2. Dissolving the SD impasse

In the disagreement we want to consider, the atheist views naturalism as rational, while the theist regards theism as rational. One way to determine which perspective is rationally superior is by examining the explanatory power of these worldviews. Suppose naturalism fails to account for the reliability of our cognitive faculties, whereas theism does not. In that case, everything else being equal, we should prefer theism over naturalism. Similarly, if there are transformative experiences, the skeptic’s naturalistic worldview can either, in principle, explain these experiences or it cannot. If these experiences are, in principle, inexplicable, those experiences become anomalies within that worldview. Indeed, some experiences require adjustments in worldview to be understood and make sense. The experience or putative evidence in question is not conventional; it is so powerful that once you are exposed to it, it transforms you and your worldview. The theist might emphasize this extraordinary characteristic of the experience as support for his worldview. To counter this strategy, the atheist would be challenged to explain this experience within his own framework. If naturalism fails to explain this powerful experience and the atheist still chooses not to be exposed to it, then he is essentially suppressing putative evidence.

Let us distinguish two scenarios. In the first, we are exposed to an experience that triggers a change in our cognitive state through brute causation. In the second, we encounter putative evidence that we evaluate in a reason-guided way; any transformation results from our uptake of its content as reasons. In the first scenario, the change occurs solely via a brute-causal trigger, not through reason-guided uptakeFootnote ¹ . Thus, the question of whether we wish to be exposed to that experience merits examination. For instance, consider The Doors of Perception by Aldous Huxley. In this book, on the one hand, Huxley argues that by taking hallucinogens like mescaline and lysergic acid diethylamide (LSD), one can gain much deeper insights into reality and experience oneness with the world, as the brain loosens its ordinary filters on perception. (Reference Huxley2009, p. 26). Huxley claims that some religious and drug-induced experiences enable our brains to transcend their limitations in perception, leading to a much richer understanding of the world. If the experience provides genuine rather than merely putative evidence that enhances our grasp of reality, then we have good reasons to be open to it. On the other hand, some argue that experimenting with these drugs can have lasting adverse effects; one may develop psychosis or experience disturbing flashbacks. Whether you choose to take the risk of using LSD, it is important to recognize that it can transform your cognitive abilities either positively or negatively. We can escalate the risk by considering a slightly different scenario in which the LSD is not obtainable from a reliable chemist, raising the possibility that it could be laced, thus amplifying the potential damage to your brain and other organs in fatal ways. An even more extreme case is where there is a chance that you take a “mad pill”. The mad pill is such that once one swallows it, one becomes mad and will be fundamentally transformed. In such cases, which are brute-causal triggers, it seems that our choice regarding whether we wish to be exposed to the experience is not so much about open-mindedness; instead, what is at stake are moral virtues and vices such as courage, cowardice, and recklessness.

In contrast, consider a case where the putative evidence is reason-guided rather than brute-causal, and we analyze it rationally. In such situations, we are responsive to reasons, and open-mindedness requires exposing ourselves to evidence we can engage with rationally. In such cases, we engage in arguments and use our rational capacities to determine which position aligns with our standards of rationality.

To explore whether open-mindedness requires exposure to the putative evidence of the SD despite the risk of corruption, we must first recognize the deep epistemic impasse at the core of Paul’s challenge. The theist claims the SD offers genuine evidence of God’s existence, while the skeptic suspects it is a form of mental corruption. The LSD case vividly demonstrates this tension: proponents like Huxley argue that such experiences go beyond ordinary cognition to reveal a deeper reality, while skeptics view the same phenomenology as neural malfunction or damage. Therefore, what is contested is not the phenomenology itself but its epistemic status; that is, whether it should be considered evidence or just putative evidence. Paul’s concern is that once standards of evaluation are themselves in dispute, there is no neutral ground; exposure to the SD would not be mere argument-reception but a causal process that the pre-transformation self would regard as epistemically corrupting.

The skeptic’s rational response is not to attempt to settle the matter by undergoing SD-phenomenology, an approach that would assume their current standards are correct, but to shift the analysis to a higher level. The strategy is to assess the status of the SD which has two components: (i) deny that SD deliverances are properly basic (or, more strongly, that there are properly basic beliefs of the relevant kind), and (ii) evaluate, on metaphysical and explanatory grounds, whether a naturalistic account can match or outperform a theistic account of the relevant phenomenology. On the first component: one may challenge the coherence, intelligibility, or meaningfulness of the SD as a faculty that produces properly basic beliefs (i.e., non-inferential, foundational beliefs not grounded in evidence or arguments). If successful, such challenges demystify the SD within a naturalistic framework, reducing it to a phenomenon that can be scrutinized and potentially rejected using pre-existing rational standards without undergoing the transformative experience itself. Consider arguments that undermine the claim that beliefs from the SD are properly basic. Koons (Reference Koons2011) contends that no beliefs are truly properly basic because all are theory-laden, depending on previously adopted background theories (pp. 842–846). Applied to the SD, any “divine sensing” is not an automatic, foundational process but is filtered through prior theoretical commitments (theistic or naturalistic). Thus, the disagreement over the SD shifts from a mysterious, brute-causal transformation to a rational debate about which background theory better explains the reported experiences. The skeptic can then, ex ante, consider the following: does naturalism (e.g., via evolutionary psychology or neurology) account for such experiences as adaptive illusions or byproducts, or does theism provide superior explanatory power?

If Koons is right, the SD loses its status as an independent faculty, becoming evaluable through standard epistemic tools like coherence with other beliefs and explanatory adequacy. This helps dissolve the impasse Paul describes, as the skeptic can rationally decide against exposure if naturalistic theories prevail, without fearing corruption because the evaluation proceeds by reason-guided causal assessment rather than brute-causal triggering.

Even granting a properly functioning SD, its deliverances do not float free of the rest of one’s beliefs. They are received and interpreted within a holistic web of beliefs, so errors outside the SD can propagate inward and misdirect how its deliverances are taken (Dormandy Reference Dormandy2020, p. 403). Hence, believers have a pro tanto epistemic reason to invite critical engagement with dissenters in order to test, calibrate, and refine both SD-related judgments and the surrounding network. This rationale supports the skeptic’s ex ante stance. Assessing SD-type claims by public abductive standards before exposure is not closed-mindedness but a safeguard against distortion, even on terms the theist can accept.

Another complementary line of critique questions the alleged basicality of theistic belief itself. Goetz (Reference Goetz1983) argues that “anyone who believes in God must acknowledge his contingency and that his knowledge of his contingent nature enables him to infer the existence of a necessary being or beings.” (p. 484). Grigg adds that belief in God lacks the automaticity and phenomenological immediacy typical of paradigmatically basic beliefs such as perception and memory, and that it often functions as psychological comfort shaped by cultural factors (1990, p. 390). Together, these points suggest that theistic belief is not a basic deliverance at all but typically arises from inference and from psychological or cultural influences, so it is something we can and should evaluate before exposure. If a belief in God is non-automatic and comfort-driven, the skeptic can infer that the transformation is not epistemically enhancing but akin to a placebo effect. Again, this enables a pre-exposure rational assessment: The epistemic value of any such belief is determined by its fit with the skeptic’s broader web of beliefs.

These theses collectively reframe the SD from an inscrutable, potentially corrosive experience into one amenable to rational analysis. Unlike brute-causal triggering, where transformation occurs regardless of reasons, the SD, once stripped of basicality, depends on debatable theoretical foundations. This distinction is key when comparing it to the LSD example. Huxley (Reference Huxley2009) claims LSD transcends perceptual limits for deeper insights (p. 26), but a skeptic might view it as mere neural damage creating illusory enhancements. Because LSD’s effects are modeled here as brute-causal triggers, there is no ex ante route to adjudicate their epistemic status by reason-guided assessment; ingestion is precisely what would bypass such assessment. In contrast, if the SD is not properly basic (per Koons, Goetz, and Grigg), it becomes evaluable through arguments about theory-dependence and disanalogies between the SD and other properly basic beliefs – tools that are accessible to the skeptic and the theist.

If no SD-deliverance is properly basic, then uptake of SD-content is not a brute-causal trigger but a reasons-responsive process: its standing depends on integration with, and comparative support from, the subject’s background theory and total evidence. This relocation matters for Paul’s corruption worry. Brute-causal triggers threaten rational agency because they bypass the subject’s pre-exposure standards. By contrast, reasons-responsive candidates are appraisable ex ante by those very standards (e.g., coherence and explanatory power).

This leads to the second part of the strategy, i.e., evaluating the explanatory power of the skeptic’s naturalistic worldview. The very existence of powerful, transformative experiences like those attributed to the SD presents a metaphysical challenge. Is the SD a persistent anomaly that threatens naturalism’s explanatory completeness, or can it be convincingly explained away? If naturalism offers robust explanations (e.g., based on cognitive science, neuroscience, and evolutionary psychology), the skeptic has an internal reason to classify the SD as a potentially misleading natural phenomenon and to dismiss it. If, however, such experiences remain deeply inexplicable within naturalism, the principle of seeking the best explanation itself provides a rational motive or even a duty to consider investigating them, despite the transformative risk. In this case, avoiding the experience would amount to shielding one’s worldview from a significant putative counter-evidence.

The choice, therefore, is not between blind faith and dogmatic refusal, but between two rational, fallible policies under uncertainty. The skeptic can make a calculated risk assessment based on the best available evidence from their current perspective:

If these strategies show that SD-type reports lack the properties theists claim, the phenomena can be reframed and studied naturalistically. Even then, the experiences may remain life-altering, but unlike LSD, the denial of basicality lets the skeptic weigh such risks ex ante. Conversely, if naturalism fails to explain the phenomena adequately, that failure itself is a reason to revise naturalism – again, on the basis of pre-exposure assessment. Thus, the skeptic is not epistemically helpless but possesses a rational framework that addresses Paul’s dilemma without succumbing to corruption or simple closed-mindedness.

Based on these considerations, a decision rule can be devised for corruption-averse skeptics. Avoid entering enabling conditions for SD-type experiences if (i) the best available naturalistic explanations render the phenomenology epistemically non-enhancing or distorting by the skeptic’s present standards, and (ii) the expected epistemic value of exposure does not outweigh a credible risk of standards-corruption; otherwise, exposure is rationally permissible.