Skip to main content

A Dilemma for Reductive Compatibilism


A common compatibilist view says that we are free and morally responsible in virtue of the ability to respond aptly to reasons. Many hold a version of this view despite disagreement about whether free will requires the ability to do otherwise. The canonical version of this view is reductive. It reduces the pertinent ability to a set of modal properties that are more obviously compatible with determinism, like dispositions. I argue that this and any reductive view of abilities faces a significant challenge: it cannot adequately explain the freedom-grounding element of this ability. The problem has the form of a dilemma. This leaves reasons-responsive compatibilists with two options: abandon theories of free will grounded in abilities or abandon reductive theories of abilities.

This is a preview of subscription content, access via your institution.


  1. 1.

    The quote is often misattributed to Viktor Frankl. See:

  2. 2.

    Wolf (1990), Fischer and Ravizza (1998), Nelkin (2011), McKenna (2013), Vargas (2013), and Vihvelin (2013), among others, each defend some version of this view despite disagreement about whether free will requires the ability to do otherwise. Sartorio (2016) is a kind of reasons-responsive theorist, but her view does not turn on abilities. “Reasons-responsive compatibilism” usually picks out views in the style of Fischer and Ravizza (1998). I use the term more broadly. I would include Susan Wolf (1990) and Dana Nelkin (2011) in this camp, who defend what is sometimes called the “Reason view”. Kadri Vihvelin’s (2013) view is often labeled a form of “dispositional compatibilism”, since her view involves the disposition to choose on the basis of reasons. All of these views understand free will is to be had in the ability to respond aptly to reasons, in spite of disagreement about, for instance, whether moral responsibility requires the ability to do otherwise.

  3. 3.

    It is possible that some event-causal libertarian views of free will have implicit commitments to the kind of reductionism at issue in this paper. (Agent-causal libertarians do not endorse reductionism, so they can be set aside). But there are tricky interpretive issues here. I will discuss these complications at length in footnotes 27 and 29. My view is that a further discussion is needed about how event-causal libertarians treat the ontological status of abilities, one that falls outside the scope of this paper. The general problem for reductionism about abilities is most clearly articulable, and most important for the ongoing debate about free will, in the context of contemporary accounts of reasons-responsive compatibilism.

  4. 4.

    It is generally accepted that there is at least one other condition on moral responsibility besides a metaphysical or control condition: a knowledge condition. It is an open question whether not the conditions for moral responsibility are exhausted by these two conditions. I discuss the problem in terms of necessary and sufficient conditions for the sake of clarity in presenting the problems. Less strictly, what we are after is a theory or explanation of what moral responsibility requires. I should also note that although it is not uncontroversial to frame the problem in terms of moral responsibility, it is also not universally accepted. For some likeminded philosophers, see: Pereboom (2001: xxii), Mele (2006: 17), and McKenna (2008: 187).

  5. 5.

    More formally, determinism is the thesis that two propositions, one describing the past at some given time, and another describing the laws of nature, together entail a proposition describing the one unique future.

  6. 6.

    I’ll omit this qualification in the discussion to follow.

  7. 7.

    McKenna (2019: 23, ft. 15) claims that Fischer and Ravizza never commit to a reductive view. Nevertheless, I believe it is the most promising interpretation of their view, since, if it succeeds, it secures the compatibility claim. McKenna’s alternative suggestion for them does not. I’ll discuss this kind of alternative in Sect. 6. Thanks to an anonymous reviewer for raising this issue.

  8. 8.

    To be clear, Vihvelin argues that free will and determinism are compatible prior to her endorsing this dispositional view of abilities by attacking incompatibilist arguments. That our abilities are explicable as bundles of dispositions is meant to explain why the ability to respond aptly to reasons is compatible with determinism. More forthcoming in Sect. 5.

  9. 9.

    Two points. First, van Inwagen means “willy-nilly” in the sense of happening like it or not, rather than happening haphazardly. I’ll use it in this sense too. Second, ordinary talk of abilities is quite permissive. We can say perfectly well of my car that it is able to handle rough terrain in virtue of its four-wheel drive and suspension. The abilities at issue in this discussion are the sort at issue in the free will debate, which are relevant to human agency.

  10. 10.

    For examples, see Martin (1994) and Lewis (1997).

  11. 11.

    One might be tempted to define indeterministic dispositions as those dispositions whose manifestation may but needn’t occur given the presence of the stimulus condition and the absence of intervening factors (cf. Clarke, 2009: 326, ft. 3). But as an anonymous reviewer rightly pointed out, these dispositions could occur in deterministic worlds. Nothing in my argument that there is a gap between dispositions and abilities hangs on how we should precisely understand the nature of indeterministic dispositions.

  12. 12.

    Sartorio (2005) argues that causation involves this kind of difference making.

  13. 13.

    Some, like Mele (2017), demand clarity about how one settles what they do in the context of the disappearing agent objection to event-causal libertarianism. Say that I am deciding on what flavor of ice cream to buy. Following McKenna and Pereboom (2016: 238), I say that you settle which flavor-buying decision—deciding to get butter pecan instead of chocolate—by settling on butter pecan. A candidate view of the free will ability is an elucidation of how an agent goes about settling. For instance, to say that free will consists in the reasons-responsive ability is to say that an agent settles what she is going to do insofar as she is able to recognize and reacting to a suitable range of reasons. Settling, then, is the prior notion, one that could plausibly be basic in giving a theory of free will. Obviously not everyone will agree with this characterization. As I say below, the difference between dispositions and the kinds of abilities at issue in the free will debate to turn on an intuitive desideratum on a theory of free will, that it explain how a free agent has the control to settle what they are going to do. The problem then is best understood as a burden shift driven by, as a matter of methodology in metaphysics, taking this intuitive desideratum seriously.

  14. 14.

    Clarke (2009: 338–339, 2015: 901) and Vetter (2019) each point to a similar problem. Clarke suggests that having many kinds of abilities, and in particular the ones at issue in the free will debate, cannot simply be a matter of having a disposition (or bundle of dispositions). Abilities involve both underlying competencies (plausibly construed as dispositions) but also something more. When it comes to the abilities at issue in the free will debate, perhaps this something more is the choice to exercise them. So, dispositions are necessary but not sufficient for having the sort of abilities at issue. Vetter argues that dispositionalists “fail to classify a range of important exercises as exercises of abilities” (2019: 07). Drawing examples from Steward (2012), she points to sub-intentional actions, like scratchings, and complex habits, like riding a bicycle, as instances of action that do not depend on the kinds of basic actions that tend to feature in dispositional analyses of abilities, like tryings (or decidings or choosings). The gap problem is a distinct worry from either Clarke or Vetter’s, but it explains why Clarke and Vetter are each right. Abilities make a difference in a way that dispositions do not. Thanks to an anonymous reviewer for pointing me in the direction of Vetter’s paper.

  15. 15.

    Thanks to an anonymous referee for raising this worry, and pushing me to clarify these points about how an agent’s activity can feature in the reductive dispositional analysis of abilities.

  16. 16.

    One could plausibly read Michael Smith’s (1997, 2003) and Michael Fara’s (2008) views in this way.

  17. 17.

    Thanks to an anonymous reviewer for helpfully suggesting this way of framing the gap problem.

  18. 18.

    They talk in terms of a “cognitive” power to recognize reasons and an “executive” power to choose. Both are general dispositions of an agent’s mechanism of action rather than abilities of the agent. This executive power is explicitly cast in terms of reacting to incentives recognized by the cognitive power (1998: 75). The successful manifestation of the cognitive disposition is the stimulus for the executive one.

  19. 19.

    For instance, Pereboom (2001: xxi) suggests that freedom and responsibility apply to decisions.

  20. 20.

    Switching to an agent-based reasons-responsive theory (e.g. McKenna, 2013) won’t resolve this problem. If we characterize the agent’s reasons-responsiveness in terms of dispositions, then we seem committed to her (willy-nilly) responding to a condition which was not settled by herself.

  21. 21.

    An initial gloss of “irrespective of the agent” might read “whether the agent likes it or not”, but as a referee helpfully pointed out, perhaps the pertinent manifestation of the disposition involves doing as one likes! The more apt comparison is again to an ordinary disposition like fragility. The fragile glass will simply break when dropped.

  22. 22.

    Vihvelin does not offer an analysis in the sense of giving necessary and sufficient conditions. Rather, she thinks of herself as offering an analysis vis-à-vis plausible ontological reduction as a kind of research program for compatibilists (2013: 170).

  23. 23.

    In spite of my criticism, I am sympathetic to Vihvelin’s defense of commonsense compatibilism. She rightly claims that the issue of determinism is orthogonal to what ordinary people mean when they say that they have abilities and opportunities. I only worry about her treating abilities like dispositions.

  24. 24.

    As an anonymous reviewer pointed out, trying often involves performing complex bodily actions, and it seems like the simple fact of acquiring an intention or forming a desire is not sufficient for trying. Vihvelin’s way around this problem, which strikes me as the plausible strategy, is to count as tryings causally effective initiators of complex sequences of goal-oriented action. Basic actions that count as tryings will likely involve one’s reasons for action, like the weighing of one’s reasons. But now we have to ask if this weighing is the exercise of an ability or the manifestation of a disposition, and hence, we run into exactly the sort of explanatory issue the gap problem raises for a view like Fischer and Ravizza (1998), as discussed in Sect. 4.

  25. 25.

    Vihvelin offers a defense of both options. She can consistently do so because she holds the view that a process counts as a trying if it causally leads to the beginning of an action. Such a process may or may not be initiated by an agent’s trying to try (2013: 176–180).

  26. 26.

    Although I know of no libertarians who offer explicitly reductive accounts of the free will ability in terms of indeterministic or chancy dispositions, it is generally assumed that you can get an event-causal libertarian view by taking an event-causal compatibilist view and inserting indeterminism into the causal chain at the right moment. And the reductive, reasons-responsive compatibilist views we have been considering are indeed event-causal. This suggests that it is possible that some extant libertarian accounts of free will also are caught up in the gap problem in terms of dispositions, and the dilemma it poses. Here’s how. Randolph Clarke (2000: 21, 22) characterizes a modest event-causal libertarian view as one that starts from the assumption that agents have “an ordinary capacity to engage in practical reasoning”, and who agrees with the compatibilist “that agent’s having and being able to exercise a capacity for rational-self-governance” but insists that some causation in the causal process that leads to action must not be deterministic. What are these capacities? Perhaps they are dispositions. Perhaps it is plausible, then, to think that the gap problem extends to event-causal libertarian views that unreflectively assume the standard reductive compatibilist commitments I believe fail due to the gap problem, and the dilemma that ensues from it. Are there unreflective event-causal libertarians? I am doubtful. My sense is that contemporary event-causal libertarians will either not offer full reductive analyses of the free will ability, like Franklin (2018), who offers a partial analysis of abilities, or simply not appeal to abilities in offering their views, perhaps like Ekstrom (2019)—see footnote 34. As a historical note about the origins of the gap problem, it is worth pointing out that van Inwagen himself nearly takes the notion of human ability as a fundamental when he rejects classic dispositional accounts of abilities (1983: 9–10). This makes sense. If you thought that the important freedom-grounding element of free will was indeterminism, you would not worry so much about appealing to (and then reducing) abilities to something that was compatible with determinism, like dispositions. However, like Dana Nelkin (2011: 75), I think that a focus on indeterminism mislocates what is special about free will.

  27. 27.

    See Mele (2006: 170) for another example.

  28. 28.

    Here again we might consider how actual libertarian views interact with the gap problem. Mele (2006, inter alia) is a well-known agnostic about the free will problem, advancing both compatibilist and libertarian theories of free agency. Perhaps the libertarian version of his view will face the gap problem. But in general, it is harder to say exactly what libertarians should think about reducing abilities to causal sequences. Let me explain with a recent example. Laura Ekstrom (2019: 141) seems to endorse a kind of reductionism about abilities when she says that “events—the occurrence of certain of an agent’s attitudes, which are considerations relevant to the decision—cause the event of the decision; that is what it is for an agent to exercise her ability to make a decision for reasons.” But in saying so, she is responding to the disappearing agent objection as Pereboom advances it, and so it seems charitable to think she is downplaying the explanatory role of abilities rather than adopting the strong version of reductionism about abilities that I am targeting in this paper. This would suggest that many event-causal libertarians have a somewhat analogous view to that of Carolina Sartorio (2016), whose I describe as advancing an “ability-neutral” view in the next section.

  29. 29.

    My sincere thanks to an anonymous referee for pointing this out. Disappearing agent objections are related to a more general worry with event-causalism as a reductive theory of agency. On an event-causal framework, where events are fundamental rather than agents, one might worry that there seems to be no agent settling anything at all, but rather, only agent-involving events (Mele, 2003; Steward, 2012; Velleman, 1992). One way to persuasively respond to this problem is, like Mele (2003), to characterize it in terms of “shrinking” agency and show how an event-causal account can explicate agency par excellence, agency that has all the features we want it to. I am inclined to think that part of agency par excellence is apt description of agency in terms of abilities and believe that these descriptions are consistent with both agent-causal and event-causal frameworks.


  1. Ayer, A. J. (1954). Freedom and necessity. Philosophical essays (pp. 3–20). St. Martin’s Press.

  2. Brand, M. (1984). Intending and acting. MIT Press.

    Google Scholar 

  3. Chisholm, R. (1964). Human freedom and the self. Reprinted in Chisholm, Roderick. 1989. On Metaphysics. Univ. of Minnesota Press.

  4. Clarke, R. (2000). Modest libertarianism. Philosophical Perspectives, 14, 21–45.

    Google Scholar 

  5. Clarke, R. (2009). Dispositions, abilities to act, and free will: The new dispositionalism. Mind, 118(470), 323–351.

    Article  Google Scholar 

  6. Clarke, R. (2015). Abilities to act. Philosophy Compass, 10(12), 893–904.

    Article  Google Scholar 

  7. Davidson, D. (1973). Freedom to act. In T. Honderich (Ed.), Essays on freedom of action. Routledge.

  8. Ekstrom, L. (2019). Toward a plausible event-causal indeterminist account of free will. Synthese, 196, 127–144.

    Article  Google Scholar 

  9. Fara, M. (2008). Masked abilities and compatibilism. Mind, 117(468), 843–865.

    Article  Google Scholar 

  10. Fischer, J. M., & Ravizza, M. (1998). Responsibility and control: An essay on moral responsibility. Cambridge University Press.

    Book  Google Scholar 

  11. Franklin, C. (2018). A minimal libertarianism: Free will and the promise of reduction. Oxford University Press.

    Book  Google Scholar 

  12. Hájek, A. (2020). Minkish dispositions. Synthese, 197, 4795–4811.

    Article  Google Scholar 

  13. Lehrer, K. (1968). Can’s without ‘if’s. Analysis, 24, 159–160.

    Article  Google Scholar 

  14. Levy, N. (2008). Bad luck once again. Philosophy and Phenomenological Research, 77(3), 749–754.

    Article  Google Scholar 

  15. Levy, N. (2011). Hard luck: How luck undermines free will and moral responsibility. Oxford University Press.

    Book  Google Scholar 

  16. Lewis, D. (1973). Causation. Journal of Philosophy, 70, 556–567.

    Article  Google Scholar 

  17. Lewis, D. (1997). Finkish dispositions. The Philosophical Quarterly., 47(187), 143–158.

    Article  Google Scholar 

  18. Manley, D., & Wasserman, R. (2007). A gradable approach to dispositions. The Philosophical Quarterly, 57(226), 68–75.

    Article  Google Scholar 

  19. Martin, C. B. (1994). Dispositions and conditionals. The Philosophical Quarterly, 44(174), 1–8.

    Article  Google Scholar 

  20. McKenna, M. (2013). Reasons-responsiveness, agents, and mechanisms. In D. Shoemaker (Ed.), Oxford studies in agency and responsibility (Vol. 1, pp. 151–184). Oxford University Press.

  21. McKenna, M. (2019). Watsonian compatibilism. In J. Coates & N. Tognazzini (Eds.), Oxford studies in agency and responsibility (Vol. 5).

  22. McKenna, M. (2008). Ultimacy & sweet Jane. In N. Trakakis & D. Cohen (Eds.), Essays on free will and moral responsibility (pp. 186–208). Cambridge Scholars Publishing.

    Google Scholar 

  23. McKenna, M. (2014). Resisting the manipulation argument: A hard-liner takes it on the chin. Philosophy and Phenomenological Research, 89, 467–484.

    Article  Google Scholar 

  24. McKenna, M., & Pereboom, D. (2016). Free will: A contemporary introduction. Routledge.

    Book  Google Scholar 

  25. McKitrick, J. (2009). Dispositional pluralism. In G. Damschen, R. Schnepf, & K. R. Stüber (Eds.), Debating dispositions: Issues in metaphyscis, epistemology and philosophy of mind (pp. 186–203). Walter de Gruyter.

  26. Mele, A. (1995). Autonomous agents. Oxford University Press.

    Google Scholar 

  27. Mele, A. (2003). Motivation and agency. Oxford University Press.

    Book  Google Scholar 

  28. Mele, A. (2006). Free will and luck. Oxford University Press.

    Book  Google Scholar 

  29. Mele, A. (2017). On Pereboom’s disappearing agent argument. Criminal Law and Philosophy, 11(3), 561–574.

    Article  Google Scholar 

  30. Mele, A., & Moser, P. K. (1994). Intentional Action. Nous, 28, 39–68.

    Article  Google Scholar 

  31. Moore, G. E. (1912). Ethics. Oxford University Press.

    Google Scholar 

  32. Nelkin, D. (2011). Making sense of freedom and responsibility. Oxford University Press.

    Book  Google Scholar 

  33. Nozick, R. (1981). Philosophical explanations. Harvard University Press.

    Google Scholar 

  34. Pereboom, D. (2001). Living without free will. Cambridge University Press.

    Book  Google Scholar 

  35. Pereboom, D. (2008). A hard-line reply to the multiple-case manipulation argument. Philosophy and Phenomenological Research, 77(1), 160–170.

    Article  Google Scholar 

  36. Pereboom, D. (2014). Free will, agency, and meaning in life. Oxford University Press.

    Book  Google Scholar 

  37. Prior, E., Pargetter, R., & Jackson, F. (1982). Three theses about dispositions. American Philosophical Quarterly, 19, 251–257.

    Google Scholar 

  38. Sartorio, C. (2005). Causes as difference-makers. Philosophical Studies, 123(1), 71–96.

    Article  Google Scholar 

  39. Sartorio, C. (2016). Causation and free will. Oxford University Press.

    Book  Google Scholar 

  40. Schlick, M. (1939). When is a man responsible? In Schlick, Problems of ethics (pp. 143–156). Prentice-Hall.

  41. Smith, M. (2003). Rational capacities, or: how to distinguish recklessness, weakness, and compulsion. In S. Stroud & C. Tappolet (Eds.), Weakness of will and practical irrationality (pp. 17–38). Clarendon Press.

  42. Smith, M. (1997). A theory of freedom and responsibility. In G. Cullity & B. Gaut (Eds.), Ethics and practical reason (pp. 293–317). Oxford University Press.

    Google Scholar 

  43. Steward, H. (2012). A Metaphysics for freedom. Oxford University Press.

    Book  Google Scholar 

  44. van Inwagen, P. (1983). An essay on free will. Clarendon Press.

    Google Scholar 

  45. Vargas, M. (2013). Building better beings. Oxford University Press.

    Book  Google Scholar 

  46. Velleman, D. (1992). What happens when someone acts? Mind, 101(403), 461–481.

    Article  Google Scholar 

  47. Vetter, B. (2013). ‘Can’ without possible worlds. Philosopher’s Imprint, 13 (16), 1–27.

  48. Vetter, B. (2019). Are abilities dispositions? Synthese, 196, 201–220.

    Article  Google Scholar 

  49. Vihvelin, K. (2013). Causes, laws, & free will: Why determinism doesn’t matter. Oxford University Press.

    Book  Google Scholar 

  50. Waller, B. (2011). Against moral responsibility. MIT Press.

    Book  Google Scholar 

  51. Watson, G. (2004). Agency and answerability. Oxford University Press.

    Book  Google Scholar 

  52. Whittle. A. (2010). Dispositional abilities. Philosopher’s Imprint, 10(12), 1–23.

  53. Wolf, S. (1990). Freedom within reason. Oxford University Press.

    Google Scholar 

Download references


My sincere thanks to Michael McKenna, Dana Nelkin, Carolina Sartorio, Terry Horgan, Mark Timmons, Jeremy Reid, Tim Kearl, Joseph Metz, Josh Cangelosi, and Phoebe Chan for comments on previous drafts of this paper. Thanks also to three anonymous referees at Erkenntnis, and a number of anonymous referees at other journals, for helpful comments and criticism.

Author information



Corresponding author

Correspondence to Robert H. Wallace.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Wallace, R.H. A Dilemma for Reductive Compatibilism. Erkenn (2021).

Download citation