The science-pseudoscience demarcation problem

philosophy bitesI finally made it! I was invited by Nigel Warburton as a guest on his Philosophy Bites, likely the most downloaded podcast in philosophy. The man has so far accumulated 352 interviews with some of the most interesting contemporary philosophers, including recently Jesse Prinz (on whether everything is socially constructed), David Owens (on duty), Kimberley Brownlee (on social deprivation), Shelly Kagan (on specieism), and many, many others.

And, now, yours truly. Our conversation focused on one of my main interests, the so-called “demarcation problem,” which is what Karl Popper called the issue of how to distinguish between science and non-science, and in particular between science and pseudo-science. For anyone wanting a more in-depth treatment, my friend Maarten Boudry and I have put out a collection of essays on this topic, published by Chicago Press.

Early on in the history of the demarcation problem people (including Popper) thought that it was going to be possible to come up with a small set of necessary and jointly sufficient criteria to define science, pseudoscience, and the like. Hence Popper’s famous idea of “falsification,” which essentially states that if a hypothesis or theory is (potentially) falsifiable — i.e., it can empirically be shown to be false, if it is indeed false — then it is scientific. It ain’t falsifiable, it ain’t science.

Subsequently Popper himself came to understand that so-called “naive” falsificationism doesn’t actually work, both because some obviously pseudoscientific ideas are indeed falsifiable (e.g., homeopathy) and because some good scientific ideas aren’t (immediately) falsifiable (e.g., superstring theory).

Larry Laudan, back in 1983, therefore proposed that demarcation is impossible, and that philosophers really ought to do something else with their time. But plenty disagreed, which in time led to the above mentioned collection of essays, the first several chapters of which are responses to Laudan’s skepticism.

What philosophers of science interested in demarcation seem to agree on, however, is that there are no necessary & sufficient criteria that can do the job, for the very good reason that concepts like “science,” “pseudoscience” and the like are inherently fuzzy, without sharp boundaries, and require therefore less rigid, clearcut treatment.

So, if you have to give it your best shot, what do you think distinguishes science from pseudoscience? And are psychoanalytic theories, as well as Marxist theories of history, pseudo-scientific, the way Popper thought? What about string theory: physics or metaphysics?


49 thoughts on “The science-pseudoscience demarcation problem

  1. Tom,

    I guess like a lot of philosophy, it is trying to reach beyond the obvious, everyday details of life, to find the deeper patterns, without making too much fools of ourselves.


  2. Hi Thomas,

    Coel,”both were best attempts to understand the universe.” Not really. This is an overstatement on your part. Astrology was simply the Oracle of Dephi in Sunday clothes.

    If we go back to the ancient Babylonian period (or ancient China or Egypt or similar), the origin of astrology was trying to understand the motions of the stars and planets in order to track the seasons and predict weather, and to make decisions such as the best time to plant crops. That was indistinguishable from “astronomy” at that time, and (as I see it) can fairly be called “scientific”, or at least proto-scientific, being a best attempt to understand and predict the world.

    If one was consulting the astrologers/astronomers about when to plant crops and what the weather might be, it would then be natural to ask them about the right time to consecrate a temple or declare war or make other big decisions. Afterall, all such things were wrapped up with “nature gods” in the understanding of those times.

    We now know that such notions were misguided; but then a lot of things in the history of science have, with the benefit of hindsight, turned out to be misguided. Thus I’m sympathetic with those ancient astrologers, they were doing their best with what they knew, just as scientists do today.

    It was only gradually that better understandings were developed, and only by comparison with those better understandings is it then fair to say that mistaken notions are “pseudo-science”.

    Liked by 1 person

  3. brodix,

    I know you are well intentioned, but Thomas’ frustration reflects my own, which goes back to the old Scentia Salon site. Please stick to the topic at hand, without going further afield (and usually in a very predictable direction).

    Also, everyone, please watch the tone and keep the conversations cordial and constructive. I really don’t want to go to moderation, but I will not hesitate if I feel the quality of the site begins to suffer. Thanks!


    “In the very early days astrology was as much a science as astronomy was”

    I think you are assuming a lot about the history of astrology and astronomy, with which neither of us is really familiar enough. But I’d go the other way and say that at the time the very categories of “science” and “pseudoscience” didn’t really apply. As you know, I have a far more restrictive concept of science than you do. I’d say what was going on was some type of proto-science (early astronomy) and some kind of confused mystical thinking (astrology), more or less entangled with each other.


    “I would also argue that without the fringes, i.e. the pseudosciences, there could be no core, i.e. the sciences. Much as without chaos, i.e. unframed activity, there would be nothing to frame and order”

    Nice analogy, but I don’t buy it. I don’t see what science has to gain, either historically or now, from the existence of pseudoscience. The latter is distracting at best, misleading and dangerous at worst. I would gladly do without it, if I could.


  4. Coel, you’ve made a good point here about “when to plant crops and what the weather might be,” but a debatable one when you conclude that it would be “natural” etc. I don’t buy into the narrative that the second is a natural outgrowth of the other or that *all* the “ancients” made no distinction between making observations regarding the seasons and the lengths of days and, say, sacrificing a virgin to prevent a volcano from exploding.


  5. Massimo:

    “I’d say what was going on was some type of proto-science (early astronomy) and some kind of confused mystical thinking (astrology), more or less entangled with each other.”

    This is well-put and does a much better job of describing my thinking than I have.

    Liked by 1 person

  6. Massimo,

    How do you have trial without error? Yes, you could argue the pseudosciences are carrying it to unnecessarily extreme and ludicrous lengths, but to prevent that you need an authority/arbitrator, which then has quite a history of being abused. The most effective solution is to simply logically present the counter argument and let them have enough rope to hang themselves. Now I can understand, per your comment to Coel, that you necessarily might take it as part of your role as a philosopher of science, to police the boundaries, but that is still one role in a much larger context. Think of me as someone who might be skirting the edges of the law. But just out of curiosity, of course.


    Much as with climate denial, for example, one would suspect there are quite a few social, cultural and possibly economic impulses behind the ancients versions of pseudo sciences, just as today. We all want the take the sword to the knot, but that is still just a short cut, if your desire is to actually try and sense what their actual thought processes on the subject are. I think the term is normalcy bias.


  7. Massimo,

    Also, as philosopher of science, per my attempt to satire, how do you falsify a theory, if every time observations don’t match predictions, another enormous force of nature is hypothesized and everyone runs off looking for it?

    Is this like with the banking system, that it is just too big to fail? Where are the police? Do we just have to give them enough rope to hang themselves?


  8. Hi Massimo (and Thomas),

    As you know, I have a far more restrictive concept of science than you do.

    Yes, true, and I think that explains why we see the demarcation issue a little differently.

    Your concept of “science” would (I presume) be along the lines of the modern “scientific method” (where that, of course, is a shorthand for a bundle of methods that isn’t easy to summarise in one sentence).

    By that account, ancient astrology would certainly not be a “science”, and early “science” would be at best a “proto-science” since it was heavily flawed in method as judged by modern standards.

    On the other hand, I would regard “science” as a more general process of trying to model and predict the world, and of continually improving ideas to that end (iterating round a loop of comparing to reality, and improving the ideas).

    By that account, today’s “scientific method” is itself a product of the process of science. Thus how to do science well is one of the things that we have discovered over time by the process of science.

    From that view it makes more sense to regard early astrology/astronomy as “science”, being a best-attempt to model and predict the world, but one that preceded a better understanding of which methods work and which don’t.

    Liked by 1 person

  9. Hi Thomas,

    I don’t buy into the narrative that the second is a natural outgrowth of the other or that *all* the “ancients” made no distinction between making observations regarding the seasons and the lengths of days and, say, sacrificing a virgin to prevent a volcano from exploding.

    I’ll happily grant that there was a spread of opinion back then. But, let’s take the concept of “nature gods” and the idea that, for example, one can ensure a good harvest by sacrificing a goat to the right god.

    Such an idea is nowadays “unscientific”, but is that because science must declare naturalism as a metaphysical axiom, and thus treat the possibility of nature gods as off-limits?

    I’d say no (since, to me, the idea of science being founded on metaphysical axioms that can’t be further justified is anathema!). Rather, I argue that naturalism and the rejection of nature-gods is itself a *product* of science, a product of doing science and working out what works. Science has thus ditched nature gods and adopted naturalism because that best models the world.

    But, back in ancient Babylonia, Egypt, China et al, that hadn’t been worked out, and the issue was still open for discussion among the philosophers and other intellectuals of the time. Thus, to me, the emergence of naturalistic science and the modern scientific method is part and parcel of the process of science itself, and thus the early stages of that process are best viewed as also being “science”.

    Liked by 1 person

  10. Popper was never a naïve falsificationist and he was suggesting the most effective way to use data to eliminate mistaken ideas. The idea was to encourage imaginative and critical thinking and to avoid getting bogged down by confirmation bias. He also suggested tolerance for “deviants” from orthodox thinking, just in case they might be onto something.


  11. brodix,

    “How do you have trial without error?”

    Do you imagine that scientists don’t advance by trial and error of their hypotheses? Do you think that they need astrology to hang around in order to make their enterprise better? Why?

    “how do you falsify a theory, if every time observations don’t match predictions, another enormous force of nature is hypothesized and everyone runs off looking for it?”

    I don’t think that’s what physicists do…


    “Your concept of “science” would (I presume) be along the lines of the modern “scientific method'”

    It’s worse than that. As most contemporary philosophers I don’t think there is a scientific method, just a bunch of different approaches and heuristics, some of which work better than other and are retained for further use.

    “how to do science well is one of the things that we have discovered over time by the process of science”

    I would agree, and yet still maintain a more narrow concept of the term than you do.


    “Popper was never a naïve falsificationist”

    Yes, I know. But it’s also clear (look at the Stanford Encyclopedia of Philosophy article on him) that he changed his mind and softened quite a bit over time. Initially, he was reacting to verificationism, the idea that science proceeds by verification (i.e., confirmation) of theories. His criterion of falsification was starkly different (because a lot of confirmations don’t show that a theory is true, while a single disconfirmation can show it false). But in the end he had to admit that one needs both, which took a lot of the bite out of his initial suggestion.

    “He also suggested tolerance for “deviants” from orthodox thinking, just in case they might be onto something.”

    True, though not as much as Feyerabend did! (Too much, in the latter case.)

    Liked by 1 person

  12. Coel, I appreciate your remarks on how the “ancients” viewed themselves. Unfortunately, we run into problems when we start tossing labels about. For example, astrology. This term came into parlance about the 14th century. Chances are, had you surfaced about the time of, say, the pre-socratics and had said something along the lines of “Hi, guys, how’s your astrology going?” you might have received a quizzical look by way of response, depending on whether you were addressing someone who took a practical interest in observing the movements of “heavenly” bodies or someone who claimed to use this knowledge as a pretext for justifying divination. This is, is I think, the point Massimo is making in one of his later remarks here. At any rate, none of us has really addressed the questions posed by Massimo in the last paragraph of his post. I think we can all agree on this much. The demarcation problem is by and large a moot point until the late middle ages when our working knowledge reached a level a sophistication that afforded a basis for questioning prevailing viewpoints. But this is largely a Occidental orientation. I lack the necessary background to even begin to compare it to an Oriental orientation.


  13. Massimo,

    “Do you think that they need astrology to hang around in order to make their enterprise better? Why?”

    Necessarily astrology was a premise that diverged from practical science, which tended toward technology and experiment, more than theory, as the split developed. Remember that back in those days, error often meant you were dead, not that your paper was rejected. The reason it continued into religion would be for more sociological reasons, such as power politics and generally keeping the masses entertained and directed. Much like science fiction today tends to reflect current sociology as much or more than technology.

    “I don’t think that’s what physicists do…”

    Do you recall when Perlmutter et al essentially announced the discovery of Dark Energy? It was that they discovered a curvature in the rate redshift decreased. In other words, the theoretical assumption was that after the Big Bang, the rate of expansion slowed at a steady rate, but instead it initially decreased fairly rapidly and then flattened out. So there was an issue of explaining what powered this closer gradual expansion, if the original energy of the initial effect dissipated quickly. So the argument became that there must be some additional energy inherent to space, causing it expand and when calculated, the amount necessary added up to three times all prior presumed mass and energy of the universe, including the dark matter used to explain why centrifugal forces don’t tear galaxies apart.

    So basically it is a patch between theory and observation.

    Now if redshift were an optical effect, such as gravitational lensing, then if this effect compounds on itself, it would explain why the redshift goes parabolic, with distance. Such that eventually visible light is shifted entirely off the spectrum, creating the assumption that these distant galaxies must be receding at the speed of light.

    Yet that still assumes that light is an otherwise objective measure of the expanding space. So, as I keep asking, what is the basis of the speed of light, if space itself is expanding?

    Now if redshift is optical, this would be a moot point.


  14. Keep in mind that redshift increases equally in all directions, effectively making us appear as the center of the universe. As well as that volume increases with radius, but presumably the universe is much more compact with age, yet the density of galaxies appears similar, even though they exist in a void where the speed of light is implied to be stable.

    Ockham’s razor would strongly suggest we do consider how redshift might be an optical effect. Here is a paper pointing out much of the reason only doppler effect is believed to cause redshift is based on single frequency tests, while cosmic light is mostly multifrequency;


  15. Here’s a data point I’d like to see plotted: On which side of the demarcation does mainstream economics fall? Science or pseudoscience?

    Liked by 1 person

  16. I think this article comes close to the answer:

    Good point.

    One way of thinking about science is that it is beliefs about the operation of Laws of Nature which are anchored in observation. These are two necessary tests. The so-called ‘behavioural sciences’ cannot be science because the presence of free will decouples them from the laws of nature. Observation can uncover generalisable patterns in human behaviour, shaped by context, but this can never reach the level of science, i.e. uncover genuine laws of nature, because free will obscures the link between behaviour and the laws of nature.

    As the article says:

    human knowledge about humans is fundamentally different from human knowledge about the natural world

    Of course scientism has reared its ugly head.

    Perhaps the most pernicious effect of the status of economics in public life has been the hegemony of technocratic thinking. Political questions about how to run society have come to be framed as technical issues, fatally diminishing politics as the arena where society debates means and ends.

    The pernicious effect of scientism is that it promotes the myth that study of human behaviour is a science, reducing us to statistics and devaluing the richness of being human.


  17. Sun, labnut,

    It might also be that the assumptions built into their models are drastically flawed. For example, money is treated as both medium of exchange and store of value, yet if you consider it biologically, blood is the medium of exchange within the organism, while fat is the basic energy store. Necessarily, it is wise not to have much fat stored in the body’s circulation system, or the result is clogged arteries, poor circulation and high blood pressure to compensate. Which is fairly analogous to trickle down economics and quantitive easing.

    Energy is more effectively stored in the muscles and what remains liquid within the circulation system and our current economic model resolves this by having government finance borrow a lot of the excess back out of the system and spend it in ways to further extend private sector investment, even if it has long term destructive consequences, such as having to pay interest back into the system and therefore only creating a larger problem for the future.

    Which is simply to say that economics might just still be in a very early stage of intellectual development and there are serious cultural forces mitigating against looking too closely into the dynamic.

    For instance, one needs to study the economics and politics of the 19th and early 20th century to get a sense of how this relationship between finance and the economy was far better understood, such as with Jackson’s campaigns against a national bank, as an institution controlled by a private banking consortium, to see the foundations of our current system, but the politicization of the topic made it impossible to objectively study and develop an objective economic model over the last century. Consequently we have “capitalism” and “socialism” as polarized opponents.

    Science and politics do not make for friendly bedfellows and given human nature, politics is inherently dominant.


Comments are closed.