The Largest-Ever Survey of Philosophers: What Do They Believe?

by Luke Muehlhauser on December 8, 2010 in Resources

Last year, David Bourget and David Chalmers conducted an exercise in the sociology of philosophy, the largest survey of philosophers ever (3000+ respondents): the PhilPapers Surveys. Now that new results have been released, let’s look back at the findings.

First, it’s worth noting, as the editors do, that (1) the survey focuses mostly on Anglophone analytic philosophers, and (2) answer choices were often too brief for respondents to know how to answer, and that (3) though the response rate of 47% was pretty good, there is inevitably some selection bias, probably toward younger analytic philosophers. More survey design thoughts here.

Basic results

The results for some of the questions of wide interest include…

(note that ‘other’ includes answers like ‘I don’t know’)

Ethics: realism or anti-realism?
56.3% moral realism
27.7% moral anti-realism
15.8% other

Ethics: deontology, consequentialism, or virtue ethics?
32.3% other
25.8% deontology
23.6% consequentialism
18.1% virtue ethics

Abstract objects: Platonism or nominalism?
39.3% Platonism
37.7% nominalism
22.9% other

External world: idealism, skepticism, or non-skeptical realism?
81.6% non-skeptical realism
9.2% other
4.8% skepticism
4.2% idealism

Free will: compatibilism, libertarianism, or no free will?
59% compatibilism
14.9% other
13.7% libertarianism
12.2% no free will

Atheism or theism?
72.8% atheism
14.6% theism
12.5% other

Naturalism or non-naturalism?
49.8% naturalism
25.8% non-naturalism
24.2% other

Truth: correspondance, deflationary, or epistemic?
50.8% correspondance
24.8% deflationary
17.5% other
6.8% epistemic

The positions with the greatest degree of consensus in the whole survey are:

  • non-skeptical realism about the external world
  • scientific realism about theoretical entitites
  • atheism
  • belief in a priori knowledge
  • switching on the trolley problem (intervening so that 1 person dies instead of 5)

So… there are fewer philosophers who believe in a god than there are philosophers who would not intervene to save 4 lives. :)

Correlations

Most of the strongest correlations are unsurprising: naturalism is correlated with physicalism, moral realism is correlated with thinking aesthetic value is objective, theism is correlated with free-will libertarianism, etc.

As for geographic effects, Australasia is associated with consequentialism and the B-Theory of time. Canada is associated with free-will compatibilism and atheism. Europe is associated with moral non-cognitivism, aesthetic subjectivism, and scientific anti-realism. UK is associated with a belief in the a priori. USA is associated with deontological ethics.

As for gender effects, being female is most correlated with the epistemic theory of truth, with not switching on the trolley problem, and with rejecting a priori knowledge. Being male is most correlated with rejecting the epistemic theory of truth, with switching on the trolley problem, and with accepting a priori knowledge.

As for age effects, being young is correlated with Humeanism about the laws of nature, with the B-Theory of time, with accepting a priori knowledge, with physicalism, and with rejecting free-will libertarianism.

But things get really interesting when you start looking at subject areas.

For example, though theism is unpopular in philosophy in general, it is more popular among philosophers of religion (72.3%) than physicalism is among philosophers of mind (61.2%). Trent Dougherty sees this as a credence-boost for theism, since the experts on the subject of God tend to believe in him. But the obvious reply is that most people aren’t going to do philosophy of religion if they don’t believe in God. There doesn’t seem to be anything about philosophy of mind in general that would show a selection effect for physicalism, but obviously philosophy of religion will show a selection effect for theism.

But then, one would expect to see a strong selection for moral realism among those who do applied ethics, right? Why would you do applied ethics if you don’t believe in morality? Strangely, the rate of moral realism among those who specialize in applied ethics is identical to the rate of moral realism among all respondents: 56.8% vs. 56.3%! Similarly, those who specialize in aesthetic philosophy are about as likely to think aesthetic value is objective as are all respondents (44.7% vs. 41%).

Consider some other correlations with philosophical orientation:

  • Atheism is most correlated with not being a philosopher of religion, with not identifying with Plato or Aristotle or Kant or Leibniz, with analytic philosophy, with identifying with Hume and Quine, and with specializing in philosophy of mind, philosophy of language, philosophy of biology, and philosophy of cognitive science.
  • Theism is most correlated with being a philosopher of religion or specializing in medieval and renaissance philosophy, with identifying with Plato and Aristotle and Kant and Leibniz, with being a continental philosopher, and with not identifying with Hume and Quine.
  • Naturalism is most correlated with specializing in philosophy of cognitive science, philosophy of biology, philosophy of mind, and general philosophy of science, with identifying with Hume and Quine, with not identifying with Husserl, Plato, Moore, and Kant, and with not specializing in philosophy of religion, metaphysics, or continental philosophy.
  • Non-naturalism is most correlated with specializing in philosophy of religion and metaphysics, with not specializing in philosophy of cognitive science, philosophy of biology, or philosophy of mind, with not identifying with Hume and Quine, and with identifying with Husserl, Plato, Moore, and Kant.

Some strange ones I found:

  • Specializing in philosophy of action is correlated with theism, and the A-Theory of time.
  • Being accurate on the metasurvey was most correlated with believing there is a priori knowledge.
  • Political communitarianism correlated most strongly with theism, while political egalitarianism correlated most strongly with atheism.
  • Specializing in philosophy of mathematics is highly correlated with Platonism, but specializing in philosophy of probability is highly correlated with naturalism.

Other details

The survey also asked philosophers to predict the results of the survey, though there were fewer respondents for this “metasurvey.” Some results of the survey were especially surprising to philosophers themselves. These are the questions on the metasurvey with the largest mean error. In general…

  • Philosophers overestimated the rate of aesthetic subjectivism by 23.4%.
  • Philosophers underestimated the rate of acceptance for the analytic/synthetic distinction by 22.1%.
  • Philosophers overestimated the rate of empiricism by 19.4%.

There is also the list of respondents who let all their answers be viewed publicly, including Derek ParfitBrian Leiter, David Chalmers, David Papineau, and many others.

Those interested in this kind of thing may be interested in a similar survey of economists.

Previous post:

Next post:

{ 32 comments… read them below or add one }

Dima December 8, 2010 at 5:22 am

Thank you, Luke!
Very interesting post.

  (Quote)

Reginald Selkirk December 8, 2010 at 6:05 am

naturalism vs. non-naturalism vs. other – I would need descriptions to even understand those categories.

  (Quote)

Joel December 8, 2010 at 6:10 am

The fact that most astrologists believe in a realist conception of astrology (i.e. that giant masses of hot gases have a correlative relationship to some primates’ lives) is no evidence for the truth of astrology.

That said, Luke, are there any signs of trends? As you mentioned, there may be a difference in belief-sets when comparing older philosophers with the younger ones. I wonder if there may even be a trend towards atheism in philosophy of religion – this may be possible if we consider the fact that not all philosophers of religion have to specialize in the field (e.g. Mackie was mainly involved in metaphysics and ethics).

  (Quote)

Reginald Selkirk December 8, 2010 at 6:48 am

What do they mean by “skepticism”? Is that classical Pyrrhonism (“nothing can be known, not even this”), or the modern concept of scientific skepticism, in which believe is apportioned according to evidence? They are two very different concepts sharing one word.

  (Quote)

Luke Muehlhauser December 8, 2010 at 6:48 am

You’re welcome, Dima.

  (Quote)

Joseph December 8, 2010 at 7:12 am

This one:

“Theism is most correlated with being a philosopher of religion or specializing in medieval and renaissance philosophy, with identifying with Plato and Aristotle and Kant and Leibniz, with being a continental philosopher, and with not identifying with Hume and Quine.”

I’ve had personal experience on many christian forums, and I can say that to be true. Also, I found a fetish love for Aquinas, which your survey doesn’t mention.

  (Quote)

Nope December 8, 2010 at 7:32 am

An interesting result…

Time: A-theory or B-theory?
Other 542 / 931 (58.2%)
Accept or lean toward: B-theory 245 / 931 (26.3%)
Accept or lean toward: A-theory 144 / 931 (15.4%)

Certainly B-Theory is more popular, but what could these “others” be? I thought the majority of time theories fell under either A or B.

  (Quote)

Luke Muehlhauser December 8, 2010 at 8:36 am

‘other’ includes answers like ‘I don’t know’ or ‘Not sufficiently familiar with the subject.’

  (Quote)

Tito December 8, 2010 at 9:33 am

Thanks much for the analysis Luke, especially the surprises and correlations.

The link between phil of religion and theism is the most interesting to me and I appreciate the link to the theist trying to use that as evidence that the people closest to the details believe in god. Reminds me of your podcast 1 with Licona where you asserted that most NT scholars were believers therefore were not a good sample of authority for agreement on the minimal historical facts argument. This is a very key point in apologetics as if the minimal facts start loosing their authority the scaffolding holding up the foundation of literal interpretation comes down.

It would be interesting to measure the theistic believe of people in phil of religion across time, especially before choosing to specialize in the field. I suspect you are right that there would be an overwhelming tendency for theists to go into the subject and non-theists to avoid it. I also wonder what & of phil of religion have changed their theism/atheism view since becoming philosophers.

  (Quote)

Hermes December 8, 2010 at 11:15 am

Luke, on A-theory and B-theory, outside of Craig’s use of the KCA is the distinction one that shows up in other philosophical discussions? I would expect the answer would be no as it doesn’t seem to apply to many questions, also I don’t remember it being discussed in philosophy classes let alone any books I’ve read on my own. If time is discussed, it’s discussed as we as individuals commonly experience it.

  (Quote)

Jeff H December 8, 2010 at 11:32 am

Excellent stuff! Very interesting results. Just wanted to add a bit of a warning, however, in terms of the correlations. With a sample size this large, and with so many questions, it’s easy to find spurious correlations that are merely illusory. I would suspect that some of the ones that Luke mentioned as “strange” might fall into this Type I error.

Now, perhaps the authors corrected for the large number of correlations they conducted – I don’t know, since I haven’t read the paper. But if not, it’s important to take the results with a grain of salt. One can generate hypotheses and then see if the data fit them, but once you start going into “let’s test random things and look for any connection!”, beware.

  (Quote)

Nope1 December 8, 2010 at 11:49 am

Luke, on A-theory and B-theory, outside of Craig’s use of the KCA is the distinction one that shows up in other philosophical discussions?I would expect the answer would be no as it doesn’t seem to apply to many questions, also I don’t remember it being discussed in philosophy classes let alone any books I’ve read on my own.If time is discussed, it’s discussed as we as individuals commonly experience it.  

It comes up when discussing McTaggart’s paradox, but that’s about the only other place. And anywhere else it might be discussed would refer to it as the Tensed theory vs. the Tenseless theory. However, most philosophers seem to discuss it as a settled question due to the success of Einstein’s relativity. A-Theorists (or Tensers) usually seem to believe that the philosophical arguments against B-Theory are more compelling than relativity.

  (Quote)

Luke Muehlhauser December 8, 2010 at 12:25 pm

Philosophy of time is a small area in philosophy, just like philosophical zombies. The major figure that started off modern philosophical discussion of time is McTaggart. Like Nope1 says, it’s usually an argument between science (relativity, B-Theory) and philosophical arguments (A Theory).

  (Quote)

Hermes December 8, 2010 at 4:03 pm

Luke & Nope1, thanks for the clarification. That would explain why Other is so high 58%; a general lack of interest. I’m surprised that A-theory didn’t fare better, though, as theists would be more likely to pick A-theory while non-theists would be less likely to be motivated to pick either. The results aren’t as clear as I would have expected, though;

Time:A-theory 0.194
A-theory B-theory
theism
47.8% (45/94)

26.5% (25/94)
atheism
20.8% (89/427)

44.9% (192/427)
Response pairs: 578 p-value: < 0.001

Source: http://philpapers.org/surveys/linear_most_with.pl?A=main%3AGod%3Atheism

It would be interesting to know of the respondents what ones have taken a college level introduction to physics or any other physics classes. Still, it’s a good survey and they have to stop somewhere.

  (Quote)

MauricXe December 8, 2010 at 4:03 pm

The link between phil of religion and theism is the most interesting to me and I appreciate the link to the theist trying to use that as evidence that the people closest to the details believe in god. Reminds me of your podcast 1 with Licona where you asserted that most NT scholars were believers therefore were not a good sample of authority for agreement on the minimal historical facts argument. This is a very key point in apologetics as if the minimal facts start loosing their authority the scaffolding holding up the foundation of literal interpretation comes down.

On this note, do we have any statistics for how many NT scholars, that aren’t Christians, that accept the NT facts?

  (Quote)

Reginald Selkirk December 8, 2010 at 4:19 pm

On this note, do we have any statistics for how many NT scholars, that aren’t Christians, that accept the NT facts?

One of the two.

I’m joking, but my point is that there are not many New Testament scholars who aren’t Christians. Why would they be interested in that?

And what do you mean by “accept the facts”? Bart Ehrman for example, accepts the “fact” of Jesus’ existence, but not the “fact” of his resurrection.

  (Quote)

nope3 December 8, 2010 at 5:54 pm

‘other’ includes answers like ‘I don’t know’ or ‘Not sufficiently familiar with the subject.’  

Yeah, I looked at the public results for the A-Theory/B-Theory question and most people seemed to indicate “agnostic/undecided” or “insufficiently familiar with the issue.” One person even commented that she couldn’t recall the meaning of the labels. I also noticed that some people responded with “both” which is utterly perplexing – how does one accept both a-theory and b-theory? I’d be very interested to know.

  (Quote)

MauricXe December 8, 2010 at 5:59 pm

Good question.

The original quote I replied to said:

“most NT scholars were believers therefore were not a good sample of authority for agreement on the minimal historical facts argument”

I was wondering how many NT scholars, that aren’t believers, subscribe to the minimal historical facts argument.

  (Quote)

Silver Bullet December 8, 2010 at 8:27 pm

I can’t believe that the question of the existence of an external world is an issue that anybody thinks is worth spending even a nanosecond on.

This is not to say that it isn’t a provocative question, but it seems to be one that is pointless, even absurd, and for which, for all intents and purposes, there is only one practical answer.

So what am I missing?

  (Quote)

Landon Hedrick December 8, 2010 at 8:57 pm

Hermes and Nope,

The A-Theory vs. B-Theory distinction also plays a role in the metaphysics of persistence over time. There is a natural fit for A-Theory with endurantism and B-Theory with perdurantism.

As for those who say they accept both A and B, it may be that what they accept is a hybrid view. The “growing block” and “shrinking block” theories of time fit this general description.

  (Quote)

Tito December 8, 2010 at 10:22 pm

About scholars, Gary Habermas (the apologist) claims to have done an exhaustive survey of 1,400 papers since 1975 to assess the acceptance of different facts by NT scholars. Here’s a link to the best info I could find about it:

http://www.garyhabermas.com/articles/J_Study_Historical_Jesus_3-2_2005/J_Study_Historical_Jesus_3-2_2005.htm

I haven’t had time to read through it yet, but wanted to pass along the source, as suspect as it is ;) I don’t think he breaks down between believers and non-believers since it’s not a yes/no question considering there are shades of just about every belief between literal fundamentalism and liberal christian mysticism.

He does claim that 75% accept the empty tomb and 25% oppose it, as I read it. We know there are non-believing scholars like Robert Price and others so liberal they deny the facts used by Habermas and Craig (e.g. Crossan and others on the Jesus Seminar)

  (Quote)

Chip December 9, 2010 at 2:21 am

You totally missed Michael Jordan’s answers.

It turns out, Michael Jordan could not even conceive of zombies existing.

  (Quote)

Joe Snow December 9, 2010 at 8:33 am

About the 73% theism rate in PoR…

I suppose this could be viewed as an indication that theism is actually true (i.e. of those who study the issue at depth, the overwhelming majority conclude theism).

But what if this percentage is compared to a control population – e.g. people in general? The theism rate among the general public is somewhere around 80-90%, right? In that case, studying PoR could be seen to actually decrease the rate of theism.

Of course, it’s probably not completely accurate to compare the general public to PoR-bound students, but it’s something to think about…

  (Quote)

Luke Muehlhauser December 9, 2010 at 8:40 am

Joe Snow, good point.

  (Quote)

Lee A.P. December 9, 2010 at 9:54 am

What is “non-skeptical realism”?

  (Quote)

bossmanham December 9, 2010 at 9:29 pm

Philosophy of time is a small area in philosophy, just like philosophical zombies. The major figure that started off modern philosophical discussion of time is McTaggart. Like Nope1 says, it’s usually an argument between science (relativity, B-Theory) and philosophical arguments (A Theory).  

Of course that’s a little too simplistic, since science relies on philosophical presuppositions one brings into their study of the theories of time, and A-theorists are fairly comfortable in engaging the scientific evidence. Not being a little biased here, are we Luke? What would Yudkowsky say?

  (Quote)

wissam December 10, 2010 at 2:38 am

Isn’t Platonism in conflict with theistic activism? Isn’t conceptualism a bit more friendly to theistic activism?

  (Quote)

Luke Muehlhauser December 10, 2010 at 2:48 am

wissam,

As I recall, the survey presented the options as either leaning toward nominalism or toward Platonism. Perhaps theists think conceptualism is ‘closer to’ Platonism, I dunno. Or that theistic conceptualism, at least, is pretty close to Platonism.

  (Quote)

Kip December 11, 2010 at 12:40 pm

belief in a priori knowledge

I don’t understand this belief. Maybe it’s just a semantic misunderstanding I’m having with these philosophers?

It seems to be counter to a correspondence view of truth. If something is true in that it corresponds to reality, then the only way we can know if it corresponds to reality is to examine reality to see if it does.

The examples that I’ve seen to support “a priori knowledge” all come back to some sort of non-contingent “truths” (e.g. something that is “true by definition”). But, I don’t think these “truths” necessarily refer to things in the real world, so I’d hesitate to call them “knowledge”.

Every time this comes up, I’m reminded of a quote by Einstein:

“As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality.”

Luke: do you believe in “a priori knowledge”? If so, can you explain it to me? Am I misunderstanding something?

  (Quote)

Luke Muehlhauser December 11, 2010 at 2:32 pm

Kip,

That’s one of many questions on which I don’t even hesitate to guess until I’ve done more reading. :)

  (Quote)

Mike Gage March 17, 2011 at 7:57 am

If you don’t already think Alistair Norcross is awesome, check out his answer to the metaphysical zombies question:

Zombies: inconceivable, conceivable but not metaphysically possible, or metaphysically possible?

Accept: metaphysically possible
Actual. See the Republican National Convention

  (Quote)

Keith Augustine November 16, 2011 at 2:44 pm

I flip-flopped on “Metaphilosophy: naturalism or non-naturalism” a few times because I wasn’t sure what “naturalism” meant as a “metaphilosophy.” If the labels had been prefaced with metaphysics, philosophy of religion, metaethics, or epistemology I would’ve known exactly what it meant (something different in each case except the first two, and maybe even slightly different there). I finally settled on the guess that it meant naturalized epistemology, though it then should have been prefaced as “Epistemology: naturalism or non-naturalism” IMO.

I finally found confirmation of my lunch here Metaphilosophy:

Metaphilosophical naturalism – which is the focus in what follows – asserts a strong continuity between philosophy and science. A common construal of that continuity runs thus. Philosophical problems are in one way or another ‘tractable through the methods of the empirical sciences’ (Naturalism, Introduction). Now, within metaphilosophical naturalism, one can distinguish empirical philosophers from experimental philosophers (Prinz 2008). Empirical philosophers enlist science to answer, or to help answer, philosophical problems. Experimental philosophers (or ‘experimentalists’) themselves do science, or do so in collaboration with scientists. Let us start with empirical philosophy.

Quine is an empirical philosopher in his approach to metaphysics and even more so in his approach to epistemology. Quine presents and urges his epistemology thus: ‘The stimulation of his sensory receptors is all the evidence anybody has had to go on, ultimately, in arriving at his picture of the world. Why not just see how this construction really proceeds? Why not settle for psychology?’ (Quine 1977: 75). Such naturalistic epistemology – in Quine’s own formulation, ‘naturalized epistemology’ – has been extended to moral epistemology. ‘A naturalized moral epistemology is simply a naturalized epistemology that concerns itself with moral knowledge’

  (Quote)

Leave a Comment

{ 1 trackback }