Tag Archives: Hugo Mercier

The Theory of Mind

Posted on by

I just finished reading The Enigma of Reason by Dan Sperber and Hugo Mercier. They look at the process by which we arrive at reasons for actions. The following thought occurs to me:

You probably assume that understanding your own mind is prior to having a “theory of mind” about other humans. However, it could be the other way around.

Sperber and Mercier do not make this sort of claim. However, I do not think that it is terribly inconsistent with their views.

A theory of mind seeks to explain why agent X performs action A. What I am suggesting is that we arrive at this theory not through introspection but instead by observing action A followed by consequence B repeatedly. After we have seen this happen enough, we develop the insight that perhaps agent X is performing action A in order to achieve consequence B. Call this the basic theory of mind, or at least a theory of what motivates others. Note that we might hold such a basic theory of mind or motivation about animals or even about an inanimate object.

Given that we have a basic theory of mind and that we assume that others have similar basic theory of mind, we can engage in a new form of teaching. If I tell you that I am performing action A in order to achieve consequence B, then you can get the point of performing action A without my having to repeat action A many times.

This explanatory form of teaching is very efficient. With cultural communication so important in humans, we have become very good at explaining to others why we do things. Moreover, explanation and justification are similar functions. We develop the ability to justify to others why we do things.

We are concerned with what others think of what we say and do. As I read Sperber and Mercier, they argue that the natural function of reason is to try to gain respect and approval of others for our actions. I think that Sperber and Mercier do not give enough credit to the role of reasons in making teaching more effective. Imagine telling a child to look both ways before crossing a street without telling the child why they should do so. The child could perform the ritual exactly as directed and then walk right in front of moving car.

But the role of reasons in teaching does not address the enigma to which Sperber and Mercier refer. The enigma is that our reasoning process evolved to be biased rather than optimized to arrive at truth. Their explanation is that our reasoning process evolved as a mechanism to explain and justify our actions to others. The goal of reasoning is not to seek Truth but to defend our status. Biased reasoning is helpful for defending status. Bias is less helpful when we are trying to make decisions, but when we make decisions we are simply adapting our reasoning tool to a less natural context.

Sperber and Mercier make another claim, which is that when we argue with one another, we arrive at more reasonable conclusions than when we reason on our own. They say that this is because when we evaluate our own reasons we lack objectivity. They think we are more objective when we evaluate others’ reasons, so that our evaluations are more reliable. I do not find that persuasive. I think that part of defending our own reasons is attacking our opponents’ reasons, and I believe that we tend to be uncharitable to those who disagree with us. I am more inclined to ascribe the benefit of arguing to exposure to reasoning that we have not considered, rather than to a greater objectivity in hearing others’ points of view than in evaluating one’s own.

If reasoning evolved to justify our actions, then how do we get to a point where we use reasoning to make decisions? I think that the most consistent application of their idea would be to say that when we make decisions we anticipate having to defend our actions. As we go through this mental process, we may decide that some actions are unwise. Anticipating my wife’s reaction should I come home drunk, I stop drinking.

It could be that people with poor self-control have difficulty engaging in this exercise. That is, they either lack the ability to anticipate the reactions of others or they are less sensitive to such anticipated reactions.

It is interesting to note that I have often advised people in the throes of making a decision to imagine explaining that decision to a variety of other people. If you are thinking of quitting your job, imagine explaining that to your family, to close friends, to co-workers, and so on. I have suggested that such an exercise can help to clarify your thoughts.

Anyway, what occurs to me is that we obtain our theory of mind “outside-in” rather than “inside-out.” That is, by observing other people and listening to their reasons, we develop a theory of how our own minds ought to work.

Human Interaction

Posted on by

Whereas reason is commonly viewed as a superior means to think better on one’s own, we argue that it is mainly used in our interactions with others. We produce reasons in order to justify our thoughts and actions to others and to produce arguments to convince others to think and act as we suggest. We also use reason to evaluate not so much our own thought as the reasons others produce to justify themselves or convince us.

That is from Hugo Mercier and Dan Sperber, The Enigma of Reason: A New Theory of Human Understanding, a book that I have just started to read.

They call their theory of human reason “interactionist.” They argue that when economists or psychologists find what appear to be errors of human reasoning, we are taking human reason out of its natural context by focusing on individual choice. The authors apparently are going to argue that when human reasoning is used interactively, it works better than one might expect based on looking at the individual capacity to reason.

Regardless of how that argument works out, I think that economists would do well to recognize the interactive nature of human behavior. Treat economics as the study not of autonomous individuals (“human action”) but as the study of humans interacting in the context of production and exchange. Models based on the autonomous individual sometimes work well, but my guess is that once you get beyond the most basic supply and demand story, models become very dependent on the prevailing beliefs, cultural norms, and laws in the society to which one attempts to apply the model.

There is no need for economists to commit to what Deirdre McCloskey derides as the Max U view of human nature. We can instead accommodate the view that man is a cultural animal, and that we learn our habits and beliefs from others.

Taking human beings as social animals subject to various departures from pure rationality, market exchange is still a very defensible mode of human interaction. Markets help to organize large-scale specialization and cooperation. Markets are effective learning mechanisms. If Mercier and Sperber are going to claim that our collective brain works in spite of (and perhaps even because of) the flaws of our individual reason, then I am prepared to claim that the market system is often the best tool for taking advantage of that collective brain.