Paper Review – “Unbelievable” (Tony Sobrado)

Tony Sobrado’s article “Unbelievable” in the June 2012 issue of “The Skeptic” treats conspiracy theories as a species of what Sobrado calls “pseudo-theories.” Sobrado runs an analogy between the demarcation of the (natural) and pseudo-sciences and the demarcation between theory (proper) and pseudo-theories. His analogy is then used to argue for the claim that conspiracy theories fail the test of being anything other than pseudo-theories, which in turn justifies our suspicion of such theories.

Obviously, given that my thesis is called “In defence of conspiracy theories,” Sobrado and I do not see eye-to-eye in this matter. Indeed, I find his argument perplexing, in part because I think he applies the wrong tool to the analysis of conspiracy theories (falsificationism) and he buys into a pejorative definition of conspiracy theory (by talking almost solely about what he calls “meta-conspiracy theories:” the ones which cite the existence of the New World Order, et cetera) ((I also think he buys into Popper’s notion of the conspiracy theory of society when he writes:

This then leads to the paradox in conspiracy theory with regards to observed social phenomena and epistemology. This is that the observed phenomenon is allocated to the realm of conspiratorial explanation. The explanation encapsulates the activity of cabalist agents operating illustrious tricks of misdirection yet simultaneously these grand tricks are not only detected by conspiracy theorists but also are fully deciphered. We are thus left with a logical dissonance because the conspirators are both fully competent and almost omnipotent like in that they can control all facets of social activity but are completely inept in the maintenance of secrecy as they leave an amassed trail of clues behind for conspirators to find. (p. 24)

Sobrado, like Popper before him, seems to think there is some kind of weird paradox between the claims of seemingly all-powerful conspirators and their plots being known about, but as Charles Pigden wrote in “Popper Revisited, or What is wrong with conspiracy theories anyway?” this thesis appears to be a bit of a strawman argument against belief in conspiracy theories.)) one that makes his argument trivially valid but skirts around the interesting epistemic issues.

Sobrado’s article is, ostenisbly, an argument about how we can apply some theories from the Philosophy of Science to the analysis of conspiracy theories. His argument is based upon an understanding of how philosophers of the Natural Sciences talk about theory and the various desiderata that are currently in vogue when we talk about such theories being adequate. However, I think Sobrado is mistaken to talk about theory sui generis in this way, because he is talking about theories in the Natural Sciences and then importing that discussion to theories in the Social Sciences.

Firstly, take one of the Holy Grails of the scientific method: prediction. These abound in the natural sciences, ranging from how, when, and why your PC will turn on to planes flying and equations of time and space. The social sciences tread a more precarious and unreliable ground. However, through the collection of data based on conceptualised variables along with statistical models of causation, predictions can be levied. Anyone who has an investment portfolio can see the benefits of employing time series and regression analysis in economics although the latest financial crises illustrates that these predictions are far from completely accurate. (p. 23)

Herein lies the first problem: what is he defining as the “Social Sciences” here? Is Anthropology, under his understanding, a social science? Is History? Because it seems he has chosen an example here (Economics) which looks, at first glance, to be a good fit/analogous with the theories of the Natural Sciences but may not be representative of the Social Sciences as a whole.

For example, if History is a Social Science, then prediction is questionable as a shared feature between the Natural and Social Sciences, given that historical explanations are often contextualised to one time and place and if you force them into a form where they can be made to appear predicative, such predictions are either hopelessly vague (“Monarchies get overthrown when the middle-classes become powerful”) or false (“Monarchies get overthrown when the middle-classes become powerful”). The same kind of question can be raised about Sociology (in which many adequate sociological theories explain the phenomenology of a social group at some particular time), Anthropology, Geography and the like. Do we expect all the adequate theories in these Social Sciences to be predicative? Some very well might be, but many aren’t, but that doesn’t seem to be a problem for many of those theories, given that they still adequately explain phenomena in the domain of that particular social science.

I think, then, that Sobrado has, presumably inadvertantly, chosen an example (Economics) which closely adheres to the standards of the Natural Sciences (although my suspicion is that most Economics is more like Sociology and the like, so the fit is partial at best). He has then inferred if Economics can be predicative, then not only should the other Social Sciences should be as well, but as conspiracy theories fail to be predicative, they are pseudo-theoretic.

I am not at all convinced. As I argue in my thesis, conspiracy theories are a kind of explanation (specifically, historical explanations) and whilst scientific explanations typically act as both predictions and retrodictions, social scientific ones (especially historical theories) are not necessarily predicative (although it’s nice when they can be). Prediction might well be a “holy grail” for theories in the Natural Sciences, but it’s not clear it is a disiderata for theories in the Social Sciences.

Sobrado also brings in talk of falsificationism, stating that as conspiracy theories cannot be falisified, they are, consequently, pseudo-theories (p. 23). For one thing, the process of falsificiationism differentiates between the theories which are scientific and non-scientific theories. If a purported “scientific theory” cannot be falsified, then it can rightly be considered “pseudo-scientific.” That being said, falsificationism isn’t necessarily a useful demarcation between adequate theories in the Social Sciences. Certainly, according to some philosophers of Science, the theories of the Social Sciences are not examples of kinds of theories we find in the Natural Sciences. Thus, while under Popper’s view such non-science theories might well be unfalsifiable, that doesn’t mean they are pseudo-scientific (say, in the way that Homeopathy is). If a theory is unfalsifiable it is, according to Popper, not a scientific theory (and thus we should not accept it in the packet of scientific theories) but just because a theory is unfalsifiable, that doesn’t make it a pseudo-theory because whilst being falisifiable is important to theories in the Natural Sciences, it isn’t clear that it’s a desiderata of theories in the Social Sciences. Lots of theories in the social sciences are unfalsifiable but still adequate ((I’m leaving to one side all the attendant criticisms of the thesis of Falsificationism which indicates that it might not be a good demarcating criterion between the Natural Sciences and the pseudosciences and that it might not even be a feature of theories in the Natural Sciences anyway.)).

Also, I have to ask, has Sobrado not read Brian L. Keeley’s seminal (at least with respect to the philosophy of conspiracy theories) “Of Conspiracy Theories?” Keeley argues that falsification is a fine thesis but just not applicable to things like conspiracy theories. To quote:

By hypothesis, the conspiracy theorist is struggling to explain phenomena that other, presumably powerful, agents are actively seeking to keep secret. Unlike the case of science, where nature is construed as a passive and uninterested party with respect to human-knowledge gathering activities, the conspiracy theorist is working in a domain where the investigated actively seeks to hamper the investigation. Imagine if neutrinos were not simply hard to detect, but actively sought to avoid detection! This is exactly the case with which conspiracy theorists contend we are confronted in the cases they seek to explain. This is why countervailing evidence and lack of evidence can and ought to be construed as supporting their theories. (p. 120)


It is not ad hoc to suppose that false and misleading data will be thrown your way when one supposes that there is somebody out there actively throwing that data at you. (p. 121)


Falsifiability is a perfectly fine criterion in the case of natural science when the target of investigation is neutral with respect to our queries, but it seems much less appropriate in the case of the phenomena covered by conspiracy theories. (p. 121)

If there are conspiracies in existence, we should expect evidence for them to be hard to come by and the very evidence which might falsify such a conspiracy theory might well be disinformation that has been put out there by the conspirators in order to hide or obfuscate evidence for the existence of said conspiracy.

In short, if it turns out that the kind of things which make theories in the Social Sciences adequate are different from those of the Natural Sciences, then Sobrado’s analogy between the pseudo-sciences and what Sobrado calls “pseudo-theories” might very well fall down. I think this is yet another article (common to the field) where the author assumes conspiracy theories must be, sui generis, unwarranted and thus constructs a case for said lack of warrant. In this instance, Sobrado applies an understanding of the Philosophy of the (Natural) Sciences to conspiracy theories without, it seems, realising that the criteria by which we analyse the Natural Sciences is not necessarily the same as the way we appraise theories in the Social Sciences (of which conspiracy theories are a subset).