I would now like to turn to another important, but often overlooked, aspect of the free will debate: the role of reference. Shaun Nichols, Manuel Vargas, and Oisín Deery have done some amazing work on this subject, and it’s Shaun’s article in my edited collection that got me thinking seriously about reference. This post is based off of my recent Philosophical Studies paper (DOI 10.1007/s11098-015-0453-x), in which I argue for free will eliminativism in the form of a response to Shaun Nichols. I will summarize my argument below and see if I can convince any of you that free will eliminativism should follow regardless of whether we adopt a descriptive account of reference or a causal-historical approach.
Shaun Nichols* has recently argued that while the folk notion of free will is associated with error, a question still remains whether the concept of free will should be eliminated or preserved. He maintains that like other eliminativist arguments in philosophy, arguments that free will is an illusion seem to depend on substantive assumptions about reference. According to free will eliminativists, people have deeply mistaken beliefs about free will and this entails that free will does not exist (e.g., Pereboom, Strawson). However, an alternative reaction is that free will does exist, we just have some deeply mistaken beliefs about it (e.g., Vargas). According to Nichols, all such debates boil down to whether or not the erroneous folk term in question successfully refers or not. Since Nichols adopts the view that reference is systematically ambiguous, he maintains that in some contexts it’s appropriate to take a restrictivist view about whether a term embedded in a false theory refers, while in other contexts it’s appropriate to take a liberal view about whether a token of the very same term refers. This, according to Nichols, affords the possibility of saying that the sentence “free will exists” is false in some contexts and true in others. Hence, according to Nichols, even if we acknowledge that our everyday conception of free will is significantly in error, there is flexibility in whether we embrace the eliminativist claim.
I would like to argue that even if we grant Nichols his pluralistic approach to reference, there is still good reason to prefer eliminativism to preservationism with regard to free will. My argument focuses on one important difference between the concept of “free will” and other theoretical terms embedded in false theories—i.e., the role that the phenomenology of free agency plays in reference fixing. I argue that once we acknowledge the importance of phenomenology, eliminativism appears to follow on whichever account of reference one adopts.
I will begin by briefly outline Nichols’s account of free will and error. Then I will argue that on either a descriptive account of reference or a causal-historical one—the two main accounts of reference Nichols discusses—eliminativism appears to follow. On a descriptive account of reference, eliminativism follows since, as Nichols himself argues, the folk concept of free will contains significant error, hence nothing satisfies the description. Yet, on a causal-historical account of reference, despite Nichols’s claims to the contrary, eliminativism also seems to follow since it is prima facie plausible to think that the concept of “free will” was originally baptized in a causal-historical story which appealed, at least in part, to our first-person experience of free agency. And since Nichols and I have both argued elsewhere that the phenomenology of free agency is best viewed in a libertarian and incompatibilist sense (see Caruso 2012; Deery, Bedke, and Nichols 2013), one that is likely in error, eliminativism awaits us down this path too.
Nichols’s argument begins with the assumption that the ordinary notion of free will is in error because it falsely presupposes that indeterminism is true. He maintains that:
A growing body of empirical evidence indicates that people think that their choices aren’t determined (Nichols and Knobe 2007; Sarkissian et al. 2010). It is not just that they don’t have the belief that their choices are determined. Rather, they positively think that their choices are not determined. And this belief is implicated in their thoughts about free will. For instance, when presented with a description of a deterministic universe, most participants say that in that universe, people don’t have free will (Roskies and Nichols 2008; also Deery et al. forthcoming). This provides reason to think that the everyday conception of free will is not compatible with determinism (see also Rose and Nichols forthcoming).
If we combine these empirical findings with the assumption that determinism is true and that we do not make indeterminist (nonrandom) choices, we get the conclusion that the ordinary notion of free will is in error.
While these assumptions are, of course, highly contested, I will grant them for the sake of this post, since the issue that Nichols and I are both interested in concerns the consequences of folk error on free will. Since “the issue can only be joined if we assume that there is something interestingly mistaken about the folk view,” I will join Nichols in assuming that the ordinary notion of free will is in error. (I apologize in advance to Eddy and others here!!) Now, one may think that the deck has been so thoroughly stacked against free will that it trivially follows that it does not exist, but Nichols disagrees. It is here that he introduces his reference pluralism and his argument against free will eliminativism.
Nichols correctly points out that it is not always clear that we should adopt eliminativism for theoretical terms embedded in false theories. If contemporary philosophical debates over the last several decades have taught us anything—including debates in ethics over error theory, in social-political philosophy over the existence of race, in philosophy of science over scientific realism, and in philosophy of mind over eliminative materialism—they have taught us that wherever you find one group of philosophers advocating eliminativism, you will usually find another group of philosophers advocating revisionism or preservationism. “In each of these debates,” Nichols notes, “eliminativists maintain that K doesn’t exist (where K might be morality, race, belief, etc.). Typically shortly after an eliminativist claim of this sort is made another group of philosophers adopt a preservationist position. In effect, they say, Ks aren’t what we thought they were”(Nichols 2013, 204). According to Nichols, then, the fact that the ordinary notion of free will has a false presupposition does not immediately entail that there is no free will.
It’s at this point that Nichols discusses the role that reference plays in such disputes. The key distinction is between theories of reference that are liberal and those that are conservative. For brevity and ease of discussion, I will follow Nichols in focusing on causal-historical accounts as the key example of a liberal theory, and a restrictive descriptivism as the key example of a conservative theory. To see how these different theories of reference can play out in such disputes, consider, for example, the disagreement between Stephen Stich and William Lycan over eliminative materialism.
Stich famously defends a version of eliminative materialism, maintaining that concepts like “belief” are part of a folk psychological theory that is massively mistaken. As Nichols describes Stich’s argument:
[Stich] begins by setting out David Lewis’ descriptivist account of how theoretical terms get their meaning and reference from the theory in which they are embedded. A theoretical term refers to whatever object or class of objects satisfies some critical set of claims in the theory. Stich then argues that the folk psychological theory that gives meaning and reference to “belief” is deeply erroneous; from this he concludes that beliefs don’t exist.
Lycan, on the other hand, rejects restrictive descriptivism (and with it eliminativism about belief), adopting instead a “liberal” view of reference fixing. Lycan writes:
I am at pains to advocate a very liberal view . . . I am entirely willing to give up fairly large chunks of our commonsensical or platitudinous theory of belief or of desire (or of almost anything else) and decide that we were just wrong about a lot of things, without drawing the inference that we are no longer talking about belief or desire. To put the matter crudely, I incline away from Lewis’s Carnapian and/or Rylean cluster theory of the reference of theoretical terms, and toward Putnam’s . . . causal-historical theory.
On Lycan’s approach to reference fixing, the reference of a theoretical term can succeed even if the theory is radically mistaken. On this liberal/causal-historical approach, the reference of a theoretical term is the entity or kind that was “baptized” when the term was introduced. Since the term continues to refer to the entity or kind that was baptized at the end of the causal-historical chain of transmission, “people can have massive misconceptions about the objects that their terms (or concepts) refer to.”
Other examples of disputes between liberal and conservative approaches to reference can be found with regard to the concepts of “whales,” “witches,” and “phlogiston.” Each of these concepts was originally enmeshed in significant error, yet some were preserved (e.g., whales) while others were eliminated (e.g., witches and phlogiston). The folksy concept “whale,” for example, falsely presupposed that whales were fish. Instead of eliminating the concept, however, a liberal approach to reference was employed and taxonomy kept the label and adjusted the concept, moving WHALE onto the mammal branch. On the other hand, “witches” and “phlogiston” were eliminated because a more restrictive descriptivism was adopted.
Nichols argues that we can view the free will debate, especially the debate over eliminativism, as similarly a debate over which reference-fixing convention we should employ. Free will eliminativists often employ a descriptivist convention for reference, according to which a natural kind term refers by means of an associated description. Preservationists, on the other hand, maintain a causal-historical convention for reference, according to which an initial baptism sets the referent of a term, even if the nature of the object referred to be misunderstood at the time of baptism. Hence, the eliminativist maintains that since the descriptive nature of the term yields a failure of reference, there is no such thing as free will. Preservationists (including revisionists) argue instead that there is free will it’s just that “everyone has merely been under some misapprehension about its nature” (Nichols 2013). How, then, should we resolve this dispute?
It is here that Nichols argues for the systematic ambiguity of reference and a pluralistic approach to free will. He introduces recent experimental evidence that suggests that speakers regularly use different referential conventions for different kinds of sentences in which natural kind terms appear. Since Nichols concludes that reference is systematically ambiguous, he maintains that when it comes to free will it is sometimes correct to say, “Free will does not exist,” since people employ the descriptivist convention in sentences like this one. Nevertheless, it is also correct to say, “Free will isn’t what we thought,” since a causal-historical convention of reference can be employed for sentences of this type. With all the pieces in place, Nichols concludes that both sides can agree that libertarian free will does not exist and the everyday conception of free will is significantly in error, without precluding the possibility of preservationism—e.g., there is a flexibility in whether we embrace the eliminativist claim.
My reply to Nichols will concede four of his main assumptions. I am willing to grant for the sake of argument that (i) the concept of “free will” is enmeshed in significant error, (ii) the free will debate depends on substantive assumptions about reference, (iii) not all theoretical terms embedded in false theories should be eliminated, and (iv) reference is systematically ambiguous. I maintain that even granting these assumptions a strong case can be made for free will eliminativism. My argument will focus on the causal-historical approach to reference fixing since this is the one typically employed by preservationists. But before discussing this liberal approach to reference fixing, let me just say a quick word about descriptivist accounts.
It’s pretty clear that on Nichols’s own assumptions the path of restrictive descriptivism leads to eliminativism. Given assumption (i) above, the assumption that the concept of “free will” is enmeshed in significant error, it follows rather directly that if we adopt a descriptivist approach to theoretical terms (according to which the meaning and reference of a theoretical term is fixed by some critical set of claims within a given theory), free will eliminativism follows since the term yields a failure of reference. Some philosophers will no doubt resist this conclusion by questioning assumption (i), but seeing as though Nichols appears to concede that this path leads to eliminativism, and since I am granting assumption (i) for the sake of argument, I will focus my attention on the second main path.
Can we avoid eliminativism if we adopt a causal-historical approach? I think it depends on how one conceives of the original baptism. It’s rather easy to imagine how concepts like “whale” and “water” were initially baptized. On a causal-historical account, the term “whale” can continue to refer despite the fact that a key component of the folk description of whales (i.e., that they were fish) turned out to be wrong, since the initial baptism most likely took place in a demonstrative way. But free will seems significantly different than whales. The same reference-fixing properties are not available for free will. How, then, are we to conceive of the initial baptism that set the referent of “free will”? Nichols is unfortunately silent on this point, but I believe there are several leading candidates. It is possible, for example, that the initial baptism was to whatever power or ability is required to justify ascriptions of desert-based moral responsibility, or to that feature of choice and action that justifies our reactive attitudes, or to a set of compatibilist-friendly capacities (e.g., reasons-responsiveness). While I cannot adequately address all these possibilities here (although I will say something about them below), my proposal is that we should look elsewhere, i.e. to the phenomenology of free agency.
I maintain that it is prima facie plausible to think that the concept of “free will” was originally baptized in a causal-historical story that appealed, at least in some significant way, to our first-person experience of free agency. This does not mean that cultural attitudes about moral responsibility played no role in reference fixing, nor does it deny that certain philosophical and religious doctrines were important in shaping our beliefs and attitudes about free will. My suggestion is simply that our first-person experience is more primitive and basic, and hence prima facie plausible as the primary candidate for reference fixing. Imagine, for example, a possible world in which we retain reasons-responsiveness and other compatibilist capacities but do not experience ourselves as acting freely—or worse still, actively experience ourselves as lacking free agency, as (say) patients with anarchic hand syndrome, Tourette’s syndrome, and schizophrenic thought insertion often do. In such a world, it is unlikely that the concept of free will would have ever been introduced, despite the fact that other candidates for reference fixing are still present.
Now one may be tempted to think that this has more to do with the voluntary vs. involuntary distinction—e.g., that free will is baptized in terms of its role in distinguishing voluntary from involuntary behavior, and that my example has more to do with involuntariness than it does with phenomenology—but I do not believe that this is the case. Agents can engage in voluntary actions (actions that are clearly distinct from involuntary actions) while not experiencing a sense of agency or free will. Consider, for example, the long distance truck driver who finds himself at his destination but does not recall how he got there. Or consider Daniel Wegner’s examples of automatisms—i.e. cases where we experience no sense of free agency while performing voluntary actions. Such automatisms not only involve a “lack of the feeling of doing an action but may even go beyond this to include a distinct feeling that we are not doing” (Wegner, 2002: 99). Common examples include automatic writing, Ouija board spelling, the Chevreul pendulum, dowsing, and the phenomenon of ideomotor action.
The fact that the phenomenology of free agency is absent in such cases is so profound that during an automatism “the person may vehemently resist describing the action as consciously or personally caused. It seems to come from somewhere else or at least not from oneself” (2002: 99). Yet despite this phenomenological fact, what’s remarkable about such cases is that they are clearly caused by the agent, are driven by goals and intentions, and involve sophisticated actions and movements (unlike the herky-jerky movements of many involuntary actions). I maintain that if we experienced all of our actions in this way, even if we retained the voluntary/involuntary distinction as a candidate for reference fixing, it is still unlikely that the concept of free will would have ever been introduced.
An additional reason for thinking that the phenomenology of free agency is important comes from the role it has played historically in libertarian arguments. Libertarians have long emphasized our feeling of freedom and our introspective abilities. In fact, many libertarians have suggested that our introspection of the decision-making process, along with our strong feeling of freedom, provides some kind of evidence for the existence of free will. As Ledger Wood describes this common form of reasoning:
Most advocates of the free will doctrine believe that the mind is directly aware of its freedom in the very act of making a decision, and thus that freedom is an immediate datum of our introspective awareness. “I feel myself free, therefore, I am free,” runs the simplest and perhaps the most compelling of the arguments for freedom. (1941: 387)
While I have elsewhere argued against this introspective argument for free will (see Caruso 2012), there is no denying its intuitive appeal. It captures the fact that for many philosophers and ordinary folk, the phenomenology of free agency is the thing to be explained and preserved, often at all costs! In fact, many agent-causal libertarians are willing to bend over backwards to preserve it, for it is our feeling of freedom that is the raison d'être of libertarianism (see, e.g., O’Connor 1995, 196; Taylor 1992, 51; Campbell 1957, 169). Conceiving of the initial baptism in terms of phenomenology therefore has the virtue of acknowledging this long historical tradition as well as its wider appeal among ordinary folk.
A third reason for focusing on the role of phenomenology in the initial baptism is that many of the other candidates for reference fixing seem historically anachronistic. Take reasons-responsiveness for example. A preservationist could argue that when the term “free will” was initially baptized, it was not to some libertarian phenomenology but to a more innocuous set of reasons-responsive capacities. It need not be the case that we recognized these capacities as reasons-responsiveness at the time, only that reference was primarily fixed to these capacities. While it is impossible to rule out this possibility completely, I believe there are reasons to doubt that the reference-fixing properties were so narrowly focused. The concept of free will stretches back to pre-scientific, pre-theoretical times when it was likely that more than just reasons-responsiveness would have been part of the initial baptism.
Reasons-responsiveness is of course essential to our phenomenology as agents, reasoners, and moral beings, but it does not capture in totality our pre-theoretical self-conception as agents. If Nichols’s own work is any indication, people pre-theoretically conceive of themselves as capable of exercising indeterminist free will (2012), as being able to do otherwise in a way that is not easily amendable to conditional analyses (Nichols and Knobe 2007), and as being introspectively aware of the key mental aspects of decision-formation (Kozuch and Nichols 2011). Paul Bloom’s work further reveals that a major component of our folk psychology (at least in the West) is the belief in dualism. According to Bloom, the belief in dualism comes naturally to children and this dualistic worldview persists in adulthood. Preservationists that want to restrict the reference-fixing properties of the initial baptism so as to preserve “free will” are being disingenuous, I maintain, by downplaying or dismissing the anti-scientific components of our pre-theoretical self-conception as agents.
Preservationists, of course, could shift their focus to more external reference-fixing properties, such as the cultural practices associated with punishment, reward, praise, and blame. Perhaps these practices could, in some way, ground an initial baptism. But problems belie this path too. Drawing on Tamler Sommer’s recent work in Relative Justice (2012), there are past and present cultures that feel completely justified in praising, blaming, and punishing people (sometimes severely) without requiring that those people satisfy the control conditions often associated with free will. Sommers, for example, describes how “honor cultures” often condone severely punishing, sometimes even killing, seemingly innocent individuals for the ‘sins’ of their relatives. Non-honor cultures, like our own, consider this fundamentally unfair and unjust. Furthermore, honor cultures shun third-party punishment, whereas third-party punishment is the norm in Western cultures. According to Sommers, these cultural variations are the product of different ways of life resulting from adaptations to local ecological conditions.
If Sommers is correct about the cultural diversity of our moral responsibility practices, then preservationists who propose that the initial baptism was to a natural kind picked out by these practices would be forced to conclude that there are no universal reference-fixing properties for “free will,” only a divergent and culturally relative set of norms and practices. If, for example, individualist cultures like our own require that a control condition be met for ascriptions of desert-based moral responsibility, while other cultures de-emphasize or lack this condition altogether, then there would be no common feature to serve as the reference fixer for “free will.” This lack of a common property (or set of norms and practices) among cultures represents a serious problem for preservationists; especially those who wish to avoid a wholesale relativism about free will.
For the reasons outlined above, I maintain that the phenomenology of free agency remains the best candidate for the initial baptism. If correct, there is good reason to conclude that free will eliminativism follows on both the causal-historical path and the restrictive descriptivist path. On a descriptivist account of reference, eliminativism follows since the folk concept of free will contains significant error, hence nothing satisfies the description. On a causal-historical approach, eliminativism likewise follows if we assume, as Nichols does, that the phenomenology of free will is illusory. Nichols has elsewhere argued, as have I, that our phenomenology is libertarian and incompatibilist in nature (see Deery, Bedke, Nichols 2013; Caruso 2012). But on the assumption that we lack libertarian free will, one of the key assumptions of Nichols’s argument, it turns out that our first-person experience of free will is illusory as well. Hence, if the initial baptism was to the phenomenology of free agency, and that phenomenology is illusory (a conclusion reached on Nichols’s own assumptions), we once again arrive at a failure of reference.
I conclude that on the set of assumptions laid out in this paper, both descriptivist and causal-historical approaches lead to free will eliminativism. This conclusion is, of course, contingent on a set of assumptions, and critics are free to challenge them. But for those who agree with Nichols that the folk notion of free will is significantly in error and that the phenomenology of free will is illusory as well, the dilemma presented above remains a serious one for anyone wishing to avoid free will eliminativism.
*Shaun Nichols (2013). Free Will in Error. In Exploring the Illusion of Free Will and Moral Responsibility (ed. Gregg D. Caruso). Lexington Books.
Lots of very interesting claims in here, Gregg! Just a quick thought:
Suppose you think there is some agentive experience, but it is not libertarian in character. And suppose you think claims about libertarian phenomenology out to be understood in terms of mistaken beliefs about our phenomenology - an inference from the lack of any experience of deterministic causation to the thought that such agentive experience as there is in deciding or acting must be indeterministic. Then what we'd need to explain is the faulty inference, and phenomenology would not be illusory (I tell a similar story about the role of conscious vision in action in a recent paper in Nous, and I'm tempted by the same story here).
Then things change a bit, right? Because then you have to wonder about the background beliefs that make the inference seem plausible, rather than talking about illusory phenomenology. I, for one, am not tempted at all by the inference (perhaps because my background beliefs are so thoroughly non-incompatibilist).
Posted by: Josh | 03/07/2015 at 08:45 AM
Gregg -
Great post! You've clearly articulated some great ideas. I've had the first inklings of similar ideas (in response to Nichols and others), but I did not have the philosophical competence to complete them and articulate them clearly!
Here are some further thoughts:
1. I'm not sure that we should limit the initial baptism to first person agency. Free will also seems involved heavily in third person attributions of freedom and responsibility. Fortunately, for the eliminativist, we know from the psychology literature that these are also filled with errors (if only because of publication bias!).
2. I've mentioned a host of different errors that, prima facie, implicate free will beliefs: the positive illusions, the illusion of control, the fundamental attribution theory, reactance, the just world phenomenon, system justification theory, anthropomorphic bias, etc. I would hesitate to limit the initial baptism in a way that prevented the eliminativist from invoking these other errors to justify his/her view.
3. As I recall, Nichols defends his theory of ambiguity in part by invoking political considerations. For example, we eliminated witches for political reasons - it was more just to kill the concept than to keep burning innocent women at the stake. Nichols draws on research about the bad consequence of eliminativism (e.g., it causes people to cheat) to suggest that we should adopt a preservationist view of free will, at least sometimes. (I apologize in advance if I'm misreading Nichols here.) The problem here is that 1. political considerations can cut both ways (eliminating free will might create benefits in policy, mercy, and justice that offset the problems) and 2. with my epistemic rationalist hat on, I'm not sure that political considerations should play any role in deciding whether X is true.
4. One thing that seems missing to me in this entire discussion is a discussion of common usage. It seems to me that the biggest factor - maybe the only factor - in fixing the meaning of a term is common usage. When most people say X means Y, then the dictionary authors say that X means Y. When most people stop saying that, then the dictionary authors change the definition of X (or delete it). I have trouble understanding how a big debate about the definition of free will can proceed without asking how most people use the term, much less marshaling evidence (such as survey evidence) about how people use the term. I suspect that, if we do this thoroughly, we would find that the term "free will" has a lot of vagueness, and is probably too vague to settle the big philosophical questions, even though compatibilists and eliminativists both make great points and insights about human nature, psychology, and responsibility - and I have always found the eliminativist points more interesting and attractive.
Posted by: Kip | 03/07/2015 at 10:11 AM
Josh, thanks for the comments. Yeah, my argument is predicated on a number of controversial assumptions--one of them being that the phenomenology is illusory. If you give that assumption up the argument definitely changes. I was trying to create a dilemma for those who share the same starting points as Shaun Nichols (and myself)--i.e., that the folk notion of free will is significantly in error and that the phenomenology of free will is illusory as well. While I think a good case could be made for both of these claims, I do not argue for them here. If these assumptions are mistaken, then of course the dilemma I propose could be avoided.
Since my next post will be about the phenomenology of free agency, perhaps I can take up your alternative reading then. Sorry for kicking the can to next week. That said, you may actually like what I have to say next week and your interpretation may have some traction.
Kip, many thanks for your comments and your kind words! Much appreciated. I agree with most of what you have to say so you won't get much disagreement from me. To comment on your third point, yes Shaun at the end of his paper makes a pragmatic case for the practical benefits of preserving the concept of free will and points to some potential dangers of free will eliminativism (although I believe he also says the evidence cuts both ways w/regard to the practical considerations). Since I am an optimistic skeptic à la Derk Pereboom, Bruce Waller, Thomas Nadelhoffer, etc. I don't find the practical argument for preservationism compelling. (I'll be posting about this toward the end of the month.) So I very much agree with you about (3).
As for your first point, I agree that turning to third part attributions of moral responsibility would not necessarily help the preservationist--in part, for the reasons you mention but also because of the great work Tamler Sommers has done on the cultural relativity of our moral responsibility practices.
Posted by: Gregg Caruso | 03/07/2015 at 10:56 AM
Very interesting post--great overview of lots of issues here.
Isn't there one phenomenon of consciousness that is universal among self-conscious people across the cultural spectrum? That is, people can imagine logical possibilities. They imagine non-actual ones in the past in counterfactual thinking, but perhaps more relevantly they imagine contrasting and even contradictory logical possibilities in the local future that seem often relevant to their decisions and actions. Conceivability is a stock-in-trade for such propositions that appear to be somehow accessible to us as alternatives. But of course the metaphysical conclusion that alternative conceivable propositions are in fact accessible to us isn't warranted. The near-universal but erroneous conclusion that conceivability entails at least real dual-ability access to such alternatives certainly must be one strong route to the baptism of "free will", and independently of any questions about morality, responsibility, etc.
Posted by: V. Alan White | 03/07/2015 at 11:45 AM
Alan, interesting proposal. So you're suggesting that the initial baptism was to our ability to conceive of alternative logical possibilities, is that correct? Is this a phenomenological feature of experience or are you conceiving of it as something different? For example, are you proposing that we experience ourselves as being capable of counterfactual thinking as well as conceiving of logically open possibilities, and it is this experience we have of such thinking that sets the reference of free will? It also sounds like you are saying that this "near-universal" phenomenon on consciousness is "erroneous." Would this, then, be helpful to the preservationist? (Remember we are also assuming that the indeterminist intuition is erroneous for the sake of this conversation.)
Posted by: Gregg Caruso | 03/07/2015 at 12:20 PM
I agree with your arguments about what follows from the assumptions you share with Nichols, but as you know I don't share all of those assumptions. I think our ability to engage in counterfactual reasoning in our deliberations, coupled with our decisions following from our overall assessment of those considerations, coupled with our actions following consistently with and from those decisions, together with our sense that of the voluntary nature of the bulk of these processes, together with tacit knowledge based on induction from previous episodes of the same, and a host of related phenomenological aspects of agentive experience, form the phenomenological basis of our folk conception of agency and I agree with your argument that the phenomenological base here is the most intuitive referent in the original baptism of the concept. As you probably know, I don't think the mere putative fact of determinism is enough to justify illusionism, nor that there is enough error in this reference base to justify illusionism.
Posted by: Rick Repetti | 03/07/2015 at 01:30 PM
Rick, thanks for the comments. I agree that if the phenomenology of free agency included only the features you mentioned, which all sound very compaibilist-friendly, the assumption of determinism alone would not be enough to justify illusionism. But I think there is more to the phenomenology than you do. I also think there are aspects of our first-person experience of agency that are undermined, not by determinism, but by a proper understand of consciousness along with recent developments in the behavior, cognitive, and neurosciences. To go into detail here, however, we take me too far afield. As you pointed out, I'm running the argument on a set of shared assumptions with Nichols. I know from our conversations that you don't share these with me ;)
Posted by: Gregg Caruso | 03/07/2015 at 02:14 PM
Hi Gregg--only a moment to respond. But my point is that anybody who concludes from the conceivability of something like future subjunctives of action that this entails an actual ability to engage in more than one possible course of action is mistaken. But most people of course will not realize that, and think such an entailment supports their belief in free will.
Posted by: V. Alan White | 03/07/2015 at 04:36 PM
Ok Gregg I'm back, after watching Eddy Redmayne's doppelganger portrayal of Hawking on BlueRay tonight.
Conceivability may have some substantive metaphysical results if that constrains what we may be able to think about possible scenarios--Chalmers has put that in play for sure. But no one has suggested that the mere fact that we can conceive of various possible scenarios in the future implies that we have actual powers to enact any such scenarios--logical possibility or conceivability cannot entail that we have any given powers or abilities in the actual world. But people largely think that they have the power to influence the future--an open future--because they can conceive of it as such. The root of incompatibilist FW is planted in our imaginations, and people world-wide have great powers of imagination. But that fact has zero rational force for thinking that we have incompatibilist-open-futured FW, and constitutes a near-universal error in thinking that we do.
Posted by: V. Alan White | 03/07/2015 at 10:22 PM
s it really true that we eliminated 'witch' because we happened to use a descriptivist amount of reference? Isn't it more plausible that because a causal-historical account would fail to preserve the properties of witches that was central to our implicit theory of witches (doing magic; being in league worth the devil), it just wasn't an option? You can see where this is going: at least if moral responsibility is a property that it matters to preserve, the historical account just isn't available.
(alternatively, were we to sever the connection between free will and moral responsibility - as Bruce Waller recommends - we get elimination of what you and I care about in any case. So far as I can tell, it would be crazy to try to use the causal historical view to vindicate moral responsibility; not, at least, on a basic desert account of the latter. You can't punish someone on the grounds that thought they don't instantiate properties that justify punishment, they instantiate properties that are in the same general region).
Posted by: Neil Levy | 03/08/2015 at 07:49 AM
Neil, excellent points! I'm curious to hear from some preservationists to see how they respond.
I especially like your point that: "it would be crazy to try to use the causal historical view to vindicate moral responsibility; not, at least, on a basic desert account of the latter. You can't punish someone on the grounds that thought they don't instantiate properties that justify punishment, they instantiate properties that are in the same general region." It reminds me of those in the marriage equality debate that want to focus on what the concept "marriage" originally meant rather than address questions of justice and fairness. Who cares about the causal-historical account of marriage!? Shouldn't the question be about what is right and just--i.e., is it just to deny a group of people equal rights based on their sexual orientation? Perhaps this is a poor analogy but I think you get my point!
Alan, I'm wondering if you see your comments as helping the preservationist or the eliminativist? I don't see how a "near-universal error in thinking" is helpful to the preservationist. Can you apply your proposal to the argument at hand so I can better understand which way you think it cuts? I like to know the traps that lay ahead before I commit to anything.
Posted by: Gregg Caruso | 03/08/2015 at 08:38 AM
Sure Gregg, I'll have a go at it.
While I think that cultural and historical and religious factors have variously contributed to the prevalence of L-like FW intuitions, I think my "false consequence argument" (FCA) is a ground-level account for why such a view percolated into thought in the first place, and will likely and stubbornly continue to contribute to its prominence in everyday life. But once we see that the FCA cannot provide justification for the existence of L-like FW, that realization undercuts a lot of what people are referring to when they claim they have it.
On the other hand, people obviously do have at least some control--even if only causally--over the future, and at least some of that control can be tied to better or worse reasons. Whether that is a better referent for "free will" than L-like FW I can't commit to--but I think it's an anchoring point for some compatibilist account of moral responsibility, though it's going to need to deal with Neil's luck arguments even there as a part of holding people responsible.
So I'm a (pragmatic) moral responsibility preservationist at least, but for me the jury is still out on FW eliminativism.
There's a reason people love "I Believe I Can Fly":
. . .There are miracles in life I must achieve
But first I know it starts inside of me,
If I can see it, then I can be it
If I just believe it, there's nothing to it
I believe I can fly. . .
Well, at least they can, by merely thinking about it beforehand, dual-ably choose, if not to fly, to go out a door to the right *and* to the left, and to have been able to have done otherwise even after the fact of doing either. And they ain't thinking in particularly compatibilist ways here.
Posted by: V. Alan White | 03/08/2015 at 12:17 PM
Thanks Alan that helps a lot. I don't disagree then with your FCA since it's perfectly consistent with eliminativism. While I am not committing to it, there is also no reason for me to reject it. It provides a plausible account of how we came to believe in L-like FW.
I also don't disagree when you say: "On the other hand, people obviously do have at least some control--even if only causally--over the future, and at least some of that control can be tied to better or worse reasons." Most free will skeptics would agree, they would just add that this is not enough to ground desert-based moral responsibility (though I know we would disagree on that latter point).
Going back to free will eliminativism, I'm glad to hear that you are open to the idea! Perhaps if we just eliminated the term free will, we could focus on things you seem to care about--degrees of control, autonomy, reasons responsiveness, etc.
Posted by: Gregg Caruso | 03/08/2015 at 12:43 PM
Thanks for the comments Gregg. Just to be clear let me outline what I currently believe:
(i) Libertarianism is false. What I have dubbed FCA doesn't support its supposed indeterminism, and no other evidence has moved me to believe that it is a clear and stable position that yields the right form of control it claims to have *pace* many who are much sharper than I that disagree (and all whom I respect).
(ii) Largely because of (i) but also because of metaphysical/constitutive luck (thanks Neil) I do not support backward-looking accounts of buck-stopping, basic-desert responsibility.
(iii) While I am open to being convinced about eliminating FW as any kind of stable concept whether compatibilist or not, I'm inclined to think that there are grounds, at least some pragmatic grounds, for linking "free will" to some appropriate, wide-ranging cluster of psychological states and behaviors as a plausible referent. (I think Vargas' BBB makes as careful and convincing case for that as I've ever encountered.)
(iv) Even if eliminativism is true, pragmatism more generally supports identifying appropriate psychological states and behaviors as proper markers of responsibility in both attributive and accountable ways (I'm a big fan of Wolf and Watson here).
(v) I'm (some sort of) a moderate realist about values, but think that the role of values is far too often neglected as what really, deeply drives arguments about FW/MR, and even those that seem far afield from such issues.
Well at least you've prompted me to come out clearly where I stand generally! Thanks.
Posted by: V. Alan White | 03/08/2015 at 03:59 PM
Alan, thanks for laying out your position so clearly! Given that you accept (ii) we may not be that far apart. Perhaps later in the month when I discuss the practical consequences of free will skepticism--including my optimism about life without free will and desert-based MR--you may come to disagree. But it sounds like we both embrace forward-looking accounts of MR, even if we reject basic desert MR.
As for this post, your (iii) is obviously the most relevant. Personally, I do not see how pragmatic considerations can settle the reference question, but since you're open to FW eliminativism perhaps I should stop while I am ahead ;)
Posted by: Gregg Caruso | 03/08/2015 at 04:24 PM
//It reminds me of those in the marriage equality debate that want to focus on what the concept "marriage" originally meant rather than address questions of justice and fairness.//
I once got into a big debate about marriage equality, and one of my opponents suggested "If they just didn't call it 'marriage' I'd be fine with it."
//Perhaps if we just eliminated the term free will, we could focus on things you seem to care about--degrees of control, autonomy, reasons responsiveness, etc.//
Ahem.
Posted by: Mark Young | 03/08/2015 at 04:31 PM
Gregg, great post and thanks for the kind words about my work. One thing I should emphasize though: my work on cultural diversity was solely concerned with moral responsibility. I don't know if there's anything close to that kind of diversity on the question of free will, free action, voluntary action etc. I mention this because you don't seem to be using 'free will' as a placeholder for morally responsible action.
On a different note: would deep cross-cultural (and cross-historical) differences on 'free will' necessarily lead to eliminativism? (I didn't think so in my book). It could also lead to relativism or pluralism, right?
Posted by: Tamler Sommers | 03/10/2015 at 03:12 PM
Tamler, thanks for your comments! I know your work is concerned with moral responsibility and not free will per se. That said, I reference your work at a point in the argument when moral responsibility practices are in fact relevant. A preservationist, for example, may try to avoid my dilemma by shifting their focus to more external reference-fixing properties, such as the cultural practices associated with punishment, reward, praise, and blame. More specifically, the preservationism may suggest that the original baptism (on a causal-historical account) is to some external set of moral responsibility practices. It's at this point that I think they would need to confront your excellent work on the cultural diversity of such practices. I was only appealing to your work in this limited way (which I think is a fair application of your work?).
I think your other point, however, about eliminativism vs. relativism is a fair one. If the preservationist adopts a causal historical account where the reference-fixing properties were to such cultural practices, they may end up facing the challenge of relativism rather than eliminativism. That's a fair enough point. Most preservationists, however, would probably find that conclusion equally unsatisfying. Hence, if preservationists want to avoid relativism and/or eliminativism, they need to provide an alternative account of how the initial baptism took place. (They would also need to address Neil's concerns above about adopting a causal-historical approach in the first place.)
Posted by: Gregg Caruso | 03/10/2015 at 05:08 PM
I see, that makes sense. Yeah, trying to locate the original baptism of free will in anything relating to blame, praise, and punishment is a non-starter (even for the relativist). The connection between freedom and blame/punishment is tenuous in many of the cultures I talked about in the book.
Posted by: Tamler Sommers | 03/11/2015 at 10:37 AM
Agreed!
Posted by: Gregg D. Caruso | 03/11/2015 at 10:41 AM