In my introduction to this series, I listed the four general strategies (as I count them) that one might take in attempting to resolve the intellectual problem that evil poses for theism. They’re not mutually exclusive strategies, and they might overlap with each other to varying degrees, but I find it helpful, at least, to categorize them in this way. The first type of general strategy I’d like to explore at greater length is that of denying the problem in one way or another.
Now I don’t personally know of anyone who simply denies the problem in a dismissive “next question please” kind of way. But there are certain types of responses which are effectively very similar. And again, they typically are paired with other kinds of responses as well, but these denial responses are aimed at undercutting the problem of evil altogether.
A common example
One of the most common such responses–certainly from lay people but from many professional theologians and philosophers as well–is to claim in some form or another that evil actually proves the theistic God’s existence. Because without this God, the very concepts of good and evil evaporate, and thus to even speak of good and evil is to presuppose theism. I’m actually going to save a more thorough discussion of this response for a later entry in this series, after discussing theodicy and skeptical theism. And here I’ll simply state that I don’t think it’s a good response. I know that’s a tease–how about this: I’ll touch at least briefly on this question when I circle back to it near the end of this post.
How deductive arguments function
But I don’t bring up and then move quickly past that example just to be a tease. Rather as a way to introduce a broad category of responses that tend to function, in effect, as denials of the problem of evil. Namely, deductive arguments for God’s existence.
It might seem like a miss and unfair on my part to characterize such arguments as denials of the problem of evil. After all, deductive arguments aim to prove that God exists, not to sidestep the questions, right? Allow me to explain what I’m getting at.
A deductive argument is such that if its premises and conclusion are structured in a logically valid way, then the conclusion follows necessarily from the premises, if all the premises are true. Such an argument–valid structure with true premises–is said to be sound. And so a sound deductive argument for God’s existence would single-handedly demonstrate that God does exist, and thus the intellectual problem of evil would appear to evaporate, because no ostensible evidence (from evil or otherwise) would properly call that proof into question. (That is, so long as we haven’t additionally identified any sound deductive arguments for God’s non-existence!)
At least, that’s the way deductive arguments are typically treated. And crippling ambiguity quickly arises when we try to tease out exactly what burden of proof there ought to be for demonstrating that the premises of such arguments are true.
Many professional philosophers place an extremely high burden here, as well they should, if they’re going to regard sound deductive arguments as decisive proofs. But with such high (though still ambiguous) burdens of proof, precious few deductive arguments–at least outside of formal mathematics–can be regarded as sound. And unless an argument can be regarded as sound, it is thought by many to provide no support for the proposition in question. So within this framework, we get endlessly bogged down in the minutia of the various premises of this argument and that, and the debates grow largely arid and stalemated before very long.
Alternately, many non-professionals place a fairly modest burden of proof here–at least for their own preferred arguments. (And I should note, most non-professionals don’t have carefully constructed formal deductive arguments in mind, in any event.) So long as such an argument appears to clear this sort of modest hurdle, then it is thought to provide decisive proof for the proposition in question. So within this framework, one might regard each of the premises of a deductive argument as probably true (ie, more likely true than false–a modest hurdle), and then proceed to regard the conclusion as decisively true, and any counter-evidence as irrelevant. Which is hugely problematic and allows ad hoc thinking and confirmation bias to run rampant.
Might I suggest that this flat sound vs. unsound binary is incredibly unhelpful. This stark dichotomy between providing decisive proof vs. providing no support at all. And the burden of proof which divides this binary is inherently ambiguous.
Instead, we should think in terms of a rich spectrum of confidence levels, ranging from 0% (irresistibly false) to 100% (irresistibly true). And it seems to me that we shouldn’t ever regard the soundness of any argument (even in formal mathematics) at either the 0% or the 100% extreme. So within this framework, we can simply throw soundness out the window, at least as a flat binary concept. Valid deductive arguments are seen to offer the prospect of evidence for or against various propositions, which may be weak or strong (or even very strong!) evidence as the particular case may be. And the bottom line is that no deductive argument can ever give us carte blanche to ignore counter evidence. No deductive argument for God’s existence can make the intellectual problem of evil simply evaporate, or can give us permission to deny the problem as irrelevant.
Here’s an example of what I’m suggesting: Bertrand Russell is to have said in a eureka moment, “Great God in Boots! The ontological argument is sound!” (More on ontological arguments in a bit.) And then later he reflected, “The argument does not, to a modern mind, seem very convincing, but it is easier to feel convinced that it must be fallacious than it is to find out precisely where the fallacy lies.”
And I think that’s good methodology. Even if we happen to accept to some degree a particular deductive argument for God’s existence–or at least, we can’t find a precise flaw in that deductive argument–it doesn’t follow that we therefore ought to conclude that God exists. That argument would serve as evidence for God’s existence, which may be weak or strong evidence, as the case may be. But not decisive proof.
So that methodological groundwork laid, I want to very quickly survey some of the common types of deductive theistic arguments.
A lightning quick survey of deductive arguments for theism
In surveying the varieties of deductive arguments for theism, one of the most important things I want to emphasize is that in most of these cases, even if the arguments are otherwise successful, they don’t actually point specifically to the God of theism.
For example, typical design arguments will point, at most, to a vague designer of some sort. But they won’t speak in careful detail to the question of who that designer might be. The morally perfect omnipotent being of theism? No, that would be a tremendous and unwarranted leap.
Philosopher Stephen Law has championed the very clever Evil God challenge. One of its underlying implications is that unless a particular theistic argument entails a morally perfect God, it can equally be marshaled in support of an evil anti-god. The Kalam cosmological argument (“Everything that begins to exist has a cause…”) for example: Even if some formulation of the Kalam is otherwise a successful argument, it doesn’t actually entail the morally perfect omnipotent being of theism. It just as easily points to a morally evil anti-god. Or to a morally ambiguous, non-omnipotent being of some vague sort. Or even simply to an impersonal process of some unknown variety, for that matter.
To put a bow on it, most of these theistic arguments–even if otherwise successful–provide, at best, evidence for theism which is very ambiguous and tenuous. And even that very modest accomplishment only carries if they can be firmly anchored in some manner to the perfect being of theism, as part of a cumulative case for theism. And the two most common attempted anchor points are the various moral arguments and the various ontological arguments.
As to moral arguments: as already mentioned, I’ll save a detailed discussion for later, after discussing theodicy and skeptical theism. The short of it–which I’ll simply state for now–is that a robust and workable concept of morality actually entails atheism; theism entails profound moral skepticism.
As to ontological arguments: Modern modal ontological arguments and so forth offer refinements of Anselm’s original, but the general idea is still communicated well by Anselm’s terminology: God is that being than which none greater can be conceived. (Incidentally that’s gotta be one of the most impossibly cool yet archaic phrases ever conceived.) And because a being which actually exists is greater than a being which only “exists” in our imagination, God must actually exist. And must be morally perfect, omnipotent, etc.
The aforementioned Bertrand Russell provided a good if perhaps not entirely satisfying response to ontological arguments old and new. Even if apparently successful, ontological arguments feel so strange and slippery that I can’t see how they can count as particularly compelling evidence for theism. And perhaps most damning to ontological arguments are the various paradoxes that they entail. As Stephen Law points out, it’s just as easy to craft an ontological argument for the existence of evil god, a being than which none more evil can be conceived (not the same hypothetical being as Satan, by the way, who is thought to be a finite being). And thus, ontological arguments entail contradictions and are therefore self-defeating.
In summary
The long and short of it all is that even at their best, the various deductive arguments for theism provide evidence for God’s existence rather than decisive proof. Evidence which must be weighed against whatever counter-evidence there might be. And in reality, that theistic evidence appears to me to be incredibly tenuous evidence, which can stand only in very weak opposition to the avalanche of counter-evidence. Much more on that counter-evidence to come.
And even if–contrary to all I’ve just argued–one were to consider a specific deductive argument as decisive proof of theism, it would still be critically important to attempt to understand what reasons God might have for allowing the evils we see around us. Lacking a strong and reasonably thorough understanding of the evils around us, we would quickly fall prey to crippling moral skepticism, as I noted a moment ago. Again however, I’ll have to save a more thorough discussion of this issue for a later entry in this series.
But before moving on to theodicy, there’s one more sub-genre of responses within my denial category that I’d like to explore. In its most sophisticated form: reformed epistemology and its handling of proper belief formation, defeaters, and so forth. Which is a refinement and more careful articulation of the “I just know deep down that it’s true” kinds of responses which are so often appealed to in attempts to justify theism.
Till next time…
[…] Deductive arguments for theism […]