Cognitive genre theory IV: More thoughts on new theory & conclusion

This is the fourth post in a series exploring a possible cognitive theory of genre. The first, appearing here, covered some basic concepts from contemporary cognitive science. The second focused on prototype theory, script theory, and relevance theory. The third began to lay out some of the issues in genre theory where attention from a cognitive perspective would be beneficial. You’ll find this post easier if you read the previous ones.

This post takes up where the last left off. It considers a place for relevance theory, argues that genre tokens function as non-hierarchical semiosis, and considers the role cognitive genre in pedagogy.

Relevance, stability and dynamism

A sometimes perplexing fact about genres is that they are both stable and changing. Relevance theory may well help to explain both the stability and dynamism of genres. Stability responds to audience expectations, and thus to minimizing processing effort; the communicative agent can count on the audience adopting the easy interpretation of a work that conforms closely to generic conventions. But assume for the moment that the communicative agent cannot achieve with the current genre conventions what she wants to achieve. Consider a company employee who prepares a periodic report that has generic status within the company. Perhaps another communicative approach would cost her less effort, better highlight her own contributions, allow her to take credit for saving the company money, gratify her desire for efficiency, or offer her some other benefit. Any effort she makes to vary from the existing genre convention is likely to be opposed with the cumulative force of other individuals to the extent that they individually perceive the increased cognitive effort of understanding the new report exceeds the benefits attendant on it. Our hero can predict this resistance (thanks to the operations described by relevance theory) and revise her strategy to address the cognitive cost/utility balance. She can even engage in epistemic actions (see Kirsh, 2009, previous post), testing the reactions of other agents whose responses she has to consider.

(Aside: It probably has not escaped your notice that I have carefully avoided making claims about whether generic knowledge can or does reside in the conscious or unconscious, whether it forms a kind of tacit knowledge, explicit knowledge, or a little of both. I’m saving those questions for another day.)

The studies on my reading list do not explore the effectiveness of individuals’ efforts to challenge generic conventions from a cognitive perspective. Berkenkotter and Huckin (1994) discuss the evolution of the periodical Reader over 11 years, noting changes from the perspective of a discourse community. Swales (1990) cites Bazerman’s (1988) study of the history of the research article. But even Bazerman’s analysis considers the broad sweep of developments while making only plausible surmises about the motivations and ideas of individual agents in his history.

Genre as non-hierarchical semiosis

Miller (1984) describes genre as being part of a complex hierarchy of semiotic levels. This part of her work has received less uptake, and this is perhaps because the logical hierarchy represents an ironic departure from her attack on a priori theoretical and logical forms earlier in her article. Nevertheless, cognitive genre would see generic practices as a form of semiosis in the sense that it sees all symbol use as semiosis: Each token of a symbol type functions to reinforce certain associations between the symbol and other symbols and certain referential relationships between the symbol and the world perceived by its users. As noted above, genre theory should explain the workings of stability and dynamism at the cognitive level.

In a sense, the cognitive meaning of a text type consists of the associations that its use produces, reinforces, and weakens. As Devitt (2004) says: “A genre reflects, constructs, and reinforces the values, epistemology, and power relationships of the group from which it developed and for which it functions…” (64). This echoes Berkenkotter and Huckin’s appeal to Giddens’ structuration theory. Agents make predictions (according to relevance theory) about what communicative tokens will be easy for audience members to interpret; these predictions are influenced by the factors Devitt sets out. But by adapting to those influences in constructing her utterances, the agent is reinforcing them; to the extent that she varies from the expectations of her audience and succeeds in winning the audience’s determination that what she is offering is still relevant, she actually changes the meaning of the genre, even if only in a very small way. The agent’s and audience’s determinations are deeply situated and driven by recognition heuristics and processes.

In a sense, this semiotic role of every utterance, either to reinforce or vary existing associations between symbols and other symbols or their referents, is a speech act attendant on the utterance. Searle (1985) contemplated direct and indirect illocutionary forces in utterances; but he did not acknowledge the semiotic force of every utterance as a speech act.

Role of cognitive genre theory in pedagogy

Time and space are running short, and one of the most important topics is so far left unaddressed. The question of what role genre knowledge should play in language teaching and writing pedagogy has been important at least since Swales (1990). Bhatia (1993), Berkenkotter and Huckin (1994), and Devitt (2004) all take it up. Bawarshi and Reiff (2010, p. 61) sum up the challenge:

[This] understanding of genres suggests that they cannot be explicated, explained, or acquired only through textual or linguistic means; they also cannot be abstracted from the contexts of their use for pedagogical purposes. Because learning genres is about learning to inhabit “interactionally produced worlds”… and social relationships, to think and act and recognize situations in a particular way, and to orient oneself to particular goals, values, and assumptions, some… scholars have questioned the value of explicit genre teaching, while others have more recently sought to develop pedagogical approaches based in genre awareness, ethnography, and situated apprenticeship.

Situated apprenticeship sounds fine: the best place to learn real genres is probably in their use. But I question the value of turning students into genre researchers in order to teach them the effective use of genres. This is a bit like training students in linguistics to teach them to be effective users of their language.

And the rhetorical genre studies folks are perhaps too quick to discount the semiotic value of simulations, something else that cognitive science is illuminating. For example, March et al. (1991), in an article charmingly titled “Learning From Samples of One or Fewer,” explore the symbol-making possibilities in organizational discussions of rare events, near-factual accounts of events, and accounts of hypothetical events. They claim that such linguistic performances have the potential to create powerful symbolic networks in individuals that can be mobilized later to deal with problems similar to those ‘studied.’ Perhaps if genre theory is recognizing the incredible ability of human cognition to use likenesses and typification in communications, it should recognize that classroom simulations can build some (though certainly not all) of the symbolic connections necessary for students to navigate professional and disciplinary genres later. Of course, as you’d expect, I advocate for empirical testing of such claims.

The classroom should also be a safe place to play with genres, providing a (partially) situated opportunity to engage in epistemic actions. Kirsh (2009) notes the classical problem-solving model’s data driven nature, which limits development and evaluation of possible solutions to those where there are existing “patterns currently in working memory” (p. 300). Kirsh uses the example of a player re-arranging Scrabble tiles in order to generate possible solutions that her mental representation might not permit given the solutions she has already represented. This is very similar in some ways to the individual agent confronted with a personal intention and a set of genres that do not allow her to achieve that intention. Epistemic actions (communicative “experiments” that allow her to assess the relevance theoretic challenges of certain modes, approaches, convention breaches, etc.) permit her to gauge the possibilities for violating genre conventions without facing intractable resistance.

De-reifying genre

You are no doubt praying this is the last section, and so it is. And by this time, it should also be obvious that I have not addressed an elephant in the cognitive genre room: I have not offered a definition of genre. In fact, I’ve tried to avoid the word, preferring instead “text type,” “genre knowledge,” or “genre convention,” to recognize generic tendencies. My resistance to defining genre stems from the fact that all communications are conventional and that defining genre tends to result in reification of it and the establishment of a priori categories. Unfortunately, I’m really out of time, and I’ll have to leave my arguments about this until another day (perhaps the oral phase of my exam?).


There is too much here left unsaid (and probably much that is at best ill-said). In the section just concluded, I have attempted to illustrate questions in genre theory that would benefit from (or even require) cognitive theories and methods for their resolution. The commitment of genre theorists since Miller to social constructionism and a view of genre that places is as a property of groups has been very productive and illuminating. But it has failed to address important questions, particularly, questions about how genre knowledge arises and works in the individual agent. I suggested in the first post that the reason genre theorists may have avoided cognitive methods in their efforts is a misunderstanding of the cognitive project(s) and a tendency to equate “cogntivism” with the Carnegie-Mellon model. The work of Flower and Hayes has been influential, certainly, but it is time to leave behind the 1981 view of cognitive science and find a place for contemporary cognitive science’s theories and methods in the genre theory conversation.


Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.