According to the noncausal libertarian view of free will, in order for a person’s action to be free, it must be uncaused. A standard criticism of this view—the control objection—is that a person cannot have control over whether an uncaused action occurs and, so, such an action cannot be free. The background to this criticism is the claim that control over action is plausibly a causal rather than noncausal matter. In this paper, I defend noncausal libertarianism against the control objection by developing a new noncausal theory of control. What emerges is not simply a defense of noncausal libertarianism against the control objection but a new theory of control more generally.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
Subscribe to journal
Immediate online access to all issues from 2019. Subscription will auto renew annually.
Tax calculation will be finalised during checkout.
There are variant noncausal libertarian views, developed for instance in initial work by Ginet (1990, 1997), according to which while a person’s free actions could be caused (albeit not deterministically), they need not be caused. The question I take up in this essay applies equally to these views: Given that such views allow that a person can act freely if her action is not caused, how, if her action is uncaused, can she have control over whether her action occurs?
Davidson (1980) is the standard development of this position.
Obviously, a complete defense of noncausal libertarianism requires giving a noncausal account of what makes an event an action in the first place, as well as responding to other objections to the view. But my aim in this paper is just to defend the view against one prominent criticism: the control objection.
I thank both anonymous reviewers for comments that led me to substantially rewrite this section of the paper to better capture the force and depth of the control objection as well as consider, in the next section, existing responses to it.
Lewis (1973), for example, proposes that: “We think of a cause as something that makes a difference, and the difference it makes must be a difference from what would have happened without it” (p. 557).
This is against the background of the claim that every action either is, or begins with, a basic action such as a decision or volition (where a non-basic, complex action is an event that’s composed of a basic action and its consequences). Thus, while a complex action will have causation within it—that is, part of what composes a complex action is a basic action’s causing some event—the (whole) event of a complex action itself is distinguished from a mere occurrence not by virtue of it—the (whole) event of the complex action—being caused in any way but, rather, by virtue of it having—via the part of it that’s a basic action—certain intrinsic features.
Ginet (1990) also argues that when a person acts freely, she cannot be subject to irresistible compulsion.
As they point out, concerns about appealing to these intrinsic features to explain the control required for an action to be free are, in the first instance, concerns about whether these features are sufficient to explain what makes an event an action per se rather than a mere occurrence. They make this latter concern vivid by arguing that an event with the intrinsic features that Ginet and McCann cite could seemingly be the result of external brain stimulation, in which case the event would not seem to be an action of the person’s at all despite these features being present.
Widerker (2018) himself suggests that a person has control over whether her action occurs in virtue of her having the power to act otherwise, where her having this power is understood as a primitive notion. Developing an account of control in terms of having the power to do otherwise seems prima facie plausible. But one reason to be cautious is that, given the well-known knotty debates about what it is for a person to have the power to do otherwise, such an account might raise as many questions as it answers.
Clarke does not endorse this proposal. Rather, he uses it as a way of responding to a criticism by Pereboom (2001, 2014) of event-causal libertarian views of free will. Pereboom argues that such views—on which a person’s free actions must be indeterministically caused by prior events or states of the agent—are false because, if a person’s actions were so caused, she would not have the power to “settle” whether her action occurs (something required for acting freely). In reply, Clarke offers the aforementioned proposal for what it is for an agent to settle whether her action occurs and argues that Pereboom hasn’t given a clear enough (or defensible enough) account of what it means for a person to “settle” whether her action occurs that would entail that she cannot settle her action in this way, something that she would be able to do even if (as required by event-causal libertarians) her action were indeterministically caused by prior events or states.
He argues that this is a logical “by” rather than a causal “by” and illustrates the difference as follows: “we have a causal “by” in “I made a C major chord sound by pressing those three keys”; we have a logical “by” in “I made a C major chord sound by making sound simultaneously a C, an E, and a G.”” (1997, p. 87).
Just as one might question, in the light of Ginet’s and Clarke’s proposals, what it means for a person to “make it the case” that her action occurs or “settle” whether it occurs, couldn’t someone also question what it means for a person to have “control” over whether her action occurs? Yes. However, for the purposes of this paper, I’ll grant proponents of the control objection that we have an intuitive grasp on this concept—at least, an intuitive grasp on it that’s sufficient to begin assessing whether such control is a causal or noncausal matter.
Clarke (2019) makes this very same point concerning what it means for a person to “make a decision.”
Why do I say that a person’s so acting or deciding constitutes her exercise of control over whether the action or decision occurs as opposed to using some other “dependence” relation, such as a grounding (i.e., such as saying that a person’s so acting grounds her exercising control over whether the action occurs)? The reason is that grounding is typically (though not universally) thought to be a relation between facts, and what is at issue here is events—the event of the person’s performing the action and the event of her exercising control over whether the action occurs. Moreover, constitution is usually thought of a relation that can hold between events (and this is true, even though the standard examples of constitution—the clay constituting the statue, for instance—involve employing it as a relation between objects not events). For instance, if a someone attacks someone by kicking her, then it is plausible to say that the event of the kick constitutes (at least in part) the event of the attack (Wasserman 2017). So, constitution seems a natural “dependence” relation to posit. I do not, however, want to be dogmatic about this. Since my aim here is to construct a noncausal account of control, and grounding is usually understood to be a noncausal relation, then I am open to grounding (or some other noncausal dependence relation) being the pertinent relation if it can be suitably developed.
I haven’t explicitly argued for claim (i)—that what a person must have control over, in order to act or decide freely, is whether her action or decision occurs. But it is a claim that all sides of the debate will agree on. Where causalists and noncausalists (and proponents and rejectors of the control objection) will disagree isn’t over the claim that what a person must have control over, in order to act or decide freely, is whether her action or decision occurs but, rather, whether her exercising control over this must be a causal matter or not. There is one question, however, concerning (i), for which I thank an anonymous reviewer: In order to act or decide freely, must the person have control over whether her action or decision occurs per se or just whether it occurs at a particular time? In what follows, I stay neutral on this, noting that my account can be developed in either way. Although I develop my account in the main text without temporal indices, it can be modified to accommodate them. Suppose that what a person must have control over, in order to act or decide freely, is not whether her action or decision occurs per se but, rather, whether it occurs at a particular time T. We can then understand my account of control as follows: Assuming that no-one else and nothing else has control over whether her action or decision occurs at T, the person herself can exercise control over whether her action or decision occurs at T simply by performing that action at T or making that decision at T, where her performing that action at T or making that decision at T constitutes her exercise of control over whether that action or decision occurs at T.
Can my account of control explain omissions? I believe so. Suppose, for instance, that a person omits to raise her arm. On the account I propose, in order for the person to freely omit to raise her arm, she must have control over whether she omits to raise it. Moreover, if no-one else and nothing else has control over whether she omits to raise her arm, then she can exercise control over whether she omits to raise it simply by omitting to raise it, where her omitting to raise it constitutes her exercise of control over whether she omits to raise it. (Of course, omissions present their own metaphysical problems—for instance, in what sense, if any, they exist. But these are problems for any account of control over omissions, not just mine.) I thank an anonymous reviewer for raising this issue.
In fact, my view isn’t committed to any view about what makes an event an action. All it’s committed to, from a noncausal libertarian perspective, is that what makes an event an action does not involve the event’s being caused in any way.
Of course, one might now ask how the mere occurrence of an event with the action-conferring features (whatever they are) could explain how a person has control over whether her action occurs. My answer to this is now familiar. Given that no-one else and nothing else has control over whether the person’s action occurs, what else would she need to do in order for her to have control over whether her action occurs other than perform it (where her performing it is just for the event of her action to occur)? And what else would need to be the case in order for her to have control over whether it occurs besides the fact that she performs it? The answer to both of these questions, I’ve been arguing, is: nothing.
I thank an anonymous reviewer for encouraging me to explain how my account avoids this version of the objection.
Graham (2017) similarly suggests that, when it comes to a person’s making something the case, this relation has, “as we ordinarily understand it” (p. 2), either a causal or constitutive sense, depending on what’s at issue.
I concede that we may not immediately realize that we ordinarily understand these notions to have two different senses (i.e., that it might take some reflection for us to realize that we do). But this is (arguably) because when we talk (in ordinary life) about making something happen or making the difference as to whether it happens (to the extent we do—or when we use cognate notions to express this), we’re usually talking about events that are the consequences of our actions (not our actions themselves) and, so, in most of our everyday usage of these terms, we use them in the causal sense only.
I thank an anonymous reviewer for raising this issue.
There’s a further issue here (helpfully raised by an anonymous reviewer): What would it be for someone else or something else to have control over whether a person performs an action? Presumably, the control at issue must involve causation, for (intuitively) at least part of how someone or something else could have control over whether another person (say) raises her arm is by causing her to raise it. Thus, am I committed to control having a causal as well as noncausal sense? The answer is yes, depending on what type of event is at issue. As I explained earlier, when it comes to a person’s own actions (or omissions), she has control over whether they occur (assuming that no-one else and nothing else does) not by causing them; in the case of her own actions, she has control over them simply by performing them. But when it comes to events that are not a person’s own actions (or omissions), a person (arguably) has control over their occurrence, at least in part, by causing them. So, given that a person’s raising her arm is not an action (or omission) of someone else’s or something else’s, then (arguably) someone or something else can have control over whether the person raises her arm, at least in part, by causing (the event of) the person’s raising it.
Not everyone would agree that such people would act in such a case or that if they did that their actions must be unfree. For instance, it might be argued that this kind of neural manipulation or severe addiction would render these people mere “puppets” or “robots” in such a way that they cannot even act at all, even unfreely. Or it might be argued, as Frankfurt (1971) might suggest, that although a person so manipulated or addicted would act in such a case, their actions need not be unfree—they would not be unfree if, for instance, they were acting as they desired to act, and that desire was one that they wanted to move them to act (thus meeting Frankfurt’s well-known condition for acting freely). In reply, I am just using these cases as illustrations of (intuitively) unfree action. Readers should feel free to substitute in their own preferred examples of unfree action.
Of course, it will also be the case according to (FW1) that in order for a person to act freely she must exercise control over whether her action occurs, and whether her action occurs cannot be under the control of someone else or something else. But this will be true by default. For according to (FW1), in order for a person to act freely, she must have control over whether her action occurs, and she cannot have this control if the action’s occurrence is under the control over someone or something else.
An anonymous reviewer raises the following interesting issue for libertarians. Suppose, as (FW2) allows, that a person can exercise control over whether her action occurs even if someone or something else also has control over this. Then it seems that a person can have control over her action’s occurrence even if it’s causally determined by someone or something else. But if a person can have control over whether her action occurs even if it’s causally determined, then it’s hard to see why she couldn’t exercise free will even if her action is causally determined. (This is all the more apparent given that, on some understandings of free will, to “have free will” just is for a person to have control over whether her action occurs.) Of course, (FW2) rules out a person’s acting freely under such circumstances, saying that if a person’s action was also under the control of someone or something else then she can’t act freely. But in the light of the reviewer’s challenge, this might seem unduly stipulative. On balance, then, it might be best for libertarians to endorse (FW1) rather than (FW2), since (FW1) says that a person simply cannot have control over whether her action occurs if its occurrence is under the control of someone or something else. As an aside, I think it’s natural for libertarians to endorse (FW1) since it seems to capture a key libertarian intuition: that if a person’s action is under the control of someone or something else, then the person herself cannot have control over it.
Clarke, R. (2003). Libertarian accounts of free will. New York: Oxford University Press.
Clarke, R. (2019). Free will, agent causation, and ‘disappearing agents’. Nous,53, 76–96.
Clarke, R., & Capes, J. (2017). Incompatibilist (nondeterministic) theories of free will. In E. N. Zalta (Ed.), Stanford encyclopedia of philosophy (Spring 2017 edition). Retrieved May 27, 2020 from https://plato.stanford.edu/archives/spr2017/entries/incompatibilism-theories/.
Davidson, D. (1980). Essays on actions and events. Oxford: Clarendon Press.
Frankfurt, H. (1971). Freedom of the will and the concept of a person. Journal of Philosophy,68, 5–20.
Ginet, C. (1990). On action. Cambridge: Cambridge University Press.
Ginet, C. (1997). Freedom, responsibility, and agency. Journal of Ethics,1, 85–98.
Ginet, C. (2007). An action can be both uncaused and up to the agent. In C. Lumer & S. Nannini (Eds.), Intentionality, deliberation, and autonomy. Aldershot: Ashgate.
Ginet, C. (2014). Can an indeterministic cause leave a choice up to the agent? In D. Palmer (Ed.), Libertarian free will: Contemporary debates. New York: Oxford University Press.
Goetz, S. (1997). Libertarian choice. Faith and Philosophy,14, 195–211.
Goetz, S. (1998). A noncausal theory of agency. Philosophy and Phenomenological Research,49, 303–316.
Goetz, S. (2008). Freedom, teleology, and evil. London: Continuum.
Graham, P. (2017). Review of van Inwagen’s. In Thinking about free will. Notre dame philosophical reviews. Retrieved May 27, 2020 from https://ndpr.nd.edu/news/thinking-about-free-will/.
Kane, R. (1996). The significance of free will. New York: Oxford University Press.
Lewis, D. (1973). Causation. Journal of Philosophy,70, 556–567.
McCann, H. (1998). The works of agency. Ithaca, NY: Cornell University Press.
McCann, H. (2012). Making decisions. Philosophical Issues,22, 246–263.
Mele, A. (2006). Free will and luck. New York: Oxford University Press.
O’Connor, T. (1993). Indeterminism and free agency: Three recent views. Philosophy and Phenomenological Research,53, 499–526.
O’Connor, T. (2000). Persons and causes. New York: Oxford University Press.
Pereboom, D. (2001). Living without free will. Cambridge: Cambridge University Press.
Pereboom, D. (2014). Free will, agency, and meaning in life. New York: Oxford University Press.
Wasserman, R. (2017). Material constitution. In E. N. Zalta (Ed.), Stanford encyclopedia of philosophy (Fall 2108 edition). Retrieved May 27, 2020 from https://plato.stanford.edu/archives/fall2018/entries/material-constitution/.
Widerker, D. (2018). In defense of noncausal libertarianism. American Philosophical Quarterly,55, 1–14.
I would like to thank Carl Ginet and, especially, two anonymous reviewers for extremely helpful comments on earlier drafts.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Palmer, D. Free will and control: a noncausal approach. Synthese (2020). https://doi.org/10.1007/s11229-020-02701-4
- Free will