The idea is that it is relatively easy to find historical (and even recent) examples of notions or fields that began within the scope of acceptable scientific practice, but then moved (or, rather, precipitated) into the realm of pseudoscience. The classic case, of course, is alchemy. Contra popular perception, alchemists did produce a significant amount of empirical results about the behavior of different combinations of chemicals, even though the basic theory of elements underlying the whole enterprise was in fact hopelessly flawed. Also, let's not forget that first rate scientists - foremost among them Newton - spent a lot of time carrying out alchemical research, and that they thought of it in the same way in which they were thinking of what later turned out to be good science.

Another example, this one much more recent, is provided by the cold fusion story. The initial 1989 report by Stanley Pons and Martin Fleischmann was received with natural caution by the scientific community, given the potentially revolutionary import (both theoretical and practical) of the alleged discovery. But it was treated as science, done by credentialed scientists working within established institutions. The notion was quickly abandoned when various groups couldn't replicate Pons and Fleischmann's results, and moreover given that theoreticians just couldn't make sense of how cold fusion was possible to begin with. The story would have ended there, and represented a good example of the self-correcting mechanism of science, if a small but persistent group of aficionados hadn't pursued the matter by organizing alternative meetings, publishing alleged results, and eventually even beginning to claim that there was a conspiracy by the scientific establishment to suppress the whole affair. In other words, cold fusion had - surprisingly rapidly - moved not only into the territory of discarded science, but of downright pseudoscience.

Examples of this type can easily be multiplied by even a cursory survey of the history of science. Eugenics and phrenology immediately come to mind, as well as - only slightly more controversially - psychoanalysis. At this point I would also firmly throw parapsychology into the heap (research in parapsychology has been conducted by credentialed scientists, especially during the early part of the 20th century, and for a while it looked like it might have gained enough traction to move to mainstream).

But, asked Maarten, do we have any convincing cases of the reverse happening? That is, are there historical cases of a discipline or notion that began as clearly pseudoscientific but then managed to clean up its act and emerge as a respectable science? And if not, why?

Before going any further, we may need to get a bit more clear on what we mean by pseudoscience. Of course Maarten, I and our contributors devoted an entire book to explore that and related questions, so the matter is intricate. Nonetheless, three characteristics of pseudoscience clearly emerged from our discussions:

1. Pseudoscience is not a fixed notion. A field can slide into (and maybe out of?) pseudoscientific status depending on the temporal evolution of its epistemic status (and, to a certain extent, of the sociology of the situation).

2. Pseudoscientific claims are grossly deficient in terms of epistemic warrant. This, however, is not sufficient to identify pseudoscience per se, as some claims made within established science can also, at one time or another, be epistemically grossly deficient.

3. What most characterizes a pseudoscience is the concerted efforts of its practitioners to mimic the trappings of science: They want to be seen as doing science, so they organize conferences, publish specialized journals, and talk about data and statistical analyses. All of it, of course, while lacking the necessary epistemic warrant to actually be a science.

Given this three-point concept of pseudoscience, then, is Maarten right that pseudoscientific status, once reached, is a "black hole," a sink from which no notion or field ever emerges again?

The obvious counter example would seem to be herbal medicine which, to a limited extent, is becoming acceptable as a mainstream practice. Indeed, in some cases our modern technology has uncontroversially and successfully purified and greatly improved the efficacy of natural remedies. Just think, of course, of aspirin, whose active ingredient is derived from the bark and leaves of willow trees, ​the effectiveness of which was well known already to Hippocrates 23 centuries ago.

Maybe, just maybe, we are in the process of witnessing a similar emergence of acupuncture from pseudoscience to medical acceptability. I say maybe because it is not at all clear, as yet, whether acupuncture has additional effects above and beyond the placebo. But if it does, then it should certainly be used in some clinical practice, mostly as a complementary approach to pain management (it doesn't seem to have measurable effects on much else).

But these two counter examples struck both Maarten and I as rather unconvincing. They are better interpreted as specific practices, arrived at by trial and error, which happen to work well enough to be useful in modern settings. The theory, such as it is, behind them is not just wrong, but could have never aspired to be scientific to begin with.

Acupuncture, for instance, is based on the metaphysical notion of Qi energy, flowing through 12 "regular" and 8 "extraordinary" so-called "meridians." Indeed, there are allegedly five types of Qi energy, corresponding to five cardinal functions of the human body: actuation, warming, defense, containment and transformation. Needless to say, all of this is entirely made up, and makes absolutely no contact with either empirical science or established theoretical notions in, say, physics or biology.

The situation is even more hopeless in the case of "herbalism," which originates from a hodgepodge of approaches, including magic, shamanism, and Chinese "medicine" type of supernaturalism. Indeed, one of Hippocrates' great contributions was precisely to reject mysticism and supernaturalism as bases for medicine, which is why he is often referred to as the father of "Western" medicine (i.e., medicine).

Based just on the examples discussed above - concerning once acceptable scientific notions that slipped into pseudoscience and pseudoscientific notions that never emerged into science - it would seem that there is a potential explanation for Maarten's black hole. Cold fusion, phrenology, and to some (perhaps more debatable) extent alchemy were not just empirically based (so is acupuncture, after all!), but built on a theoretical foundation that invoked natural laws and explicitly attempted to link up with established science. Those instances of pseudoscience whose practice, but not theory, may have made it into the mainstream, instead, invoked supernatural or mystical notions, and most definitely did not make any attempt to connect with the rest of the scientific web of knowledge.

Please note that I am certainly not saying that all pseudoscience is based on supernaturalism. Parapsychology and ufology, in most of their incarnations at least, certainly aren't. What I am saying is that either a notion begins within the realm of possibly acceptable science - from which it then evolves either toward full fledged science or slides into pseudoscience - or it starts out as pseudoscience and remains there. The few apparent exceptions to the latter scenario appear to be cases of practices based on mystical or similar notions. In those cases aspects of the practice may become incorporated into (and explained by) modern science, but the "theoretical" (really, metaphysical) baggage is irrevocably shed.

Can anyone think of examples that counter the idea of the pseudoscience black hole? Or of alternative explanations for its existence?