Submitted by Daniel Pi on Thu, 11/26/2015 - 09:35

Protective Tin Foil HatsIn a much talked about and frequently cited paper, Cass Sunstein and Adrian Vermeule explore the topic of conspiracy theories as a social problem, speculating on possible remedies to their emergence and proliferation (Note: Sunstein has also recently published a book touching on the topic—however I shall restrict my remarks to the 2008 paper only). I should be the last person to defend the activities of conspiracy theorists, insofar as I regard their practices to be epistemically unsound, profoundly unscientific, and unbearably annoying. Moreover, I am inclined to regard such individuals with extraordinary personal contempt, for as any sane person who has had the misfortune of engaging one of these specimens in conversation will already know, they possess a singularly potent combination of aggressive obnoxiousness and howling ignorance. 

And yet perversely, I find myself compelled now to argue against Sunstein and Vermeule, scholars for whom I have much respect, and in favor of the conspiracy theorists, lunatics for whom I feel nothing but loathing. 

I have two points of contention. First, Sunstein and Vermeule claim that conspiracy theories can cause harm. Second, Sunstein and Vermeule approach the phenomenon from the perspective of behavioral economics and behavioral psychology. Neither of these seem to me quite right.

Do conspiracy theories cause harm?

Of course, Sunstein and Vermeule are careful in pointing out that many (perhaps most) conspiracy theories are—however annoying—essentially harmless. They restrict their discussion to the problem of potentially harmful conspiracies. In Section II (A) of their paper, the authors articulate their case that at least some conspiracy theories can cause serious harm. They provide several examples, claiming that the harmful consequences of allowing conspiracy theories to fester include (1) fascilitating Al-Qaeda recruitment; (2) inspiring the bombing of the Oklahoma City Federal Building; (3) engendering resistance to post-9/11 antiterrorist policies; (4) encouraging a general distrust of governments.

Do these examples really illustrate "harmful consequences"? In the first case, suppose it were true that Al-Qaeda recruiters employ conspiracy theories about the West's plans for global domination in their pitch to potential recruits. It is not entirely clear to me that claims of that sort should even count as "conspiracy theories." Historically, it was never really a secret plan that the West sought to conquer the world during the era of colonial expansion. Nor today is it really a secret plan of Western nations to spread Western culture (such as democracy, free markets, and McDonald's restaurants). Yet let us suppose there's more to the Al Qaeda recruiter's story than merely observing the open machinations of the West. Let us imagine the impoverished, uneducated, angsty 20-something Muslim man, disaffected and powerless in his society, fanatical (if ignorant) in his religious devotion, seeking some source of purpose in his otherwise miserable life. He objects to the influence of Western values, and he sees modern Americans and Europeans as no different from the colonialists of centuries past. Are we really to believe that some additional conspiracy narrative is needed to heat this volatile brew of craziness past the boiling point?

In short, there's no doubt that terrorist recruitment is a harmful thing, though it requires rather more work than Sunstein and Vermeule offer to claim that conspiracy theories have been instrumental at all here. The only authority they cite is an anonymous State Department official in charge of anti-disinformation (who would have some personal interest in playing up the potential harmfulness of disinformation), who claimed, "a great deal of harm can result 'when people believe these lies and then act on the basis of their mistaken beliefs.'" Such vague allusions to potential harms are difficult to take seriously and strike me as little more than scaremongering.

On the second example, again, it seems implausible to me that the conspiracy theory portion of Timothy McVeigh and Terry Nichols' nutty beliefs were instrumental in their path toward violence. No doubt they believed in conspiracy theories. Yet their violence seems to have been motivated less by these false beliefs than by their political preferences. Once again, while I will easily grant that the example showcases harm, I question whether it is a consequence of a belief in conspiracy theories.

On the third and fourth examples, I am unconvinced by the harm portion. Is it such a bad thing after all for people to be suspicious and distrustful of their government? No doubt that it is in many cases wasted energy. To the extent that government officials harm the public interest, I am thoroughly convinced it is due not to nefarious ulterior motives, but rather almost always due to incompetence and ignorance. The stupidity of human beings is almost always a more likely explanation for bad government actions than elaborate, wilfull conspiracies against the common good. But is such distrust and suspicion harmful in the sense that Sunstein and Vermeule seem to be interested? It is hard to cash out the claim in any concrete way. 

At best, Sunstein and Vermuele fail to convince on this rather essential point. When flakey lunatics believe conspiracy theories, they seem merely to be annoying and harmless. When dangerous lunatics get a hold of such beliefs, it seems likely they'd be dangerous with or without the conspiracy theories. In either case, it does not seem that belief or disbelief in conspiracy theories is likely to have much bearing on whether harm will arise. 

Moreover, though I am loath to admit their positive value, it is worth noting that on some rare occasions, the conspiracy theorists have been right. What are now well-known, well-documented historical events were at some point "secret government plots." For example, the Tuskegee Experiment, the Manhattan Project, Project MK-Ultra, the Congressional Testimony of Nayirah, Operation Mockingbird, Watergate, NSA spying, and the Iran-Contra Affair. Though the lunatic conspiracy theorists did little to expose these genuine conspiracies, a general attitude of tolerance toward such speculation has obvious benefits—at least insofar as we should like to expose genuine real-world conspiracies.

Presumably, Sunstein and Vermeule, inasmuch as they restrict their discussion to harmful conspiracy theories, also restrict their discussion to false conspiracy theories. Yet it is rather the point that we don't know which ones are true and which ones are false. Sensible people can be reasonably certain that the vast, vast majority of conspiracy theories floating around in the wide world are false. I do not suggest that we ought to take an agnostic attitude toward conspiracy theories, merely because 0.01% of them are true. However, if we are taking stock of their costs and benefits, I am simply pointing out that benefits, however marginal, are non-zero. Combined with my suspicion that the costs are zero, it seems as though we have everything to lose and nothing to gain by acting against conspiracy theories.

Are behavioral economics and behavioral psychology the right way of looking at conspiracy theories?

The second claim that Sunstein and Vermeule implicitly make is that various cognitive biases and psychological effects can explain the emergence and proliferation of conspiracy theories. Yet it is difficult to believe that people are taken in by such elaborate narratives (as conspiracy theories tend to be) due to mere cognitive bias. Cognitive bias influences behavior subtly and unconsciously. The classic examples (e.g., Kahneman and Tversky's framing effects) do not seem to be the sorts of things that can lead a person to believe a complex narrative so committedly. They may cause inconsistencies in one's everyday choices, but they do not seem to be the kind of thing that would lead a person to adopt a complicated false narrative about the world. It does not take much to recognize a fundamental difference between, say, the anchoring effect and the belief that Elvis Presley faked his death.

I do not claim that nothing is wrong with conspiracy theorists—there surely is much wrong with them. But I think it is not so plausible that cognitive biases offer much of a diagnosis. What seems the simpler and more straightforward explanation (and for me, the one that rings truer) is simply that people are stupid and easily taken in by argumentative fallacies. 

When someone infers: (premise 1) Socrates is mortal, (premise 2) All men are mortal, therefore (conclusion) Socrates is a man, this is indubitably a fallacious argument, but it is not a bias. There is no cognitive effect at play, unless we count simple, mundane stupidity as a cognitive effect. Such a person simply lacks either the intelligence or the training to detect the logical fallacy (employing the false inferential rule that if X is P, and all Y's are P, therefore X is a Y).

To hear the conspiracy theorist's rant, one is struck not by how afflicted his reasoning is with cognitive biases, but rather by the string of unending logical fallacies issuing from his big mouth. To the extent that such conspiracy theories spread, the mechanism seems far stupider and simpler than Sunstein and Vermeule imply. A stupid listener hears the reasoning of the stupid speaker and becomes convinced of the fallacious inference. Believing that he is now privy to some secret knowledge (as all conspiracy theories are supposed to reveal), he repeats the information to yet more stupid people, who are convinced by the same bad reasoning (with possible emendments and amendments of more bad reasoning), who then repeat it for the same reasons. Secrets of any sort (whether true or false) seem to be higher value topics, for reasons that elementary microeconomics economics is well-equipped to explain. And thus the conspiracy theory is born, and thus it spreads.

Learning to Live with Lunatics

And yet suppose Sunstein and Vermeule are right on these foregoing points. It seems to me that any attempt to combat conspiracy theories will merely add fuel to the flame. Sunstein and Vermeule address the potential for backlash effects, but I find these prevarications unconvincing. Any proactive effort on the part of the government, whether coercive or (as Sunstein and Vermeule propose) subtly nudging, will invariably be taken by conspiracy theorists as a clear confirmation of their suspicions. Indeed, Sunstein and Vermeule's paper seems to be cited by conspiracy theorists at least as often as it is by scholars, with the accusation that they too are part of "the conspiracy" (See, e.g., this and this). The conspiracy theorists have in particular seized upon one of Sunstein and Vermeule's suggested remedies, which they refer to with the ill-chosen term, "cognitive infiltration," which has, unsurprisingly, only managed to stoke the flames of speculation. Never mind how implausible and ineffective such a proposal might be (I believe it would be quite), given the perverse consequence of merely suggesting it, it can only be imagined what response actual implementation would engender. 

Suffice it to say, as horrible as they are, my position is that we must simply learn to live with the crazies and their crazy ideas. For the foregoing reasons, I am doubtful that conspiracy theories are terribly harmful after all, and I doubt that their causes are anything more than routine human stupidity and the perceived high informational value of "secrets." And insofar as that's right, I can think of no promising remedy, even if one were felt to be warranted. It would be easier to hold the ocean back with one's hands than to stop stupid people from believing in false things.