Alternative Title: Douthat’s Achilles’ Heel
(Note: This is the fourth blog post in a multi-part series.)
When are conspiracies real? We have been reviewing Ross Douthat’s recent New York Times column on this question, “A Better Way to Think about Conspiracies,” which contains a multi-part test for deciding which alleged conspiracies to keep an open mind about. Aside from simplicity and stochastic selectivity, Douthat also tells us to “avoid theories that seem tailored to fit a predetermined conclusion.” This key criterion appears to be inspired by Sir Karl Popper’s famous falsifiability principle. (Professor Popper, pictured below, was an influential 20th-Century philosopher of science who introduced the concept of falsifiability in his 1934 book Logik der Forschung, which was further revised and translated into English in 1959 as The Logic of Scientific Discovery.)
In brief, the Popperian concept of falsifiability refers to “testability” or “refutability”: the capacity for some proposition, statement, theory, or hypothesis to be tested and proven wrong, i.e. contradicted by evidence or “falsified.” As a result, conspiracy theories that are designed to confirm our pre-determined conclusions don’t deserve our respect; theories need to be falsifiable before they can be taken seriously. At first glance, Douthat’s version of this falsifiability criterion–which was originally used by Karl Popper to separate science from pseudo-science–would appear be an especially useful technique to distinguish “legitimate” or plausible conspiracy theories from imagined or invented ones. But upon closer inspection, one can make a psychological or “Kuhnian” objection to Popper’s falsifiability principle in the context of conspiracy theories. Such theories are more like religious beliefs: they are often a product of people’s deep-seated intuitions and implicit assumptions about the world, and those intuitions, beliefs, and assumptions are generally impossible to test or “falsify”!
As it happens, Douthat himself concedes in his NY Times essay that “to be a devout Christian or a believing Jew or Muslim is to be a bit like a conspiracy theorist, in the sense that you believe that there is an invisible reality that secular knowledge can’t recognize ….” In other words, religious beliefs, like many conspiracy theories, are usually the product of one’s private intuitions, not rational deliberations. These intuitions often reflect one’s most deeply-held beliefs and thus cannot be tested or falsified in any meaningful sense. To return to my favorite historical example, consider the German “Stab-in-the-Back” Myth from the Weimar Republic era (1919 to 1933). Not only is this conspiracy theory relatively simple and selective; it also solves a major mystery: why did Germany lose WWI? Alas, this myth is not amenable to rational analysis because it is unfalsifiable. If you really believe that the German Army was stabbed in the back by internal enemies, no amount of new evidence will be able to refute this narrative of Imperial Germany’s defeat.
Either way, whether we are in the domain of religion or the domain of politics, the main problem with Douthat’s analysis of conspiracy theories is that all such theories are, by definition, tailored to fit a pre-determined conclusion; that’s what makes it a conspiracy theory! As a result, conspiracy theories and alternate realities are impossible test or otherwise falsify. Like Freudian psychoanalysts or Marxist critics of capitalism, conspiracy theorists will always update their priors in favor of their pre-existing beliefs whenever they are presented with new or additional information. Put another way, no amount of evidence will be able to convince a “true believer” that a particular conspiracy theory or alternate reality is contrived. Perhaps, then, we should take a different approach. Instead of asking which conspirary theories we should keep an open mind about, what if we were to ask a different question. Specifically, what if we asked, Why are conspiracy theories so popular in the first place? I will proceed to address this question in my next set of blog posts starting on Monday, March 22.