Alvin Plantinga is not your average apologist. He does not reach for sentiment or tradition or the argument from personal testimony. He is a rigorous analytic philosopher, a former president of the American Philosophical Association, and he has spent decades constructing the most technically sophisticated defence of Christian theism available in contemporary philosophy. If you are going to argue against religious belief on epistemological grounds, Plantinga is the person you have to answer. Most critics of religion do not bother. That is a mistake.
His argument from proper basicality is the one I want to address here, because it is the argument that most directly challenges the epistemological case against religious belief, and because it is far more interesting, and far more revealing, than its defenders typically acknowledge.
Plantinga’s claim, stated plainly, is this. Not all rational beliefs require evidential justification. Some beliefs are what he calls “properly basic,” meaning they are rational to hold without being derived from evidence or argument. He gives examples: my belief that I am looking at a tree, my belief that other people have minds, my memory that I had breakfast this morning. I do not infer these from a chain of evidence. I simply find myself with them, and they are rational because of the faculty that produces them, not because of the evidence that supports them.
Plantinga then argues that belief in God can function in exactly the same way. A person who finds themselves, in moments of beauty or moral awareness or contemplation, with a sense of God’s presence, is not being irrational. They are operating a cognitive faculty, what Calvin called the sensus divinitatis, that produces belief directly, without inference. The belief is properly basic. Demanding evidence for it misunderstands its epistemic status, in the same way that demanding inferential justification for my belief in other minds would be a category error.
It is a genuinely clever argument and I want to resist the temptation to dismiss it quickly, because quick dismissal is what Plantinga anticipates and has already answered. The serious objection is this.
The argument only works if the faculty producing the belief is reliable. My perceptual faculties are generally reliable in producing beliefs about the external world. We have extensive evidence of this, from the internal coherence of perception, from intersubjective agreement, from the practical success of acting on perceptual beliefs. My memory is imperfect but broadly reliable, and we have independent means of checking it. The properly basic belief is rational precisely because the faculty producing it can be shown, by other means, to be trustworthy.
The sensus divinitatis cannot be shown to be trustworthy by any independent means whatsoever. Plantinga’s response to this is to say that the faculty is functioning properly if God exists and has designed it to do so. But this is circular. The reliability of the faculty is conditional on the truth of the belief it produces. You cannot use the faculty to justify belief in God and simultaneously use belief in God to justify the faculty. That is not a properly basic belief. That is a closed loop.
There is a further problem that Plantinga’s framework generates and does not adequately resolve. If the sensus divinitatis is a genuine cognitive faculty producing properly basic beliefs, there is no principled reason why it should produce specifically Christian theism in one person, Islamic theism in another, animist belief in a third, and nothing at all in a fourth. The phenomenology of religious experience is radically diverse and frequently mutually contradictory. Properly basic perceptual beliefs show remarkable cross-cultural convergence because they track a shared external reality. Properly basic religious beliefs show the opposite pattern. They track the tradition into which the believer was born with a reliability that is embarrassing to the faculty that supposedly produces them.
Plantinga’s answer to this is that sin and cultural interference can impair the sensus divinitatis, producing false or absent beliefs. But this is precisely the kind of unfalsifiable auxiliary hypothesis that the argument was supposed to rise above. Any belief-forming faculty that produces the right results when it works and can be excused when it produces the wrong results by appeal to interference is a faculty that no longer has any predictive or evidential content. It explains everything, which means it explains nothing. Plantinga has a further argument worth noting: he contends that evolutionary naturalism faces an equivalent circularity problem, since any cognitive faculties produced by blind natural selection have no guaranteed reliability either. This deserves a full response and will get one. For now it is enough to observe that two unreliable systems do not make one reliable one, and that pointing at the neighbour’s window does not repair your own.
What Plantinga has actually constructed, with considerable philosophical sophistication, is an elaborate permission structure. He has built a framework in which theistic belief is declared immune from the normal demands of evidence, on the grounds that it is a different kind of belief, produced by a different kind of faculty. And when that faculty fails the tests you would normally apply to any belief-producing mechanism, the failure is attributed to noise in the system rather than defects in the faculty.
This is not the same as “God works in mysterious ways.” It is more careful, more technically dressed, and far more difficult to dismiss in a sentence. But it arrives at the same destination. The belief is placed beyond the reach of ordinary epistemic standards, and any challenge to it is reclassified as a misunderstanding of its nature.
The difference between Plantinga and the parish priest is not that one has solved the epistemological problem. It is that one has built a more elaborate gate around it.
Philosophers sometimes speak of the “God of the gaps,” the theological strategy of placing God wherever scientific explanation has not yet reached, retreating as the gaps close. Plantinga’s properly basic God is something more sophisticated: a God of the epistemological gaps, placed precisely where the demand for justification is most inconvenient, and defended by the claim that inconvenience itself is a kind of category error.
The serious question his work raises is not whether properly basic belief is a coherent concept. It is. The serious question is whether a belief-forming faculty that cannot be independently validated, that produces radically divergent results across persons and cultures, and that retreats to claims of interference whenever it fails, is one that any honest epistemology should take seriously.
The answer, I think, is that Plantinga has given us not a defence of theism but an extremely precise description of its problem. The sensus divinitatis, as he has defined it, is a faculty that works when you already believe and fails when you do not, that produces conveniently different results in every tradition that invokes it, and that is protected from falsification by design. He has described, with more rigour than almost anyone, a mechanism for belief that is systematically insulated from the possibility of being wrong.
That is not a properly basic belief. That is a properly unfalsifiable one. And the sophistication of the argument does not make the destination any less familiar.
The Permanent Rebuttal 