Home        Center for Naturalism       Applied Naturalism        Spirituality       Philosophy 

[back to Epistemology, Atheism]


Too Good to Be True, Too Obscure to Explain:
Cognitive Shortcomings of Belief in God

 Thomas W. Clark, Center for Naturalism

This essay originally appeared in 50 Voices of Disbelief: Why We Are Atheists, edited by Russell Blackford and Udo Schuklenk and published by Wiley-Blackwell, 2009. It is reprinted here by permission of Wiley-Blackwell. My sincere thanks to Russell and Udo for their hard work in putting this diverse, insightful and international collection together. Please see here for contents, reviews and purchasing information at Amazon, here for the book's John Wiley & Sons publication page.

For a philosophical and scientific naturalist such as myself, the traditional Christian god is ruled out simply because the existence of the supernatural in general is ruled out. If you stick with science as your guide to what’s ultimately real, and critique your assumptions in open philosophical inquiry, there are no good reasons to believe that reality is split between two categorically different realms, the natural and the supernatural. Instead, science reveals that the world is of a piece, what we call the natural world. Disbelief in God, therefore, is a corollary of the rationally defensible claim that nature is all there is, the basis for the worldview known as naturalism.

Epistemic commitments of naturalism

Naturalists are driven by the immodest desire to plumb the depths of reality, to know what objectively exists, to understand how things fundamentally work, and to have maximally transparent explanations of phenomena. In this project our primary commitment is epistemic, to a philo-scientific way of knowing that we justifiably believe gets us reliable beliefs about the world. I call this a philo-scientific epistemology because it combines openness to philosophical critique with a reliance on scientific criteria of explanatory adequacy as vetted by that critique and the actual practice of science. Naturalism holds that science and philosophy are continuous, interpenetrating and collaborative in our investigation of reality; neither is foundational to the other. The naturalist mainly wants not to be deceived, not to make errors of logic or method or assumptions when understanding the world. Science, kept presuppositionally and methodologically honest by philosophy and real-world experience, has given us increasingly reliable explanations of how things work as judged by our growing capacity to predict and control phenomena. Such is the naturalist’s pragmatic test of knowledge: we are not deceived because we successfully predict.[1] 

Because naturalists are driven by the quest for reliable knowledge, we are not in the business of defending a particular picture of what finally exists, a particular ontology. If the best, most transparent explanations of a phenomenon, for instance consciousness, should end up in some sort of mental-physical dualism, so be it. If, in our astrophysical explorations, we discover that a race of super-beings created the observable universe, so be it. We are ontological non-dogmatists, letting the ontological chips fall where they may just so long as the theory specifying the ontology is the best one going. We jealously reserve the right to be mistaken in our view of what exists, given that theories often change under pressure from further investigation.

The unity of scientific explanations

Even though an ontological dualism might conceivably surface in our investigation of the world (there is thus far no indication that it necessarily will), the investigation itself militates against the possibility of a root metaphysical divide between the natural and supernatural. This is because whatever is brought within the orbit of philo-scientific cognition is necessarily shown to be in relation, either proximate or remote, to everything else within its purview. What scientific theories aim to do, after all, is show the causal, temporal and structural relationships that obtain between various levels and domains of phenomena, for instance between the atomic and the chemical, chemical and biological, and eventually, perhaps, between the biological, psychological, behavioral and economic. The diversity of the animal kingdom, the complexities of human thought and culture, even consciousness itself – all this can in principle, and increasingly in practice, be traced back through biological, geological, stellar and cosmic evolution to the Big Bang. Because the empirical understanding of the world is inherently unifying, those engaged in it without prior metaphysical attachments, such as belief in God, are often led to naturalism: that reality is constituted by a single, interconnected whole, however disparate its domains, levels, parts and characteristics. We call this whole, this reality, nature. Note that such naturalism isn’t a philosophical bias imposed on science by naturalists, as some anti-naturalists like to claim,[2] but rather an entailment of the cognitive commitment to science as the basis for reliable beliefs. The overarching and underlying unity of nature is produced by the philo-scientific project of seeking transparent, reliable, prediction-generating explanations.

The explanatory poverty of the supernatural

From the point of view of this project, the existence of the supernatural is simply unmotivated, an explanatory non-starter or superfluity. The supernatural, after all, is just that which cannot find a place in an empirically well-supported theory. If it did, it would cease to be supernatural – it would be immediately naturalized by its observational and theoretical connections to other natural phenomena, those entities and processes that do have a place in the theory. The project of gaining secure empirical knowledge therefore undermines the plausibility of and need for supernatural explanations. Indeed, the single most salient point (perhaps the only point!) of agreement among philo-scientific naturalists – an argumentative, fractious crew – is about the non-reality of the supernatural.[3] If, as the naturalist contends, the most reliable grounds for believing in something’s existence is that it plays a role in our best, most predictive explanations, then there’s no good reason to believe in the supernatural, since nothing supernatural plays such a role. Of course the naturalist doesn’t claim to be able to disprove the existence of the supernatural, but lack of disproof is not proof of existence. If it were, one’s ontology would necessarily expand to include all logically conceivable entities, however scant the evidence for them – an unwieldy universe indeed.

The traditional Abrahamic god, a prime exemplar of the supernatural, is a patently unexplained explainer and thus necessarily absent from an ontology driven by the demand for explanatory transparency. Whether God is brought in to explain the creation of the universe or the design of life, in neither case can the supernaturalist provide an account of God’s nature or how he operates. But good explanations don’t simply posit the existence of some entity or process to fill a purported explanatory gap, in this case a creative, designing intelligence; they must supply considerable additional information to achieve explanatory adequacy. A good theistic explanation would have to, for example, supply concrete specifications for God – his motives, characteristics, powers and modes of operation – to shed light on how and why he created certain species and not others, for instance. It would also have to show his relationship to antecedent and surrounding conditions: his historical provenance, his ontological status (mental, physical, or what?), and, not to put too fine a point on it, his current location. Further, an adequate theistic explanation would have to provide independent intersubjective evidence for God’s existence beyond his posited role as creator-designer. Without such evidence, in principle available to any impartial observer, there are no reliable grounds to suppose he exists.

Theists are unable to meet these basic criteria of explanatory adequacy, criteria which according to naturalists should apply to all entities in good ontological standing.[4] This makes God an ad hoc gap filler, an evidentially and theoretically unwarranted excrescence. No wonder then that, despite the claims of creationists and proponents of intelligent design, God plays no role in scientific accounts of human and cosmic origins. Those wanting clear explanations can’t abide the spurious explanatory completeness that God supplies; such completeness is patently bought by sacrificing understanding, when after all understanding is the whole point! No, naturalists are happy to admit that in some cases – many cases actually, including the origins of existence itself – we don’t understand what’s going on. Far better an honest admission of naturalistic unknowing than a premature claim to knowledge that invokes the supernatural. Belief in God, a cognitive cul-de-sac, is ruled out by the naturalist’s desire for explanatory transparency, a transparency exemplified by science.

The demands of objectivity

But defenders of God sometimes argue that the naturalist’s commitment to science, however philosophically sophisticated, is too narrow. Were we to expand our epistemic horizons and use non-scientific as well as scientific modes of knowing, we would find that nature is not all there is.[5] From this standpoint, the argument about God’s existence boils down to an argument about epistemic norms, about our standards for having confidence in our beliefs. Anti-naturalists are more epistemically liberal than naturalists, granting objective warrant to beliefs certified by, for instance, personal intuition and revelation, folk psychology, religious traditions, and textual (e.g., Biblical, Koranic) authority. In addition, some anti-naturalist philosophers such as Alvin Plantinga put relatively more stock in purely rationalistic proofs against naturalism (and thus for God’s existence), while downplaying the need for observational evidence for God.[6]

Both sides, however, are making claims about how the world objectively is, as contrasted with merely subjective appearances. Neither side will admit to being systematically deceived or otherwise misled in picturing reality. Both naturalists and anti-naturalists should agree, therefore, that our modes of cognition should, as much as possible, insulate factual claims from the influence of bias, wishful thinking, and other motivational contaminants. If they do not, then we’re at risk of projecting our human hopes and categories onto the world instead of grasping its true nature. Call this the insulation requirement. Meeting this requirement is incumbent on any worldview that purports to represent reality objectively, and thus applies equally to naturalism and all varieties of anti-naturalism – theism, supernaturalism, paranormalism, New Age worldviews, etc. The inescapable demand of any claim to objectivity is that we do our level best to separate how we wish things would be from how they actually are.

Projecting God

The question about God’s existence then becomes the following: is there any reason to prefer naturalism over theism, or visa versa, on the basis of which epistemic norms most respect the insulation requirement, and thus more reliably confer objectivity on factual claims?  Which norms, those of theists or those of naturalists, best guard against projecting our human hopes and fears onto the world when constructing a worldview?

To raise this question is almost immediately to answer it, since the raison d’etre of science, in collaboration with philosophy, is to achieve as far as possible a 3rd-person, intersubjective and therefore maximally objective understanding of the world. In principle and almost always in practice, any honest scientist will (eventually, sometimes after considerable controversy about methods and data) reach more or less the same conclusions in a well-researched domain of inquiry as another scientist, whatever their original differences. Why?  Because over the last 350 years experimental methods and criteria of explanatory adequacy have been selected precisely for their bias-reducing properties, for their capacity to filter out subjective hopes and expectations when picturing reality. The predictive, explanatory successes of science, not to mention its practical applications, compel consensus about matters of fact no matter what we wish were the case. Science abstracts away from the motivated human perspective to give us, as far as possible, what philosopher Thomas Nagel called “the view from nowhere.”[7] That this view is often not particularly to our liking (no evidence for God, heaven, the soul or immortality) suggests that science isn’t projecting our wishes onto the world.

Traditional theism, on the other hand, seems to specialize in defending the prospect that our fondest dreams – for life everlasting, reunion with loved ones, a purposeful cosmos headed up by a benevolent intelligence – might be fulfilled. Far from seeking to limit the distorting effects of human hopes in picturing the world as it objectively is, religion panders to them. God and his powers, exercised on our behalf, are exactly what we needy, fragile, all too mortal creatures would most want to exist. Theistic religions make their living by offering existential reassurance, and much modern theology, however sophisticated and cognizant of current science and philosophy, is essentially an apologetics on behalf of a desired conclusion: that God exists. Likewise, the standard justifications for belief in God – the authority of sacred texts and religious officiants, personal revelation and intuition, the various rationalistic armchair proofs – are all quite the opposite of science’s open-ended, corrigible empiricism. They are modes not of investigation, but of confirmation. God is the vigorously defended projection of our deepest hopes onto the world.

Nature is enough

If you’re interested in objectivity, the choice between the relatively rigorous epistemic demands of naturalism and the more relaxed demands of theism is obvious. If you want a picture of the world more or less as it is, insulated as much as possible from the distorting effects of your own all-too-human psychology, you will stick with science. Not that science is infallible, but it fully recognizes and tries to reduce the influence of wishful thinking when representing reality. Such caution in service to objectivity helps to keep explanations transparent, since only well-evidenced entities and processes get to play a role. By contrast, theism and theology, despite their claims to objectivity, manifestly fail to respect the most basic cognitive requirement involved in such claims: that we should leave our hopes behind when investigating the world. This failure is reflected in the obscurity and spurious completeness of theistic explanations: they involve unexplained explainers and questionable evidence which function to protect a cherished image of what the world must be like. God, a mystery, must move in mysterious ways to keep our dreams of a benign, divinely ordered universe alive.

The naturalist’s off-the-cuff challenge to the traditional theist might be that God is simply too good to be true and too obscure to explain. As we see what’s involved in claims to objectivity, and what’s required for transparent explanations, this challenge holds up well. Wanting to know what’s real, naturalists acknowledge that in some respects reality may not be to our liking; we therefore guard against projecting our wishes onto nature. Wanting clear explanations, we seek reliable, intersubjective data about the world; and we construct a plausible story – a theory – to explain the data as transparently as possible. As far as we can tell there is no role for God, humankind’s fondest hope, in such a story. But the absence of God and the supernatural simply highlights the presence of nature. For the naturalist, nature is all there is, and therefore it’s enough.



Beilby, J.K, ed.,  Naturalism Defeated?: Essays on Plantinga’s Evolutionary Argument Against Naturalism, New York: Cornell University Press, 2002.

 De Caro, M., Macarthur, D., eds, Naturalism in Question, Cambridge: Harvard University Press, 2004

Hahn, L., and Schilpp, P., eds., The Philosophy of W.V. Quine, La Salle, Ill.: Open Court, 1986.

Haught, J.F., Is Nature Enough?, Cambridge: Cambridge University Press, 2006.

Hunter, C. G., Science’s Blind Spot: The Unseen Religion of Scientific Naturalism, Grand Rapids: Brazos Press, 2007.

Nagel, T. The View From Nowhere, Oxford: Oxford University Press, 1986.


[1] As W. V. Quine put it: “Naturalized epistemology does not jettison the normative and settle for the indiscriminate description of ongoing processes. For me normative epistemology is a branch of engineering. It is the technology of truth-seeking, or, in more cautious epistemic terms, prediction.” From Quine’s "Reply to White," in The Philosophy of W.V. Quine, L. Hahn, & P. Schilpp, editors, pp. 664-5.

[2] See for instance Craig Hunter, Science’s Blind Spot: The Unseen Religion of Scientific Naturalism and, accessed August 4, 2008.

 [3] The diversity of views on naturalism, both among naturalists and anti-naturalists, is well documented in Naturalism in Question, M. De Caro & D. Macarthur, editors. Even the claim that science necessarily shows reality to be a single, natural realm is contested; see the last section of my review of J.P. Moreland’s The Recalcitrant Imago Dei: Human Persons and the Failure of Naturalism, in particular footnote 13.

 [4] For a non-exhaustive list of such criteria, see, accessed August 4, 2008.

 [5] See John F. Haught, Is Nature Enough?, reviewed at, accessed August 4, 2008.

 [6] See Plantinga’s so-called “evolutionary argument against naturalism” discussed in Naturalism Defeated?: Essays on Plantinga’s Evolutionary Argument Against Naturalism., J. Beilby, editor, and discussed at, accessed August 4, 2008.

 [7] Thomas Nagel: The View From Nowhere.


[back to Epistemology, Atheism]

Home        Center for Naturalism       Applied Naturalism        Spirituality       Philosophy