Reflection: Formulaic expressions across varieties of English

There can be differences in the frequency of occurrence of at least some routine expressions across different varieties of English, and in some cases these arise through generalized conversational implicatures that have become somewhat fossilized. The routine expressions tell me about it and you bet, for instance, occur more frequently in American English compared to British or Australian English.

Another instance is the use of rhetorical questions in response to polar (i.e. yes/no) questions. Schaffer offers a long-list of such rhetorical questions that are commonly used as retorts in American English.

Speakers can implicate through such responses not only “yes”, as in the two examples in listed in [4.19], but also “no”, as in the two examples listed in [4.20]. However, in being implicated via a tautology, that is, as (supposedly) having an “obvious” answer, the speaker also implicates that the original question was inapposite or inappropriate in some way. Given that the basic form of these utterances follows a recognizable and recurrent pattern, they are, as Schaffer (2005) argues, a strong candidate for treatment as a kind of utterance-type meaning. Interestingly, it was the “Pope question” that Bouton (1988) found L2 learners had difficulty interpreting – they clearly lacked the relevant utterance-type meaning. The fact that there are many such routine expressions, or preferred ways of saying things, is often relatively opaque to native speakers of English. However, it is all too familiar to those learning English as a second language, or even to those coming across routine expressions that are particular to a certain variety of English.

In sum, the neo-Gricean contribution to our understanding of pragmatic meaning representations, as we have discussed, has been two-fold. On the one hand, emphasis has been placed on the importance of a literal notion of what is said as a meaning representation that is potentially accessible but not necessarily accessed by users. On the other hand, the importance of a layer of presumptive or utterance-type meaning representation has been highlighted. While some scholars, particularly Relevance theorists as we shall see, have argued that these two levels of meaning representation are unnecessary (Carston 2002), such a conclusion, we would suggest, is premature given the way in which speakers can retreat to what is literally said in particular circumstances, on the one hand, and given the extent to which formulaicity abounds in discourse and interaction, on the other.

However, it is worth noting that Levinson’s (2000) move to shift the theorization of generalized conversational implicatures out of a strictly neo-Gricean account of speaker-intended meaning into a broader account of what is communicated through the proposal of hearer corollaries for the maxims has been the source of considerable subsequent debate (see Carston 2002). Indeed, there is a growing body of experimental work that attempts to examine the cognitive processes and mechanisms which underpin pragmatic meaning, including the question of whether default inferences are involved. As this is primarily a matter of how users understand meaning, however, we will return to discussing this line of experimental work.