Thursday, June 30, 2011

Argument, Truth and the Social Side of Reasoning By GARY GUTTING NYTimes

The Stone is a forum for contemporary philosophers on issues both timely and timeless. Arguments, Philosophy, reasoning

The Stone is featuring occasional posts by Gary Gutting, a professor of philosophy at the University of Notre Dame, that apply critical thinking to information and events that have appeared in the news.

Philosophers rightly think of themselves as experts on reasoning. After all, it was a philosopher, Aristotle, who developed the science of logic. But psychologists have also had some interesting things to say about the subject. A fascinating paper by Dan Sperber and Hugo Mercier has recently generated a lot of discussion.

Reasoning is most problematic when carried out by isolated individuals and most effective when carried out in social groups.

The headline of an article in The Times about the paper— echoed on blogs and other sites — was, “Reason Seen More as Weapon than Path to Truth, ” a description, that implied that reason is not, as we generally think, directed to attaining truth, but rather to winning arguments. Many readers of the Times article thought that this position amounted to a self-destructive denial of truth. The article itself (though perhaps not the abstract) suggests a more nuanced view, as the authors tried to explain in replies to criticism. In any case, we can develop an interesting view of the relation between argument and truth by starting from the popular reading and criticism of the article.

Sperber and Mercier begin from well-established facts about our deep-rooted tendencies to make mistakes in our reasoning. We have a very hard time sticking to rules of deductive logic, and we constantly make basic errors in statistical reasoning. Most importantly, we are strongly inclined to “confirmation-bias”: we systematically focus on data that support a view we hold and ignore data that count against it.

These facts suggest that our evolutionary development has not done an especially good job of making us competent reasoners. Sperber and Mercier, however, point out that this is true only if the point of reasoning is to draw true conclusions. Fallacious reasoning, especially reasoning that focuses on what supports our views and ignores what counts against them, is very effective for the purpose of winning arguments with other people. So, they suggest, it makes sense to think that the evolutionary point of human reasoning is to win arguments, not to reach the truth.

This formulation led critics to objections that echo traditional philosophical arguments against the skeptical rejection of truth. Do Sperber and Mercier think that the point of their own reasoning is not truth but winning an argument? If not, then their theory is falsified by their own reasoning. If so, they are merely trying to win an argument, and there’s no reason why scientists — who are interested in truth, not just winning arguments—should pay any attention to what they say. Sperber and Mercier seem caught in a destructive dilemma, logically damned if they do and damned if they don’t.

Philosophical thinking has led to this dilemma, but a bit more philosophy shows a way out. The root of the dilemma is the distinction between seeking the truth and winning an argument. The distinction makes sense for cases where someone does not care about knowing the truth and argues only to convince other people of something, whether or not it’s true. But, suppose my goal is simply to know the truth. How do I go about achieving this knowledge? Plato long ago pointed out that it is not enough just to believe what is true. Suppose I believe that there are an odd number of galaxies in the universe and in fact there are. Still, unless I have adequate support for my belief, I cannot be said to know it. It’s just an unsupported opinion. Knowing the truth requires not just true belief but also justification for the belief.

Rational agreement, properly arrived at, is the best possible justification of a claim to truth.

But how do I justify a belief and so come to know that it’s true? There are competing philosophical answers to this question, but one fits particularly well with Sperber and Mercier’s approach. This is the view that justification is a matter of being able to convince other people that a claim is correct, a view held in various ways by the classic American pragmatists (Peirce, James and Dewey) and, in recent years, by Richard Rorty and J├╝rgen Habermas.

The key point is that justification — and therefore knowledge of the truth — is a social process. This need not mean that claims are true because we come to rational agreement about them. But such agreement, properly arrived at, is the best possible justification of a claim to truth. For example, our best guarantee that stars are gigantic masses of hot gas is that scientists have developed arguments for this claim that almost anyone who looks into the matter will accept.

This pragmatic view understands seeking the truth as a special case of trying to win an argument: not winning by coercing or tricking people into agreement, but by achieving agreement through honest arguments. The important practical conclusion is that finding the truth does require winning arguments, but not in the sense that my argument defeats yours. Rather, we find an argument that defeats all contrary arguments. Sperber and Mercier in fact approach this philosophical view when they argue that, on their account, reasoning is most problematic when carried out by isolated individuals and is most effective when carried out in social groups.

The pragmatic philosophy of justification makes it clear why, even if we start from the popular reading of their article, Sperber and Mercier’s psychological account of reasoning need not fall victim to the claim that it is a self-destructive skepticism. Conversely, the philosophical view gains plausibility from its convergence with the psychological account. This symbiosis is an instructive example of how philosophy and empirical psychology can fruitfully interact.

No comments: