When discussing fringe scientific topics, people often use the term “pseudoscience” to describe something that doesn’t quite meets the high standards of scientific proof. Basically, a claim can be said to be pseudoscientific when it presents itself as being the product of empirical research and the scientific method, but demonstrably fails to live up to these claims.
In identifying pseudoscience, it helps to be aware of what makes something scientific in the first place. In broad terms, a scientific theory has the following characteristics:
Empiricism: Knowledge is acquired through systematic, carefully noted observation.
An important aspect of empirical claims, as put forth by Karl Popper, is the quality of falsifiability – that is, they can be supported or refuted by real-world data; things outside of the realm of observation (such as religion/spirituality) are inherently unfalsifiable and therefore do not fall under the prevue of science.
Objectivity: Observations and interpretations of the data are unbiased.
Public: Available to all; papers are usually subject to careful scrutinization by colleges (“peer review”) prior to publication in scientific journals, with studies with major flaws in methods or conclusions being rejected.
Replicable: the findings must be able to be confirmed by repeating the study (sometimes with a whole different team/equipment) and obtaining the same results – this is to rule out flukes.
Now that we have a good understanding of what real science is, let’s look at some of the characteristics of pseudoscience and see why they don’t hold up:
Methods are not sufficiently rigorous (lacks precision, poor methodology, etc.)
Pseudoscience experiments often lack important aspects of true science like effective controls groups, boundary conditions, operational definitions, and parsimony (i.e. seeking the explanation that requires the fewest additional assumptions). Claims are frequently vague, exaggerated, or untestable, and often the burden of proof is shifted from the claimant (the one usually responsible for providing evidence of a new theory) to the skeptic.
Evidence is often anecdotal and relies on confirmation over refutation
Anecdotal evidence (such as selected cases and personal experience testimonials) is a problem in scientific research because it is often subject to selection bias (that is, picking and choosing evidence which supports one’s claims while rejecting conflicting data). Experiments which rely on self-selection and self-reporting are often called “quasi-experimental” because they lack the element of random assignment – without participants being randomly assigned to experimental or control groups, there is no way of knowing if some other factor than the one being measured/manipulated was responsible for the results.
Lack of objectivity and openness to review/replication
Pseudoscience advocates tend to avoid subjecting their research to peer review, often claiming either that peer review is bias towards established scientific paradigms, or that their assertions cannot be adequately evaluated with existing scientific methods; substantive debate on the issue by knowledgeable participants of all viewpoints is routinely discouraged. There are frequently little-to-no efforts at making a results replicable, since the experimenters likely simply conducted the experiment over and over until getting the results they wanted. Sometimes there is also a non-so-hidden bias on the part of the experimenters or those funding the research which casts further suspicion onto their results (for example, a fossil-fuel company producing a study which “debunks” climate change research).
Claims are often delivered in scientific-sounding terminology (“technobabble”), usually by a respected and trusted authority figure (i.e. a “doctor” in a lab coat).
Complicated, scientific-sounding terms are often used to add weight to the claims by exploiting the average lay-person’s lack of understanding of scientific principals, usually backed up by having the claim delivered by an “expert” – frequently a person in a lab coat. Knowledgeable critics can usually spot this sort of thing, since the “technobabble” will usually be used in strange and idiosyncratic ways that betrays a lack of understanding and familiarity with the term’s mainstream use (e.g. using the full chemical name of a common substance, or attaching the word “quantum” to make things sound more “metaphysical”).