Shaun Nichols

Shaun Nichols (University of Arizona) is presenting “The Rise of Compatibilism: A Case Study in the Quantitative History of Philosophy." The two invited commentators are Eric Schwitzgebel (University of California--Riverside), and Kelby Mason (Rutgers University--New Brunswick).

Download shaun_nichols.pdf

Download comments_on_nichols.pdf

Comments

[Paul Torek]

There is another hypothesis which would explain the empty cell in the table, without reference to motivation. We could posit that these philophers were forced to reject at least one member of an inconsistent triad, each of which was otherwise suggested by available evidence. The triad being determinism, incompatibilism, and free will/moral responsibility.

One could of course reject more than one of the triad, but by hypothesis, this would have been foolish (in that cognitive context) if it were not necessary, and it was not necessary.

[Cihan Baran]

What I don't understand is that the majority of compatibilists are not, in Kadri Vihvelin's words, Hobartian compatibilists - the idea that the moral responsibility is incompatible with indeterminism.

Most compatibilists think that if you have the right sort of indeterminism, this would still be compatible with moral responsibility.

So the names that occupy the upper left cell may also be occupying (in this sense I have outlined above) the upper right cell - it's just that they are not explicit about it.

[Kip Werking]

This paper was thrilling to read. Like Nichols & Knobe's previous empirical work on free will (which I think is the most important paper written about free will in, say, the last ten years), it suggests a new empirical method for undermining compatibilism.

Unfortunately, unlike the previous work, this new research is of the form "if X is true, then we would expect Y, Y is true, therefore X is supported." But this method is obviously very weak. Just consider: if compatibilism is motivationally motivated, then I expect the sun to rise today. The sun is rising, therefore my thesis is supported.

I don't think Nichols' work here is as problematic as the sun example. The work is fascinating, but I'm sure yet how much it shows. For example, I wonder whether people are just more likely to be determinists on independent grounds, and I wonder whether people are likely to be compatibilists on independent groups, and both of these trends might go a long way towards explaining the emptiness of that cell (but cannot, I suspect, entirely explain it). Perhaps this just points in a good direction for future research.

Here's an experiment I'd like to see: are compatibilists more likely than incompatibilists to engage in wishful thinking? Considering how empirically shaky libertarianism is, I doubt this. It is true that incompatibilists include hard determinists, but these represent a small fraction of the population (of both philosophers and especially the folk; they seem disproportionately represented in the early modern period). As someone who tends to be more of a hard determinist, I am more interested in exploring the differences between realists and non-realists about free will.

Finally, I have to say: Locke and Hobbes were not compatibilists. And I do not mean this in the shocking "Kant was not a Kantian" way that it sounds. Consider first Locke:

"If this be so, (as I imagine it is,) I leave it to be considered, whether it may not help to put an end to that long agitated, and, I think, unreasonable, because unintelligible question, viz. WHETHER MAN'S WILL BE FREE OR NO? For if I mistake not, it follows from what I have said, that the question itself is altogether improper; and it is as insignificant to ask whether man's WILL be free, as to ask whether his sleep be swift, or his virtue square: liberty being as little applicable to the will, as swiftness of motion is to sleep, or squareness to virtue. Every one would laugh at the absurdity of such a question as either of these: because it is obvious that the modifications of motion belong not to sleep, nor the difference Of figure to virtue; and when any one well considers it, I think he will as plainly perceive that liberty, which is but a power, belongs only to AGENTS, and cannot be an attribute or modification of the will, which is also but a power."

Now consider Hobbes:

"And therefore if a man should talk to me of a Round Quadrangle; or Accidents Of Bread In Cheese; or Immaterial Substances; or of A Free Subject; A Free Will; or any Free, but free from being hindred by opposition, I should not say he were in an Errour; but that his words were without meaning; that is to say, Absurd."

These quotes are straight from the Project Gutenburg copies of Leviathan and An Essay Concerning Humane Understanding, respectively. They both sound more like Richard Double than Daniel Dennett. I think you'll also find strong non-realist sympathies, if not explicit endorsements, in other compatibilist writings, such as Hume's writing on God and liberty, early Mill, Dennett's hints of revisionism, Fischer's denial of the ability to do otherwise, and so on. This is in stark contrast to libertarians, who were always very clear about believing in "free will" or "freedom of the will"---because that term sounds metaphysically ambitious and (only) their metaphysics can deliver on that promise.

Why do people regard Hobbes and Locke as compatibilists? One answer is: because they so clearly thought that other valuable things were consistent with determinism. Locke seems to think that "free agents" are real but "free will" is nonsense. Hobbes denied the existence of free will but asserted that natural liberty or freedom is consistent with determinism. There is usually the further claim "what's inconsistent with determinism isn't valuable and what is consistent with it is valuable", *therefore* the given philosopher must be a compatibilist.

But neither of these moves gets you to compatibilism. Compatibilism is the doctrine that free will, or freedom of the will, is compatible with determinism. It is not the claim that liberty or free agents are consistent with determinism---at least not when these philosophers explicitly distinguish between free will and these other things. It is also not the claim that the only kinds of freedom worth wanting are consistent with determinism. It's the doctrine that free will, itself, is consistent with determinism---whether it is valuable or not, whether we can be free in other senses or not. Unless I am somehow wrong about understanding Hobbes and Locke this way, I really hope people will stop pretending that there were more famous compatibilists, and fewer non-realists, than there really were.

[Eddy Nahmias]

Kip, imagine a philosopher, Dell, who has "ambitious" metaphysical aims and seeks to show how immaterial minds exist. Dell is eager to prove this since he believes that the existence of minds is incompatible with materialism, that real minds simply cannot exist in a fully material world.

Now, another philosopher, Matt, comes along and says the idea of immaterial minds is incoherent or highly implausible at best. Matt thinks Dell's conception of minds is mistaken and aims to explain how material minds are possible (indeed, actual).

Sure, you could then say that Matt is a skeptic about minds if you wanted. But that would be silly. Matt thinks minds exist. They just aren't what Dell says they are.

Crucially, Matt thinks that his conception of minds captures everything (or almost everything) that matters. His view of minds may revise certain peripheral theoretical ideas about minds that have seeped into ordinary thinking (perhaps from religion or philosophy)--namely, that minds are non-physical--but Matt's view is meant to capture ordinary mental experiences and ordinary beliefs about such experiences.

I am suggesting that Hobbes and Locke and most other compatibilists think of free will like Matt thinks of minds. Sure, the crazy libertarian notion of self-creation doesn't exist any more than Dell's notion of minds. But what matters does exist, call it what you will (free will or free action or liberty or whatever--the powers of agents that make them autonomous agents subject to judgments of responsibility...).

You can't just assert that someone doesn't count as a compatibilist unless he or she believes that determinism is consistent with a type of free will that has the properties incompatibilists assign to it.

[Kip Werking]

Eddy,

Thanks for responding to my comment. I'm a big fan of your work.

I understand the point your making. It's a revisionist point (and it might be a lot better if compatibilists such as yourself and Fischer, amongst others, would call yourself revisionists if that is actually what you are).

And, if Locke and Hobbes showed any sign of being revisionists, then I would agree with you. Consider your analogy. Just as Matt still wants to use the label "mind", Locke and Hobbes would have to keep using the label "free will".

But they *don't* do that. They *jettison* the label free will and take up some other label, whether than is freedom, free agents, natural freedom, natural liberty, or some other thing. They're explicit in asserting that "free will" is non-sense and refers to nothing real. Hobbes and Locke are *not* revisionists about free will. This is the crucial difference between Matt and Locke/Hobbes that prevents me from agreeing with you.

If Locke said "we used to think free will was this non-sensical thing but actually it is this weaker, sensical thing that captures more of what we used to think, and I am going to therefore keep using the label free will", then I would be more sympathetic. [Note that I still think this revisionist move is problematic; it attempts to "change the subject while insisting it isn't" as Richard Double said.] But Locke doesn't even do this much.

Of course, most contemporary philosophers agree with you in categorizing Hobbes and Locke as compatibilists. That's why Nichols categorized them that way in his paper. I just suspect, however immodestly, that they're all wrong.

[Eddy Nahmias]

This is a lengthy response to Shaun (I may come back to Kip's response later):

I find Shaun Nichols’ methodology fascinating and certainly worth a try in the free will debate and others. Like Eric Scwitzgebel, I think the “psychology of philosophy” is an under-explored field, probably because, as Shaun would say, we have motivational reasons not to explore it—we philosophers prefer to think we are able to transcend our idiosyncratic psychological tendencies to discover unadulterated truths (and to be fair, our methods are designed in part to allow just this sort of movement towards objectivity). The free will debate is a particularly ripe field for such investigation, filled as it is with appeals to intuitions and with the “dialectical stalemates” that ensue when these intuitions butt heads. Having said this, I think that the data Shaun presents may be better interpreted in a different way than he interprets them.

Notice that almost all the philosophers Shaun lists in the determinist column are not only determinists but also *naturalists* (i.e., philosophers who were impressed by the methods and discoveries of modern science—Galileo, Newton, Bacon, etc.—and who believed that all processes and events, including human decision-making, are natural events governed by natural laws). OK, being motivated to make my point, I just pulled the claims in the last sentence out of my … assimilation of the history of philosophy. But they do seem “truthy” to me—certainly, Hobbes, Hume, D’Holbach (and LaMettrie were he on the list) fit this description of naturalism, and Locke and Spinoza (as a monist) seem to kinda fit. I have no idea about the others, and Leibniz and Pascal may be problematic given their theism (but remember some theists are happy to be compatibilists, *motivated* to find a common response to the problem of determinism and the problem of divine foreknowledge or divine preordination).

Now, at least for the naturalists in the group comprising Shaun’s left column, the story might go like this:
1) First they come to believe naturalism is true, perhaps being impressed by the rise of modern science or for other reasons.
2) Then, they come to believe that naturalism entails determinism. They believe this based on the Newtonian view of natural laws. I think I’m right that the very idea of *indeterministic* natural laws or indeterministic causation was considered implausible, if not simply incoherent, until the advent of quantum physics. Indeed, these philosophers might further believe that the falsity of determinism would require some breach of naturalism (perhaps a miracle or something).
3) So, given this commitment to naturalism, these philosophers would then realize that if we are to be free and responsible, free will and moral responsibility must be compatible with naturalism *and hence* compatible with determinism.
4) At this point, some of the naturalists (determinists) embraced compatibilism, perhaps motivated to preserve freedom and responsibility. Other philosophers embraced hard determinism (hard naturalism), but in some cases they seemed to be arguing that naturalism rules out the sort of religious libertarian view the philosophers on their right (on Shaun’s graph) accepted. But note that they do not thereby give up on the concepts of free and responsible action (Spinoza offers a nice explanation of how increased self-knowledge allows for increased freedom and responsibility of a valuable sort). Certainly, those who are labeled compatibilists are rejecting the libertarian conception of free will without rejecting the idea that we are free and responsible agents (see my previous post).

Meanwhile, the non-naturalists do not accept that science and natural law can explain everything—at a minimum it can’t explain certain human actions (notice how Descartes, also impressed by modern science, applies its mechanistic explanations to everything, including all animal behavior, except for the human mind). Were these thinkers driven to reject naturalism in order to preserve human freedom and responsibility? Possibly. But it’s not that they rejected *determinism* per se in order to preserve freedom and responsibility. If they had been presented with a naturalistic *indeterministic* alternative (e.g., quantum theory), I suspect they would have rejected that alternative as well (i.e., they would not have followed the path Robert Kane carves out).

Now, we can see why, *given the period Shaun focuses on*, the top right square remains empty. If one was a determinist, it was most likely because one was a naturalist. One’s commitment to naturalism preceded one’s views about the compatibility question. If one answered the compatibility question as a compatibilist, there would be no reason to embrace indeterminism to preserve free will and every reason not to embrace it since embracing naturalism brought determinism along with it.

Meanwhile, if one rejected naturalism, one was rejecting the determinism that was thought to go with it. So, there was no reason to be a compatibilist about *determinism* and free will given one’s prior rejection of determinism.

But along comes the 20th century and the connection between naturalism and determinism was severed by quantum theory—it is now considered at least coherent (if not likely) that everything could be understood in terms of natural laws and science and yet *not* be deterministically caused, entirely predictable in principle, etc. So, now there is no reason to embrace determinism if one has embraced naturalism. James’ term “soft determinist” is antiquated (as is “hard determinist”) because there is no reason for a compatibilist to be a determinist (and many contemporary incompatibilists are skeptics about free will who think indeterminism does not help).

This allows for a prediction: now that naturalism and indeterminism are compatible, there should be compatibilists who accept indeterminism.

Well, I, for one, am a compatibilist who also thinks indeterminism is quite possibly true (even at the macro level, e.g., human decision-making), and I can think of at least one other who would join me in the top right box (indeed, it’s nice to share logical space with David Lewis). What the advent of quantum mechanics has made clear is that we need to separate clearly two questions which, I propose, were conflated during the modern period (and are still conflated by some):
1) the compatibility of free will/responsibility with determinism (or with indeterminism)
2) the compatibility of free will/responsibility with a naturalistic world view—for example (but not the only example), the causal closure of the physical.

[Neil Levy]

Like Eddy, I was struck by the fact that the combination of compatibilism plus belief in indeterminism is much more common, post quantum mechanics, than Shaun thinks. Indeed (unlike Eddy) I think it's the dominant view: most people I talk to think the Copenhagen interpretation is probably true, that compatibilism is true, and that we are free (ie, that free will is compatible with determinism and with some degree of indeterminism). Now, of course we can wheel in a motivational explanation for this combination of views, too. So it's no great problem for Shaun.

Here are two problems, one small, one larger. First, libertarianism seems at least as vulnerable to the wishful thinking charge as compatibilism. Libertarians think that they can tell what the nature of physical reality is, just by introspecting! So the question is: what explains the contrastive fact that wishful thinking is more often expressed in compatibilism, in the history of philosophy, than libertarianism? Shaun holds beliefs about determinism fixed, but there seems no better justification for holding than fixed than for holding questions about, say, compatibility, fixed.

Like I say, that's a small problem. I can imagine several (prima facie plausible) responses. Here's a slightly bigger one. Among scientists (I think; proper quantitative methodology should be used to confirm) the combination of determinism and incompatibilism is a common one, indeed, the default. So why are scientists less vulnerable to this particular bit of wishful thinking (especially given that it's outside the area of their expertise)? My guess is that motivational explanations get a grip here, but that the explanations refer more to what is regarded as hard-nosed, serious, and so on, within a field - and that precisely the same factors play a MUCH bigger role in the psychology of philosophers in the free will debate than Shaun acknowledges.

BTW - and in full acknowledgement that the plural of anecdote is not data - I hold the following combination of beliefs: free will is compatible with determinism and with whatever degree of near determinism is likely to be true of the actual world; free will is metaphysically impossible. So there.

[Roman Altshuler]

Oddly, Kant has not received his own discussion in the comments so far, and I would like to chime in with the obligatory "Kant was not an indeterminist" claim. This is not, really, a matter of interpretation; I don't know how much clearer Kant, or anyone else, could have been than this:

"all the actions of the human being in appearance are determined in accord with the order of nature by his empirical character and the other cooperating causes; and if we could investigate all the appearances of his power of choice down to their basis, then there would be no human action that we could not predict with certainty." (Critique of Pure Reason, A549-550)

Kant was also pretty clearly a naturalist. What he did not accept, however, was the idea that naturalism could be internally justified. Justifying the legitimacy of the sciences, in turn, required an extra level of analysis, which opened a place for freedom. (On the libertarian conception, unless of course that conception is taken to involve alternate possibilities.)

This is old news, and Kant is a particularly odd case (although, one can similarly ask whether Leibniz is really a compatibilist, in anything like the sense common today). But the question of interpretation raises some issues:

First, as Kip Werking points out (and as Shaun Nichols mentions in the paper), there are controversies about which categories a number of these figures fall into. Many do not fall neatly into the current divisions. This is because these divisions did not yet exist in their current form.

Second, because the major representatives of modern philosophy frequently do not fall neatly into contemporary categories, decisions on where to place them will largely be biased by contemporary views. Interpreters are generally not "pure" historians of philosophy--they too have been trained within a climate in which particular divisions, and ways of applying these divisions (c.f. the points made above by Werking and Nahmias re: which features are sufficient for, and which exclude someone, from being a compatibilist) are the norm.

Third, to extend on this last point--which does get into the psychology of philosophy, or the psychology of the history of philosophy--philosophers with compatibilist tendencies are, I'm guessing, more likely to interpret their favorite historical figures as compatibilists. Similarly for incompatibilists. Etc.

That is, this sort of approach, while certainly interesting, has to contend with the bias built into the commitments of the interpreters. Nichols mentions that "Historians of philosophy notoriously disagree about how to interpret the philosophers." But this isn't the only problem--the problem is that the very same psychological factors that structure one's commitments on the free will issue are also likely to structure one's interpretation of various historical figures' commitments. Taking the dominant reading doesn't resolve the problem: it is possible, for example, that indeterminists are disproportionately drawn to Kant scholarship, or compatibilists to Hume scholarship.

Is it possible to compensate for this bias? Maybe, but I don't see how one could do that without performing the same sort of study on the historians of philosophy themselves.

[Derk Pereboom]

To the “competing explanation” themes raised by Eddy and Neil’s comments, I’d like to add this: for the seventeenth and eighteenth century philosophers, the only serious motivation for indeterminism derives from incompatibilist intuitions about free will; and determinism is strongly motivated by physical theory, and also, to a certain degree, by theology. It would seem that as a result of this alone, one would expect a low incidence of compatibilists who are indeterminists. Now that (since the development of quantum physics in the1920’s and 30’s) we have a motivation apart from incompatibilist intuitions about free will to accept indeterminism, as Eddy and Neil note we find compatibilists who are indeterminists, and compatibilists who are at least open to its being true.

(Motivation distinct from incompatibilist intuitions about free will was present in the ancient world. Lucretius, the 1st century BCE Epicurean writes: … when the atoms move straight down through the void by their own weight, they deflect a bit in space at a quite uncertain time and in uncertain places, just enough that you could say that their motion has changed. But if they were not in the habit of swerving, they would fall straight down through the depths of the void, like drops of rain, and no collision would occur, nor would any blow be produced among the atoms. In that case, nature would never have produced anything.” (Lucretius, On the Nature of Things, 2.216.) But the Epicureans were incompatibilists about free will, so they don’t provide us with examples of compatibilist indeterminists. No motivation for indeterminism of this sort, to my knowledge, is present in the 20 seventeenth and eighteenth century philosophers Shaun lists.)

Another potentially significant factor is Hume’s contention in the Treatise of Human Nature that moral responsibility requires determinism. The Treatise dates from 1739/40, and the influence of this view has the potential of reducing the incidence of compatibilism together with indetermimism for the rest of the period

Is Descartes an incompatibilist indeterminist? Scott Ragland (University of St. Louis) has recently published a number of splendid articles on this issue, and he contends that, on balance, this reading of Descartes is correct. But as he acknowledges, there are recalcitrant texts. I think that one of the strongest indications of determinism is in his letter of October 6, 1645 to Princess Elisabeth:

...it seems to me that all the reasons that prove the existence of God, and that he is the first and immutable cause of all the effects that do not depend upon the free decision of men, likewise prove in the same way that he is also the cause of all those that do depend on it. For one could not demonstrate that God exists save by considering him as a being sovereignly perfect; and God could not be sovereignly perfect if something could happen in the world that did not come entirely from him... philosophy alone suffices to give us the knowledge that the least thought cannot enter the mind of man if God had not wished and willed from all eternity that it enter therein. (AT IV 313-4/B162)

On Kant, the A549-550 passage Roman quotes expresses his determinism about the phenomenal, and it is uncontroversial among commentators that Kant is a determinist about the phenomenal. But Kant counsels us to believe that as intelligible (noumenal) agents we are free in a way he describes as follows: “By freedom in the cosmological sense… I understand the power of beginning a state of itself (von selbst) – the causality of which does not in turn stand under another cause determining it in time in accordance with the law of nature” (A533/B561). I agree with those commentators who contend that Kant has in mind an incompatibilist conception of freedom. But others, for example Ralf Meerbote and Hud Hudson, interpret this notion of freedom as compatibilist. However, Kant does not think that we can establish that we as noumenal agents are free in this sense; “we have not been trying to establish the reality of freedom.” In fact, he claims that “we have not even tried to prove the possibility of freedom.” Rather, he hopes only to show that “nature at least does not conflict with causality through freedom” (A558/B586). His claim is, in effect, that while we should for practical reasons believe that we are free in this sense, we can’t show that we really are, or that it is metaphysically possible that we are. We can only show that our conception of ourselves as free in this sense is not internally contradictory, and that it does not contradict anything else we can establish. So while Kant in a sense aspires to incompatibilism and indeterminism, he does not think that this position can be shown to be true, or even to be metaphysically possible.

[Joe]

Neither Descartes nor Kant was an indeterminist.

[Kip Werking]

Joe,

Your statement about Descartes and Kant seems pretty confident. Is there no room left to doubt their belief in determinism?

Consider Descartes. It's uncontroversial that Descartes believed that the mind intervened in the causal order of the world through the Pineal Gland and thereby controlled the body. Perhaps someone could interpret that in a fully deterministic way. But it sounds a lot like agent causation to me.

Regarding Kant: Kant seems to have been a determinist about the phenomenal world. But was he a determinist about the noumenal world? Isn't the noumenal world the one in which in hinged his hopes for free will?

It seems that in both of these cases, the philosopher had a largely mechanical or deterministic understanding of the actual world, but was also careful to make an exception for the human mind or soul. They carved out a niche in which each person might escape from this deterministic world, and it was only in virtue of this escape that the person might have, or hope to have, free will.

If this sketch of their views is right, or largely right, then it seems to me awkward at best to describe them as determinists. In any case, I would like to hear you elaborate on why you think they are determinists.

[Joe]

Kip,

Part of the problem is that there is no univocal use of the expression 'free will.' You seem to have one, narrow definition in mind whereas I tend to use the expression in a broader way, similar to the way indicated in Eddy's post: "free will or free action or liberty or whatever--the powers of agents that make them autonomous agents subject to judgments of responsibility..." Clearly, in this broad sense, Locke is a compatibilist.

The Locke quote you cite -- which is one of my favorites in all of philosophy -- is, it seems to me, taken completely out of context. The basic point is that the will is a power and freedom is a power and, since it doesn't make sense to say that a power has a power, it doesn't make sense to talk about 'free will.'

This is NOT a criticism of free will in either your narrow sense or my broad sense. It strikes me as a comment on the expression 'free will,' not a criticism of any conception of free will. I don't know of anyone who ever held the view that to have free will is to have a will that is free.

I'll try to read the rest of the posts more carefully and make a more substantive contribution soon.

[Justin Coates]

In Gideon Yaffe's "Liberty Worth the Name: Locke on Free Agency" there is an interesting passage (pgs 21-27) that's devoted to the quote that Joe and Kip are discussing.

Yaffe suggests that this passage is motivated by a desire to "remedy the mistake of homunculizing the will - a mistake that the existence of the phrase "freedom of will" encourages - Locke thinks that we must always keep in mind that it is not a person's will that is free or unfree, but the person herself" (Yaffe, 24).

To avoid such a homunculization, Yaffe points out that Locke abandons the question "is the will free?" in favor of "are agents free to will?" (26).

I think that Locke answers the latter question affirmatively even if determinism were true, so it seems to me (for whatever that's worth) that Nichols is not mistaken when he puts Locke in the company that he does.

[Joe]

Kip,

My last comment was to a previous post of yours on Locke – which I must have been writing as you were posting your comments about Descartes and Kant.

As for Descartes, I wrote a paper on his views of freedom and determinism: “Descartes on Spontaneity, Indifference and Alternatives,” New Essays on the Rationalists, eds. Gennaro and Huenemann (Oxford UP, 1999). I’d be happy to send you a copy.

I agree with what Derk has to say on this subject, e.g., that Descartes’s commitment to determinism follows for theological reasons. For instance, Descartes believed that God is “the total cause of everything” (CSMK: 272), which (given some assumptions about God’s nature – immutability, etc.) seems to entail that determinism is true.

You write: “It's uncontroversial that Descartes believed that the mind intervened in the causal order of the world through the Pineal Gland and thereby controlled the body. Perhaps someone could interpret that in a fully deterministic way. But it sounds a lot like agent causation to me.”

First, it doesn’t follow from your last claim that Descartes was not a compatibilist, nor that he was not a determinist. Second, the dualist story only complicates matters. Are we to suppose there is some wiggle room in our mental life that can never be manifested in the physical world? That is hardly a helpful kind of freedom.

As for Kant, I think it depends on whether you give what my colleague Harry Silverstein calls the ‘two-worlds’ or ‘one-world’ interpretation of Kant. I accept the latter. It isn’t as if the noumenal realm is a different place, distinct from our phenomenological descriptions of it. And, according to Kant, we can – in fact we must – describe that world as determined. The world is determined yet freedom – full-blown agent causation – is possible. Ergo, compatibilism is true.

Obviously, these issues are debatable. Derk made a good case for an alternative conception of Kant above. My flippant comment was a reaction to seeing Descartes and Kant placed in, what I understand to be, the wrong boxes. I have the terrible habit of responding to unsupported, dogmatic claims with opposing unsupported, dogmatic claims.

I have to get back to writing my commentary on Derk’s paper for next week’s conference.

[shaun nichols]

Let me begin by thanking everyone for reading my outrageous paper and offering such helpful comments! I’m particularly grateful to Eric for such a delightful commentary. And Thomas & Eddy – thanks for all the excellent work putting together the OPC!

I have a lot to say in response, so I’ll make a few separate posts. In this post, I’ll focus on the motivational account of compatibilism, and in a later post I’ll take up issues about history, method, and the psychology of philosophy.

One important initial comment is that for this discussion, I will focus centrally on free will rather than on moral responsibility. This is because although I think that incompatibilism is intuitive both for FW & MR, I am tempted to a kind of prescriptive compatibilism about moral responsibility. But I am not attracted to that kind of view about choice.

I’m not sure whether Derk was coming to my assistance in his first paragraph, but I hope so! Because I was planning to respond to Eddy & Neil partly by saying pretty much what Derk said in that paragraph. (More generally, I’d like to be on the same side as Derk. Hard incompatibilism is close to my heart.)
Let me first rehearse how I view the dialectic. The incompatibilist claims that the intuitive, commonsense view is that we have a kind of choice that is incompatible with determinism. The compatibilist denies that this is the intuitive view of choice. He proceeds to point out that many great philosophers have maintained that indeterminism is not part of the natural view of choice. (Let’s leave aside interpretations of historical figures as revisionists, at least for the moment.) This generates a worry for the incompatibilist. How can I explain the responses of those great philosophers if I am right that the natural, intuitive view of choice is indeterminist? Answer: the move to compatibilism is driven by motivational factors. This is the answer that a lot of incompatibilists are inclined to give, but to make it more than an ad hominem, I wanted to investigate a prediction made by the motivational account. The prediction, again, is that determinists will be more likely than indeterminists to be compatibilists. (Note: Eddy, Neil & Derk are right to point out that this doesn’t work in the contemporary scene. More on that in a later post.)
The pattern that emerged from looking at the history of modern fit the prediction very well indeed. Every position was well represented except indeterminism + compatibilism.

Now, I admit that there are alternative interpretations of the pattern. I consider one alternative in the paper. But others were suggested in the comments by Eric, Paul, Eddy, and Neil. In evaluating alternative interpretations, I want to stress two features of my prediction:
i. It is a directional prediction – it predicts a distinctive pattern of responses.
ii. The prediction is generated by a standing theory. That is, many incompatibilists are antecedently inclined to a motivational story for why people endorse compatiblism. That story generates a certain prediction, and it is this prediction that is up for the test. So the prediction is not a post-hoc explanation. (Indeed, when I generated the prediction, I really did not know how it would turn out for many of the philosophers in the major 20. My knowledge of the history of philosophy doesn’t go that deep.)
I think that both of these features weigh in favor of the motivational account: the prediction was not post-hoc, and it predicted a specific pattern which was borne out by the history. Other explanations that fail to meet these conditions are, other things equal, less good. So let’s look at the other explanations.

I agree with Paul that philosophers who become determinists face an inconsistent triad: determinism; incompatibilism; free will. They are forced to revise their beliefs. But then we have to ask “what makes them revise it in favor of compatibilist free will?” Here the motivational hypothesis has a clear story to tell. So Paul, I’d want more detail about why you predict the specific pattern.

Eric’s alternative hypothesis is that the directional explanation I offer – that determinists are motivated to be compatibilists – isn’t directly shown by the table. For the table itself is actually symmetrical, so one can equally well interpret the table as saying that only incompatibilists are motivated to be indeterminists. (Ron Mallon also pointed this out to me, but I swore him to secrecy. Too late to try that with Eric!) This move clearly meets the condition of explaining the pattern. Eric just offers a different directional explanation that conforms to the very same table. This reveals an inherent limitation of this kind of correlational analyses (about which I’ll say a bit more in the next post). However, without more detail this seems like a post-hoc explanation rather than one that is inspired by a standing theory. So I think that my motivational hypothesis is preferable.

Eddy’s proposal, that naturalistic philosophers will be more inclined to determinism, seems plausible to me. And it might explain why compatibilists are determinists and nonnaturalists are indeterminists. But what I don’t see is how it explains why there is good representation of naturalists who are incompatibilists. Again, the motivational hypothesis neatly accommodates this aspect of the table.

Neil adds some ammunition that might be used to fortify Eddy’s position. He says that the lure of being “hard-nosed” would explain why people embrace hard determinism both in science and philosophy. So it’s actually motivationally attractive to be a hard determinist. I admit this never occurred to me, and I’m not sure how to evaluate whether it explains the phenomenon. In my own case, I remember that when I was first introduced to the problem my response was something like “This is terrible – we don’t have free will. That’s really depressing.” It didn’t seem to me like being hard-nosed was much compensation for the depressing news. To make my anecdotes plural, let me add one more on the topic. A psychologist who defends free-will eliminativism told me recently that he had never even heard of compatibilism until philosophers started criticizing his work. I suspect this is really the standard view in the sciences – compatibilism about free will doesn’t even occur to most scientists who are innocent of philosophy.

I’ll try to tackle quantum mechanics later.

Thanks again for all the comments!
Shaun

[Kelby Mason]

Thanks to Shaun for his paper, Eric for his comments and everyone in the virtual audience for their comments so far. You may be wondering where the second set of comments are. Apparently, not everyone read the formal comments at last year's OPC--shocking, but true--so I've left my comments for now. I'll keep them shortish, to make it even likelier that people might skim them, at least.

Like Eric, I'm rather more interested in Shaun's methodology and background research program than in the nuts and bolts that have been occupying the commenters so far. If possible, I'd like to get some more discussion of that as well.

Eric calls Shaun's project "psychology of philosophy". That's right in one sense, but it gives a rather misleading impression about the most interesting parts of Shaun's project. Here's the bits of Shaun's paper that jumped out at me:

1) Shaun's attempt to give a causal history of a certain pattern in philosophy--i.e. to give a partial causal explanation for why certain philosophers at certain periods came to hold certain views

2) The invocation of "external factors" as part of that causal history--now, what do I mean by "external factors"? Here's a first pass: they're (among the) causal reasons for an agent's belief, but the agent wouldn't endorse them as a justifying reason for holding that belief.

3) The use of empirical methods to test (1)

Now, in this instance, Shaun has invoked an external factor which is psychological--namely, motivated cognition. But Shaun could just as well have invoked non-psychological external factors, e.g. economic or social factors, and his paper would still have been just as striking, and as unlike much else in mainstream history of philosophy. (Although he would probably have had to use a different case study).

So, although his use of psychology *is* interesting, I don't think it's the most important part of the paper. (Sorry, Eric! And maybe Shaun himself would disagree?) Rather, what's really important is (1), the attempt to do causal history of philosophy, and from that (2) and (3) naturally follow.

I myself like Nietzsche's word for this sort of history: genealogy. But we could just as well call it "naturalistic history of philosophy", "empirical history of philosophy" or "causal history of philosophy".

Such genealogies are absolutely standard elsewhere in the historiography of ideas. In the one area of historiography that I know a (very small) bit about, the historiography of science, there was actually a heated dispute through the latter half of the twentieth century about "external" versus "internal" factors. As I read the literature, the externalists largely won--that is, it's now generally considered kosher to invoke external factors when you're doing history of science. Depending on which journals you read, you might even suppose that external factors are positively preferred.

By contrast, causal histories of *philosophy* are exceedingly thin on the ground. Anyone should feel free (please!) to point me in the direction of the voluminous and rigorous literature that I've missed--but, unless I have missed such a literature, it seems to me that the vast majority of history of philosophy is what we can call "close reading" or "textual analysis".

Such histories are concerned with who thought what, and what justification they had. Textual analysis either isn't concerned with the actual *causes* for someone's holding a belief, or tacitly assumes that the causes are internal.

Now, as we've seen in the comments so far, genealogy and textual analysis can happily coexist. I'd go further, and submit that they need one another. Shaun's coding system, for instance, is only as good as the encyclopedias he consulted, and they rely on textual analysis. Several of the comments so far show how deeper textual analysis might change Shaun's data. But I suspect that textual analysis can benefit from genealogy too, for genealogy can reveal patterns in the historical data that might otherwise be missed.

But nobody likes an irenist, so let me end on rather on an ironic and polemical note. The problem with textual analysis, Shaun's paper seems to suggest, is that it's history of philosophy without the history.

What Shaun offers instead is history of philosophy without the philosophy. And I, for one, approve.

[Randy Clarke]

Suppose we understand determinism in a standard way: A complete statement of the laws of nature, together with a description of the total state of the world at any given time, entails every truth about how the world is at any other time. Then I don't see that Descartes is a determinist, for I don't see that he takes there to be any laws of nature applicable to the mind, including its decisions.

Derk and Joe suggest that he is nevertheless a theological determinist. A different kind of determinism, no?

[Eddy Nahmias]

Shaun, I don't understand why your hypothesis explains incompatibilist determinists any better than my hypothesis that some naturalists (who are therefore determinists) are incompatibilists. Can you explain?
I would argue that the "hard determinists" take that view because they think that the "strong" concept of free will invoked by many philosophers and theologians is impossible given a naturalistic world view. I don't know any of these thinkers very well, though if they are like Spinoza or like many modern "hard incompatibilists," the difference between them and the compatibilist will often look like a terminological dispute (this is related to Kip's point). They all agree that the libertarian conception of free will (self-creator, etc.) does not exist. But the ones we call compatibilists are more likely than the ones we call hard determinists to say that this conception of free will is not the right one--it is not the one tied to ordinary usage or beliefs and/or it is not the one tied to what we care about, such as practices of moral responsibility. So, my own motivational thesis is that the motivation to be a naturalist comes first and then they try to sort out free will and responsibility within that commitment, usually making sense of those ideas in a way that denies the need for the "silly libertarian view." But notice that people like Spinoza and Derk Pereboom do not think it's the end of the world because we can't have the sort of free will libertarians demand.
My own motivational hypothesis would predict that non-naturalists are motivated to take that view because of their theological commitments or their view that we can't be special (no free will, morality, meaning) unless we are unlike the rest of nature. That's why we don't see non-naturalist determinists (again, I'm just asserting this claim but I'd like a historian to tell us if it's true).
You are right that most scientists are ignorant of the compatibilist position. I think that is because they think the ordinary view is that free will exists and that naturalism is false, and they assume there is a conflict between the two views. Hence, science (naturalism) shows free will does not exist. These scientists (and it's not all of them) don't understand compatibilism about free will and *naturalism* (doesn't science show there is no soul, no "real" mind, it shows free will is an illusion?). I think most scientists don't understand that the philosophical debates about free will are not about the compatibility of naturalism and free will but about determinism, a thesis which is consistent with naturalism and with non-naturalism (notice Descartes could have been, perhaps was, a dualist determinist). And importantly, naturalism is consistent with indeterminism.

[Joe]

Randy,

What is a law of nature? It is a universal generalization that has some modal import -- or something like that. If, as Descartes suggests, God is the "total cause of everything," then I find it hard to believe that there aren't necessary universal generalizations connecting propositions about the state of the world at any instant with propositions about the state of the world at any other instant.

Such laws might take this form:

(L) Given the world is such-and-such at this time, it will be so-and-so at that time.

Given other assumptions that Descartes held -- God's essential immutability, God's eternal foreknowledge, predestination, etc. -- it seems that Descartes would hold that there was a complete set of propositions of the form (L) that were necessarily true and that linked propositions about the past with propositions about the future, etc.

You note that Descartes's dualism complicates this matter and I agree. I wouldn't say that the argument above is a knockdown argument either. But as I said in a previous post it would be an odd theory to suppose that there is some wiggle room in our mental life that can never be manifested in the physical world. It can't be that human actions result in physical events that do not conform to laws of form (L), for then God would not be the total cause of everything. Humans would be the cause of some things. What do we make of this?

The best bet, I think, is to suppose that Descartes held a compatibilist view similar to Davidson's anomalous monism. Some commentators reject this interpretation because they think that compatibilism is incoherent. But that is like saying that Descartes is not a dualist because that position is incoherent, also. Even if incompatibilism is true, the most we should assume is that Descartes held another incoherent view. Yet it is still the view that fits best with the text, I think.

I'm sensitive to Roman's comments above that "the major representatives of modern philosophy frequently do not fall neatly into contemporary categories." Nonetheless I think that the case for Descartes's determinism is stronger than the case for his indeterminism.

[Joe]

[I've been preparing the section below for the last hour or so but couldn't help responding to Randy's post before posting it. Sorry for the dual, lengthy posts!]

I apologize in advance for commenting on Shaun's paper without knowing much about experimental philosophy. I had dinner recently with Jonathan Weinberg and he convinced me that my ignorance was indeed lamentable but I have yet to have time to correct the flaw.

Like Eric, Kelby, Eddy, and others I think that the “psychology of philosophy” is important, interesting, and worthwhile and, thus, I’m supportive of Shaun’s project. Though I have more than a few complaints about the paper.

Let me begin with a few criticisms of the science of Shaun’s paper. First, the methodology that Shaun uses to collect his data is bad. Deciding what philosophers believe based on encyclopedia entries is, in at least some cases, only slightly better than picking their positions out of a hat. Leibniz, Hume, and perhaps Spinoza are relatively clear-cut but the views of the other Magnificent Seven on this matter are, as previous postings suggest, a matter of great debate. In the case of Kant and Descartes, for instance, if you look at a variety of entries, you’ll get a range of answers. Furthermore, the fact that Descartes gets placed in his cell because Richard Taylor said so – in an article entitled “Determinism,” no less – is just plain absurd. This is the kind of thing that will prevent the project from getting any serious attention from historians, which is a sham since it is an interesting and important project.

And there is simply no reason to collect the data in this way. The main point of the data is to show that there are no philosophers from the modern era in the top right box, and this is a point about which there is no doubt. You could just poll all of the folks who do history of modern and ask them, “Were there any indeterminist compatibilists in the modern era?” The overwhelming answer would be “No.” The collection of the data is seriously flawed but that fact is relatively insignificant as far as Shaun’s thesis goes since any reputable method of collecting the data would yield similar results. But why, Shaun, do you want to leave yourself open to irrelevant criticisms?

Second, I don’t see how Shaun can jump from data about the modern era to a general conclusion like (MD): “Determinists will be more likely than indeterminists to be compatibilists” (7). There are particular facts about that era of history that apply to it only, which suggests, I think, a better explanation of the data than the one that Shaun offers. (Personally, I like the explanation that Derk gives in the first paragraph of his comments above; Eric’s explanation, in section (iii) of his comments, is also plausible.) Furthermore, the thesis is complicated by the fact that it is demonstratively false when applied to contemporary, post-quantum theory philosophers, as Neil and Eddy note. Like Eddy and David Lewis, I am an indeterminist, given quantum theory. Thus, determinism cannot play a role in why I’m a compatibilist. In fact, I’ll go one step further (though not as far as Neil) and note that I am a near agnostic when it comes to the free will thesis. I believe that I have free will but I’m aware that I have no good reason to think so. The only thesis that I feel confident about is the thesis of compatibilism but even this is based on the mere fact that I don’t think that the arguments for incompatibilism are very good. It is not as if I have an argument for compatibilism, nor am I sure what such an argument would look like. Since I think that all arguments for free will denialism presuppose incompatibilism (or theses which lead to incompatibilism), I don’t see any reason to deny the free will thesis. But it isn’t as if I worry that it might turn out to be false. Given the work of Derk and Saul Smilansky, especially, I think things would be fine either way.

In any event, I don’t know of ANY contemporary philosophers who would classify themselves as soft determinists and I know of only one philosopher who falls loosely under the label of ‘hard determinist’ (Ted Honderich). Mind you, I think that I know about as much about the views of contemporary philosophers on this subject as anyone.

As I’ve noted before (“Freedom and Determinism: A Framework,” Freedom and Determinism, eds. Campbell et. al.), the main areas in the free will debate are no longer the ones noted by William James. Rather, they are incompatibilism, compatibilism (represented entirely, I think, by indeterminists or folks who are agnostic about determinism), and successor views (Kane’s term for views that are successors to hard determinism, including pessimism, free will denialism, hard incompatibilism, restrictivism, and whatever you’d like to call van Inwagen’s mystery view).

Again, the only determinist working in free will that I know of is Honderich and he classifies himself as neither a compatibilist nor an incompatibilist (“After Compatibilism and Incompatibilism,” Freedom and Determinism”). Thus, a more accurate thesis than (MD), applied to contemporary philosophers, would be: Determinists are likely to be neither compatibilists nor incompatibilists.

I’d be interested in knowing if I am wrong about this, e.g., if anyone knows of ANY contemporary philosophers who would call themselves ‘determinists’ and take a stand, either way, on the compatibilism/incompatibilism debate.

To summarize, I think that Shaun’s methodology is flawed, his argument for (MD) is fallacious, and the thesis itself is demonstratively false, at least when applied to contemporary philosophers. Nonetheless, the project is fascinating and important and I encourage him to stick with it since we all have something to learn from his approach!

[Randy Clarke]

I seem to recall seeing attributed to Descartes the view that, in matter, quantity of motion is conserved but direction can be altered without violating that conservation principle. That would leave something physical undetermined by the physical. I confess I don't know if that is really his view.

[Joe]

That sounds interesting, and plausible. I should confess that the Descartes paper I mentioned above was published in 1999, so its been awhile since I looked at these issues in detail. Since then there's been a lot of good work -- like the Ragland papers that Derk mentions -- which I need to read more carefully. I'll let you know when and if I change my mind!

[shaun nichols]

I’m flattered by the continued discussion here. Sorry I’m so slow! In this post I want to discuss two problems that I didn’t get to in the previous post.

The first problem, raised by Eddy and Neil (but it also comes up in Cihan and Derk’s posts), is that many many philosophers now accept both compatibilism and indeterminist physics. Hence my table doesn’t represent the current climate. I mention the problem of indeterminist physics in a footnote in the paper. What I say there is:
“the incompatibilist can offer a plausible explanation for why the philosophical geography changes with indeterminist physics. According to the incompatibilist, the reason people are indeterminists in the early modern period is precisely because they believe in free will. With quantum mechanics, a different reason for indeterminism emerges and this complicates the philosophical space.”
Derk’s first paragraph makes a related claim (for a different point), and I think that it does help to parry the blow somewhat. As an incompatibilist, I’m trying to explain why many great philosophers in the modern era were compatibilists. Those philosophers did, I suggest, start with a notion of choice that is incompatibilist and indeterminist. Why then did some of my faves endorse compatibilism? Because they were motivated by their commitment to determinism to do so. When indeterminist physics came along, there came a reason to believe indeterminism that was disconnected from choice. As a result, the table can no longer be restricted to the domain of choice.

As I say, I think this parries the blow somewhat, but not entirely. Even if this defense is right, I’m convinced by the comments from Eddy, Neil, Derk and Joe that I need to change the way I frame my prediction. I need to say up front that the prediction is not supposed to hold after the advent of indeterminist physics.

Now let me turn to the other serious related difficulty. Although I think Derk’s first paragraph helps me with the problem of indeterminist physics, he aims to offer it as a competing explanation of the pattern I report. The idea is that we can explain the empty cell as a product of two facts: modern philosophers were indeterminists because of incompatibist considerations and modern philosophers were determinists because of naturalistic or theological considerations.
My first stab at a reply is as follows. It’s true that modern philosophers believed in indeterminism only because they thought of choice as indeterminist. But it was open to them to think that, as a matter of fact, choices are indeterminist, but this is irrelevant to what makes them free choices. That is, they might have thought that indeterminism is true of choice, but inessential to free choice. To offer a naïve example, suppose that it turns out (as it might) that only earth has intelligent life, then it’s a fact that all choices occur in our solar system. But that doesn’t mean that it’s essential to being a free choice that the choice happens in our solar system. It’s a fact that is uninteresting for the metaphysics of choice. Similarly then, one might hold that every choice happens to be indeterministic, but this just isn’t an important fact about free choice. It was open to compatibilists to say this – choice happens to be indeterministic but that is irrelevant to free will. But apparently that’s not what any of them said.
I offer this reply with some trepidation given the vastly superior expertise of the people who have been blogging. But I want to agree with the claim that when modern philosophers were indeterminists it was because they thought of choice as indeterminist, but at the same time I want to maintain that this did not necessitate that they be incompatibilists about free choice.

[Joe]

This clears up a lot, Shaun. Thanks!

But I still like Derk's explanation better than the one you offer above. I think that what you're missing in Derk's response is that, in the Modern period, determinism was the default position, the one that philosophers began with. Some philosophers abandoned it in light of considerations of incompatibilism, which is evidenced by the prevalence of arguments similar to the Consequence Argument in the writings of these philosophers. (Another reason why I think that Descartes was a compatibilist is that he considers such an argument and then rejects it.)

On your explanation, the order is reversed: “Those philosophers did, I suggest, start with a notion of choice that is incompatibilist and indeterminist. Why then did some of my faves endorse compatibilism? Because they were motivated by their commitment to determinism to do so.” The problem is that I don’t see any evidence to suggest this. Nor is there much textual support that “they might have thought that indeterminism is true of choice, but inessential to free choice.”

But your hypothesis is interesting enough to make me want to investigate the matter more fully this summer.

[shaunnichols]

In this post, I’ll focus on methodology and history rather than the issue of compatibilism per se. (I worked most of this up before Kelby’s post, so I’ll post something about Kelby later.)

Kip! Thanks for the kind words! Where should I send the payola?
Kip worries that the method I use in this paper has the disadvantage of having the following structure: "if X is true, then we would expect Y, Y is true, therefore X is supported." He is of course right that one would prefer more direct support for a hypothesis. Such scruples play a major role in psychology, where the gold standard of methodology is to introduce a factor (a dependent variable) to see whether it affects outcome. One needs to introduce a manipulation to show that the factor causally influences the outcome as opposed to merely being correlated with the outcome. (Indeed, in some communities, it doesn’t count as an ‘experiment’ unless there is some such manipulation.) But obviously in the case of history this option is unavailable. Until we invent time machines, the best we can hope for is correlations. (And even if we do invent time machines, the IRB probably wouldn’t let us manipulate the past anyway. Jerks.) I comfort myself in the fact that this is quite a general feature of historical enterprises. Our best hypotheses about dinosaurs are all generated without introducing manipulations. And dinosaurs are obviously really interesting and worthy of study, as evidenced by the fact that they captivate 5 year olds. The history of philosophy is also really interesting (though usually only to post-pubescents), and to understand it better, I think we need to cut ourselves some slack and accept that the best evidence we can get is correlational.

As for the historical details (Is Hobbes a compatiblist? Is Kant an indeterminist?), I am not equipped to enter those debates. But in any case, I think it’s important for me not to enter into the debates but to rely on the claims of experts. Otherwise, I risk being guided by my own motivated reasoning. So I’m not even going to try to weigh in on the discussions of particular historical figures, though I have learned a lot from them.

Roman raises some more general worries about interpretation. He’s surely right that for some disputes, our contemporary categories could not be applied to historical figures. But I would have thought that for many of the central philosophical disputes that are most engaging (and least technical), it is fair to interpret philosophers with our categories. That is one central reason for studying those philosophers – because they speak to the same issues that we care about today. In the case of free will, there are historical periods in our tradition (e.g. ancient Greece) and other cultural traditions (classical Chinese philosophy) in which the issue was not formulated with anything like our categories. But I take it that the basic categories we use today (determinism/indeterminism and compatibilism/incompatibilism) were handed down to us from the modern era. For what it’s worth, for most of the modern philosophers I investigated, there were encyclopedia entries that described the philosopher’s view on free will. So it seems to be taken for granted by some experts (those that write for encylopedias) that we can categorize modern philosophers on these issues. Once again, I will defer to the experts.
Roman has a further worry - that there is so much bias in the interpretations given by the historians of philosophy. This does concern me a great deal. But anyone who consults the secondary literature has to contend with this. Indeed, I suspect that it’s a problem for historical enterprises generally. But I am loathe to give up on history because of this problem.

Eric challenges my use of statistics. I admit that my use of stats here is unorthodox, and I acknowledge this issue in a footnote in the final version of the paper (which wasn’t the one that got posted here – I might have sent Thomas the wrong one, which conforms to the fact that I am an idiot). Here’s the text from that note:
“There is some question about the propriety of using statistical tests on this kind of material. For certain assumptions about the standard statistical tests are not met. For example, one assumption for a chi-square test is that the sample is randomly selected. That’s not the case here, of course, since the philosophers were not randomly selected by the anthologizers. On the other hand, the selection was not biased by the hypothesis that we are exploring. If we think of statistics as a tool for uncovering patterns, as opposed to an end in itself, then we might think it worth at least considering the results of some simple analyses.”
I do think that the results are useful, though Eric is quite right that there are confounds that make it worrisome. Once again, I think we have to make some general decisions about what standards and expectations to apply to historical data. At a minimum, I think we need to be somewhat less demanding.

Eric describes my paper as a work in the psychology of philosophy. I do regard it as part of that general endeavor. And I think Eric is right that there is a wealth of issues to explore here. Drawing on Sulloway’s work on birth order, for instance, strikes me as really promising. Eric himself has recently been doing work on the psychology of philosophy. I encourage people to look at his delicious investigations of the ethics of ethicists (http://schwitzsplintersethicsprofs.blogspot.com/), which he’ll be presenting at the SPP this year.