. computational and non-cancellable vs. cancellable. One very significant contribution to the discussion is Sperber and Wilson’s (1993, reprinted in Wilson and Sperber 2012) hypothesis of cognitive foundations of the distinction. They characterise conceptual vs. procedural information in terms such as accessibility to consciousness vs. inaccessible to consciousness (Wilson and Sperber 1993; Wilson 2011). These features of conceptual and procedural information find their roots, on the one hand, in the parallel that has been made between natural language and the ‘language of thought’ and, on the other hand, in the ‘massive modularity hypothesis’ (Sperber 2005; Carruthers 2006). De Saussure (2011) proposes a methodological criterion to distinguish between what is conceptual and what is procedural. In his words, an expression is procedural if it triggers specific inferences that cannot be predicted on the basis of a conceptual core to which general pragmatic inferences (loosening and narrowing) are applied. Escandell-Vidal and Leonetti (2011) propose rigidity as the major feature of procedural information. Their hypothesis is that conceptual information is flexible while procedural information is rigid. Curco (2011) argues that procedures are context-insensitive, categorical, arbitrary and language-specific. Moeschler et al. (2012) suggest that another feature of conceptual vs. procedural information is easily translatable vs. translatable with difficulty. Their suggestion is based on the assumption that conceptual information encoded by linguistic expressions is linked to conceptual representations, which are constituents of language of thought. As language of thought exists beyond specific languages, this should facilitate the translation from one language to another.
The description of conceptual and procedural information in these terms is without any doubt accurate and reflecting the linguistic and cognitive reality of language users. There are however two important limitations of the existent studies. The first is that they are purely theoretical and have not been tested experimentally. One of the very few exceptions is Zufferey (2014), who brought evidence from online (self-paced reading tasks) and offline (acceptability judgments tasks) experimentation that communication of given information is part of the procedural instructions conveyed by some connectives like the French puisque. The second limitation is that existing studies make use uniquely of qualitative features and lack quantitate and objective measures. One recent suggestion is Grisot and Moeschler (2014), who propose to use the Kappa statistical coefficient (Carletta 1996; Artstein and Poesio 2008) for measuring the inter-annotator agreement rate in experiments with linguistic judgment tasks. Their hypothesis is that, based on the existent qualitative features, judging conceptual information results in high rates of inter-annotator agreement whereas judging procedural information results in low rates of inter-annotator agreement (also Cartoni et al. 2013 for connectives in a cross-linguistic perspective).
One crucial question that arises and which can be answered only through experimental studies (on-line techniques such as ERP and eye-tracking) regards the brain’s processing of linguistic expressions encoding procedural and/or conceptual information, as well as the types of inferences involved. Specifically, at this moment, we have no knowledge if these two types of information involve two different cognitive processes or the same cognitive process but different types of inferences. Conceptual information (assumed as easily accessible to consciousness) is expected to be processed faster than procedural information (assumed as inaccessible to consciousness). However, procedural information is described as facilitating the hearer’s task by constraining the possible contextual assumptions and therefore, sentences containing procedural information are processed faster than sentences without it.
In this workshop, we would like therefore to bring together researchers interested in the conceptual/procedural distinction and who aim at investigating it from theoretical and empirical (corpus work, L1 and L2 acquisition, grammaticalization, on-line and offline experimentation) complementary perspectives. We suggest the following research questions:
What is the relation between conceptual and procedural information and which are the consequences and predictions of their theoretical descriptions in terms of cognitive processing?
If procedural meaning is language-specific, which are the constraints imposed by the semantics of particular languages? Are there other constraints (i.e. non- language specific) on procedural information? Are there procedural universals?
What are the means that natural languages use when linguistic elements (tense, connectives, aspects) are not available (for example, encoding of temporal information in tenseless languages), and what is the impact of these strategies on cognitive processing?
Cristina Grisot, Joanna Blochowiak, Jacques Moeschler
Jacques Moeschler
Relevance Theory claims the main function of procedural meaning is to guide the interpretative process (Blakemore 1987), and as a result, to reduce its cognitive load. The cognitive effects resulting in following procedural instructions is to balance the addition of linguistic items. In this respect, the communicative principle of relevance triggers the relevance heuristics: “follow a path of least effort” (Wilson & Sperber 2004). Conversely, the achievement of relevance is dependent on specific cognitive effects launched by procedural meaning. Relevance Theory predicts that these positive effects are either the addition of new information, the strengthening or the suppression of old information (Wilson & Sperber 1986).
This paper questions the difference in achieving relevance between discourse segments with and without procedural markers, as discourse connectives. (1) contrast with (2), but produce the same cognitive effects (3) and (4):
(1) a. Abi pushed Axel. He fell.
b. Axel fell. Abi pushed him.
(2) a. Abi pushed Axel and he fell.
b. Axel fell because Abi pushed him.
(3) The event ABI PUSHED AXEL temporally precedes the event AXEL FELL
(4) The event ABI PUSHED AXEL causes the event AXEL FELL
A fine-grained analysis of the role played by procedural information in the interpretation process will be proposed, to go beyond the current picture where procedural markers make discourse interpretation easier and provides as a result one of the three cognitive effects. Two claims are made.
1. It is assumed that connectives represent additional information explaining the difference between the treatment of a single vs. two discourse segments (Carston 2002). Another assumption is that connectives are complementary to conceptual information encoded in lexical items like ‘push’ and ‘fall’, which triggers a causal ‘push-fall’ rule (Lascarides & Asher 1993). I argue that connectives are more than additional and complementary information. My assumption is that connectives constrain the interpretative process, i.e. limit the interpretation process by disregarding unintended possible interpretations, and are predicted to occur when no sufficient clues are given to ensure the intended interpretation. When, connectives are redundant with conceptual information (2), their main function is the strengthening of contextual effects that could have been achieved without them.
2. The second claim is that these contextual effects are not language specific but rather discourse marker specific, i.e. they arise in the presence of discourse markers and there is no perfect cross-linguistic correspondence among them (Zufferey & Cartoni 2012). Moreover, their cognitive effects can be ranged into different types of inferential meaning. One challenge is to define the locus of semantic and pragmatic layers meanings between close connectives. In causal connectives like ‘parce que’ (‘because’), ‘donc’ (‘so’) and ‘et’ (‘and’), the differences do not lie in their causal meaning but in their semantic and pragmatic locus, like entailment, explicature, implicature (Moeschler 2015). As an example, the causal meaning is an explicature (not defeasible) with ‘parce que’, and an implicature (defeasible) with ‘et’ and ‘donc’, whereas the set of entailments are not identical: P and Q for ‘parce que’ and ‘et’, only P for ‘donc’.
References
Blakemore D. (1987). Semantic Constraints on Relevance. Oxford, Blackwell.
Carston R. (2002). Thoughts and Utterances: The Pragmatics of Explicit Communication. Oxford, Blackwell.
Lascarides A. and Asher N. (1993). “Temporal interpretation, discourse relations and commonsense entailment”. Linguistics and Philosophy 16, 437-93.
Moeschler J. (2015). “Argumentation and Connectives. How do discourse connectives constrain argumentation and utterance interpretations?”. In Capone A. and Mey J. (eds.), Interdisciplinary Studies in Pragmatics, Culture and Society. Cham, Springer.
Sperber D. and Wilson D. (1986). Relevance: Communication and Cognition. Oxford, Blackwell.
Wilson D. and Sperber D. (2004). “Relevance theory”. In Horn L.R. and Ward G. (eds.), The Handbook of Pragmatics. Oxford, Blackwell, 607-632.
Zufferey S. and Cartoni N. (2012). “English and French causal connectives in contrast”. Languages in Contrast 12(2), 232–50.