Abstract
Daniel Dombrowski contends that: (1) a number of versions of the ontological argument [OA] are sound; (2) the deity whose existence is most well established by the OA is the deity picked out by Hartshorne’s neoclassical concept of God; (3) skeptics who insist that the OA only shows that “if God exists, then God exists necessarily” are contradicting themselves, and (4) the OA is worth a great deal since it effectively demonstrates the rationality of theism. I argue that theses (2) and (3) are clearly false and offer a presumptive case for thinking that (4) is false, since, absent an independent proof of God’s existence, the theist appears to be in no position to rationally assert (1). I also show that the Anselmian OA harmonizes rather poorly with a Hartshornean neoclassical conception of God. I conclude by assessing the philosophical and dialectical worth of ontological arguments vis-à-vis establishing the rationality of theism.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
See Dore (1989).
It is worth stressing that the logic of perfection requires that we conceive of the greatest and most perfect being conceivable period, not the greatest and most perfect being that we can conceive of whose attributes happen to be consistent with the sorry state of the world.
That COND has any air of paradox or apparent contradictoriness derives from the fact that it is stated in the object language. To speak carefully and accurately, all existential claims should be put metalinguistically. There is nothing paradoxical or contradictory about either of the following more accurate ways of stating COND:
COND*: If the concept of “God” is instantiated in the actual world, then the concept of “God” is instantiated in all possible worlds.
COND**: If the word ‘God’ refers in the actual world, then the word ‘God’ refers in all possible worlds.
According to Plantinga, maximal excellence is a world-bound property in that whether or not a being is maximally excellent in W depends only upon its (non-world-indexed) properties in W (1974a, p. 214).
D2 defines maximal greatness across possible worlds because, as Plantinga sees it, a being’s greatness in W is a function not only of its excellence in W, but also of its excellence in every other world (1974a, p. 214).
D1 and D2 are somewhat oversimplified. Strictly speaking, Plantinga maintains that maximal greatness entails maximal excellence in every possible world and maximal excellence entails omniscience, omnipotence and moral perfection (where a property P entails a property Q if there is no world in which there exists an object x that has P but lacks Q) (1974a, p. 214). Since nothing in the paper turns on this subtlety, I will treat the definiens in D1 and D2 as equivalences of their respective definienda, rather than as mere necessary conditions.
Properties exist whether they are instantiated or not. The property of being round and square exists, but it is an impossible property, because it is impossible for any figure to be simultaneously round and square. So, to speak of property possibility is not to speak about whether it is possible for that property to exist (it already exists); rather, it is to speak about whether it is possible for that property to be instantiated.
For, by definition, x is necessarily ϕ iff x is ϕ in every possible world in which x exists.
For people for whom the rational acceptability of POA1 depends upon the prior rational acceptability of theism, Plantinga’s argument does no more to increase the rational acceptability of theism than the following argument:
Either 7 + 5 = 13 or God exists
7 + 5 ≠ 13
Therefore, God exists,
which Plantinga himself considers and rightly dismisses as “dialectically deficient” (1974a, pp. 217, 218).
Hacking (1975, p. 325). As stated, Hacking’s analysans is only offered as a sufficient condition of the ϕ-possibility of A’s x-ing, but as far as I can tell it is also a necessary condition of such possibility, and that is how I will understand Hacking’s suggestion in the rest of the paper. Lloyd Reinhardt, following Hacking, makes a somewhat similar suggestion. See Reinhardt (1978, p. 217).
For a formal demonstration that consistency is not sufficient for possibility, see Loewer (1978, p. 108).
Rowe (1987, p. 71).
In asserting “For all we know, M is instantiated, and for all we know, L is instantiated,” I am assuming that no one knows (at least no one knows with certainty) that God exists and that no one knows (with certainty) that God does not exist. Certain knowledge of the former would epistemically preclude L, and certain knowledge of the latter would epistemically preclude M. Note: To rule out the e-possibility of either M or L in this way would require prior knowledge of the truth-value of POA’s conclusion (POA8).
Note: In order for Plantinga’s argument to be sound, M must be a μ-possible property. In order for his argument to be rationally persuasive, it must be rational to believe that M is a μ-possible property. The fact that M is an e-possible property is of no use to Plantinga, because in the present context (where L is also e-possible) the e-possibility of M is no indication of M’s μ-possibility.
One such property which Plantinga himself discusses is that of near-maximality, a property “enjoyed by a being if and only if it does not exist in every possible world but has a degree of greatness not exceeded by that of any being in any world” (1974a, p. 218). It should be obvious that near-maximality is μ-non-compossible with M.
One remark of Plantinga’s suggests that this might, indeed, be his view: “And [POA1], the central premise of [POA] is not of this sort; a sane and rational man who thought it through and understood it might none the less reject it, remaining agnostic or even accepting instead the possibility of no-maximality” (1974a, p. 220).
The underlying idea being expressed here is nothing novel. It is simply the age-old philosophical precept that one is rationally required to apportion one’s belief to the evidence.
Barry Loewer has shown that no formal demonstration of the μ-possibility of M can be given which does not presuppose the actuality of M. See Loewer (1978, pp. 108, 109).
This invalidity is obvious when we juxtapose the two premises:
POA-e1 There is an e-possible world W in which maximal greatness is instantiated.
POA-μ2 If there is a μ-possible world W in which maximal greatness is instantiated, then there exists a being G in W who possesses maximal greatness in W.
The consequent of POA-μ2 does not follow from these two premises because the antecedent of POA-μ2 differs from POA-e1.
To think otherwise is to commit an argumentum ad ignorantiam.
It should be obvious that the for-all-I-know-existence of some being B is compatible with B’s nonexistence. For all I know, there exists life on another planet in our galaxy. That is, of course, compatible with there being no life on any other planet in our galaxy.
I borrow this terminology from Peter Klein who makes a similar point with respect to skeptical arguments predicated on the closure principle. He claims that such arguments “virtually beg the question” because one of the premises in closure-based skeptical arguments can only be supported by a subargument that employs the conclusion of the main skeptical argument as a premise. See Klein (1995, pp. 213–236).
Strictly speaking, Gödel defines being Godlike as follows:
D10′ X is Godlike iff X has as essential properties those and only those properties that are positive.
He then argues that being necessary is a positive property and concludes that being necessary is included in the property of being Godlike. For ease of exposition, I’ve built this step into the D10 definition of being Godlike.
For a complete presentation of Gödel’s OA, see Oppy (2019).
I’m assuming that being Godlike, as defined in D10, is a consistent property.
Oppy (2009) discusses a similar property—i.e., the property of having no world-mate that is a necessarily existent, essentially omnipotent, essentially [omniscient], essentially perfectly good being—though he puts it to a different use. Oppy appeals to this property to create an atheistic parody of Pruss’s OA, whereas I appeal to the property of being in undivine company [U] to argue that we are not rationally entitled to believe that being Godlike is a μ-possible property unless we have an independent reason to think that U is a μ-impossible property.
References
Dombrowski, D. (2006). Rethinking the ontological argument. Cambridge: Cambridge University Press.
Dore, C. (1989). God, suffering, and solipsism. New York: St. Martin’s Press.
Hacking, I. (1975). All kinds of possibility. The Philosophical Review,84, 321–337.
Klein, P. (1995). Skepticism and closure: Why the evil genius argument fails. Philosophical Topics,23, 213–236.
Loewer, B. (1978). Leibniz and the ontological argument. Philosophical Studies,34, 105–109.
Maydole, R. (2003). The modal perfection argument for the existence of a supreme being. Philo,6, 299–313.
Oppy, G. (1995). Ontological arguments and belief in God. Cambridge: Cambridge University Press.
Oppy, G. (2009). Pruss’s ontological arguments. Religious Studies,45, 355–363.
Oppy, G. (2019). Ontological arguments. In E. N. Zalta (Ed.), The stanford encyclopedia of philosophy. Retrieved at https://plato.stanford.edu/archives/spr2019/entries/ontological-arguments/. Accessed 23 May 2019.
Plantinga, A. (1974a). The nature of necessity. Oxford: Clarendon Press.
Plantinga, A. (1974b). God, freedom, and evil. New York: Harper and Row.
Pruss, A. (2009). A Gödelian ontological argument improved. Religious Studies,45, 347–353.
Reinhardt, L. (1978). Metaphysical possibility. Mind,87, 210–229.
Rowe, W. (1987). Modal versions of the ontological argument. In L. Pojman (Ed.), Philosophy of religion: An anthology (pp. 69–73). Belmont: Wadsworth.
Acknowledgements
An early ancestor of the present article was presented at a Society for the Study of Process Philosophy group meeting of the American Philosophical Association. I would like to thank Dan Dombrowski and the others in attendance for their comments and questions. Special thanks to Tomis Kapitan and an anonymous referee for this journal for their very helpful recommendations. The present article is much improved as a result.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Engel, M. What ontological arguments don’t show. Int J Philos Relig 88, 97–114 (2020). https://doi.org/10.1007/s11153-019-09718-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11153-019-09718-x