The Exemption That Holds Everything Together
There is a move in theological argument that appears so frequently, and in so many different guises, that most people who make it have long since stopped noticing they are making it at all. It goes like this: you establish a principle that applies universally, and then, at the precise moment that principle threatens to undermine your central claim, you announce that the central claim is exempt from it. Everything in the universe requires a cause, except God. Evidence of absence is not absence of evidence, except when the absence in question is God’s. A being of perfect love would not allow suffering without purpose, except when the being in question is God and the purpose is inscrutable. The principle is applied with rigour to the entire cosmos, and then suspended, without argument, the moment it points at the thing you most need it not to reach.
Logicians call this special pleading. In formal terms, it is the fallacy of applying a principle selectively, exempting a case from a rule without providing a principled reason for the exemption. It is one of the oldest identified fallacies, and one of the easiest to recognise once you have been shown what you are looking for. In theological discourse, however, it has an unusual status: it is not a peripheral error committed occasionally by careless apologists. It is the load-bearing wall. Remove it and the entire structure comes down, because there is no version of the classical theistic argument that does not, at its critical juncture, require God to be placed beyond the standards demanded of every other claim in every other domain of inquiry.
That is the thesis of this piece, and it will be developed carefully, because the claim deserves more than assertion. The argument will proceed through the three most important theological domains in which special pleading operates: the cosmological argument, the hiddenness argument, and the problem of evil. It will then return to the structural question of whether a concept that can only be defended by suspending the rules of evidence and inference is a concept doing any genuine intellectual work at all. The connection between this question and the burden of proof matters enormously, and readers who want the fuller account of how that burden operates should consult the earlier piece on the proof that never comes. The analysis here, however, stands on its own.
What Special Pleading Actually Is
Before examining the theological instances, it is worth being precise about the fallacy itself, because imprecision here allows the apologist a convenient escape route. Special pleading is not the same as making an exception. Exceptions are perfectly legitimate, provided there is a principled, independently justifiable reason for them. We do not apply criminal law to children under a certain age in the same way we apply it to adults, and this is not special pleading because there is an independently verifiable reason, rooted in cognitive development, moral understanding, and neurological maturation, that justifies the different treatment. The reason does not derive from wanting to protect children from legal consequences; it derives from observable facts about what children are and how they reason.
Special pleading occurs when the exemption is granted simply because the claimant wants it granted, and the only justification offered for the exemption is that the exempted thing is special. If a defendant argued in court that the laws of evidence should not apply to the charges against him because he is a particularly important person, that would be special pleading. The importance of the person is not an independently grounded reason to suspend evidential standards; it is just a restatement of the desire to avoid those standards, dressed up as a reason.
This is precisely the structure of theological exemption. When an apologist argues that God does not require a cause because God is, by definition, the uncaused first cause, they are not providing an independently grounded reason for the exemption. They are inserting the desired exemption into the definition of the thing being examined, which is a circular move, and then presenting the circular definition as though it were a substantive argument. The exemption is justified by the specialness of God, and the specialness of God is partly constituted by the exemption. The wall holds itself up by leaning against itself.
Arthur Schopenhauer put the philosophical principle with characteristic precision: “Philosophy is a science, and as such has no articles of faith; accordingly, in it nothing can be assumed as existing except what is either positively given empirically, or demonstrated through indubitable conclusions.” Theology, when it engages with philosophical argument at all, routinely violates this constraint by treating the existence and nature of God as an article of faith that the argument is then deployed to support, rather than as a conclusion that the argument independently establishes. Special pleading is what happens at every point where the argument would otherwise fail to close.
It is also worth noting what distinguishes this fallacy from adjacent forms of bad reasoning. The appeal to ignorance says that something must be true because it has not been proved false. The argument from personal incredulity says that something must be false because the speaker cannot imagine how it could be true. Special pleading is more architecturally fundamental than either: it does not merely mishandle evidence or fail to engage the imagination. It constructs a two-tier system of epistemic standards, one for the central claim and one for everything else, and insists that this asymmetry is justified without providing an asymmetric reason. Identifying it matters precisely because it is harder to see than the cruder fallacies, and because it operates at the level of argument structure rather than at the level of individual claims.
First Exemption: The Uncaused Cause
The cosmological argument is the oldest and most persistent argument for the existence of God, and it exists in many forms: the Kalam version, associated with al-Ghazali and more recently with William Lane Craig; the Leibnizian version, concerned with sufficient reason; the Thomistic version, concerned with the impossibility of an infinite regress of movers. Despite their differences, all versions share a common structure, and all versions arrive at the same junction where special pleading becomes structurally necessary.
The argument begins with an observation that carries genuine intuitive force: everything we have ever encountered in the universe has a cause, or at minimum a sufficient explanation. Nothing appears to bring itself into existence from nothing. If you trace any causal chain backward, you seem to be committed to one of two uncomfortable conclusions: either the chain is infinite, stretching backward without terminus, or it terminates in something that was not itself caused. The argument then proposes that this terminating thing is God, specifically the God of classical monotheism, complete with all the relevant attributes of omnipotence, omniscience, and moral perfection.
The special pleading arrives immediately and unmistakably. The argument begins with the premise that everything requires a cause. It then concludes that there is a thing, namely God, which requires no cause. If the questioner asks why God is exempt from the principle that opened the argument, the apologist has precisely two options, and both reveal the problem. The first option is to say that the principle was never “everything requires a cause” but rather “everything that begins to exist requires a cause.” This reformulation is the Kalam version, and it attempts to ground the exemption by claiming that God is eternal and therefore never began to exist. But this simply relocates the special pleading rather than resolving it. The claim that God is eternal is not an independently established fact; it is a theological stipulation that serves precisely the purpose of avoiding the demand for a cause. The eternal attribute is not derived from evidence; it is assigned in order to generate the exemption.
The second option is to say that God is, by definition, a necessary being: a being whose non-existence is impossible, which therefore requires no external explanation. This is the Leibnizian move, and it is philosophically more sophisticated, but the special pleading is no less present. The concept of a necessary being is not established through observation or through logic that is independent of the desired conclusion. It is a philosophical category created specifically to house God, and the claim that God occupies that category is, once again, circular. We are told that God does not require explanation because God is the kind of thing that requires no explanation, and we are told that God is the kind of thing that requires no explanation because that is what God is. The definition does not escape the circle; it is the circle.
Notice what the cosmological argument never does. It never considers whether the universe itself might be the kind of thing that requires no external cause. Quantum mechanics has, since the early twentieth century, provided a robust framework in which events occur without prior classical causes, confirmed by experimental results of extraordinary precision. The notion that the universe might be self-sufficient, or that causation as we understand it breaks down at or near a Big Bang singularity, is not an eccentric speculation but a live scientific possibility actively investigated by physicists including Stephen Hawking, James Hartle, and Alexander Vilenkin, each of whom has proposed models in which the universe has no beginning in the relevant sense or requires no prior cause in the everyday sense. The apologist dismisses these possibilities without extended engagement, because engaging with them seriously would mean applying to the universe the same analytical generosity applied to God. The universe gets no exemptions. Only God does.
This is not a minor logical footnote. It is the entire structure of the argument laid bare. The cosmological argument works by establishing a universal principle, applying that principle to produce a gap, and then filling the gap with a concept that has been pre-equipped with an exemption from the principle that generated the gap. The exemption is not argued for; it is built into the definition of the gap-filler. And then the gap-filler is presented as a conclusion requiring our assent. This is not inference; it is verbal sleight of hand, and recognising it as special pleading dissolves the argument entirely, not as an act of dogmatic dismissal, but as a straightforward application of the standards the argument itself invokes.
There is a further wrinkle worth noting. Even if the cosmological argument succeeded in establishing that there is an uncaused first cause, it would not thereby establish the existence of the God of classical monotheism. The uncaused cause could, on the argument’s own terms, be anything: a mindless quantum field, a brute physical fact, a fluctuation in a prior void. The leap from “there is something that requires no cause” to “that something is a personal, omniscient, omnibenevolent being who takes an interest in human affairs” is an enormous one, and it is made by assumption rather than argument. The cosmological argument, at its very best, establishes a gap in our understanding. It then smuggles God into the gap by definition, and calls the smuggling a proof.
Second Exemption: The Hidden God
The problem of divine hiddenness is one of the most philosophically serious challenges to theism, and it deserves to be developed carefully before the special pleading response is examined. The argument runs as follows. If God exists, and if God is a person in the morally relevant sense, specifically a being capable of love, of communication, of relationship, then we would expect God to make his existence known in some reasonably clear way, particularly to those who are genuinely open to relationship with him. An attentive parent does not hide from a distressed child who is searching for them. A loving friend does not allow a relationship to lapse into silence while the other party calls out and receives nothing. The persistent non-existence of clear, universally accessible evidence for God’s existence is therefore exactly the kind of state of affairs we would expect if there were no God, and not at all the kind we would expect if there were a loving personal God.
The philosopher J. L. Schellenberg has developed this argument with considerable rigour across a body of work spanning several decades, and it has proven genuinely difficult to answer. The standard apologist responses are instructive precisely because they all involve granting an exemption that no other claim would receive. The first response is that God hides in order to preserve human freedom: if God’s existence were obvious, people would believe by compulsion rather than by free choice, and the freely chosen relationship has greater value than the compelled one. This is a coherent enough idea in the abstract, but it immediately invites a question that the exemption cannot answer: why does God apply this principle to himself and not to everything else whose existence bears on human belief? The laws of physics are obvious. The existence of other people is obvious. The historical record, such as it is, is accessible. None of these facts are thought to compromise human freedom, even though they are things people believe by overwhelming evidence rather than by the kind of ambiguous faith apparently required for God. The hiddenness exemption applies to God alone, which means it is not a principled account of the relationship between evidence and freedom but an ad hoc rationale constructed to explain away a problem that arises specifically and only because God is the subject.
The second response is that hiddenness is itself a form of communication: God is present in the still small voice, in the sense of wonder, in the moments of inexplicable peace. This response moves the goalposts rather than answering the question. The problem of hiddenness is not that people never have subjective religious experiences; plainly they do, and those experiences are real as psychological events. The problem is that those experiences are not distributed in a way that distinguishes between a world with a loving personal God and a world without one. People have intense experiences of divine presence across every religion and across mutually contradictory theological systems. They also have equally intense experiences that are explained by neuroscience, psychology, and the well-documented human tendency to find patterns and agents in ambiguous stimuli. If the evidence for God’s presence is indistinguishable from the evidence we would expect in God’s absence, then citing that evidence as proof of presence is a form of special pleading: the experience counts as evidence for God, but the equal and opposite absence of experience does not count as evidence against him.
Consider how any other empirical claim would fare if we applied the same standard. Suppose someone claimed that a particular medical treatment was effective, and that evidence of improvement counted as confirmation while evidence of no improvement was reinterpreted as the treatment working in subtle ways not visible on standard metrics. We would correctly identify this as a failure of the evidential standard required of medical claims. The treatment gets no exemption from the requirement to produce results that differ from what we would observe in its absence. God, according to the hiddenness defence, is permanently exempt from this requirement. His presence explains the religious experiences people have, and his presence also explains the absence of religious experiences that others have, because hiddenness is itself a divine communication strategy. An explanatory framework that can accommodate both the presence and the absence of its predicted effects is not providing evidence; it is demonstrating its own unfalsifiability.
This brings the question of unfalsifiability into direct view, and it is a question that will run through the rest of this analysis. A claim that cannot in principle be falsified is not a claim that has survived testing; it is a claim that has been removed from the domain of testing altogether. The hiddenness exemption achieves exactly this removal: whatever the evidential state of the world, the absence of clear evidence for God can be explained by divine hiddenness, and the presence of ambiguous experiential evidence can be explained by divine communication through subtlety. The concept accommodates any possible state of the evidence, which means it is not in contact with the evidence at all.
Russell’s Teapot and the Invisible Dragon: What These Analogies Actually Demonstrate
Bertrand Russell proposed, in an essay written in 1952 for a magazine that declined to publish it, a now-famous analogy. Suppose he were to assert that there is a china teapot orbiting the sun in an elliptical path between Earth and Mars, too small to be detected by even the most powerful telescope. He could not be disproved. No observatory could scan the solar system finely enough to confirm the teapot’s absence. Should we therefore consider the teapot’s existence a live possibility deserving of serious consideration? Russell’s point was that the burden of proof rests with the one making the positive claim, not with those declining to accept it. The impossibility of disproof is not evidence of existence. This piece links to the fuller treatment of the burden-of-proof question for readers who want the complete argument, but the connection to special pleading is worth drawing out here in its own right.
The teapot analogy is not simply about burden of proof. It is about what special pleading looks like when applied to a claim that has been deliberately insulated against evidence. The teapot’s undetectability is not a feature of the teapot that demands our credulity; it is a feature of the claim that marks it as unfalsifiable, and therefore as making no genuine contact with reality. When the apologist argues that God’s hiddenness is compatible with his existence and his love, and that the absence of evidence is not evidence of absence, they are making precisely the teapot move: they are insisting that their claim be taken seriously while simultaneously removing it from the domain in which claims are tested. The special pleading is the insulation, and the insulation is the problem.
Carl Sagan’s invisible dragon in the garage, developed in “The Demon-Haunted World,” makes the same point with greater granularity and is worth examining at some length. Suppose Sagan’s neighbour claims to have a dragon in their garage. Sagan asks to see it. The neighbour says it is invisible. Sagan proposes to pour flour on the garage floor to detect footprints. The neighbour says the dragon floats in the air. Sagan proposes infrared sensors to detect the dragon’s heat. The neighbour says the dragon is heatless. At every point where an evidential test is proposed, the dragon acquires a new property that exempts it from that test. The result is a concept of a dragon that has been progressively redefined to escape every possible form of detection, and the question Sagan asks is direct and devastating: what is the difference between this dragon and no dragon at all?
Jerry Coyne, drawing on Sagan’s analogy, put the epistemic conclusion with the directness the question deserves: “Putting all this together, we see that religion is like Sagan’s invisible dragon. The missing evidence for any god is simply too glaring, and the special pleading too unconvincing, to make its existence anything more than a logical possibility. It’s reasonable to conclude, provisionally but confidently, that the absence of evidence for God is indeed evidence for his absence.” This is a carefully calibrated statement and it rewards unpacking. Coyne is not claiming certainty; he is claiming that the repeated deployment of special pleading to explain away the absence of evidence changes the epistemic status of the claim. A claim that survives only because it has been exempted from every test is not a claim that has been confirmed by those tests; it is a claim that has been made immune to them, and an immune claim is one that no longer functions as a genuine description of reality.
The dragon analogy also illuminates the specific mechanism of theological special pleading in a way that the abstract description does not. The dragon acquires its exemptions not through evidence about dragons but through the argumentative necessity of avoiding falsification. The same is true of God’s attributes. God’s omniscience, omnipotence, omnibenevolence, and incorporeality are not properties derived from observation; they are properties assigned because the alternative, a limited and therefore testable deity, would be far more vulnerable to the evidence we actually have. An omnipotent God can do anything and therefore cannot be refuted by pointing to things that are not done. An omniscient God knows everything and therefore cannot be refuted by pointing to apparent ignorance. An incorporeal God has no physical presence and therefore cannot be refuted by the failure to detect one. The classical attributes of God are, viewed from this angle, not descriptions of a discovered being but a set of exemption certificates, each one protecting the concept from a different line of evidential attack.
The theological tradition has, over centuries, refined these exemptions to a remarkable degree of sophistication, and the sophistication should not be mistaken for substance. A concept that has been insulated against evidence by a more elaborate set of moves is not thereby confirmed; it is merely more elaborately insulated. The dragon with an additional property that escapes an additional test is not more real than the dragon with fewer properties; it is more carefully protected against investigation, which is a different thing entirely. Sophistication in the service of avoidance is still avoidance.
Third Exemption: The Problem of Evil and the Inscrutable Plan
The problem of evil is the most emotionally powerful and the most philosophically persistent objection to classical theism, and the theological response to it represents perhaps the most structurally significant instance of special pleading in the entire tradition. The problem is this: a being of perfect power, perfect knowledge, and perfect goodness would have both the capacity and the motivation to prevent gratuitous suffering. Gratuitous suffering exists in enormous quantity. Therefore, a being of perfect power, perfect knowledge, and perfect goodness does not exist. The argument has been refined and extended by philosophers including Mackie, Rowe, Draper, and Schellenberg, and in its evidential rather than logical form it is now widely considered by professional philosophers to be a serious problem for classical theism rather than a merely rhetorical challenge.
The theodicy tradition represents theology’s sustained attempt to dissolve the problem, and the special pleading at its core is both obvious and instructive. The most commonly invoked response is the free will defence: suffering exists because God respects human freedom, and free beings inevitably produce suffering. The problems with this response are well rehearsed. It accounts only for moral evil, the suffering caused by human choices, and leaves natural evil entirely unaddressed. The child born with a degenerative disease, the earthquake that buries a village, the tumour that develops in the body of a person who has made no morally significant choice: none of these are produced by human freedom, and the free will defence has nothing to say about them. The apologist then reaches for a different tool: soul-making theodicy, most associated with John Hick, which argues that suffering is necessary for human moral and spiritual development.
The special pleading in the soul-making theodicy is worth examining with some care, because it is embedded in what appears to be a coherent developmental argument. Hick’s claim is that a world without adversity would produce morally shallow beings, and that genuine virtue requires genuine temptation and genuine suffering. This is not entirely implausible as an account of human moral psychology: there is something to the observation that difficulty can produce character, that compassion often develops through shared suffering, that a life of unbroken comfort might not develop the moral seriousness we most admire. The problem is not with the observation but with what is required of God to make the theodicy work.
If God is omnipotent, then God can produce beings who are morally mature without requiring them to suffer the death of children from preventable diseases, the slow destruction of dementia, the systematic cruelty of genocide. The soul-making argument implicitly assumes a limitation on God’s capacity to achieve developmental ends without the means currently employed. But an omnipotent God has no such limitation by definition. The apologist is therefore simultaneously maintaining that God is omnipotent and that God cannot achieve morally significant human development without the current level and distribution of suffering. These claims are in direct tension, and the tension is resolved only by granting God an exemption: an omnipotent being, in the theological account, is apparently constrained by the requirements of soul-making in a way that a genuinely omnipotent being would not be. The exemption is from the logical consequences of the attribute the tradition itself insists upon, and the insistence upon the attribute and the exemption from its consequences operate simultaneously without any acknowledgement of the contradiction.
The deeper and more general version of this exemption is the appeal to inscrutability: God’s ways are higher than our ways, his plans are beyond human comprehension, and what appears to us as gratuitous suffering may serve a purpose visible only from the divine perspective. This response is theologically ancient and philosophically familiar, and it is the purest form of special pleading in the theodicy tradition. No other claim in any other domain of inquiry is permitted to invoke inscrutable purposes as a defence against the evidential challenge posed by its own predictions. If a pharmacologist said that a drug appears to be killing patients but that the mechanism by which it is actually helping them is beyond ordinary understanding, we would not accept this as a satisfactory response to the clinical evidence. We would demand a comprehensible account of the mechanism or withdraw the claim. God alone is permitted to operate by mechanisms that permanently exceed understanding, and then to have this inscrutability cited not as a problem for the claim but as a feature of the claimant, a mark of divine transcendence rather than a failure of theological explanation.
The smug certainty that can accompany this move deserves its own treatment, and readers interested in the psychology of that certainty should look at the earlier post on the smug certainty of the theist. For the purposes of this argument, the critical point is structural: the appeal to inscrutability is not a solution to the problem of evil but an exemption from the requirement of solving it. Once the exemption is granted, the problem of evil ceases to be a problem not because it has been resolved but because the concept of God has been rendered immune to it. This is not theological sophistication; it is intellectual avoidance wearing the costume of humility, and it is important to say so plainly rather than to treat the ancient dressing as a mark of profundity.
There is a further dimension to the problem of evil that the theodicy tradition consistently underweights, and it concerns the scale and distribution of suffering rather than merely its existence. If suffering serves a purpose of soul-making or divine pedagogy, we would expect it to be distributed in proportion to the developmental needs of those who suffer. We would not expect the most excruciating suffering to fall on infants who cannot learn from it, on the victims of dementia who cannot retain it, or on the inhabitants of regions whose histories have produced the least moral improvement. The distribution of suffering across the world looks nothing like the distribution we would expect from a being of perfect knowledge and perfect purpose. It looks rather like the distribution we would expect from a universe indifferent to human welfare, operating by physical processes that have no investment in outcomes. The theodicy tradition does not engage seriously with this distributional problem, because engaging with it seriously would require conceding that the evidence looks far more consistent with the absence of divine oversight than with its presence. The exemption from distributional accountability is granted silently, which is the worst kind of special pleading: the kind that is not even acknowledged as such.
The Pattern Underneath All Three Exemptions
Step back from the three domains examined above and the structural pattern becomes unmistakable. In each case, there is a standard that would, if applied consistently, undermine the theistic claim. In each case, that standard is applied rigorously to everything except the theistic claim. And in each case, the exemption for the theistic claim is justified not by an independently grounded reason but by some feature of God that has been assigned precisely in order to generate the exemption. The uncaused cause gets no cause because it is defined as the kind of thing that needs none. The hidden God’s hiddenness is not evidence against him because God is defined as the kind of being whose absence is compatible with his presence. The suffering-permitting God is not refuted by suffering because God is defined as the kind of being whose purposes exceed our comprehension. In each instance, the definition does the work that the argument cannot do honestly, and the definition is not established independently but is constructed for this purpose.
This is why special pleading is not merely a common feature of theological argument but a structurally necessary one. The god of classical theism is not a hypothesis that has survived evidential testing; it is a concept that has been progressively elaborated to avoid it. Every attribute assigned to the classical God serves a protective function. Omnipotence insulates the concept from arguments about what God could not achieve. Omniscience insulates it from arguments about what God would not know. Omnibenevolence insulates it from arguments about what God would not permit if capable of preventing it. Incorporeality insulates it from physical detection. Timelessness insulates it from arguments about causal history. The result is a concept that is internally consistent in the sense that its various exemptions do not visibly contradict one another, but whose internal consistency is purchased at the cost of any genuine contact with the world the concept is supposed to explain.
Compare this to the way in which productive explanatory concepts behave in other domains. A good scientific hypothesis makes specific predictions, predictions that could in principle be falsified, and the accumulation of confirmed predictions is what gives the hypothesis evidential support. The concept of natural selection predicted that transitional fossil forms would be found, and they were found in abundance. It predicted that the genetic code would reveal patterns of relatedness consistent with the morphological evidence, and it did so with a precision that astonished even its advocates. It predicted that artificial selection over short timescales would produce the kind of variation consistent with natural selection over long ones, and experiment confirmed this repeatedly. At every point, natural selection was making contact with evidence that could in principle have gone the other way. No such predictions are available from classical theism, because the classical attributes have been chosen to ensure that the evidence cannot go the wrong way. This is what unfalsifiability looks like when it is systematic rather than accidental, and it is what special pleading produces at scale.
Responding to the Apologist’s Defence
Several objections to this analysis are standard enough to deserve direct engagement, and engaging them honestly is more useful than ignoring them.
The first objection is that God is, by nature, a unique being, and that unique beings may legitimately receive treatment different from ordinary ones. This is true in principle but question-begging in practice. The uniqueness of God is not independently established; it is part of the claim under examination. To say that God deserves special treatment because God is unique is to assume that God exists with the relevant properties in order to justify the exemption that protects the claim that God exists with the relevant properties. This is circular in the same way that the cosmological argument’s appeal to necessary being is circular, and it is detected by the same diagnostic question: would any evidence count against the claim? If the answer is no, the circularity is complete and no genuine argument has been made.
The second objection is that all explanatory frameworks rely on some basic postulates that are not themselves derived from prior evidence, and that holding theology to a different standard in this respect is itself a form of special pleading against theology. This is a more interesting objection, because it has some genuine content. It is true that scientific reasoning relies on foundational commitments, such as the reliability of sense experience, the uniformity of natural law, and the validity of inductive inference, that cannot themselves be proved without circularity. The philosophical tradition of foundationalism has explored these questions with considerable sophistication, and the objection cannot simply be dismissed.
The response, however, is that there is a decisive difference between foundational commitments and theological exemptions. The foundational commitments of scientific reasoning are minimal, intersubjectively accessible, and pragmatically validated: they are the commitments required to do anything at all, and their validity is evidenced by the extraordinary success of the practices that rest on them. The theological exemptions, by contrast, are not minimal; they are elaborate and specifically tailored to a particular claim. They are not intersubjectively accessible; they are available only to those who already accept the tradition. And they are not pragmatically validated; the practices that rest on them do not reliably produce the outcomes they predict. The difference between a foundational commitment and a special plea is precisely the difference between a bedrock assumption required for any inquiry whatsoever and an ad hoc exemption designed to protect a specific conclusion from a specific line of objection.
The third objection is that science itself uses models and idealisations that do not correspond to observable reality: frictionless planes, perfect spheres, point masses. These are exemptions from observable conditions granted for explanatory convenience, and they are not criticised as special pleading. The response here is that the idealisation in scientific modelling is explicitly recognised as such, is granted for specific, acknowledged purposes of tractability, and is always understood to be correctable in light of observations that reveal its limitations. The frictionless plane is an idealisation that the physicist knows will require correction when applied to real surfaces, and the correction is built into the methodology from the outset. The theological exemption is not offered as an idealisation to be corrected; it is offered as a permanent feature of the concept that places it beyond correction. The difference is between acknowledged simplification and systematic insulation, and only the latter is special pleading.
Readers wanting to see how these patterns of evasion appear across a broader range of theological fallacies will find the fuller catalogue at the religious fallacies debunked post, which covers the argumentative landscape with more breadth than is possible here. The chess-with-a-pigeon problem, relevant to understanding why these exemptions are so difficult to contest in direct debate, is treated separately in the piece on evidence versus faith.
Why This Matters Beyond the Seminar Room
One objection to this entire line of inquiry is that it is merely academic: a logical point about argument structure that makes no practical difference to whether people live well or badly. This objection deserves a direct answer, because the stakes of special pleading in religion extend well beyond the philosophy seminar.
Special pleading in theological argument is not merely a logical error; it is an intellectual habit, and intellectual habits are not confined to the domains in which they are first acquired. A person trained from childhood to accept that the most important claim in their worldview is permanently exempt from the standards of evidence and inference applied to every other claim is a person being trained in a mode of reasoning that has consequences outside theology. The willingness to exempt preferred conclusions from evidential scrutiny, to accept appeals to inscrutable authority as adequate responses to reasonable objections, to treat the impossibility of disproof as equivalent to evidence of existence: these are not just theological positions. They are epistemic dispositions, and epistemic dispositions shape how people evaluate claims in every domain, including medicine, politics, history, and science.
The harm is not hypothetical. The creationist movement in the United States and elsewhere has deployed exactly the same special pleading structure as classical theism: the theory of evolution is held to an impossibly high standard of completeness while the alternative, divine creation, is exempted from any evidential demand at all. The alternative is presented as requiring no mechanism because the mechanism is inscrutable, no evidence because the evidence is in the design visible all around us, and no falsifying test because divine action cannot be limited by human experiment. This is theological special pleading migrated into a domain, natural history, where its consequences are measurable and damaging. Children educated in frameworks that normalise this kind of asymmetric evidential standard are children less equipped to evaluate the claims that will actually govern their lives, including claims about medical treatments, about the reliability of public information, and about the criteria for trustworthy authority.
The same structure appears in faith healing communities that exempt prayer from the clinical trial standards applied to every other proposed medical intervention, in communities that exempt scriptural authority from the standards of historical textual criticism applied to every other ancient document, and in political movements that exempt their foundational claims from the scrutiny they apply to their opponents. Special pleading, once established as an acceptable mode of reasoning in the most important domain of a person’s intellectual life, does not stay there. It spreads. The load-bearing wall of theology, if it holds, holds a great deal more than the roof of the church.
This is why the evidential standard matters, and why its consistent application is not an exercise in philosophical pedantry but a question of intellectual integrity with practical consequences. The demand that religious claims meet the same standards of evidence and inference applied to all other claims is not an expression of hostility to religious people; it is an expression of respect for reasoning itself, and for the people who deserve to encounter the world’s genuine complexity rather than a framework that has been specially insulated against it. The point is not to wound but to clarify, and clarification in the service of honest thinking is the most respectful thing one can offer.
The God of the Gaps and the Vanishing Exemptions
There is a historical dimension to the special pleading problem that theologians rarely acknowledge but that the historical record makes clear. The domains in which God’s exemption from ordinary causal explanation has been invoked have steadily contracted over the past four centuries, and they have contracted not because theology has voluntarily relinquished territory but because natural inquiry has occupied it.
Newton believed that God was required to adjust the orbits of the planets periodically to maintain the stability of the solar system, because Newton’s gravitational mechanics predicted instabilities that only divine intervention could correct. Laplace demonstrated, a century later, that the solar system is self-stabilising through mechanisms entirely internal to celestial mechanics, and Napoleon’s famous exchange with Laplace, in which Laplace explained that he had no need of the hypothesis of God in his model, captures the moment precisely. The divine exemption from natural causal explanation was invoked for a specific physical problem, and when that problem was solved, the exemption was no longer needed and was quietly dropped, without any theological reckoning with what the dropping implied.
The same pattern has repeated itself across the history of natural inquiry: the origin of species, the structure of the genome, the neural correlates of consciousness, the large-scale structure of the universe, each has passed from the domain of divine exemption into the domain of natural explanation as inquiry has advanced. The origin of species was, for most of human intellectual history, a domain in which the exemption was invoked with complete confidence: life in its diversity and complexity could not be explained without a designer. Darwin’s work, and the century and a half of confirming evidence that has followed it, closed that gap with a thoroughness that remains the most powerful single instance of natural explanation superseding divine exemption in the history of science.
The theological response to this historical pattern has been to retreat, each time, to a more fundamental gap: not the stability of planetary orbits but the initial conditions of the solar system; not the initial conditions but the origin of life; not the origin of life but the fine-tuning of physical constants; not fine-tuning but the very existence of the universe. Each retreat preserves the structure of the exemption while moving it to ground where natural explanation has not yet fully arrived. This is the God-of-the-gaps pattern, and its theological deployment is itself a form of special pleading: the claim is that God’s causal role is real and demonstrable, but the domain in which that role is exercised conveniently shifts to wherever current natural explanation is thinnest.
The honest inference from this historical pattern is not that God is the explanation for whatever natural inquiry has not yet reached; it is that God has been the explanation-of-last-resort for every gap that has subsequently been closed by natural explanation, which is exactly the track record one would predict from a concept that fills explanatory gaps rather than from a concept that corresponds to a real causal agent. Each closure has been followed not by a theological revision of the core claim but by a retreat to a smaller, more remote gap, and the exemptions have been correspondingly redefined. The special pleading continues precisely because it must: without the exemption from evidential standards, the concept of God would need to make specific, falsifiable predictions and meet the confirmatory evidence that is required of every other explanatory claim. It has not met that standard in any domain that has subsequently been investigated, which is why the exemption must never be surrendered and the retreat must never be acknowledged as a defeat.
Applying the Standard Consistently
There is a simple diagnostic test for special pleading that can be applied to any theological argument, and it is worth making explicit here for readers who encounter these arguments in conversation, debate, or their own thinking. The test is this: identify the standard that the argument uses to evaluate the competing claim, and then apply that same standard to the theological claim itself. If the result is that the theological claim would fail the standard it applies to its competitors, you have found a special plea.
The cosmological argument uses the standard of causal sufficiency, the demand that things have causes adequate to explain them. Apply that standard to God: does God have a cause adequate to explain the existence of a being of infinite power, knowledge, and goodness? The apologist says no, because God is self-sufficient. But why is this self-sufficiency not available to the universe, whose existence the argument treats as requiring an external cause? If self-sufficiency can be invoked to terminate the causal regress at God, it can equally be invoked to terminate it at the universe, the quantum vacuum, or any other proposed terminus. The standard is being applied selectively, and the selection is made not on principled grounds but on the basis of which application serves the conclusion.
The hiddenness argument uses the standard of proportionality between love and effort to communicate. Apply that standard to God: a being of perfect love would make extraordinary efforts to communicate with beings who seek relationship. The apologist invokes hiddenness as divinely intentional. But apply the same standard to the hiddenness itself: if a person claimed to love their child but explained the absence of any communication as a deliberate strategy to preserve the child’s freedom, we would question both the claimed love and the claimed strategy with considerable force. The standard is suspended precisely and only for God, and the suspension is granted without argument, as though the specialness of the case were self-evident rather than the very thing requiring demonstration.
The problem of evil uses the standard of moral responsibility: a being with the power and knowledge to prevent suffering, who allows it anyway, bears moral responsibility for it unless there is a compelling and comprehensible reason. Apply that standard to God: the compelling and comprehensible reason is replaced by inscrutable divine purpose, which is an exemption from the standard rather than a satisfaction of it. The standard would not be accepted as satisfied by “inscrutable purpose” in any other domain of moral evaluation. A judge who explained a puzzling and apparently unjust verdict by appealing to considerations beyond the understanding of the court would not be praised for humility; they would be asked, correctly, to make their reasoning transparent. The inscrutability exemption is accepted in theology because theology claims it, and the claim goes largely unchallenged because the tradition of claiming it is long. Length of tradition is not a justification.
Once the diagnostic test is applied, the special pleading is visible in every case, and its visibility dissolves the argument. This is not because the questions theology addresses are unimportant; they are among the most important questions a human being can ask. It is because those questions deserve honest engagement rather than arguments that win by exempting their central claim from the standards of reasoning. The questions deserve better than the answers currently being offered, and recognising special pleading for what it is, the load-bearing wall of a structure that cannot stand without it, is the first step toward asking them with greater honesty.
What Remains When the Exemptions Are Removed
The natural and fair question, at the end of this analysis, is what happens to the concept of God if the special pleading is removed. If the cosmological argument’s uncaused cause must meet the same standard as every other proposed terminus of the causal chain, if God’s hiddenness must be held to the same standard of evidential significance as any other predicted absence, and if divine inscrutable purpose must meet the same standard as any other proposed justification for harm, then what does the concept of God look like without its exemptions?
The honest answer is that the classical God of monotheism does not survive the removal of its exemptions. A God who requires a cause becomes indistinguishable from another natural entity requiring explanation, and the regress problem re-emerges with full force. A God whose hiddenness counts as evidence against divine existence looks, by that standard, very much like a God who does not exist. A God whose purposes must be comprehensible to satisfy the moral standard applied to everything else looks, given the evidence of suffering, like a God of limited power, or limited goodness, or both. None of these conclusions is metaphysically impossible, but none of them corresponds to the God of classical monotheism that the arguments were originally constructed to establish.
What remains, when the exemptions are removed, is a much more modest claim: that there might be something rather than nothing, that the existence of the universe admits of no obvious natural explanation at its deepest level, that subjective experience has features not yet fully accounted for by neuroscience. These are genuine puzzles, and engaging with them honestly is intellectually serious work. But they are not arguments for the God of Abraham, Isaac, and Jacob, for the personal God of prayer and intervention, for the omniscient and omnipotent and perfectly loving being of classical theism. The distance between “there are things we do not fully understand” and “therefore the God of classical monotheism exists” is precisely the distance filled by special pleading, and once that pleading is removed, the distance must be crossed by evidence. No such evidence is available, and no amount of argumentative sophistication can substitute for it.
This is the point at which the unfalsifiability of the classical God concept becomes fully visible. The repeated deployment of special pleading to explain away the absence of evidence changes the epistemic status of the claim, as Coyne’s summary makes plain, to something no stronger than a logical possibility. A logical possibility is not nothing, but it is not enough to warrant assent, any more than the logical possibility of Russell’s orbiting teapot warrants the purchase of a second cup in its honour. The evidential standard demands more than logical possibility, and meeting that standard requires abandoning the exemptions. The concept of classical theism cannot be maintained without them, which is not an observation about the limits of human understanding but an observation about the structure of the argument being made.
There is also something worth noting about the asymmetry between what the exemptions cost and what they purchase. Each exemption costs a principle of reasoning that is otherwise applied without exception across the whole of human inquiry. The price is intellectual consistency, the willingness to follow an argument wherever it leads rather than wherever you need it to go. What is purchased by paying that price is the continued plausibility of a pre-existing commitment. This is a remarkably bad trade in epistemic terms, and it is made not because the arguments are genuinely compelling but because the commitment is prior to the arguments and the arguments are constructed to serve it. Recognising this dynamic is not an act of contempt; it is an act of clarity, and clarity is the only honest starting point for the questions that theological argument purports to address.
Conclusion: The Wall and What It Supports
Special pleading is not the most glamorous of logical fallacies. It lacks the dramatic structure of the ad hominem or the intuitive appeal of the false dilemma. It operates quietly, embedded in the technical vocabulary of philosophical theology, disguised as a concession to God’s unique nature or a recognition of the limits of human comprehension. But it is the fallacy on which everything in classical theism depends, which is precisely why recognising it matters so much, and why the theological tradition has such a strong incentive to keep it invisible.
Remove the uncaused cause exemption, and the cosmological argument dissolves into a restatement of the problem it was meant to solve. Remove the hiddenness exemption, and the silence of God becomes evidence of exactly the kind that it would be in any other domain: the absence of a predicted effect when the conditions for that effect are present. Remove the inscrutability exemption, and the suffering of the world becomes the kind of evidence that, applied consistently, points toward the absence of a perfectly good and perfectly powerful overseer. The exemptions are not peripheral rhetorical devices; they are the structural elements that hold the argument upright, which is what it means to say that special pleading is religion’s load-bearing wall.
The demand that religious claims meet the same standards of evidence and argument applied to every other domain of inquiry is not a demand made in hostility to the people who hold those claims. It is the minimum courtesy owed to serious questions by serious interlocutors. The questions about existence, meaning, suffering, and moral responsibility that theology addresses are among the most important that human beings can ask. They deserve arguments that stand without exemptions, and conclusions that are reached rather than assumed. The load-bearing wall has borne the weight long enough. What is needed now is not a new set of exemptions but the intellectual honesty to find out what the building looks like when it has to stand on its own.
Further reading: Jerry Coyne, “Faith Versus Fact: Why Science and Religion Are Incompatible” (2015); J. L. Schellenberg, “The Hiddenness Argument: Philosophy’s New Challenge to Belief in God” (2015); Bertrand Russell, “Why I Am Not a Christian” (1927); Carl Sagan, “The Demon-Haunted World: Science as a Candle in the Dark” (1995); J. L. Mackie, “The Miracle of Theism” (1982); Arthur Schopenhauer, “The World as Will and Representation,” Vol. I (1818).