Platonism and the theists

March 2011

Theistic activism is one response to a problem arising from the concurrent postulation of both God’s1 existence and the existence of certain platonic entities. Examples of the latter include numbers, propositions, relations, and possible worlds, entities thought to exist necessarily, as does God. This essay considers the thesis as formulated by Morris and Menzel (1986). I examine two arguments against theistic activism, presented by Davidson (1999) and by Bergmann and Brower (n.d.), and I explain why I think Bergmann and Brower have the more cogent argument.

According to Morris and Menzel, “A theistic activist will hold God creatively responsible for the entire modal economy, for what is possible as well as what is necessary and what is impossible. The whole platonic realm is thus seen as deriving from God” (p. 356). Augustine proposed a version of the thesis, which Morris and Menzel have reworked to incorporate insights from modal logic. Their basic idea, like Augustine’s, is that abstracta are concepts in God’s mind. Anything that necessarily exists, exists in all possible worlds, and God creates all possible worlds in addition to the actual world, and so the abstracta derive their necessary existence from God’s creative activity, which is a manifestation of his intellectual activity (pp. 356-58).

A problem for the thesis arises with the realization that abstracta include certain of God’s own properties. Nothing is identical with its properties, and God creates everything not identical with himself. God has his properties essentially and so he would not exist if they did not exist. From this it seems to follow that God depends on himself for his own existence, i.e. he creates himself (Davidson 2005). Morris and Menzel’s solution is to deny the problem: “It just seems to us that there is nothing logically or metaphysically objectionable about God’s creating his own nature in precisely the way indicated” (Morris and Menzel, p. 360). That he creates his own nature, they argue, does not entail his creating himself, and so long as he isn’t doing that, nothing absurd is going on (Ibid.).

Davidson finds this unsatisfying, and so do Bergmann and Brower. The latter suggest that Morris and Menzel’s failure to see the circularity is owing to excessive vagueness. When properly clarified, they say, the problem becomes fatal to theistic activism. If Christian theism is true, they argue, then platonic abstracta do not exist (Bergmann and Brower, p. 17). Davidson declines to go there. He concludes instead that it just is not the case that all abstracta depend for their existence on God while acknowledging the difficulty of finding a principled distinction between those that do and those that don’t (Davidson 1999, p. 290).

Bergmann and Brower present their case in terms of logical priority. Under platonism, for any true predicate of the form “a is F,” the property a is exemplified in F, and that makes F logically prior to a. This being so, they argue, the conjunction of theism and platonism entails the following contradictories:

C1: God’s creating an exemplifiable is logically prior to the exemplifiable being able to create an exemplifiable.

C2: The exemplifiable being able to create an exemplifiable is logically prior to God’s creating an exemplifiable. (p. 12)

And so, since abstracta cannot be things that God creates, abstracta cannot exist if theism is true.

Davidson’s argument focuses just on those abstracta that are exemplified by God himself. He begins with an analysis of the dependence relationship held to obtain between God and all things (including abstracta) that exist (1999, pp. 278-286). He argues that it seems to be a causal relationship (p. 281), though causation itself must be taken as a primitive notion (p. 286). He then observes:

The theistic activist claims that God causes properties such as ‘being omniscient’, ‘being omnipotent’, ‘existing necessarily’, ‘being able to cause abstracta to exist’, and ‘having cognitive activity’ to exist. She also claims that God causes His own haecceity . . . ‘being God’, to exist. However, to claim this is to get the dependence relationship backwards. Surely, God’s being able to cause abstract objects to exist must be posterior to His having properties like the ones mentioned above. And if God has these properties, they must exist. But, the proponent of this theory is committed to the existence of properties being posterior to God’s causing them to exist. (p. 287)

Davidson considers various responses by defenders of theistic activism and finds them all inadequate, and so the thesis leads us inevitably to the notion that “we have the Divine causing His own existence; God is pulling Himself up by His own bootstraps” (p. 289). Among the assumptions that force this conclusion is the proposition that all abstracta depend on God for their existence, and that is the assumption that Davidson discards in order to eliminate the causal circularity.

As noted initially, the problem arises from the joint assertion of two premises: (1) God exists and has the characteristics attributed to him by conventional theism; (2) platonic abstracta exist. For this essay I have no interest in questioning the first. Bergmann and Brower’s argument is to the conclusion that the second is false. Davidson’s argument concludes that the first must be modified, since its traditional construal holds that everything without exception depends on God for its existence. I find nothing in the arguments themselves on which to base a preference for one over the other, but I think Bergmann and Brower’s conclusion is independently justifiable.

We begin by asking: Why believe either premise? We can assume for the sake of this discussion that the answer in both cases is: Intuition. Most of us find the existence both of God and of abstracta to be just too obvious to doubt. But we also have learned, if we have lived long enough and well enough, that our intuitions occasionally mislead us; and so, when we discover that a particular couple of intuitions lead to a contradiction, we try to figure out which of them needs to be either revised or abandoned in order to resolve the conflict. I suggest that in this case, platonism should be abandoned, and not only for the reasons adduced by Bergmann and Brower. But as they note (p. 21), a denial of platonism raises a problem more specific than “It’s counterintuitive.” Among the motivations of platonism is the notion that it grounds certain of our ideas about truth. We make sense of statements such as “That rose is red” by supposing not only that the rose exists but also that something called redness exists and that the rose has that something. More generally, for any property that we attribute to any object, we think the attribution cannot be true unless the property actually exists. Furthermore, we find ourselves talking, quite naturally, about properties and other abstractions as if they existed. Mathematical discourse is a paradigmatic example. How, we wonder, could it possibly be true that the sum of two and three is five if two, three, or five—or sums, for that matter—don’t even exist? Space precludes a thorough response, and so some apparent hand-waving is unavoidable, but I suggest that an answer is to be found in the origins of our own existence, quite regardless of what God’s role in those origins might have been.2

At this point in scientific history, we do not know much about the details, but evolution produced our brains for the same reason it produced all our other organs: They help us survive. Any organism capable of locomotion needs to make decisions. If nothing else, it needs to decide at any given moment whether to move or not move. For the simplest animals such as protozoa, subcellular chemistry suffices. But one of the earliest divisions of cellular labor dedicated some cells to data processing. Certain data about the environment were input—food this way, danger that way—and the output was movement in this direction or that direction. Primitive organisms would have had no concepts of food, danger, or any other sort of abstraction, any more than a thermostat has a concept of heat or cold, much less of any idea as purely subjective as comfort. None of that mattered. Only results mattered. Nervous systems, and the DNA sequences causing those systems, survived if the organisms hosting them survived. Evolution continued, and in due course we humans came along. We have no reason to suspect there were any discontinuities. At some point certain of our ancestors’ brains got complicated enough to produce the sensation we call self-awareness. We became able to think about thinking, and as soon as we started doing that, we started doing philosophy. We are not even close to figuring out how this all happened or even could have happened—how computational complexity alone could have been sufficient to produce sentience and abstract thought as we know it. So far as we can tell, though, it is just a matter of complex computation. Our every thought is just some data in our brains. At least, in some sense, that is all it is. I’m not here endorsing a pure identity theory of mind, just the notion that when we experience a thought, that thought is represented by some data physically encoded in our neurophysiology, and if the data aren’t there in our brains, then we don’t think them or about them.

It has been suggested that our brains’ neurophysiological data could have no meaning if the brain was nothing but a computer if, qua computer, it was doing no more than executing a program (Searle 1983). The argument, though, rests on an assumption about the insufficiency of any syntactical system to generate semantic content, and that assumption has been challenged (Churchland and Churchland, 1990). It thus seems at least arguable that what we call abstractions are just data corresponding to relationships or patterns that we observe in, among, or about the physical objects in our environment. Those data exist at least as the neurochemical states that encode them, and that is all the existence they need in order to make statements communicating them true, provided only that their correspondence with the real environment is sufficiently accurate.

Our intuitions, then, are just data produced by certain processes of which we are not entirely aware. Our brain generates them, and we don’t know how, but there they are. The existence of minds other than our own might be such a datum, for one instance. But we also know that our brains are capable of generating data, which we experience as thoughts, that do not correspond to anything real. Sometimes we know that this is happening, and when we know it, we usually suppose that something about our cognitive faculties is not functioning as it should. But the brain’s proper function is just to facilitate our survival. Assuming God’s actual existence, our intuitive awareness of his existence presumably facilitates our survival in some way, but in any case, we have our intuition because, through most of our evolutionary history, it accurately tracked reality often enough to be favored by natural selection.

In the case of abstracta, what had to happen was that the way we talked about them had to be efficient for communicative purposes. If I see a cat to my right, a cat to my left, and a cat in front of me, I can say just that—“There is a cat to my right, and a cat to my left, and a cat in front of me”—or, if their spatial distribution happens not to be immediately relevant, I can say “I see three cats.” My ability to say the latter is not contingent on the existence of some entity we decide to call three. What matters to the truth of my saying it is that there exist entities x, y, and z such that x is a cat, y is a cat, x ≠y, etc. Their existence is a datum, their separate identities is a datum, the concatenation of those data is another datum, and we can call that additional datum three without supposing any increase in the number of things that actually exist outside our heads. On this account, reality is the only truthmaker, and if all reality is dependent on God, then God is the maker of all truths. Theists need no other.


1The God in question is that of the currently predominant monotheistic religions, Judaism, Christianity, and Islam, although most of the relevant literature is by Christian philosophers. The arguments seem to suffer no loss of generality on that account.

2Not all theists will agree. I’m assuming a readership whose theism is consistent with modern science.



Bergmann, Michael, and Jeffrey E. Brower. n.d. A Theistic Argument Against Platonism (and in Support of Truthmakers and Divine Simplicity). Papers/Theism%20and%20Platonism.pdf.

Churchland, Paul M., and Patricia Smith Churchland. 1990. Could a Machine Think? Scientific American 262 (January):32-39.

Davidson, Matthew. 1999. A Demonstration against Theistic Activism. Religious Studies 35 (3):277-290.

Davidson, Matthew. 2005. God and Other Necessary Beings. In The Stanford Encyclopedia of Philosophy, Edward N. Zalta, ed. Stanford, CA: Stanford University. http://

Morris, Thomas V., and Christopher Menzel. 1986. Absolute Creation. American Philosophical Quarterly (4).

Searle, John R. 1983. Can Computers Think? In Minds, Brains, and Science, pp. 28-41. Boston, MA: Harvard University Press. Reprinted in Philosophy of Mind: Classical and Contemporary Readings, David J. Chalmers, ed. 2002. New York: Oxford University Press.

Site homepage

Philosophy index