Skip to main content

Timelessness and freedom


One way that philosophers have attempted to defend free will against the threat of fatalism and against the threat from divine beliefs has been to endorse timelessness views (about propositions and God’s beliefs, respectively). In this paper, I argue that, in order to respond to general worries about fatalism and divine beliefs, timelessness views must appeal to the notion of dependence. Once they do this, however, their distinctive position as timelessness views becomes otiose, for the appeal to dependence, if it helps at all, would itself be sufficient to block worries about fatalism and divine beliefs. I conclude by discussing some implications for dialectical progress.

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


  1. For a discussion of similar arguments, see the introduction to Fischer and Todd (2015). The first premise and the conclusion of this argument include what we might call a “no choice operator” (cf. van Inwagen 1983), which may be read as claiming that a certain fact obtains (e.g., in the conclusion, the fact of your reading this paper at t) and that you had no choice about that fact’s obtaining. The conclusion of the argument concerns the freedom to do otherwise; not having a choice about the fact that you perform some action X is roughly equivalent to not being able to do otherwise than X. (As should be clear, there is nothing special about the times and action in question, so the argument generalizes to the fatalistic conclusion that we never have a choice about anything we do.) Arguments like this one rely on a “transfer principle” that allow us to infer (3) from (1) and (2): “you have no choice about what necessarily follows from what you have no choice about” (Todd and Fischer 2015: p. 3). I will not evaluate this principle here.

  2. Here and throughout the paper, whenever a claim about one of God’s beliefs is exactly parallel to a claim about a true proposition, I will use the same number for the claim but will add an asterisk.

  3. As an anonymous reviewer points out, one might worry that a timelessness theorist who took God to be necessarily timeless might reject the validity of the argument. On this timelessness view, God’s believing something at a time is impossible, and since we do not have a choice about anything impossible, this would apparently make the first premise true. Moreover, the second premise would be trivially necessarily true since the conditional would have a necessarily false antecedent. Finally, the conclusion is false (or at least the timelessness theorist hasn’t been given any new reason to think that it’s true).

    I believe that this concern can be avoided by stipulating that the “no choice operator” at work in the first premise and the conclusion be factive. The recipe for this alleged counterexample to the validity of the argument requires that you have no choice about the “fact” that God believed 1000 years ago that you would read this paper at t. But, according to our timelessness theorist, it is not a fact that God believed 1000 years ago that you would read this paper at t. So if it is impossible that God believed 1000 years ago that you would read this paper at t (because, say, God is necessarily timeless), then the first premise of the argument is false, and thus there is no counterexample.

    If one finds this response unsatisfactory, one could replace the second premise, which is a conditional, with the following biconditional: Necessarily, God believed 1000 years ago that you would read this paper at t iff you read this paper at t. Even if (contrary to what I’ve just argued) God’s being necessarily timeless would render the first premise true, this modified second premise would be false on this view, and thus the alleged counterexample would disappear. (Thanks to the same anonymous reviewer for suggesting this solution.) As nothing in what follows hangs on which of these versions of the premise is used, I will keep the original formulation of the premise.

  4. Thanks to an anonymous reviewer for encouraging me to clarify. As this reviewer also points out, it is difficult to see the difference between these two views, and this may suggest that endorsing a timelessness view is insufficient to escape the fatalist’s worry, a point that I will defend with a different argument in what follows.

  5. For a discussion of a similarly modified argument, see Zagzebski (1991, chapter 2).

  6. There are other responses to both worries, but these deny that there are now true propositions about future contingents, or that God has exhaustive foreknowledge. These are sometimes called the “Aristotelian” and “Open Theist” responses, respectively. There are also certain compatibilists (semicompatibilists) who need not respond to these arguments since, on their view, we can be free in the sense required for moral responsibility even if we lack the freedom to do otherwise, which is the freedom that is at stake in these arguments.

  7. Not all authors use the “having a choice about” locution in articulating this view, but, for reasons I explained in note 1 (about the sense of “having a choice about” in the argument for fatalism), I do not think these differences in articulations amount to a real difference in views. For example, in developing his version of this view (which appeals to explanatory dependence), Swenson’s “principle of the independent past” refers to what an agent “can” do: “An agent S can (at time t in world w) do X at t only if there is a possible world w* in which all of the facts in w up to t that do not explanatorily depend on S’s choice(s) at t hold and S does X at t” (2016: 662). What an agent can, is able, or is free to do in a certain world and at a certain time is precisely what is at stake in evaluating what an agent has a choice about (in the sense relevant to the argument for fatalism).

  8. This move is also developed by Swenson (2016) and Wetsphal (2011).

  9. The answer, I argue below, is that the account must be supplemented with an appeal to dependence. As it turns out, Ayer goes on to endorse a weak notion of dependence (namely counterfactual dependence) to secure freedom, indicating that he saw the need to say more than that propositions are outside of time. In my view, Ayer was right to see that a claim about dependence was needed, though, as we have seen, counterfactual dependence turns out to be too weak to do the job.

  10. In Sect. 2, I mentioned a few ways that responses to fatalistic arguments might appeal to dependence, and I noted that the notions of counterfactual dependence and the dependence involved in the entailment relation are arguably too weak. In what follows, I use ‘dependence’ to refer to one of the stronger notions of dependence mentioned in Sect. 2, at least as strong as Merricks’s notion of dependence.

  11. Perhaps this is why van Inwagen (1983) uses “has a choice about p” and “can render p false” interchangeably throughout.

  12. An alternative way to motivate (8) would be, first, to consider examples of actions that someone putatively has a choice about and, second, to show that the truth of propositions about that person’s actions therefore depends on what she does. I set aside this strategy here.

  13. Whether there is any further explanation for this dependence relation (or whether it is brute) is a separate question; I am only claiming that the correlation between the content of certain propositions, on the one hand, and what you do, on the other, calls out for explanation.

  14. This is not to say, of course, that there are not reasons independent of worries about fatalism for preferring a timelessness view of propositions to the view that propositions are in time. All I have argued is that appealing to timelessness is otiose in responding to worries about fatalism.

  15. Thanks to Andrew Law for helpful discussion of this potential objection to my argument.

  16. Alternatively, you could suppose that I form a true belief about what you will do at t. See Merricks (2009: p. 35) for this suggestion.

  17. Suppose someone says: “No, your claim is not true/false, since only propositions are the bearers of truth/falsity, and these are outside of time. Instead, your claim merely expresses a timelessly true proposition, which is not true in the past.” In response, I would ask, is it true that my claim expresses, before t, a timelessly true proposition? Clearly it is, but that fact about the past (the fact that my claim expressed, before t, a timelessly true proposition) does not imply that you have no choice about reading at t.

  18. For developments and defenses of this view, see Stump and Kretzmann (1981, 1991) and Leftow (1991).

  19. In my own view, my conclusion should be welcomed by timelessness theorists, for it seems the most natural way to explain their frequent appeal to God’s “awareness” of what occurs in time. In fact, Stump and Kretzmann (1991: pp. 418–419) come very close to endorsing a dependence account when explaining how God’s timeless beliefs are innocuous with respect to the freedom of human actions:

    For God, who timelessly sees contingent events future to us when and as they are temporally present, those events have the sort of inevitability that accompanies presentness, and only that sort. For us, relative to whom they are future, those events are as evitable now as the presently occurring contingent events were evitable when they were future. Nothing in God’s relationship to those events determines them in advance any more than our observing [some human agent’s] present actions would render those actions of hers unfree.

  20. For some of these objections, see Todd (2013) and Fischer and Tognazzini (2014). I argue in an unpublished manuscript (“Atemporalism”) that the timelessness view about God’s beliefs is vulnerable to exactly these objections.

  21. In particular, classical theists may deny that God depends in any way (even perhaps in what he knows) upon anything else. In addition, according to certain views of divine providence (such as Molinism), God’s knowledge of our behavior is in some sense “prior” to our actual behavior (and so does not depend on what we actually do). Of course, such theists will need to address the threat to freedom from divine beliefs, but perhaps they will not be inclined to take either the sempiternalist nor the timelessness version of the view that appeals to dependence. (Insofar as they reject certain alternatives, such as Open Theism, however, such theists may, in the end, need to give up the type of freedom at issue in this paper.) I am indebted to an anonymous reviewer for raising these points.

  22. Again, see Mackie (2003) and Merricks (2009) for typical presentations.

  23. Again, as I noted in note 6, there are alternative responses to these worries (the “Aristotelian” and “Open Theist” positions) that deny that there are now true propositions about future contingents, or that God has exhaustive foreknowledge, but we have set these aside here.


  • Ayer, A. J. (1963). Fatalism. In A. J. Ayer (Ed.), The concept of a person and other essays (pp. 235–304). London: Macmillan.

    Chapter  Google Scholar 

  • Fischer, J. M., & Todd, P. (Eds.). (2015). Freedom, fatalism, and foreknowledge. New York: Oxford University Press.

    Google Scholar 

  • Fischer, J. M., & Tognazzini, N. (2014). Omniscience, freedom, and dependence. Philosophy and Phenomenological Research, 88, 346–367.

    Article  Google Scholar 

  • Leftow, B. (1991). Time and eternity. Ithaca: Cornell University Press.

    Google Scholar 

  • Mackie, P. (2003). Fatalism, incompatibilism, and the power to do otherwise. Noûs, 37, 672–689.

    Article  Google Scholar 

  • Merricks, T. (2009). Truth and freedom. The Philosophical Review, 118, 29–57.

    Article  Google Scholar 

  • Stump, E., & Kretzmann, N. (1981). Eternity. The Journal of Philosophy, 78, 429–458.

    Article  Google Scholar 

  • Stump, E., & Kretzmann, N. (1991). Prophecy, past truth, and eternity. In J. Tomberlin (Ed.), Philosophical Perspectives (Vol. 5, pp. 395–424). Atascadero: Ridgeview.

    Google Scholar 

  • Swenson, P. (2016). Ability, foreknowledge, and explanatory dependence. Australasian Journal of Philosophy, 94, 658–671.

    Article  Google Scholar 

  • Todd, P. (2013). Prepunishment and explanatory dependence: A new argument for incompatibilism about foreknowledge and freedom. The Philosophical Review, 122, 619–639.

    Article  Google Scholar 

  • Todd, P., & Fischer, J. M. (2015). Introduction. In J. M. Fischer & P. Todd (Eds.), Freedom, Fatalism, and Foreknowledge (pp. 1–38). New York: Oxford University Press.

    Google Scholar 

  • van Inwagen, P. (1983). An essay on free will. Oxford: Clarendon.

    Google Scholar 

  • Wetsphal, J. (2011). The compatibility of divine foreknowledge and freewill. Analysis, 71, 246–252.

    Article  Google Scholar 

  • Zagzebski, L. T. (1991). The dilemma of freedom and foreknowledge. New York: Oxford University Press.

    Google Scholar 

Download references


For comments on earlier drafts of this paper, I am grateful to Andrew Law, Jonah Nagashima, Michael Nelson, and Jared Smith. Thanks especially to Andrew Law and Jonah Nagashima for helpful discussion throughout the process of writing this paper.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Taylor W. Cyr.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Cyr, T.W. Timelessness and freedom. Synthese 197, 4439–4453 (2020).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI:


  • Dependence
  • Divine beliefs
  • Fatalism
  • Propositions
  • Timelessness