Skip to main content

The Timing Objection to the Frankfurt Cases


According to the principle of alternative possibilities (PAP), a person is morally responsible for what he has done only if he could have done otherwise. Pereboom (Living without free will, Cambridge University Press, Cambridge, 2001, Midwest Studies in Philosophy 29:228–247, 2005) has developed an influential version of a Frankfurt case, known as “Tax Evasion,” which he believes is a counterexample to PAP. Ginet (Journal of Ethics 6:305–309, 2002) raises a key objection against Pereboom’s case, known as “the timing objection.” The main claim of the timing objection is that we need to pay close attention to the precise time at which people act in order to determine what they can, and cannot, be morally responsible for. Recently, Pereboom (The philosophy of free will, Oxford University Press, New York, 2012) has defended his position against this objection by developing a new Frankfurt case called “Tax Cut.” In this paper, I assess this new development. I argue that Tax Cut is just as vulnerable to the timing objection as Pereboom’s original case, Tax Evasion. Thus, Pereboom’s response to the timing objection fails, leaving PAP intact from the threat of both of his Frankfurt cases. Along the way, I further motivate and develop the timing objection, and explore the distinction between derivative and non-derivative responsibility.

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


  1. Pereboom initially presented this example in his 2001 book, Living Without Free Will, but he regards his later statement of the example, in his 2005 paper, as its definitive presentation.

  2. In his most recent work, Pereboom (2008, 2012) has amended his robustness condition. In his 2008 article, he argues that an alternative possibility is robust if and only if the individual “could have willed something different from what she actually willed such that she understood that by willing it she would be, or at least would likely be, precluded from the moral responsibility she actually has” (p. 5). In his 2012 article, Pereboom gives a similar condition for robustness (p. 12). But these amendments do not affect my argument. My argument undermines Pereboom’s case using either of the versions of his robustness conditions.

  3. Why does Pereboom concede that Joe has the alternative possibility to bring to bear moral reasons? Can’t he argue that Joe doesn’t even have this alternative possibility? After all, can’t we imagine that Black can detect Joe’s being about to bring to bear moral reasons at t1 and—having noticed this—intervene and “make” him decide to evade taxes at t1 instead? The answer is, “no,” given the stipulations of the case. Since Joe is, generally speaking, a libertarian free agent, none of his decisions or actions are causally determined. So, in the counterfactual situation in which Joe bring to bear moral reasons at t1 rather than—as he actually does—decide to evade taxes at that instant, there will be no antecedent deterministic sign indicating what Joe will do at t1, so no way for Black to “know,” prior to t1, what Joe will do at t1.

  4. Though I won’t develop the point here, this “timing objection” to Pereboom’s case would (if sound) also undermine a structurally identical Frankfurt case developed by David Hunt (2005). In fact, I think it’s plausible to argue that this timing objection will (if sound) undermine any Frankfurt case in which it’s within the agent’s power to do otherwise than act as he does at that very time.

  5. Notice that Joe can’t avoid deciding to vote for the tax cut by t2 by not making any decision up to t2 and then imagining the no promotion scenario at t2. This is because (i) he’s committed with certainty to deciding one way or the other by t2, and (ii) he understands that having any non-decision thought at t2—like imagining the no promotion scenario—precludes him from deciding one way or the other at t2.

  6. In the paper so far, I’ve argued that Joe can’t be responsible for A-ing simpliciter (hereafter, simply A-ing) since he couldn’t have done otherwise than A, though he can be responsible for A-ing at t since he could have done otherwise than A at t. In reply, suppose that Joe satisfies all plausible freedom-relevant considerations for A-ing: he’s not coerced or compelled; perhaps he’s an agent-cause of his A-ing or perhaps his A-ing is not caused at all (thus accommodating proponents of agent-causal or non-causal views of free will), and so on. Given this, it’s plausible to think Joe freely A-ed. But if he freely A-ed, and if he satisfied all the plausible non-freedom-relevant considerations for responsibility (e.g., he knew that he was A-ing, and so on), then why shouldn’t we say that he’s morally responsible for A-ing?

    I see two ways to respond. First, we might adopt a fine-grained account of action-individuation according to which the time at which a person acts is an essential part of that act token and, hence, deny that Joe freely A-ed and insist that, at most, he freely A-ed at t. On this account of action-individuation, it doesn’t make sense to speak of Joe freely A-ing ‘outside’ of the time at which he A-ed. For the time at which the action was performed, in part, makes it the particular action that it is. Alternatively, we might grant that Joe freely A-ed but insist that he can only be responsible for A-ing at t (rather than for A-ing) because saying this allows us to preserve two key intuitions: that moral responsibility requires alternative possibilities (i.e., PAP) and that there’s something for which Joe is morally responsible in Pereboom’s example. I concede that both of these responses require further development (which I hope to undertake in future work). But I hope they’re sufficient to parry the criticism, at least in the sense that they show I have things to say in response, and that the criticism is not, therefore, “knock-down.”

    An anonymous reviewer helpfully suggested that the proponent of the timing objection might revise PAP as follows and develop a modified timing objection as a result: a person is morally responsible for A-ing (rather than for A-ing at t) only if, at or immediately prior to the time he A-ed, it was within his power to do otherwise than A at that time. Pereboom’s case is not a counterexample to this version of PAP for although Joe can be responsible for A-ing (and not just for A-ing at t) it was within his power at t to do otherwise than A at t. On this version of the timing objection, the agent is morally responsible for his action (and not just for performing his action at a certain time), but he could have done otherwise at that time than perform that action at that time (even though he would eventually perform the action as a result of Black’s intervention if he hadn’t have performed it at t).

  7. Why is it if Joe waits to decide to vote for the tax cut at the very last instant at which it’s possible for him to do so—namely, at t2—then he will lack a robust alternative to deciding to vote for the tax cut? It’s because in order for him to decide to vote against the tax cut, it’s necessary that he vividly imagine the no-promotion scenario. And, crucially, Pereboom assumes that Joe cannot decide to vote against the tax cut at the very same time at which he imagines the no-promotion scenario. If this were possible—that is, if it were possible for Joe, at t2, to both imagine the no-promotion scenario and decide to vote against the tax cut—then Joe would have a robust alternative at t2 after all. For the sake of argument, I will assume, with Pereboom, that Joe’ imagining the no-promotion scenario must temporally precede his deciding to vote against the tax cut. And, so, I’ll grant Pereboom the claim that Joe does not have a robust alternative, at t2, to deciding to vote for the tax cut.

  8. It’s plausible to think that given Joe’s commitment to deciding one way or the other by t2, and by his failure to make a decision before then or to bring to mind the thoughts needed for him to decide otherwise, it is causally determined that Joe decides to vote for the tax cut at t2. Now, he might be derivatively responsible for deciding at t2, as I have suggested, but given that Joe is causally determined, incompatibilist supporters of PAP will argue that Joe cannot be morally responsible for deciding at t2, derivatively or otherwise.

  9. David Hunt and Seth Shabo (2012) offer a further argument against the claim that the times at which people act are an essential or necessary part of the thing for which they’re responsible. I plan to address their argument in future work. My aim in this article is just to respond to Pereboom’s criticism of the timing objection.


  • 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.

    Google Scholar 

  • Ginet, C. (2002). Review of Pereboom’s Living without free will. Journal of Ethics, 6, 305–309.

    Article  Google Scholar 

  • Hunt, D. (2005). Moral responsibility and buffered alternatives. Midwest Studies in Philosophy, 29, 126–145.

    Article  Google Scholar 

  • Hunt, D., & Shabo, S. (2012). Frankfurt cases and the (in)significance of timing: A defense of the buffering strategy. Philosophical Studies. doi:10.1007/s11098-012-9874-y.

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

    Book  Google Scholar 

  • Pereboom, D. (2005). Defending hard incompatibilism. Midwest Studies in Philosophy, 29, 228–247.

    Article  Google Scholar 

  • Pereboom, D. (2008). Defending hard incompatibilism again. In N. Trakakis & D. Cohen (Eds.), Essays on free will and moral responsibility (pp. 1–33). Newcastle: Cambridge Scholars Press.

    Google Scholar 

  • Pereboom, D. (2012). Optimistic skepticism about free will. In P. Russell & O. Deery (Eds.), The philosophy of free will: Selected Contemporary Readings. New York: Oxford University Press. Available at (forthcoming).

  • Van Inwagen, P. (1983). An essay on free will. Oxford: Oxford University Press.

    Google Scholar 

Download references


I would like to thank two anonymous reviewers for very helpful comments on an earlier draft of this article.

Author information

Authors and Affiliations


Corresponding author

Correspondence to David Palmer.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Palmer, D. The Timing Objection to the Frankfurt Cases. Erkenn 78, 1011–1023 (2013).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Moral Responsibility
  • Moral Reason
  • Alternative Possibility
  • Timing Objection
  • Frankfurt Case