Confirmation bias

While doing my part to keep the print media alive, I read Sharon Begley’s “The Limits of Reason .” Yes, I do see the irony in linking to the online version.

Begley begins her discussion by pointing out the obvious: humans are bad at reasoning. While she notes that psychologists have been documenting this from the 1960s, I would be remiss not to point out that philosophers have been discussing this since the beginnings of philosophy.

According to Begley, a new wave in philosophy and cognitive science is the view that such failures of reason have a purpose in that they enable us to “devise and evaluate arguments that are intended to persuade other people.”  She notes that  Hugo Mercier and  Dan Sperber claim that this poor reasoning is, in fact, an effective strategy aimed at winning arguments. This point is, of course, something that has been made by teacher of logic and critical thing for quite some time. For example, when I taught my first logic class a student asked me why people use fallacies. I  still use the two answers I gave back then. The first is that people are generally bad at reasoning. The second is that it works (as a way to persuade).

In philosophy, it is generally assumed that the goal of argumentation is to establish truth. However, Begley considers the idea that argumentation is about overcoming opposing views. That is, the goal is to persuade rather than to prove. The idea that people argue in the informal sense in order to persuade is, obviously enough, nothing new. However a standard approach in critical thinking is to distinguish between the goal of argumentation (truth) and persuasion (to get someone to believe). Part of making this distinction involves  pointing out that people often confuse persuasion and argumentation. As such, to say that the goal of argumentation is to  overcome opposition is to merely call persuasion by the name “argumentation.” Since there are two distinct goals and methods, it certainly makes excellent sense to maintain a distinction in terms.

To anticipate some obvious objections, arguments can be used to persuade and persuasive techniques can be used in arguments. However, the fact that a hammer can be used to pound in screws does not make it a screwdriver.

Begley then turns to a specific error, that of confirmation bias. As folks who read blogs have surely noticed, people tend to focus on evidence that supports their view and ignore that which goes against it. She notes that this serves a useful purpose when “arguing” because  “it maximizes the artillery we wield when trying to convince someone…” Mercier even claims that “it contributes to effective argumentation.”

Having observed this numerous times, I do agree that it can be an effective persuasive tool. Of course, it depends on the opposition not being prepared with evidence and also on the ignorance of the target. Someone who is aware that the person using this artillery is selectively focusing on supporting evidence will hardly find this approach convincing (either logically or rhetorically).

As a tool of argumentation in the proper sense, it is obviously not effective. After all, falling victim to the confirmation bias is not an effective way to establish truth. If someone wants to say that the goal of argumentation is persuasion, then that is fine. However, we will need a new term for what it is that we do when we try to establish truth. Sticking with the spirit of the thing, perhaps we should call that “persuasion.”

Begley moves on to note the value of motivated reasoning. An example of this is when a person looks very hard for flaws in a blog that supports a position she disagrees with. Another example is when people dismiss evidence that goes against their view. From a logical standpoint, falling victim to this is an error since it will impede an objective assessment. However, as Begley points out it does have its advantages. Someone who falls victim to this will tend to be more effective in finding flaws. Of course, there is the concern that the flaws might be imagined as opposed to real. There is also the concern that evidence will simply be ignored (as in Begley’s example of the Birthers who refuse to accept Obama’s birth certificate as real).

Begley finishes with a last example, what she calls the “sunk-cost fallacy” (often presented as a slippery slope variant). This fallacy occurs when a person believes that she should or must follow a course of action because she is already embarked on that course.  This, as she notes, is a rather effective persuasive device. For example, this sort of fallacy was used to “argue” in favor of re-electing George Bush. As another example, this fallacy is sometimes called the “Vietnam fallacy” and that war nicely illustrates how persuasive it can be. However, it is clearly bad logic.

While Begley does not go into any detail, the subtitle of the essay “Why evolution may favor irrationality” suggests her overall point. The idea seems to be that the dominance of bad reasoning can be explained on the grounds that bad reasoning confers an evolutionary advantage.

Based on my own experience studying and teaching critical thinking, I can attest to the persuasive power of poor reasoning and fallacies. As I mentioned above, I tell my students that one of the main reasons people use fallacies is because they work marvelously as persuasive devices. This, of course, gives those who effectively use such methods an advantage in terms of convincing others. Presumably this provides a reproductive advantage so that people who are bad at reasoning but good at persuading tend to be selected.

However, fallacies and poor reasoning are obviously not very effective when it comes to getting things right.  In fact, the fallacies and errors Begley used as examples tend to lead people towards disaster and death.  For example, the sunk-cost fallacy can keep people stupidly grinding away on a failed plan, war, or way of life.  In a nutshell, our greater persuasive skills can overcome our inferior logic skills and convince us to do remarkably unwise and stupid things.

Obviously, poor reasoning has not killed off the species…yet. However, it is interesting to speculate what the long term consequences might be if the hypothesis that Begley considers is correct.

While the evolution folks tend to focus almost entirely on what they think are the evolutionary advantages of our traits, they should give due consideration to the negative aspect of natural selection. To be specific, perhaps our tendency to reason poorly and to be persuaded by poor reasoning are traits that will result in our species being selected out of the evolutionary game. Perhaps a long time hence clever academics from whatever species succeeds us will be writing essays about how evolution weeded out the irrational animal known as man.  I imagine one of the sentences would be something like this: “Homo sapiens became extinct largely because humans were very good at persuading each other to believe very stupid things and very bad at telling what was, in fact, very stupid.”