Tag: rationality

Was it over when the Germans bombed Pearl Harbor?

Apparently so.

The NYT has a book review that treads the familiar ground of why Americans are so dumb when it comes to supposedly vital information about the world. The old X% of Americans support a war in Iraq, but only Y% can actually find Iraq on a map… This is especially upsetting to the classic, old-school curmudgeons who think that a classic education involves knowing a lot of key names, dates, and classical cultural references.

This particular thing, however, does not bother a number of my colleagues (I’ll admit I’m a bit more on the fence about it than others…).* The argument goes: such random trivia, like where Iraq is on a map or who was the 13th President of the United States** is readily deposited in great knowledge warehouses such as Google. If you need to know it, just look it up. More important to teach and more important to learn is the skill to look up this bit of information when you need it. Teach the skill and students will always be able to get what they need. In fact, such skills are part of the “skills revolution” that actually makes American kids/workers smarter than other kids/workers around the world.

But such pedagogy seems in the minority. The more troubling trend in American culture is that:

But now, Ms. Jacoby said, something different is happening: anti-intellectualism (the attitude that “too much learning can be a dangerous thing”) and anti-rationalism (“the idea that there is no such things as evidence or fact, just opinion”) have fused in a particularly insidious way.

Now you have talking heads on TV screaming at each other, often seeming to make stuff up to support a point, justified under the guise of “opinion” where every opinion counts (that’s your opinion, and you’re entitled to it, but I believe otherwise…).

The anti-intellectualism / anti-rationalism combination undermines the knowledge gathering skills that make the need for memorization of random facts less important, thereby removing any and all facts from the evaluation of argument. When facts are widely known and memorized, they constrain a discussion to those sets of facts. Classically educated individuals talk about the classics. They also set truth conditions for claims–conforming to the shared facts.

The skills revolution of what our library folks call information literacy (the ability to locate, recognize, and evaluate relevant information) performs a similar task, albeit on another level. It too sets constraints on argument, though from a more methodological perspective. But, it still sets truth conditions for claims–demonstrated collection of relevant and accurate information from appropriately authoritative sources. To take Global Warming, for example (Gore’s movie was on tonight and I watched the last hour of it as I ate dinner….) information literacy skills lead you to the point that scientists accept global warming as a fact, as all peer reviewed scientific papers agree to this. If you know how to search for and evaluate scientific information, this fact is easily obtainable and the position is hard to dispute. But, when you ignore this skill, and adopt the anti-intellectual, anti-rationalist position lamented here, Global Warming becomes an opinion, still open for debate.

All of which is to say, I don’t worry so much about the fact that kids don’t know X or Y fact. But, I do worry when kids can’t tell the difference between analysis based on a logical arrangement of well researched facts and opinion based on personal proclivities with facts selected to fit that view.

*I think that a number of so called facts are actually more conceptual than we give them credit for, and thus, relatively important to know. Knowing where Iraq “is” is not just about a spot on a map, it also is linked to questions of strategy, culture, politics, and the like. Certain other claims, like “Iran is a player in Iraq” are greatly aided if you know that Iran and Iraq share a rather large land border.

**Millard Fillmore–I bet you didn’t know either until you googled it.

***Germans? Forget it, he’s rolling….


Maybe they’re more like us than we thought–UPDATED

[UPDATED — see last paragraphs below the fold, which take into account Bush’s latest comments on Iran.]

Startling news out of the American intelligence community today:

Iran’s “decisions are guided by a cost-benefit approach rather than a rush to a [nuclear] weapon irrespective of the political, economic and military costs.”

Hmm. After months, even years, of “Ahmadinejad, like all of those people, is a crazy ideologue who does what he does because he hates our freedoms,” maybe things are changing a little bit? Maybe we’re on a track towards regarding Ahmadinejad to be — wait for it — human, and therefore amenable to the kinds of diplomatic techniques that work so well when we area dealing with other human beings? Perish the thought.

Sarcasm aside, this is actually a fascinating shift. It’s important in policy terms, obviously, because it makes a great deal of difference whether we think we’re dealing with someone who is strategic and cost-sensitive or with someone who is fanatical and cost-insensitive. Politically, it says something interesting that we’re now officially — at least at the level of the National Intelligence Estimate — regarding Ahmadinejad, or at least the country that he heads up, to be more or less rational. But what I find most interesting here is less the shift itself and more the somewhat misleading way that the shift presents itself: it claims to be an empirical conclusion, when it’s actually an ethical presumption.

There’s been a lot of controversy about this in IR over the past several decades, so let me spell this out a bit so that I’m not misunderstood. One of the perennial problems in American social science is that scholars have a tremendously pervasive tendency to conflate presumptions and results, especially when it comes to issues like “rationality.” Why is this? Well, it’s probably overdetermined, but one of the key reasons in my estimation is that American social scientists are, by and large, “irrational Lockeians,” which means both that they generally equate ‘freedom’ and ‘agency’ per se with their liberal individualist variants, and that they generally believe that such notions can be empirically validated somehow. Of course, outside of political theory, most American social scientists don’t generally say this explicitly; instead, they illustrate their commitments by deploying models of strategic action that depend on individuals making self-interested calculations, reducing social forces to individual-level decisions and (especially cognitive) mechanisms, and (mis-)treating analytical constructs like “the rational actor” as though they were empirically falsifiable hypotheses.

This last one is particularly important to the present issue. Somehow, American social science has gotten it into its collective head that we can somehow prove that a given actor is rational, when this is in fact a silly and misleading exaggeration of the powers of empirical research. One cannot prove than an actor is rational, because rationality is an interpretive presumption that we often use to make sense of the things that actors do in the first place. We assume that actors are goal-directed, and that they calculate various means of achieving their ends with an eye to achieving those ends more or less effectively — indeed, we would have a very hard time making sense of any social action if we didn’t make that assumption, because virtually nothing that we would see would cohere in any meaningful way.

Now, if one wants to define “rationality” in more formal and precise terms, then yes, one can collect evidence about (say) the consistency of an actor’s preferences and use that as a way of determining empirically whether an actor is rational. But the NIE is not, as far as I can tell, operating at that level of subtlety; instead, it’s distinguishing between rationality defined as cost-sensitivity, and irrational motivations like ideological commitment. And here my point about rationality remains unshaken, since there is no way to definitively say whether an actor is behaving rationally or irrationally. If you point to a piece of evidence that you cite as representing cost-sensitivity and some kind of calculation of self-interest, I can point to the same piece of evidence and argue that it’s really generated by the actor’s ideology and her or his (or its) understanding of that ideology. And neither of us would be “right,” since neither of us can pry open minds and see what’s taking place there that is generating the social action(s) that we’re trying to explain.

So: you say “rationality,” I say “ideology.” This is strengthened if we eliminate the straw-man argument that ideological causation means a fanatical pursuit of a single goal regardless of cost or consequence; virtually no actor is that fanatical, especially since outside of “total institutions” like prisons and militaries and some cults, ideological commitments are susceptible to a range of interpretations. Indeed, if you ever listen to a group of ideologues debating what to do about something, what they are generally going to be talking about are the subtle nuances of their shared commitments — and they’re going to be dwelling on differences and distinctions that might seem incomprehensible to an outsider, but those distinctions point in different directions for action. You probably won’t be able to justify any course of action in the terms of a given ideology (indeed, if you could, then ideology would be completely epiphenomenal and more or less irrelevant), but you can generally justify a few different courses of action that might bear a family resemblance but are different and distinct all the same. To a rationalist, this looks like actors calculating their interests and proceeding; to a culturalist, it looks like contingent combinations of rhetorical resources entailing different ways of proceeding. There’s no way to empirically rule out one or the other.

And no, you can’t do this by predicting the kind of action that an actor would take if they were “really” motivated by one or another set of concerns, and then seeing what they actually do; a committed partisan of one of the other model of actor-hood and social action can always, always make sense of action in their preferred terms. To take the matter at hand, suppose we are dealing with a state that is spending money on the construction of nuclear reactors, is involved in the support of various insurgent groups, has declared an intention to develop nuclear weapons, and is engaged in a regional power-struggle with other states, some of which have nuclear weapons themselves. We might predict that, under the circumstances, a rational state might try to acquire or develop nuclear weapons if the cost wasn’t too high; based on that prediction we might apply various kinds of pressure to stop the state from acquiring nuclear weapons; and if the state does in fact refrain from going nuclear, we might pat ourselves on the back for “getting their motivations right” and using them to plan our strategy. But this is quite an an unjustified conclusion, inasmuch as a culturalist could explain the state’s actions in terms of changing social norms, the importance of international legitimacy to prop up a domestic regime, and the ambiguous symbolic value of nuclear weapons themselves. Predictions based on interpretive presumptions simply can’t be used to falsify the initial presumption, because that presumption is built into the very reading of the evidence itself.

Now, as I said at the outset, it makes a great deal of difference whether we assume that a given actor is cost-sensitive or not. If we don’t think that another actor is calculating any sorts of costs or benefits, then our actions towards them are not likely to be designed to alter those calculations; instead, they would probably involve either implacable opposition or warm embrace, depending on whether the course of action was a “good” (from our perspective) one or not. But we can’t know with certainty what motivates any particular actor we encounter, an observation that leads to the classical realist recommendation to always take care of your own security and to be on guard against possible attacks from any quarter. But even classical realists presume that states, or their leaders, are generally “rational” . . . and this opens the possibility of dropping one’s guard somewhat, depending on the circumstances. In practice, almost all of us make presumptions about other actors’ motives and intentions all the time, despite the fact that we can’t ground those presumptions on any definitive empirics. Since those presumptions don’t come from empirical observation, where do they come from?

Ultimately, I’d say that these presumptions are ethical rather than empirical, and have a lot more to do with how we think about various self-other relationships than they do with particular cases. Tzvetan Todorov argues that the question of whether we think that the other is like or unlike us is an entirely separate component of the self-other relationship than the question of how much we know (or don’t know) abut the other; the two can’t be collapsed into one another. I can know little about another actor and still presume that she or he or it functions more or less the same way that I function. Indeed, I’d posit that whatever data about the other that I generate based on that presumption can’t even in principle give me any sense of whether the presumption is empirically accurate: if I really think that you operate the way that I do, and if I am concerned to evaluate that characterization, then I’ll keep finding evidence that “proves” that you do in fact operate this way. And ditto, in reverse, if we start out thinking that the other is completely unlike us.

Given this, I think we have to ask a different question about things like this than we conventionally ask when formulating foreign policy to deal with other actors. Instead of asking what “they” are thinking or what motivates “them,” we have to ask whether we are comfortable with the ethical stance that we’re taking. And that ultimately means asking about our own identity as an actor. Are we comfortable treating Iran as a “wholly Other” that doesn’t think about costs when pursuing a weaponized nuclear capability? What kind of social and political work is an assumption like that doing in our present society? Conversely, what is at stake ethically in regarding Iran to be basically like us in the way that it makes nuclear policy decisions? Which can we live with better: the older estimate of Iran as fanatical, or the new one of Iran as rational? (Can we maybe have a third one: Iran, like all actors at whatever scale, as culturally strategic rather than formally rational or fanatically irrational?) Yes, it makes a difference which we choose, since they point to different policies. But we can’t choose on empirical grounds, so we’re left choosing on ethical grounds. It would be best, I think, if we were more explicit about that.

UPDATE 5 December: According to the New York Times, Bush had this to say about Iran at a news conference yesterday:

“Look, Iran was dangerous, Iran is dangerous, and Iran will be dangerous, if they have the knowledge necessary to make a nuclear weapon,” Mr. Bush said during a news conference dominated by questions about the fallout of the assessment, known as a National Intelligence Estimate. “What’s to say they couldn’t start another covert nuclear weapons program?”

I can think of no better illustration of the epistemological point I was trying to make. Determinations about the fundamental and underlying character of an actor aren’t empirical considerations, but ethical ones: once that determination has been made, empirics can be interpreted, but there’s no way to logically go in reverse order even if the grammar of the statements made might lead one to believe that this is the case. Bush didn’t look at the evidence and decide that Iran was behaving rationally; Bush decided that Iran was a fanatical, ideologically-motivated opponent of the USA (and “our freedoms,” presumably) and then placed the empirical evidence into that framework (so that the mere potential that Iran might at some unspecified point in the future develop the capacity to produce a nuclear weapon appears sufficient reason for concern). This is no different than what all of us do all the time when making sense out of any actor’s actions, but it does point to the inadequacy of assuming that these debates are somehow purely based on morally neutral “factual” evidence.

Indeed, Bush’s comment might even be implicitly referencing a specifically Christian theology. The cadence of “Iran was dangerous, Iran is dangerous, and Iran will be dangerous” sounds strikingly similar to a well-known Christian doxology called the Gloria Patri, part of which runs:

. . . as it was in the beginning, is now, and ever shall be . . .

Even if this is stretching a bit (and in Bush’s case I don’t think that it is), the point remains: these judgments are never made on empirical grounds alone.


Assume A Rational Actor….

This was just too good to pass up.

With immigration a growing political issues, The New York Times looks back and asks— remember NAFTA, like 14 years ago? Wasn’t that supposed to stop illegal immigration by building up the Mexican economy through free trade?

Why didn’t Nafta curb this immigration? The answer is complicated, of course. But a major factor lies in the assumptions made in drafting the trade agreement, assumptions about the way governments would behave (that is, rationally) and the way markets would respond (rationally, as well).

Neither happened…

Ahh, the dreaded rationality assumption, rearing its ugly head again.

First, it was assumed that the government would respond rationally to the new incentives provided by NAFTA:

When Nafta finally became a reality, on Jan. 1, 1994, American investment flooded into Mexico, mostly to finance factories that manufacture automobiles, appliances, TV sets, apparel and the like. The expectation was that the Mexican government would do its part by investing billions of dollars in roads, schooling, sanitation, housing and other needs to accommodate the new factories as they spread through the country.

It was more than an expectation. Many Mexican officials in the government of President Carlos Salinas de Gortari assured the Clinton administration that the investment would take place, and believed it themselves, said Gary Hufbauer, a senior fellow at the Peter G. Peterson Institute for International Economics in Washington who campaigned for Nafta in the early 1990s.

“It just did not happen,” he said.

Next, it was assumed markets (ie farmers in the agricultural market) would rationally respond to the new incentives offered by NAFTA:

The assumption was that tens of thousands of farmers who cultivated corn would act “rationally” and continue farming, even as less expensive corn imported from the United States flooded the market. The farmers, it was assumed, would switch to growing strawberries and vegetables — with some help from foreign investment — and then export these crops to the United States. Instead, the farmers exported themselves, partly because the Mexican government decided to reduce tariffs on corn even faster than Nafta required, according to Philip Martin, an agricultural economist at the University of California, Davis.

“We understood that the transition from corn to strawberries would not be smooth,” Professor Martin said. “But we did not think there would be almost no transition.”

Two key assumptions, both based on a particular model of a rational, homo economicus, model of actorhood. The State pours investment into areas that lead to the greatest benefit in terms of national product growth. Farmers shift crops to farm what provides them the greatest return at the market. Any rational actor facing these market pressures and incentives would choose this course.

Except that they didn’t. The Mexican government–and here’s the key: despite individuals within that government individually believing that they would–never was able to reform its domestic spending priorities. Farmers, rather than shift to farming a new crop, simply followed other relatives into the United States, creating an immigration network.

Finally, the steady flow of Mexicans to the United States has produced a momentum of its own — what Jeffrey Passel, a demographer at the Pew Hispanic Institute, calls a “network effect,” in which young Mexicans travel to the United States in growing numbers to join the growing number of family members already here.

This brief NY Times article reveals the core difficulty and flaw of rationality assumptions and the real consequences of building policy on social science theory based on rationalist models of human behavior. Quite simply, as the scholarship of most of the contributors of this blog (and a large number of our friends and colleagues) has shown, identity and therefore rationality are social constructs that depend on rules of legitimacy. Government leaders and farmers ask not what offers the greatest return, but rather, “Who am I?” and what do I do now? Absent a domestic political climate that could reform the rules and legitimacy of government spending practices, the investment envisioned by NAFTA couldn’t take place. Governments don’t just “rationally” decide to reallocate funds. Anyone who has ever looked at a defense spending bill can tell you that. Its a political process, and winners and losers in politics are not determined by the same rules as rational returns on investment in economics.

Questions of livelihood are approached with the same process. Farmers suddenly unable to farm corn don’t just say well, what crop would sell. No, they say I’ve lost my livelihood, what do I do now. The look to others who define their identity–family–for opportunity, and see it in America. Hence, the network patterns of immigration. Its not a “rational” response to incentives, rather, its a network push and pull bringing certain people to the US and not others.


“We have indeed had one disappointment after another on this score,” Mr. Rodrik said, noting that the same assumption about government spending is part and parcel of the agreements, now before Congress, with Columbia, Peru and Panama.

Perhaps Congress and the USTR should not build in such assumptions to these deals–rather, appreciate that other logics might inform government spending, migration, investment, and trade patterns, and allow for enough flexibility in the deal to address this.

And, perhaps some of the social scientists out there who continually trumpet theories and policies based on those theories that assume a Rational Actor should take a time out and really think about the consequences of what they’re doing.


© 2021 Duck of Minerva

Theme by Anders NorenUp ↑