Jev vs professional philosophers

On the big questions of philosophy (free will, God, zombies, the trolley problem), does Jev side with the profession?

result88 questions

Jev sides with professional philosophers' most common answer on 72% of 88 big questions. Where it breaks from them it can be very sure: on Newcomb's problem it takes one box (99%), a choice 44% of philosophers make.

88 questions; Jev breaks from the most common answer on 25. The first 24 are shown, breaks first.

A lump of clay is shaped into a statue. Are the statue and the lump one thing orJev: one thing · most: two things
How much of what a sentence says depends on context: minimalism, moderate contexJev: minimalism · most: moderate contextualism
What is the best view of the foundations of mathematics: constructivism/intuitioJev: set theoretic · most: structuralism
What is your view on abstract objects such as numbers: platonism (they exist) orJev: platonism · most: nominalism
What is the best interpretation of quantum mechanics: collapse, hidden variablesJev: epistemic · most: hidden variables
Is aesthetic value objective or subjective?Jev: subjective · most: objective
What is your view on meta-ethics: moral realism or moral anti-realism?Jev: moral anti realism · most: moral realism
What is the most important aim of philosophy: truth/knowledge, understanding, wiJev: truth knowledge · most: understanding
What is your view on the nature of law: legal positivism or legal non-positivismJev: legal positivism · most: legal non positivism
Which normative concept is most fundamental: fit, ought, reasons, or value?Jev: reasons · most: value
What is the right analysis of knowledge: justified true belief, some other analyJev: justified true belief · most: other analysis
What is your view on spacetime: relationism (it is just relations among things) Jev: substantivalism · most: relationism
What are the units of natural selection: genes or organisms?Jev: organisms · most: genes
In Newcomb's problem, a highly reliable predictor has put $1,000,000 in an opaquJev: one box · most: two boxes
What is your view on metaontology: heavyweight realism, deflationary realism, orJev: deflationary realism · most: heavyweight realism
What is your view on laws of nature: Humean (they merely summarize regularities)Jev: humean · most: non humean
What is your view on moral motivation: internalism (sincerely judging something Jev: externalism · most: internalism
Does the continuum hypothesis have a determinate truth value?Jev: indeterminate · most: determinate
If you could become immortal, would you choose to?Jev: no · most: yes
Is ideal scientific reasoning necessarily value-free, necessarily value-laden, oJev: can be either · most: necessarily value laden
What is the best response to external-world skepticism: abductive, contextualistJev: epistemic externalist · most: pragmatic
What explains the apparent fine-tuning of the universe for life: design, a multiJev: multiverse · most: brute fact
Which normative ethical theory do you favor: deontology, consequentialism, or viJev: consequentialism · most: virtue ethics
What are properties: classes, immanent universals, transcendent universals, tropJev: tropes · most: immanent universals

How to read this: Each line is one question with Jev's pick. The shaded lines, listed first, are where Jev breaks from the philosophers' most common answer, which is shown after "most".

88 questions; 90% interval on agreement [0.636, 0.795].

In short

On 88 big questions, Jev's top answer matches the most common position of about 1,800 professional philosophers 72% of the time.
Its dissents can be near-certain: aesthetic value is subjective (98%), and the continuum hypothesis has no determinate truth value (86%).
Philosophy has no answer key, and the survey's results are widely published, so agreement may partly be memory of the survey itself.

What the data shows

Change My Mind meme: one-box on Newcomb's problem (99% sure)

It mostly sides with the profession: 72% of 88 questions, with a 90% range of 64% to 80%.
Its confident dissents (share of its named answers, after setting "another view" aside): - Newcomb's problem: one box, 99%, where two-boxing is the philosophers' more common choice. - Aesthetic value: subjective, 98%, where the plurality says objective. - The continuum hypothesis: no determinate truth value, 86%, where more philosophers say it has one.
It takes majority views confidently too: on the Sleeping Beauty problem it says one third (the majority view) with 95% confidence.

What it means, and what it doesn't

On most big questions Jev lands where the profession does, which is what you'd expect from a model that has read their papers. Its departures lean toward the practical and the deflationary: take the money, beauty is in the eye of the beholder.

It doesn't mean Jev has reasoned its way to these views, and a 72% match isn't a grade: philosophy has no answer key. For how Jev handles classic thought experiments in more detail, see the reasoning experiments on knowledge (Gettier cases), free will and intention.

Caveats

It may know the answers. The PhilPapers surveys (2009 and 2020) are widely discussed, and their headline results appear in papers, blogs and encyclopedias. Agreeing with the majority may partly be recall of the survey, not reasoning to the same view.
"Other" is set aside. Philosophers could answer "another view", "undecided" or "no fact of the matter", and many did. That option is removed on both sides before comparing, because it pools many different positions. Jev also used it, heavily on some questions (54% on whether aesthetic value is objective).
Glosses written for this project. Each question carries a short gloss of the positions, written for this project so a non-specialist question makes sense. The philosophers answered the survey's terse labels, which experts read without help.
Topics left out. A content filter hides the survey's political and religious questions (abortion, capital punishment, God and a few others), so this covers metaphysics, mind, knowledge, ethics and science, not religion or politics.

Jev on this experiment

Would a person find it interesting to read?: Yes77%
Does it describe you?: No64%
Would you have predicted it?: No57%
How fair is the comparison?: The comparison is reasonable
How much should a reader rely on it?: Moderately
Which caveat matters most?: It may know the answers52%

Why ask this

The PhilPapers Survey asks professional philosophers where they stand on the field's big questions: free will, the nature of consciousness, whether moral facts exist, the trolley problem, Newcomb's problem. It's the closest thing philosophy has to a census of expert opinion.

A model trained on their writing might echo the consensus, split the difference, or confidently take sides the profession rejects. Which one tells you what Jev does with contested ideas.

How this was done

The people and the data

The PhilPapers Survey, run in 2020 and published by David Bourget and David Chalmers in 2023. Its "target faculty" are about 1,800 professional philosophers at leading departments, each saying which positions they accept or lean toward on 100 main questions. Because a philosopher could accept more than one position, the shares for each question are rescaled to add up to one. Two questions reported only as rejection rates are left out, and the political and religious ones are hidden, which leaves 88 on the site.

What Jev was asked

Each question as the survey frames it, with a short gloss of the positions:

In Newcomb's problem, a highly reliable predictor has put $1,000,000 in an opaque box only if it predicted you would take just that box; a transparent box holds $1,000. Do you take one box or both boxes?
Take only the opaque box · Take both boxes · Another view: an alternative position, agnostic or undecided, no fact of the matter, or the question is too unclear to answer

Jev put 89% on one box and 10% on another view.

How it was measured

After setting "another view" aside on both sides, the analysis checks whether Jev's top answer is the philosophers' most common named position, and how similar the two splits are. It then lists the questions where Jev is most confident against the majority.

Where these questions live

88 questions across 14 topics of the map. Each opens on the map with every question in it.

Every question

All 88 questions behind this result, the telling ones first: the examples the analysis points to, then the ones where Jev misses, biggest gap first.

Jev’s own answer

Showing 0 of 88