The Free Will Debate and the Limits of Philosophy

Are there limits to philosophical method as a way of rationally adjudicating disputes? One thought is that such limits are illustrated by several disputes within the free will debate. For a first, manipulation cases are designed by incompatibilists to sway intuitions in a dialectical situation in which prior reasoning has not broken a standoff between compatibilists and incompatibilists. The incompatibilist can hope that intuitions about such cases will generally go her way. She can bolster her case by arguing, for instance, that it’s inappropriate to privilege intuitions about ordinary cases that don’t involve manipulation. But she cannot justify the claim that when in the end the compatibilist’s intuitions don’t conform to hers, he is irrational. One plausible diagnosis is that for the compatibilist, reflective equilibrium involves affirming responsibility in any properly constructed manipulation case (one in which all of the defensible compatibilist conditions on responsibility are satisfied) and retaining responsibility in the ordinary deterministic case, while for the incompatibilist reflective equilibrium goes the other way, denying responsibility all the cases instead.

Another illustration involves a move the leeway theorist can make against Frankfurt examples. As I see it, in a good Frankfurt example there will be a trigger for intervention whose occurrence the agent could have but does not cause.  It’s open to the Frankfurt-defender (e.g., Widerker 2006) to argue that the agent’s moral responsibility is derivative of his responsibility for not causing the occurrence of the trigger. I’ve argued that invoking derivative responsibility at this point is suspect because the good Frankfurt example differs relevantly from the paradigms of derivative responsibility, in which the agent knowingly makes it the case that he will not satisfy uncontroversial conditions on responsibility, and he foresees that he might then perform the action at issue. But the intuition that an agent can only be blameworthy if she has an exempting alternative is strong. For many, it might be so strong that reflective equilibrium involves retaining this intuition and affirming that the agent in the Frankfurt example is only derivatively responsible despite the departure from the paradigm case. But for many others, reflective equilibrium involves going Frankfurt’s way instead.

Are we stuck at this point? One option for manipulation cases is for the compatibilist to claim that the incompatibilist needs to provide an additional argument that the agent is not morally responsible in the manipulation case, or for the incompatibilist to contend that the compatibilist needs to provide an additional argument that the agent is morally responsible in the manipulation case. But ex hypothesi, the point of such examples is to adjudicate philosophical standoffs just by intuitions about cases, and if the intuitions aren’t forthcoming, the standoff remains in place. And in addition, I’m thinking that the only way philosophy provides for adjudicating standoffs in these kinds of situations is by intuitions about cases. So if the hoped-for intuition isn’t generated in the best sorts of cases, or if it will always be swamped in the process of reaching reflective equilibrium, it seems that philosophy has reached its limits for adjudication, and we need to live with philosophically irresolvable disputes. Is this right?

Comments

[Tamler Sommers]

Yes.

[Kip]

There are two things that prevent philosophers from making more progress in the free will debate:

1. semantic ambiguity
2. cognitive illusions

About 1: terms like "free will" and "moral responsibility" are not defined precisely enough, to do the kind of hair-splitting work that philosophers want them to do. We each have our pet favorite definitions, but no way to force others to agree with us. Too few seem to do intensive empirical work on how people actually use these terms (if they do use them). And many philosophers resist the idea that, even if the empirical data supported one definition, that data would necessarily govern the definition.

About 2: there are numerous (overlapping) biases (fundamental attribution error, positive outcome bias, reactance, illusion of control, just world phenomenon, victim blaming, the empathy gap) that prevent people from accurately seeing reality on the topic of free will. Even worse, there are also a lot of biases that prevent people from seeing the ways in which determinism is not, per se, threatening to free will. These biases are persistent, by definition.

A third problem is that many issues in the free will debate invoke underlying issues about personal identity. For example, a lot of manipulation arguments invoke issues about how much you can change someone with, or without, actually killing them (even if a new, living, breathing person remains). These issues are related to concerns about constitutive luck, and how much the concept of constitutive luck even makes sense.

My not-so-modest suggestion is that philosophers aren't going to be able to make much progress until they seriously grapple with all three of these issues. I tend to think that these problems (at least the first two) infect philosophy in general, and not just the free will debate.

[James Laird]

We’ll need to live with philosophically irresolvable disputes until we embrace a new idea: human thoughts exert new emergent forces. When we grasp that, we’ll open a whole new path forward.

[Ben Matheson]

Hi Derk

Really cool post. From a compatibilist perspective, I think one way to make progress is ask *why* incompatibilists get the intuitions they do about manipulation cases. If it can be shown that the incompatibilist intuition stems from a dubious source, then I think that will count against that intuition. For instance, someone might hold that Professor Plum is not morally responsible because he has been named after a fruit. If that were the source of the non-responsibility intuition, then the non-responsibility intuition is going to be useless in an argument against compatibilism. Of course, you might say that the source of the non-responsibility is the fact that Plum has been causally determined by factors outside of his control. But perhaps it’s the case that the source of the intuition is something else. My thought is that if it can be shown that there is at least a competitor to the incompatibilist’s diagnosis of the source her intuitions about moral responsibility in manipulation cases (and cases of mere causal determinism), then (I think) we cannot be sure what the source of the non-responsibility intuition is. And if that competitor diagnosis has a dubious source (perhaps, say, implausible beliefs about the nature of persons, or something), then it seems the non-responsibility intuition isn't going to be helpful in an argument against compatibilism. Of course, it’s open to incompatibilists to make the same move against compatibilists. But it seems like we're making progress.

[Justin Capes]

Here's one of my favorite David Lewis quotes:

"The reader in search of knock-down arguments in favor of my theories will go away disappointed. Whether or not it would be nice to knock disagreeing philosophers down by sheer force of argument, it cannot be done. Philosophical theories are never refuted conclusively. (Or hardly ever. Gödel and Gettier may have done it.) The theory survives its refutation—at a price. Argle has said what we accomplish in philosophical argument: we measure the price. Perhaps that is something we can settle more or less conclusively. But when all is said and done, and all the tricky arguments and distinctions and counterexamples have been discovered, presumably we will still face the question which prices are worth paying, which theories are on balance credible, which are the unacceptably counterintuitive consequences and which are the acceptably counterintuitive ones. On this question we may still differ. And if all is indeed said and done, there will be no hope of discovering still further arguments to settle our differences." (Lewis, Philosophical Papers Vol. 1 p. x)

Lewis's remarks seem especially applicable to the free will debate.

[Justin Capes]

I should say that although I think the Lewis quote I posted earlier is right on, I don't think all the tricky arguments and counterexamples have been made in the debate over manipulation cases or Frankfurt-cases. But once they have been, the limits of philosophy for adjudicting disputes will be reached, and we will (I suspect) still disagree.

[John Fischer]

Great post, Derk.

YES. I think that reflective equilibrium is the only game in town (really--but, in Norman Daniels' term, "wide" reflective equilibrium), and yes, it is plausible that people will just go different ways in this process. So philosophy has limits. We *can* however make significant progress, showing how things fit together, pruning out inconsistencies, discovering our deeper principles, and so forth. But there are limits here--not unlike in other areas, such as normative ethics. That's life, and I think we just have to deal with it.

One point I would make is that a proper WRE methodology should not just focus on intuitions and principles about a narrow issue, going back and forth and seeking equilibrium about that particular issue. It should cast its net more broadly (or, I guess, "widely".) So, just as an example (admittedly close to my heart), an evaluation of the Frankfurt-style cases as putative counterexamples to PAP should also include consideration of general frameworks or explanations of the "value" of moral responsibility (or "pictures" of moral responsibliity). So the proponent of PAP has a "make-a-difference" picture, and the proponent of the FSCs against PAP might have a "make-a-statement" picture. Of course, more would need to be said, and perhaps there are make-a-difference pictures that are consistent with the rejection of PAP. And so forth. But it can be helpful to seek illumination by considering framework principles.

Still, of course, it is all about consistency, at least in my view. That still allows for lots of flexibility in rethinking intuitions and principles. But in the end, what else is there? This is the methodology Rawls has defended in ethics, but also (roughly) the methodology David Lewis has defended for metaphysics (and other branches of philosophy). Lewis wrote that somtimes we can't clear out the fog entirely to achieve clear blue skies, but we can shift the location of the fog in ways that are theoretically advantageous.

[Brent]

Isn't it possible that the most interrupting biases are not those regarding definitions and intuitions *within* the debate, but rather those regarding our intuitions about which possible world would turn out better (one in which strong free will is affirmed or one in which it is debunked). Perhaps it is our (unconscious?) commitment to this related but very different question (which world we think would turn out better) that results in the stalemate going on in specific debates.

If we are like icebergs, the exposed 10 percent that floats around above water represents our specific free will positions. The 90 percent that lurks beneath represents our incompatible worldviews (either the concept of free will is the only thing between us and chaos *or* the concept of free will has always retarded human progress, and is unjust on the side) that keeps our top parts from truly engaging in the first place. Viewing it from the surface, we find it notable, as Derk has in this post.

I think this idea is supported by the fact that, comparatively speaking, this larger question (of which world would be better off) is rarely seen or intended to be the center of the conversation. Indeed, it is often thought of as distracting from the real, more modest and seemingly answerable specific free will questions (compatibility, luck, etc.).

Perhaps if the underlying question is attacked, the trench warfare that is the free will debate would be likewise disturbed.

[Andrew]

I think Kip and Brent are onto something about 'cognitive (and affective) illusions' lurking underneath. But following Taylor and Brown (1988), I also think there's an inevitable trade-off between tracking and living with inconvenient truths and maintaining a basic, day-to-day healthy mental function. Even those who can apparently handle the cold hard truth of the matter (Coyne, Harris, Greene etc) don't advocate applying it strictly and consistently to your personal life. Realist utilitarians (Singer) are probably in the same boat.

[Paul Torek]

Kip, Ben, and John all flirt with a point that I would put as: wide reflective equilibrium is wider than you'd think. In the spirit of Hofstadter's Law - I propose Torek's Rule: reflective equilibrium is wider than you think, even when you take into account Torek's Rule. And that wideness allows the landscape a lot of power to force answers to apparently unanswerable questions - or to dissolve apparently indispensable questions.

[Sofia Jeppsson]

I pretty much agree with Derk; at the end of the day, people's reflective equilibria will probably still differ to some extent, and there's nothing we can really do about that. And I'm also sympathetic to Justin's Lewis quote.

I try to steer clear of thought experiments as much as possible for this reason. I try to discuss what kind of function various moral terms fill instead, as a way of getting around some of the old stale-mates - but if everyone were to follow me, new stalemates would probably just appear. :-)At the end of the day, any method invoked will depend on intuitions to some extent, and the reaching of a reflective equilibrium.

We're probably influenced by all kinds of bias. The thing is, if that's true for one camp in the debate, it's probably just as true for another. I think there's a tendency among people who embrace "radical" ideas, ideas that are contrary to some kind of common sense, to think that they must be less prone to bias than most people. And perhaps people with "radical" ideas are less prone to conforming-to-the-group-bias, that much seems plausible enough, but that doesn't preclude them having a bias towards the radical; to be terribly fond of the idea of themselves as truth-sayers, as real intellectuals who dare pull even the most radical of conclusions and embrace views that most people can't bring themselves to accept.

This is, of course, an empirical-psychological question, but I would be surprised if it turned out that any particular philosophical camp were more prone to bias than any other, although different kinds of bias may rule different camps.

[Sofia Jeppsson]

I should add that I do think Kip makes an important point about personal identity, though. There's a difference between
a) A manipulates B, and B is therefore not morally responsible
and
b) A plays around with a second body which is essentially A:s puppet - there's only one agent here, which is A, and A can be morally responsible.
There's also a difference between
a) A manipulates B, and B is not morally responsible for what she does after manipulation,
or
b) A interferes with B in a way that destroys B and creates a new agent, C, instead. C can be morally responsible for what C does, but can't be held morally responsible for what B did before the manipulation.

I think these scenarios tend to get muddled up in many thought experiments.

[Neil Levy]

I hereby endorse Torek's rule. It is worth noting that the right test for whether an argument has succeeded (in changing intuitions) is not - just - whether it changes minds. Rather it what happens to the intuitions of the next generation. Science advances, one funeral at a time; same thing true here.

[Ben Matheson]

Sofia, I agree about the personal identity stuff. But notice that in your second set of cases you are already seem to making assumptions about personal identity - namely that a psychological approach to personal identity is true. If a biological/bodily approach of some sort were true, then(psychological) manipulation would not create a new agent no matter how thorough it was. Perhaps the reason that many don't see a difference between these two cases is that they implicitly hold some sort of biological (or other substance-based) approach.

[Sofia Jeppsson]

That's a good point, Ben.

[John Fischer]

Sofia writes:

"I pretty much agree with Derk; at the end of the day, people's reflective equilibria will probably still differ to some extent, and there's nothing we can really do about that. And I'm also sympathetic to Justin's Lewis quote.

I try to steer clear of thought experiments as much as possible for this reason. I try to discuss what kind of function various moral terms fill instead, as a way of getting around some of the old stale-mates - but if everyone were to follow me, new stalemates would probably just appear. :-)At the end of the day, any method invoked will depend on intuitions to some extent, and the reaching of a reflective equilibrium."

I think we are in basic agreement, especially with the points that at the end of the day we can't avoid an appeal to intuitions, and also the additional point that we are all prone to bias--no camp, even those who depart more radically from common sense, is immune to bias.

But I do wish to point out that David Lewis himself was no enemy of thought-experiments! His work is filled with them. I don't really see how one can use the WRE methodology without to some extent employing thought-experiments--they are analogous to controlled experiments in science. And, whereas one can try to get results from simply studying "the function of moral terms", how far can one really get by simply making a descriptive study of these functions? I think that one will inevitably be making recourse to thought-experiments--as Sofia does in her post on the various cases involving manipulation!

Of course, WRE will employ different elements, of which thought-experiments are just one. And they need to be used in moderation and carefully--often the abstraction from real-world details distorts our intuitions. But how to do without them? Even Wittgenstein was fond of them! Go figure!

[Tamler Sommers]

I like Torek's rule too but you've got to do more than just emphasize the wideness of WRE to make to it plausible that an objectivist or universalist account of MR is on the horizon. After all, someone could point to Torek's rule in defense of objectivism about taste judgments, aesthetic judgments, or humor judgments. Sure it's possible that this apparently irresolvable dispute will be find a definitive resolution. But that's true for any domain of inquiry. What's needed is an argument for why the MR debate is different from debates that we don't think will submit to an objectivist analysis.

[Robert Allen]

I'm just wondering how much hand-wringing goes on at the end of the day? Do these disagreements cause any of us to doubt our philosophies? Should they? Or, instead of giving in to despair, are we more prone to dig in our heels? Why should I be bothered by the fact that my colleagues disagree with me? Especially when it comes to free will, I know the score and from an unimpeachable source: my own mind. I am in complete control of myself, seasoned captain of my own volitional ship; no moral voyages are undertaken by that vessel without my express written consent. ("The illusion of control": is that like when I force myself to think about philosophy instead of the upcoming World Baseball Classic, but it's really not me having on effect upon myself, it just seems that way, the actual cause being some hidden neural process? Or, when I manage to divert my attention from some pressing personal concern long enough to teach a class, it wasn't really a case of self-discipline; rather, my neurons again came to the rescue? You know in a more tough-minded time, depression was considered a sin, so much did people then believe in their own spiritual invincibility.) As Reid cautioned, once one doubts free will, it's Katy bar the door regarding other forms of skepticism.

Sanguinity is to be found only in St. Anselm's dictum "I believe so as to understand." I don't doubt my philosophy- Agent Causalism- for an instant. My task, then, is to find the fallacies lurking within the objections its critics raise, which, again, I have no doubt are there to be unearthed.

[Gregg Caruso]

Great question Derk (and I apologize for joining the conversation late). I agree that the traditional stalemates in the free will debate are often the result of competing intuitions, even after applying WRE. I also think, however, that Richard Double’s work on the role of metaphilosophy is relevant here. According to Double, metaphilosophical views condition competing theories of free will. If correct, then both arguments for and against FW and MR will be persuasive only if one adopts supporting meta-level views of what philosophy is. I can, of course, make a case for my preferred package of meta-level views and substantive free will theory (i.e., free will skepticism), but if one does not share my metaphilosophy they are likely to find their intuitions going elsewhere. I feel like every paper on free will should start with (a) a list of definitions (what does one mean by “free will” and “moral responsibility”?) and (b) a statement of one’s meta-level viewpoint—e.g., are they adopting something like Philosophy as Conversation, Philosophy as Praxis, Philosophy as Underpinnings; Philosophy as World View Construction (along with intermediate-level principles that work in combination with the metaphilosophies—e.g., skeptical vs. non-skeptical epistemic standards, realism vs. instrumentalism, ontological conservativism vs. ontological liberality, strict vs. liberal requirements on explanations, etc.). Perhaps these meta-level and intermediate-level viewpoints are responsible, at least in part, for many of our disagreements?

[Sofia Jeppsson]

John, you're right, at the end of the day, one always has to say something about thought experiments too, and I do so too. However, I think it's interesting and fruitful to see what arguments it's possible to come up with that do NOT rely on "agent X in situation S is clearly responsible/not responsible, and since X is analogous to ordinary agents under determinism, the same goes for them". In part because philosophers never agree on whether agent X really is responsible or not.

When it comes to Richard Double's meta-philosophy argument, I've only read the one paper on it that was part of the Oxford Handbook, and I'm not sure I understood it. He says his view is that philosophy is continuous with science, but what does that mean really? I think philosophy neither competes with nor contradicts empirical science, since philosophy is concerned with non-empirical questions (although, in many areas, both philosophical and empirical questions tend to pop up). Does this mean I agree with Double? And are there people in the free will debate disagreeing with this picture?

I do think, however, that meta-ethics is often relevant for the free will/moral responsibility debate. Whether you think an agent being or not being morally responsible is a mind-independent moral fact or depends on a system of morality that we have somehow constructed for ourselves, for instance, seems to have implications for a lot of the debate.

[Derk Pereboom]

I agree with John that it seems right to use the method of wide reflective equilibrium in thinking about free will and moral responsibility. Is it possible to go too wide? G. A. Cohen (2008) argues against Rawls’s (1971) recommendation to use WRE to determine principles of justice on the ground that if considerations of justice together with the wider reflections suggest that we should adopt certain rules to regulate our institutions, those rules may not be principles of justice anymore because some of what will justify our adopting those rules are not relevant to justice per se. Similarly, if we use WRE for determining the nature of moral responsibility, it could be that because we’ve brought in concerns that are irrelevant to the nature of moral responsibility, what we get in the end is not a theory of moral responsibility but, for example, a theory about how best to treat each other. This issue is relevant to Bruce’s concern for Manuel’s position that arose in the “What Are We Fighting About?” thread. It could be that what we get after applying WRE in Manuel’s view is not a theory of moral responsibility, but a theory that dispenses with moral responsibility in favor of a different notion. Specifically, Bruce’s concern is that the view really rejects moral responsibility and instead recommends treating people as if they were morally responsible. As Sofia remarks, metaethical issues arise here. And as Kip and Gregg argue, it’s important to define terms so that the resulting controversies are not merely verbal. I would argue that there are different senses of ‘moral responsibility,’ and that Manuel’s view preserves an important one, but maybe not the basic desert kind.

[Gregg Caruso]

Sofia, I would recommend checking out Double's book "Metaphilosophy and Free Will." There he explains how many arguments and positions within the free will debate depend upon meta-level assumptions. I think that book is one of the most underrated and under-appreciated in the extent philosophical literature. While I do not agree with all of it, I DO think it may be relevant to Derk's question about the limits of philosophical argument and the role of intuitions.

[Sofia Jeppsson]

Thanks for the advice, Gregg.

[Paul Torek]

Sofia, on meta-ethics, very well said.

[Patrick Mayer]

I don't know if I am making something up here, but it seems like people are focusing on the methods and limitations of philosophy as the factor that prevents consensus on how to evaluate manipulation cases and Frankfurt cases. It might be that peculiar features of philosophy are partially to blame, but I think it is worth noticing that the situation we are in is not so unlike what you find in the sciences after a new theory comes along.

Newton was not accepted on the Continent until well after his death, it took decades for people to accept Darwinism even though the alternative theories (multiple cycles of divinely caused creation events, and things like that) about the distribution of traits across species were just terrible, a committed core of physicists still refuse to accept the presence of indeterminacy at the quantum level, etc. These people were/are not being obviously irrational. They were all quite smart, quite well educated, and usually quite serious about their jobs. They revised their theories to accommodate new experimental results that suggested the theory was false, or they questioned the viability of the measuring techniques in the experiments themselves, or they simply dug in their heels and claimed that the wealth of evidence for the status quo position gave them the right to assume some explanation, they knew not what, would come along to make their well confirmed theory cohere with the observational evidence.

I think the first place to go for an explanation of the recalcitrance of our fellow philosophers is whatever causes such recalcitrance in the wider intellectual world. And I have a hunch that whatever mechanisms are in place in physics and biology are going to do all or nearly all of the work in philosophy. I think this because I don't think we are so much worse off than they are. Consider all the progress in the free will debate from before the 1950s till now. I think this shows up best with compatibilists. Nowell-Smith and Ayer were the cream of the compatibilist crop and were making arguments from social utility for holding people responsible. Hasn't the free will community reached a consensus that this kind of defense fails? Similarly haven't we done away with CA Campbell type appeals to the fact that can introspectively know that we have free will? Many of the top people in the free will sub-field either were already working philosophers when manipulation cases started coming out, or were educated by someone who was. It takes a bit longer than that even in fields of study that don't feel the need to constantly worry that its fundamental methodology is suspect.

[Sofia Jeppsson]

Do you think John Perry's distinction between weak and hard laws of nature is relevant not just for leeway incompatibilism (that's the context in which I've seen him discuss the issue) but also for a view like Derk's? It seems to me that it probably is. If laws of nature are weak in Perry's sense it seems as if determinism ceases to be a threat not just to our ability to do otherwise, but also to our being the original sources of our actions. And the idea that indeterminism would somehow rob us of control seems to make less sense.

(I remember that Tim O'Connor discussed the possibility of weak laws of nature at the Budapest University summer school in 2010, only to dismiss the idea as "crazy metaphysics", but I don't remember now whether it was in the context of leeway incompatibilism or incompatibilism period.)

Any thoughts on that?

[Derk Pereboom]

I agree with Gregg that Richard Double’s work is really interesting, and has a potentially serious upshot for the debate. One of his main lines of reasoning is that the claim that we have the free will required for moral responsibility can’t be true, since it assumes that there is a mind-independent feature of human agency that is necessary and crucial for moral responsibility. The core reason he cites for there being no such mind-independent feature is that the concept of free will is internally incoherent. One supporting argument is that the debate between compatibilists and incompatibilists is irresolvable, and that the compatibilist and incompatibilist conceptions are free will are incompatible and are part of the concept of free will. Another supporting argument defends an irrealist metaethical position about moral judgments more generally, then contends that 'moral responsibility' and 'the sort of free will required for moral responsibility' are moral notions, and concludes that claims that agents are morally responsible and have this kind of free will also cannot be true.

I’m not so sure about Richard’s argument from irresolvable disputes. Probably many the major disputes in philosophy are irresolvable, and I’d be reluctant to accept irrealism in each case. Another explanation for irresolvability is that we don’t have a method for resolution. The debate between physicalists and ant-physicalists in philosophy of mind seems as irresolvable as any, but irrealism about this issue doesn’t seem plausible, and my guess is that the irresolvability here is due to the unavailability of a method of resolution.

I wonder if metaphilosophy can make a difference to how, for example, the manipulation and Frankfurt case arguments are evaluated. I’m initially skeptical. But maybe someone has a pair of contrasting metaphilosophies in mind that would occasion different responses to one of these arguments.

[John Fischer]

I agree that Richard Double's work on the relationship between methodology and our free will/moral responsibility literature is interesting and somewhat under-appreciated. I would also point to Manuel Vargas's new tome, *Building Better Beings*, that also has important reflections on the metaphilosphy of free will, as it were.

Does "metaphilosophy" make a difference for resolving debates about manipulation cases and Frankfurt-style cases? Great question. Dunno. But here are some inchoate reflections.

First, how one comes out on the Frankfurt cases might depend on how conservative one is. So PAP is a well-entrenched, central principle, and our actual conceptualization of both the forward-looking and backward-looking aspects of free will/moral responsibility reflects an acceptance of PAP. Whether one gives it up in light of FSCs might depend in part on how conservative one is--or how philosophically adventurous one is. But I admit that this is a slightly tendentious way of framing the issues. I myself do not think one should give up PAP solely on the basis of cases, but rather, a careful, holistic evaluation that includes alternative principles and alternative frameworks or "pictures" of moral responsibility (as I mentioned above). Still, how one comes out might depend in the end on one's philosophical personality. Also, of course, just as a practical matter, it is challenging to give up a position one has defended in print or wants to defend--so some of us have "stakes" in seeing our particular views succeed. This is a psychological factor that no doubt plays a role, even in the Weltanaschauung of a Sincere Semicompatibilist, hard as one tries to be fair.

Here however is another methodological reflection. How one is inclined to come down on manipulation cases might depend crucially on WHERE ONE STARTS. (A plug here for my ANALYSIS paper, "The Zygote Argument Remixed". If you start with horrifying and mortifiying and terrifying manipulation cases and work back to "ordinary casual determinism", you might think that, since you are obviously not responsible in the manipulation cases, you aren't in the ordinary deterministic cases as well (please allow the use of "ordinary" here, but you can keep track of that if you'd like). But if you start with ordinary cases (even under causal determinism) and you work back to the putatively horrifying manipulation cases, you might conclude that the agent can be morally responsible in at least CERTAIN of the so-called manipulation cases. (I distinguish, and I think it important to distinguish, manipulation from initial design cases, although some deem the latter to be manipulation cases; let me be clear: I do think that Ernie is morally responsible in Al Mele's famous Zygote Argument, and so I do think this sort of "manipulation"--initial design--is compatible with moral responsibility. (Of course, I admit that this is highly contentious, and reasonable people can and do disagree here.)

How exactly should we conceptualize this "path-dependence" of our intuitions? It shouldn't matter where we start, but it does seem to matter, at least as a psychological matter of fact. In Social Choice Theory, there is the norm or constraint of "path-independence", and I think something like this is or should be a norm in our reasoning. But which starting point is hegemonic? Or how should we think about these dialectical contexts that appear to involve an infelicitous path-dependence??

[V. Alan White]

I second (or third) study of Double's work. His metaphilosophical approach mirrors crude thoughts kicking around in my classes for years and helped push me toward free will pragmatism, favoring compatibilism or maybe semi-compatibilism, which I think best characterizes my stance these days. The mountains of literature on incompatibilism strongly echo my shouts of "what does it all mean?" back, and additionally encourages me to leverage the pragmatics of morality against metaphysical stalemates (HT to John for pressing such issues). Note that I do not think that the role that pragmatism plays entails that morality itself is pragmatic--at least every other day I am a moral realist. But ultimately what we need to do with one another morally in terms of outcomes and improvement trumps the bit more exotic question of what we ought to do with each other that sometimes regards outcomes and improvement as irrelevant. I can't see that.

This was an inspired post Derk. Thank you.

[Robert Allen]

John,

I never thought of myself as starting off at one end of the spectrum rather than the other. I began, instead, at both ends and worked my way to the middle. That is, I was a Compatibilist who was also certain that extreme forms of manipulation were inconsistent with FW. Once I got to the middle I looked around to see what was still left from the ordinary cases and lacking in the manipulation cases already dismissed. What I discovered was Mele's notion of the ability to "shed" influences in the form of beliefs and attitudes. I then touched up that idea to meet a counterexample of Kapitan and had what I thought was a viable form of Compatibilism.

[Bruce Waller]

Derk, Thanks for a fascinating question; and thanks also for making clearer the point I was muddling toward in my response to Manuel in your earlier question. John, your remarks about where we start from in dealing with these questions is very interesting; I wonder if it is influenced by the framing effect (Dan Kahneman talks about that a great deal in his work); my favorite example is of two priests, arguing about whether it is OK to pray and smoke at the same time. The first reports a meeting with the Pope, in which he asked the Pope whether it was permissible to smoke while he prayed; the Pope answers that it would be wrong. The second priest says that he is surprised, because when I questioned the Pope, I got exactly the opposite answer; but I asked the question in a different manner: Is it permissible for me to pray while I smoke? Let me add one more strong endorsement of Double’s work; it is not only very insightful, but a wonderful read: I still laugh every time I think of “sheepish subjectivism.” On the intriguing question of how metaphilosophy influences our views on specific questions, Tamler Sommers has some very interesting ideas in Relative Justice; and Neil Levy, in Hard Luck (I think it’s the penultimate chapter) has a remarkable argument about Frankfurt examples that places the issue in the larger context of internalist vs. externalist/environmentalist views in epistemology and philosophy of mind (whether that rises to the level of a metaphilosophical concern I’m not quite sure).

[John Fischer]

Bruce,

Nice points--and yes, I think that the path-dependence to which I pointed is indeed similar to the Kahneman/Tversky "framing effects". And framing effects are by definition irrational. Somehow we need to become aware of these effects (as in the work of Peter Unger, especially in *Living High and Letting Die*, and in the work of various X-phi folks) and minimize their effects on us.

I like the idea of requiring in our WRE a kind of "path-independence" or our considered intuitions. That is, there should be a constraint requiring resiliency with respect to the path one takes to a case. If there is an initial path-dependence, we should rethink our views and seek path-independence.

But, yikes, sometimes it is unclear which is the hegemonic starting point. Here's where holism plays a big role. So, for me, since I think that compatibilism (of the Semi-variety) has considerable virtues and indeed tremendous advantages--it can allow us to avoid the menacing Consequence Argument and worries about Theological Determinism, it can avoid having our status as morally responsible agents hang on a thread, similarly, it avoids having meaningfulness in life, friendship, love, and so forth hang on a thread--I am willing to start say with intuitions about "ordinary" cases under causal determinism and "work back" to Ernie. (And, by the way, I think it is independently plausible that the distal intentions of a creator, such as Diana, say a billion years ago, shouldn't matter for Ernie's responsibility--and it wouldn't matter, similarly, if it were just thirty years ago.) But, of course, reasonable people can take a different dialectical path here.

[Robert Allen]

John,

I still don't see why being a Compatibilist (of any sort) allows one to initially avoid the CA, TD or worries about meaningfulness, love, etc.. Nor did I put Ernie on the back burner. These issues were salient all along, shaping the formulation and defense of (my version of) Compatibilism.

[John Fischer]

Robert,

The CA and TD threaten our status as free to do otherwise. But Semicompatibilism holds that we can be morally responsible, and have meaningful lives, friendship, love, and so forth, even if we (and others) lack freedom to do otherwise.

That's how it is supposed to work, anyway. I have however heard rumors that at least suggest that not everyone is (fully) convinced...

[Derk Pereboom]

On John’s thoughts about holism and WRE -- here’s a worry about giving our current practices and commitments heavy weight in reaching reflective equilibrium. Sometimes skeptical threats to a practice arise that ordinary people take very seriously. I think skepticism about free will and moral responsibility is one of them. It’s natural to turn to philosophy to adjudicate the skeptic’s claim. But if the method of philosophy is wide reflective equilibrium, and if this method accords heavy weight to our current practices and commitments, then it looks like we’ll be able to tell in advance how philosophy will adjudicate the question. And even if we can’t, it seems that philosophy initially favors the anti-skeptical side, and for that reason is untrustworthy. This will motivate those who take the threat seriously to look elsewhere for advice.