Why do irrational beliefs mimic science?

Creationism debateI have recently co-published a paper, together with my collaborators Stefaan Blancke and Maarten Boudry, entitled “Why Do Irrational Beliefs Mimic Science? The Cultural Evolution of Pseudoscience,” that I think readers of this blog will find interesting.

In the paper, we develop and extend an epidemiological framework to map the factors that explain the form and the popularity of irrational beliefs in scientific garb. These factors include the exploitation of epistemic vigilance, the misunderstanding of the authority of science, the use of the honorific title of ““science” as an explicit argument for belief, and the phenomenon of epistemic negligence. We conclude by integrating the various factors in an epidemiological framework and thus provide a comprehensive cultural evolutionary account of science mimicry.

Faking and imitating science, exploiting its cultural and epistemic authority, evidently constitutes a profitable strategy. People are somehow duped into believing that pseudoscience constitutes the real thing, or provides a worthy alternative to actual science. But why are people so easily misled?

In previous papers, we have explained the popularity, persistence and typical features of weird beliefs in terms of cognitive and cultural evolutionary processes in which ideas and beliefs adapt to particular susceptibilities of the human mind and to withstand falsification and criticism (see, for instance, here). In the new paper, we investigate why some of these beliefs or belief systems, such as creationism, homeopathy and astrology, assume the cloak of science or pretend to be on equal footing with science.

The first concept we zoom in is that of “epistemic vigilance.” In a landmark paper published in 2010, Sperber and colleagues have explored a human capacity for what they call epistemic vigilance. They argue that such vigilance can be targeted both at the person who is communicating (the source), and the information itself (the content). As to the source, one can rely on two criteria, honesty and competence. The first criterion deals with the intentions of the informant, the second with whether he or she is capable of providing correct information. An informant can be dishonest or incompetent, both, or neither, but only in the last scenario should we trust him. Hence, it is important that we can detect and identify reliable sources, as the opposite of epistemic vigilance is not trust, but blind trust.

If we apply these considerations in the context of expertise, we see that human cognition is clearly equipped with the mechanisms to discriminate, at least in principle, between genuine and false experts. We have the ability to check — within limits — whether an expert is reliable, and whether the information he or she provides is consistent and coherent with our background beliefs.

However, in the case of science and pseudoscience, things tend to be a bit more complicated. Scientific beliefs are often too difficult to comprehend for lay people, which makes content evaluation impossible. This leads people to accept, or reject, scientific concepts mainly on the basis of trust. Now, deferring to experts is not necessarily a bad thing, and indeed, it would be impossible to navigate everyday life without such trust (e.g., think of mundane actions like going to a dentist, or to a car mechanic). Even without the possibility of content evaluation, one can be epistemically vigilant (e.g., checking the credentials of said dentist or car mechanic). One just needs to discriminate between real and false experts, and this latter issue is the target of the next section our paper.

In 2001, Goldman has argued that, even though novices or lay people do not have epistemic access to a particular domain of knowledge, they can rely on five sources of evidence to find out which experts they can trust.

Firstly, one can check the arguments that experts bring to the discussion. Lay people may not be able to grasp the arguments directly, but they can check for what Goldman calls “dialectical superiority.” This does not simply mean that one looks for the best debater — although debating skills can certainly add to the impression that one is an expert — but that one keeps track of the extent to which an alleged expert is capable of debunking or rebutting the opponent’s claims.

Secondly, a novice can check whether and to what extent other experts in that field support a given (alleged) expert’s propositions. It will be more reasonable to follow an expert’s opinion if it is in line with the consensus.

Thirdly, lay people can distinguish between experts on the basis of meta expertise, in the form of credentials such as diplomas and work experience. For example, an expert with a PhD in a relevant field can in general be considered to be more reliable — ceteris paribus — than an amateur.

Fourthly, a novice can check for biases and interests that affect an expert’s judgement. If an expert has a stake in defending a particular position, it will raise the suspicion that he is not interested in providing correct information, which will undermine his credibility. Of course, nobody can be free of biases, which also counts for scientists. Hence the question is not whether there is bias (there always is), but how much, where it comes from, and how one can become aware of and correct it.

Fifthly, a novice can assess an expert’s past track record. The more an expert has been right in the past, the more he has demonstrated that he has indeed access to some expert domain. As such, he will probably be right again in the future.

These five sources of evidence can help novices to tell genuine from false experts, even if they do not understand the substance of the arguments. The problem, of course, is that purveyors of pseudoscience put considerable effort in mimicking each of these tell-tale signs. Take so-called “scientific” creationism, for instance. For decades, creationists of all stripes have been inviting evolutionary biologists to take part in public debates. In the 1930s, creationist Harry Rimmer seized on every occasion to engage scientists in public debates for large crowds and humiliate them. Later, in the 1960s and 1970s, young-earth creationist Duane Gish built himself a reputation out of debating evolutionary scientists (including yours truly, five times).

The creationist strategy has clearly paid off. In a debate setting, a creationist with good rhetorical skills can demonstrate his “dialectical superiority” over less prepared scientists (who tend to underestimate their creationist opponents). Creationists also make a habit of pointing out that many scientists support their cause, and they boast their academic titles and credentials, even if these do not apply to the field they pretend to be experts on (as is very frequently the case). Moreover, creationists often publicly accuse evolutionary scientists of having a hidden political agenda, while they present themselves as unbiased seekers of truth (even though, of course, they are anything but). And, finally, creationists boast of having excellent explanations for biological phenomena. In sum, pseudoscientists give novices a hard time in identifying genuine experts.

The central section of our paper attempts to answer in more detail the question of why pseudoscience is still around, despite a number of cognitive mechanisms available to human beings to spot bullshit, so to speak. I will not attempt to summarize it here (feel free to download the full paper, linked above, if you are interested), but will only mention that there we cover the following topics: error-prone mechanisms and heuristics, exploitation of epistemic vigilance, conceiving of science as an argument, the phenomenon of epistemic negligence, and the existence of what we call “stabilizing factors” (such as confirmation bias and anti-expertise attitudes).

Given all the above, why is cultural mimicry of science so successful? Irrational beliefs become more relevant by dressing up as science, in the sense that they are more likely to grab people’s attention, to be remembered and cognitively processed, and to be disseminated. In the paper we identify and discuss several factors that we think affect the relevance of pseudoscience. Here is a summary of the specific contribution of each factor in this process of cultural mimicry, within the larger framework of cultural epidemiology.

The reason why pseudoscience exists is not that people are stupid or ready to believe anything that they are told. In fact, humans have a suite of mental mechanisms that enable them to filter the information that they receive from others. They can assess the quality of the source by checking competence and honesty, and the quality of the content by checking for consistency and coherence. But in the case of science, things become a bit more complicated. Because people tend to be epistemically vigilant, and critical about sources of information, irrational beliefs need to pretend to originate from a source that people tend to deem trustworthy, i.e., science. However, people do not fully understand or appreciate the epistemic authority of science. They either ascribe authority to science on the basis of its effects or simply because of its reputation. The resulting confusion makes it easier for irrational beliefs to mimic science. In sum, the mechanisms of epistemic vigilance, an environment in which science is regarded as an epistemic authority, and a public who lacks an understanding of that authority, together constitute sufficient conditions for pseudoscience to emerge.

This deception, we show in the paper, largely occurs by persuasion. Purveyors of pseudoscience explicitly use scientific publications, language and typical features such as graphs and formulas, to convince people that they are dealing with genuinely scientific and thus reliable information. Reason-based choice explains why people tend to prefer irrational beliefs that mimic science to non-mimicking ones, because the former provides them with at least one additional argument, i.e., an extra reason for belief.

The relevance, and ensuing pervasiveness, of pseudoscientific beliefs, however, is not only a matter of argumentation, but also of motivation. People are not interested in impartial truth, but in finding and supporting beliefs that make intuitive sense. This lack of concern for truth becomes exacerbated by “epistemic negligence”: people do not invest time and energy in understanding and sustaining highly counterintuitive scientific concepts that are practically useless. The resulting shaky notions of scientific concepts come closer to pseudoscience than science, so that any difference between the two starts to blur. Consequently, psychological factors such as confirmation bias and anti-expertise allow pseudoscientific beliefs to stabilize.

To summarize, pseudoscientific concepts are pervasive: (1) because posing as science works as a tool of persuasion, and (2) because people lack the motivation to correct their intuitive beliefs, but instead seek to confirm them and, simultaneously, distrust genuine scientific expertise.

At the level of individual chains of communication, a complex picture emerges. In chains of transmission, each of the factors discussed above can be present to varying degrees. Several factors, or perhaps even all of them, might be involved at the same time. For instance, purveyors of pseudoscience try to persuade their audience by using science as an argument, while members of the audience readily accept pseudoscientific claims because these tend to corroborate their intuitive, pre-scientific beliefs. Zooming out, however, chains of transmission that might differ depending on the factors involved, when instantiated a sufficiently large number of times, will have the effect of making irrational beliefs converge around particular cultural attractors, namely irrational beliefs that mimic science. As such, the various factors that affect the micro-level processes in which pseudoscientific beliefs become relevant, result in a relatively stable cultural evolutionary process through which irrational beliefs turn into pseudoscience.

107 thoughts on “Why do irrational beliefs mimic science?

  1. couvent2104

    The point is rather that it would display a shocking degree of philistinism and incuriousity.

    No, because every physicist knows that Newton would have used the techniques developed by Lagrange et al. if he would have had them at his disposal. It has absolutely nothing to do with philistinism. One can appreciate and enjoy the majestic splendour of classical mechanics without having read Newton.

    Liked by 1 person

  2. milesmutka

    The notation used today is more Leibniz than Newton. Stephen Wolfram has studied Leibniz originals in his recent blog post, very interesting stuff. Not sure if it the best of all possible notations, and it is hard to edit on a computer, but it looks pretty cool.


  3. Coel

    Hi Massimo,

    That said, Coel, you are excusing Hawking for awfully inaccurate writing on the basis of his health? Besides the fact that his health hasn’t been good for a long time, no, it doesn’t fly.

    His ability to communicate has declined over the decades (and declined a lot since he wrote A brief history of time). At the time of the book in question it was down to one word per minute (it has since increased a bit owing to an improved computer interface). His input would have been targeted at particular areas, leaving his co-author to flesh out the book and do most of the writing. (Books with the word “Hawking” on the cover sell; publishers know that.)


  4. couvent2104

    Most of modern classical mathematics was developed on the continent

    That should have been modern classical mechanics, not mathematics.

    “Modern” isn’t entirely accurate either. There are more modern approaches nowadays, using techniques from differential geometry. Perhaps we could call it Newton’s revenge? His techniques were also mainly geometric … Although in a form of geometry that’s very, very far from differential geometry. These more modern approaches are (in my opinion) closer to Lagrange et al. than Lagrange was to Newton.


  5. Daniel Kaufman

    As I said, I am shocked, though not surprised. Indeed, it confirms many of my impressions about many contemporary scientists, with whom I interact on a daily basis at the university, not to mention online. And it’s also what makes those who are more historically literate stand out so much from the rest. I was talking with a very historically literate mathematician the other day about the recently discovered/translated Archimedes’ text, and it was a thrilling conversation, as well as refreshing, given how rarely it happens. Thank goodness there are at least a few such people out there who refuse to give into the lazy and arrogant notion that the history of ideas is irrelevant to contemporary scholarship.



  6. Daniel Kaufman

    Coel wrote:

    “Physics derives from empirical evidence, not from past writings about physics.”


    No, duh. But given that observation is theory-laden, one still has to know the history of theory. Now, you can get it through secondary sources, but I think that the use of primary sources is preferable for many reasons. That is the reason I do not use textbooks.


  7. couvent2104

    I was talking with a very historically literate mathematician the other day about the recently discovered/translated Archimedes’ text, and it was a thrilling conversation, as well as refreshing, given how rarely it happens.

    I did some Archimedes too, when I wanted to find out how he used the method of exhaustion. My admiration for him and for someone like Eudoxus of Cnidus has no bounds. Eudoxus came as close as possible to the properties of real numbers, given the mathematics of his time (I always feel he knew all the relevant properties without actually defining real numbers – and that’s something so essentially mathematical for me that I doubt I can explain it to a layman).

    But no, the original text has no special importance.
    That’s the way it is.
    It has nothing to do with being literate, an engineering mentality or philistinism. Sometimes the truth, the beauty, the importance and the relevance of something are to a surprising degree independent of the original text.

    A bit provocatively, I would even say that you only can appreciate the profoundness of Eudoxus when you’ve studied the modern definition of the reals with Dedekind cuts.


  8. Coel

    Hi Dan,

    That is the reason I do not use textbooks.

    Not using textbooks would be impossible in physics. The process of distilling and refining knowledge into textbooks (tertiary sources) is itself a major part of the process of physics. The primary sources are too disjointed, bitty, incremental, error-prone and un-refined to be used instead of textbooks. When someone like Weinberg produces something like his three-volume Quantum Theory of Fields that in itself is a major contribution to physics. Everyone else will then learn from a work such as that, rather than from the primary sources that were distilled into it.

    Your approach may well be optimal and appropriate for your discipline, but fields such as physics are genuinely different, and the history of the subject really is a lot less important to understanding the subject.

    Liked by 3 people

  9. couvent2104


    In the article in ScienceNews the author writes

    In this case, it did work, but it remained for Newton and Leibniz to figure out how to make the argument mathematically rigorous.

    Newton and Leibniz did not make the argument rigorous. That was only done in the 19th century by people like Cauchy, Darboux, Riemann, Stieltjes etc.


  10. Daniel Kaufman

    couvent: I’m afraid we disagree entirely on this point. And saying “that’s the way it is” is foot-stamping, not an argument.

    And not every scientist or mathematician agrees either. One of my great uncles was a brilliant Hungarian physicist/mathematician who did a lot of work on Einstein’s notebooks and was a great historian of science. He told me that it was as important to see how the great scientific and mathematical minds did their work as it was to know the conclusions they came to and the evidence on which they were based, and this is only possible in confronting the real texts. I can teach Aristotle’s theories or Descartes’ and even act as a secondary source and describe the reasoning they used to reach their conclusions. But I cannot speak in their voices and do things the way they did them. The value of reading primary sources is not just the thrill and excitement of seeing a great mind at work — it is a means by which ones own mind may be moved to look at and do things in different ways.

    The view that the history of a subject and primary sources are irrelevant to contemporary practice is pernicious and reflects, in my view, a very self-satisfied brand of progressivism. That entire disciplines have adopted this posture is no argument for it and to say “well, it works” is unpersuasive, without knowing the relevant counterfactuals. For my money, I don’t think mathematics or science would be worse off for having more of the people who do it be like my friend, wrestling with Archimedes, or my great uncle poring through Einstein’s notebooks.

    Liked by 1 person

  11. brodix

    With all everyone has read, has anyone here given much thought to what the “fabric of spacetime” entails?

    Ptolomy was no slouch at math either, but when there is a problem with the basic assumptions, the explanations are going to be more complex than necessary.


  12. brodix


    Before you chastise me for bringing up pet theories again, this is a discussion about irrational science and I am one of, possibly the only remaining crank, so I would think it an excellent opportunity to pick apart my arguments and before you say Coel already did so, Coel’s and my point of contention was whether it is logical to say space expands, while still assuming an otherwise stable and distinct speed of light, from this expanding space.
    Just think of me as a creationist, arguing the universe was created a mere 13+ billion years ago. Oh, wait….


  13. SocraticGadfly

    I think Couvent is right on the issue of “ur-texts” in the natural sciences vs. not only philosophy, but other humanities. Facts are facts. But, when writing philosophy, or history, or political science, etc. analysis and interpretation ride high. One can insist on trying to get a proper angle on Hume, for example, on either empirical philosophy, or on British constitutionalism in his history-wearing hat.

    Or to take an example of spanning the two worlds (and not C.P. Snow):

    Linus Pauling.

    One need not read his exact writing on chemistry to accept the chemistry work that won him that Nobel.

    But one should indeed read what Pauling wrote, not what somebody else wrote, about nuclear weapons issues, to understand what won him the Nobel Peace Prize.

    Liked by 2 people

  14. Daniel Kaufman

    socratic: That’s fine, but it really doesn’t address the substantive case that I made on its behalf, as articulated by my Great Uncle. (With regard to whom, I should add, in case anyone might think that he was just some musty historian, beyond his theoretical and historical work, he also worked for Grumman for decades and thus, was also a hands-on, practicing scientist and engineer.)

    Liked by 1 person

Comments are closed.