Humorous exploitation of discourse patterns
Recycling and recontextualisation for humorous purposes can apply freely in the case of sequences with fixed positions and typically conventionalized, salient meanings attacking conventional use while at the same time keeping the canonical pattern intact. The playful recycling of the relevant, (otherwise) completely fixed formulae, such as the core sequences of telephone call openings, for instance, has been shown to aim at challenging linguistic conventions (Antono- poulou and Sifianou 2003). One point at which the proposed analysis differs from those of different frameworks is the status of these expressions and their relation to the rest of the lexicogrammar of a language.
In discussing the notorious ‘doctor wife’s joke’, Simpson (2003: 35) shows that its humor originates in an opposition between ‘discursive units’, since it is a departure from the formula of service encounters. The relevant trigger here is the first utterance of the character in the joke described as speaking with a ‘bronchial’ voice: “Is the doctor in?” This introduces the service encounter setting (as Simpson points out) and gives rise to the first script (patient-doctor) which will then be opposed by the second one (the lover script) activated by the response of the doctor’s “young and pretty wife” “No, come right in”. Simpson (2003: 36) notes: “the discursive shift [from a public to a private domain] suggested here is predicated upon a culturally driven formula”. In Simpson’s understanding, discoursal units are juxtaposed to cognitive ones. Within a cognitive framework, there is no cognitive-discoursal incompatibility. In fact, we are drawing attention to the importance of subsuming interactional aspects of discourse to cognitively oriented studies (Brone, Feyaerts and Veale 2006: 211). In Langacker’s terms (2008: 457), although discourse is often considered a separate topic, “requiring different methods and descriptive constructs, the contrast with lower levels is at most a matter of degree”. In cognitive grammar, discourse is understood as the “very basis for language structure and [...] thus essential for understanding grammar” (Langacker 2008: 457). Conventional patterns of language engendered through recursive occurrence of similar usage events involve all the specific level characteristics of utterances, along with the expression’s full conceptual understanding (Langacker 2008: 457-458).
In short, since discourse units signal conceptual domains, they have ipso facto cognitive status. As already mentioned in section 1, from a Cognitive Linguistics perspective (Langacker 2001:143), lexical structures of whatever length are “instructions to modify a certain discourse state”, allowing for a unified treatment of discourse expectations (Emmott 1997: 270-271). This is precisely the approach taken here where the focus is on conventional units of language linked with specific discoursal properties; in humorous discourse, these may in turn lead to socio-cultural incongruities treated as a side effect of yet another type of constructional property.
In fact, the type of data we examine here can be seen as instances of “encoding idioms” first discussed in Fillmore, Kay and O’Connor 1988 (see also Croft and Cruse 2004; Bybee 2006). These are expressions, which although interpretable by the (syntactic and semantic) rules of the language, they are still idiomatic as associating a particular form with a meaning (e.g., answer the door, bright red, etc.). In the same way, the interpretation of the expressions and passages we discuss relies on ordinary rules of semantic composition, but the discoursal setting or context may a) create conventional expectations as to the lexical items used or the way of presentation and/or b) conventionally constrain the possible interpretations or strongly prime some of these as opposed to others. Interaction offers a good opportunity for testing conventional expectations, because it builds on pre-existing (explicit or implicit) context and shapes subsequent turns. We are thus interested in conventionalised linguistic units carrying with them prospective and retrospective suppositions (Langacker 2008: 460).
The following examples instantiate the humorous exploitation of both texts belonging to well described genres, such as those of classroom discourse, and texts which may have not achieved genre status but have salient discoursally identifiable characteristics. The latter we propose to also accommodate in the theory on a par with the former. We actually suggest that unless such discoursal features are identified and their status appropriately established, the humorous effect they trigger would have to be accounted for in an ad hoc manner, impoverishing both humor theory and cognitive linguistic theory (of any denomination).
All the examples discussed in section 4 belong to well established genres. The first one (4.1) parodies the conventions of literary editions and is therefore an instance of conflicting frames on the written level. In 4.2 we consider the humorous exploitation of classroom discourse in a filmscript, thus moving to instances of directly represented spoken interaction. Indirectly represented oral discourse is exemplified with cartoon data in 4.3 and 4.4.