Springer Nature is making SARS-CoV-2 and COVID-19 research free. View research | View latest news | Sign up for updates

Three Recent Frankfurt Cases


Three recent ‘state of the art’ Frankfurt cases are responded to: Widerker’s Brain-Malfunction-W case and Pereboom’s Tax Evasion cases (2 & 3). These cases are intended by their authors to resurrect the neo-Frankfurt project of overturning the Principle of Alternative Possibilities (PAP) in the teeth of the widespread acceptance of some combination of the WKG (Widerker-Kane-Ginet) dilemma, the Flicker of Freedom strategy and the revised PAP response (‘Principle of Alternative Blame’, ‘Principle of Alternative Expectations’). The three neo-Frankfurt cases of Pereboom and Widerker shown to be insufficient for their intended purpose. Of central importance to any account of responsibility is that this applies at the level of the Right and not the Good. Arguments of Carlos Moya are expanded and augmented by considerations from Chisholm, Lucas, Dummett and Lockie (2003) to show that a number of severe problems remain for anyone attempting to resurrect the Frankfurt project.

This is a preview of subscription content, log in to check access.

Fig. 1


  1. 1.

    “Free will and / or moral responsibility…” Henceforth I shall restrict myself to the latter, as this is where the Frankfurt debates are most centrally situated; but this is to flag that one may also be concerned with these debates as they apply to the former, and may be thus concerned not simply from the standpoint of an incompatibilist—the semi-compatibilist position precisely came to diverge from more mainstream versions of compatibilism in conceding that we could not be free to do otherwise in a determined world, yet could be responsible. Only in recent decades have these debates commonly circumvented the issue of metaphysical freedom to engage with issues of responsibility direct.

  2. 2.

    Some of the core arguments of this paper are prefigured in the important work of Carlos Moya (2006, 2007, 2010, 2011)—which I commend to the reader and have fully acknowledged in the text—along with the work of Otsuka, Widerker-Kane-Ginet & Copp. However, these arguments are augmented by and integrated with original contributions involving material in sections below entitled “Necessary, Sufficient & Counterfactually Uncertain”, “Self-Absorption and the Aims of Responsible Conduct”, “The Intensionality of Intentions” and “Non-moral Reasons? Why We Won’t be Considering Tax Evasion 3 (Much)”—and throughout, the distinction between the right and the good (especially the section entitled “Robustness, Intensionality, Right vs Good: A Tension”). However, before getting to these original contributions it will be necessary to rehearse and reincorporate Moya and others’ important work.

  3. 3.

    One that is rather different to those mentioned already is Copp’s (2003) Argument From Action-Guidingness—that the point to moral requirements is to guide agents’ decisions among their alternatives.

  4. 4.

    Contrast the position that Widerker’s thought experiment leads him to, with the position of such figures as Owens (2000) and many others in the epistemic literature. Widerker’s actionally-coupled freedom of decision is seen by him as specifically problematic—in ways in which his more purely mental deliberation is not. Yet for a really major strand of thought within the epistemic literature this view of matters is precisely inverted: a relatively unproblematic freedom of action is specifically contrasted with a supposedly impossible aspiration to attain a freedom of thought. (Cf. The doxastic voluntarism debates, among others). This matters because the free will debates are starting to have a major impact in the epistemic deontology debates—debates in which issues of ‘ought’ implies ‘can’ are of central importance.

  5. 5.

    To say “I deliberate to Φ” is an odd locution, (“deliberate” sounds like a process, not a success state) but it’s a locution forced on me by Widerker’s choice of vocabulary.

  6. 6.

    Two core objections shortly to be made against Tax Exemption 2 (“The Basic Response to Tax Evasion 2” and “Question is Begged in Describing the Device as Compelling Joe’s / Jones’ Choice”) will clearly generalise to other aspects of Tax Exemption 3; so I am not just saying “I don’t understand a key aspect of this example” and leaving it at that. Nevertheless, this section’s dilemma argument is as much an attempt to bring critical pressure to bear towards clarifying this example as it is an attempt at knock-down refutation of it. Of note is that Pereboom (2012) returns to Tax Exemption 2 (‘moral reasons’) despite, in 2009, having stated that Tax Exemption 3 was his preferred variant of the case.

  7. 7.

    This is one of the responses Moya (2011) advances against Tax Evasion 3. In advancing this response Moya is building on the argument he first used in Moya (2007) against Widerker—as remarked upon briefly in the section entitled “Decide vs Deliberate” above: that Widerker’s agent Jones could not be seen as morally responsible if Jones was not capable of responding to reasons (reasons to visit his uncle rather than stay with Mary).

  8. 8.

    In another case (‘Tax Cut’) Pereboom (2012) similarly envisages such a non-moral, yet (one assumes) prior-stage moral decision point: of his agent, Jones, being able to prosthetically bring before his imagination the non-moral considerations that he realises can alone move him (his boss finding out his immoral course of action).

  9. 9.

    Actually these matters are complicated—with this a fairly unguarded statement requiring a level of qualification which it can only partly get here. A person may be objectively ‘obliged’ to do that which he cannot do, but this objective and unattainable sense of obligation pertains to the good and not the right, and thereby not to responsibility itself and as such (hence the scare-quotes—these issues will be discussed further below). It is also true that there is a venerable and morally serious tradition in ethics, particularly in certain Christian traditions, of taking obligations as an avowedly unattainable (but action-shaping) ideal. And it is true that one may have an obligation that is currently unattainable but within one’s ‘zone of proximal development’. These are large issues and this is the nearest I can get to qualifying them here.

  10. 10.

    I owe this point to an anonymous referee; and see also Pereboom (2009, 2012)—on ‘dialectically unsatisfying’ responses to his position—of which more below. Fully dealing with this point requires a consideration of the material forthcoming in the section entitled “Robustness, Intensionality, Right vs Good: A Tension”.

  11. 11.

    A point I owe to an anonymous referee is that Frankfurt may have demurred from this recently (cf. Frankfurt 2003: 344). However, whatever Frankfurt’s current view, endorsing OIC whilst rejecting PAP was a view he held, and is a view that Kant, Wolf, and Nelkin have held and do hold. “[B]eing psychologically determined to perform good actions is compatible with deserving praise for them, but .. being psychologically determined to perform bad actions is not compatible with deserving blame” (Wolf 1990: 79), Kant 1933: A547/B575 in Nelkin 2011: 99ff & passim). This divergence between OIC and PAP most centrally has to do with issues concerning praise, not blame. Frankfurt’s point, as developed by Wolf (1990) and Nelkin (2011) is that PAP requires us to maintain that doing as we ought (being praiseworthy, behaving responsibly—even commendably) requires the ability to do otherwise: what Haji (1999) calls ‘dual control’ and McKenna sees as requiring Fischer’s regulative control. That we require such control may be called into question (early Frankfurt, Wolf and Nelkin do call it into question). This tradition sees the ‘ought’ implies ‘can’ maxim as saying merely that doing as we ought requires the ability to do as we ought, not to do otherwise than we ought: what Haji (1999) calls ‘singular control’ and McKenna sees as requiring only Fischer’s guidance control.

  12. 12.

    The point made in the sections entitled “Robustness, Intensionality, Right vs Good: A Tension” & “Intentional Objects and Action in Frankfurt Cases” below is that indeed, this is not in fact something he does—he doesn’t get as far as this decision intensionally and intentionally so-described.

  13. 13.

    We can of course, be held morally responsible for culpably unforeseeable consequences of our acts (self-deception, etc.) and were Widerker and Pereboom’s cases to incorporate such features it is likely that their and my intuitions about moral culpability would coincide—but not, I think, to the detriment of PAP or any other leeway principle. Apropos remarks above it is unclear to me whether Pereboom’s difficult cases feint in the direction of such culpability through wilful ignorance; but were they to do so, this ignorance would be something his agent, Joe, would be responsible for.

  14. 14.

    I am not committed to any account of what, epistemically, occurs at any given one of these temporal stages—this is for Pereboom / Widerker to stipulate, though if they leave it under-specified then I will try various disambiguations. If Frankfurt theorists invoke temporal stages, then to stipulate a type of epistemic access occurs at one such stage which is less than Joe deciding not to pay his taxes, then the agent—here Joe—is morally responsible to whichever level of epistemic engagement he is stipulated (by such accounts’ authors) to have reached at that stage. More of which below.

  15. 15.

    I should clarify that Pereboom clearly isn’t committed to this either—he holds that had the neurological device not been in place and Joe paid close attention to his moral reasons, he might or might not have then evaded taxes (I owe this point to an anonymous referee). But Pereboom does argue against his leeway opponent that becoming more attentive to moral reasons isn’t a robust alternative to deciding to evade taxes precisely on the grounds that one might become more attentive and still not pay one’s taxes. “What is the motivation for thinking that becoming more attentive to the moral reasons now becomes a robust alternative to deciding to evade taxes? Agreed: it is the next best action available to him. But it is not per se exempting for deciding to evade taxes, since, without the device in place, even if he did become more attentive he could decide to evade taxes” (Pereboom 2012: 304). I am arguing here that it is not a requirement on me to oppose Pereboom’s claim that “without the device in place, even if he [Joe] did become more attentive he could decide to evade taxes”—I am claiming that this becomes a counterfactual space of epistemic and metaphysical ‘maybes’ over which the indeterminist of all people does not have to take a stance. I am claiming Joe is culpable for non-attention to his moral reasons, not for non-payment of taxes.

  16. 16.

    There are very well-worked debates concerning moral luck around this point. Nagel’s German company man who is transferred to Argentina in 1929 is lucky enough to avoid being in a position where he might have become a Nazi collaborator by the 1940’s. In these cases everyday circumstances, not an evil neuroscientist, ensure that we condemn him for, say, ‘tracing case’ aretaic tendencies to conformist ultra-nationalism—but not for what might (possibly, not certainly) have become complicity in murderous acts beyond this. Were one to hypostatise a Dummettian conception of realism to condemn him for counterfactual counterpart time-lines (say, to the point where circumstantial moral luck shades into constitutive moral luck) then “who should ’scape whipping?”

  17. 17.

    See remarks on the right versus the good and the intensionality of intentions. We knew already, long before the Frankfurt debates, that consequentialism has tended, for good reason, to be the moral theory of determinism, as deontology has tended to be the moral theory of libertarianism. Forty-five years of Frankfurt arguments must establish more than that were we unable to do otherwise, we might yet be ‘responsible’ for a given level of consequentialist outcome (the good).

  18. 18.

    Pereboom, though a hard incompatibilist, is explicit across many publications that he regards agent-causal libertarianism as conceptually possible and, were it actual, as capable of underwriting a responsibility-relevant free will—so unlike, say, Galen Strawson, he’s no ‘impossibilist’. However, he has hitherto regarded the agent-causal position as untenable given our current best physical theories (though there are intriguing recent signs he may be changing his views positively in the direction of agent-causation). Various agent-causal theorists do see their theories as committing them to PAP however—so they will feel that the Pereboom thought experiments’ specific conceded libertarianism does not fix an argument that applies against their position. There is reason to believe Pereboom has in his sights an event-causal libertarianism; but note that Pereboom in his (2001) appears to have been using ‘event-causal’ in a sense that now we would understand as ‘acausal’. Certainly the language Pereboom uses in 2009 and 2011 to describe the freedom of his agent, Joe, has clear aspects of acausal libertarianism and also aspects of an older, less specific ‘freedom of indifference’ (the classic compatibilists’ nowadays more-or-less straw-man libertarianism). Widerker is an acausal teleological-intelligibility libertarian but he acknowledges (Widerker 2009: 101) that he can envisage “other versions of Frankfurt-friendly Libertarianism” (his term for a source-and-not-leeway position). The question for Widerker and Pereboom is not whether they can envisage other species of libertarianism (clearly they can and do) but whether their attempted conceptual-level refutations will work against any and every notion of PAP libertarianism their opponents envisage—and advocate.

  19. 19.

    To be blunt: there is a suspicion that the core target of this argument is simple positive indeterminism (whether acausal or event-causal, but substantially unanalysed beyond being positively undetermined: the classic ‘freedom of caprice’ or ‘indifference’—a conception of liberty that now borders on the straw-man). Any such indeterminism is already, surely, well refuted by the Mind argument. A sophisticated modern leeway incompatibilist (who may also be a source incompatibilist) may be an agent-causal theorist or a complex event-causalist or anything else. Any such theory will hold that a specific sub-set of morally relevant ‘alternative possibilities’ are (sometimes) necessary for freedom and/or moral responsibility; not just any old non-moral alternative possibilities are necessary and sufficient for such.

  20. 20.

    If this ‘levels’ terminology obscures rather than assists understanding, then please assimilate the points made without it.

  21. 21.

    Pereboom considers and rejects the view that Joe is only responsible for culpable inattention to his moral reasons, not tax avoidance, maintaining that “this is a dialectically unsatisfying response to a Frankfurt example, for the reason that it explicitly cites a leeway position in support of its verdict … Joe is non-derivatively morally responsible only for not deciding to be more attentive to the moral reasons, for only relative to this decision is a robust alternative available to him” Pereboom (2012: 306). In addition to the points made in “The Basic Response to Tax Evasion 2” section above, note: a) everyone in such examples is arguing from their intuitive judgements about moral responsibility: Pereboom is and so are his opponents. I have no intuitive judgement that in these cases Joe is morally responsible, intensionally so-described for tax avoidance—but see remarks on the confusing nature of these cases above. b) In drawing judgements as to responsibility it matters a great deal that we be precise in these judgements. Such judgements (and not just in the specialised Frankfurt literature) are drawn at the intensional and intentional level, for it is at this level that we appraise people for the right rather than the good—and it is a basic principle of justice (e.g. Copp’s ‘argument from fairness’, Otsuka’s PAB—or a cognate principle) that we do so. Pereboom (2001) himself bases his hard determinism most generally on such principles (that it would be unfair to blame people in a determined world). c) Pereboom’s case is meant to reanimate the anti-PAP literature to victory not stalemate. d) Pereboom himself argues for non-robustness (for him, a crucial technical term/device) on the basis of the epistemic opacity (intentional and intensional) of the agent’s actional outcome to himself: Joe cannot foresee that his action would absolve him of moral responsibility. For all these reasons I think the claim that this response is dialectically unsatisfying is itself dialectically unsatisfying.

  22. 22.

    Were we dealing with versions of these arguments developed prior to Frankfurters carefully seeking to obviate the WKG dilemma, both this ‘could’ and that of the last sentence would have been ‘would’s. Note that, in light of the need of the Frankfurters to de-fang the ‘question-begging’ horn of that dilemma, even the later cases’ Frankfurt agents do have to be credited with possessing the strong libertarian power stated in the text (“Question is Begged in Describing the Device as Compelling Joe’s / Jones’ Choice” section after Lucas): the power not to kill.

  23. 23.

    Despite being in sympathy with Otsuka’s (1998) position generally, I would, for these reasons be wary of embracing his choice of terminology (‘PAB’: principle of avoidable blame).

  24. 24.

    I should like to record my gratitude to Carlos Moya who generously made his (then) unpublished work available for me to read and whose philosophical influence (and in many cases, priority) should be clear in this paper. I should also like to record my gratitude to Dana Nelkin for reading and providing generous comments on an earlier version of this paper. Finally, I should like to record my gratitude to an anonymous referee for this journal for acute, incisive comments on earlier drafts.


  1. Aristotle. (1999). Physics Book VIII, Clarendon Aristotle Series, tr. Daniel W. Graham, Oxford: Clarendon Press.

  2. Bealer, G. (1996a). A priori knowledge and the scope of philosophy. Philosophical Studies, 81, 121–142.

  3. Bealer, G. (1996b). On the possibility of philosophical knowledge, Noûs, vol. 30, Supplement: Philosophical Perspectives, 10, Metaphysics 1–34.

  4. Chisholm, R. (1964). Human freedom and the self. In G. Watson (Ed.), Free Will, Oxford Readings in Philosophy 2nd edn. Oxford: Oxford University Press 2003: 3–15.

  5. Copp, D. (2003). “Ought” implies “can”, blameworthiness, and alternative possibilities. In D. Widerker & M. McKenna (Eds.), Freedom, responsibility and agency: Essays on the importance of alternative possibilities (pp. 265–299). Aldershot: Ashgate Press.

  6. Dummett, M. (1978/1963). ‘Realism’, in his Truth and Other Enigmas Cambridge, MA: Harvard University Press 145–65.

  7. Frankfurt, H. (1969). ‘Alternate possibilities and moral responsibility’; in his (1988).

  8. Frankfurt, H. (1983). ‘What we are morally responsible for’; in his (1988).

  9. Frankfurt, H. (1988). The importance of what we care about. Cambridge: CUP.

  10. Frankfurt, H. (2003). Some thoughts concerning PAP. In D. Widerker & M. McKenna (Eds.), Freedom, responsibility and agency: Essays on the importance of alternative possibilities (pp. 339–345). Aldershot: Ashgate Press.

  11. Ginet, C. (1996). In defense of the principle of alternative possibilities: Why I don’t find Frankfurt’s argument convincing. Philosophical Perspectives, 10, 403–417.

  12. Haji, I. (1999). Moral anchors and control. Canadian Journal of Philosophy, 29(2), 175–203.

  13. Kane, R. (1996). The significance of free will. Oxford: Oxford University Press.

  14. Kant, I. (1933). Critique of pure reason, tr. N. Kemp-Smith, Basingstoke Hants & London: Macmillan.

  15. Lockie, R. (2003). Depth psychology and self-deception. Philosophical Psychology, 16(1), 127–148.

  16. Lucas, J. R. (1970). The freedom of the will. Oxford: Oxford University Press.

  17. Mele, A. (1997). Real self-deception. Behavioral and Brain Sciences, 20, 91–136.

  18. Mele, A. (2001). Self-deception unmasked. Princeton: Princeton University Press.

  19. Moya, C. (2006). Moral responsibility. Oxford: Routledge.

  20. Moya, C. (2007). Moral responsibility without alternative possibilities?’. Journal of Philosophy, 104(9), 475–486.

  21. Moya, C. (2010). ‘Blame, moral obligation and alternative possibilities’, research talk: Spanish Society for Analytic Philosophy (SEFA). Tenerife: Universidad de La Laguna.

  22. Moya, C. (2011). On the very idea of a robust alternative. In Crítica, Revista Hispanoamericana Filosophía, 41 (August): 3–26.

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

  24. Otsuka, M. (1998). Incompatibilism and the avoidability of blame. Ethics, 108(4), 685–701.

  25. Owens, D. (2000). Reason without freedom. London: Routledge.

  26. Pereboom, D. (2000). Alternative Possibilities and Causal Histories. Philosophical Perspectives 14, papers on Action and Freedom, James E. Tomberlin (ed.): 119–37.

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

  28. Pereboom, D. (2009). Further thoughts about a Frankfurt-style argument. Philosophical Explorations, 12(2), 109–118.

  29. Pereboom, D. (2011). Free will, skepticism and meaning in life. In R. Kane (Ed.), Oxford handbook of free will (2nd ed.). New York: Oxford University Press.

  30. Pereboom, D. (2012). Frankfurt examples, derivative responsibility, and the timing objection. Philosophical Issues: A Supplement to Nous, 22(1), 298–315.

  31. Smart, J. J. C. (1959). Sensations and brain processes. Philosophical Review, 68(2), 141–156.

  32. Watson, G. (2003) (ed.) Free Will, Oxford Readings in Philosophy 2nd edn. Oxford: Oxford University Press.

  33. Widerker, D. (1995). Libertarianism and Frankfurt’s attack on the principle of alternative possibilities. Philosophical Review, 104(2), 247–261.

  34. Widerker, D. (2000). Frankfurt’s attack on the principle of alternative possibilities: a further look. Philosophical Perspectives 14, papers on Action and Freedom, James E. Tomberlin (ed.): 181–201.

  35. Widerker, D. (2006). Libertarianism and the philosophical significance of Frankfurt scenarios. Journal of Philosophy, 103, 169–187.

  36. Widerker, D. (2009). A defense of Frankfurt-friendly libertarianism. Philosophical Explorations, 12(2), 87–108.

  37. Widerker, D. (2011). Frankfurt-style libertarianism. In R. Kane (Ed.), Oxford handbook of free will (2nd ed.). New York: Oxford University Press.

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

Download references

Author information

Correspondence to Robert Lockie.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Lockie, R. Three Recent Frankfurt Cases. Philosophia 42, 1005–1032 (2014). https://doi.org/10.1007/s11406-014-9530-1

Download citation


  • Moral responsibility
  • Alternative possibilities
  • PAP
  • Frankfurt
  • Flicker of freedom
  • Right versus Good