1 What kind of claim is a religious claim?

Before we can ask whether religious beliefs are justified, we need to know what kind of claims they are — or whether claim is even the right category. “God is love” does not look like “water boils at 100°C.” But does this mean it is a lesser claim — less serious, less true — or a different kind of claim altogether? And there is a prior question worth flagging: many religious traditions hold that doctrinal propositions are secondary to practice, community, and transformation — that “religion” is more a matter of what one does and becomes than of what one asserts.¹ The lesson that follows works largely with the propositional framing because that is what most of the philosophical literature has worked with; readers should keep the prior question alive.

1.1 Cognitive vs. Non-Cognitive Accounts

The most basic division in the philosophy of religious language is between cognitive and non-cognitive accounts of religious claims.

A cognitive claim aims to be true or false. “There is a God” is cognitive: it purports to describe how things are. If cognitive, the claim is subject to the normal demands of evidence and justification. If true, it is true in the same way that “there is a planet with rings around Saturn” is true — as a fact about reality.

A non-cognitive claim does not describe but expresses, prescribes, or performs. “God is love” might, on a non-cognitive reading, express a commitment, articulate a value, or perform an act of trust — without being literally true or false in the way factual claims are.

The appeal of non-cognitivism: it seems to protect religious claims from empirical refutation. If “God exists” is not really a factual claim, then no scientific discovery can threaten it. The corresponding question: what kind of truth a non-cognitive utterance is taking on, if it is no longer reporting a fact — and whether the answer (expressive, performative, ritual) preserves or empties the religious claim.

1.2 Flew’s Parable and the Falsification Challenge

The 1955 symposium “Theology and Falsification,” published in Antony Flew and Alasdair MacIntyre’s New Essays in Philosophical Theology, pressed the demarcation question into religious language. Flew’s parable, originally presented at the Oxford Socratic Club in 1950: two explorers come upon a clearing in the jungle.⁶ One insists that an invisible gardener tends the clearing; every test designed to catch the gardener — bloodhounds, electrified wire, overnight watch — produces no evidence, but the believer keeps qualifying the claim: the gardener is also silent, odourless, intangible. “What is left of your original assertion?” Flew asks. “Just how does what you call an invisible, intangible, eternally elusive gardener differ from an imaginary gardener or even from no gardener at all?”⁷

Flew’s challenge does not assume religious claims are false; it asks what they assert at all. A claim that excludes nothing predicts nothing — and a claim qualified into compatibility with every possible observation has, on this reading, “died the death of a thousand qualifications.” Hare and Mitchell offered the two main lines of reply. Hare denied that religious belief is the kind of thing that should be evaluated by falsifiability at all, introducing his blik account (treated below). Basil Mitchell offered a counter-parable about a partisan who trusts a stranger despite mounting evidence against him — granting Flew’s evidential structure but arguing that a religious commitment can be evidentially responsive without being cleanly falsifiable.⁸ The symposium gave twentieth-century philosophy of religion its central problem: not whether religious claims are true, but what their grammar is — what kind of claim is being made, and whether the standards of empirical demarcation apply to it at all.

1.3 Hare’s Bliks

R.M. Hare (in his contribution to the 1955 symposium “Theology and Falsification”) introduced the concept of the blik — a fundamental framework for viewing the world that is not itself open to empirical refutation.⁹

Hare’s illustration: a student is convinced that all university dons want to murder him. No amount of evidence — friendly dons, reassuring behaviour, the absence of actual attacks — changes his mind. He has a blik about dons. Is the student’s blik wrong? Hare says it is insane — but there is no neutral standpoint from which to prove it false, because it determines how he interprets all the evidence.

Hare’s point: we all have bliks. The sane man’s blik — “dons are basically well-intentioned” — is also not provable from scratch. The difference, on Hare’s account, is not that one is empirically grounded and the other is not, but that one is more appropriate to the available evidence. Flew and others reply that this “appropriateness,” if it is to do any work, must collapse back into the empirical-grounding notion Hare claimed to have abandoned. Whether that collapse goes through is what the Wittgensteinian defenders of bliks (Phillips, later) deny.

Religious belief, Hare suggests, is a blik of this kind: a fundamental interpretive framework, not a hypothesis that can be tested. This sidesteps the falsification challenge — but at the cost of making religious belief look like a very peculiar thing.

1.4 Wittgenstein: Religion as a Form of Life

Ludwig Wittgenstein’s later philosophy offers a different account. In Philosophical Investigations (1953) and the posthumously published Lectures and Conversations on Aesthetics, Psychology and Religious Belief (1966), Wittgenstein argued that language is not a single system with one set of rules, but a family of overlapping language games, each embedded in a form of life.¹⁰

Religious language is not a defective or primitive attempt at scientific language. It is a different language game with different rules. The mistake is to treat religious claims as if they were empirical hypotheses and then complain that they can’t be tested:

“The point is that if there were evidence, this would in fact destroy the whole business… Suppose somebody had seen not only Jesus, but also the Last Judgement, and he makes his report. You would say there must be something wrong.” — Ludwig Wittgenstein, Lectures and Conversations on Aesthetics, Psychology and Religious Belief (1966), “Lectures on Religious Belief,” Lecture I¹¹

Wittgenstein’s point: if a Christian belief in the Last Judgement were just a prediction about a future event that could be confirmed or disconfirmed by observation, it would not be a religious belief. Its grammar is different. It governs an entire way of life.

Critics — including many theologians — have argued that this protects religion only by emptying it. If the resurrection is not a claim about whether a body left a tomb, then orthodox Christianity, on its own self-understanding, has been misdescribed; if “God exists” is not a claim about reality independent of believers, then the believer cannot rationally pray to be wrong about the existence of God. The “Wittgensteinian fideism” charge was named by Kai Nielsen, but the position Nielsen attacks is best read against D. Z. Phillips’s actual Wittgensteinian theology (e.g., The Concept of Prayer, 1965, and Religion Without Explanation, 1976) — Phillips denies that he is a fideist at all, and the back-and-forth between them is the place to study the dispute, not Nielsen’s attack alone.¹²

1.5 Tillich’s Ultimate Concern

Paul Tillich (Dynamics of Faith, 1957; cf. Systematic Theology I, 1951) offers an account of faith that cuts across the cognitive/non-cognitive divide. Faith, for Tillich, is “the state of being ultimately concerned” — the orientation of the whole person toward that which concerns them unconditionally. God, in this account, is “the ground of being,” not a being among beings — and on Tillich’s view it is precisely as a being among others that God cannot exist, because that would make God finite.¹³

“Faith is the state of being ultimately concerned: the dynamics of faith are the dynamics of man’s ultimate concern.” — Paul Tillich, Dynamics of Faith (1957), Chapter 1¹⁴

The language of faith, on this account, is necessarily symbolic — but Tillich is emphatic that a symbol is not a mere sign. A symbol participates in the reality to which it points: “One should never say ‘only a symbol,’ but one should say ‘not less than a symbol.’”¹⁵ On Tillich’s own self-understanding this is not a reduction of God to a human projection but a refusal to make God an object alongside other objects.

Tillich’s appeal is plain: religious belief is not a detachable compartment of life but a pervasive orientation, and naive theistic personalism (God as a very large invisible person) misses what the classical tradition meant. Whether anything determinate is left of “God” once the personalism is gone is what realist theologians and analytic critics deny — and Tillich’s defenders affirm.

Critics press the question from two directions. From analytic philosophy, Alvin Plantinga (Does God Have a Nature?, 1980) asks: if “God” no longer names an entity that could in principle have failed to exist, what does talk of God’s actions, knowledge, and will refer to? Symbolic theology, on this reading, risks becoming indistinguishable from a sophisticated atheism that retains the vocabulary.¹⁶ From within Christian theology, Hans Urs von Balthasar and later evangelical critics worry that Tillich’s God is a philosophical abstraction rather than the personal God of biblical revelation: trinitarian and incarnational claims require a more determinate metaphysical commitment than “ground of being” supplies.¹⁷

Defenders reply that the realist objection presupposes the very picture of God as one entity among others that Tillich was trying to dismantle. The afterlife of the position is its own evidence: John A. T. Robinson’s Honest to God (1963), which popularised Tillich for an Anglican readership, became one of the best-selling theological books of the twentieth century — and provoked exactly the realist/orthodox split just described.¹⁸

1.6 Questions to Argue About

• If religious claims are non-cognitive (not true or false), does that make them more or less important? Can something be deeply meaningful without being true?
• Wittgenstein says judging religious language by the standards of scientific language is a category mistake. Is this a compelling defence of religion, or a way of insulating it from legitimate criticism?
• Hare says we all have bliks — fundamental interpretive frameworks that can’t be empirically refuted. Does this mean the difference between religious and non-religious worldviews is just a difference in bliks? Is that troubling?
• Can you articulate what kind of thing would count as evidence for or against a religious claim? What does your answer tell you about the nature of that claim?

2 Can faith be rational?

Faith and reason are often treated as opposites — the two great territories of human thought, separated by a border that each side regularly accuses the other of crossing illegally. But this is too simple. Underneath the territorial dispute is a question about epistemic permission: is it permissible — or even required — to believe things for which you lack decisive evidence? Philosophers have given very different answers, and the disagreement runs deep enough to touch the foundations of both science and religion.

2.1 Clifford’s Ethics of Belief

W.K. Clifford’s essay “The Ethics of Belief” (1877) opens with a case: a shipowner who, despite misgivings about his vessel’s seaworthiness, convinces himself that all is well and sends the ship to sea. It sinks; the passengers die. Clifford’s verdict: the shipowner is morally responsible for those deaths, even if his belief was sincere — because the belief was not adequately grounded in evidence.²²

Clifford generalises this to a universal maxim:

“It is wrong always, everywhere, and for anyone, to believe anything upon insufficient evidence.” — W.K. Clifford, “The Ethics of Belief” (1877)²³

Clifford’s argument is epistemic consequentialist: our beliefs, held in common, form the fabric of shared knowledge on which social life depends. To believe carelessly — to let wish or tradition or emotional comfort do the work that evidence should do — is to corrupt that fabric. It is not merely foolish; it is wicked.

2.2 William James: The Will to Believe

William James responded directly to Clifford in “The Will to Believe” (1896). James agrees that beliefs should track evidence — but argues that Clifford’s rule is itself a practical choice with costs.

Consider a genuine option: a choice between two hypotheses that is living (both are real possibilities for you), forced (you must choose — not choosing is itself a choice), and momentous (the stakes are high, the choice is irreversible). In such a case, James argues, the risk of error is not one-sided. If you demand evidence before believing in something, and the something is true, you may miss a real truth through over-caution.

“Our errors are surely not such awfully solemn things. In a world where we are so certain to incur them in spite of all our caution, a certain lightness of heart seems healthier than this excessive nervousness on their behalf.” — William James, “The Will to Believe” (1896), §VII²⁴

James’s specific claim about religious belief: if God’s existence is a live option for you, and if the existence of God cannot be settled by the available evidence, then you have the right to believe — and to let that belief shape your life — without this being epistemically irresponsible.

2.3 Pascal’s Wager

Blaise Pascal (Pensées, c. 1660, published 1670) anticipates James’s structure. Suppose God either exists or does not — a binary choice. What should you bet?²⁵

The expected utility calculation is obvious: belief dominates. Even if the probability of God’s existence is very small, multiplied by infinite gain, it outweighs any finite costs.

Three standard objections (each contested in the literature):

1. Which God? Pascal assumes the Christian God. But there are many candidate gods, some of whom might reward sincere disbelief and punish calculating belief.
2. Authenticity: Can you genuinely believe as a result of a strategic calculation? Pascal thought you could cultivate belief through practice — what he actually advised the hesitant unbeliever was to “take holy water, have masses said, and so on” until belief followed (the popular “kneel and pray, and faith will come” is a paraphrase, not a quotation);²⁶ critics think this is self-deception.
3. The many-gods problem: If Pascal’s argument works for any religion with infinite rewards and punishments, it gives no reason to choose Christianity over Islam, Hinduism, or any other.

2.4 Plantinga’s Reformed Epistemology

Alvin Plantinga’s Warranted Christian Belief (2000) argues that the question “Is belief in God rational?” is misconstrued. The correct epistemological question is whether belief in God has warrant — the property that makes true belief into knowledge.²⁷

Plantinga’s central move: not all knowledge is based on evidence and argument. Some beliefs are properly basic — held without inferential support from other beliefs, and rationally held in this way. Perceptual beliefs (“there is a table in front of me”) are properly basic: I don’t believe in the table because I have evidence for it; I simply see it.²⁸

Plantinga argues that belief in God can be properly basic in the same way. If there is a God, and if God has designed human beings with a sensus divinitatis — a faculty for perceiving God — then belief in God produced by this faculty would be rational, warranted, and potentially constitute genuine knowledge.²⁹

Plantinga’s critics (Linda Zagzebski, Paul Draper, others) press two challenges. First, the warrant Plantinga claims is conditional on the existence of God, so the argument cannot convince a non-theist that theistic belief is warranted — only show that if theism is true, theistic belief is rationally held. That is a smaller conclusion than apologetic uses of Plantinga sometimes suggest. Second, the parity problem: if the sensus divinitatis model warrants Christian belief in God, a parallel “sensus” model would warrant the Hindu’s experience of Brahman and the Muslim’s of Allah. Plantinga’s reply — that not all such sensus-claims are equally well-grounded — requires the kind of inter-tradition evaluation Reformed epistemology was designed to bypass.³⁰

2.5 Questions to Argue About

• Clifford says it is wrong to believe without sufficient evidence. Is this a claim about ethics, epistemology, or both? Are they separable?
• James says that in forced options on questions where evidence is unavailable, we have the right to believe. Is religious belief actually a forced option — can you genuinely remain agnostic?
• Pascal’s wager assumes that if you believe strategically, God will reward you. But if God is omniscient, would God not know you believed for strategic reasons? Does this undermine the wager?
• Plantinga says belief in God can be properly basic — rational without evidence, like perceptual belief. Does this analogy hold? What is the relevant difference between believing in a table and believing in God?

3 Does God exist — and can philosophy settle the question?

The arguments for and against God’s existence have been refined over a thousand years — sharpened, challenged, revised, sharpened again. It says something about the difficulty of the question that no version has ever achieved consensus. It says something different about the seriousness of the enterprise that some of the finest minds in human history have thought it worth trying. This lesson examines the best versions of those arguments — not to settle the question, but to understand what kind of question it is, and what “settling” it would even look like.

3.1 The Ontological Argument

Anselm of Canterbury (Proslogion, 1078) constructed an argument for God’s existence from the concept of God alone:³⁴

1. God is “that than which nothing greater can be conceived.”³⁵
2. Something that exists in reality is greater than something that exists only in the understanding.
3. Therefore, if God existed only in the understanding, we could conceive of something greater — a God that also existed in reality. But this contradicts premise 1.
4. Therefore, God must exist in reality.

Descartes offered a compressed version: existence is a perfection; God is a perfect being; therefore God necessarily exists.³⁶

Kant’s objection (Critique of Pure Reason, 1781): existence is not a predicate. “The tallest building is tall” tells you something about the building; “the tallest building exists” tells you that there is such a building, but adds nothing to the concept of the tallest building. Anselm’s argument illicitly treats existence as a property that enhances a being.³⁷

Gaunilo (Anselm’s contemporary) offered a parallel parody: “I can conceive of the greatest possible island. Such an island would be greater if it existed than if it did not. Therefore the greatest possible island exists.” This is absurd. If the argument works for God, it should work for perfect islands — so something is wrong with it.³⁸

3.2 The Cosmological Argument

Aquinas (Summa Theologiae, c. 1265–74) offered five ways of demonstrating God’s existence.³⁹⁴⁰ The most discussed:

The First Way (Motion): Everything that moves is moved by something else. This regress cannot go on to infinity; there must be a first, unmoved mover. This is God.

The Second Way (Causation): Everything has an efficient cause. Nothing can be its own cause. An infinite regress of causes is impossible. Therefore there must be a first, uncaused cause. This is God.

The objection: Why can’t the universe itself be the uncaused cause — the brute fact at the bottom of explanation? The standard Thomist reply is that Aquinas’s argument is not “everything has a cause” (which would apply to God) but “every contingent being has a cause” — God being necessary, not contingent. Whether anything answers to “necessary being” is then the contested question, and the cosmological argument must be evaluated on that ground rather than dismissed by misreading premise 1. The contemporary literature does not converge on the strongest formulation: the Leibnizian PSR-based argument (Pruss, Rasmussen), Feser’s neo-Thomist reading of Aquinas’s per se causal series, and Craig’s Kalām argument (structurally different, resting on the impossibility of an actual infinite past) all have serious defenders and serious critics.⁴¹

Bertrand Russell made the contrary point in his debate with Father Copleston (BBC, 1948): “The universe is just there, and that’s all.”⁴² Russell’s celestial teapot (from his 1952 essay “Is There a God?”) extends this into a methodological principle: if he claimed there was a china teapot orbiting the Sun between Earth and Mars, too small for any telescope to detect, no one could disprove him — but it would be absurd, he said, to expect rational discourse to suspend disbelief on that basis.⁴³ The image was not offered as a refutation of theism. It was offered as a clarification of where the burden of proof lies in disputes about the existence of unobservable entities, and it became one of the most widely cited philosophical images of the twentieth century. Russell extended the same negative argument across the classical proofs in his 1927 lecture “Why I Am Not a Christian,” in which he surveyed and rejected the cosmological, teleological, and moral arguments.⁴⁴

Alvin Plantinga’s three-decade reply, developed across God and Other Minds (1967) and Warranted Christian Belief (2000), challenges Russell’s implicit premise that belief in God requires positive evidential justification of the kind a teapot would. If belief in other minds, the external world, or the past requires no prior proof — if these are properly basic beliefs that we are entitled to hold without an evidential argument — then on Plantinga’s view belief in God can be properly basic too, produced by what Calvin called the sensus divinitatis operating in conditions where God exists. Russell and Plantinga frame the two opposing answers between which everything else falls: that theism must meet a higher evidential bar than ordinary beliefs and fails to meet it (Russell), or that the demand for that higher bar is itself a contestable epistemological assumption (Plantinga). The teapot illustrates the first; the parity argument with other-minds belief illustrates the second.

3.3 The Teleological Argument

William Paley (Natural Theology, 1802) offered the most famous version: if you found a watch on a heath, its complexity would compel you to infer a maker. The universe exhibits similar or greater complexity. Therefore the universe has a maker.⁴⁵

Hume’s objections (Dialogues Concerning Natural Religion, written 1750s, published 1779) are powerful — though contemporary defenders of design-style arguments (Richard Swinburne, fine-tuning theorists) hold that they survive Hume in modified form:⁴⁶⁴⁷

1. The analogy is weak: the universe is not like a watch. We have experience of watches being made; we have no experience of universes being made.
2. The inference is too strong: from the premise that the universe shows some design, we cannot infer an omnipotent, omniscient, omnibenevalent designer. The evidence might support a much more limited designer, or a committee of designers, or a junior deity.
3. The regress: if complexity requires a designer, a complex God requires a designer in turn.

Darwin’s theory of natural selection (1859) offered an alternative account of biological complexity that required no designer — random variation plus selection pressure generates apparent design.⁴⁸ Whether Hume’s Dialogues anticipated the structure of this argument before Darwin had the mechanism is itself debated in Hume scholarship; the family resemblance is real, the question of whether Hume saw it is open.

3.4 Reading the Arguments as Arguments

The arguments for God’s existence are not positions to appreciate from a distance. They have logical structures — premises and conclusions — and the correct philosophical response is to identify those premises and decide which to attack. What follows is a map. Use it.

The Ontological Argument (Anselm)

P1: God is defined as the greatest conceivable being.
P2: A being that exists in reality is greater than one that exists only in the mind.
P3: If God existed only in the mind, we could conceive of a greater being — one that also exists in reality. But this contradicts P1.
C: Therefore God exists in reality.

P1 is stipulative: it defines God into a particular category of greatness. This is not obviously illegitimate, but it invites the question of whether “greatness” can include existence as a component without begging the conclusion. P2 is where Kant’s objection lands directly: existence, Kant argues, is not a predicate that adds to a concept — it is the condition for any predicate applying at all. Gaunilo’s parody applies the same structure to a perfect island and obtains an absurd result, suggesting the argument form is invalid regardless of what is plugged into it.

If you accept P1 and P2, the conclusion does follow. So: which premise will you reject, and why? Do not say both are wrong — pick the premise that fails first and explain the failure precisely.

The Cosmological Argument (Aquinas)

P1: Everything that exists has a cause of its existence.
P2: An infinite regress of causes is impossible.
P3: Therefore there must be a first, uncaused cause.
C: This first uncaused cause is God.

The argument has two obvious attack points. P1 is too strong as stated: if everything has a cause, then God requires a cause, which contradicts C. The defender typically revises P1 to “everything that contingently exists has a cause,” distinguishing necessary from contingent beings. This revision needs defending — why should we believe any necessary being exists? P2 is also contested: Hume and Russell deny that an infinite regress is impossible, or that the universe cannot simply be a brute, uncaused fact. The transition from P3 to C is also a leap: even granting a first cause, nothing in the argument establishes that this cause is personal, omniscient, or omnipotent rather than simply the universe itself at its earliest moment.

Which premise will you reject? State your choice and the argument.

The Problem of Evil (Mackie, against theism)⁴⁹

P1: God is omnipotent (can prevent any evil).
P2: God is omniscient (knows of all evil).
P3: God is omnibenevolent (wants to prevent all evil that does not serve a greater good).
P4: Evil exists which serves no greater good.
C: Therefore a being satisfying P1–P3 does not exist.

The bracketed qualifier in P3, and its mirror in P4, are where the theodicy debate actually happens; without them the argument can be defused trivially by the observation that some evils serve goods (a vaccination hurts but prevents disease). Mackie’s original (1955) formulation did not include the qualifier, and Plantinga’s free-will defence exploits exactly this looseness.

The theist must deny at least one of P1–P4. The free will defence denies that P1 + P3 together require the prevention of all evil — God permits moral evil because free creatures are a greater good than moral automatons.⁵⁰ The sceptical theist denies that P3 entails an obligation to prevent every evil — God may have reasons inaccessible to us. Note that denying P4 is not available; no serious theist takes this route. The argument from gratuitous evil strengthens P4 by specifying a subset of evil — the suffering of children from disease, with no apparent connection to free will or soul-making — that theodicy has the greatest difficulty addressing. Ivan Karamazov’s “rebellion” in The Brothers Karamazov is precisely this argument, made not as a logical demonstration but as a moral revolt: even granting the logical possibility of a theodicy, some suffering feels like it exceeds any possible justification.⁵¹

A fourth response — distinct from the free-will defence, soul-making theodicy, and sceptical theism — denies P1 itself. Process theology (Alfred North Whitehead, Process and Reality, 1929; Charles Hartshorne, The Divine Relativity, 1948; later John Cobb and David Ray Griffin) argues that the classical-theist God of unrestricted omnipotence is itself the philosophical mistake on which the problem of evil rests.⁵² On the process account, God is genuinely affected by what happens in the world (Hartshorne calls this the dipolar God), persuades rather than coerces creatures, and works with the creative possibilities of an open universe rather than determining its outcomes. The price is high — process theology revises the doctrine of omnipotence and accepts a God who suffers — but the dividend is that gratuitous evil ceases to be a contradiction in the divine attributes. Whether the price is worth paying, and whether the resulting “God” is recognisable as the God of either philosophical theism or the Abrahamic traditions, is the question process theology’s critics (analytic theists like Plantinga, classical theists like David Bentley Hart) press in different ways.

Which premise is weakest? State the attack.

3.5 Questions to Argue About

• Kant says existence is not a predicate. Does this objection destroy the ontological argument, or can the argument be reformulated to survive it?
• Aquinas says the regress of causes must stop somewhere. Why must it stop at God rather than at the universe itself?
• Hume says the analogy between a watch and the universe is too weak to carry the teleological argument. Can you strengthen the analogy, or is Hume right?
• Does the existence of suffering count as evidence against God’s existence? If so, how strong is it? If not, why not?

4 Can science and religion be reconciled?

The conflict between science and religion is one of the defining intellectual stories of modernity — told, retold, and occasionally distorted beyond recognition by both sides. Richard Dawkins and the Archbishop of Canterbury have rather different accounts of what the conflict is about, and the distance between them is not merely temperamental. But is the conflict real — a genuine contradiction between two systems of claims — or is it a misunderstanding about what each is doing?

The Dawkins-versus-Archbishop framing is the populariser version of the dispute, and it tends to debate a strawman that no serious philosophy of religion has held since the early modern period — the picture of God as one more competing scientific hypothesis. The contemporary disciplinary debate, conducted by figures such as Alvin Plantinga, Alister McGrath, John Polkinghorne, and Sarah Coakley, is normally classified — following Ian Barbour — under four headings: conflict, independence (Gould’s NOMA), dialogue, and integration.⁵³ Dawkins is one position in this typology, not the science-side default.

4.1 Gould’s NOMA

Stephen Jay Gould proposed the principle of Non-Overlapping Magisteria (NOMA) in his essay “Nonoverlapping Magisteria” (1997) and book Rocks of Ages (1999).⁵⁹ Science and religion, Gould argued, occupy different magisteria — domains of authority — and therefore cannot conflict:

“Science tries to document the factual character of the natural world, and to develop theories that coordinate and explain these facts. Religion, on the other hand, operates in the equally important, but utterly different, realm of human purposes, meanings, and values.” — Stephen Jay Gould, “Nonoverlapping Magisteria” (1997)⁶⁰

On this account, the apparent conflict is always the result of one discipline trespassing on the other’s territory: scientists making claims about meaning and value; religious authorities making claims about natural history.

The objection: many religious traditions make factual claims — about the age of the earth, the origin of species, the occurrence of miracles, the survival of consciousness after death. These claims are not in the realm of meaning and value; they are empirical claims subject to scientific investigation. NOMA cannot be the whole story unless we significantly revise what religious claims mean.

4.2 The Galileo Case

The 1633 trial of Galileo Galilei before the Inquisition is the canonical case of religious authority suppressing scientific knowledge.⁶¹ Galileo had published the Dialogue Concerning the Two Chief World Systems (1632), which effectively supported Copernican heliocentrism over the Ptolemaic geocentric model endorsed by the Church.⁶² The contested claim was specifically that the Sun is motionless at the centre of the cosmos and the Earth moves — a claim the Church held to be contrary to scripture (Joshua 10:13; Psalm 93:1) on the literal reading then standard.

Cardinal Robert Bellarmine had already, in his April 1615 letter to the Carmelite Paolo Antonio Foscarini, drawn the distinction the trial would later turn on: a mathematician may treat heliocentrism as a useful supposition for prediction, but to assert it as physically true, against scripture, is to require demonstrative proof — which (Bellarmine maintained) Galileo did not have.⁶³ On 22 June 1633 the Inquisition found Galileo “vehemently suspect of heresy” — the technical formula for a serious-but-not-formal heresy charge — and required a public abjuration in which he was made to “abjure, curse, and detest” the Copernican opinion.⁶⁴ He was sentenced to imprisonment at the Inquisition’s pleasure, commuted the next day to house arrest, which he kept until his death in 1642. He was not tortured; the procedural threat of torture (territio verbalis) was made, but no torture was applied — and the legendary muttered “eppur si muove” (“and yet it moves”) is a story first attested over a century later.⁶⁵

The history is more complicated than the standard narrative suggests. Galileo had friends within the Church hierarchy, including Pope Urban VIII. The conflict was partly political, partly personal, and partly scientific: Galileo’s proofs for heliocentrism were not, by the standards of the time, conclusive — stellar parallax, the observational confirmation he would have needed, was not measured until 1838. Whether the Church’s position is best read as a methodologically conservative demand for stronger evidence (the revisionist reading: Lindberg, Numbers) or as the institutional defence of scriptural authority against Reformation polemic dressed in epistemic clothing (the older reading) is itself a live historiographical dispute.

The revisionist literature on Galileo (David Lindberg, Ronald Numbers, John Hedley Brooke) reads the conflict between science and religion as historically real but historically specific — not a simple clash between reason and faith, but an entanglement of institutions, politics, and the sociology of knowledge. The contrary reading — that the revisionism understates a structural conflict between empirical method and dogmatic authority that the Galileo affair does in fact illustrate — retains its defenders.⁶⁶

4.3 Lemaître and the Big Bang

Georges Lemaître was a Belgian Catholic priest and physicist who in 1927 proposed what became the Big Bang theory — the hypothesis that the universe began from a primeval atom in an extremely dense state.⁶⁷ Einstein initially rejected the idea (“Your mathematics is correct, but your physics is abominable”⁶⁸), then accepted it.

In his address Un’Ora to the Pontifical Academy of Sciences on 22 November 1951, Pope Pius XII argued that contemporary cosmology had “succeeded in being a witness to that primordial Fiat Lux” and that science had now confirmed “the contingency of the universe and the well-founded deduction” that the cosmos issued from the Creator’s hand at a definite moment.⁶⁹ Lemaître objected on three grounds: the primeval-atom hypothesis was at the time still empirically contested (the steady-state alternative would not be decisively ruled out until the 1965 cosmic microwave background discovery); a physical “beginning” is not the same kind of claim as a theological “creation” ex nihilo; and a working cosmologist who was also a Catholic priest stood to be discredited in his own field by the appearance of having done physics to prop up doctrine. He approached Daniel O’Connell, S.J., director of the Vatican Observatory (and possibly Msgr. Angelo Dell’Acqua of the papal Secretariat of State, with the substantive intermediary apparently being Msgr. Tardini, who drafted the Pope’s addresses), and together they persuaded Pius XII to remove the primeval-atom identification from the discourse he had prepared for the inaugural session of the eighth assembly of the International Astronomical Union; in the resulting address at Castel Gandolfo on 7 September 1952, the cosmological-creation identification was excised — though, as Lambert notes, the underlying conception of the relation between faith and science was not essentially changed.⁷⁰ He wanted science left to scientists.

Lemaître is a concrete challenge to the simple conflict narrative: he was simultaneously a committed Catholic priest and the originator of one of the most important cosmological theories of the 20th century. He saw no contradiction, but he was careful to keep the domains separate.

4.4 The 1755 Lisbon Earthquake

On 1 November 1755 (All Saints’ Day), an earthquake devastated Lisbon, killing between 30,000 and 50,000 people.⁷¹ Fires and a tsunami compounded the destruction. Catholic churches, packed for the festival, collapsed. Brothels survived.

The theological reaction: was this divine punishment? For what? Voltaire’s response was savage. He wrote the poem “Poem on the Lisbon Disaster” (1756)⁷² and embedded the problem in Candide (1759), where the character Pangloss — a parody of Leibniz’s optimism that “this is the best of all possible worlds”⁷³ — continues his philosophical cheerfulness in the face of catastrophe.⁷⁴

Rousseau’s reply was instructive: the deaths were not God’s fault but humanity’s. If people had been living as they should — dispersed in the countryside — they would not have been concentrated in a city that collapsed on them.⁷⁵ Rousseau was already thinking about the harms of urban civilisation.

Kant himself responded with three short scientific essays in the first months of 1756, treating the earthquake as a natural phenomenon to be explained by subterranean cavities, gases and chemical reactions rather than as divine retribution — an early case of the science-religion split between explanation and meaning.⁷⁶

The earthquake forced a public confrontation with theodicy that the polished arguments of philosophical theology had not prompted. It is one of the first events that visibly damaged optimistic theology in the European public sphere.

4.5 Intelligent Design

Intelligent design (ID) — the claim that certain biological features (e.g., the bacterial flagellum) are too complex to have evolved by natural selection and must have been designed — presents itself as scientific. It was argued before a US federal court in Kitzmiller v. Dover Area School District (2005), where Judge John E. Jones ruled that ID was not science but religion, and could not be taught in public school science classes.⁷⁷

The key issue was methodological naturalism: on the standard account, science as a practice is committed to naturalistic explanations and does not invoke supernatural causes. On this account, methodological naturalism is not a metaphysical claim that the supernatural does not exist; it is a constraint on what counts as a scientific enquiry — and ID, in invoking a designer, breaches it. For Behe, Dembski, and Plantinga, this begs the question: methodological naturalism is itself a contested philosophical commitment rather than an analytic feature of science, and the Kitzmiller ruling assumes precisely what is at issue.⁷⁸

The deeper philosophical question: is methodological naturalism an arbitrary restriction, or is it constitutive of what makes science science? Both answers have serious defenders.

4.6 Questions to Argue About

• Gould says science and religion cover different magisteria. But many religious traditions make factual claims about history, biology, and cosmology. How should NOMA handle this?
• The Galileo case shows that even within the Catholic Church in the 17th century, the relationship between science and theology was complicated. What does this suggest about treating “science vs. religion” as a simple conflict?
• Is methodological naturalism a fair ground rule for science, or does it beg the question against supernatural explanations from the start?
• Voltaire attacked theodicy through Candide. Is satire a legitimate form of philosophical argument? What can it do that systematic argument cannot?

5 What is religious experience — and what does it prove?

Millions of people across centuries and cultures have reported experiences of the divine — feelings of overwhelming presence, union with the infinite, voices, visions, the dissolution of the boundary between self and world. These reports are not the province of the credulous or the simple. Julian of Norwich, Teresa of Ávila, al-Ghazali, Simone Weil⁷⁹ — serious, scrutinising intellects who were not at all sure what had happened to them but were equally not at all sure it was nothing. What, if anything, do these experiences tell us about reality?

5.1 Hildegard of Bingen and the Migraine Theory

Hildegard of Bingen, the twelfth-century abbess, composer, and polymath, began experiencing vivid visions from childhood. She described them in extraordinary detail in her trilogy Scivias,⁸³ illustrated under her supervision with images of spinning discs, concentric rings of light, and radiant points surrounded by flickering halos. In the twentieth century, neurologist Oliver Sacks examined Hildegard’s visual descriptions in Migraine (1970) and argued that the imagery was consistent with the scintillating scotomas characteristic of migraine aura.⁸⁴ Several illustrations correspond precisely to the fortification spectra that migraine sufferers report. The reductive explanation does not settle the philosophical question: even if Hildegard’s experiences had a neurological cause, it remains open whether a divine agent could work through neurological mechanisms, whether the contents of experience can be explained away by explaining their substrate, and whether a vision that transformed an intellectual and musical tradition is adequately described as “a headache.” James’s “medical materialism” critique cuts the same way against Sacks’s twentieth-century reduction as against the present-day serotonin-receptor account.

5.2 The Johns Hopkins Psilocybin Trials (2006–)

In 2006 the psychopharmacologist Roland Griffiths and colleagues at Johns Hopkins University published “Psilocybin can occasion mystical-type experiences having substantial and sustained personal meaning and spiritual significance” in the Journal of Psychopharmacology.⁸⁵ Thirty-six healthy volunteers, none with prior hallucinogen experience, received in double-blind sessions either psilocybin (the active alkaloid in Psilocybe mushrooms) at 30 mg/70 kg or methylphenidate as an active placebo. Outcomes were assessed using Walter Pahnke’s Mystical Experience Questionnaire (1969), which operationalises James’s four marks of mystical experience: ineffability, noetic quality, transiency, passivity. Twenty-two of the thirty-six participants in the psilocybin condition met the criteria for a “complete mystical experience”; at the fourteen-month follow-up, roughly two-thirds rated the session among the five most spiritually significant experiences of their lives.⁸⁶ Subsequent Hopkins studies (Griffiths et al., 2016; the 2018 paper comparing trial-induced and spontaneous mystical experiences) extended the work to terminally ill cancer patients, where the experiences produced sustained reductions in depression and existential distress, and found no statistically discernible difference between trial-occasioned mystical experiences and the spontaneous ones recorded in the religious-studies literature.⁸⁷ Griffiths himself — an atheist for most of his career — told The New York Times in 2016 that the volunteer reports had unsettled him not because they were exotic but because they were exactly the same kind of report, in vocabulary and long-term effect, as the historical mystics had given.⁸⁸ The Hopkins programme is the controlled, randomised continuation of the question that Hofmann’s bicycle ride opened in Basel sixty-three years earlier — and Switzerland is also the rare jurisdiction where licensed psilocybin- and LSD-assisted therapy has been authorised in selected clinical settings since 2007 (Peter Gasser, Solothurn).⁸⁹ Three readings of the same data sit on the file. Reductive: psilocybin’s serotonin 5-HT₂A-receptor agonism produces, by an identifiable neurochemical mechanism, the entire phenomenology James catalogued; the historical mystics’ claims to encounter divine reality are exhaustively explained by neurochemistry. Insider: a sacrament that humans across cultures (the Aztec teonanácatl, the Mazatec velada, the Eleusinian mysteries, possibly some Vedic soma) used to access a real domain has been recovered in a clinical setting; the molecular mechanism is the route of access, not the replacement for what is accessed. William James’s: tracing the psychophysical origins of religious experience does not address its cognitive validity — the question of what, if anything, such experience puts the experiencer in contact with.⁹⁰

5.3 William James on Mystical Experience

William James’s The Varieties of Religious Experience (1902) is the foundational text.⁹¹ James collected hundreds of accounts of religious experience and identified four marks of mystical experience:⁹²

1. Ineffability: the experience cannot be adequately described in language — it must be directly experienced to be understood.
2. Noetic quality: the experience carries a sense of having knowledge — of having been shown something true about the nature of reality.
3. Transiency: the experience does not last; it cannot be sustained indefinitely.
4. Passivity: the mystic does not produce the experience by effort of will; it arrives and takes hold.

“Our normal waking consciousness, rational consciousness as we call it, is but one special type of consciousness, whilst all about it, parted from it by the filmiest of screens, there lie potential forms of consciousness entirely different.” — William James, The Varieties of Religious Experience (1902), Lecture XVI⁹³

James was cautious about metaphysical conclusions, but he argued that the pragmatic significance of religious experience was undeniable: it transformed people’s lives, produced lasting changes in character, and could not be dismissed as merely pathological.

A live debate about James’s four marks. James himself, and W. T. Stace after him (Mysticism and Philosophy, 1960), treated these features as evidence for a perennial mystical core common across religions — Christian, Hindu, Buddhist, Sufi, Daoist mysticism all converging on the same underlying experience, differently described.

Steven Katz’s contextualism, articulated in “Language, Epistemology, and Mysticism” (in Mysticism and Philosophical Analysis, 1978), is the post-1980 consensus alternative: there are no pure, unmediated experiences. The Christian mystic experiences a Christian God; the Buddhist experiences emptiness; the Sufi experiences fana with Allah. The doctrinal framework the mystic brings constitutes, not just labels, the experience. The “convergence” Stace claimed is, on Katz’s reading, a projection produced by reading different traditions’ descriptions through one perennialist filter. Most contemporary scholars of mysticism work with some version of Katz’s claim — though Robert Forman and others have argued back for a “pure consciousness event” in the meditative traditions specifically.⁹⁴

5.4 Julian of Norwich

Julian of Norwich received a series of revelations on 8 May 1373, while near death from illness. She recorded them in Revelations of Divine Love — the oldest surviving book in English written by a woman.⁹⁵

Her visions centred on the passion of Christ. Out of them came her most famous formulation:

“All shall be well, and all shall be well, and all manner of thing shall be well.” — Julian of Norwich, Revelations of Divine Love (long text), Chapter 27⁹⁶

Julian’s “all shall be well” is not easy optimism. She reached it after extended contemplation of the problem of sin and suffering. She held it together with unflinching attention to the physical reality of Christ’s suffering. It is a theological conclusion, not a platitude.

Julian’s account raises the insider/outsider problem in sharp form: from the outside, her experience can be described as a near-death fever-delirium. From the inside, it was a direct encounter with divine love that restructured her entire understanding of reality.

5.5 Muhammad’s Revelation at Hira

Islamic tradition records that in 610 CE, Muhammad was meditating in the Cave of Hira when the angel Jibril (Gabriel) appeared and commanded him to “Iqra!” — recite, or read. Muhammad initially resisted, saying he could not read. The revelation continued; the words that came to him became the first verses of what would become the Qur’an.⁹⁷

Muhammad’s response, as reported in Sahih al-Bukhari, was not confidence but terror. He returned to his wife Khadijah shaking, saying “Cover me!” He feared he had been visited by an evil spirit. Khadijah reassured him; her cousin Waraqah ibn Nawfal, a learned Christian, identified the experience as the same kind of divine revelation that had come to Moses.⁹⁸

This account is significant for epistemological purposes: the revelation came unsought, produced fear rather than elation, and was authenticated not by Muhammad himself but by the testimony of others who interpreted the experience through an existing framework of prophethood.

5.6 The Insider/Outsider Problem

Phenomenologists of religion distinguish between emic accounts (insider perspectives, describing religious experience from within the tradition’s own categories) and etic accounts (outsider perspectives, describing the same experience in secular, usually scientific, terms).

The etic accounts are not automatically superior to the emic. They explain the experience by reference to factors the experiencer would not recognise as relevant:

• Freud (The Future of an Illusion, 1927): religious experience is the projection of the father-figure onto the cosmos — an infantile wish-fulfilment. The sense of union with something greater is a recovery of the pre-ego “oceanic feeling” of infancy.⁹⁹
• Marx: religious experience is ideological — it reconciles people to their suffering by pointing to supernatural consolation, making material injustice tolerable.¹⁰⁰
• Durkheim (The Elementary Forms of Religious Life, 1912): the power felt in religious experience is real — but it is the power of society experienced as supernatural. When worshippers feel God, they are feeling the collective force of their community.¹⁰¹

The challenge for reductive accounts — one Dennett’s Breaking the Spell (2006) addresses — is that explaining the origin of an experience does not by itself show that the experience tracks nothing real. A neurological account of the brain states accompanying mystical experience does not settle whether those brain states are responses to something. Parsimony is the reductivist’s lever: once the natural mechanism accounts for the phenomenon, the burden of proof lies with whoever wants to add a transcendent referent on top of it.¹⁰² Whether parsimony can bear that weight here is itself contested.

5.7 The Neuroscience of Religious Experience

Andrew Newberg and colleagues (Why God Won’t Go Away, 2001) conducted brain imaging studies of Tibetan Buddhist meditators and Franciscan nuns during intense prayer and meditation. During peak experiences, they found consistent patterns: decreased activity in the parietal lobes (associated with the sense of self and spatial boundary), and altered activity in the prefrontal cortex.¹⁰³ The subjects described these neurological changes as experiences of boundlessness, unity, or divine presence.

Newberg’s findings are philosophically double-edged. On one reading: religious experience has identifiable neurological correlates — it is a brain event, explicable in natural terms. On another reading: of course a real encounter with divine reality would produce neurological effects, since the brain mediates all conscious experience. The presence of a neurological correlate no more proves that the experience is merely a brain event than the neurological correlate of seeing a red apple proves that red apples do not exist.

The question Newberg’s research forces is not “does religious experience happen in the brain?” — of course it does. The question is: does the brain produce religious experience, or does it register it?

5.8 Non-Western Epistemologies of Religious Experience

The James-centred account focuses on individual mystical episodes. But several non-Western traditions have developed systematic epistemologies of religious experience that present a genuinely different account of how religious knowledge is acquired.

Sufi epistemology in the Islamic tradition — developed most fully by Ibn Arabi (The Bezels of Wisdom, 13th century) — posits the qalb (heart) as a distinct faculty of knowing, irreducible to rational intellect (aql) or sensory perception. The heart, properly purified, becomes a mirror that receives divine self-disclosure (tajalli).¹⁰⁴ This is not a claim about feelings — it is a claim about a cognitive faculty that has its own object (divine reality) and its own discipline (spiritual practice). Ibn Arabi’s epistemology is not a pre-rational mysticism; it is a sophisticated account of a mode of knowing that is post-rational, reached through and beyond rational discipline.

Theravada Buddhist epistemology distinguishes multiple levels of consciousness, each with distinctive objects. At the deepest levels of samadhi (meditative concentration), the meditator is said to have direct access to the arising and passing of mental and physical phenomena at a level of resolution unavailable to ordinary consciousness.¹⁰⁵ The Abhidharma tradition — the systematic philosophical psychology of early Buddhism — treats this not as a mystical claim but as a form of empirical investigation: the meditator is the instrument, the mind is the field of inquiry, and the findings are communicable and teachable.¹⁰⁶ What is known through meditation is not supernatural — it is a more refined perception of natural processes. But it is knowledge unavailable to uninstructed consciousness.

In both cases: there is a faculty of knowing, there is a discipline for developing it, there are criteria for evaluating whether it is functioning correctly, and there are objects that are known through it. These are epistemological frameworks, not mere testimony about inner states.

Shūsaku Endō’s novel Silence (1966) puts a related question in a harder form.¹⁰⁷ The Portuguese missionary Rodrigues is made to watch Japanese Christians die one by one, under torture, while he refuses to apostatise. His God does not intervene; his God does not speak. Rodrigues interprets that silence in two irreconcilable ways across the course of the novel: as the silence of an absence (there is no God), and as the silence of a presence that cannot be described in the terms Catholic Europe gave him (the God he was taught to expect is not the God who is there). The novel does not resolve the question, and should not. What it shows is that a religious epistemology is never only a theory of how knowledge is acquired; it is also a theory of what the absence of the expected datum can mean.

5.9 Questions to Argue About

• James says mystical experience carries a “noetic quality” — a sense of having genuine knowledge. Is this sense a reliable guide to actual knowledge? Compare: dreams also feel real while we’re in them.
• Julian of Norwich’s experience can be described as a near-death hallucination or as a divine revelation. Is there any evidence that could distinguish between these two accounts? Does the absence of such evidence tell us something?
• Durkheim says the power felt in religious experience is real — it is the power of society. This means religious experience is not illusory, but it does not require a supernatural cause. Is this a satisfying account?
• The insider says “I encountered God.” The outsider says “You had an intense psychological event.” Can they both be right? Can either be wrong?
• Ibn Arabi posits the qalb (heart) as a cognitive faculty that knows divine reality. Buddhist epistemology posits levels of meditative consciousness that access processes invisible to ordinary perception. Are these claims falsifiable? If not, what follows for their epistemic status?

6 Does religion provide the foundation for ethics?

One of the oldest questions in the philosophy of religion is whether morality depends on God. Dostoevsky’s Ivan Karamazov put the challenge with characteristic ferocity: if God does not exist, is everything permitted? But Socrates — twenty-two centuries before Ivan — had already identified the deeper problem: even if God exists, does that actually settle moral questions? The Euthyphro dilemma is waiting, fully armed, for anyone who thinks “God commands it” is a sufficient answer.

6.1 The Lisbon Earthquake of 1755

On the morning of 1 November 1755 — All Saints’ Day — an earthquake of approximately magnitude 8.5¹¹¹ struck Lisbon, killing between 30,000 and 60,000 people.¹¹² The disaster struck at 9:40 a.m., when the churches were full for the holiday Mass, and the fires that followed destroyed much of what the earthquake and subsequent tsunami had spared. Voltaire’s long poem Poème sur le désastre de Lisbonne — published within weeks — asked the theodicy question with furious directness: how could providential design explain the simultaneous death of tens of thousands of worshippers in their churches on the holiest morning of the year?¹¹³ Rousseau replied that the death toll was a consequence of human choices — people had chosen to build a dense city of tall stone buildings — rather than divine malice or indifference.¹¹⁴ The exchange between Voltaire and Rousseau is the first major public philosophical debate about the problem of evil in modern European history. J. L. Mackie’s 1955 “Evil and Omnipotence” formalised, two centuries later, the argument that the existence of evil is logically incompatible with an omnipotent, omniscient, and perfectly good God.¹¹⁵ The Lisbon earthquake is the event Mackie’s argument was always really about; the 2004 tsunami is its contemporary repetition.

6.2 The Problem of Evil

The question of whether religion provides the foundation for ethics cannot be separated from a prior question: is the God that exists — if God exists — morally acceptable? The problem of evil is not merely an argument for atheism. It is a moral charge against theism.

J.L. Mackie stated the logical problem of evil with precision in “Evil and Omnipotence” (Mind 64, 1955): God is omnipotent (can prevent any evil), omniscient (knows of all evil), and omnibenevolent (wants to prevent all evil); evil exists; these propositions are jointly inconsistent. At least one must be false.¹¹⁶ The problem was already formulated in antiquity — attributed to Epicurus via Lactantius: “Is God willing to prevent evil, but not able? Then He is not omnipotent. Is He able, but not willing? Then He is malevolent. Is He both able and willing? Then whence cometh evil? Is He neither able nor willing? Then why call Him God?”¹¹⁷

Theodicy — the attempt to justify God’s ways to humans — has produced several responses:

• Free will defence (Plantinga, God, Freedom, and Evil, 1977): God permitted moral evil because the existence of genuinely free creatures is a greater good than a world of moral automatons. But this covers only evil caused by human choices, not natural evil — disease, earthquakes, the death of children from conditions entirely unrelated to human agency.¹¹⁸
• Soul-making theodicy (John Hick, Evil and the God of Love, 1966): evil is the necessary condition for moral and spiritual development. Without suffering, no courage, compassion, or resilience would develop. God’s goal is not pleasure but the formation of fully human souls.¹¹⁹
• Sceptical theism: we are not in a position to judge whether God has reasons for permitting evil that exceed our comprehension. The inference from “we can see no good reason for this suffering” to “there is no good reason” requires an epistemic confidence in our own cognitive reach that is unjustified.

The textbook ordering presents these three as competing answers; in the primary literature they answer different problems. Plantinga’s free-will defence targets Mackie’s logical problem (the claim that theism is internally contradictory); Hick’s soul-making theodicy targets the evidential problem (Rowe-style arguments from apparently gratuitous suffering); sceptical theism (Wykstra, Alston) is meta-level — it questions whether human cognitive reach is wide enough to license the inference from “no good reason visible” to “no good reason.”

Eleonore Stump (Wandering in Darkness, 2010) and Marilyn McCord Adams (Horrendous Evils and the Goodness of God, 1999) refuse the triplet altogether: Stump grounds her response in narrative rather than in a general justification, and Adams argues that only direct beatific union with God could redeem horrendous evils that wreck the meaning of an individual life.¹²⁰

Against all theodicies: the argument from gratuitous evil, formulated most influentially by William Rowe.¹²¹ If even one instance of suffering — a child dying in agony from a disease with no connection to free will, no moral lesson, no soul formed — serves no conceivable redemptive purpose, theodicy fails. Whether such suffering exists is contested. Rowe takes some cases (children’s cancers, animal suffering long before humans) as paradigmatic; sceptical theists (Wykstra, Alston) reply that the inference from “we cannot see a good reason” to “there is no good reason” requires more cognitive reach than human beings possess. The argument’s force depends on which side of that inference one stands on.

Dostoevsky’s Ivan enters here — not as an atheist making a logical argument, but as a believer making a moral revolt. His position is one option among several; readers should evaluate it against the theodicies above, not adopt it by default.

6.3 The Euthyphro Dilemma

Plato’s dialogue Euthyphro (c. 399 BCE) contains one of the most powerful arguments in the history of philosophy.¹²² Socrates asks Euthyphro:

“Is the pious loved by the gods because it is pious, or is it pious because it is loved by the gods?” — Plato, Euthyphro, 10a (translated by G.M.A. Grube)¹²³

Translated into the monotheist context: is something morally good because God commands it, or does God command it because it is good?

The dilemma has two horns:

Horn 1 (Divine Command Theory): Morality is constituted by God’s commands. What is good is good because God says so. Problem: this makes morality arbitrary — God could command cruelty, and cruelty would become good. It also makes moral reasoning pointless: if you want to know what is right, you just look up what God commands.

Horn 2 (Independent Moral Standard): God commands things because they are good — there is an independent standard of goodness that God recognises and endorses. Problem: this makes morality independent of God. God becomes a moral authority we can in principle disagree with; God’s commands are good insofar as they track the independent standard.

Most sophisticated theistic philosophers bite Horn 2: they hold that God’s commands are necessarily in accord with God’s nature, which is necessarily good. But this raises the question of why God’s nature is the standard of goodness rather than some other.

6.4 Divine Command Theory

Some theologians have accepted Horn 1 and its consequences. William of Ockham held that God could command us to hate God, and that would make hating God morally obligatory.¹²⁴ Modern proponents include Robert Adams (Finite and Infinite Goods, 1999), who attempts to soften Horn 1 by grounding divine commands in God’s essential nature of love.¹²⁵

The intuitive objection: if God commanded the torture of children for entertainment, would it be good? Answer “God would never command that,” and critics will say you have tacitly appealed to an independent standard of goodness. Adams and William Lane Craig escape the move by relocating the appeal: not to an independent standard, but to God’s essential nature, which is necessarily loving.

6.5 Dostoevsky’s Challenge

The phrase “If God does not exist, everything is permitted” is often attributed to Dostoevsky’s Ivan Karamazov — but it does not appear verbatim in The Brothers Karamazov. It is Sartre’s paraphrase, from Existentialism is a Humanism (lecture 1945, published 1946), of Ivan’s philosophical position.¹²⁶ What Ivan actually says is more complex and more interesting than the slogan.

In The Brothers Karamazov (1880), Ivan tells Alyosha about the suffering of children — specific, documented cases of cruelty he has been collecting from newspapers — and then poses his challenge. It is not, he insists, that he refuses to accept God; it is that he refuses to accept the world God has created. He says he wants to forgive, wants to embrace, wants no more suffering — but if the price of the eternal harmony is the tears of even one tortured child, then the price is too high, and he respectfully returns the ticket. Some readers (the most influential being the Russian theologian Sergei Hackel, and Rowan Williams in Dostoevsky: Language, Faith, and Fiction) have read Ivan’s position as a moral revolt that presupposes the God it accuses — not atheism but argument with the divine world order. Other readers take Ivan as Dostoevsky’s nearest portrait of intellectual atheism, with the ticket-returning as a rhetorical device. On either reading the problem of evil here is not first an argument that God does not exist; it is at minimum a moral accusation directed at a God who does.¹²⁷

Sartre transformed this into his existentialist claim: without God, there is no fixed human nature, no pre-given purpose, no transcendent standard. Humans must create their own values — “existence precedes essence.”¹²⁸ This is not nihilism but existential freedom — and responsibility.

6.6 Natural Law: Aquinas

Aquinas (Summa Theologiae, Q. 91, 94) offered a third path. Morality is not constituted by divine commands (Horn 1) nor entirely independent of God (Horn 2) — it is grounded in the rational nature that God has given to created beings. Natural law is accessible to human reason without direct revelation:

“Natural law is nothing other than the participation of the rational creature in the eternal law.” — Thomas Aquinas, Summa Theologiae, I-II, Q. 91, Art. 2¹²⁹

The practical content of natural law (preserve life, pursue truth, live in society) can be worked out by reason alone. Divine law (revealed in scripture) supplements, clarifies, and extends natural law.

The appeal: it grounds morality in reason accessible to all, while maintaining a connection to God. The problem: the specific moral conclusions Aquinas draws (e.g., the wrongness of contraception, the naturalness of hierarchy) are derived through philosophical reasoning that his critics do not find compelling.

6.7 Secular Alternatives

Sam Harris (The Moral Landscape, 2010) argues that morality is entirely a scientific question: “wellbeing” is a natural fact, and claims about what increases wellbeing are empirically testable. There is a landscape of possible human experiences, and some configurations of the world produce more wellbeing than others. God is not needed.¹³⁰

Harris’s critics argue that “wellbeing” is itself a value that cannot be derived from science alone. You cannot get an “ought” from an “is” — Hume’s famous is-ought gap. Even if we knew everything about human neurology and social psychology, we could not derive from those facts alone that we ought to maximise wellbeing.¹³¹

6.8 Questions to Argue About

• Does the Euthyphro dilemma definitively show that morality cannot be grounded in God’s commands? Or is there a version of divine command theory that survives?
• Dostoevsky’s Ivan says “if God does not exist, everything is permitted.” Is this a logical consequence, or can secular ethics ground moral obligations as firmly as theistic ethics?
• Aquinas says natural law is accessible to reason without revelation. If that’s true, what extra work does God do in Aquinas’s ethical framework?
• Harris says wellbeing can ground a scientific ethics. Does Hume’s is-ought gap defeat this? Or is there a way to make the move from facts to values legitimate?

7 How does religious language work — and what can it mean?

Two questions sit on top of each other: how religious language works, and how to think about religious diversity. They are not the same, but they turn on the same prior question: what kind of cognitive content do religious claims have? If “God is wise” is analogical or apophatic rather than straightforwardly descriptive, that affects what it would mean for the Christian and the Buddhist to disagree — and whether their traditions can be said to track the same thing under different concepts. The lesson works on the language first, then carries that work into the question of competing traditions.

Religious language makes extraordinary claims: that the infinite exists, that the transcendent is personal, that what lies beyond experience can be spoken of. The philosophy of language poses a sharp challenge: can such language be meaningful, and if so, how does it work?

7.1 The Logical Positivist Challenge

The Vienna Circle (Moritz Schlick, Rudolf Carnap, and others, 1920s–30s) developed the verification principle: a statement is meaningful if and only if it is either analytically true (true by definition) or empirically verifiable.¹³⁴ A.J. Ayer popularised this in Language, Truth and Logic (1936):

“No statement which refers to a ‘reality’ transcending the limits of all possible sense-experience can possibly have any literal significance.” — A.J. Ayer, Language, Truth and Logic (1936), Chapter 6¹³⁵

On this account, “God exists” is not false — it is meaningless: it makes no difference to any possible experience whether it is true or false, so it says nothing. The same applies to claims about the soul, the afterlife, divine love.

The logical positivists intended this as a decisive weapon against metaphysics of all kinds — not just theology. But the verification principle itself turned out to be difficult to state without self-destruction. Is “a statement is meaningful only if it is empirically verifiable” itself empirically verifiable? If not, it seems to fail its own test.

7.2 Analogical Language: Aquinas

Aquinas recognised that we cannot speak of God literally in the way we speak of things within the world. When we say God is “wise,” we do not mean the same thing as when we say a professor is wise: God’s wisdom is infinite, unconditioned, not acquired.

But nor is it purely equivocal (a completely different word that happens to sound the same). Aquinas argued for analogical predication: the term “wise” applies to God and to humans in related but distinct ways — there is a similarity-in-difference. God’s wisdom is the archetypal case of which human wisdom is a faint image.¹³⁶

The analogy of attribution: “healthy” is said primarily of an organism, and secondarily of food (which causes health) or of urine (which indicates health). “Good” is said primarily of God, and secondarily of created things that participate in God’s goodness.¹³⁷

7.3 Apophatic Theology

The apophatic or negative theology tradition — prominent in Pseudo-Dionysius the Areopagite (5th–6th century CE),¹³⁸ Maimonides (Guide for the Perplexed, 1190),¹³⁹ and Meister Eckhart (c. 1260–1328)¹⁴⁰ — holds that God cannot be described in positive terms at all. Every positive description limits and falsifies God. The only adequate theology is one that says what God is not.

Maimonides argued rigorously that even attributes like “powerful,” “knowing,” and “existing” apply to God only in an analogical sense so attenuated that they amount to negation: God does not lack power; God is not not-powerful; but “powerful” in the human sense does not apply.¹⁴¹

The tradition connects to mystical approaches: if God is beyond language, then union with God is also beyond language — hence James’s ineffability criterion.

A structurally similar move appears in a tradition that has no God to be silent about. Nāgārjuna’s Mūlamadhyamakakārikā (c. 2nd century CE) — the founding text of the Mādhyamaka school of Mahāyāna Buddhism — argues that all phenomena lack any independent essence (svabhāva), and that this lack — emptiness (śūnyatā) — is itself empty: not a hidden essence behind appearances, but simply the way conventional reality is.¹⁴² Mādhyamaka therefore distinguishes two truths:

“The Buddha’s teaching of the Dharma / Is based on two truths: / A truth of worldly convention / And an ultimate truth.” — Nāgārjuna, Mūlamadhyamakakārikā 24:8 (trans. Jay L. Garfield)¹⁴³

The famous next move — MMK 24:18 — identifies emptiness with dependent origination and with conventional designation: “Whatever is dependently co-arisen / That is explained to be emptiness. / That, being a dependent designation, / Is itself the middle way.”¹⁴⁴

Mādhyamaka is not apophaticism in the Christian sense — there is no transcendent referent being protected from inadequate predication. But it raises the same epistemological question more sharply: what kind of claim is “all things are empty,” given that the doctrine itself disclaims the kind of independent standing that would make it a positive description? Mādhyamaka’s answer — that the teaching is itself a conventional designation, useful as therapy against the reifying tendencies of the mind — runs parallel to Wittgenstein’s remark that his propositions are a ladder one throws away after climbing.¹⁴⁵

7.4 The Tao Te Ching

The Tao Te Ching of Laozi (6th or 4th century BCE) opens with one of the most famous sentences in world philosophy:

“The Tao that can be told is not the eternal Tao; the name that can be named is not the eternal name.” — Laozi, Tao Te Ching, Chapter 1 (translated by Gia-Fu Feng and Jane English)¹⁴⁶

The Tao (the Way, the fundamental nature of reality) is that which generates and sustains all things. But it is beyond characterisation: as soon as you name it, you have missed it. The Tao Te Ching itself is a sustained attempt to point at what cannot be directly stated — through paradox, negation, and metaphor.

7.5 The Psalms and Religious Poetry

Not all religious language aims to make propositional claims. The Psalms of the Hebrew Bible are poems of praise, lament, and petition — forms of address to God rather than descriptions of God:

“My God, my God, why hast thou forsaken me? why art thou so far from helping me, and from the words of my roaring?” — Psalm 22:1 (King James Version)¹⁴⁷

This cry of desolation is quoted by Jesus in the Gospels at the moment of crucifixion.¹⁴⁸ It is not a theodicy — it does not explain or justify. It is an expression of abandonment, addressed to the one who has abandoned. What kind of truth-claim, if any, is it making?

Wittgenstein’s later work suggests that demanding a propositional answer here is a grammatical confusion: the language has a different use from the one the question assumes. Critics object that the Psalmist plainly does assume God exists in a way that goes beyond a use-of-language; the cry would be empty if it did not. The Wittgensteinian reply, that “exists” too has a different use here, is where the dispute parks.

7.6 What Does Narrative Do That Argument Cannot?

Religious traditions are primarily narrative before they are doctrinal. The Hebrew Bible, the Qur’an, the Mahabharata, the Pali Canon — these are overwhelmingly stories, not propositions. The Genesis account of creation is not best understood as a defective cosmological hypothesis. It is a story about guilt, responsibility, the structure of desire, and the discovery of mortality. Its truth, if it has any, is not the truth of a scientific claim — it is the truth of a story that illuminates something about what it is to be human.

A question opens here that the propositional/non-propositional debate does not quite capture: what does narrative know that argument does not?

Aristotle’s Poetics argued that poetry is more philosophical than history because it shows what could happen rather than what did happen — it deals with universals through particulars.¹⁴⁹ The parable of the Prodigal Son¹⁵⁰ says something about forgiveness, self-deception, and love that a philosophical essay on forgiveness does not say — not because it contains more information, but because it works differently. The reader inhabits the story, identifies with its characters, feels the return and the reception — and in feeling it, knows something.

Paul Ricoeur (Time and Narrative, 1984–1988) argued that narrative is not a decorative addition to human understanding — it is one of its fundamental structures. We understand time, identity, and causation through narrative before we understand them through analysis. Religious narrative does not primarily make claims to be evaluated; it structures a world to be inhabited.¹⁵¹

This does not mean religious narratives are immune to criticism. It means the appropriate criticism is not simply “is this literally true?” but “what does this story make visible, and what does it conceal? Does it illuminate or distort human experience?”

7.7 From Language to Diversity: Exclusivism, Inclusivism, Pluralism

The previous sections asked how religious language works. This one asks a different question: given that the traditions speak — analogically, apophatically, narratively — what should we make of the fact that they speak differently? Aquinas’s analogy, Maimonides’ apophaticism, the Tao Te Ching, and Nāgārjuna’s two truths each say, in their own idiom, that the ultimate cannot be straightforwardly named. But they do not say the same thing about it. Brahman is not Yahweh; śūnyatā is not the Tao; the personal creator of Genesis is not the Trikāya of Mahāyāna. The traditions disagree — and they disagree about what is, by their own account, the most important thing there is. What should we make of that?

Michael Peterson and his co-authors set out the standard typology: “Philosophical approaches to religious diversity can be classified under three broad headings: exclusivism, pluralism, and inclusivism.”¹⁵²

Exclusivism holds that one tradition is salvifically true and the others mistaken (its sharpest historical form is the Catholic dogma extra ecclesiam nulla salus, “outside the Church, no salvation”). Inclusivism holds that one’s own tradition is fully true but other traditions implicitly grasp the same reality — Karl Rahner’s “anonymous Christians” is the canonical formulation. Pluralism, in John Hick’s developed version, treats the traditions as diverse human responses to a single transcendent reality, and grounds that reading in an empirical claim:

“I maintain that so far as we can tell this salvific transformation is taking place — and also failing to take place — to more or less the same extent within each of the great world faiths. There is no one religion whose adherents stand out as morally and spiritually superior to the rest of the human race.” — John Hick, An Interpretation of Religion (2nd ed. 2004), Introduction §4 (“The religious criterion”)¹⁵³

Hick borrows from Kant’s noumenon/phenomenon distinction: there is the Real-an-sich (the Real as it is in itself), and the Real as it is humanly and culturally experienced — the personal God of the Abrahamic traditions, the impersonal Brahman of Advaita Vedānta, the śūnyatā of Mādhyamaka. The traditions describe the phenomenal Real, conditioned by their concepts; none has direct access to the Real-an-sich. The illustration — already old in Buddhist literature, in Udāna 6.4 — is the blind men and the elephant: each touches a real part of one animal and reports honestly, but no one sees the whole.¹⁵⁴

Pluralism is contested from both sides. The exclusivist replies that Hick’s hypothesis is itself a substantive metaphysical claim — that there is a Real-an-sich, that all traditions track it, that none does so privileged — and is no less a contested first-order religious view than the ones it claims to harmonise.

William Alston’s parity argument cuts in the opposite direction: if the Christian’s experience of God is rationally credible to her, then by parity the Muslim’s, the Hindu’s, and the Buddhist’s experiences are rationally credible to them — but parity does not yield agreement, only symmetric standoff, and the question of who, if anyone, is right remains open.¹⁵⁵ Pluralism, exclusivism, and inclusivism are alike confronted by the question James called forced: one cannot indefinitely suspend.

7.8 Questions to Argue About

• Ayer’s verification principle says religious language is meaningless. But the principle itself seems difficult to verify. Does this refute logical positivism, or just show that meaningfulness is more complex than the positivists thought?
• Aquinas’s analogical language says we can speak of God by way of similarity-in-difference. Is this different in principle from the way we speak of anything we don’t directly experience — subatomic particles, for instance?
• Apophatic theology says we can only say what God is not. Is negative theology coherent? Can you have a meaningful concept constituted entirely by negations?
• The Tao Te Ching says the Tao cannot be named — then spends 81 chapters pointing at it. Is this paradoxical or profoundly appropriate? What does it suggest about the relationship between language and what lies beyond it?
• Hick’s pluralism says all the great traditions track the same Real-an-sich under different concepts. Is this a generous synthesis or a sneaky exclusivism — a meta-religion that claims to know what the first-order traditions only partially see?

8 Is religion a social construction — or does it track something real?

The sociology of religion offers a deflationary account: religious beliefs and practices are products of social processes, serving social functions. But does explaining the origin of religion explain it away? Can something be both socially constructed and true?

8.1 Durkheim’s Social Account

Émile Durkheim’s The Elementary Forms of Religious Life (1912) is the foundational text of the sociology of religion.¹⁵⁸ Durkheim studied the religion of Australian Aboriginal peoples — the arunta — and proposed a general theory:

The fundamental distinction of religion is between the sacred and the profane. Sacred things are not necessarily supernatural; they are set apart, surrounded by prohibitions, treated as an entirely different order from the ordinary (profane) world.¹⁵⁹

But what is the source of the sacred? Durkheim’s radical proposal: society itself. When people perform collective rituals — corroborees, ceremonies, festivals — they generate a collective emotional energy (what he called effervescence) that they experience as overwhelming and external. On Durkheim’s account this experience is the experience of the collective, projected onto totems, ancestors, gods.¹⁶⁰

“The god of the clan… can be nothing other than the clan itself, but the clan transfigured and imagined in the physical form of the plant or animal which serves as totem.” — Émile Durkheim, The Elementary Forms of Religious Life (1912), Book II, Chapter VII¹⁶¹

Durkheim’s framing is not that religion is straightforwardly false. On his reading it is true in a different sense: religion is a real social force; it really binds communities; its symbols really represent the group’s values. What religion-as-it-understands-itself gets wrong, on his account, is its theory of itself — it mistakes social power for supernatural power. Whether that mistake-thesis is itself the right reading of religious experience is what defenders of religion (and many anthropologists since) have contested.

8.2 Marx: Opium of the People

Marx’s account is harsher. In the introduction to A Contribution to the Critique of Hegel’s Philosophy of Right (1844):

“Religion is the sigh of the oppressed creature, the heart of a heartless world, and the soul of soulless conditions. It is the opium of the people.” — Karl Marx, introduction to A Contribution to the Critique of Hegel’s Philosophy of Right (1844)¹⁶²

The “opium” metaphor is sometimes read as purely dismissive, but Marx’s formulation is sympathetic as well as critical. Religion is a genuine response to genuine suffering — it is how people who cannot change their material conditions make those conditions bearable. The critique is systemic: religion perpetuates oppression by redirecting energy that could transform social conditions toward supernatural consolation.

The practical consequence of Marxist analysis: to criticise religion without criticising the social conditions that make religion necessary is superficial. The target is the conditions, not the symptom.

8.3 Weber: Religion and Capitalism

Max Weber’s The Protestant Ethic and the Spirit of Capitalism (1905) reverses the Marx-Engels picture.¹⁶³ Rather than the economic base determining the ideological superstructure (including religion), Weber argued that religious ideas could independently shape economic behaviour.

Calvinist theology — with its doctrine of predestination (God has already determined who will be saved) and the associated anxiety about whether one is among the elect — produced a characteristic response: intense, disciplined, worldly activity as a sign of election. Hard work, frugality, and reinvestment of profits became indicators of God’s favour. This, Weber argued, provided the cultural preconditions for industrial capitalism.¹⁶⁴

The argument is not that Calvinism caused capitalism (a simplification Weber explicitly warned against), but that the relationship between religious ideas and economic structures is not simply one of determination from below.

8.4 The Saffron Revolution

In August–September 2007, Myanmar’s Buddhist monks led mass protests against the military junta of Senior General Than Shwe, marching through the streets of Rangoon in their saffron robes. The trigger was an overnight five-fold rise in fuel prices in mid-August; the protests escalated after security forces beat monks in Pakokku on 5 September.¹⁶⁵ The monks’ weapon was the ritual of patam nikkujjana kamma — formally overturning the alms bowl — which refused the military’s offerings and thereby excluded the junta from the merit-economy that gave a Buddhist regime its legitimacy.¹⁶⁶

This was simultaneously a political act and a religious one. The military junta, which claimed Buddhist legitimacy, found itself spiritually condemned by the custodians of Buddhist authority. The monks were using the internal logic of the religious system to make a political claim — and the claim had force because the system was real, not merely because the monks had political opinions.

The Saffron Revolution illustrates Durkheim’s point in a way Durkheim did not anticipate: the social force of religion can be directed against the social structure that produced it.

8.5 The Evolutionary Cognitive Science of Religion

Durkheim, Marx, and Weber all offer social accounts of religion. A different kind of naturalistic account comes from evolutionary cognitive science: why does religion appear in every known human culture? What cognitive architecture makes it not just possible but apparently natural?

Pascal Boyer (Religion Explained, 2001) and Scott Atran (In Gods We Trust, 2002) argue that religious concepts are “minimally counterintuitive”: they violate just enough ordinary expectations to be attention-grabbing and memorable, while remaining close enough to intuitive ontologies to be easily understood.¹⁶⁷ A god who is invisible but can hear your thoughts, who exists outside time but intervenes in history, who is all-powerful but responds to petitions — this is a conceptual structure that is maximally memorable because it violates expectations (omniscience, immateriality) while operating through familiar social categories (listening, responding, caring).

On this view, religion is not an anomaly requiring special explanation — it is the expected output of cognitive systems that evolved for other purposes (agent detection, social cognition, narrative cognition) operating in conditions where there is genuine uncertainty about agency, causation, and the social world.

This does not debunk religion. It is not an argument that religious claims are false — only that they are cognitively “sticky” in ways that explain their universality. But it does pose a sharp question: if religious concepts are universal partly because of cognitive biases that evolution selected for in ancestral environments, does this count as evidence for or against their truth? Compare: our tendency to believe there are physical objects that persist when unobserved is also a product of evolution. Most of us think it is also true.

8.6 Can It Be Both Constructed and True?

The philosophical question underlying these accounts: does explaining the social origins of religious belief show that religious belief is false?

Compare: our beliefs about mathematics are socially conditioned — shaped by education, cultural tradition, the history of the discipline. Does this mean mathematical truths are merely social constructions? Most people would say no: the social processes by which we come to believe mathematical truths do not determine whether those truths are true.

The sociologist of religion can show how religious beliefs arise and what social functions they serve without thereby settling whether those beliefs are true. The origin of a belief and its truth are separate questions.

But the constructivist can press: religion is not like mathematics. Mathematical claims are intersubjectively testable in ways that religious claims are not. The cross-cultural diversity of religious belief — with incompatible theologies sincerely held by different cultures — is, on the constructivist reading, better explained by social construction than by truth-tracking.

The realist replies that cross-cultural diversity is also a feature of moral and aesthetic convictions, where most people hesitate to conclude there is no fact of the matter; and that disagreement among traditions about how to describe the ultimate is compatible with there being an ultimate to describe (Hick’s pluralist move, treated in the language-and-diversity lesson). The dispute is not settled.

8.7 Questions to Argue About

• Durkheim says religion is society worshipping itself, but this does not mean religion is false. Is this a satisfying account? What would it mean for religious practices to be meaningful if the god worshipped is actually society?
• Marx says religion is “the heart of a heartless world.” Is this a sympathetic account of religion? Does it change how you evaluate Marxist criticism of religion?
• Weber argues that Calvinist theology helped produce capitalism. Does this mean the way people understood God had real historical consequences independent of economics? What follows for the relationship between ideas and material reality?
• The Saffron Revolution shows monks using religious authority to challenge political authority. Does this challenge Durkheim’s account (religion as social glue that reinforces existing structures), or is it compatible with it?
• Can something be both socially constructed and true? Give examples from outside religion, then return to the religious case.

9 Media

These novels, films, and artworks illuminate the problems of the Religion unit — not by teaching doctrine but by dramatising the difficulty of living with and without religious conviction.

• Fyodor Dostoevsky, The Brothers Karamazov (1880) — The great novel of religious doubt. Ivan Karamazov’s “rebellion” — his rejection of God’s world on the basis of children’s suffering — is the most honest statement of the atheist case in world literature. Father Zosima’s counter-witness is equally serious. The Grand Inquisitor chapter is one of the indispensable texts of Western thought.
• Shūsaku Endō, Silence (1966) — A 17th-century Portuguese Jesuit missionary is forced by the Japanese inquisition to apostatise — to trample on an image of Christ. The novel is about the silence of God in the face of suffering, the nature of faith under extreme pressure, and what it means for God to speak through apparent defeat. One of the greatest religious novels of the 20th century.
• Ingmar Bergman (dir.), The Seventh Seal (1957) — A knight returning from the Crusades plays chess with Death and struggles to find evidence that God exists. Bergman’s film is the definitive cinematic treatment of the absence of God. Shot in 35 days on a small budget; philosophically serious in a way few films manage.
• Marilynne Robinson, Gilead (2004) — A Congregationalist minister in Iowa writes letters to his young son, knowing he will die before the boy grows up. A novel about Calvinist theology, doubt, grace, and what it means to live in a world that may or may not be watched by God. The most philosophically engaged religious novel of recent decades.
• Graham Greene, The Power and the Glory (1940) — A whisky priest — cowardly, sinful, alcoholic — is the last Catholic priest in a Mexican state that has outlawed religion. The novel asks whether holiness can inhere in an apparently unholy person, and whether institutional religion can be destroyed. Greene’s Catholicism gives him access to questions that secular novelists cannot reach.
• Andrei Rublev’s Trinity (c. 1411) — A Russian icon depicting the three angels who visited Abraham (Genesis 18). The three figures face each other across a shared meal. The icon is a visual theology: it depicts the divine not as power but as relationship, not as a single overwhelming presence but as a community of three looking toward each other. Theologically and aesthetically extraordinary.

10 Bibliography

11 Notes