Science, Non-Science, and Pseudoscience

John Wilkins points to a post by Nick Smyth with the delightful title Science, Pseudoscience and Bollocks on the demarcation problem and what Smyth thinks is a flawed tactic in the war against ID/creationism. Smyth’s basic argument is that labeling ID/creationism as non-science, or as pseudoscience, fails because there is no clear demarcation between science and non-science; we do not have criteria such that we can draw a bright line between science and all other human intellectual enterprises.

However, ID/creationism was once “science” – William Paley used the best science available to him in his Natural Theology. But ID/creationism is a failed scientific endeavor; Philip Kitcher calls it “dead science” and calls its proponents “resurrection men.”

Unfortunately, Smyth’s post implicitly equates “pseudoscience” and “non-science.” That conflation leads him to mistake the core issue and to reverse the nature of the appropriate approach. He thinks the lack of a clear demarcation criterion that tells us what “science” is makes it impossible to classify, say, ID/creationism as non-science or pseudoscience. But the task as I see it is not to define science in such a way that one can point to something and say “That’s science.” It seems to me that it’s quite possible, in principle at least, to argue that ID/creationism, for example, is “pseudoscience” without the necessity of reference to a demarcating definition of science. If one starts with the complementary problem, namely defining pseudoscience independent of any particular definition of science, then we might make more progress. That is, it may be more fruitful to demarcate pseudoscience than to attempt to demarcate science.

This approach has the advantage that it does not require us to worry about questions like whether history is a science or an art (the topic of a faculty development workshop I attended eons ago). We don’t have to wonder if the conjecture of the existence of multiple universes is ‘real’ science or fanciful speculation. We can focus on the pathology and ask what its defining properties are.

So, what are some of the properties of “pseudoscience”? We are not here engaged in defining “science” or finding a demarcation criterion that allows us to point at something and say “That’s science” or “That’s not science”, but rather are attempting to define pseudoscience in a way that allows us to reliably identify instances. I can think of a few identifying marks and scars that don’t depend on having a crisp definition of science or that refer to or explicitly contrast with genuine science. Note that I’m not here attempting to develop something like John Baez’s crackpot index, though it’s undoubtedly related. I’m not after generic cranks, but after a specific question: How can we tell that some view is pseudoscience. I take this to be something akin to medical diagnosis, where we can at least tentatively identify disease states without necessarily having a full definition of what it is to be completely healthy.

Here’s my first cut at some diagnostic properties of pseudoscience, generated right off the top of my head and listed in no particular order. I welcome additions, corrections, or amendments.

  1. Inflationary Credentialism. Habitually inflating credentials or citing proponents’ credentials that are irrelevant to a view as though they lent authority to pronouncements about the view. For example, in the Discovery Institute’s Dissenters from Darwin list, the institutional affiliation listed for a signer is often that of the most prestigious institution with which the signer has ever been affiliated at one time or another, not the current affiliation as of the signing. Thus Stephen Meyers’ affiliation is listed as “Cambridge University,” where he got his Ph.D., and not the Discovery Institute, his current employer, or Palm Beach Atlantic University, the conservative Christian institution that IIRC was his employer in the period when the list was being constructed. Many of the signers have no credentials relevant to a question in biology, but are recruited to the cause anyway. The description of William A. Dembski as “The Fig Isaac Newton of Information Theory” by Robert Koons is another example of inflationary credentialism.

  2. Perseveration with demonstrably false arguments. This is illustrated for creationism by the ability to construct an Index to Creationist Claims which describes the plethora of such false arguments, rebutted over and over in the scientific literature but persisted in by creationists. A contemporary example is William A. Dembski’s perseveration in his misrepresentation of a simple illustration of cumulative selection, a misrepresentation now lasting nearly a decade.

  3. Perverse mistreatment of evidence: Empirical evidence is perpetually re-interpreted, misrepresented or ignored in order to preserve a view. This, of course, is closely related to #2. Pseudoscience is characterized by the persistent mistreatment of evidence, mistreatment that is not merely a matter of differing interpretations that can be resolved by finding new evidence or of drawing alternative implications from incomplete evidence that can also be resolved by gathering further data that are agreed by the disagreeing parties to be relevant. Pseudoscientists admit of no evidence contradicting their position and do not identify potential research or data that would contradict that position.

  4. Claims of unfair exclusion or even persecution. The best recent example, of course, is Expelled: No Intelligence Allowed. IIRC, in the McLean trial creationists made the claim that their papers were unfairly rejected from mainstream scientific journals. When examples of such papers were requested, none appeared. That claim re-surfaced in the Ohio State Board of Education battles in 2002-2003.

  5. Claims of maverick status: Related to #4. Pseudoscientists often explicitly cast themselves as mavericks, claiming that they are bucking the tide of “mainstream” science and creating new scientific “paradigms” (a term that should be cast into the outer darkness; Thomas Kuhn has a lot to answer for there). The Wedge document illustrates this sort of phenomenon with its goal “To replace materialistic explanations with … theistic understanding …”.

  6. [Your contribution here]

Clearly there is an implicit contrast with the properties of ‘real’ science, but notice carefully that all those properties could characterize a position even if ‘real’ science did not exist and even if we had no conception of what ‘real’ science might be. They are independent of any definition of ‘real’ science. Now, we clearly would assign value or utility to a view described by them on account of their difference from ‘real’ science, but I don’t think we require a crisp-set definition of real science in order to diagnose pseudoscience. We don’t need to know what perfect health is in order to diagnose illness.