Suppose we observe an event E in the universe. A theist says that God explains this event E. An atheist then makes the following argument: “whatever you posit God as an explanation for, I can posit an alternative explanation with simpler attributes than God.”

Now of course, theists will say that nothing can be simpler than God. But let’s, for argument’s sake, say that this is false. Let’s say that God is necessarily more complex than whatever it purports to explain. For example, let’s assume that the universe is more simple than God.

Now, let’s go back to an event E. Either that event E is necessary or contingent. If that event E is necessary because of God, the atheist can say that this event E is necessary because of some law of nature that we haven’t figured out yet that is impersonal. If the event E is contingent because of God (say, by God’s free will), then the atheist can simply say that the event E was a brute contingent fact (I.e. the chances just played out that way through some natural mechanism such as quantum mechanics and the rest of physics).

Let’s now suppose the event E is the beginning of the universe. The theist might posit God to explain E. But the atheist can simply posit an impersonal cause to explain E. The theist might argue that this cause cannot be a physical law and must be something else since all laws start at the beginning of the universe. But the atheist can simply assert that there’s some other mechanism by which event E arises that is impersonal but doesn’t contain the complex attributes of God such as omnipotence, omniscience, etc. After all, if a theist can posit a cause with X number of attributes to explain event E without evidence, the atheist can also posit a cause with <X number of attributes without evidence to explain the same event E.

The question then is: does this strategy work? Is there any philosophical literature behind the general notion of nature or impersonal causes explaining everything that God can but without the additional complexity? Graham Oppy talks about how naturalism and theism have the same explanatory power with the difference being that theism posits more kinds of entities. Are there any other philosophers that talk about this?

  • I think that the concept of theism having more kinds of entities is arbitrary; for naturalism you have to account for the mathematical entities too, since the ontology of physics depends on the ontology of mathematics. Commented Nov 11, 2023 at 15:27
  • 1
    One problem with this line of thinking is that it depends on a pre-theoretic notion of simplicity/complexity, one that flies in the face of both much of theology as well as modern mathematical understanding of the notion. If God is "too simple to be simple" par excellence (as is often effectively supposed), how do we compare and contrast divine simplicity with other kinds? Is it possible to meaningfully assert a simplest explanation as such? Perhaps there are going to be trade-offs regardless. Commented Nov 11, 2023 at 15:28
  • 2
    "whatever you posit God as an explanation for, I can posit an alternative explanation with simpler attributes than God." that would not be a properly educated atheist. The proper answer would be "i don't know, but the fact that I don't know does not imply your God explanation"
    – armand
    Commented Nov 11, 2023 at 15:32
  • @IoannisPaizis Isn’t there a debate on whether mathematical entities are mind independent or not? But either way, wouldn’t theism have to account for them too? Mathematical and physical entities would have to be accounted for in theism. But God with all His attributes does not have to be accounted for under atheism.
    – user62907
    Commented Nov 11, 2023 at 15:37
  • @armand Fair enough but I suppose I’m talking about the kind of atheist that positively affirms that reality does not contain the kind of god described by most theists.
    – user62907
    Commented Nov 11, 2023 at 15:38

7 Answers 7


Many believers (not necessarily most) posit a sort of sensus creatus, a consciousness of having been "created" somehow. Perhaps this can be traced back to mere human parentage, of course; if the "sense datum" of "being created" is supposed to run deeper than that, or does run deeper betimes, then one question here becomes, "Is a universal creator a better explanation for the sensus creatus than mere parentage, or mere parentage coupled with transcendental illusions of deeper significance, here?"

  1. Mere parentage might seem to be the simplest explanation, but is it the most adequate to the case?
  2. An actual deity is one thing; mere parentage coupled with transcendental illusion seems like two things; but is an actual deity a simpler explanation, then, here?

Again, consider the mathematical phenomenon of being too simple to be simple modulo talk of God in terms of divine simplicity. Do we here define God as the simplest of all possible things? Or is simplicity still relativized, with God as the greatest exemplar of deific simplicity, but not the greatest exemplar of other kinds of simplicity? For example, one might say that the simplest relation is a null one, so that a being that relates to other things in the simplest way is a being that does not relate to anything else. This would not be the living God of religion, however, since such an entity is widely held to both logically and metaphysically relate to Its creations.

There is also an ambiguity to the claim of a relation between intentional/intelligent design and the probability of life:

  1. "If there is an intelligent designer, this being will probably design life."
  2. "It is probable that life was intentionally designed."
  3. "Life is more probable if it was intelligently designed."

(3) seems true, but perhaps irrelevant (it is more probable that I will win the lottery if I buy 10,000 tickets instead of one, but if I win the lottery having bought only one ticket, yet it doesn't become probable that I bought 10,000 tickets).

Another consideration: one might interpret deities and multiverses as reciprocally weak explanations of reality overall, per this question about empty-vs.-trivial solutions to explanatory problems.

  • 1
    Yeah I suppose my argument would be moreso against the traditional conception of God, the kind that does relate to His creations. Anyhow, the distinction between how simple and complex an object is does seem to be an endless debate. However, I just can’t see how the traditional conception of God, the kind of God without which everything in the world would not exist, is simple.
    – user62907
    Commented Nov 11, 2023 at 15:45
  • @thinkingman in Dante's Paradiso, both descriptions are given: in the ninth heaven, God appears as a zero-dimensional point of light at the center of the angels, in the tenth heaven God appears as an infinite-dimensional abyss of light encompassing the angels (and the created universe). Perhaps this is a contradiction, or at least an antinomy. But divine simplicity is at least supposed to be partwise simplicity, i.e. God can't be broken into multiple spatial/temporal parts (the more extreme properties-as-parts view has theistic proponents and detractors, e.g. Plantinga). Commented Nov 11, 2023 at 16:09
  • Explanatory simplicity pertains to God's relations, though, then, it would seem. So God is simple in the partwise sense, but to be simple in that way, God must be too metaphysically simple to be explanatorily simple (it would also seem), and the dialectic continues... Commented Nov 11, 2023 at 16:10

Yes. In connection with why any unexplained event happened, where the theist says 'God just is' you can say 'It just did'. The point about these arguments is not whether one is more or less logical than another, but about whether humans, as a species, are more susceptible to one sort of belief than another.


You introduce the hypothetical situation that both the theistic and the atheistic position have the same explanatory power, but that theism posits more entities.

That’s the classical situation where Occam’s razor applies "Entities are not to be multiplied without necessity" (Non sunt multiplicanda entia sine necessitate). See Occam’s razor for an introduction and some historical remarks. According to this principle of heuristic one should prefer the position with fewer entities.

But before arguing with Occam’s razor one needs an agreement that between both positions actually have the same explanatory power.

If the discussion is about the origin of the world I see

  • the theist positing the existence of a creator god which the atheist considers an ad-hoc argument employing a metapher from human situations,

  • while the atheist must admit that he cannot the answer the question. Possibly he even doubts that we already have the right concepts to deal with the question.

Hence the premisses of Occam’s razor are not fullfilled.

  • The point of the argument is to say that positing God as an explanation doesn’t answer the question either. For starter’s, it doesn’t tell us how God created the world. If the theist can assert an entity ad hoc to explain X, then the atheist can do the same, except make the entity even simpler without losing explanatory power.
    – user62907
    Commented Nov 11, 2023 at 16:07
  • I think that you apply Occam’s razor to an arbitrary thesis; see my comments (2 first). Commented Nov 11, 2023 at 16:09

This is an old debate

You ask: "Is there any philosophical literature behind the general notion of nature or impersonal causes explaining everything that God can but without the additional complexity?"

Yes, you are in effect talking about the cosmological argument.

Works for and against this argument are plentiful.

One of the most prominent for the argument is William Laine Craig's book Kalam cosmological argument.

Notable opponents against the argument are David Hume and Paul Edwards.

One might note though that should the claim — that a magical entity creted the known space and time that we call "our universe" — turn out to be true, this does not imply that...

  • theists have a mandate to speak for this entity
  • this entity knows of our existence
  • this entity has our best interest in mind
  • anyone is obliged to follow the will of this entity
  • this entity is supernatural
  • this entity was not itself created by someone/something else

In any case, it is for us humans — with our extreme myopia concerning time, space and the laws of nature — at this present stage entirely impossible to say that such an entity is the only possible explanation for any observed phenomenon. We simply do not know enough to say what is knowable, and with that we cannot rule out all other explanations in such a way that "God did it" is the only one remaining.

And so far, no observed phenomenon demands any kind of personal intervening deity — of the sort described by theists — in order to be explained.


In the context of creation, the difficulty is in understanding the initial materialization of existence as without a cause, or in other words, how to explain that a cause can be embedded/implanted in its actualization without an external intervention or design. Unfortunately this is a trap for (modern) naturalists which can be overcome only by applying randomness or "hiding" the cause in a yet-to-discover principle or explanation.

  • Any physicalist worth her salt simply thinks that causation in this sense is speculative and to some degree unknowable. That's not a trap so much as acceptance that there are limits to human knowledge.
    – J D
    Commented Nov 15, 2023 at 15:17
  • @J D, the trap is in reductionism, (vs logos) : the non-acceptance of the emergence of properties inside a system that are not directly caused by the interactions of their parts. Commented Nov 15, 2023 at 16:01
  • Logos is a rhetorical component of persuasion, and reductionism is an approach to grounding theories in theories. And an adequate scientific description of the origin of the universe is simply not possible, at least currently. The origins of timespace are speculative without more empirical access.
    – J D
    Commented Nov 15, 2023 at 16:05
  • logos is misunderstood. logos is the concept that the meaning of things is embedded in themselves; that cause and effect are interrelated as a manifestation of the same thing. Commented Nov 15, 2023 at 16:08
  • My definition was Aristotle's. Whose is yours? en.wikipedia.org/wiki/Logos
    – J D
    Commented Nov 15, 2023 at 16:11

Case 1: If God does exist, then explaining God without God would be pretty much impossible, assuming the typical definition of God as a necessary, eternal, all powerful and personal being. The only way I see you could attempt to explain God without God would be by changing the definition of God so that God(s) could emerge by natural causes, perhaps along the lines of an extremely advanced alien civilization that achieves a God-like level of mastery and control over the laws of nature. Therefore, under the classical definition of God, if God exists, naturalistic efforts to account for all that exists are bound to fail, because at the very least they will not be able to account for God itself.

Case 2: If God does not exist and only nature exists, then everything has a natural explanation, and therefore any theistic attempts to account for everything that exists will be bound to posit additional entities with respect to their simpler atheistic counterparts.

Thus, having a more complex theory that posits that God exists is not a problem if God actually exists (Case 1). If reality is more complex, then your theory ought to be more complex.

  • Of course, but we have no evidence for either. If we are undecided between god or not existing, and have no independent evidence for god existing, it is practical to simply assume god does not exist. This is because all explanations without evidence are equally explanatory, so you might as well believe in the explanation that posits the fewest entities.
    – user62907
    Commented Nov 15, 2023 at 17:07
  • @thinkingman The only evidence we have for the existence of nature is our subjective experience of nature. In the case of God, there are individuals who claim to have a subjective experience of God. See this answer.
    – Mark
    Commented Nov 15, 2023 at 17:10

Let's imagine that our atheist A, and our theist T, are arguing over event Y as caused by unknown factor X. T claims that X is God, or God's will. A claims that X is an unknown, simple law of nature. Neither has proof that will convince the other, so it essentially comes down to a statement of belief on either side.

It's worth noting this is a point-missing argument for most theists. A God that could be replaced by entirely naturalist explanations would essentially be a Deist God, one who is effectively distanced from the universe. Nearly every religion includes some notion of connecting to God, either through prayer, rituals or corporate worship. A substantial portion of believers experience some kind of personally convincing, subjective evidence of a relationship with God. So this argument would only have impact against the minority of believers whose belief in God is solely intellectual.

  • Actually, it would have an impact against the majority of believers who do have a belief in God with complex attributes. Simply posit a God with fewer attributes than this God. Let’s call him Mod. Mod explains everything God does but is much simpler. The theist has no reason (from a theoretical virtue standpoint) to prefer God over Mod. One can make Mod so simple that it becomes a naturalistic explanation. Since the theist cannot rule that out, he now has no reason to believe in God.
    – user62907
    Commented Nov 16, 2023 at 3:12

You must log in to answer this question.