Blog Coordinator

Knobe's X-Phi Page

X-Phi Grad Programs

« Do people draw a distinction between reality and appearance in the case of pain? | Main | Registration Open: Intuitions, Experiments and Philosophy »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Geoff Holtzman

1. Suppose scientists figure out the exact state of the universe during the big bang, and figure out all the laws of physics as well. They put this information into a computer, and the computer perfectly predicts everything that has ever happened. In other words, they prove that everything that happens, has to happen exactly that way because of the laws of physics and everything that's come before. In this case, is a person free to choose whether or not to murder someone?

2. Suppose you drive to the local baseball stadium with some friends, and try to buy tickets at the door. There are 7 of you, but there are only 6 tickets left. You can either drive everyone to a nearby bar, which will be a lot less fun than being at the game, or 6 of you can go in, and 1 of you can take the bus home and miss the game entirely. Is it most fair for everyone to go to the bar?

3. Suppose a mad scientist takes out your brain, and puts it in your best friend’s head. During the same operation, the scientist takes out your friend’s brain, and puts it in your head. Now your body has your friend’s brain, and your friend’s body has your brain. Your heroic mother storms into the room to save you, but not your friend, who she believes got you into this mess. Is the person with your body still you, her son?

4. Suppose neuroscienists are able to identify every part and every connection in the human brain. Working with a team of computer scientists, they then build a robot that has a complete electronic replica of the human brain. Could this robot experience love?

5. Suppose that all you know about Einstein is that he developed the Theory of Relativity. But suppose it turns out that Einstein actually stole the idea from some guy named Moynahan, who nobody has ever heard of. In this case, when you use the name ‘Einstein,’ are you actually referring to Moynahan?

6. Suppose you hear the sound of your cell phone, so you reach in your pocket and answer the call. Your landlord is on the line, but you realize later that your ringer was off, and the sound you heard was actually someone else’s phone. When you heard that other person’s phone ring and mistook it as your own, did you actually know someone was calling you?

7. Suppose you meet a man from the future who knows everything there is to know about science. He tells you that he doesn’t like apples, and says that though he has never eaten one, he has figured out what apples taste like just by studying the relevant science. Could he know what apples taste like without ever having eaten one?

8. Suppose scientists are able to use stem cells to grow lungs that breathe without being connected to a body. They then grow a heart that pumps without being connected to a body. If they can do all this, can they create a brain that thinks without being connected to a body?

9. Suppose a runaway train is coming down a track, and is certain to kill five workmen who can't get out of the way. You're standing next to the controls and can switch the train to the other track, but if you flip the switch, one man working on that track is sure to die. Should you flip the switch?


This is very exciting research. Awhile back I started a project to challenge the philosophical community to (a) come to grips with the possibility that many of our beliefs correlate with neurological properties and (b) to volunteer for studies that will shed light on that possibility. I have yet to be in a position to begin the research myself, but I have had the chance to present the hypothesis. I get rather mixed reviews.

During one poster presentation, someone said to me, "So you're one of the philosophers who wants to kill philosophy?"

Another said, "That is scary. Soon enough people will be saying that certian philosophical beliefs correlate with medical or psychological conditions" (paraphrase).

While I choose to remain agnostic about the specific implications of this project, I find your work very enlightening. I wonder what type of response you have received from your work.


Neil Levy

Hi Geoff,

Can I ask you for your evidence for this claim:

A great deal of literature suggests that moral disapprobation covaries with negative emotions, even when those emotions are not the result of moral considerations.

Geoff Holtzman

Hi Nick and Neil,

I'll respond to your posts more fully after I get some sleep. In the meantime, here are just a couple citations in response to your question, Neil. The second one focuses specifically on the effects of emotions that arise independently of moral considerations.

Greene, J.D., Sommerville, R.B., Nystrom, L.E., Darley, J.M., & Cohen, J.D. (2001). An fMRI investigation of emotional engagement in moral Judgment. Science, 293, 2105-2108.

Wheatley, T., & Haidt, J. (2005). Hypnotically induced disgust makes moral judgments more severe. Psychological Science, 16, 780-784.

Disc personality testing

Personality plays a big role in the enhancement of your career. Thanks a lot for sharing this information. I am looking forward to see more updates about this matter.

Jonathan Jenkins Ichikawa

This looks interesting.

Why might this sort of data show that philosophy is 'intrinsically subjective'?

Neil Levy

Geoff, the effect sizes in Wheatley and Haidt are tiny. With the exception of one case, the induced disgust does not alter the judgment, it merely intensifies it, to a small extent. The one case in which they succeeded in generating a possible confabulated judgment carries all the weight, but even in that case the effect was only on a minority. I used to put a lot of weight on this study myself. I am now convinced that this was a mistake.


Very interesting work. I've read some earlier work that claims to have detected a genetic basis to the Big Five. This would make sense from an evolutionary psych perspective. If that is the case, then philosophical beliefs may indeed be related, in some way not yet understood, to genetics. This may challenge some prevailing views about how beliefs develop and spread in communities. Ordinarily we attribute belief acquisition to enculturation - the individual is raised to hold certain beliefs consistent with the culture of their upbringing. This is a nurture argument. But nature, on the level of predispositions, may also play a role. It would be difficult to disentangle these relationships, as they are probably reciprocal. In some cultures, some personality traits may be more valued (and therefore encouraged) then others, thereby indirectly supporting certain cultural beliefs by acting on natural predispositions from below, as well as enculturation from above. Fascinating ideas, but still very speculative.

Neil Levy

I should credit Jonathan McGuire with calling my attention to the problems with Wheatley and Haidt. I was reminded when I also recalled his work on Greene et al. I'm afraid that Jonathan's work on RTs in this study, together with Kahane and Shackel's work on it, has not left much of the original hypothesis. Indeed, I thought Greene himself conceded that - not that he was wrong, but that the 2001 paper is not strong evidence for the hypothesis. In any case, the hypothesis is not the one you want. You want evidence that emotions that are not caused by moral considerations covaries with negative judgments, not that emotions cause negative judgments.

Sorry for the double post.

Jason S

Hi Geoff,

The significant correlations flagged above, are they positive or negative? (I am guessing some of the correlations are positive and some negative.)


Geoff Holtzman

Jonathan: That's a good question, and a contentious point that most of the real philosophical legwork in the article is committed to defending. I know that many, and probably most readers will reject my claims of subjectivity, even if they accept the psychological findings as valid. It's a hard point to defend, but the objections I've heard are have not been well-defensed either. I think maybe an analogy is the fastest way to illustrate why I use the word "subjective."

Suppose it were discovered that the more neurotic a doctor is, the more likely she is to recommend prophylactic surgery, or that the more agreeable she is, the more likely she is to dispense painkillers to anyone who asks. These are obvious cases of subjective bias in medicine, ones so serious that they seem to present real ethical concerns.

Now, this does not give us reason to claim that medicine is a subjective discipline, because empirical results can, in theory, reveal objective truth. But philosophy focuses on specifically those questions that are not available to empirical study-- once philosophical questions do become (arguably) objectively determinable, they're co-opted by other disciplines (psychology, physics, etc.). To the extent that medical opinion can be measured against the objective metrics of empirical results, medical opinions may be subjective without medicine being subjective. Without any impersonal check on personal opinion, I think the subjectivity of practicing philosophers can rightly be called the subjectivity of philosophy.

Again, very open to and interested in feedback and dissent on this point (as long as it's not from journal referees!).

Geoff Holtzman

Hi Neil,

First, I should mention that the present article focuses on subjectivity and does not delve into either etiological model described in my original post-- I included those models to get valuable feedback for when l do try to publish something in that regard. So I can concede that that theory may not be ready for primetime.

But in regards to my claims of subjectivity, I think what I want is evidence that emotions (or personality traits) that are not *entirely* caused by moral (or philosophical) considerations covary with negative (or philosophical) judgments, and this is what I found. Moral differences are caused by emotional differences that are not the result of moral considerations. Therefore, identifying the "best" moral view is tantamount to identifying the "best" personality. This is not to say that there is no right moral view, just that its identification is a matter of subjective values. The same strikes me as true for the other findings in this study.

Geoff Holtzman


If what you meant is that the studies I *cite* don't show what I want them to, I'll agree with you. That's why I had to take matters into my own hands! But of course, those studies were invaluable to my own work, and probably a lot more important in other ways.

Eric Schwitzgebel

Cool. Did you happen to collect Area of Specialization info? I'm curious how ethicists vs others self-rate on some of the subscales that seem to have a moral dimension.

Jonathan Jenkins Ichikawa

Let's assume for the purpose of argument that philosophical claims are not available to empirical study. (I'll register dissent, but assume it for now.) Now how do we get from data like this to the subjectivity of philosophy?

The verificationists thought that only empirically verifiable claims were candidates for objective facts. Of course, if that's your view, then given the assumption, philosophy can't be dealing in objective facts -- regardless of any psychological data about what predicts what.

If you're not a verificationist, then you think there's no very straightforward connection between objectivity and empirical confirmation. You think that there are objective medical facts, whether or not there is a test that could determine them with observations. You might think there are objective arithmetical facts, even though arithmetic disputes are not resolved against the objective metrics fo empirical results. If this is so, then the consistency of systematic bias among medical judgment with the objectivity of medicine doesn't really have anything to do with empirical verifiability. So I don't see any pressure here against the (fairly natural) view that there are objective philosophical facts, even though various contingencies of our psychologies can influence our judgments about them.

So I still don't see why data of this sort would push someone towards denying the objectivity of philosophy.

Geoff Holtzman


That's a really interesting question. I'm sort of curious if you would find an effect, or if the potential correlation between personality factors and ethical views would wash out any overall effect from ethicists, since they would run the gamut in terms of views. This, in turn, brings up the question of whether people in a particular AOS have different belief tendencies within their own field (or other fields) than people who do not specialize in that field. Unfortunately, I didn't collect AOS info, so I can't answer any of these questions.

Geoff Holtzman


You make a good point, and in my paper and in these posts, I've been careful to remain neutral towards the question of whether or not there are objective philosophical facts. What I am questioning is the ability of philosophical methods to discover these facts, if they do exist.

There is a fact of the matter about what medical treatment will lead to a better outcome in a given case, and there might also be a fact of the matter about whether a robot could feel love. If over the years, doctors make partially subjective recommendations of either surgery of physical therapy for a certain type of slipped disc (I don't know if this is true), eventually treatment outcomes will show which intervention is generally the appropriate one. But if philosophers make partially subjective recommendations about whether or not a robot could feel love (which I think I have shown), how can we ever see which answer is appropriate?

Geoff Holtzman


Yes, the correlations go both ways. Is there one you are particularly interested in? I may be able to update the chart in my original post when I get the time and am back at my computer.

Marcus Arvan


Intersting stuff (I wish I had thought of running a study like this!). However, I worry a bit about the significance thresholds you used. As I understand it, in exploratory studies Bonferroni corrections are more or less standard -- but only 2 of your findings pass a Bonferroni corrected value. Anyway, very cool stuff, but that worries me a bit.

Geoff Holtzman

Hi Marcus,

I think your worries are reasonable, and as I mentioned, these statistics are sort of tricky, and there's not really consensus about how to interpret them. I have found that philosophers tend to worry more about the numbers than psychologists do. Bonferroni corrections provide the most conservative and least powerful test, and many psychologists feel that Bonferroni corrections are way to strong. The most inclusive test would be to calculate the odds of finding 4/9 or 10/45 results significant to p < .05-- and the odds that all of these findings were due to chance is vanishing. A middle of the road, and more popular solution, is to include only those specific correlations that fit into overall models that are significant to p < .05.

I chose to take that middle route in determining my main results, but give a nod to both other methods in the article. Interestingly, the least conservative test finds effects for 6 of 9 questions, the most conservative test for 2 of 9 questions, and the in-between test for 2 of 9 questions.

One of the most important things to take away from this isn't the exact results, but the overall, fairly broad-reaching pattern, which suggests that *some* philosophical beliefs are subject to personality effects-- even if not all those marked with an * are. The other important thing to keep in mind is that hypothesized findings are generally held to be exempt from post-hoc tests. In this case, that means the connections between neuroticism and the trolley problem, love reductionism, and the knowledge argument. Because I did not have room in the article (or this post) to delve meaningfully into specific hypotheses though, I treated personality effects for the trolley problem, and for the knowledge argument, as insignificant. So the 4 of 9 effects actually *exclude* two of the three hypothesized and discovered effect.

Daniel Levine

Does the full paper consider the possibility of the cause underlying the correlation going the other way?

E.g., if I come to be convinced that thinking involves my whole body and not just my brain, might I become more agreeable and open?

Geoff Holtzman


Yes, the paper does consider that. Given that personality develops before philosophical belief and remains relatively stable throughout life, and that folk intuitions about questions they've likely never thought about before also correlate with personality, this seem probably untrue-- or at least true in addition to my claim, and not instead of it. Or, it could be that both personality and philosophy are commonly subject to some other influence, but this would still suggest that subjective factors influence belief.

While proving the order of causation might technically be an empirical, it seems to me it would be an intractable one. Short of a longitudinal study, I'm not sure how you would prove anything. But how do you run a longitudinal study looking at lay people who become philosophers, let alone do that without influencing your subject pool and running into self-selection biases? Furthermore, how would you even analyze that data to see what causes what? So I address your point briefly, but it is a tough question to answer definitively.

Jason S

Hi Geoff,

I would be interested in knowing the direction (and strength!) of the correlation for all the questions, but in particular, I would be most interested in knowing more about the nature of the correlation between those correlations you hypothesized to exist--that is, the correlation between trolley and neuroticism and also the correlations between love and neuroticism and love and conscientious.

Geoff Holtzman


In regards to the trolley problem, it's been found that induced negative emotion correlates with moral condemnation, and that good and bad judgments correlate to a certain extent (that's an oversimplification, it's sort of complicated) with differential activation of working memory and emotional areas, respectively, in the case of trolley problems specifically. So the idea was, if emotional induction and emotional brain activity would be expected to predict condemnation in the trolley problem, so would lifelong trait neuroticism. And that's what I found: The more neurotic, the more negative the moral judgement.

In the case of reducing love, it's known that neuroticism and conscientiousness correlate with reports of the intensity and passion of love. Given that my broad thesis is that subjective experience influences philosophical belief, I expected that people who experience a phenomenon more strongly than others would be more likely to rate them as "above and beyond" physical characterization. That's also what I found: Neuroticism and conscientiousness (and according to studies, therefore intensity/passion of love experience) correlated negatively with belief in reductionism about love.

Işık Sarıhan

An interesting and inspiring study, thanks. Though I would be happier to see it not being presented as a preliminary evidence for the strong conclusion that philosophy is subjective, but rather as a way to spot the psychological factors that are obstacles to be dealt with on our way to objectively agreeable conclusions.

As some people around here would be interested, I should mention an interesting survey that this work reminded me of, "Intuitions and Introspections about Imagery: The Role of Imagery Experience in Shaping an Investigator’s Theoretical Views", by Reisberg, Pearson and Kosslyn from an 2003 issue of 'Applied Cognitive Psychology', where they claim to have found that "[the philosophers and scientists] who experienced their own visual imagery as vivid and picture-like recall being more sympathetic in 1980 to the view that, in general, images are picture-like."

The comments to this entry are closed.

3QD Prize 2012: Wesley Buckwalter