Discover more from Overcoming Bias
People and institutions usually prefer to explain their behaviors in self-serving and self-flattering ways. For example, we usually explain human abilities to create and evaluate chains of reasoning in terms of truth – by reasoning we can better see what is true (including truths about what we want to do).
I’m a little late to the response party, but back in April Mercier and Sperber published their theory that reasoning is designed more to help people persuade others, than to infer truth:
Reasoning is generally seen as a means to improve knowledge and make better decisions. … Our hypothesis is that the function of reasoning is argumentative. It is to devise and evaluate arguments intended to persuade. … A wide range of evidence in the psychology of reasoning and decision making can be reinterpreted and better explained in the light of this hypothesis. … Reasoning is not only for convincing but also for evaluating arguments, and that as such it has an epistemic function. (more; ungated)
Many of their critics, however, noted that reasoning could serve even more functions. Mercier and Sperber responded that such other functions were of only minor importance:
Several commentators, while agreeing that argumentation may be an important function of reasoning, suggest that it may serve other functions, as well. … Our claim is that argumentation is the main function of reasoning. …
Dessalles and Frankish suggest that argumentation could have evolved as a means to display one’s intellectual skills. Indeed, argumentation can be put to such a use. However, … reasoning is more like a crow’s than a peacock’s tail: It may be a bit drab, but it serves its main function well. Its occasional use, for instance, in academic milieus, to display one’s intellectual skills is unlikely to contribute to fitness to the point of having become a biological function, let alone the main function of reasoning. …
Pietraszewski … draws attention to a … class of cases … [where] who is arguing should be just as important as what they are saying when considering the ‘goodness’ of an argument” … The main relevance of a communicative act may be … in the very fact that it took place at all; it may have to do with … signaling agreement and disagreement. This can be done in particular by using arguments not so much to convince but to polarize. …
Frankish points out that reasoning can be used to strengthen our resolve by buttressing our decisions with supporting arguments.
Notice that, relative to the usual story of reasoning’s function, Mercier and Sperber offer a less flattering than usual explanation for argument speakers, but not for argument listeners. That is, Mercier and Sperber accept the self-flattering story of those who hear arguments, that they mainly just want to figure out what is true about the content of the topics argued.
So what might listeners of arguments be up to instead? As the critics above suggest, listeners could be trying to gauge speaker impressiveness, or the social support the speaker can muster in his or her conflicts. Also, listeners could be trying to figure out what they will say in response, in argumentation contests with many possible criteria for who wins. And argument listeners might try to gauge what positions will become accepted by a wider community, to help them decide what positions to personally support.
Once you give it a bit of thought, you can see many possible and even plausible explanations for human reasoning abilities, beyond the simple self-flattering story that we are trying to figure out what is true about the topic of our reasoning.
What Is Reasoning For?
I agree with piddlesworth's point, though I do not find some of the provided examples. In particular, I do not perceive that tool use requires "reasoning", any more than the creosote bush needed to "reason" in developing toxins which repel competitors and predators, or a skunk needs to reason out its response to a threat. All that's required is that there is a feedback loop which differentially rewards some "behaviors" (in a broad or narrow sense) with expanded market share and a mechanism to persist the more successful. DNA appears capable of this for some amazingly complex behaviors, and mimicry can expand upon that capability on shorter timescales.
However, what characterizes these mechanisms is that the feedback loop must include manifesting some alternatives in the physical world within which outside influences provide the differential selection pressure. What we call "reasoning" is distinguished by using internal abstractions to model the physical world. With reasoning, it becomes possible to adopt a behavior based on (at least vaguely) anticipated results which have not yet been tested with real world feedback - something which DNA and mimicry based adaptive feedback loops cannot do. Of course, often this is combined with real world feedback, in that reasoning generates hypotheses which still need external validation.
This "pre-emptive" generation and winnowing of potential adaptive behaviors provides both more diverse and complex options, and a much faster adaptation timescale - obvious evolutionary advantages which do not necessarily involve persuasion.
Chimpanzees do exhibit some primitive "reasoning" when viewing a problem never before encountered and then choosing an appropriate tool based on geometry, anticipated weight, etc - rather than using just blind trial and error. They show evidence of having an internal model of the physical world, by whose rules they can refine possible actions before taking them. This involves no persuasion.
The incorporation of "reasoning" into human persuasion (along with the often stronger emotional components) would be a natural outcome, once the modeling and rule inferring/using mental abilities exist, and once the "environment" for which humans need to optimize survival strategies includes interpersonal and social structures. As such, the human ability to reason may well have been to some degree shaped by the adaptive value of its use for persuasion. But it does not seem credible to credit persuasion with the origins of reasoning, and the case for that purpose having since become the primary driver seems weak.
A great deal of current research shows that reason and rationality play a smaller part in our individual and collective decision making and in our influence upon each other, than we might like to believe. Confirmation bias and the backlash effect greatly limit the impact of even well reasoned arguments, especially in areas felt to be important to survival and thriving within the tribe. A relatively small portion of the populace develops reasoning as a primary tool in explaining the world to themselves and to others; as others have pointed out, contemporary politics demonstrates that amply.
Nevertheless, reasoning is sufficiently survival positive, that it retains at least some influence in persuasion. However, the internalized models and rules of social systems are much harder to reality-test than the models of the physical environment which generated reasoning as adaptive behavior. Very different models of "how humans work" (politically, economically, socially, psychologically) can coexist in the population more or less indefinitely without selective pressure winnowing them to a "most successful" answer. Hence the many political positions which almost never persuade each other despite massive attempts to show by "reason" that their model is more correct - and the very low interest in facts or reason by adherents to the various ideologies.
Is any of this an advance on Hume?
We’ve probably all had the experience of being on the verge of acting from anger or jealousy, when someone advises us to act reasonably. A typical picture of motivation for action is one in which emotions or desires drive us one way and our reason drives us in another. I have a desire for a tasty but unhealthy dessert, and the voice of reason tells me that I ought not to eat it. I don’t feel like helping at the food bank on Saturday, but conscience tells me that I ought to fulfil my obligation. On this picture, the morally upstanding or prudent person follows the lead of reason, while the morally deficient character caves into desire or emotion.David Hume, in A Treatise of Human Nature, rejects this traditional characterisation of action and its evaluation, offering a remarkable theory in response. He defends the views that the ends or goals of our actions in all cases are given by our “passions,” not by reason, and that the practical role of reason is to figure out how to fulfil these goals. He makes the astounding declaration that “Reason is, and ought only to be the slave of the passions, and can never pretend to any other office than to serve and obey them...[Hume] characterised reason not as some mysterious power of grasping truth or of intuiting connections between ideas or thoughts, as some philosophers did, but as the ability either to offer demonstrations or proofs or to make causal inferences.So, first, he shows that reason engaged in demonstration can never be a motive to action. Demonstration is deductive reasoning using necessary truths. Demonstrations are the proofs we use in mathematics and geometry. Mathematics can be applied to the world in the way that engineers use it to solve problems in their work, but knowing the truths of mathematics only, without the addition of a goal or purpose, will not produce a motive to any particular action. Second, Hume asks whether causal reasoning by itself can give motivate action. Causal reasoning, which requires the gathering and assembling of observations, allows us to form beliefs about the world. Do these factual beliefs supply us with motivation to act in particular ways? Say I’m sleepy and I believe coffee can stimulate me. It seems this belief can motivate me to drink a cup of coffee. If so, factual beliefs based on causal reasoning can on their own produce motives. However, Hume notices that such beliefs would have no practical effect on us if we didn’t also have some sort of attraction to the goal achieved by the motivated action – in this instance, staying awake when I’m feeling sleepy. Reason informs me that consuming the caffeine in coffee keeps me awake, but reason didn’t tell me what desires to have. It simply gives me a piece of causal information. Factual beliefs have no influence on our behaviour if they are about things in the world of no concern to us.”
Elizabeth Radcliffe, Ruling Passions