Readers of this blog know very well by now that, despite (or is it because of?) being both a scientist and a philosopher, I have often defended the idea that science and philosophy are distinct disciplines, and I am critical in particular of those who I think display a scientistic (i.e., intellectually imperialistic) attitude in wanting to expand the scope of science to pretty much everything that is worth knowing, usually at the expense of humanistic disciplines, philosophy in particular.
But, one could reasonably ask, why bother? Why try to explain to the likes of Sam Harris and Michael Shermer that ethical quandaries cannot be resolved by science, be that neurobiology or evolutionary biology? Why argue with Jerry Coyne and Richard Dawkins that when they think they have rejected “the god hypothesis” or the idea of free will on scientific grounds they have, in fact, smuggled in quite a bit of philosophy to make their case? Who cares which discipline is doing what, isn’t the outcome of our inquiry what matters?
And yet, I wager that all four of the above mentioned people (and others regularly criticized here for similar positions, including Lawrence Krauss and Neil deGrasse Tyson) would not at all object to a different demarcation problem, the one separating science from pseudoscience. Indeed, the new collection on the latter topic that Maarten Boudry and I have put together for Chicago Press (out in July) includes a chapter by Shermer himself, who clearly saw value in that particular demarcation problem.
And yet, the demarcation between science and philosophy (S-Ph) has a lot of characteristics in common with the one between science and pseudoscience (S-Ps) (and no, that’s not because philosophy is a pseudodiscipline!). The similarities are to be found along three dimensions, as we shall see in a moment, and together these help explain why both projects are worthwhile. 
Before examining the reasons why S-Ps and the S-Ph demarcations are analogous, however, I need to reiterate once more what seems to be a frustratingly common misconception about such projects: when it comes to concepts as complex as science, philosophy and pseudoscience, we will not find simple and sharp dividing lines. There is no small set of necessary and jointly sufficient conditions that identify any of these fields: overlap and gradation is the name of the game. But just because night slowly yields to day it obviously doesn’t mean that there are no differences between night and day. Nor does it mean that, most of the time, we don’t know perfectly well that it’s night (or day). Rather, a better image is that of a complex intellectual landscape where there are peaks corresponding to the concepts of interest, with the peaks gradually extending from their center and overlapping with nearby peaks. And to make things more complicated, you also need to imagine this intellectual landscape as fluid in time: what was once considered science may have turned into pseudoscience, philosophy into science, and so forth.
Let’s then start with the S-Ps demarcation peoblem, to which most of my readers will likely not object. Why does it matter? There are, I submit, three classes of reasons: intellectual, practical-financial, and practical-consequentialist.
On strictly intellectual grounds, we like to have a reasonable theory that explains why a certain way to carry out inquiries that we call “scientific” seems to be working so well (with the usual exceptions) while another class of activities, those we label “pseudoscientific,” doesn’t. The answer can’t just be that if it works it’s science and if it doesn’t it’s pseudoscience. Aether and the planet Vulcan turned out not to exist, but we don’t think they were pseudoscientific notions when they we initially proposed. Similarly, astrological charts and alchemical methods were never scientific, regardless of the fact that they didn’t pan out. Indeed, even when pseudoscientific notions may turn out to work (as it seems to be the case, in a limited fashion, for acupuncture), they are still pseudoscientific. The difference between the two classes of activities lies in the methodologies employed by their practitioners, both at the theoretical and at the empirical levels. So there is a genuine intellectual problem at play here, one that requires the methods of philosophy, history and sociology to be tackled.
Te second reason why the S-Ps problem is interesting is because there are resources at stake, particularly money and time. The label of “science” indicates a worthwhile activity, while the label pseudoscience indicates something that is not worth pursuing. That is why scientists get academic jobs and research grants, while pseudoscientists (usually) don’t. That is also why it is problematic that the National Institute of Health keeps spending money on “alternative” medicine, or that the British “royal family” (sorry, I just can’t bring myself to use those terms with a straight face) supports homeopathic remedies. In fact, when scientists and skeptics object to such funding, or when they decry the occasional university that confers degrees in parapsychology, they do so precisely by invoking a meaningful demarcation between science and pseudoscience.
Finally, when people cannot distinguish between science and pseudoscience there are consequences: parents do not vaccinate their children, governments in Africa don’t distribute anti-retro viral drugs to people affected by HIV-AIDS, and no meaningful political action is taken to stem the rise of climate change.
Again, none of the above requires a sharp demarcation between the two classes of activities concerned, nor is such demarcation necessarily stable over time. But intellectually, financially, and in terms of practical consequences, the demarcation makes a difference.
A very similar discourse applies to the S-Ph pair. From an intellectual point of view, philosophy is a different kind of activity from science, just like mathematics is in turn different (and yet shares similarities) with both. Literary criticism, say, is even further removed from the philosophy-science-math cluster. Recognizing such differences gives us an appreciation of the diversity of scholarly activities human beings are capable of and interested in pursuing, and leads to a healthy respect for the skilled practitioners within each field. [I suspect, incidentally, that most of those who think scientistically about philosophy have not actually read a single technical paper in philosophy, so that they literally don’t know what they are talking about. If they had, they would immediately appreciate that technical science is actually done in a very different way from technical philosophy, a fact readily explained by the idea that there is indeed a meaningful demarcation between the two.]
What about consequences in terms of allocation of resources? Here it is particularly puzzling to see some scientists’ acrimony against philosophy, considering that the natural sciences already command a majority of faculty positions on most campuses, and can certainly count on grants that are order of magnitudes larger than those for which humanists compete. This, by the way, is as it should be. Having done research in both science and philosophy I know first hand the disparity of resources necessary for scholarly engagement within each field. But belittling philosophy and the other humanities does carry the danger of convincing university administrators — already bent on running the academy as if it were a for-profit business — that they should cut anything that doesn’t bring in money (though, somehow, it never occurs to them to apply the same logic to athletic programs). Rest assured that programs in philosophy, languages, history and the like would then be the first ones on the chopping block (as indeed has happened over and over in the past several years).
Lastly, the practical consequences? Ah!, I can hear you say, “practical” consequences of eliminating, or even simply belittling, philosophy? Yes, there are consequences, and they are not good for the health of a democratic society. For all the (justified) complaining that our students and citizens are woefully illiterate in science, they are lacking just as much in their ability to write, comprehend complex texts, exercise critical thinking, be aware of their own cultural history and of the meaning and functioning of the laws of their own society. These are, of course, the elements of a liberal arts education, and they are vital for a meaningful democracy (which is why they are constantly under attack from reactionary forces). Education, in a society where we care about the flourishing of our citizens isn’t just a matter of acquiring skills to enter the workforce (as necessary as those are), it is also a matter of helping young minds to mature and develop in a way that will allow them to make wise choices in their own lives, as well as to contribute to society with more than just their labor. All of this is helped by a healthy recognition of, and respect for, the methods, goals and practices of different intellectual disciplines — from the sciences to the humanities.
These, then, are the reasons why demarcation projects are important. By all means, let’s disagree about the criteria that separate science from philosophy, and both of them from pseudoscience (and pseudophilosophy — there is such a beast). But let’s do that in a productive and mutually respectful fashion, not with silly declarations such as “if it has to do with empirical evidence it’s science,” which makes no more sense than to claim that everything that has to do with thinking counts as philosophy. And of course, let’s not neglect the many bridges or borderline areas between science, pseudoscience, philosophy and all the rest. Those areas may very well turn out to be interesting and fruitful in their own right.————
 This post is not about how to separate science from philosophy, or science from pseudoscience per se. On the first topic I have written several times on RS and in other places, and the second has been covered broadly in my Nonsense on Stilts, and will be revisited more in depth in the forthcoming edited volume, Philosophy of Pseudoscience. Here, I am concerned with the distinct question of why such demarcation problems are worthwhile to begin with.