Some Unscientific Demographics

Josh Habgood-Coote recently asked on BlueSky if there's any way of finding the most cited articles in philosophy overall. He noted that a recent informative Weatherson post, who was looking at a slightly different thing - citation networks within philosophy journals. What Habgood-Coote ended up doing was extracting the ten most cited papers from the 2019-2023 period using google scholar, and linked them here. Using my blog to brag a bit, I am happy to see Vindicating Methodological Triangulation there! But I decided I would like to look at something slightly different while we are here having this conversation.

I wanted to see what topics philosophers are writing on that are actually getting attention. What are the areas of interest that drive conversation in or through our field, where is it that our work seems to get recognition and uptake, or at least spark conversation? That I thought I could do by looking at some of the highly cited philosophy journals and looking at a smattering of the highly cited papers therein, seeing what subfields they fall within. And while I was there I would take a butcher's at the author demographics, as a little treat.

I set myself a very tight deadline though, and that created problems. Various ways in which I made arbitrary decisions going into this: initially I intended to just combine the two obviously Philosophy coded lists on google metrics - this one and this one - into a top ten by h5-number, defined as "h5-index is the h-index for articles published in the last 5 complete years. It is the largest number h such that h articles published in 2019-2023 have at least h citations each." But then when I was doing that I got cold feet about Phenomenology and the Cognitive Sciences, Mind & Language, and Review of Philosophy and Psychology (all of which would have appeared in that list) since I worried the overlap with psychology was so great that maybe they didn't get at what I was interested in; maybe they were driving conversation by somehow borrowing from the glitz of on going conversations in psychology, somehow making them not-representative? By the time I had finished actually gathering the info I actually regretted that decision and no longer think it was a valid concern, but given my deadline combined with my laziness I couldn't go back. Add in some other fudging I did to deal with ties and wanting an even number for aesthetic or platonic-mystical reasons, and I ended up with 14 journals I actually looked at:

Synthese
Philosophical Studies
Noûs
Philosophy and Phenomenological Research
The British Journal for the Philosophy of Science
Philosophy Compass
Mind
Philosophy of Science
Studies in History and Philosophy of Science
Topoi
Inquiry: An Interdisciplinary Journal of Philosophy
Philosopher's Imprint
Erkenntnis
European Journal for Philosophy of Science

They have in common that they are among the philosophy journals with the highest h5-numbers, but I don't think they form a natural class and there are other journals that one might reasonably have put in here that might easily be correlated with stuff I am going to talk about. 

From these journals I then took the top 5 most cited papers and by eye-balling authors and abstracts coded what subfield I think they are in and the race (per US categories that I am familiar with from past work) and gender of authors. I think there is a fairly large chance I misgendered some NB people, so apologies on that front. With ethnicity I was largely more confident except in two cases (Felipe Romero, who I counted as a white male, and Nurbay Irmak, who I counted as an Asian male). And in one case there were so many authors on one of the highly cited papers that I simply didn't count it in any of the analyses; so much for even numbers. The very scientific method of me deciding what sub-field categories I would put things in based on vibes in the abstract, on the other hand, I can see no flaws with and feel no need to apologise for. I ended up with these sub-field categories, for what it is worth, based on what I was seeing:

Philosophy of Science
Social Epistemology
Traditional Epistemology
Metaphysics
Philosophy of Computing
Language
X-Phi
Metaphilosophy
Mind
Ethics
Political Philosophy
Aesthetics

All in all I think the message has to be - don't take this too seriously. I am in a rush to get this out because I am trying to stick to one blog post a month, and part of the point of doing things on blog posts is that they do not all have to be held to journal standards. But just to be very upfront on the matter I do want people to be aware of the limitations So: do not induct from this, do not form any detailed opinions on general demographics based on this, your inference will not be warranted. Are we clear? I hope we are clear.

Here's a table with what I found based on such rigorous procedures:

Note that what is being counted here is - papers with at least one author of such and such demographics, or which I judge to be in that subfield but possibly among others. Hence the numbers can add up to more than 70/69 (nice).  Some thoughts on this, bearing in mind all the caveats about being wary of drawing any general inferences - so they are more just thoughts sparked or prompted, rather than justified, by this list.

1) Philosophy of Science is over-represented for basically the same reason that I got nervous about including the Cognitive Science journals (and hence why I should probably have just included them) - the fastest way to being well cited in philosophy is to have impact outside of philosophy. Go look at the papers on Habgood-Coote's list and see who is citing them if you don't believe me! This is in line with my general optimistic sense that for all philosophers angst about no-one outside of philosophy listening to us, there's actually quite a healthy degree of co-operation between philosophy and other sciences. I think this is reflected in where analytic philosophy has clearly done well at influencing broader intellectual culture in a positive fashion, for instance. A lot of the philosophy of mind included here was also at an evidently thriving intersection of science and philosophy, and clearly if I had included the Cog-Sci journals that would only have increased.

2) While the specific proportions of people from various groups shouldn't be taken too seriously, my strong bet is that the point re people of colour being a fairly tiny proportion of authors on the highly cited papers list will turn out to be representative. This is because it coheres with other things I've seen. Usually when I am looking at people of colour in well cited works I always end up having to thank Jaegwon Kim for basically holding up team-rest-of-the-world on his shoulders like Atlas. So in that spirit I will give Rima Basu the shout out here for kinda doing the same for non-white women!

3) Whereas women are authors on around 40% of the papers in this weirdly constructed data-set, which in some sense over-represents them relative to the general population of professional philosophers. And the thing is, the journals I excluded but probably should have included were ones with overlap with psychology - which actually has more women, so if anything I think this data-set is unrepresentative in that it undercounts women. For good reason taking citation as a measure of quality is something no one would advocate for straightforwardly, but at the same time I don't think it is totally unrepresentative. So my belief that women publish less in part because they are investing the time in doing better work (which receives some better empirical support here) is, at the least, not challenged by this data.

4) I specifically went in looking for Philosophy of Computing as its own little subfield - the specific number is again not that important, because there is of course arbitrariness. Does working on misinformation and fake news count as doing something computing related? Not intrinsically, but it's also clear that social media is what has given the topic its impetus. How about using simulation methods in your work, even if the topic you are using those simulations to talk about has nothing especial to do with computers? I ended up saying "no" on the second of those questions and "yes" on the first, for what its worth. But it being around 1/10 papers in this weird data-set is, I think, probably indicative of its genuine rise as a field - in this case I think so because it matches what I and every grad student has noticed about the job market in our discipline these past few years. The philosophy world, at least, is behaving as if something is afoot with the increased prevalence of automated-thinking-machines in our lives and work.

5) But the actual stand out thing was the strong prevalence of epistemology. Even the "Philosophy of Science" category was very often composed of writings on scientific method that could reasonably be thought of as epistemological, simply focussed on the epistemology of a particular scientific field. I didn't always cross classify those as epistemology, so the two epistemology sections combined under-count the total number of epistemology. Likewise the "ethics" and "politics" pieces were often co-categorised as epistemology, which I think is a big part of the "applied turn" that Warmke has commented on (my own previous thoughts on that here). Again, don't draw any conclusions just from this data, so what I would say is: my working hypotheses is that the things currently driving engagement with published works by professional philosophers concern epistemological issues. For whatever reason it is that sub-area, that domain of our expertise, which most often generates comments and commentary.

I look forward to seeing whether more systematic and better carried out work would actually confirm or deny this hypothesis!