You Can't Infer Motives From Behavior

I occasionally comment on memes and other internet oddities that catch my eye. Usually these are political and/or tribal in nature and strike me as saying something the maker or distributor didn't intend to say. So it was that the following came across by FB feed this past weekend:

Now, this is a pretty typical piece of ideological tribalism. In fact, all it really is is a quote from one of Thomas Sowell's columns - which I'm sure are readily available elsewhere on the internet for his devoted readers - gussied up with some pictures of Democrats. As memes go, it's pretty unremarkable, although it does provide a FB- and twitter-ready way to spread around a sound byte that others might not see if they don't go track down Mr. Sowell's work and read it in its entirety.

What struck me about this otherwise ordinary bit of internet flotsam was the claim that it makes up front: the author claims to know what the real motives of a vaguely-defined group of people are, separate from what they themselves say about their motives. Leaving aside the question of whether a large group ("liberals") can all have the same set of motives, this claim violates one of the first rules I learned in graduate school: you can't infer motives from behavior.

The problem of motive has vexed political and social scientists for generations. All of us make inferences of motive all the time. We believe that the person who cut us off in traffic is a jerk who likes to push other people out of the way. We claim that the other political party has a hidden agenda not revealed in their public explanations of their policy preferences. We argue that Iran is building nuclear weapons so that it can annihilate Israel. And so on up and down the line.

The problem in every one of these cases is that there is always another explanation. The car in traffic may be driven by a desperate mother rushing her toddler to the ER. The folks writing that political party platform may genuinely believe that their policy prescriptions will lead to better outcomes for the welfare of the country. And Iran might be pursuing weapons capabilities for deterrence purposes in an effort not to change the status quo but to maintain it against what it perceives are aggressive outside forces.

The answers to the motives question obviously matter. As Robert Jervis pointed out more than a generation ago in Perception and Misperception in International Politics, there is no "safe bet" - misinterpreting aggressive motives as defensive ones is just as bad as seeing a defensive actor where there is in reality an aggressive one. There is no "play it safe" - all miscalculations can lead to ruin.

Unfortunately, the only thing we can see - the other actors' behavior - doesn't tell us what their motives are, because different motivations can lead to the same behaviors. Moreover, as Jervis and many others have pointed out, we tend to be very selective about which behaviors we see and how we interpret them. We tend to ascribe the motives to the other side that we want to. Our judgments are filled with wishful thinking.

This is actually what Mr. Sowell is engaged in. He would very much like what he says to be true, because it strengthens his own group identity and sense of superiority. It also makes him a great deal of money to say these kinds of things, which is motive enough for many kinds of behavior. This is not to say that he doesn't believe what he says - he may very well be sincere. But it does cast doubt on whether his claim should be taken seriously.

Indeed, nearly all claims about "the motives of the enemy" should be taken with a bag or two of salt. Sussing out motivations is a difficult and time-consuming task undertaken only with extensive data and rigorous analysis as removed as possible from wishful thinking and other biases - a task that requires extensive training and practice. And even then, practitioners of this art are sometimes wrong. Until we invent the machine that reads minds (and maybe even not then), this is the best we can get.

So what's the average layman to do? Political science can't teach people to make better judgments about motive, not unless people are prepared to undertake both a lot of study and a significant amount of personal introspection to control their own biases. But there is a much shorter punchline that everyone can reach quickly. If you care about actually being right about politics, ignore any and every statement that starts with "the real motives of X are..." Stop and consider the full range of possible motives, not just the ones that confirm your biases. It's a small step, and not an easy one. But if you care about the toxicity of political discourse, it's a necessary step for all of us.