On the term 'free will': What's up with that?

So Bruce Waller and Justin Capes are pleading for something to discuss in the month of March, and they've asked me to start early (I am not slated to begin until April).  Sorry, I cannot do much right now, but figured I could pose a question about our use of terminology. 

Here is a simple question: How ought we to make use of the term 'free will'?  Is it merely a term of art?  Does it pick out something philosophically interesting in folk usage, something that x-phi-ers can test?  Does it gain content by reflection upon the phenomenology of agency, as my colleague Terry Horgan contends?  Should we just forgo use of it and speak in terms of free action, or acts issuing from a relevant sort of control?  What's at issue?  Nothing?  Is this mere bookkeeping?  Or is there something substantive at stake?  Can we unproblematically move from arguments of those like John Martin Fischer and Mark Ravizza to counter-arguments of those like Robert Kane?  Recall, Fischer and Mark Ravizza write in terms of control and avoid altogether use of the expression 'free will', whereas Kane insists that there's something important lost if we do not attend to a special faculty picked out by the expression 'free will'. 

My own opinion is that 'free will' is merely a term of art whose content is fixed by philosophical tradition(s).  This, combined with the fact that there is not just one traditional philosophical problem which is "the" free will problem, suggests that we philosophers can have some latitude in how we use the term, so long as we are clear about how we are using it. 

I myself define 'free will' as the unique ability of persons to exercise control over their actions in the strongest manner necessary for moral responsibility.  I then understand free actions as those actions issuing from an exercise of that ability.  (Of course, we need to talk about non-deviance, and so forth.) 

But others might say I'm smokin' crack (and I'm not necessarily denying that I am).  They might object that free will just is the ability to do otherwise, and that it is folly to begin with a definition that is normatively charged as mine is.  Now, my reaction to this objection is just to allow that we can use the term that way (that is, as the ability to do otherwise), since there is a distinct philosphical puzzle here understood as "the" free will problem--a problem that is just about the relation between determinism (or indeterminism) and the ability to do otherwise.  But I deny that we must use it that way. 

What do all of you think?  Joe Campbell, you gave a nice paper at Arizona last year arguing that we should understand 'free will' in terms of what is up to a person at a time.  Are you still committed to that?  Others, what do you think?  Bruce?  Justin?  Have at it! 

Best,

Michael 

Comments

[Kip]

I like to think that people (i.e. the rabble) use "free will" consistently enough to shed light on two things:

1. whether it is consistent with determinism
2. whether it exists

I like to think that is true, even if people do not use the term consistently enough to fully define it (or paint a picture of it).

As far as I can tell, there has been far too little research on how people actually use the term "free will." And this is a big obstacle to making progress, because compatibilists and incompatibilists tend to use different definitions.

But, if/when that research is done, I will not be surprised if people do not use "free will" consistently enough to make any progress. We can call this the "Richard Double" threat. I worry that, in the end, Double was right, "free will" is not properly defined, and so we're all wasting a lot of time, arguing about whether it exists (or is consistent with determinism).

I don't like the idea of setting up a pet definition of free will, even if you clearly announce your pet definition, and then doing philosophical work with that definition. That strikes me as a self-indulgent kind of word game, thoroughly divorced from the millenniums-old tradition of the "free will" problem. That word game has practically no relevance to any philosopher who disagrees with your pet definition. If we're not trying to solve the free will problem as that term was defined historically, and widely used today, then what the hell are we doing?

[Side note: I don't like to define free will in terms of moral responsibility. I see those terms as flip sides of the same ambiguous coin, so defining one in terms of the other solves nothing. I also think that free will can exist, even if objective morality, and moral responsibility, cannot exist, for the same reasons that Double outlined in The Non-Reality of Free Will.]

But I hope Double was wrong.

[Eddy Nahmias]

Thanks for getting the ball rolling Michael! (And if you were not alluding to the SNL skit "What's up with that?" you should be.)

I think your definition is (a) about right, (b) about what ordinary people have in mind, and (c) what philosophers should care about when debating whether free will exists, what might threaten it, and (too much neglected) how much humans possess in general, how much different humans possess, and how much individuals exercise in different situations.

You define free will as "the unique ability of persons to exercise control over their actions in the strongest manner necessary for moral responsibility." I agree, except I'm not sure what the 'unique' is meant to do and perhaps you should go ahead and flesh out 'moral responsibility' to avoid the ambiguity Kip mentions and others worry about (e.g., justifiably deserve praise and blame).

My guess is that this understanding of free will is (properly) essentially tied to people's understanding of what capacities are required for persons to make choices, to feel like they control their actions (and to feel free), to be able to choose otherwise (but not in any incompatibilist way!), and for an agents actions to be 'up to' them.

The existing evidence suggests that the term "free will" is very closely associated with the term "choice"; my data suggests that people respond very consistently in response to statements or questions about 'free will', 'deserving blame (or praise),' 'making a choice,' and 'actions being up to the agent'. Some of my recent work on Frankfurt cases suggests that people associate 'making choices' with free will more than 'having choices'.

[V. Alan White]

"Free will" is a term like "death"--so heavily value-laden that it is useless or absurd to try to apply it in critical contexts as if it had reliable reference apart from such association. But that comparison is instructive. "Death" must have empirical descriptive properties to be meaningful. So it must pick out something functional as related to our biology--the heart, the whole brain, higher-brain consciousness--that it then endows with the value of being the determinate function of when life irreversibly ceases (Veatch has long advocated this point). Such values vary wildly among different groups of people for lots of historical, religious, and sociological reasons. So death as a concept evolves at least in the sense that values evolve to attach to objective criteria as superceding previous ones. In the pragmatic sense of current US law, the values that attach to whole-brain criteria have generally superceded those of more holistically somatic or a purely cardiac sense of death. We can disconnect ventilators by declaring death for the comatose now pretty regularly, where we could not have done so in the early 60s.

Someone's having "free will" likely is declared as a relevant feature of human nature on a similar basis. We attach a value-laden meaning to that term as we pick out properties of an empirical or metaphysical nature as more important (and much of the dispute between compatibilists and incompatibilists is thus ultimately about whether empirical or more abstract metaphysical properties are more valuable and usually established on outlying philosophical world-view criteria).

This is why I'm gravitating to a pragmatic view: I think in fact the averaged reasoned view of what is valuable about free will has evolved to favor more empirical characteristics than purely metaphysical ones. While the tortured metaphysical arguments about incompatibilism go on, focus on reasons, attributability, and other emprically identifiable psychological characteristics have gained favor for demarcating matters of responsibility, both philosophically and in practical legal statutes. I for one take van Inwagen's retreat to mysterianism as a watershed: one of the best philosophically gifted minds has declared a stalemate on that metaphysically argumentative issue, and so (I would say) the tide turns to what values ultimately matter as attaching to what we can know about how minds work with respect to our applying the term "free will" to that empirical question.

I think the question of what values actually matter as applied to the mind's working is a reasonably informed pragmatic affair. The very fact that values have in fact ranged over a wide spectrum of disagreement of what matters gives rise to Double's skepticism about the use of the term "free will". But the ongoing fact that we have had (and still have) such a wide spectrum of values about death, yet have come to a reflective equilibrium about what it means thus far to be actually dead, gives me hope that "free will" may yet have a reasoned if pragmatic value-laden definition based on some citable and I think empirical criteria. So can we have a reflective concept of free will that achieves the standard of equilibrium at some given state of understanding of its value? I think so. But if "death" is our comparative standard of definition for it, that definition of "free will" may yet evolve.

[James Laird]

Michael,

Great question. I don’t see FW as a term of art – I see it as something much more concrete.

The issue with the definition of the term “free will” is fundamentally about control. Is there any true sense of freedom about our actions, or is there only one possible path forward and that path is predetermined *solely* by the four fundamental forces of physics (4FFOP). I’m thinking there’s likely a mix between science and philosophy that’s needed in order to find the truth.

Yes, we can say that free will exists in the “weak sense”, because a person obviously has the capability to make a choice. But the really big question at hand is: can a person make a choice that isn’t predetermined *solely* by the 4FFOP (i.e., does FW in the “strong sense” exist). My opinion is yes, but in order to believe that’s true, we need to believe in something that humans haven’t formally recognized yet: human thoughts exert new emergent forces that add into the mix and are part of what determines the path of reality (i.e., forces that don’t result simply from a direct sum of pre-existing forces).

I believe that mankind can prove FW exists in the strong sense, because the intelligence associated with the interaction of two human thoughts isn’t innate to the 4FFOP. In other words, if all events in material reality are controlled *solely* by the 4FFOP from the bottom-up in a predeterministic manner, then that includes everything that happens within your physical brain. So how is it possible that one thought within your brain can affect another thought within your brain in an intelligent manner, if all of the control is from the 4FFOP? Is it a miracle that your thoughts make sense, or is it more reasonable to believe the intelligence associated with the interaction is an emergent property of billions of neurons firing in a coordinated manner? It doesn’t matter how you model the detailed mechanics of the interaction between two human thoughts; the fact remains that some kind of intelligent interaction happens, and the intelligence isn’t sourced directly from the 4FFOP.

[Tamler Sommers]

Michael, great question. I agree that it's term of art that comes out of philosophy. But the term also has taken hold of the public imagination in a way that 'moral responsibility' and certainly 'desert' have not.

I realized a while back I only cared about moral responsibility, and like you took free will to be whatever freedom was required to have it. In my book, I barely talk about 'free will' at all because I didn't want to add an unnecessary layer of complication. So I wanted to leave 'free will' out of my subtitle. I thought it was false advertising. But the editors strongly recommended that I keep it in. Otherwise, I would lose a lot of readers. So I did.

[Bruce Waller]

Michael,
On behalf of Justin and myself, thanks for responding to our desperate pleas; and with such a great topic. I’ve always admired your very straight forward way of framing a question, and your proposed definition of free will as the “unique ability” etc. is a great way of getting to the core of the issue. But whatever the “folk” (Kip’s “rabble”, Hume’s “vulgar” – those terms seem a trifle harsh) might say about free will and its relation to moral responsibility, that seems pragmatically a less than ideal way of starting – for it is indeed “normatively charged”. The result seems to be that we can use a moral judgment (about whether one justly deserves blame) to draw a conclusion about the powers of human animals; as in Van Inwagen saying that it is as good a justification as can ever be given for free will to say that free will is necessary for moral responsibility, and obviously we do have moral responsibility. Also, as you note, it makes free will a “unique ability of persons” (and here “persons” seems to imply humans); and it seems to me that one of the best ways of studying free will would be to examine it (or at least its close approximations) in the behavior of other animals; I would agree with your old colleague Michael Ruse: if something is very important to human life (Ruse is speaking of ethics, but the same would apply to free will) then we should expect to find elements of it in other animals. And indeed we do find very important elements of free will in the behavior of other animals, especially if we think of free will (as Dennett does, for example, and as Eddy, Alan, and James all seem to agree is central to free will) in terms of control; after all, some of our most important insights on the psychological importance of exercising control come from studies with dogs and mice and rats. So wouldn’t it be better to start with our best understanding of free will (and that will involve all sorts of psychological and biological investigations of why control and a sense of control is important for human animals, and the sorts of investigations that Eddy and Alan describe so well), and then see whether our best account can support claims and ascriptions of moral responsibility? Michael, thanks again for a wonderful topic, and a great “fix” for those of us suffering withdrawal pains; I think this makes you an “enabler”; but of course I don’t think you deserve blame for your moral shortcomings, nor reward for your much more considerable virtues.

[Eddy Nahmias]

Bruce, picking out the capacities required for MR (or desert), in part by exploring ordinary (I've let go of "folk") conceptions and practices, can (and should) be done without thinking that such work *establishes* what powers humans have. Rather, it tells us what powers to look for. Then we do science (and metaphysics) to see whether we actually have those powers (or capacities).

And then we also know what to look for when we are considering (1) the evolution of these capacities (e.g., by doing fun comparative ethology to see what our primate cousins can do), (2) the development of these capacities in children, as well as (3) the various neuropsychological deficits that diminish these capacities (and hence mitigate responsibility). Just as there has been way too little work on how to understand degrees of freedom and responsibility, there has been too little discussion of the evolution of the relevant capacities (Dennett and Baumeister both say some interesting things, and I've got a little story I tell in my dissertation.)

[Justin Capes]

Michael,

Your understanding of free will is a sort of functional definition. Assuming (as we should) that Frankfurt-style cases don't work, it might turn out that you your definition is extensionally equivalent to the "could have done otherwise" definition.

I prefer to use "free will" in such a way that it at least entails "could have done otherwise" because this is the way I think it has historically been used , and so, unless it could be shown (e.g., by a Frankfurt-style case) that the association between free will and responsibility is mistaken, to alter our usage only serves to muddy the waters.

[Roger Turner]

I'd like to see someone take seriously the (allegedly) distinct notions of liberum voluntas and liberum arbitrium (i.e. freedom of decision and freedom of the will). These are two notions that, at least, Aquinas and Augustine took seriously. For what it's worth, they both seem (to me) to be of the opinion that the former is the sort of capacity we have now, but the latter is reserved for the eschaton.

I'm among the tribe of people that conflate these two ideas; so, I want to know why two titans like this thought they were distinct. I think if any sense can be made of this, it might be helpful in the current debate. It might help us, in other words, figure out what we *mean* by 'free will'.

[Bruce Waller]

Eddy,
Sorry, shouldn't write these things when I'm rushed; I didn't mean that exploring the ordinary conceptions is not useful (I believe it is very useful); rather, I don't like starting from the idea that whatever free will is, it must support moral responsibility.

[Christopher Haugen]

Coincidentally, I was working on a somewhat similar topic. I claimed that the answer to the question “What is free will?” has an answer that is necessarily true, but the question of whether we are free is not necessarily true.

To return to your question, which is a good one, a few thoughts crossed my mind. While I contend that the question “What is free will?” has a necessarily true answer, it seems that there are many other questions that we can more easily answer. They are “What do people mean by free will?”, “What happens in our brains when we make decisions?”, “What are the causal relations involved in making a decision?”, “What are the objects that are causally related in making a decision?”, “What is the impact of experience on making decisions?”, “What is the impact of the intellect on making decisions?” and so on. Notice that none of these questions necessarily answer the question “What is free will?” Answering these questions merely tells us what is actually going on in the world—not whether it is an instance of free will. An account of free will will have to do that. However, we can make plenty of decisions about how to make decisions and how to treat other people in light of their decisions regardless of whether the act is free. For example, experience has a great deal of impact on a person’s decisions—especially when the person in question is young. In light of this, one might allow a young person to make a serious mistake with deep, but not permanent, consequences in order to provide a corrective lesson. The question of whether one is free does not impact this whether one uses the corrective lesson or not. That is, even if the youngster is not free, one will allow the lesson anyway to correct behavior.

This is a simple example, and there is more to consider. However, I think this shows that answering “What is free will?” need not have as much impact as we think it does.

[John Fischer]

Bruce,

Right: we should not start with the assumption that any analysis of "free will" will *support* moral responsibility. But it is not implausible to suppose that "Free Will" picks out, or could be construed helpfully as picking out or seeking to pick out, the control linked to moral responsibility. Of course, it would be then a separate question whether we have that kind of control.

Justin: you write that we should assume that the Frankfurt cases "don't work". Well, I certainly concede that it is highly contentious whether they work, and on behalf of what precise point they do work, if they do. No question about it: reasonable people, and even philosophers, can and do disagree here. But were you just joking, or do you really think it is ok just to assume a certain view about the FSCs in this context? And why would one need to make such an assumption in order to understand what "Free Will" picks out (in anything)?

[Justin Capes]

John: I was joking.

[John Fischer]

Ok, ha/ha. I'm hoping that you are not joking when you say, "John: I was joking."

[Joe Campbell]

Good comments! I like your definition, Michael: "the unique ability of persons to exercise control over their actions in the strongest manner necessary for moral responsibility." Well, sort of. I think that criticisms by Bruce and James apply and here is one more criticism of this definition: It makes the debate about whether free will entails moral responsibility a trivial debate whereas it strikes me as a substantive debate.

For similar reasons, I wouldn't include "the ability to do otherwise" as part of the meaning of the phrase "free will" either. My own view is something like this: Free will is the most fundamental power relevant to action (I think I stole this from Al Mele but it is close to Michael's). In my book, I say this power is "up-to-usness" (borrowing the term from Saul). And then I ARGUE that if an action is up to me, then I must be able to do otherwise. This latter is (I think) a substantive conclusion and not part of the meaning of "free will."

[Bruce Waller]

John, I agree that there could be such a free will-moral responsibility connection; but I'm curious about how strong that connection would have to be. Suppose you come up with a splendid account of free will, but you conclude that it could not support moral responsibility (actually, that's the reaction I sometimes have in reading your marvelous work: you give me wonderful insights into free will, but I'm never quite convinced they support MR; but I digress); would that cause you to conclude that your account of free will is fatally flawed? (You might still decide that there are possible grounds for MR; just that this new wonderful account of free will cannot provide them. Mike (and of course others), I would also be most interested in your responses.

[Michael McKenna]

Hi Everyone:

Thanks for your comments! Again, please forgive me for not participating much. As I explained, I really cannot do much right now, given some of my other commitments. But I was worried about all of you out there needing a Flickers of Freedom fix. So I hope this helped a little.

I cannot really reply to all of the interesting points brought up, and I am sorry about that. But I will quickly respond to a few of the points at which people directly addressed my proposal.

First, Eddy, the point about the ability at issue being unique to persons was meant to capture the idea that only beings of a certain sort are candidates for the freedom at issue, and here I assume that the relevant sort is captured by the concept of personhood. Think, for example, of Frankfurt's remark in his concept of a person paper that only of persons can we think that the freedom of the will could be a problem. This seems correct to me. And, I might add, it seems consistent with Bruce's inclination to look further down the rungs in the evolutionary ladder for critters that aspire to the more elaborate kinds of features we persons happen to have. (We can do this in terms of abilities and capacities falling shy of but coming close to the ones we possess and are interested in examining for, as Alan suggested, pragmatic reasons.)

Justin, I take it as an *advantage* of my proposal that it is an open, substantive question whether free will requires (is) the ability to do otherwise, whereas on your preferred definition, it is a simple analytic point. And it might well turn out that we get an extensional equivalence on my proposal, as you observe. But as I know you are aware, that is not itself decisive reason to think we have located sameness of meaning. Also, note that you seem to assume here in your claim about extensional equivalence that whatever this sort of freedom comes to, even if it requires the ability to do otherwise, it is simply *exhausted* in the ability to do otherwise. But, even granting that free will requires the ability to do otherwise, mightn't there be further conditions for this contested sort of freedom other than a condition of the ability to do otherwise? What if one such condition is, say, also being suitably responsive to reasons, as Fischer might have it, or being at least able to identify with the will upon which you act, as Frankfurt might insist?

Joe, I confess, I do find your lean version appealing, and it shares affinities with Kane's basic approach. But as I was understanding my proposal, free will would not entail moral responsibility (MR), as it (FW) is merely one necessary condition for MR. Rather, MR would entail FW. The way I am thinking of it is that the freedom at issue is one that an agent can exercise in non-moral contexts. And it is one that might well be possessed by beings at worlds in which no one is morally responsible for anything. It is just that it is of the sort that is required (in the strongest sense necessary) for moral responsibility at worlds in which there do exist agents who are MR for what they do.

Eddy, I do have some further worries about the relation between x-phi research on how the folk understand the meaning of the expression 'free will' given that in my estimation it *is* a term of art. But maybe we can leave that for a distinct post. (So as not to mislead, my worry is not the dismissive one of rejecting x-phi work in this area generally. It is about whether work on folk understanding of 'free will' will be as informative for us as philosophers as in comparison with concepts like 'deserve' and 'blame' and 'act freely', all of which have fairly familiar folk applications.)

Let me just say, as regards whether 'free will' is best understood as a technical philosophical term, that a survey of many of the influential writers working on this topic treat the expression pretty much as a term of art settled by our discipline's own tradition. Furthermore, there is SO MUCH diversity in how writers use the expression that, it seems to me, contra what Justin had suggested, the waters can only get "unmuddied" if we specify what technical way we wish to use the term, and that we are entitled to some latitude here in our manner of unmuddying, given the actual variability of the way the term is used.

To substantiate this, I'll now follow this comment up with a distinct comment to this thread devoted just to a quick survey of how various philosophers treat (or avoid) the term when working on what they are all willing to call "the free will problem."

Best,
Michael

[Michael McKenna]

Carl Ginet: “By freedom of will is meant freedom of action. I have freedom of action at a time if more than one alternative is then open to me.” (On Action, 1990: 90) [Note in this last sentence Ginet only offers a sufficient condition, whereas the ‘is meant’ suggests a definition.]

Peter van Inwagen: “I use ‘free will’ out of respect for tradition. My use of the term is not meant to imply that I think there is such a ‘faculty’ as ‘the will’. When I say of a man that he ‘has free will’ I mean that very often, if not always, he ‘he has to choose between two or more mutually incompatible courses of action…each of these course of action is such that he can, or is able to, or has it within his power to carry it out…. ‘Free will, then, is to be defined in terms of ‘can’. (An Essay on Free Will, 1983: 8)

Robert Kane: “Free will, in the traditional sense I want to retrieve… is the power of agents to be the ultimate creators (or originators) and sustainers of their own ends or purposes. The notion should be distinguished from freedom of action, and no simply because free will is a power. To act freely is o be unhindered in pursuit of your purposes…; to will freely, in this traditional sense, is to be the ultimate creator (prime mover, so to speak) of your own purposes.” (The Significance of Free Will, 1994: 4)

Randolph Clarke: “Following many other writers on a basic characterization of this freedom [free will], I shall say that when an agent acts freely (or with free will), she is able to do other than she does then.” (Libertarian Accounts of Free Will, 2003: 3)

Timothy O’Connor: “’Free Will’ is a philosophical term of art for a particular sort of capacity of rational agents to choose a course of action from among various alternatives. Which sort is the free will sort is what all the fuss is about… Most philosophers suppose that the concept of free will is very closely connected to the concept of moral responsibility. Acting with free will, on such views, is just to satisfy the metaphysical requirement on being responsible for one's action.” (Stanford Encyclopedia of Philosophy, “Free Will”, 2010 updated version, opening sentences)

Ishtiyaque Haji: “In what follows, when I speak of ‘free action’ the freedom at issue is the freedom that moral responsibility requires… I take free will simply to be the power or ability to act freely.” (Incompatibilism’s Allure, 2009: 18)

Alfred Mele: “A comment is in order about how I use ‘free will’ in this book. Whatever, exactly, free will is, it is, most fundamentally, the power or ability to act freely. So one can try to understand free will by ascertaining what it is to act freely. One can develop an account of free action and define free will as the power or ability to perform actions that satisfy the account. For a combination of reasons, including the following, I would like to think this approach is viable. First, I often cannot tell what authors mean by ‘will’ in ‘free will’; second, I am blameless for this ignorance, as far as I can tell; third, I seem to have been able on various occasions to write about acting freely without using the expression ‘free will’. In any case, if free will may be simply defined in terms of free action—as the power or ability to act freely—one can go about the business of trying to understand free action without worrying about what (the) will is supposed to be. I find that thought liberating.
I close this section with an announcement about my use of ‘free action’ and it cognates… My interest is in what might be called moral-responsibility-level free action—roughly, free action of such a kind that if all the freedom-independent conditions for moral responsibility for a particular action were satisfied without that sufficing for the agent’s being morally responsible for it, the addition of the actions being free to this set of conditions would entail that he is morally responsible for it.” (Free Will and Luck, 2006: 17)

Harry Frankfurt: “Now freedom of action is (roughly, at least) the freedom to do what one wants to do. Analogously, then, the statement that a person enjoys freedom of the will means (also roughly) that he is free to want what he wants to want.” (“Freedom on the Will and the Concept of a Person,” section III, paragraph 6) AND “A person’s will is free only if he is free to have the will he wants to have. .. But although this question [how to understand ‘could have done otherwise’ in pertinent philosophical contexts] is important to the theory of freedom it has no bearing on the theory of moral responsibility…. It is a mistake… to believe that that someone acts freely only when he is free to do whatever he wants or that he acts of his own free will only if his will is free.” (same article, section IV, paragraphs 7&8) [Note that, like Kane, and unlike several others quoted here, Frankfurt thinks it important to preserve the distinction between ‘free’ as modifying action and ‘free’ as modifying will. Note also that earlier on, when he explains that it means to say a person enjoys freedom of the will, he qualifies with ‘(roughly)’ and then later in the essay, we find more, precise, technical usages (suggesting terms of art) ‘acting of one’s own free will’ which does not require the ability to do otherwise, and acting with ‘freedom of the will’ which does.]

Derk Pereboom: He offers no definition of free will in his 2001 book, but in response to this email question from me, “Do you have a definition of how you use 'free will' anywhere in print? If so, where, and what is it? If not, I take it that it would be accurate to say that you write of it in terms of the "freedom of the sort required for moral responsibility" 2001, p.xxii, where this is in turn understood in the basic desert-entailing sense.” Pereboom replies: “Yes, that's right -- in Living without Free Will, and everything since then, I argue that the notion of free will at issue in the historical debate is the sort of freedom required for moral responsibility in the basic desert sense -- and I tie this to my definition of basic desert. I never claim that this is the only notion of free will we have.”

John Martin Fischer (from his 1994 book) and Fischer with Mark Ravizza (from their 1998 book): In neither book does the expression ‘free will’ occur in the index, and one of the two books is titled, *The Metaphysics of Free Will*. In neither is there any working definition of ‘free will’ given. In the 1994 book, after reporting of his book “It is a book on free will” (1994:3), Fischer then writes “It is generally thought that a person must possess (among other things) a certain sort of control” (1994:3), and the mention of ‘free will’ pretty much evaporates; the discussion is framed in terms of control. Later in the book we find a distinction between ‘guidance control’ and ‘regulative control’ where the former does not involve the ability to do otherwise and the latter does. This occurs again in 1998, with Ravizza, and the control condition for moral responsibility is motivated by reference to Aristotle from Nicomachean Ethics (1998: 12-13).

Kevin Timpe: “…in what follows I will use the term ‘free will’ to refer to the kind of control required for moral responsibility rather than equating it with one or other of these two free will conceptions.” (Free Will: Sourcehood and Its Alternatives, 2008: 12) The two other conceptions Timpe is referring to are i.) free will as having the ability to do otherwise; and ii.) free will as being the ultimate source of one’s actions (2008: 11)—roughly van Inwagen’s and Kane’s proposals.

[Joe Campbell]

Thanks Michael! I expressed my point incorrectly.

The point is: On your view it is analytic that moral responsibility requires free will. But, following some points made by Justin, one might argue that historically the phrase "free will" picks out the ability to do otherwise (I held this view at one time) BUT the ability to do otherwise is not necessary for moral responsibility. This is a substantive debate and should not be folded into the definition of "free will."

[Justin Capes]

Michael: You're right. My earlier remarks did suggest that all there is to free will is "could have done otherwise." I should have been more careful. I don't think that's all there is to it. That's why I said later that I take free will to at least entail "could have done otherwise."

I agree, of course, that people use "free will" in lots of different ways. My reason for thinking that free will has historically been used in a way that at least entails "could have done otherwise" is that many of the various controversies about free will are hard to make sense of otherwise. Why, for instance, should anyone have thought that divine foreknowledge threatened human free will unless they assumed that free will required the ability to do otherwise?

Also, like Joe, I think it is a substantive issue whether MR requires FW.

But perhaps there are just several different conceptions of free will. There is the "up to you" view (e.g., van Inwagen), the "source" view (e.g., Kane), and the "control-required-for-MR" view (e.g, McKenna). Perhaps there are others. And perhaps it's fine to use "free will" in all of these ways, as long as you're clear about how your using it.

[John Fischer]

Clearly, people use "Free Will" in different ways; it is helpful to offer a catalogue of some of these ways. I don't think there is a fact of the matter about how this term *should* be used.

I have tended to think of it as purporting to pick out the freedom or control required (on most views) for moral responsibility. Of course, on some views, such as Angie Smith's "Rational Relations" view, and others, no sort of freedom or control is required. But on many views, *some* sort of freedom or control is required for moral responsibility. It is contentious what kind of freedom or control this is, and what its conditions are. I have tended to use the term, "Free Will", on those rare occasions when I've used it (except in book titles!), to purport to pick out the freedom or control required for moral responsibility, whatever it turns out to be (and however it is to be analyzed).

I think it would be Procrustean to suppose that there is one correct way of using "Free Will". I simply commend this way. Also, it has the virtue of not presupposing any particular account of the relevant kind of freedom, and not presupposing any sort of faculty picture of the Will (although it is compatible with such a picture).

[John Fischer]

Bruce,

You are always so nice and amazingly generous that it is REALLY hard to disagree or take issue with you. (LET THIS BE A LESSON TO OTHERS... HA/HA)

Well, there are forms of freedom or control that are genuine and the possession of which would *add* to one's overall freedom, but which are not necessary for moral responsibility: regulative control, for instance. It is cool to have, but (arguably, at least) not necessary for moral responsibility.

I *do* think that guidance control is necessary for moral responsibility. But I agree that it is not sufficient: we need to satisfy an epistemic condition as well (and guidance control is construed via the Tracing Condition).

Does one need more in the way of freedom? I don't think so. We do need a more robust kind of freedom for AUTONOMY, but it is important to distinguish autonomy from moral responsibility, and guidance control, I would argue, is all the freedom we need for moral responsiblity. But, argh, I'm not betting this will convince you...

[V. Alan White]

Following on Joe's observations about disagreements on what matters in using the term "free will", I reiterate from my post above that what constitutes proper use of the term must (like "death") (i) pick out something of an empirical or metaphysical sort for reference (substantial capacities for reason or dual-ability, for instance) and (ii) attach value to that reference which allows the term to coordinate with relevant larger issues of a moral nature (and which have existential merit positively like Kane or negatively like Pereboom). I can't rightly assess if this translates into "free will" being a term of art, but if it is sufficiently like the way we use "death", its importance in making real-life moral judgments should weigh in favor of a similar regard for how we use "death"--which has real medical and legal consequences, and thus indicates that it at least is more than a term of art.

[Paul Torek]

This thread strikes me as an invitation to metaphilosophy. Here's some of mine.

Psychologically, whatever else they may also have or do, concepts have a prototype structure. Each one has features that are strongly associated and others that are less so. An item (a particular penguin, say) can lack an important feature (flight) and still fall under the concept (bird), but he is a bit less birdy than a sparrow.

As we learn more about the world, we often succeed in pinpointing a structure - a particular branch of the evolutionary tree, say - which explains enough* birdy features of all* the birds to allow us to wrap the whole concept into a nice tight little package. (*Asterisk: not necessarily quite all of what we had thought were birds, and not necessarily all of the features.) If we can pull off such a feat (and sometimes we can't), by all means we should. In that case we can discover necessary and sufficient conditions for bird-hood.

To attempt that feat for "free will", we need to test hypotheses, looking into the implications of various proposed definitions. Can the phenomenology of agency help? It's certainly worth a try. Can X-Phi help? It's certainly worth a try. Evidence has to come from somewhere.

Might FW be definable as the unique ability of persons to exercise control over their actions in the strongest manner necessary for moral responsibility? It's certainly worth a try. Might it be definable as up-to-usness? It's certainly worth a try. It's also certainly worth noting that these definitions need not be competitors: there could be a lawlike coextensiveness between them.

Even terms of art have prototype-structured concepts. Consider how "mass" fared in the transition from Newtonian to Relativistic physics. At least one of the features that had been explicitly tied to "mass" - the idea that it was an intrinsic feature of each object - was given up. So why do we still have "mass"? Or why isn't rest mass just called "mass" and what we now call mass called something else? Because F still equals m times a, among other things; and those features are more important. A select group of scholars can coin a term, but they are still human, and still think in prototypes. A surprising discovery can knock out a feature or two that has traditionally been associated with such a concept. And yet the concept may survive.

One substantive comment: V. Alan, I really liked your analogy.

[V. Alan White]

Thank you so much Paul. Michael's OP inspired my rethinking of Double's skepticism; I taught Veatch's concept of death this week in my bioethics class and it made me think about similar metaphilosophical issues in the FW debate.

[Eddy Nahmias]

Alan, I like the analogy too, but now how are we to understand the term: "The Death of Free Will"??

Putting together some of the ideas mentioned above, I think another nice definition of "free will" is "the abilities that allow agents to *make choices* that are *up to* them and to *control* their actions accordingly."

But since animals and children can make choices that are up to them and control themselves accordingly, we are still going to want to amp up the senses of choice, up to, and control to whatever levels are required to justify reactive attitudes and attributions of desert-level responsibility. In my view, such amping up is not going to involve new metaphysical powers but instead uniquely well-developed cognitive capacities (e.g., for reflection and conscious deliberation).

[Randolph Clarke]

It's an interesting question what it takes in the way of control of one's behavior for one to be morally responsible for what one does.

It's an interesting question what it takes for it to be the case (as I think it commonly seems to us to be the case) that often it's up to oneself whether one wills to do a certain thing (where deciding to A counts as willing to A -- as setting one's will, as Kane puts it).

It's an interesting question how these two issues are related.

It doesn't strike me as a substantive question which of these first two phenomena should be called free will (given that there's some controversy over whether they come to exactly the same thing).

[V. Alan White]

Isn't "control" thus contextually normative as is "free will"? So a billiard ball causally controls what the second ball it hits does, and with minimal axiological "impact" (unless in further context of a bet placed on the behavior of the second ball). But causal control involving neurological patterns associated with mental states might make such mental states more directly related with those causal outcomes by stipulated value association of the two: this might constitute (for example) guidance control. And that is a value attached to such association as normative for futher concerns (as being responsible for resultant acts, or, if the mental states are deemed somehow inappropriate, not responsible for the same). But this all fits the metaphilosophical pattern I present above. So "control" is just another instance of (i) cited interesting empirical or metaphysical conditions of the unfolding of events (whether causal or not) and (ii) the stipulated value of that for further moral reasoning by assigning exactly what that value is in some positive or negative way.

Is this descriptive metaphysics? I'd say no--the explicit value-addition at crucial points of the description makes it at least pragmatic metaphysics. And distinct from metaphysical or axiological skepticism.

[Robert Allen]

Free will is, as Sts. Augustine and Anselm thought, the decision making faculty's ability to maintain its rectitude for its own sake. It involves being dually disposed- willing happiness as well as rectitude- and the efficient cause of one's "choices." (It thus entails PAP satisfaction and what Gary Watson aptly calls "self-determination.") There is a normative or teleological element here as well: the latter disposition is supposed to be given pride of place should both be manifested. To will freely is to achieve a certain goal- rectitude maintenance for its own sake.

The term FW is Catholic in origin and arose in the context of discussions of the Last Judgment. Thus, if you are talking about FW, then you are discussing something required to make sense of that notion. Now I guess you can water down the above definition, a la Hobbes, Hume, Fischer, and Frankfurt, and then use the term FW to justify all sorts of other interesting and important practices. The "revisionists" Perry and Vargas recommend doing precisely that, having no philosophical qualms about changing the subject. (Perry: "Augustine has seriously misled us about FW." In correspondence.) But, for whatever it’s worth, this philosopher at least would be intellectually very disappointed by such a tack.

[V. Alan White]

But Robert (if I may) such "revisionists" as you call them therefore demur on what is really important as to what features in the world pick out FW. You have declared your world-view values. That moves the question to the level of world-views as to the justification of such. Good luck to all of us on that.

[Robert Allen]

Alan (if I may),

They call themselves 'revisionists' and Perry, at least, hardly "demurs" when it comes to defining free agency (eschewing altogether the term 'FW', for the reason given above) favoring a Humean approach. (Cf. “Wretched Subterfuge: A Defense of the Compatibilism of Freedom and Natural Causation,” Proceedings and Addresses of the American Philosophical Association, volume 84:2, November 2010, pp. 93-113. It seems that Professors Perry and Vargas had a meeting of the minds several years ago when the latter traveled from the lovely environs of USF to the equally gorgeous campus of Stanford University to sit in on the former's free will seminar.) The term 'free' on his view picks out the property possessed by agents who have opportunities of which they know how to take advantage. In his example, I am free to buy whole milk instead of skim (satisfying PAP) because whole is also stocked and I'm capable of putting a gallon of it in my cart. Now I WON'T buy the whole milk, because I'm in the habit of pleasing my wife who has requested skim, but nevertheless I can and, thus, freely (so as to be responsible and slightly praiseworthy) purchase the skim.

I'm just saying that if we want to discuss FW in the original sense of that term then we must follow the medievalists. Even some secularists acknowledge as much. They heuristically use the notion of a Final Judgment, maintaining that FW exists only if we possess a power that would justify God, were He to exist, in consigning human beings to Heaven and Hell, i.e., REALLY holding us responsible for the things we do and fail to do. McKenna seems to be saying something along those lines.

Thank you for your response.

[Randolph Clarke]

Micheal Frede's A FREE WILL: ORIGINS OF THE NOTION IN ANCIENT THOUGHT traces the idea to Stoics, and prior to Christianity. And their idea had more to do with following reason than with being able to do otherwise.

In any case, words change in meaning, and etymology is no good guide to philosophical analysis.

'Free will' has been used to talk about a variety of things. You can use it without fault for any of them. Just tell us which one you're using it for.

[Robert Allen]

The term 'will' is a part of the locution so it would be inappropriate to use FW to talk about a phenomenon that is not subjective. If folks want to talk about free agency or freedom in action, then they should abjure the term free will, which presupposes the medieval notion of the mind as a system of "faculties," one of which is the will.

It should surprise no one here that Augustine and Anselm drew on the work of earlier thinkers- philosophers, as I tell my students, don't work in a historical vacuum, their claims of originality notwithstanding. (I tend to agree with Whitehead's assessment of our work.) The titles of their works, however, "popularized" the term free will. Those works themselves then became touchstones for free will theorists throughout the Middle Ages and beyond. That is quite a bit of linguistic baggage, Professor Clarke, more than I would care to slough off in my use of a term.

[Joe Campbell]

There are descriptive and normative questions. Which definition deserves the name “free will” given usage or historical usage (or some other criteria used to settle the expression’s meaning)? Isn’t this a purely descriptive question? We might ask a normative question, like the one that Randy doesn’t care about: Which definition of “free will” should one use?

I want to make the case that this isn’t a silly question. But I want to make clear what I am not claiming first, and the sense in which I agree with John and Randy that this is a silly question.

I’m not saying that everyone should always use a particular definition of “free will” in, say, their academic writings about free will. That would be silly. Even if there is a single definition that is the best definition, one might still use the term in some different way, as long as one stipulates what one means by the term. What is wonderful about John, and Randy, and Eddy, and frankly most if not all of us is that though we may use the term “free will” differently, we make it pretty clear beforehand what we mean by the term. I’ve defined it in several different ways, but always remained consistent within the given work. You can’t use the term “free will” in any old way, but there is a range of ways in which you could use the term in academic writings, as long as you make things clear.

Nor do I want to say that the definition of “free will” that should be used is determined only by the descriptive data: empirical research on actual usage, historical usage, etc. Randy’s report of Michael Frede’s account – which traces the term by to the Stoics, generally regarded precursors to the source theory (Derk takes this line, I think) – is important. It rules out certain popular definitions of free will but that allow won’t tell us what definition we should adopt.

Choosing the definition of “free will” that we should adopt is more like choosing the scientific theory we should accept. Sure the empirical data is important but it isn’t the whole story. We want the definition to embrace theoretical virtues like simplicity and fecundity. Maybe that is what Al’s initial comment is getting at.

With all this in mind, I’ve tweaked Eddy’s definition: Free will is the most fundamental ability, or set of abilities, that allows an agent to perform an action that is genuinely up to the agent.

[Randolph Clarke]

Not exactly the TERM 'free will' that Frede traces to Stoics. For that matter, I doubt we'll find the expression in Augustine or Anselm. The first use noted by the OED is from the 13th century.

[Robert Allen]

Randy,

I checked the OED myself and was surprised because of the titles of the 2 As works: On Free Choice of the Will and On Free Will. I'm guessing that they don't count because they were not written in English? In any event, wouldn't you concede that the debate over FW begins in earnest during the Middle Ages? When Hobbes provided his famous redefinition, wasn't it the notions of the medievalists at which he was taking aim, the ones that couldn't be true of us were Determinism true? I have always read him that way.

Joe,

How can it be a good thing that one term is used with lots of different meanings? Doesn't this practice invite the very sort of confusion that Frege admonished us to remove from our notations? Can't we at least take Professor Kane's advice and reserve 'FW' for a subjective phenomenon, using 'free agency' when we want to talk about bodily movements or activities that meet a certain standard? Professor Perry, a revisionist, takes this approach. After all, the term 'will', mental in connotation, is a part of 'FW'.

[Randolph Clarke]

Robert, I'm not sure what to say about when the debate really begins, but I suppose I'd trace it to the first philosophical discussions. Not really sure when these took place.

[Robert Allen]

Randy:

The history of that discussion is related here:

http://www.newadvent.org/cathen/06259a.htm

It is basically as a I said: an attempt by Catholic philosophers such as Augustine and Anselm to square the belief in the moral freedom required to justify the Final Judgment with the doctrines of divine omniscience and grace.

[Eddy Nahmias]

Robert, suppose it turns out that most people (including most philosophers!) no longer use "free will" to refer to "whatever capacities would justify final judgment" [and perhaps help make sense of the problem of evil?], but instead use it to refer to, e.g., "whatever capacities for choice and action allow an agent's actions to be up to her in such a way that she can deserve credit or blame." If so (and assuming those two definitions are not synonymous), then why should we use (and theorize) about "free will" as used by the medievalists?

[Randolph Clarke]

Robert, I'm not sure why you discount the discussions of free will that pre-date Christianity. Anyway, I don't.

[Robert Allen]

Eddy,

I don't have a problem with your definition ito "capacities" (although it is sketchy, contains a promissory note, compared to Anselm's exposition in which those capacities are spelled out). I also think that the medieval notion and yours ARE virtually synonomous. As I noted above, many secularists define FW a ala the medievals, saying it is that freedom that would justify eternal reward or punishment, were such things to exist.

My beef is with those folks who would use 'FW' to designate something extra-mental, having to do with behavior: let's say Hobbes had not spoke of liberty but FW or Professor Perry had not abjured 'FW' in his Humean analysis in favor of 'free agency'. I like discussions of FW to focus on something that goes on in a person's mind around the time at which he is contemplating and deciding upon various courses of action. They miss the mark if they are considering the activity that ensues subsequent to the taking of a decision.

But thank you very much for your response. It is always good to hear from you.

[Robert Allen]

Far be it from me to discount those discussions, Randy; Frede's essay is now on my reading list. I just think that the debate really heats up in the MA, due to the above mentioned religious concerns. For Augustine et al the matter wasn't merely philosophical but intimately related to their most pressing real life concern. Beginning with OFCW and running through the work of Anselm, Scotus, and Aquinas, you have a systematic exploration of a single notion: they are all tweaking the same idea. It is my sense that when historians and early modern philosophers referred to FW it was with their discussion in mind, which is why Frede's research is groundbreaking. No doubt, then, that "notions," of FW to use Frede's apt term, existed prior to the MA, but it was the medievals, I believe, who assembled those ideas, along with others, in a nice, neat philosophical package.

Thanks again for your continued interest in this topic, the only game in town until Mike's tenure officially begins and MLB starts up again.

[Alan White]

Robert I certainly understand your perspective. The tradition of western law has certainly supported your point of view with its focus on mens rea. Commentaries on the role of free will qua mental dual-able capacity have peppered legal writings for centuries. But concepts surely evolve as much as anything else in the universe (except maybe the laws of physics, and there's controversy even about that). My remarks about how we think about death were made to make just that point: concepts are only as good as facts and values inform them, and "free will" surely is a concept thus doubly informed, and subject to the evolution of that interplay. I've said this before but it bears repeating: when the Feds adopted the Model Code in the 60s for its jurisdictional adjudication of insanity, it included two questions related to mens rea. The first involved the substantial capacity to appreciate wrongfulness, and the second a similar capacity to conform behavior to the law. That second question was the so-called voluntary prong--FW much as you deem it rightly used to determine responsibility (though certainly also not inconsistent with compatibilist views). Either question could exculpate, making FW (of some sort) a full partner in determining responsibility.

Then John Hinckley was found not guilty, and the law changed to reflect society's horror.

Now the Feds have dropped the voluntary prong entirely along with other reforms. Many states followed suit as well.

Four states abandoned the insanity defense entirely--that is a complete step away from associating guilt with mens rea.

There has been no discernible outrage about all this. Legally questions of FW have been increasingly swept into a dustbin of law cleansed of such concerns. In some way this reflects an evolution of values that embraces questions of action over intention in assessing responsibility. But then recall that evolution need not entail progress--just change. But then that also brings us right back to the matter of world-views in evaluating such a thorny meta-issue as what constitutes possible progress.

Ultimately I think world-views clash when we discuss FW issues, and maybe it'd be best if our discussions moved there. But since facts and values determine the very meanings of our terms like FW, we probably need a complex reflective equilibrium that tries to fit particular claims about facts and values in the real world within a larger rational picture of what that real world is.

[Bruce Waller]

Now is the winter of our discontent made glorious summer by this sun of Tuscon; many thanks to Michael for starting this discussion, which has carried me through the stubborn season and on into spring, eagerly awaiting the “official” McKenna discussion which will start in little more than a week. This has been a great thread, and I’ve learned a great deal. Just a couple of things. Again, thanks to Randy for recommending the Frede book, as he did in an earlier post; for anyone interested in free will – and that must cover just about everyone who ever visits Flickers – it is a very rich resource. On Robert’s comments on when the free will debate “really heats up,” certainly it gets quite warm with Augustine and Aquinas; but it seems to really burst into flames with Valla, who influenced Luther, and then Pico proposed his heretical solution; and because this was a major point of contention between Protestants and Catholics, the flames – sometimes literally – burned hotter. Alan, your post was extremely thought-provoking; had never really thought about all these issues in connection, the way you pulled them together. I remember the post-Hinckley period quite well, and the political/legal reaction to the not guilty verdict. I was living in North Carolina, teaching at Elon at the time, and one of my friends on the faculty there was a state senator. We were having lunch, and I had just read in the Raleigh paper that the state legislature was considering a change in the state law regarding insanity verdicts: If a person who committed a crime was determined to be not responsible by reason of insanity, then the person would be confined in a mental institution for the criminally insane until the person was ruled sane; at that time, the person would be transferred to prison, to begin serving his or her full sentence for the crime committed. I brought this up with my friend, thinking it was one of the worst ideas I had ever heard – the person is not responsible because of insanity, but if the person ever becomes sane then we’ll punish them severely – and that my senator friend would agree; to my amazement, he thought it was a great idea, and he intended to vote for it, and all my arguments against it were to no avail. In retrospect – and with the help of Alan’s comments – this now seems to me a sort of “natural experiment” that confirms the experimental results reported by Doris and Knobe and by Nichols and Knobe, on variations in beliefs about moral responsibility depending on whether one is given a theoretical or “actual” case to consider: when considered in the abstract, most people (including most legislators) don’t hold people morally responsible when they do things outside their rational control; but when they deal with a dramatic case (as Hinckley surely was) they insist on moral responsibility, no matter what.

[Alan White]

Bruce--thanks so much for your kind remarks and that rather chilling story. I agree that all these post-Hinckley "reforms" were various knee-jerk reactions rather than careful deliberations--shame on us as a so-called 1st-world educated society for standing by and letting such philosophical debacles happen. I really like your description of all this as a "natural experiment" and connecting it with X-Phi studies. I'd never thought of it that way!