The arguments for and against God's existence, honestly assessed

Why the arguments matter

The case for God’s existence is not one argument. It is a stack. Classical theologians like Thomas Aquinas, modern philosophers of religion like Richard Swinburne and Alvin Plantinga, and contemporary apologists like William Lane Craig all converge on the same move: no single argument is decisive, but together they raise the probability that a God of some description exists high enough to justify belief. The atheist who engages the strongest versions of each argument is not engaging one claim, but a cumulative case.

This page exists because most popular writing on both sides caricatures the other. Atheists dismiss the cosmological argument as “everything has a cause, so God,” which is not what any serious proponent argues. Theists dismiss the problem of evil as “people suffer, so no God,” which is also not what any serious atheist argues. The point here is to give each side credit for the strongest version of its own case and then show what the responses actually are.

If you came to this site because you are deconstructing, you do not need to feel intellectually cornered by arguments you have never seen stated seriously. And if you came here as a skeptic who thinks it is all obvious, the classical arguments have kept some of the best minds in philosophy busy for two thousand years. That is worth respecting before dismissing.

Arguments for God’s existence

1. The cosmological argument

Steelman. Every contingent thing in the universe — every thing that could have failed to exist — depends on something else for its existence. The universe as a whole is a contingent thing. Therefore the universe depends on something for its existence, and that something cannot itself be contingent, or the regress continues forever. The only stopping point is a necessary being, something that exists by its own nature rather than by depending on anything else. Classical theists identify that necessary being with God.

Response.The argument from contingency is far stronger than the “first cause” caricature, but it still trades on assumptions that can be contested. Why should the universe as a whole be the kind of thing that has an explanation outside itself? (This is the Hume-Russell objection.) Why should the necessary being be personal, let alone good? (A brute mathematical structure, or the laws of physics themselves, could play the same role.) And even if the argument succeeds, it takes you only to a necessary something — not to any particular God of any particular religion. For the fuller treatment, see the cosmological argument.

2. The Kalam cosmological argument

Steelman. A modern variant of the cosmological argument, revived by William Lane Craig and drawing on medieval Islamic philosophy. Its simplest form: (1) whatever begins to exist has a cause; (2) the universe began to exist; therefore (3) the universe has a cause. Craig supplements the second premise with arguments from cosmology (Big Bang theory, the Borde-Guth-Vilenkin theorem) and from the philosophy of infinity (an actual infinite cannot exist; a beginningless temporal sequence would constitute an actual infinite).

Response. The first premise is the Achilles heel. It looks intuitive, but our intuitions about causation were formed by watching things rearrange themselves inside the universe, not by watching universes begin. Quantum events routinely come into existence without prior causes of the kind Craig has in mind. The second premise depends on a contested reading of cosmology: the Big Bang is the beginning of our observable cosmos, but it is not clear it is the beginning of everything. See the Kalam cosmological argument for a deeper walkthrough.

3. The ontological argument

Steelman.Originating with Anselm of Canterbury in the 11th century and reformulated by Descartes, Leibniz, and most recently Alvin Plantinga in modal logic, the ontological argument tries to show that God’s existence can be derived from the concept of God alone. Anselm’s version: God is, by definition, the greatest conceivable being. A being that exists in reality is greater than one that exists only in the mind. Therefore God must exist in reality, or God would not be the greatest conceivable being.

Response.The argument feels like a magic trick, and most philosophers since Kant have argued that it is. Kant’s objection is still the canonical one: existence is not a predicate, not a property a thing can have or lack like being red or round. To say God exists is not to add something to the concept; it is to say the concept is instantiated. Plantinga’s modal version is more sophisticated — it argues from the possibility of a maximally great being to its actual existence — but it relies on a premise (it is possible that a maximally great being exists) that the atheist has no reason to grant. The ontological argument page walks through both versions in more detail.

4. The design argument (teleological)

Steelman.Living things show extraordinary functional complexity — eyes, wings, immune systems, ecosystems that hang together across billions of interacting parts. Paley’s famous watchmaker analogy: if you found a watch on a heath, you would infer a watchmaker from the fit between the parts and the function. Living systems display vastly more integration than any watch. The most natural explanation is that they too have a designer.

Response. Darwin. The theory of natural selection is the definitive answer to the biological design argument, and a comprehensive one: variation within a population, differential reproductive success, and enough time will produce the appearance of design without any designer. This is not controversial among working biologists. The design argument from biology is largely abandoned in academic philosophy of religion; the living version is the fine-tuning argument, which moves the design intuition up a level to the laws of physics themselves. For more, see the argument from design.

5. The fine-tuning argument

Steelman. The physical constants of our universe — the strength of gravity, the mass of the proton, the cosmological constant, a dozen others — appear to fall within an extraordinarily narrow range compatible with the existence of complex structures, stars, planets, and life. Tweak any one of them slightly and the universe produces no chemistry, no stars, no observers. This improbable coincidence demands explanation. Either it is a brute fact, or it is the product of a multiverse, or it is the result of deliberate fine-tuning by a designer.

Response.The fine-tuning observation is real; the inference to a designer is not. The anthropic principle explains why we necessarily find ourselves in a universe compatible with observers — if it weren’t, we wouldn’t be around to comment on it. The multiverse hypothesis takes the sting out of the improbability: if many universes exist with different constants, at least one will look like ours, and we are the ones in that one. The critic can also push on what counts as “fine-tuned” — we don’t know what range of constants is even possible, let alone what probability distribution over them is correct. See the fine-tuning argument for a longer treatment.

6. The moral argument

Steelman. Objective moral truths exist. Torturing children for fun is wrong, not just disapproved of by most people. Cowardice is worse than courage regardless of whether anyone notices. But in a purely material universe of particles and forces, what could ground an objective moral truth? There is nothing in physics that forbids cruelty. If objective moral values exist, they must have a ground outside the natural order, and the most plausible candidate is a personal moral lawgiver.

Response. The argument has two live responses. The first denies the premise: moral anti-realism holds that moral claims are expressions of preference, commitment, or community standards, not descriptions of mind-independent facts. The second accepts the premise but rejects the inference: moral naturalism argues that objective moral facts can be grounded in features of the world — the capacity to suffer, the preconditions for cooperation, the nature of rational agents — without needing a deity. The Euthyphro dilemma is also still alive: is something good because God commands it, or does God command it because it is good? If the first, morality is arbitrary; if the second, morality is independent of God anyway. See the moral argument and morality without God.

7. The argument from religious experience

Steelman. Across cultures and centuries, millions of people have reported experiences that seem, from the inside, to be encounters with a transcendent reality — mystical union, answered prayer, the presence of a loving other, conversion visions. Taken at face value, these are data about reality. We accept sense experience as reliable evidence about the physical world on the principle of credulity: absent positive reason for doubt, trust reports of experience. Apply the same principle to religious experience and you have prima facie evidence for the divine.

Response.The principle of credulity is reasonable but not unlimited. Religious experiences are vastly more diverse than sense experiences and frequently mutually incompatible — the Christian experiences the triune God, the Hindu experiences Krishna, the Muslim experiences Allah, and each experiences their own tradition’s theology rather than some neutral numinous other. That divergence is hard to explain if the experiences are tracking the same underlying reality. And the psychology and neuroscience of mystical states is now well enough developed that we have non-divine explanations for almost every feature of them. See the argument from religious experience.

8. The argument from miracles

Steelman. If a genuine miracle occurred — a breach of the ordinary order of nature produced by divine action — it would be direct evidence for a God. The gospels report the resurrection of Jesus with multiple independent witnesses; Catholic tradition has continuously generated reports of healings investigated by medical commissions; contemporary charismatic Christianity has thousands of first-person testimonies. A blanket refusal to consider miracle claims is not skepticism; it is assumption.

Response.David Hume’s classic essay still applies: no evidence for a miracle is ever stronger than the evidence against it, because the evidence against any particular violation of the natural order is the entire history of unbroken regularity on which we base our understanding of that order. The modern response adds probability theory: given a base rate of credible miracle reports (low) and the known frequency of honest mistake, memory distortion, and deliberate embellishment in eyewitness testimony (high), the rational prior on any given miracle claim is near zero, and the evidence required to overturn that prior is correspondingly enormous. See the argument from miracles.

9. The argument from scripture

Steelman. The Bible (or the Quran, or the Bhagavad Gita, depending on the tradition) contains content that could not plausibly have been produced by human authors alone — fulfilled prophecies, scientific foreknowledge, moral teachings that transcend their cultural context, a coherent narrative across dozens of authors separated by centuries. The best explanation is supernatural inspiration.

Response. Critical scholarship has spent two centuries examining the biblical text with the tools of literary, historical, and linguistic analysis, and the result is a detailed, non-supernatural account of how the texts came to be: multiple authors with distinct theological agendas, contested canonization, translation errors, scribal interpolations, and post-hoc editing to make predictions look fulfilled. Every specific claim of scientific foreknowledge in the Bible turns out, on inspection, to depend on stretching the original Hebrew or Greek beyond what the text will support. The argument from scripture page walks through specific cases.

10. The argument from consciousness

Steelman.Subjective experience — the felt quality of seeing red, tasting coffee, feeling grief — is the hardest thing in the universe to explain in purely material terms. David Chalmers calls this the “hard problem” of consciousness. Neurons firing in a brain might explain behavior, but they do not, on the face of it, explain why there is something it is like to be the person whose brain is firing. If consciousness cannot be explained by physics, a non-physical explanation is required, and a personal God who created conscious beings is one candidate.

Response. The hard problem is a real open question, but it is not a knockdown argument for theism. Several secular approaches take the problem seriously and offer non-theistic answers: panpsychism, biological naturalism, emergentist theories of mind. Invoking God is one more hypothesis alongside these, and it inherits all the usual problems — adding a non-physical mind to explain non-physical minds is not obviously progress. See consciousness and the soul.

Arguments against

The cumulative case for theism has to be weighed against the cumulative case the other way. Here are the main arguments the atheist side of the debate is built on.

The problem of evil

Steelman. A God who is all-powerful could prevent any suffering. A God who is all-knowing would be aware of every instance. A God who is perfectly good would want to prevent unnecessary suffering. But gratuitous suffering is everywhere — the child with bone cancer, the tsunami that drowns a village, billions of years of animal predation. Either God cannot prevent it, does not know, or does not want to. Any of the three undermines the classical picture of God.

Response. The free-will defense (moral evil is the necessary price of genuine freedom), the soul-making theodicy (suffering builds virtue), and skeptical theism (we are not in a position to judge which evils are gratuitous) are the three standard replies. None is universally accepted, but they block the argument from functioning as a proof. The atheist version then shifts to the evidential problem of evil: even if theodicies block the logical problem, the sheer scale of suffering still looks like strong evidence against theism. See the problem of evil for the full treatment.

Divine hiddenness

Steelman.The philosopher J.L. Schellenberg’s formulation: if a perfectly loving God existed, no person who was open to a relationship with God would fail to be in one. Non-resistant non-belief happens — people who sincerely want to believe, who would welcome evidence, who end up unable to. The existence of such people is incompatible with the existence of a God whose defining property is self-giving love.

Response. Theists reply that God might have reasons to remain hidden that are compatible with perfect love — to preserve libertarian free will, to build character through searching, to prevent belief from being coerced by overwhelming evidence. The atheist counters that none of these reasons explain why God would be hidden specifically from sincere seekers. See divine hiddenness.

The argument from societal harm

Steelman. A theory should be judged in part by its fruit. The historical record of institutional religion includes inquisitions, holy wars, suppression of scientific inquiry, sexual abuse cover-ups, institutional homophobia, and the coercive indoctrination of children. Secular societies with low religiosity consistently outscore highly religious ones on well-being, education, gender equality, and violence reduction. If religion were grounded in a God of love, we would expect the fruit to be better, not worse.

Response.The defender of religion points to the charitable, educational, and moral contributions of religious traditions — hospitals, schools, civil rights movements, the abolition of slavery, the founding of most of the world’s universities — and argues that the critic is cherry-picking. The empirical question of whether religion is on balance beneficial or harmful is genuinely hard to answer; serious work in the sociology of religion continues. See religion and societal harm.

Meta: how to tell a good argument from a bad one

After working through the arguments above, you might notice that they are not all doing the same kind of work. Some (cosmological, fine-tuning) try to draw conclusions from general features of reality and rely on contested philosophical premises. Some (miracles, scripture) try to draw conclusions from specific empirical claims and rely on the reliability of historical testimony. Some (moral, religious experience) try to draw conclusions from features of our own inner lives and rely on whether those features require a theistic explanation. These are different kinds of bets, and evaluating them calls for different skills.

A few practical rules for reading any of them:

• Steelman first, critique second. If the argument you are refuting is not the strongest version, you are wasting your time. Neither side benefits from beating up caricatures.
• Notice where the weight is being carried.Most arguments have one crux premise. Find it. If the argument turns on the premise “whatever begins to exist has a cause,” the rest of the chain is secondary.
• Distinguish logical from evidential arguments. Logical arguments try to show that something is impossible. Evidential arguments try to show that something is unlikely. Do not conflate them — a response that defeats the logical version may leave the evidential version entirely untouched.
• Be suspicious of the argument that moves you most. Motivated reasoning works both ways. The argument that confirms what you already want to believe is the one you should interrogate hardest.

Where to go next

Each argument above links to its own page on this site, where the history, technical detail, and academic literature are laid out more carefully. If you want the opposite framing — the ethical and psychological side of leaving faith rather than the philosophical side — start with the deconversion section or the individual first-person stories.

For the thinkers who have shaped these arguments, see the profiles of notable figures— Richard Swinburne and William Lane Craig on the theist side, Graham Oppy and J.L. Mackie on the atheist side, and philosophers like Sam Harris, Daniel Dennett, and Alex O'Connor who have pushed the public conversation.

For short definitional entries on the technical terms used above — theodicy, fine-tuning, natural theology, apophatic theology, presuppositionalism — see the glossary.