An overview of the problem of evil | Denying the problem: reformed epistemology

Did I say something about wrapping this series up “before 2016 comes to a close”? Maybe I meant 2020? Anyway, if you missed it, or forgot it all in the lengthy interim, here’s the “Overview of the problem of evil” series thus far:

• Introduction

“Denying the problem” section:

• Deductive arguments for theism

Today’s entry is the second and final entry in the “Denying the problem” section of the series. Here I’ll focus on 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, evidence be damned” kinds of responses which are so often appealed to on behalf of theism. This family of responses shares a functional similarity with the family of deductive arguments for theism, in that both types of responses are often thought to undercut the problem of evil (and any problems for theism, really) in one fell swoop. Here’s how one of the more prominent Christian apologists in the world, William Lane Craig, summarizes the issue:

What I’m claiming is that even in the face of evidence against God which we cannot refute, we ought to believe in God on the basis of His Spirit’s witness. Apostasy is never the rational obligation of any believer, nor is blasphemy against the Holy Spirit. God can be trusted to provide such powerful warrant for the great truths of the Gospel that we will never be rationally obliged to reject or desert Him.

Now, plenty of Craig’s fellow apologists would take issue with his sweeping summary here, even some within the reformed epistemology camp, and I’ll get into that before all is said and done. But my contention here will be that despite whatever protests there might be to a summary statement such as Craig’s, reformed epistemology as actually practiced (ie, if it’s to be of much help to theism) amounts to very nearly the same thing.

It suffers from the same basic problem as the “I just know that it’s true” epistemology of the layperson. Namely, it boils down to: anything goes. Or, on certain versions, anything goes so long as you can wrap it within a suitable just-so story; so in effect, it still boils down even on these versions to anything goes. It’s radically permissive of nearly any belief, so much so that it compares with radical skepticism though in the opposite direction. And such radical permissiveness is no more workable than is radical skepticism, so it ends up manifesting as a cherry picking of beliefs in a radically ad hoc manner, and we all end up on segregated islands of belief with little hope of rationally bridging the differences between us.

There’s also an underlying problem of principle, which I’ll have to circle back to later in this series, after discussing skeptical theism. In brief: reformed epistemology assumes as a given that if the theistic God exists, then surely this God chooses to communicate with humans. But this isn’t a tenable assumption, and so the entire enterprise fails to get off the ground.

But I’m getting ahead of myself, so I want to step back now and take a look at reformed epistemology from the ground up. Of course I can’t even begin to handle this topic comprehensively in a single essay, nor do I have the expertise to do so, but what I can attempt instead is to get down and dirty and to try to summarize the important issues in a *comparatively* quick and straightforward manner.

And not by tracking the historical development of the field, which would be far more convoluted than necessary: 95+% of that task would be irrelevant for my purposes here anyway, as I’m willing to grant nearly the entire epistemic framework for the sake of argument. (But for a taste of that, here’s a 102-minute interview with epistemologist Tyler Wunder, tracing the development of Alvin Plantinga’s work. And though Plantinga is perhaps the most notable figure in the field, and I’ll structure much of my essay here around his work, that’s not even to touch directly on the work of William Alston, Nicolas Wolterstorff, and others.) Instead, I’ll proceed in a topical manner and get down to the bottom line as I see it, as quickly as I reasonably can.

Epistemic foundations

Epistemologists have long realized that if we are to avoid falling prey to a sweeping and crippling and ultimately incoherent skepticism, we have to make certain assumptions at the very outset of our epistemic journey. The most common strategy has been to identify certain foundational pieces/sources of knowledge which are taken for granted, rather than rigorously justified, lest we stumble along an infinite regress or fall into circular thinking.

One such foundational assumption involves our sensory perceptions. We assume that these perceptions correspond with our external environment in a generally accurate manner.

[Pedantic note #1: that’s not necessarily to make assumptions about the underlying ontology of our environment. For example, are we living in a computer simulation? Possibly, though that’s not relevant here. But I’m not only trying to be pedantic by mentioning this; I’ll actually revisit this possibility near the end of this essay, in a different context.]

So, back to our sensory perceptions: unless in a given case we have a compelling reason to doubt the deliverances of our senses, we are justified in trusting those deliverances. For example then, if one sees/hears/feels/smells/tastes(?!) a cuddly, purring kitten on their lap, the simple occasion of perceiving the kitten is all the justification they need for believing that there is indeed a kitten on their lap. This kitten belief can be termed properly basic, or directly justified, with no further justification needed than the simple perceptual experience itself.

This kitten belief is defeasible, however. Say one comes to realize that they were strongly under the influence of hallucinogenic drugs whilst perceiving said kitten. Or, say compelling photographic evidence were later to indicate no such kitten. In such cases, one would indeed have good reason to be suspicious of their associated sensory perceptions. Such reasons for compelling doubt can be termed defeaters to this kitten belief. In other words then, this kitten belief is innocent until proven guilty, and such beliefs must be considered innocent until proven guilty if we’re to avoid radical, unworkable, self-defeating skepticism.

And so, as one example, simple perceptually-informed beliefs as “there’s a cuddly, purring kitten on my lap” can be termed properly basic absent defeaters.

[Pedantic note #2: I’m not very fond of this properly basic absent defeaters terminology. It encourages a framework that’s too flat and dichotomous. A belief is typically thought to be either properly basic or else it’s not, as if it falls cleanly into one of these either/or categories. Likewise, an occasion for doubt is either a defeater or else it’s not, as if there’s a clear dividing line. My inclination is to try to assign weight to various pieces of evidence and counter-evidence (perceptual or otherwise), along a rich spectrum. What if the ostensible kitten was roughly twenty yards away in a dimly lit area, and one was under the influence but only mildly so? What then? Does it make any sense to speak of this kitten belief as rigidly either/or properly basic? Defeated? Nah, let’s think instead of an evidential spectrum with a healthy agnostic zone smack in the middle.]

Reformed epistemology and God perception

None of this is terribly controversial nor unique to reformed epistemology (I’ll call it RE from this point, and it’s so-named because several of its chief figures have come from and have drawn from the reformed tradition within Protestant Christianity, though RE has no necessary association with this reformed tradition). Where RE begins to depart strongly from mainstream epistemology is in claiming that belief in God can be properly basic absent defeaters. Put another way, and at least in certain cases: God-beliefs are claimed to function in roughly the same immediate, innocent-until-proven-guilty manner as do simple perceptually-informed beliefs (eg, the kitten belief described above), and such beliefs remain justified so long as no insurmountable defeaters are in view. So if one *perceives* God (in some way), then the simple occasion itself of this perception serves to justify theistic belief. No external evidence or argument required, though one must remain on guard against potential defeaters.

So, what might such God perceptions actually feel like? How might one perceive God in a manner which feels sufficiently analogous to kitten perception? And not just a vague God-belief, but moreover a decidedly monotheistic belief, and even a highly specified theistic belief of, say, an Islamic or Christian variety?

In my own case, the sole occasion on which I’ve had a powerful and unmistakeable mystical experience roughly akin to kitten perception (which is one more such occasion than most theists have experienced, I would guess), the episode had a decidedly non-theistic flavor to it, even though I was very much a theist at the time and thus was likely pre-disposed to having experienced it in a theistic manner. And it most certainly didn’t include any esoteric propositional content along the lines of, “Jeff, the Bible is inerrant! (By that I–the Holy Spirit of the Triune God–am hereby generally affirming The Chicago Statement on Biblical Inerrancy, but with several caveats: First let me draw your attention to Section II, Article XVII, paragraph one…).”

Of course I’ve spoken anecdotally just now, but it’s quickly apparent that if one is trying to ground theism in personal mystical experience, theism will therefore be undermined for the large majority of people. And the situation becomes exponentially more dire as layers of theological complexity are added. A far more permissive process of belief formation is needed, clearly, if RE is to be of much help to theism.

Plantinga, the sensus divinitatus, and the “internal instigation of the Holy Spirit”

A common move, then, has been to speak of a sensus divinitatus (general in-born sense of God) and of an internal witness of God’s spirit (specific promptings from God). So, what are these processes supposed to feel like? The aforementioned Plantinga describes in Warranted Christian Belief how these processes might hypothetically feel under ideal circumstances:

Someone in whom [the sensus divinitatus] was functioning properly would have an intimate, detailed, vivid, and explicit knowledge of God; she would have an intense awareness of his presence, glory, goodness, power, perfection, wonderful attractiveness, and sweetness; and she would be as convinced of God’s existence as of her own.

A description which sounds similar to what continuous, profound mystical experience might plausibly feel like for some people. And despite my anecdotal note above that my own mystical experience had a strongly non-theistic flavor to it, I would have a fair amount of sympathy for any theist who might find themselves in the sort of ideal circumstances here described by Plantinga, although I’d still contend that it would nonetheless be a mistake to infer theism even under such circumstances. But in any event, it certainly seems that very few, if any, theists actually find themselves in such ideal circumstances on any regular basis. Again from Plantinga:

The sensus divinitatis has been heavily damaged by sin; for most of us most of the time the presence of God is not evident. For many of us (much of the time, anyway) both God’s existence and his goodness are a bit shadowy and evanescent, nowhere nearly as evident as the existence of other people or the trees in the backyard.

So then, how are these processes actually supposed to feel under normal (ie, “heavily damaged by sin”) circumstances? From Plantinga yet again; this time speaking of specifically Christian belief, although a similar description would seemingly apply to more general theistic beliefs as well:

What is said [ie, in hearing/reading Christian “Scripture”] simply seems right; it seems compelling; one finds oneself saying, “Yes, that’s right, that’s the truth of the matter; this is indeed the word of the Lord.” I read, “God was in Christ, reconciling the world to himself”; I come to think: “Right; that’s true; God really was in Christ, reconciling the world to himself!” And I may also think something a bit different, something about that proposition: that it is a divine teaching or revelation, that in [John] Calvin’s words it is “from God.” What one hears or reads seems clearly and obviously true and (at any rate in paradigm cases) seems also to be something the Lord is intending to teach.

Now, if you’re immediately wondering, “Well, why can’t anyone do that, on behalf of any belief, no matter how outrageously far fetched?” then you’ve asked a very sensible question indeed, which many of Plantinga’s professional peers have also asked in one form or another. These are the so-called “Great Pumpkin” types of objections (eg, “Why couldn’t Linus simply avail himself of a similar process in trying to justify his belief in the Great Pumpkin?”). Here’s how Plantinga replies regarding a hypothetical practitioner of voodoo attempting to adopt a similar model of justification:

It doesn’t follow, of course, that the voodoo epistemologist is also warranted in claiming that voodoo belief is properly basic with respect to warrant. For suppose voodoo belief is in fact false, and suppose further that it arose originally in some kind of mistake or confusion, or out of a fearful reaction to natural phenomena of one sort or another, or in the mind of some group hoping to gain or perpetuate personal political power. If so, then those original voodoo beliefs did not possess warrant. Suppose still further that these voodoo beliefs were passed on to subsequent generations by way of testimony and teaching. Now if a testifier testifies to some belief p that has no warrant for her, then p will also have no warrant for anyone believing it on just the basis of her testimony. If p has no warrant for the testifier, then it has none for the testifiee either.

Now, if you’re immediately wondering, “Well, why can’t I object to theism in exactly the same way?” then you’ve asked another very sensible question indeed. Might it potentially be the case that most theistic beliefs trace similarly back to “some kind of mistake,” and therefore might most theists similarly need to justify their beliefs by sifting carefully and extensively through the chain of experiences and evidences which produced such beliefs?

The critical difference, according to Plantinga, is that voodooists can’t appeal to any processes suitably analogous to the sensus divinitatus and “internal instigation of the Holy Spirit” (or IIHS). So voodooists have a considerable burden of proof to meet; a burden from which Plantinga exempts his fellow theists (as he lists them, in addition to Christians: “[adherents of ] Judaism, Islam, some forms of Hinduism, some forms of Buddhism, some forms of American Indian religion”).

In other words, according to Plantinga: theists are in a unique epistemic position such that all they need to provide by way of justifying their theistic beliefs is a fuzzy, “[This] simply seems right; it seems compelling.” [Yes, he does admit of the possibility of defeaters for theistic belief–more on that in a bit.] But despite his hasty assertions to the contrary, nearly anyone can craft a suitable just-so story to pull a similar kind of move, including voodooists, flat earthers, metaphysical naturalists, and others on his naughty list. I’ll return to this point near the end of this essay, but first I want to take a detour and introduce one other variety of RE:

Rauser and properly basic testimonial belief

Randal Rauser is an intellectual protégé of Plantinga, even having studied directly under him for a stint. Following the groundwork laid out by Plantinga, Rauser has developed a version of RE which is similar to but also importantly different from Plantinga’s. And I know Rauser’s work rather intimately, having spent far too much time over the course of several years engaging directly with him (he’s a generous host who engages very interactively with his online community, and he seems a decent fellow with whom I’ve got quite a lot in common).

Though Rauser doesn’t object in principle to the sensus divinitatus and IIHS processes of direct divine testimony-of-sorts, he’s turned his own attention primarily toward the “less exotic” (to use his words) process of simple human testimony. Following many mainstream epistemologists, he considers testimony to confer proper basicality upon testified-to beliefs, under the right conditions (eg, when it comes from trusted sources, etc).

[Pedantic note #3: It makes no sense to me to speak of properly basic testimonial belief, as that would seem to imply that the simple brute fact of the testimony itself serves to justify the testified-to belief. But of course we’re always weighing testimony in an evidential way, as is clearly implied by the “under the right conditions” proviso attached by Rauser and other epistemologists. For our purposes here, evidence need not be crudely restricted to direct corroborative sensory experiences, but can instead encompass a wide variety of indirect and inferential corroborative processes (eg, appealing to expert consensus), and it’s often assessed by cognitive heuristics of which we may not be actively aware. So I can’t see how we ever accept testimony in an entirely non-evidential way. But anyway, no matter, let’s grant testimony as a potential source of properly basic belief, for the sake of argument. There’s no need here to get sidetracked on the endlessly stale debates about various candidate “criteria for proper basicality.”]

So, on Rauser’s view, so long as one is testified to under appropriate conditions, that testimony itself is all the justification one needs for adopting the associated beliefs. For example, then, if one receives testimony from a trusted source (eg, parents, teachers, religious leaders, etc) that theism is true, or even that specifically Christian theism is true, or even that specifically Reformed Baptist Church of God, Reformation of 1879 -styled Christian theism is true, then the simple fact of the testimony itself serves to justify the associated beliefs, and no further corroboration is needed. And although many other epistemologists would agree with Rauser that testimony can confer proper basicality, Rauser casts the “under the right conditions” net far more widely than most.

To his credit, Rauser doesn’t seem to shy away from admitting that he’s radically permissive here on the front end of the belief formation process. He’s happy to grant that a similar process of belief formation might be properly adopted even by those Plantinga has tried to exclude. Here’s how Rauser puts it in The Swedish Atheist…, in a mock conversation between himself and a fictional discussion partner named Sheridan:

“Look, let me be blunt [says Sheridan]: What separates your God beliefs from good old-fashioned magic? If you can believe that an invisible sky God made the sunset, then why can’t I just believe that a giant pink dragon created the sunset? If you give up requiring people to defend their beliefs with evidence, it becomes a free-for-all where everyone can believe what they want.”

“I take your pink dragon idea to be a form of reductio ad absurdum. That is, you’re trying to argue against my view by claiming that it leads to absurd consequences. Thus, if I give up the obligation to provide evidence that God created the sunset then I have to accept the absurd consequence that another person could believe without evidence that a pink dragon created the sunset.” Sheridan nods.

“Okay,” I say, “I guess I don’t have a problem conceding that. In principle it’s possible that the belief that a pink dragon created the sunset could be properly basic. This shouldn’t be too surprising, since all sorts of beliefs that appear to be extraordinary at first blush can be properly basic.”

Remember the full properly basic absent defeaters terminology from earlier in this essay? Rauser is radically permissive of the properly basic (ie, innocent until proven guilty) half of that equation, focusing more on the defeaters half of the equation as a way to separate the wheat from the chaff. And despite my pedantic side-note complaints that I don’t especially like this properly basic absent defeaters terminology in the first place, I’m willing to grant nearly the entire RE framework (typically: a proper functionalist Reidian externalist moderate foundationalism, phew!), so long as we can at least treat defeaters in a reasonable manner. In a manner, in other words, which actually can effectively separate the wheat from the chaff, the guilty from the not guilty, on the back end of the belief formation/retention process.

Defeaters

So what exactly is a defeater? There are, according to Plantinga: rebutting defeaters, undercutting defeaters (which together constitute rationality defeaters), warrant defeaters, purely epistemic defeaters, defeater defeaters, meateater defeaters (alright, jk on that last one), the list goes on.

A barrage of jargon and dubiously helpful distinctions.

But what’s almost entirely missing from the work of Plantinga and Rauser is a rich discussion of how to weigh beliefs and defeaters against one other in a useful way. I’ve personally pressed Rauser on a number of occasions for any sort of useful operational definition of defeater, but never with any luck. Here’s how Plantinga defines a “defeater simpliciter”:

D is a defeater of B for S at t if and only if (1) S’s noetic structure N (i.e., S’s beliefs and experiences and salient relations among them) at t includes B, and S comes to believe D at t, and (2) any person (a) whose cognitive faculties are functioning properly in the relevant respects, (b) whose noetic structure is N and includes B, and (c) who comes at t to believe D but nothing else independent of or stronger than D would withhold B (or believe it less strongly).

Bless analytic philosophers in their attempts at precision, but that seems a needlessly dense way to say, basically, that a defeater is a consideration which ought to rationally compel one to abandon their associated, now-defeated belief.

Which initially sounds fine and all, but upon closer inspection it’s simply not a helpful definition. As I complained earlier in this essay, it’s all so very flat and dichotomous: a belief is either defeated or it’s not, as if there’s a stark binary distinction to be made; and no precision whatsoever is offered regarding where that dividing line ought to be placed. I’ll give Plantinga a little bit of credit for nuancing this with a mention of the possibility of “believ[ing] less strongly,” but I can’t see that he puts such nuances into meaningful practice when it comes to his core theistic beliefs.

I could more or less be appeased if defeaters were to be handled in the following way: a defeater is a consideration which–given one’s current web of beliefs/experiences/available information–renders an associated belief as probably false; that is, as more likely false than true. And therefore as a belief which ought to be abandoned, at least until new countervailing considerations come to light.

It still comes across as fairly flat, rather than encouraging a rich spectrum of verdicts ranging from almost certainly false to almost certainly true, with a large and healthy agnostic zone in between. But at least the basic insight is captured: that we’re dealing with probabilities here; that we’re weighing various evidences and counter-evidences against one another in a probabilistic way.

So in other words, fine, knock yourself out: your theistic beliefs are properly basic absent defeaters. You don’t need evidence to initially adopt them. Ok, if you say so. But the moment potential defeaters come into view–which for most contemporary folks ought to be nearly right away, so long as they’re at least marginally inquisitive–then the epistemic process boils down anyway to the weighing of evidence.

This is perhaps why Rauser, Plantinga, and others within the RE camp appear to be so resistant to treating defeaters in the manner I’ve just proposed, at least regarding the perceived core of their theistic beliefs. They seem not to want the process to be probabilistic, evidential. Of course they typically handle fairly mundane beliefs in a sufficiently evidential manner, with a comparatively modest threshold for what might constitute a defeater in such cases. But as the beliefs in question approach nearer to the core of their Christian theism (ie, the “great things of the gospel” as Plantinga puts it, borrowing a phrase from Jonathan Edwards), the hurdle is raised to increasingly dizzying heights. To the eventual point that–and here’s the crux of the problem–only clear and compelling demonstrations of thoroughgoing impossibility (or something very close to that threshold) might be considered adequate to dislodge the associated beliefs.

Let me note that it isn’t terribly uncommon, for a wide variety of people in a wide variety of contexts, to similarly privilege beliefs perceived to be at or near the core of their view of the world. And within reasonable limits, this sometimes can be epistemically appropriate: Ideally, beliefs near the center of one’s “plausibility structure” enjoy that status because a huge collective weight of confirming evidence has gradually placed them there, and thus a correspondingly huge weight of counter-evidence would need to surface to finally dislodge such beliefs. (Though I say “ideally” because it’s incredibly easy, for humans generally, to confuse emotional weight with rational weight and to severely distort and belittle any counter evidence.)

But the striking case is this: reformed epistemologists typically contend that their core theistic beliefs require no initial support beyond mere testimony (whether human or, ostensibly, divine), and furthermore, that they’re under no epistemic obligation to abandon these beliefs unless and until they’re demonstrated to be altogether logically impossible, or very nearly so. And it’s the conjunction of these two positions that in particular is so terribly problematic. It’s not only that the hurdle on the front end of the belief formation process is radically low, and it’s not only that the threshold of disproof on the back end is almost impossibly high; it’s that both ends of the process are thus so drastically weighted toward one’s favored beliefs, whatever those beliefs might happen to be.

I know that might very well sound contentious: that Rauser, Plantinga, and others in the RE camp have adopted something approaching logical impossibility as the standard of disproof for their core theistic beliefs. I haven’t known Rauser or Plantinga, for example, to put it in terms quite this stark, but I hope to demonstrate here that this is indeed how they appear to handle such matters in actual practice. (And it’s not just them, or those within the RE camp; as I’ll argue later in this series, theism appears to fundamentally require such a standard, if there’s any hope of salvaging it.)

Rauser and disproof

In The Swedish Atheist… Rauser briefly discusses how one might try to provide a defeater to the belief “God loves me.”

Let’s first take a moment to consider how we might typically demonstrate that some other human doesn’t love us. If, for example, that other person were to willfully allow us to suffer in terrible ways (say, by allowing us to be burned horribly in an accident that they could easily have anticipated and prevented), and for no apparent outweighing beneficial purpose, then it would be a pretty slam dunk case to infer that they don’t love us, at least not in any actually-meaningful way. Of course, it would still be remotely possible that they really do love us, and that they allowed us to suffer the terrible accident for a sufficiently good and appropriate purpose (a purpose they’re not willing/able to share with us for some odd reason). But as it stands, that would be nothing more than a remote possibility, which we should reject as almost certainly false. We would have made an evidential, probabilistic case (and a very strong one at that) that they do not love us.

I don’t think I’ve made any very controversial claim in the preceding paragraph. This is exactly how we would typically argue such a case when only human agents are involved. But for some reason, the rules of the game are thought to change the moment God is substituted into the picture. This evidential, probabilistic process is discarded in favor of an absurdly high standard of disproof wherein only a demonstration of logical impossibility (or something very close to that) would count as a defeater. Rauser offers the following two example possibilities for what might constitute a defeater for the belief “God loves me”:

• A demonstration that God simply doesn’t exist, by way of a demonstration that theism is logically incoherent/impossible.
• A demonstration that theism is entirely the product of wish fulfillment. Unfortunately Rauser doesn’t go on to outline how this might in principle be done (he simply mentions it in passing as a possibility), but if you’re familiar with his style of argumentation, you’ll know that he’s looking here for a case that’s immune to any counter examples, or very nearly so.

What’s notably absent is anything along the evidential lines I proposed a moment ago, in my example case involving two humans. It’s as if–in order to establish that some other human doesn’t meaningfully love you–you’d need to first establish that that other person simply doesn’t exist. Which is a bizarre demand, to say the least. The evidential weighing of the so-called “facts of evil” around us never enters Rauser’s discussion.

One more example from Rauser’s work: His one-time coauthor Justin Schieber asserted that “Heaven is a place where free will is preserved yet has no evil. Such a world is possible for God to create from the start. He chose not to.” Rauser replied, “If [Schieber] wants to present an objection to theism he needs to show that God doesn’t have morally sufficient reasons [for choosing not to create “heaven” from the start].”

It might seem a bit weak sauce to pull an example from a tweet, but it’s a relatively recent example which prompted a direct exchange between Rauser and me (and a feisty one at that–both parties guilty!). I asked him:

Randal, it is your position/practice–is it not?–that for something to count as a defeater to your theistic views, it must demonstrate that theism is very nearly impossible?… You’re clearly implying [in the exchange with Schieber quoted above] that to count even as an objection to theism, Justin’s argument must demonstrate that theism is logically impossible.

His reply:

So here’s an analogy. Imagine that you see Dave spreading manure on his lawn. Since you’re unaware of any reason why Dave would do that you conclude that Dave must be inept at lawn care. I then come along and point out the obvious: before you conclude that Dave is inept at lawn care you need to provide evidence that he doesn’t know what he’s doing. That’s just good sense (and common courtesy).

The same courtesy that you extend to Dave should be extended to God, and those who believe in God. And your attempt to shirk that burden of proof by protesting that folks are demanding “an impossibly unreasonable standard of (dis)proof” is nothing more than handwaving.

My reply:

So in your view, how might one go about demonstrating that God does not have a morally sufficient reason for x, y, z? By demonstrating a logical inconsistency? Or something else?

His reply:

Theism is no harder to falsify than that hydra-headed beast known as “naturalism”.

But it is interesting to see the atheist asking the theist for help with critiquing the theist’s beliefs. I’ll let you figure that out. But your frustration that you’re having trouble proffering evidence to undercut or rebut a truth claim is not, in itself, evidence against the truth claim or those who hold it.

My reply:

No, I’m not asking for help with critiquing theism. I’m pointing out that the only standard of disproof you’ll accept is that of logical impossibility, or something very close to that.

Regrettably at that point, the discussion fell completely off the rails (and again, I’m partly guilty, hehehe). But had I been barking up the wrong tree with this whole logical impossibility thing, it would have been so easy for Rauser to simply set the record straight. That he chose not to (and has always chosen not to, in similar previous exchanges) seems quite telling to me, and as close to a confirmation of my suspicion as I’m likely to get.

Plantinga and disproof

Plantinga follows a slightly different route to arrive at approximately the same destination.

He delves more into the language of evidence and probabilities, examining and praising, for example, the “subtle and sophisticated” evidential arguments from evil of William Rowe and Paul Draper, which nevertheless “fail resoundingly,” as must nearly all proposed evidential defeaters to theism.

First he considers an example of what might demonstrate to his satisfaction that theism is simply false: namely, an argument showing a straightforward logical contradiction between the existence of God and the presence of evil:

Now from an atheological point of view, the old argument for inconsistency in Christian belief had a lot to be said for it. It was short and sweet; if there is a contradiction in Christian belief, then Christian belief is false, and that’s all there is to it. It doesn’t matter what else is or isn’t true, and it doesn’t matter whether there are any good arguments or evidence of other kinds for Christian belief: if it is inconsistent, it’s false, and that settles the matter. Furthermore, once you see that a proposition is false, you can’t rationally continue to believe it; so such an argument would show at one stroke that Christian belief is false and that it is irrational, at least for those apprised of the argument.

He concludes however that “it is [now] widely conceded that there is nothing like straightforward contradiction or necessary falsehood in the joint affirmation of God and evil.” And I’ll agree–I can’t see an obvious logical contradiction here. It’s a bold and difficult task to demonstrate a clear logical contradiction in something so conceptually complex as theism.

So he then turns his attention toward evidential arguments from evil. Arguments, in other words, which try to establish that the existence of God is unlikely (rather than impossible) given the facts of evil. Specifically, he considers whether theism might be evidentially challenged, as he puts it. A proposition (eg, “God exists”) is evidentially challenged if there’s an alternate proposition (eg, “God does not exist”), which is on approximately the same epistemic footing, and which better fits the relevant data. In what kinds of cases might such an evidential challenge be problematic?

Here we think first of scientific hypotheses. I propose a hypothesis H* to explain the behavior of gases: you point out that certain data are more probable with respect to another hypothesis H’ incompatible with mine; that certainly seems to be a strong prima facie reason to doubt my hypothesis.

However:

It is an enormous assumption to think that belief in God or, more broadly, the larger set of Christian (or Jewish or Muslim) beliefs of which belief in God is a part, is in this respect like a scientific hypothesis. Not only is this assumption enormous: it is also false. The warrant for these beliefs, if they have warrant, does not derive from the fact (if it is a fact) that they properly explain some body of data. For most believers, theistic belief is part of a larger whole (a Christian or Muslim or Jewish whole); it is accepted as part of that larger whole and is not ordinarily accepted because it is an explanation of anything; hence its rationality or warrant, if it has some, does not depend on its nicely explaining some body of data.

He gives the following analogy:

Compare the case of Maynard [an example he introduced earlier] and my belief that he is a cat. You point out that this belief suffers from an evidential challenge: that he likes cooked green beans is much less likely on his being a cat than on his being a Frisian. I agree, but am undeterred, continuing in full rationality to believe that he is indeed a cat [because his belief that Maynard is a cat is grounded in properly basic sensory perception].

And then concludes:

There is of course no cognitive malfunction involved in my continuing to hold a belief with significant warrant from such sources as memory, perception, IIHS, and the like, even when I learn that the belief is subject to an evidential challenge.

All of that is to say, in other words, that because theistic belief is (according to Plantinga) analogous to sensory perception, and it’s not subject to a strict logical contradiction, it can therefore withstand any evidential challenge. Other than, perhaps, in two kinds of cases he can imagine:

First, suppose I am a theist, am rational in accepting this belief, but hold it with little firmness and furthermore think my reasons for it are absolutely minimal—barely sufficient for holding the belief rationally. Then if I learn that theism is subject to an evidential challenge, perhaps I have a defeater for it. I say ‘perhaps’ advisedly; the situation isn’t really clear.

The second sort of situation is clearer. Consider a belief B I accept because I think it the best explanation of a certain range of data D [in other words, contra most theists, one holds theism not primarily on the basis of ostensible processes like the sensus divinitatus/IIHS, but rather, because it is thought to best explain some data set]; B has no warrant apart from its properly explaining D, and I am aware of this fact. Finding that B is subject to an evidential challenge, one thinks, gives me a defeater for it—provided that the belief that is more probable with respect to the alternative hypothesis is one that B is supposed to explain.

So to rewind and recap: On the basis of a fuzzy, “[This] simply seems right; it seems compelling,” theistic beliefs are very nearly immune to evidential challenges, according to Plantinga. I’ll repeat the William Lane Craig quote I provided near the outset of this essay:

What I’m claiming is that even in the face of evidence against God which we cannot refute, we ought to believe in God on the basis of His Spirit’s witness. Apostasy is never the rational obligation of any believer, nor is blasphemy against the Holy Spirit. God can be trusted to provide such powerful warrant for the great truths of the Gospel that we will never be rationally obliged to reject or desert Him.

I haven’t found in Plantinga’s work any single quote that’s quite so bombastic, but for the life of me I can’t see how his position differs in any important respect from Craig’s. It’s epistemically disastrous.

Just-so stories

What the RE framework boils down to, at least in most of its varieties, is a just-so story. As long as one can craft a suitable story about how one comes by their theistic beliefs (“It’s the heavily-sin-damaged sensus divinitatus!” “The IIHS!”), and so long as such a story isn’t subject to something approaching a logical contradiction, then anything goes. In order to undermine such beliefs, it is thought, one must simply start from the outset with the assumption that such a story is false. And that’s hardly fair, right?

Plantinga seems heartened by the fact, as he sees it, that theism provides the only suitable just-so story (of course, he wouldn’t call it a “just-so story”). So it might be difficult to rationally bridge the differences between, say, Christians and Muslims, but at least non-theistic beliefs are excluded from such justifications, like voodooism, flat earthism, and metaphysical naturalism. Oh, and Great Pumpkinism. (Actually, all of these save metaphysical naturalism could easily be grafted into theism; or at least, a theism-of-sorts in the case of voodoo.)

But it’s really not hard to construct a suitable story for all of these -isms as well. Let’s throw a little love Linus’ way:

Let’s say that Linus is a Great Pumkinist, voodooist, flat earthist, metaphysical naturalist, all rolled into one. (It might be a little awkward to reconcile voodoo with metaphysical naturalism, but perhaps he adopts a subtle, liberal version of voodoo that’s compatible with naturalism?) In any event, on the basis of Linus’ proposed sensus pumpkinatus and IIPS (internal instigation of the Pumpkin Spirit), Linus believes that the “great things of the Good Pumpkin News (GPN)” simply seem right; they seem compelling.

And what are the great things of the GPN? First and foremost: that the Great Pumpkin will visit, one of these Halloweens. Also: that the earth is flat, that voodoo is the Truth, that metaphysical naturalism is true, and perhaps other items as well. I’m not quite sure how these items all relate to one another, but the Great Pumpkin’s ways are higher than ours.

And I can’t see any clear logical contradiction in any of these claims, so it would appear to me that Linus’ epistemology is fully on the up and up by the standards of RE. No doubt Plantinga wouldn’t at all be satisfied with such a story, because it’s not obviously analogous–based on my description so far–to the ostensible epistemic situation in the case of theism (ie, an ESP-like communicative process from a suitably trustworthy, knowledgeable, epistemically well-placed source). Now, I’m not personally convinced–for reasons I won’t get into here–that any more actually needs to be said in favor of the epistemic status of Great Pumpkinism, assuming the standards of RE. However, I know it’s all a bit much to take in (“Wait, who’s this Great Pumpkin and why should we take the IIPS seriously?”), so let me try to throw a little sheen on it and elaborate on the story, to show how it could be sufficiently analogous to the case of theism:

Let’s say we’re actually living in a computer simulation (told you I’d get back to this, and it’s really not so far fetched as it sounds). Let’s say further that our simulation is actively, externally superintended by someone who communicates with a chosen few of us within the simulation, in subtle ways. And specifically, this superintendent (aka the “Great Pumpkin”) communicates primarily with Linus (so Linus says), the high priest of Great Pumpkinism, who then shares the great things of the GPN with the rest of us.

We don’t quite know for sure why this superintendent is so enamored of pumpkins; why they’ve chosen to allow mass modern confusion as to the flatness of our simulated earth; what exactly this subtle, liberal version of voodoo has going for it; and how exactly they could know that metaphysical naturalism is true.

But whatever the case, we Great Pumpkinists have sometimes directly, and more often indirectly through Linus, tasted of the deep wisdom and goodness of the Great Pumpkin. And we know that our epistemic horizon within the simulation is radically impoverished relative to the epistemic horizon enjoyed by the Great Pumpkin, who sits external to our simulation. So although the Great Pumpkin might like to explain all the mysteries of the GPN to us, we simply don’t enjoy enough epistemic overlap with the Great Pumpkin for that to be feasible. Accordingly, we take the great things of the GPN on faith in the good and wise Great Pumpkin.

Are there actually any Great Pumpkinists? Hehe, not that I know of. But if there are, they might enjoy helping themselves to three heaping scoopfuls of reformed epistemology.

Up next

Next time (so, maybe 2019?) I’ll be leaving the “Denying the problem” section of this series and diving into the problem of evil proper. Discussing theodicy, which seeks to provide explanations for why, assuming God exists, we should actually expect the various evils we see around us. I know you’re waiting with bated breath!