Religious Experience as Evidence for God

Experience as Evidence

Abstract

The affirmative case holds that experiences as of God are, epistemically, on a par with sense-experiences as of physical objects: both present their objects as simply there, and both are innocent until proven guilty. James supplies the phenomenology — mystical states are marked, noetic, and authoritative for their subjects. Swinburne supplies the epistemic principle — the Principle of Credulity — that turns such seemings into evidence. Alston supplies the framework — Christian mystical perception is a "doxastic practice" as rationally engaged in as sense-perception. Together they mount not a proof but a defeasible perceptual argument: the burden lies on the critic to show why this class of experience, uniquely, must be discounted.

Formal statement

1. Religious experiences "are in all relevant respects like sensory experiences; sensory experiences are excellent grounds for beliefs about the physical world; so religious experiences are excellent grounds for religious beliefs" (Webb 2022, SEP §3).
2. By the Principle of Credulity, "one is justified in believing that what seems to one to be present actually is present, unless some appropriate defeater is operative" (Webb 2022, SEP §3).
3. In mystical experience it seems, with the force of direct perception, that a higher reality is present — the states carrying "a curious sense of authority" and being "as direct perceptions of fact for those who have them as any sensations ever were for us" (James 1902, Lectures XVI–XVII).
4. It is rational to trust an established belief-forming practice unless it is shown unreliable; the Christian mystical-perceptual practice is so established (Alston).
5. Therefore the subject of such experience is prima facie justified in the corresponding belief in God.

Key evidence / textual basis

James's marks. James defines mystical states by four marks: ineffability ("no adequate report of its contents can be given in words"); noetic quality — they "seem to those who experience them to be also states of knowledge... illuminations, revelations... [carrying] a curious sense of authority for after-time"; transiency; and passivity, in which "the mystic feels as if his own will were in abeyance, and indeed sometimes as if he were grasped and held by a superior power" (James 1902, Lectures XVI–XVII). The noetic mark is the load-bearing one: it is precisely because these states present themselves as cognitive — as disclosures of fact rather than mere feeling — that the question of their evidential worth arises at all.

Their authority. James's verdict on that authority is deliberately three-part: "(1) Mystical states, when well developed, usually are, and have the right to be, absolutely authoritative over the individuals to whom they come. (2) No authority emanates from them which should make it a duty for those who stand outside of them to accept their revelations uncritically. (3) They break down the authority of the non-mystical or rationalistic consciousness... They open out the possibility of other orders of truth" (James 1902, Lectures XVI–XVII). The affirmative case builds on (1) and (3): the experience is self-authenticating for its subject, and the mere fact that it lies outside the "rationalistic consciousness" is no strike against it, since "our own more 'rational' beliefs are based on evidence exactly similar in nature to that which mystics quote for theirs" (ibid.).

Swinburne's Principle of Credulity. Swinburne generalizes the point into an epistemic principle. On the Principle of Credulity, as SEP summarizes it, "one is justified in believing that what seems to one to be present actually is present, unless some appropriate defeater is operative" (Webb 2022, SEP §3). Swinburne then canvasses the circumstances that would be defeaters in the ordinary sensory case and argues that "those defeaters do not obtain, or not always, in the case of religious experience" (ibid.). The dialectical effect is to shift the burden: "anyone who accepts the principle has excellent reason to accept the deliverances of religious experience, unless he or she believes that defeaters always, or almost always, obtain" (ibid.). The critic must therefore show not that religious experience might mislead, but that it must — that some defeater is universally operative.

Alston's doxastic-practice defense. Alston reframes the debate away from arguments-by-analogy toward the rationality of practices. A doxastic practice is a "socially established" way "of forming and epistemically evaluating beliefs" that is "beholden to an 'over-rider system'" of background beliefs against which its outputs are checked (Jones 2022, SEP 'Mysticism' §9.1). His crucial move is that the justification of every such practice — sense-perception included — "is 'epistemically circular'... its reliability cannot be established in any way independent of the practice itself" (ibid.). Since "we cannot avoid engaging in doxastic practices," it is "a matter of practical rationality to engage in the doxastic practices we do engage in providing there is no good reason to think they are unreliable" (ibid.). The Christian mystical practice, with an over-rider system "consisting of scriptures, Christian dogma, and guidelines resulting from the past history of the... practice," meets that condition; so "it is rational for a person engaged in such a practice to take its belief outputs as true unless the practice is shown to be unreliable" (ibid.). Its scriptural echo is the community's gift of "discernment" (1 Cor 12:1-11 (bib)) — the over-rider system in operation.

Plantinga's parallel. Alvin Plantinga supplies the Reformed-epistemology variant. Beliefs formed by an experience of God ("M-beliefs") are properly basic: "If beliefs formed by sense-experience can be properly basic, then beliefs formed by this faculty" — the Calvinian sensus divinitatis — "cannot, in any principled way, be denied that same status." His warrant theory adds that "if the beliefs are true, then they are warranted," so "one cannot attack claims of religious experience without first addressing the question as to whether the religious claims are true" (Webb 2022, SEP §3). Note the divergence from evidentialism here: where Swinburne treats the experience as public evidence to be weighed on a probability scale, Plantinga treats theistic belief as basic and not grounded in argument at all. SEP's 'Natural Theology' entry registers the resulting boundary dispute — the Continental Calvinist tradition harbors "a revulsion against arguments in favor of theism," so many locate the sensus divinitatis outside natural theology altogether (SEP 'Natural Theology' §3.3).

Leading proponents

• William James — The Varieties of Religious Experience (1902; in corpus) — the phenomenology of mystical states and their authority for the subject.
• Richard Swinburne — The Existence of God (1979; via SEP) — the Principle of Credulity and its companion Principle of Testimony.
• William P. Alston — Perceiving God (1991; via SEP) — the doxastic-practice argument.
• Alvin Plantinga — Warranted Christian Belief (2000; via SEP) — properly basic M-beliefs and the sensus divinitatis.

Strongest counter-arguments

The heaviest blow to the analogy is the checkability disanalogy, and it is pressed hardest not by James but by analytic critics of Alston (see the Naturalist / Skeptical Reply below): with sense-perception "we can cross-check by employing inductive methods... we can 'triangulate' an event... and we can discover causal mechanisms," none of which "are available for checking on experiences of God" (Jones 2022, SEP 'Mysticism' §9.3.1). A second blow is Alston's own admitted exportability: he "admits that such an argument might be equally available to other religious practices" (Webb 2022, SEP §3) — the Muslim and Vedantin may "sit tight" too (see the Pluralist Complication). A third is James's own two-level verdict, which caps how far the evidence reaches: the experience is authoritative "for me, but not for you" (Webb 2022, SEP §3).

Responses

On checkability, Alston replies that the demand for public verification is not epistemically neutral: "it is not enough... to show that religious experiences do not typically allow for independent public verification, unless one wants to give up on other perfectly respectable practices, like rational intuition, that also lack that feature" (Webb 2022, SEP §3). On exportability, the reply is that it "is not refutation": "from within a tradition, a person acquires epistemic resources not available to those outside... just as travelling to the heart of a jungle allows one to see things that those who have not made the journey can't see"; so "even if people in other traditions can make the same argument, it is still reasonable to say that some are right and the others are wrong" (ibid.). On James's transmission problem, the affirmative case grants that testimony to religious experience may carry less force than the experience itself, while denying that this shows the first-person case defective: whether "the differences are great enough to disqualify religious testimony always and everywhere" is precisely what "is not clear" (ibid.). And James insists the states "must be sifted and tested, and run the gauntlet of confrontation with the total context of experience, just like what comes from the outer world of sense" (James 1902, Lectures XVI–XVII) — an empirical program, not a fideist retreat.

Assessment

Assessment: Strong — the Principle of Credulity is widely conceded to shift the burden onto the critic, and Alston's practical-rationality argument is generally regarded as showing that religious experience cannot be dismissed merely for lacking public checkability without collateral damage to intuition and memory. What keeps it from being decisive is that both defenses turn on "some degree of similarity with sense-experience" (Webb 2022, SEP §3), leaving the checkability and diversity defeaters genuinely live.

The Naturalist / Skeptical Reply

Abstract

The naturalist grants that religious experiences are real experiences and denies only that they are perceptions — presentations of an external object. Two lines converge: a structural disanalogy (religious experience lacks the cross-checking that makes sense-perception a reliable route to truth) and a causal debunking (psychological and neurological mechanisms suffice to explain the experiences without positing their objects). If either succeeds as a universal defeater, premise 3 of the master argument fails and the Principle of Credulity is silenced for this domain.

Formal statement

1. An experience-type yields perceptual knowledge only if it admits cross-checking — triangulation of the object, correlation with independent effects, identification of causal mechanisms (Fales, Gale, Martin).
2. Experiences of God admit none of these: God "does not exist in space and time," so there could "never be evidence that a person had experienced God" as opposed to having "an 'O-ish-impression'" (Jones 2022, SEP 'Mysticism' §9.3.2).
3. Independently, sufficient naturalistic causes (psychodynamic, socio-economic, neurological) are available for the experiences.
4. Where a sufficient natural cause is in view, "we have no grounds for positing anything beyond that naturalistic cause" (Webb 2022, SEP §3).
5. Therefore religious experience is not evidence for its purported object.

Key evidence / textual basis

The checkability disanalogy. Evan Fales "argues that cross-checkability is an integral part of any successful perceptual epistemic practice. Therefore, the perceptual epistemic practice in which mystical experiences of God are embedded is severely defective." Richard Gale adds that such experiences lack "agreement between perceivers" and any way "of checking whether the perceiver was in the 'right' position and psychological and physiological state for a veridical experience"; C.B. Martin concludes that claims to have experienced God are "'very close' to subjective claims like 'I seem to see a piece of paper' rather than to objective claims like 'I see a piece of paper'" (Jones 2022, SEP 'Mysticism' §9.3.1). William Rowe sharpens the asymmetry: "God may choose to reveal himself to one person and not to another," so the failure of others to perceive God "does not impugn the validity of the experience" — but then "we have no way of determining when an experience of God is delusory. If so, neither can we credit an experience as authentic" (ibid.). The very feature that protects the believer from disconfirmation (God's freedom to withhold) also strips the experience of the error-detecting structure that makes perception trustworthy.

The debunking mechanisms. Freud held that "religious experiences can be adequately explained by psychological mechanisms having their root in early childhood experience and psychodynamic tensions"; Marx, that religious belief traces to "materialistic economic forces"; both concluding "there is no need to suppose... that the beliefs are true" (Webb 2022, SEP §3). The contemporary version is neurological: brain events "during meditative states and other religious experiences are very similar to events that happen during certain kinds of seizures, or with certain kinds of mental disorders, and can also be induced with drugs," so "there is nothing more to religious experiences than what happens in seizures, mental disorders, or drug experiences" (ibid.). Anthropologically, Guthrie "argues that religion has its origin in our tendency to anthropomorphize phenomena... seeing agency where there is none" (ibid.).

James as insider witness for the skeptic. The most damaging testimony against the common-core version of the argument is James's own retraction. Having asserted "an eternal unanimity" among mystics, he immediately confesses that "if we take the larger mass as seriously as religious mysticism has historically taken itself, we find that the supposed unanimity largely disappears": mysticism "is dualistic in Sankhya, and monistic in Vedanta philosophy... the great Spanish mystics are anything but pantheists," and "the mystical feeling of enlargement, union, and emancipation has no specific intellectual content whatever of its own" (James 1902, Lectures XVI–XVII). Worse, there is "diabolical mysticism, a sort of religious mysticism turned upside down," springing from "the same mental level" as the classic sort: "seraph and snake abide there side by side. To come from thence is no infallible credential" (ibid.). If the same subliminal source yields saints and psychotics indiscriminately, its productions carry no presumption of veridicality on their own.

Leading proponents

• Evan Fales and Richard Gale — the cross-checkability requirement (via SEP 'Mysticism' §9.3.1).
• C.B. Martin and William Rowe — the subjective-claim and no-disconfirmation objections (via SEP 'Mysticism' §9.3.1).
• Sigmund Freud, Karl Marx, Stewart Guthrie — the debunking mechanisms (via SEP 'Religious Experience' §3).
• William James — the insider witness whose retraction of "eternal unanimity" arms the common-core skeptic (Lectures XVI–XVII).

Strongest counter-arguments

The theist answers the debunking arm with a symmetry charge. Freud's and Marx's theories "are put forward as hypotheses, not as established facts. The proponent assumes that the experiences are not veridical, then casts around for an explanation" (Webb 2022, SEP §3) — Freud's account "has few adherents" and Marx's "has all but been abandoned" (ibid.). The neurological arm faces Ellwood's parity objection: "every experience, whatever its source, is accompanied by a corresponding neurological state. To argue that the experience is illusory because there is a corresponding brain state is fallacious. The same reasoning would lead us to conclude that sensory experiences are illusory" (ibid.). A correlated brain state no more debunks the experience of God than the visual cortex debunks seeing a tree.

On checkability, the theist's reply is twofold. First, the demand may prove too much: rational intuition and introspection also resist triangulation yet remain "perfectly respectable practices" (Webb 2022, SEP §3). Second, checkability is not wholly absent — proponents note "mystical procedures for getting into position for a mystical experience of God" that "others can take up... to try to check on the subject's claims" (Jones 2022, SEP 'Mysticism' §9.2), a communal analogue to learning to see.

Responses

The skeptic holds the line on both fronts. Ellwood's symmetry point is granted but redirected: the naturalist's best case is not the crude "there is a brain state, therefore illusion" but the disanalogy that sense-perception, unlike mystical perception, licenses further independent checks that a brain-state correlation cannot supply — the neurological data merely remove one reason to treat mystical states as exceptions. And on checkability, the rejoinder is that a practice whose own practitioners cannot use it to catch and correct their errors is not vindicated by the errors it happens to avoid: the "mystical procedures" reproduce the disposition to have the experience, not a procedure for detecting when the experience is delusory — which is the very feature Rowe showed to be missing when "God may choose to reveal himself to one person and not to another" (Jones 2022, SEP 'Mysticism' §9.3.1). The rational-intuition parallel is conceded to blunt the strongest form of the objection (public verification is not a universal requirement) without rescuing the analogy to sense-perception on which Swinburne's and Alston's positive arguments depend.

Assessment

Assessment: Live — the checkability disanalogy is the single most discussed objection in the literature and is not defused merely by the intuition/memory parallel, since those practices, unlike mystical perception, still admit internal error-correction. The debunking arm is weaker (Ellwood's symmetry point and the hypothesis-not-fact charge blunt it), but the two arms together keep the naturalist reply firmly in play.

The Pluralist Complication

Abstract

Distinct from the naturalist, the pluralist need not deny that religious experiences are veridical. The worry is that their contents conflict: subjects report "nothingness as the ultimate reality," "a vast impersonal consciousness," and "an infinitely perfect, personal creator" (Webb 2022, SEP §3) — objects that cannot all be as reported. If contradictory experiences carry equal prima facie warrant, they threaten either to cancel one another (skeptical outcome) or to force the conclusion that all apprehend one Reality under culturally shaped guises (Hick's pluralist outcome). Either way the experiences no longer support the specific God any tradition worships. This article keeps the complication distinct from the standalone debate in Religious Pluralism; here the question is narrowly what diversity does to the evidential force of experience.

Formal statement

1. The Principle of Credulity, applied impartially, confers prima facie warrant on every sincere religious experience, whatever its tradition.
2. The objects reported are mutually incompatible — personal creator, impersonal Brahman, śūnyatā (Webb 2022, SEP §3; Jones 2022, SEP 'Mysticism' §9.6).
3. "If the different practices produce experiences the contents of which are inconsistent with one another, one of the practices must be unreliable" (Webb 2022, SEP §3).
4. No neutral ground privileges one tradition's experiences over another's.
5. Therefore either the experiences mutually defeat one another, or they must be reinterpreted as diverse responses to a single trans-categorial Reality (Hick) — and in neither case do they evidence the particular God of any one creed.

Key evidence / textual basis

The conflict. The diversity of reported objects is the datum. In the history of religions "we find innumerable gods, with different characteristics," plus "experiences of non-personal ultimate realities, such as the Brahman of Advaita Vedanta. Brahman cannot be an ultimate reality if God is... since it is devoid of all personal features"; and "different theistic faiths claim experience of the one and only God, ostensibly justifying beliefs that are in contradiction with one another. If theistic mystical experiences lead to such contradictory results, they cannot provide evidence in favor of their validity as experiences of God" (Jones 2022, SEP 'Mysticism' §9.6). James had already located the fault line: beneath "the warring gods and formulas of the various religions" lies only "a certain uniform deliverance" — an "uneasiness" and "its solution" (James 1902, Lecture XX) — while the tradition-specific revelations, because they "corroborate incompatible theological doctrines... neutralize one another and leave no fixed result" (ibid.).

Hick's resolution. John Hick converts the problem into a hypothesis. Rather than let the experiences cancel, he proposes that "the great world faiths embody different perceptions and conceptions of one reality that Hick christens 'the Real.' The Real itself is indescribable and is never experienced directly but has 'masks' or 'faces' that depend on how a particular culture or religion thinks of the Real" (Jones 2022, SEP 'Mysticism' §9.6). The machinery is Kantian: "the object of these experiences, in itself, is one and the same reality, but it is experienced phenomenally by different people differently," so "one and the same object can be experienced in ways that are completely incompatible with one another" (Webb 2022, SEP §3). The evidential upshot for our question: experience may warrant belief in the Real, but not in Yahweh, the Trinity, or Allah as opposed to Brahman.

The perennialist/constructivist stakes. Whether diversity is fatal depends on a prior dispute the mysticism literature calls essentialism versus constructivism. Essentialists claim "a 'common core' to all mystical experiences independent of culture" (Jones 2022, SEP 'Mysticism' §5); perennialists posit "universal mystical doctrines... a common core of esoteric doctrines that are expressed differently in different religions" (SEP §6); constructivists counter that "context penetrates mystical experiences themselves," so "there is no way to separate mystical experiences completely from their interpretations since our conceptual apparatus shapes our very experience" (SEP §7). If hard constructivism is right, there is no cross-tradition core to appeal to and the conflict is irreducible; if essentialism is right, the diverse reports may share enough structure to count as experiences of one reality — which is exactly the ground Hick and James's "MORE" occupy.

Leading proponents

• John Hick — An Interpretation of Religion (1989; via SEP) — the pluralistic hypothesis and the Kantian "Real."
• Steven Katz — hard constructivism, on which diversity is irreducible (via SEP 'Mysticism' §7).
• William James — the "neutralize one another" verdict on tradition-specific revelation (Lecture XX).

Strongest counter-arguments

Swinburne answers with an ascent to generality: "conflicting descriptions of the objects of religious experience pose a challenge only to detailed doctrinal claims, not to general claims of having experienced a supernal being" (Jones 2022, SEP 'Mysticism' §9.6). On this view diversity trims the content the experiences can support down to a generic supernaturalism, but does not cancel it — and generic supernaturalism is already a substantial defeat for naturalism. A second reply notes that "if the reports are at all similar, then it may be reasonable to conclude that there is some truth to the testimony, at least in broad outline" (Webb 2022, SEP §3); James's "uniform deliverance" is precisely such a broad-outline residue. A third is the particularist reply carried over from the affirmative view: from within a tradition one has "epistemic resources not available to those outside," so it remains "reasonable to say that some are right and the others are wrong" even when rivals run the same argument (ibid.).

Against Hick specifically, the theist objects that the Kantian solution "is only as plausible as the Kantian framework itself is" and, more tellingly, that it "leaves the problem untouched": if the practices yield inconsistent contents, "one of the practices must be unreliable" — relabeling the incompatibility as phenomenal appearance of one noumenal Real does not remove it (Webb 2022, SEP §3).

Responses

The pluralist grants the ascent-to-generality point but presses its cost: what survives is only a "supernal being" or a "MORE... operative in the universe" (James 1902, Lecture XX), which "grounds none of the doctrines any tradition actually preaches." A generic supernaturalism is compatible with deism, pantheism, and Hick's Real alike — so the argument from experience, purified of its conflicting specifics, delivers far less than the creeds need even if it delivers something. Against the "some are right" reply, the pluralist joins the peer-disagreement literature: once it is admitted that adherents of every tradition can with equal right claim tradition-internal resources, the claim confers no discriminating support on any one of them. And James's own two-level verdict is conceded on both sides: the experience remains "authoritative over the individuals to whom they come" (James 1902, Lectures XVI–XVII) even as the doctrines built on it cancel — which is either a stable resting point or an unstable halfway house, depending on whether one thinks first-person authority can float free of public warrant.

Assessment

Assessment: Live — diversity is, with checkability, one of the "two most important defeaters" (Webb 2022, SEP §3), and it bites hardest against the specific God of any creed while leaving generic theism comparatively intact. Whether Swinburne's ascent-to-generality is a successful containment or a quiet surrender of the doctrinal stakes is exactly the unresolved question; its outcome turns on the still-open essentialism/constructivism dispute.