Hostname: page-component-745bb68f8f-v2bm5 Total loading time: 0 Render date: 2025-02-06T09:34:17.700Z Has data issue: false hasContentIssue false

Reasoning, argumentation, and cognition

Published online by Cambridge University Press:  29 March 2011

Keith Frankish
Affiliation:
Department of Philosophy, The Open University, Milton Keynes MK7 6AA, United Kingdom. k.frankish@open.ac.ukhttp://www.open.ac.uk/Arts/philos/frankish.htm

Abstract

This commentary does three things. First, it offers further support for the view that explicit reasoning evolved for public argumentation. Second, it suggests that promoting effective communication may not be the only, or even the main, function of public argumentation. Third, it argues that the data Mercier and Sperber (M&S) cite are compatible with the view that reasoning has subsequently been co-opted to play a role in individual cognition.

Type
Open Peer Commentary
Copyright
Copyright © Cambridge University Press 2011

I agree with Mercier and Sperber (M&S) on the need for a dual-process perspective that distinguishes intuitive inference and explicit reasoning, and I find M&S's case for the argumentative theory of reasoning attractive. The theory is also a salutary corrective to the tendency of dual-process theorists to see explicit (“type 2”) reasoning as unbiased and normatively correct. (Henceforth, I shall follow M&S in using “reasoning” for explicit thought processes, as opposed to non-conscious intuitive inferences.)

Here I shall add some brief, broadly sympathetic comments. First, I want to outline a further reason for thinking that reasoning evolved for social, argumentative purposes. M&S claim that reasoning is a personal-level activity, involving the construction of complex arguments in response to intuitive beliefs about the strength of the component steps. This view, which I have myself defended in various places (e.g., Frankish Reference Frankish, Carruthers and Boucher1998; Reference Frankish2004; Reference Frankish, Evans and Frankish2009), requires that reasoners have intuitive-level knowledge of rules of inference, such as modus ponens, which guides their explicit reasoning. (Rules here includes rules of thumb; the arguments we construct need not be demonstrative; see Frankish Reference Frankish2004.) Now, there are two main forms these rules could take: They could be abstract, defined over concepts and logical structures, or they could be linguistic, defined over the words and structures of a natural language. (Rules of the latter kind can be applied in constructing arguments, provided the arguments are expressed linguistically, either overtly or in inner speech.) They are far more likely to take the latter form, however, especially in early human societies. Knowledge of linguistic rules can be obtained relatively easily in the course of exposure to, and engagement in, argument with one's peers. (It might even be purely procedural, embedded in routines for manipulating linguistic structures.) Knowledge of abstract rules, by contrast, would require explicit teaching, or at least abstraction from previously acquired linguistic rules. (These considerations are set out at more length in Frankish Reference Frankish, Carruthers and Boucher1998 and Reference Frankish2004.) Note that I am assuming here that the knowledge that guides explicit argument construction is largely learned. The case for this is strong; reasoning strategies are known to be responsive to verbal instruction and to vary dramatically between individuals (e.g., Stanovich Reference Stanovich1999). Note, too, that this view is compatible with M&S's claim that there is a specialized intuitive mechanism for representing arguments. It merely implies that the mechanism operates primarily on linguistic representations and that its rule database is acquired.

If this is right, then it supports the view that the original function of reasoning was social and argumentative. For it suggests that individual reasoning is an internalized version of overt argumentation, conducted in inner speech and guided by knowledge acquired in the course of public argumentation. (There are other reasons, too, for thinking that conscious reasoning is language-involving; see, e.g., Carruthers Reference Carruthers1996 and Reference Carruthers1998.) And this in turn suggests that public argumentation predated individual reasoning and that whatever adaptations we have for reasoning originally evolved to serve the former. (We can tell a bootstrapping story about the origins of the knowledge that guides public argumentation, involving a process of cultural evolution.)

Second, I want to say something about the function of public argumentation. M&S argue that this is to make communication more efficient and reliable. I agree that argumentation does this, but it may not be its only, or even main, function. As the data M&S cite make clear, group reasoning is often a strikingly effective process of inquiry, which zeroes in on the correct solution. Reasoning may thus have evolved primarily for collective cognition, if not for the individual variety. (It would not be surprising if evolution preferred collective cognitive mechanisms. Early human societies would have depended heavily on collective action and problem solving, whereas individual projects might have disrupted group cohesion.) Argumentation may have other important social functions, too. For example, displays of argumentative prowess may serve to attract mates, providing evidence of health and mental acuity.

Third, while I agree that reasoning evolved to serve public argumentation (and still does serve it), I suspect M&S underestimate the extent to which it has subsequently been co-opted to play a role in individual cognition. For the demands of argumentation and individual cognition may not be as different as M&S suppose. There are two points here. First, sound arguments tend to be convincing and optimal decisions defensible, so an argumentative mechanism will often deliver the same results a cognitive mechanism would. Secondly, the biases natural to argumentation may also affect much individual reasoning. For this too may often have a quasi-argumentative structure, with individuals seeking to defend their attitudes and decisions against criticism from their future selves.

I shall expand on the latter point. The idea is that, even as private reasoners, we have other motives besides epistemic ones. We often accept propositions for pragmatic reasons, because we find them comforting, or because they are popular with our peers. Moreover, we value psychological stability for its own sake; we like to have a settled set of opinions (to know our own minds) (Frankish Reference Frankish2004). We may therefore be inclined to bolster our opinions with arguments, so that our future selves will be able to resist counter-evidence and avoid distressing changes of mind. Thus, even in private we may tend to display confirmation bias and motivated, proactive reasoning. Something similar goes for decision making. Intentions serve to coordinate future planning (e.g., Bratman Reference Bratman1987), and in many cases it is better to form and stick with an intention, even if it is not optimal, than to remain undecided or keep changing one's plans. Given this, people may tend to settle on decisions they find easy to justify to themselves, and to buttress them with arguments, so that they will not be tempted to revise them later. Hence, they may still tend to exhibit reason-based choice, even when they do not have an eye on the reactions of their peers.

Thus, while the results cited by M&S may show that reasoning is not well adapted for individual intellectual inquiry (which, as M&S rightly stress, we find very hard), they do not show that it is not adapted to other roles in individual cognition, broadly construed. Of course, as M&S note, motivated reasoning and reason-based choice often have unwelcome consequences (especially, perhaps, in modern technological societies), but, if anything, this tends to support the present suggestion, since the more functions these biases have, the more gains there are to offset the costs.

References

Bratman, M. E. (1987) Intention, plans, and practical reason. Harvard University Press.Google Scholar
Carruthers, P. (1996) Language, thought and consciousness: An essay in philosophical psychology. Cambridge University Press.CrossRefGoogle Scholar
Carruthers, P. (1998) Conscious thinking: Language or elimination? Mind and Language 13:457–76.CrossRefGoogle Scholar
Frankish, K. (1998) Natural language and virtual belief. In: Language and thought: Interdisciplinary themes, ed. Carruthers, P. & Boucher, J., pp. 248–69. Cambridge University Press.CrossRefGoogle Scholar
Frankish, K. (2004) Mind and supermind. Cambridge University Press.CrossRefGoogle Scholar
Frankish, K. (2009) Systems and levels: Dual-system theories and the personal-subpersonal distinction. In: In two minds: Dual processes and beyond, ed. Evans, J. St. B. T. & Frankish, K., pp. 89107. Oxford University Press.CrossRefGoogle Scholar
Stanovich, K. E. (1999) Who is rational? Studies of individual differences in reasoning. Erlbaum.CrossRefGoogle Scholar