Mercier and Sperber's (M&S's) analysis of reasoning as designed for argumentation represents another blow to certain long-held assumptions about cognitive processes: That reasoning is the abstract application of a propositional calculus used to determine what is true. Instead, M&S suggest that reasoning is the output of argumentation psychology, a suite of cognitive systems designed to handle incommensurate representations between people. This is courageous and provocative because it suggests that enterprises such as science are handled by a psychology designed for argumentation. Insofar as reasoning can be defined as an interestingly coherent natural category, M&S are likely correct.
However, the argument can be taken further. If reasoning is for argument, what is argument for? While M&S allude to this, there is some value in explicitly addressing the function of argument because it directly speaks to how argumentation psychology should work.
Consider the case of the evaluation of factual or policy claims. It is tempting to think that argument's proper domain in such cases is to determine the truth or accuracy of incommensurate representations–a natural consequence of information being distributed nonuniformly across bodies–that some people have access to information that others do not, and that, given imperfect information, each person is expected to prefer one's own data and conclusions and be wary of others. On this view, even if reasoning is for argument, then the ultimate logic of argument is the same as the classical view of reasoning–as a way of determining truth–albeit in a way that diverges from a rational view, by virtue of the division of information access in the real world.
However, a consideration of the selection pressures for argument suggests that this is not a complete picture of the function of argument and thus not a complete picture of the psychology of argumentation, even when dealing with claims ostensibly about truth. Here's why.
Communication exists because it affects behavior. This makes communicated information a particularly powerful way to leverage others' capacities. This creates opportunity for collaboration, as well as exploitation, and as such there will be selection on both senders and receivers of argument. M&S do not ignore that conflict of interest is inherent in communication–suggesting the existence of epistemic vigilance for ensuring that only “reliable,” “effective,” “quality” information is acted on. But what constitutes reliable, effective, or quality information? Certainly, as M&S imply, a criterion of accuracy applies to a class of claims. “The bear is on this side of the river” and “The bear is on the other side of the river” are factual claims warranting investigation of truth value. The bear is somewhere, after all, and argument will help determine where.
However, while there is reason to think that there is strong selection for discovering the accuracy for certain classes of claims, there is also good reason to think that there is little selection for searching for the accuracy of many others. Instead, if signaled information is capable of causing others to act, there is selection to broadcast representations that will cause others to modify their current or future actions with respect to the sender. Because utility and accuracy are ultimately separable, even for factual claims, there is not always selection for accuracy, but instead for a psychology that fixes representations–in oneself and in others–along actuarially beneficial dimensions.
This suggests at least two broad classes of selection pressures and, subsequently, two broad classes of argument psychology:
First, is a class of argumentation psychology designed to handle conflicts of interest over self-regarding and other-regarding behaviors, the goal of which is to change representations of the payoff structure of pursuing certain future plans. This view of argument has already uncovered a highly-specific “grammar” of argument structure in both senders and receivers (Petersen et al. Reference Petersen, Sell, Tooby, Cosmides and Høgh-Olesen2010; Sell Reference Sell2006; Tooby et al. Reference Tooby, Cosmides, Sell, Lieberman, Sznycer and Elliot2008).
Second, is a class of argumentation psychology designed around social coordination. Because shared mental content is a consequence of coordinated activities, and because cooperation requires a meeting of the “minds,” shared mental representations can be used as markers and facilitators of social coordination. In other words, the exposition of claims–and the signal of agreement or disagreement with them–can be used as a social instrument to mark affiliation. Agreement and disagreement therefore become commodities in themselves as a way of signaling the coordination strength and challenging others. This class of argumentation psychology should be designed to conflate evaluations of the argument with the source and social context of the argument; who is arguing should be just as important as what they are saying when considering the “goodness” of an argument. Additionally, the motivation to argue, and the choice of argument content itself, should be at least in part the result of strategic nonconscious assessments of the local social world. This view of argument has already led to the discovery of evidence that the mind treats certain classes of claims as markers of social affiliation (Pietraszewski et al., in preparation).
These are not aberrant uses of argument, but, rather, these functions lie at the core of how the human psychological competence of argument–and thus how “reasoning”–works. The valuation of social coordination, for example, is likely built right into the sinew and bone of argumentation–both in terms of the criteria for generating arguments and for the criteria of assessing argument. This suggests that reasoning is not simply based on argument, but on argument designed for negotiating conflicts of interest and signaling social coordination rather than exclusively fixing truth.
Mercier and Sperber's (M&S's) analysis of reasoning as designed for argumentation represents another blow to certain long-held assumptions about cognitive processes: That reasoning is the abstract application of a propositional calculus used to determine what is true. Instead, M&S suggest that reasoning is the output of argumentation psychology, a suite of cognitive systems designed to handle incommensurate representations between people. This is courageous and provocative because it suggests that enterprises such as science are handled by a psychology designed for argumentation. Insofar as reasoning can be defined as an interestingly coherent natural category, M&S are likely correct.
However, the argument can be taken further. If reasoning is for argument, what is argument for? While M&S allude to this, there is some value in explicitly addressing the function of argument because it directly speaks to how argumentation psychology should work.
Consider the case of the evaluation of factual or policy claims. It is tempting to think that argument's proper domain in such cases is to determine the truth or accuracy of incommensurate representations–a natural consequence of information being distributed nonuniformly across bodies–that some people have access to information that others do not, and that, given imperfect information, each person is expected to prefer one's own data and conclusions and be wary of others. On this view, even if reasoning is for argument, then the ultimate logic of argument is the same as the classical view of reasoning–as a way of determining truth–albeit in a way that diverges from a rational view, by virtue of the division of information access in the real world.
However, a consideration of the selection pressures for argument suggests that this is not a complete picture of the function of argument and thus not a complete picture of the psychology of argumentation, even when dealing with claims ostensibly about truth. Here's why.
Communication exists because it affects behavior. This makes communicated information a particularly powerful way to leverage others' capacities. This creates opportunity for collaboration, as well as exploitation, and as such there will be selection on both senders and receivers of argument. M&S do not ignore that conflict of interest is inherent in communication–suggesting the existence of epistemic vigilance for ensuring that only “reliable,” “effective,” “quality” information is acted on. But what constitutes reliable, effective, or quality information? Certainly, as M&S imply, a criterion of accuracy applies to a class of claims. “The bear is on this side of the river” and “The bear is on the other side of the river” are factual claims warranting investigation of truth value. The bear is somewhere, after all, and argument will help determine where.
However, while there is reason to think that there is strong selection for discovering the accuracy for certain classes of claims, there is also good reason to think that there is little selection for searching for the accuracy of many others. Instead, if signaled information is capable of causing others to act, there is selection to broadcast representations that will cause others to modify their current or future actions with respect to the sender. Because utility and accuracy are ultimately separable, even for factual claims, there is not always selection for accuracy, but instead for a psychology that fixes representations–in oneself and in others–along actuarially beneficial dimensions.
This suggests at least two broad classes of selection pressures and, subsequently, two broad classes of argument psychology:
First, is a class of argumentation psychology designed to handle conflicts of interest over self-regarding and other-regarding behaviors, the goal of which is to change representations of the payoff structure of pursuing certain future plans. This view of argument has already uncovered a highly-specific “grammar” of argument structure in both senders and receivers (Petersen et al. Reference Petersen, Sell, Tooby, Cosmides and Høgh-Olesen2010; Sell Reference Sell2006; Tooby et al. Reference Tooby, Cosmides, Sell, Lieberman, Sznycer and Elliot2008).
Second, is a class of argumentation psychology designed around social coordination. Because shared mental content is a consequence of coordinated activities, and because cooperation requires a meeting of the “minds,” shared mental representations can be used as markers and facilitators of social coordination. In other words, the exposition of claims–and the signal of agreement or disagreement with them–can be used as a social instrument to mark affiliation. Agreement and disagreement therefore become commodities in themselves as a way of signaling the coordination strength and challenging others. This class of argumentation psychology should be designed to conflate evaluations of the argument with the source and social context of the argument; who is arguing should be just as important as what they are saying when considering the “goodness” of an argument. Additionally, the motivation to argue, and the choice of argument content itself, should be at least in part the result of strategic nonconscious assessments of the local social world. This view of argument has already led to the discovery of evidence that the mind treats certain classes of claims as markers of social affiliation (Pietraszewski et al., in preparation).
These are not aberrant uses of argument, but, rather, these functions lie at the core of how the human psychological competence of argument–and thus how “reasoning”–works. The valuation of social coordination, for example, is likely built right into the sinew and bone of argumentation–both in terms of the criteria for generating arguments and for the criteria of assessing argument. This suggests that reasoning is not simply based on argument, but on argument designed for negotiating conflicts of interest and signaling social coordination rather than exclusively fixing truth.