“[B]ut to put compulsion on the gods against their will - no man can do that.”
— Sophocles, Oedipus Rex.
Abstract
It is now the majority view amongst philosophers and theologians that any world could have been better. This places the choice of which world to create into an especially challenging class of decision problems: those that are discontinuous in the limit. I argue that combining some weak, plausible norms governing this type of problem with a creator who has the attributes of the god of classical theism results in a paradox: no world is possible. After exploring some ways out of the paradox, I conclude that the classical theist should accept Marilyn Adams’s view that no norms (of morality or of rationality) apply to gods.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
I owe thanks to many people for helpful comments and conversations about this paper, including Eddy Chen, Philip Swenson, Andrew Moon, Sam Lebens, Nevin Climenhaga, Brian Cutter, Blake McAlister, Liz Jackson, John Barker, anonymous referees, and audiences at the 2015 Midwest Society of Christian Philosophers, 2015 St. Thomas Summer Seminar, 2016 Eastern APA, Rutgers Center for Philosophy of Religion Reading Group, and Keith DeRoses’s Freedom and Foreknowledge Seminar. I owe special thanks to Dean Zimmerman, Robert Adams, Marilyn McCord Adams, and David Black for extensive feedback on multiple drafts.
Thanks to David Black for showing me how this can be done.
Even deontologists and virtue ethicists must make choices. Using the tools of revealed preference theory, we could by offering constant decisions recover their rankings. The rankings would not be determined by the normative properties their decisions are sensitive to, and not intrinsic goodness, of course, but they would provide the required formal structures.
See Buchak (2014) for an illuminating discussion.
Rowe (1994).
These can be thought of as consistent, maximal-modulo-God situtations.
Arntzenius et al. (2004).
Bartha et al. (2014).
Roughly, this says: if a is a choice such that some course of action excluding it dominates every course of action including it, don’t choose a.
Barker et al. characterize these general puzzles more precisely. Letting the {A\(_i\}\) be a series of strategies, we are in the following sort of situation (where \(\rightarrow\) takes us to the limit):
$$\begin{aligned}&A_0< A_1< A_2< A_3 <\cdots \rightarrow A;\,\hbox {but}\\&\hbox {lim}_{i\rightarrow \infty }U(A_i) > U(A) \end{aligned}$$Which is to say: the utility of the limit of the series of strategies is lower than the limit of the series of the utilities of the strategies. Consequently, there is no non-dominated strategy (and a fortiori no optimal strategy) available to Eve or to the recipient of St. Peter’s Offer.
Because all we have is a comparative ordinal ranking, choice of 0 is arbitrary. I’ve chosen the lonely world because it’s a ‘baseline’ from which God’s creation either adds or subtracts value.
Where \(>_D\) means dominates.
A referee suggested an alternative formulation: a world w is possible iff w could be actualized supposing God wanted to. I suspect that my argument would go through using this formulation as well. The main contribution of classical theistic plenitude is to rule impermissible worlds impossible, and therefore the strategies which involve creating them unavailable. This depends primarily on the doctrine of divine impeccability, and so long as that doctrine is in place, both formulations of the principle do the needed work.
As Sect. 7.1 notes, an argument similar to mine could be run using only the essentiality of divine goodness and rationality as a premise. Thus, certain denials of classicla theistic plenitude will not completely solve the problems raised. I will stick with the full version in this presentation for simplicity and because it is the more common commitment.
Readers interested in working out the details can find the mathematical background in Jech (2006).
We did not use iia in this presentation, but it is required for the version of the argument applied to non-total orders, dense orders, or in dealing with limit ordinals.
I owe this extension of the argument to David Black.
It is likely that possible world-seeds will receive their values in the same way that lotteries do. This can get mathematically strenuous when dealing with infinite sets of worlds or worlds of infinite value. See Colyvan and Hajek (2016).
Swinburne (1994).
Howard-Snyder and Howard-Snyder (1994, 1996) and Rowe (1994). In brief, Rowe argues: if there is no best world, then God must create one that is less than best. So God must have a standard for what counts as minimally creatable. Whatever standard God adopts is improvable: it might have excluded more worlds. Better Gods have better standards. But since there is no best God, there is no best standard. So God can’t be perfect, and therefore does not exist. My own response to this argument should be clear: God neither has nor needs a standard of creatability. Everything is fair game.
Hurka (1990).
We could, as the Howard-Snyeders do, try more sophisticated definitions of the threshold. Perhaps we simply choose a world as ‘good enough,’ and allow any action that leaves the world as good as or better than that world. Coupled with the ordinal structure of the goodness of worlds, this will make finding infinitely-ascending chains of worlds below this one to construct something like a modified blank check case difficult. But it will have other unhappy results. Suppose my world is very good, so good that nothing I can do could push it below the threshold. Perhaps I inhabit a large multiverse, an infinity of individual universes within which surpass the threshhold, but in which there are only a few people in my own spacetime. The rest of the multiverse is so good that even if I kill everyone I can - producing a large net negative of evil - enough wonderful disconnected universes should be able to balance it off. So I am free to behave as badly as I please. But living in a good world is no excuse for poor behavior. Hurka (2004) presses a line similar to this one in arguing that satisficing only seems desirable when ‘subjective’ values are concerned. And clearly, in the case we care about, objective values (if any there be) are involved.
References
Arntzenius, F., Elga, A., & Hawthorne, J. (2004). Bayesianism, infinite decisions, and binding. Mind, 113(450), 251–283.
Arrow, K. L. (1963). Social choice and individual values (2nd ed.). New Haven: Yale University Press.
Bartha, P., Barker, J., & Hajek, A. (2014). St. Petersburg, Saint Peter’s Offer, and Satan’s Apple. Synthese, 191(4), 629–660.
Buchak, L. (2014). Risk and rationality. Oxford: OUP.
Colyvan, M., & Hajek, A. (2016). Making ado without expectations. Mind, 125(499), 829–857.
Howard-Snyder, D., & Howard-Snyder, F. (1994). How an unsurpassable being can create a surpassable world. Faith and Philosophy, 11, 260–68.
Howard-Snyder, D., & Howard-Snyder, F. (1996). The real problem of no best world. Faith and Philosophy, 13, 422–25.
Hurka, T. (1990). Two kinds of satisficing. Philosophical Studies, 59(1), 107–111.
Hurka, T. (2004). Satisficing and substantive values. In M. Byron (Ed.), Satisficing and maximizing: Moral theorists on practical reason. Cambridge: Cambridge University Press.
Jech, T. (2006). Set theory. New York: Springer.
Kraay, K. (2013). Can god satisfice? American Philosophical Quarterly, 50, 399–410.
Langtry, B. (2008). God, the best, and evil. Oxford: OUP.
McCord Adams, M. (1999). Horrendous Evils and the goodness of god. Ithaca: Cornell University Press.
McCord Adams, M. (2006). Christ and Horrors: The coherence of christology. Cambridge: Cambridge University Press.
Rowe, W. (1994). The problem of no best world. Faith and Philosophy, 11, 269–71.
Sen, A. (1970). Collective choice and social welfare (2nd ed.). New York: North Holland.
Sullivan, M. (2013). Peter van Inwagen’s defense. In D. Howard-Snyder & J. McBrayer (Eds.), The Blackwell guide to natural theology (pp. 396–410). Oxford: Wiley.
Swinburne, R. (1977). The coherence of theism. Oxford: OUP.
Swinburne, R. (1979). The existence of god. Oxford: OUP.
Swinburne, R. (1994). The Christian god. Oxford: OUP.
van Inwagen, P. (2006). The problem of evil. Oxford: OUP.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Rubio, D. God meets Satan’s Apple: the paradox of creation. Philos Stud 175, 2987–3004 (2018). https://doi.org/10.1007/s11098-017-0991-5
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11098-017-0991-5