Hey there, a reader on Daily Nous suggested that the final plot of degrees conferred versus jobs posted wasn't that bad with the former relativized to the latter.

The lower 'ratio' line is degrees conferred per job posted. It looks fairly stable, hovering around 1.3 PhDs per j0b posted. There's a downward trend from 2020-2022, but it's too soon to say whether that's a new trend or a fluke. But we can make the analysis a little more sophisticated. The values for 'ratio' don't take the backlog of PhDs into consideration. Here's an idealized model: every PhD who didn't get a job in year T will apply again in year T+1. (This is idealized because some PhDs will stop looking for jobs, many of these jobs are fixed term, some PhDs will look for new jobs, etc.) In the idealized model, the backlog of PhDs accumulates over time, and it hovers between 1.5 and 1.6 PhDs per job.

These idealized models suggest that the market is bad (on average, 2 jobs for every 3 philosophers) but it's not getting worse over time.

Greetings! As the fall term gets underway, it's a good time to take a look back at the job market: job ads from January-June of 2024 in particular. As a reminder, I break the application cycles into primary (July-December) and secondary (January-June). We'll look at the numbers for the secondary cycle of 2024 and then put them in wider context. You can find the dataset and code under "Blog Data."

Here's the big picture:

Junior faculty leads (206), followed by postdocs (108), and visiting positions (55). Now we'll take a look at contract types.

There are far more fixed-term positions offered than tenure/tenure-track, which is standard for this cycle. Onto the intersection of contract types for junior and postdoc openings:

Now that we've seen what's going on now, let's add some historical context.

On the upside, there's been an overall increase in the number of jobs advertised over the last decade with the exception of open contracts. Open contracts have decreased slightly over time. For fixed-term positions, the last two cycles have seen a cooling-down after a post-COVID bump. But how do junior and senior faculty fare in all this? Let's look at junior and postdocs first.

What's interesting about this is that fixed-term positions for both junior faculty and postdocs have been increasing over the last 10 years, but openings for postdocs have had greater increases relative to fixed-term junior positions. What could account for this? One possibility is that more philosophers are winning grants, so the demand for postdocs is increasing. Another is that departments are getting specially-allocated funds for postdocs from private foundations or donors. I don't really know but those seem like plausible guesses.

Independently of the causes, the increase in postdoc demand is a double-edged sword. On one side, freshly-minted PhDs can have some more training and support as they get their careers off the ground, especially new PhDs from programs outside the Philosophical Gourmet Top N (take your pick of an appropriate value for N). On the other, postdocs are typically paid worse than junior faculty.

Looking at positions other than junior faculty and postdocs, there's a noticeable trend for senior faculty hires as well as visiting positions.

Both are trending downwards, and visiting fellowship positions have more or less leveled off for the last four years. To explain the decline, my only guess is general belt-tightening: senior faculty are more expensive than junior faculty.

Finally, let's compare hiring in the primary and secondary cycles.

This plot is disheartening. The trend line for tenure-track jobs in the main hiring cycle is flat over the last 10 years while the trend line for fixed-term jobs is increasing in both cycles. In fact, the average number of jobs per year for fixed-term positions is 145 (primary cycle) and 157 (secondary cycle). The average number of tenure-track jobs is 184 (primary) and 33 (secondary). Let's combine both cycles to get the bigger picture.

While there have always been more fixed-term than tenure-track jobs advertised in the last 10 years, the former is growing at a faster rate than the latter. We've noted this in previous analyses, and the pattern continues.

Finally, let's look at the same data but for junior and postdoc positions only.

No surprises: the same pattern holds.

One final plot before closing. These data are especially sobering when compared with the cumulative number of philosophy PhDs conferred every year.

The data for PhDs conferred comes from the Survey of Earned Doctorates. (Go to "Build a Table" to find the historical data for philosophy.) The plot here ends at 2022 because that's the last year for which the data from the SED are available. From 2014-2022, not only have there been more PhDs conferred than junior jobs advertised, that gap has been growing.

Recently, the APA's Committee for Non-Academic Careers has set up office hours to talk with philosophers in industry, shared by DailyNous. Given the data, folks working on their PhDs (as well as newly-minted PhDs) might do well to take advantage of those office hours.

On that note, let me know if there's any other analyses you'd like to see.

Greetings from the Pacific APA! There are lots of great papers and folks--it's just day 2 and the sessions on Phil of Psychiatry, the History of Analytic Phil, and Care/Forgiveness/Trust have been just phenomenal.

To the task at hand: Over the course of last semester, I sent out requests to heads of search committees to fill out a brief survey to figure out how many people were on the job market this past cycle.

First, a HUGE thanks to Gonzaga undergrad Kate Ferrell and Gonzaga philosophy grad student Tobias Propst for collecting the names and email addresses. Without them, this would have taken a lot longer.

I sent out 239 emails and gathered 79 completed responses. Thank you to everyone who filled it out!Here's the breakdown of responses by Carnegie classification:

Certificate:  1
Bachelors: 25
Masters : 6
Doctorate : 43

I didn't collect identifying information about the institutions or programs. I asked for number of applicants, AOS, Carnegie classification (R1, R2, MA, BA, certificate), teaching requirements, and some subjective impressions about volume of applications proportion of teaching and research.

(Quick note about terminology: from here on, I use "status" as shorthand for the category of highest degree awarded. It seems like a neutral-ish and short term for the category.)

Ok, so let's get to it. The big picture: the average for numbers of applicants across all responses is 136. Given the responses and the number of invitations sent out, the margin of error is +/- 9% (with 95% confidence level). So the average for number of people on the market is probably somewhere between 125 and 147. That being said, there is a lot of variation.

The number of applicants to jobs at doctoral institutions was higher on average than any other, but there was a lot more spread around that average. This is clear from the table, but also look at the following plots. Differences among numbers of applicants to BA-granting institutions are roughly linear. But for PhD-granting institutions, it's closer to exponential. (There were relatively few MA-granting institution responses, so I'm hesitant to draw conclusions.)

Now let's pull together a bunch of info about applications to various programs: the fewest, average, median, and greatest number of applicants by program status.

It looks like jobs with PhD programs are getting a lot more applications than any other. In fact, the values for max and average applications decreases as the status of the program goes from PhD to MA to BA. (It's hard to draw conclusions about jobs at certificate-granting institutions since there was only one response.) A first hypothesis: Applicants preferentially apply for jobs at PhD-granting institutions. Two things to check: (i) whether the trend holds for R1 and R2 and (ii) if the values are a result of open-rank hires.

For (i), the trend doesn't hold for differences between R1 and R2 jobs among respondents, but it's worth noting that there were relatively few R2 respondents.

For (ii), let's restrict the data to TT jobs only. That way, our numbers aren't inflated by open-rank hires.

I think our initial hypothesis is right: applicants are biased towards applying for jobs at PhD-granting programs but the data don't suggest that this bias holds for R1 vs R2 programs.

Let's look at the numbers breaking out by AOS and program status. An important caveat: our 9% margin of error was for all responses to all surveys sent out. Once we begin dividing up the pool of responses, the margin of error increases and sometimes substantially. Suppose that the split between jobs advertised for BA- and PhD-granting institutions was 50/50. The MoE for the BA-granting average jumps to 17% and for the PhD-granting average to 12%. So once we begin parsing into subgroups, I'm saying less about all jobs for this year and more about the responses. I'll try to be explicit about that where appropriate.

 Here's how to interpret the plot. The vertical grey line is the average for each AOS. The error bars give the minimum and maximum values for each status and AOS. The point in the error bar is the average for AOS and status. The tables for these values are here.

This plot pulls together a lot of what we've previously mentioned. Jobs at PhD-granting institutions tend to get more applications than elsewhere, particularly for open and history/tradition AOS. The jobs that on average had the fewest applications among respondents were history/tradition jobs at BA-granting institutions. In fact, in 2 out of 5 cases, the max number of applications for BA programs is less than the average for that AOS.

My interpretation of PhD-programs having more applicants is a manifestation of two sides of prestige bias. On one side, it's plausible that many candidates want to apply almost exclusively to prestigious jobs, where being a PhD-granting program tracks prestige. The other side of the coin is that candidates sell themselves short, thinking that there's no way they could make it past the first round at an R1 or R2.

That being said, if there are other plausible interpretations, please feel free to share!

One of the takeaways restricted to the respondents (but is perhaps indicative of the wider field of applicants and openings) is that if you're applying to R1's, you'll likely find yourself with a lot more competition relative to MA- and BA-granting programs, particularly for open, science/language/math, and history/tradition AOS's.

Two more tables before calling it a day: one about impressions of applicant volume and another about time allotment for the job.

So that's good to know: typically, committee chairs have decent expectations for the numbers of applicants. How does this compare to the numbers of applications received?

The "fewer" bar all the way to the right? That's 515 applications for an open AOS, open rank position at an R1. With those parameters, I'd have guessed there would be more too.

That's useful to know: the typical response is that time spent teaching and researching is roughly equivalent. The breakdown for that number is:

BA-granting institutions: 8
MA-granting institutions: 4
PhD-granting institutions: 22

...which I found pretty darned interesting. As you might imagine, "mostly research with some teaching," was largely PhD-granting institutions (12) and "some research and mostly teaching" was largely BA-granting institutions (10). The lesson, I think, is that job applicants can't afford to neglect teaching or research in their applications but have to discern which to amplify for the job applied to. I think that's common sense, but it's nice to have some info to support it.

If you'd like the raw data and script, you can find them under the "blog data" tab.

Well, that's all. If there are any other analyses you'd like to see, please let me know!

Hello again, friends! The numbers are in, and let's take a gander at the job market this past cycle. As a reminder, I define "primary cycle" as running from July to December. Let's start with the overall picture.

The biggest groups were tenure-track and postdoc, followed by open rank hires. Here are the numbers for each.

Now for the historical perspective. The solid lines are the regression.

Some depressing info here. For the primary job cycle, tenure-track jobs for junior faculty have shown no growth since 2014. But things look different if we leave out 2020. (More on that below.) Excluding that awful awful year, tenure-track jobs trend upwards. Huzzah!

Fixed-term junior jobs are trending slightly downwards. But given what the data suggest about more fixed-term jobs being advertised in the secondary cycle, this isn't too surprising.

What jobs are rising most quickly? Postdocs. That's good news for freshly-minted PhDs as far as paying next month's rent, but those jobs are for a limited time only. Additionally, it's no secret that postdoc gigs tend to be relatively low-pay for high-labor. Some postdocs are pushing back. (That article looks at science postdocs and not philosophy ones. If readers want to chime in with their experiences, please do.) One explanation is that more stable and high-paying jobs are being replaced by postdocs, which saves universities a lot of scratch--the research version of relying on adjuncts to meet demand.

Finally, let's take a look at junior positions over time.

Fixed term and tenure-track junior positions are below and above average (respectively) but still within one standard deviation. But what if we exclude 2020 as an outlier? Here's what we see:

Tenure-track jobs did unusually well in 2019 and 2022 primary cycles, and 2023 is below average but not terribly so. Fixed term is lower than one SD and postdocs are higher than one SD--both of which, I think, are concerning trends even with this more optimistic view of the market.

So what to conclude? I think the big question is, should 2020 be excluded from the data as an outlier? Doing so gives a rosier view of the market. However my own hunch is to say "no," and not just because I haven't had my afternoon coffee. An oft-shared joke on social media is that we're experiencing once-in-a-lifetime events more frequently. The recession of 2007-2009 hit the market hard. Just over a decade later, COVID hit the market hard again.  It is, of course, a challenge to predict global catastrophes, but there will surely be more as the effects of climate change are felt more acutely. As a matter of personal preference, given the choice of being optimistic or guarded about the future of the philosophy job market, I play it conservative.

Still, I'm open to arguments that 2020 really is an outlier and can be (relatively) safely excluded.

That's all. Thanks, as always, for reading. If there are other analyses you'd like to see, please let me know!

To my fellow R users:
If you want to start playing with new color palettes, try out coolors.co. It has some great tools, and it's how I got the colors for these plots. And if you want to try out new fonts (these plots use Source Sans 3 from Google), this is a great guide.

Hello! This post is going to be a long one. So get some coffee or tea or beer and let's get to it.

First some preliminaries. Counting AOS's is hard. Here's how I'm doing it. There are ads that had one AOS: history/traditions, open, etc. But there are others that are disjunctions: history/traditions or metaphysics/epistemology. Since we're looking at major trends, I counted each member of the disjunction separately. So if the AOS was metaphysics/epistemology or value theory, then the total count for metaphysics/epistemology goes up by one and same for value theory. Breaking up and counting disjunctions in this way tells us more about disclipinary trends than either (i) dropping one-off disjunctions or (ii) keeping disjunctions as their own categories. If you're interested in more fine-grained analyses, go ahead and check out the data set. It's available in the "blog data" section of the website.

Second, there are some AOS that don't fit neatly into any of the categories on PhilJobs, like philosophy of technology. I did my best in each case to figure out what the job was looking for and then apply the relevant tags. Readers are welcome to follow up with the data under the "Blog Data" tab.

Third, the numbers in each plot are not a count of jobs. If an ad has A or B or C as main AOS's, each AOS counts in the tally towards the number of A's, B's, and C's in the data. And if we're treating the numbers as counts for jobs, then that one job gets counted three times. So use the values as comparisons within a plot, not across plots.

Finally, AOS's here are divided into two kinds. There are main AOS's. Those are: (1) open, (2) value theory, (3) metaphysics/epistemology, (4) history/traditions, and (5) science/logic/mathematics. Under each of these (except "open"), there are other, more detailed AOS's, like ethics, aesthetics, and philosophy of religion under value theory. "Main AOS" will be used to pick out one of the five top-level categories. "Detail AOS" will be used to pick out second-level categories.

One of the themes I'll come back to throughout this post is looking at detailed AOS's alone versus with other AOS's. It seems this can provide some insight into the structure of the discipline. First, a working hypothesis: Ads featuring multiple main AOS's would be more likely to hire someone who ticks as many boxes as possible, all other things being equal. Why? Suppose an ad asks for specialization A or B. If you have two candidates who are equal in all other respects except one specializes in A while the other specializes in A and B, the latter candidate is more desirable than the former. If that's right, then the presence of multiple AOS's in an ad is a proxy for the desirability of candidates with fingers in several AOS pies and the popularity of crossing intradisciplinary boundaries.

Ok, let's do it.

First the big picture:

Value theory shows up most often in AOS, which (to be honest) is surprising. I would have guessed "open" but here we are. The reason is that value theory showed up in a lot of disjunctions: value theory or metaphysics/epistemology; value theory or history/traditions, etc--that is, lots of jobs were for value theory alongside some other area. Now, if we don't break up the disjunctions (against what I said above), "open" has the greatest number of ads. So the story here is that value theory had a very good year, showing up a lot both as a solo AOS and alongside other AOS's. Metaphysics/epistemology, science/logic/math and history/traditions are roughly on par with one another.

Personally, I find metaphysics/epistemology bringing up the rear to be kind of surprising. When I was in grad school (*cue sounds of a rocking chair*), LEMMings were all the rage. Maybe not so much anymore.

Ok, now let's zoom in on each of these

Here are the more detailed areas for ads in value theory.

Fairly self-explanatory, but a few things of note. First, for the top 11 detailed AOS (ethics through business ethics), the overwhelming majority of ads feature value theory alone as the main AOS. Second, there's nonetheless a long tail on the data, so value theory shows up with a lot of other AOS's.

The dominance of ethics and social-political philosophy isn't too surprising. These are mainstays of gen ed requirements.

You'll notice some that seem strange on a list of value theory: philosophy of computing, epistemology, and philosophy of mind (among others). Let's parse out the data to see differences in ads that are value theory alone and value theory with other AOS's. First we look at value theory all by its lonesome:

The structure of the data (more or less) reflects the big picture above: ethics, soc-pol, and applied ethics taking the lead with other value theory categories trailing behind.

Here's how to interpret this plot: each of the AOS's on the left appeared with an ad tagged "value theory." The colors are the other main AOS's that the keyword appeared with. For example, look at "philosophy of mind": when it appeared in ads tagged with "value theory," those ads were also tagged with history/traditions, metaphysics/epistemology, and science/logic/mathematics. Of those other tags, metaphysics/epistemology showed up most often.

The big trends are interesting. Whenever "ethics" shows up in the context of other main AOS's, it's most often with metaphysics/epistemology. So one would expect that ethicists who also do some metaphysics or epistemology would be better positioned for these kinds of jobs than folks doing only ethics.

Likewise, notice under "applied ethics" that it occurred most often with the main tag "science/logic/math." On reflection, this makes sense given the rise of ads looking for philosophers working in philosophy of tech and/or AI (discussed below).

When "philosophy of race" shows up with other main AOS tags, it's exclusively with history/traditions. This reflects what we find in the data: ads looking for philosophers of race typically mention non-Western philosophy in the ad.

A word on the data: whenever an ad listed a string of non-Western traditions, I used the label "trad_non_west." This covers anything that isn't Western philosophy (including American and Continental). The only time I didn't use this tag for a string of non-Western AOS's is if there were only 2 or 3 members of the disjunction (e.g. "Latin American or Native American").

A few things of note. Ancient and modern are the most looked-for. Continental and the history of western philosophy are just about even. And fairly far behind is any kind of non-western tradition: Africana, Asian, Arabic/Islamic, Native and Latin American. Bringing up the rear for traditions, we find Native American and Latin American. Now, they're included under the generic "non-western" label, so not all hope is lost. The prudent interpretation is that departments in the last year are more interested in Native American and Latin American as members of non-Western traditions rather than as traditions in their own right.

Out of curiosity, let's compare tags for Western philosophy vs non-Western.

There are a lot more ads mentioning Western philosophy (75) than non-Western (25). The lion's share of non-Western tags are for anything in a non-Western  tradition. Native American philosophy appeared once last year. So while non-Western traditions are appearing more and more on the contemporary scene (e.g. Chalmers's book Reality+, Peter Adamson's fantastic podcast), they're still lagging far behind in job-ad mentions.

Now to look at history/tradition filtering out any ads that included other main AOS's:

Looking at history/traditions detail AOS, where only "value theory" is the main AOS, we see the history of Western philosophy is most advertised-for: either ancient, modern, or just in general. Ads under "hist_trad_other" include detailed AOS's that don't fit neatly under any other history/tradition category, like Kantian, post-Kantian, phenomenology & history of philosophy, and digital humanities.

Now let's look at history and traditions where they appear with other main AOS's.

First, notice the relatively small number of mentions: history and traditions don't typically show up with other main AOS's.

Here's a noticeable trend: history/traditions ads, when they appear with other main AOS's, are typically coupled value theory (21). It doesn't seem that departments are looking for history/tradition (8) or metaphysics/epistemology (10) alongside someone doing something in science, logic, and math.

The big picture:

Phil science dominates. Under "slm_other" we find ads for philosophy of AI, science and technology studies, complexity/emergence, and technical methods (e.g. agent-based modeling). The least sought-after specialization in this subfield is decision theory. But it's not as though philosophy of cognitive science or philosophy of social science are exactly dominating either.

Philosophy of science dominates, but as David Wallace pointed out in a comment on Daily Nous way back when, other specializations can apply for philosophy of science jobs. But again, we're looking at trends here and not necessarily numbers of jobs, and it's worth noting that more departments are looking for philosophers of science rather than philosophers of some particular science.

When science/logic/math ads appear with other main AOS's, metaphysics/epistemology dominates (59), with value theory trailing (35), and history/traditions a distant third (8).

Something of note: there is a sizable number of co-occurrences of philosophy of computing with "value theory" tags and there are quite a few "applied ethics" showing up with science/logic/math ads. A lot of this can be chalked up to burgeoning interest in ethics of technology and AI. Let's dig into that.

This table tells us how often "AI" and/or "technology" appeared with each of the main AOS's since July 2022 (i.e. the last job cycle).

(If you're wondering about "open", those are ads that had an open AOS but tech/AI for AOC.) By and large, departments looking for philosophers of tech/AI are looking for ethicists rather than folks working on (say) computing theory and AI or epistemology and AI.

Now for the historical perspective. For this last job cycle, 54 ads had "AI" or "technology" appearing in either the AOS or AOC description. (AOC is included because some AOS were value theory but the AOC explicitly mentions technology or AI.) Let's take a look at this by calendar year

Clearly, ads for philosophy of technology and/or AI are increasing steadily. Who knows when it will plateau? Of course we don't want Disco Stu projections...

...but it seems like phil of tech/AI is here to stay.

Last but not least: metaphysics and epistemology. First, the big view:

Now let's dig into M&E ads without any other main AOS's:

Well well well....looks like we found our LEMMings.

Under the "other" tag, we have (i) an ad looking specifically for a social epistemologist and (ii) an ad for the epistemology and/or cognitive science of religion. (And the "meta_epist" tag is where the ad's AOS was "metaphysics or epistemology.")

Of those ads featuring M&E, they occur more often with science/logic/math (22) and value theory (16) than history/traditions (11).

So what can we conclude from all this?

First, in comparing solo main AOS with multiple main AOS in ads, only metaphysics/epistemology has fewer solo instances than joint instances. This suggests that metaphysics/epistemology is sought after but mostly in conjunction with other areas (e.g. mind and cog sci or epistemology and philosophy of science). There was less demand this year for straight-ahead metaphysics/epistemology.

Second, non-Western traditions weren't all that sought after this last year compared to Western traditions.

Third, philosophy of technology/AI is on the rise, particularly in connection with moral questions.

That's all for now. If you have any suggestions or analyses you'd like to see, please email me

Hi friends! Now that we're in the dog days of summer, it's time to look at the job market for the secondary cycle. (Friendly reminder: the primary cycle runs July to December; the secondary cycle runs January to June.) We'll look at the numbers and focus on a couple of interesting trends.

First up, let's look at the volume of jobs posted for each of the job types on Philjobs.

As you might expect, junior faculty and postdoc jobs are the most prevalent, followed by visiting and then open rank. But lets put this in historical context.

The colored lines had, at least once, at least 25 postings in a secondary cycle. So this is heartening: ads for both junior jobs and postdocs have been going up over the last decade. Huzzah! All other jobs have either declined slightly or stayed flat. Let's zoom in on junior jobs

 This is less heartening, but probably expected. Fixed-term jobs are increasing annually at a faster rate than tenure-track jobs. Again, this is for the secondary cycle--I think received wisdom is that fixed-term jobs are more prevalent here than in primary cycle. So let's check the data for this. (BTW we're more interested in general trends now, so I'm going to drop annotating the lines with numbers.)

What we have here breaks out jobs by: (1) year, (2) cycle, and (3) contract type. This confirms our earlier suspicions: more TT jobs are posted in the primary than the secondary cycle and slightly more fixed-term jobs are posted in the secondary cycle.

But there's something odd here: the trend line for TT jobs for the primary cycle is decreasing while the trend for fixed-term jobs for both cycles is increasing. This suggests that there are fewer TT jobs and more fixed-term jobs as time goes by. Let's check this. First, we're going to exclude 2020--that was a terrible year for the job market and perhaps that's dragging the TT primary cycle trend line down.

That's not much comfort. Excluding the worst year for the market since 2008, TT jobs tend to stay steady while fixed-term jobs increase.

Another possibility is that the trend lines are affected by the number of jobs posted each year. More fewer posts means fewer TT and fixed-term jobs. Maybe scaling TT and fixed-term jobs to number of posts tells a different story?

Ok so the story here looks slightly worse. Relative to all postings in a year, there are progressively fewer TT jobs and more fixed-term jobs for the primary cycle. Fixed-term jobs remain relatively stable for secondary cycles over the last decade while TT jobs are increasing slightly during second cycles. Let's switch back to raw (instead of relative) counts for a moment  and look at fixed-term and TT jobs by cycle.

Indeed the trend seems to be fewer TT jobs and more fixed-term jobs advertised in the primary cycle.

I'm as much of a Pessimistic Paulie as the next philosopher, but maybe, maybe there's reason for a guarded optimism. Take a look at the trends in the differences between TT and fixed-term jobs advertised by year and cycle.

This suggests that, over the last decade, the differences between the number of TT and fixed-term jobs generally decreases. What can we conclude from all this? First, we shouldn't expect the number of fixed-term jobs to overtake the number of TT jobs, at least not in the near future. Second, this is consistent with more fixed-term and fewer TT jobs advertised. (Note that this does not include adjuncts. It's hard to find reliable data (or any data, for that matter) on adjunct positions.) The safe bet is this: there will continue to be fewer TT jobs and more fixed-term jobs in the near future, but long term there will continue to be more TT than fixed-term jobs. At least, that's the hope.

That's all. If you have any other analyses you'd like to see, please let me know! I've posted the data and code for these analyses under the "Blog Data" tab.

​Hello! I'm Charlie Lassiter. You may remember me from such blog posts as "The Job Market: We're All Doomed!" and "What Does Brian Leiter Blog About?" I'm here today to help answer the question, "what are newspapers saying about philosophy?"

This was inspired by my colleague Mark Alfino, who keeps tabs on trends in higher ed. I says to myself I says, "I wonder how academic philosophy is talked about in newspapers?" I'm typically worried by articles like this one from Axios. And while I plan to get to online sources soon, newspaper turned out to be the easiest to tackle first.

I used LexisNexis to get all American newspaper articles from the last 10 years mentioning "philosophy." To make sure I was getting returns on academic philosophy instead of the makeup brand or things like "her investment philosophy is...", I first filtered by "liberal arts" and then by "philosophy." Also, I limited the search to national and local newspapers. This means leaving out Chronicle of Higher Ed as well as some other higher-ed-specific sources. This gave me a batch of 686 articles. After that, I filtered out obituaries; they were throwing off the  results.

Sentiment analysis is exactly what it sounds like: getting an analysis of the positive and negative sentiments in a text. There are a few different ways to do this. The AFINN corpus from Finn Arup Nielsen tags words with a value between -5 ("bastard", "motherfucker") and 5 ("outstanding", "superb"), excluding 0. We'll be using that. Nielsen assigned these tags by hand and then checked his ratings against other corpora. You can find his paper discussing methodology and validation of his ratings here. I like it because it gives some nuance between bivalent "positive" and "negative" assignments, like you find with Bing Liu's lexicon.​

Hey all. I've been wanting to look at AOS's but it's a bit of a pain since the data has to get cleaned. No matter. It's done. Here's the big picture:

A note on method: In cases where multiple areas were listed, I've counted those separately. So this isn't a tally of all jobs on the market. Rather, it's a tally of all the jobs advertised in one of the five main categories on Philjobs. So (e.g.) there are a hair under 50 jobs in metaphysics/epistemology. But some ads listed meta/epist or history of phil as the AOS they were looking for. Here, each of those was counted separately. I opted for this method of counting because I'm assuming that any disjunctions in AOS's for job ads can draw from two different applicant pools: that is, that AOS pools are exclusive. (I know this isn't always true but go with it as a simplifying assumption for the model.) For (for instance) a job ad looking for an epistemologist or an ethicist gets listed here as a job for an epistemologist and as a job for an ethicist. So if Jack is an epistemologist and Jill is an ethicist, the job ad counts as a listing for Jack and for Jill individually. In a nutshell, I'm counting jobs by AOS from the point of view of job seekers under the assumption that multiple AOS's have mutually exclusive candidate pools.

Let's zoom in on a few of these, starting with value theory.

Ethics dominates, followed by social and political philosophy. We'll get to "other" in a moment. Sadly, phil of art is at the bottom with two ads.

Value_other is a catch-all for specific areas that don't fit neatly into any of the other categories. It was a hodgepodge of seemingly grant-specific areas, e.g. citizenship, media ethics, etc. But there were 3 ads for AI ethics and 2 for ethics and tech. So there are more job ads for ethics and tech/AI than there are for phil of art or phil of sex and gender.

Let's take a closer look at history and traditions to see what's up there:

Carving up this data was a bit trickier. Some ads just had "history of philosophy" without any detail. Others said "non-Western". For the former, I listed ancient, medieval, and modern as covering a generic "history of philosophy." And for non-Western, I listed Asian, Africana, Indigenous, Indian, and Latin American as proxies. So again, keep in mind that these aren't total job counts but rather (roughly) jobs for which one could apply if one were an expert in (say) ancient or Latin American phil. Given these assumptions, ancient is in greatest demand. 

Non-western traditions are on the map, but not in huge demand. "Africana" showed up in 13 ads while "indigenous" showed up in four. Though it's worth noting that Africana is in higher demand than a number of subcategories in value theory.

Now science, logic, and math:

Philosophy of science is leading the way with relatively few opportunities for other subfields. What's up with the other category? This is a grab bag of: history of science or medicine or technology; AI; the metaphysics of science; postdoc ads that are too specific to classified anywhere else (e.g. a project on episodic memory)

The last category: metaphysics and epistemology.

Seven jobs for philosophers of mind? Nine for epistemologists? Yeesh. what's in the "other" category? Five philosophy of technology positions, a social epistemology, and a couple of animal cognition.

What's interesting about this from my view is when there are repeated instances in the "other" category. This suggests the emergence of an important subfield that isn't yet counted among the standard options, or at least not in any obvious way. Decision theory has a subcategory on Philjobs, but philosophy of technology doesn't. Nor does history of science/STS, even though phil tech and history of science each had more ads than decision theory. I'd say that history of science/STS and phil tech are established subfields. A colleague of mine who works in phil tech once described it as fringe. Kirby, if you're reading this, I don't think it's fringe any more. 

Let's now filter AOS by junior positions and post-docs. Here's what that looks like:

So trends for junior positions largely follow that of all positions. Most are in value theory, followed by open and history/traditions. Things are a bit different for postdocs, with the greatest number of positions being advertised as open.

In talking with my friend Nader Shoaibi, I wonder if digging into the details makes the picture a bit drearier than the numbers here say. I'll use myself as a test case. If I were going on the market, I'd apply for open jobs and jobs in philosophy of mind. I'd apply for everything but would really be looking at tenure-track jobs--that's where the security is at. Filtering for those values, there are 36 jobs: 33 open and 3 in philosophy of mind. I got my PhD at Fordham, so let's be realistic about my job prospects. I know, I know...that one person went from Oklahoma to Harvard (apologies I can't remember their name). Fantastic for them but clearly an outlier. At a non-prestigious university, I can cut the Ivies and lots of R1s from the list. And let's take jobs outside the US off the table because I'm not a superstar and a university outside the US is unlikely to pay for the costs of hiring a foreigner. That leaves me with 16 openings, all advertised as open. 

That's all to say that the numbers don't look great in the big picture and things only go downhill when getting into the weeds. 

I'm hoping to get some help soon for looking at AOS trends since 2015. It's a lot of data to clean. But I'll be sure to let you know when that's available. 

As always, if there are any other analyses you're interested in, please let me know!