![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
Grice’s analysis of human communication has proven to be highly influential among many philosophers and cognitive scientists, both past and present. At the same time, it has long been recognized that his analysis faces some difficult objections. In particular, a number of theorists have objected to the account Grice provides of the mental states and processes of those engaged in communication. For these theorists, communication as conceived of by Grice has seemed too mentally demanding and complex to be a good general model of human communication. In this article, I consider this challenge afresh from the perspective of Dan Dennett’s intentional stance theory. More specifically, I consider some recent remarks Dennett has made in this area, seeking to pin down and clarify his view. I then argue that while Dennett’s thinking is on the right track, his view stands in need of both a substantive adjustment and more positive detail. I seek to provide such improvements here. Finally, I consider some of the implications of this improved Dennettian account of Gricean communication for how great ape and human communication are related.
Disclosure statement
No potential conflict of interest was reported by the author.
Notes
1. See Moore (2017) for a discussion and defense of this minimal formulation. Also see Neale (1992).
2. Some theorists instead contend that the metarepresentational beliefs in question need only be potentially inferable (see, e.g., Wilby, 2010). A main challenge for such a view is to explain how such merely potential beliefs can make a genuine difference to actual utterance production and interpretation behavior.
3. See I. Apperly (2011) for an excellent overview.
4. See R. J. Planer (2021) for a recent discussion of the evidence.
5. Though spontaneous pointing in wild bonobos has been reported (Veà and Sabater-Pi (1998); Douglas and Moscovice (2015). Specifically, bonobos point to their swollen genitals to initiate “genito-genital rubbing.” But this is obviously quite different from, e.g., pointing out a tool for an agent which they can use to achieve a current goal (as even very small children will do).
6. This article’s argument picks up on and seeks to extend some ideas that were sketched in R. Planer and Sterelny (2021). And as pointed out to me by one of the referees, it also has a partial (though only partial) precursor in the highly innovative work of Tad Zawidzki (see, especially, T. W. Zawidzki, 2013, 2015, 2018). Briefly put: Zawidzki sees Dennett’s intentional stance theory as a good model of our quotidian mindreading practices, including whatever mindreading we engage in during communication. He then argues that the metarepresentational demands of taking the intentional stance have been seriously overestimated. In particular, in Zawidzki’s view, taking the intentional stance toward some agent does not require the attribution of full-blown propositional attitude states to them (states which, among other things, represent their contents as such, and not merely extensionally). In turn, Zawidzki argues that human communication, including human linguistic communication, can run on a much more modest set of mindreading capacities, such as the “System 1” or “minimal” mindreading capacities proposed by Apperly and Butterfill (see, e.g., Butterfill & Apperly, 2013; I. A. Apperly & Butterfill, 2009; I. Apperly, 2011). As will become clear below, I too shall leverage Dennett’s intentional stance theory to argue that the mindreading demands of human communication have been overstated. But I arrive at this conclusion via quite a different route. In addition, my focus is on the role of higher order intentional states in (Gricean) communication, as opposed to how, exactly, such states represent their contents. And finally, the account I shall develop incurs few, if any, of the commitments of Zawidzki’s account, or the broader framework it is couched within (e.g., about the role of mindshaping in human social life). (For a brief, but comprehensive overview of Zawidzki’s views, see, especially, Zawidzki (2018).).
7. Roughly: a representation plays the desire role when it functions to motivate the agent to change the world so that the representation comes to be an accurate representation of the world. In contrast, a representation plays the belief role when the agent – or better: processes inside the agent – treat that representation as accurately representing the world. These notions will not be important in what follows, so I shall not elaborate on them further.
8. Though not entirely neutral. Dennett is clear that a successful intentional description of some system does imply that the animal is capable of picking up and adaptively responding to certain types of information from its environment (see, especially, Dennett, 2017 on this issue). This tells us at least something about how it must be physically constituted. No system could sort objects based on their color without the capacity to physically register different wavelengths of light.
9. For an especially rich and thorough examination of Dennett’s intentional stance theory, and its broader theoretical role in Dennett’s work, see T. Zawidzki (2018).
10. An allusion to Anscombe’s (1957) remarks about Aristotle’s modeling of our syllogistic reasoning.
11. Or at least as a general rule.
12. A fortiori: I can intend this without having a state of my brain meaning you recognize that I intend that you believe that berries are delicious, where this state causes my real-time communicative behavior.
13. Which would explain why such children struggle to pass tests of higher-order mindreading for years to come, including tests of a more procedural variety.
14. As one referee pointed out to me, whether even younger children (i.e., less than 18 months old) might legitimately be said to be Gricean communicators is a further question. This question is not an idle one, as a number of theorists do wish to characterize infants’ communicative behaviors during the first year of life – in some cases as early as 9 months – in Gricean terms. I agree with this referee that additional evidence is necessary to establish the empirical accuracy of such a characterization. I take no stance on the issue here (though I would not be surprised if such evidence can indeed be provided).
15. Another good example to think about here is infant directed speech. (For a nice review, see Cristia, 2013.) The tendency of adults to slow down and exaggerate their speech stream in a variety of ways when talking with a young child can be understood in terms of the adult wanting to make it as easy as possible for the child to understand what they are saying.
16. In part because of the breadth of the potential evidence base. Also, identifying evidence for Gricean behavioral patterns in chimps or some other non-human great ape species is complicated by the fact that they obviously lead very different lifeways, and labor under very different social, cognitive, and physical constraints, compared to humans. As such, we need be especially careful about using the absence of evidence as evidence of absence in this area. (For example, it is now well-known that chimps tend to perform quite differently on tasks depending on whether the task is inherently cooperative or competitive. Were we to ignore how they behave in the latter contexts, we would form a seriously impoverished picture of their sociocognitive capacities.).
17. See Tomasello (2010) for a discussion of these behaviors.
18. On another, i.e., non-Dennettian, perspective, one might accept this claim but still hold that there are other reasons to conceive of chimps as acting with Gricean communicative intent. In particular (as suggested by one of the referees), one might see the relatively neat correspondence between chimp gestures and the attention getters that (sometimes) proceed them, and the two clauses of Grice’s analysis, respectively, as sufficient grounds for saying that chimps act with Gricean communicative intent. I agree that this is a possible view (though not one I find very compelling). In any case, I again emphasize that my goal here is limited to considering the question of (possible) Gricean communication in non-human great apes from a Dennettian perspective, and not any other perspective.