An Introduction to Discourse Analysis - Journal Article Review [7]

Title                            : Processing Conversational Implicatures: Alternatives and Counterfactual Reasoning

Researchers               : Bob van Tiel and Walter Schaeken

Year of Publication   : 2016

Source                         : Cognitive Science: A Multidisciplinary Journal, 41, Pages 1119–1154. 
DOI:10.1111/cogs.12362

Objectives                   : 1. To describe the four kinds of inferences (Bott and Noveck: 2004) in more detail which are scalar inferences, free choice inferences, conditional perfection, and exhaustivity in “it”-clefts.

               2. To spell out how these inferences are predicted to behave.

Method                       : The method used in this research is quantitative approach. The results were analyzed by fitting linear or binomial mixed models predicting reading times, decision times, and choice proportions. These models were constructed in R, a programming language and environment for statistical computing (R Development Core Team, 2006) using the lme4 package (Bates & Maechler, 2009). The corresponding p-values were calculated by approximating denominator degrees of freedom using the Satterthwaite procedure as implemented in the lmertest package (Kuznetsova, Bruun Brockhoff, & Haubo Bojensen Christensen, 2013).

Results                       : The computation of scalar inferences was associated with an increase in decision times. The results for the other three types of inferences differed significantly from the pattern for scalar inferences in that their computation facilitated decision times. In addition, there was a significant difference in the processing profiles of conditional perfection and exhaustivity in “it”-clefts, with the former having a greater facilitation effect than the latter.

Following that, however, it would be an exegetical and conceptual mistake to interpret Grice’s approach as an account of utterance processing.

Grice’s approach is situated at the computational level, explaining what inferences listeners are licensed to draw and why. It should not be confused for an algorithmic account of how these inferences are computed. The results of this study indicate that such an algorithmic account should allow for parallel processing of enriched meanings (e.g., Chierchia, 2004; Geurts, 2010; Levinson, 2000; Recanati, 1995).

The finding also warns us against using presence of a processing cost as a diagnostic for conversational implicatures. More broadly, our findings indicate that sometimes pragmatic inferencing is not associated with a processing cost.

Comments                  : For me, this journal article has a comprehensive explanation which every point is explained clearly and deeply. I think this journal article could be one of reference to help the reader understand more about conversational implicature because it provides lots of information about conversational implicature, utterance, and inferences. As we know that utterance and inference are tied with conversational implicature.