Arguments relating to the existence of God — AQA A-Level Philosophy

Anselm argues a priori from the concept of God to God's existence. In the Proslogion he defines God as 'that than which nothing greater can be conceived'. Even a fool who denies God has this concept in his mind, so it exists in the understanding. But if it existed only in the understanding, a greater being could be conceived that also exists in reality; that would contradict the definition. So God must exist in reality. Anselm then argues that God cannot be conceived not to exist: a being that cannot be thought away is greater than one that can, so necessary existence belongs to God. Learn the steps in order, the role of the definition, and the distinction between existing in the understanding and existing in reality.

Descartes argues a priori in the Fifth Meditation from the idea of God as a supremely perfect being. He claims existence is a perfection, so a supremely perfect being must possess it, just as a triangle must have three angles equal to two right angles. To deny God's existence, he says, is like conceiving a mountain without a valley: the ideas cannot be separated. Descartes also holds that existence belongs to God's essence necessarily, not contingently. Learn the analogy with geometrical essences, the claim that existence is a perfection, and the necessity involved. Note that the argument moves from the content of the idea of God to God's existence, and be ready to explain why Descartes thinks the denial of God's existence is self-contradictory.

Malcolm revives Anselm's argument by distinguishing two kinds of existence-claim. Something may be conceived not to exist: a unicorn is conceivable, and its non-existence is not a contradiction, so its non-existence is contingent. By contrast, the non-existence of a being than which nothing greater can be conceived would be a logical impossibility, because a necessary being cannot coherently be thought away. Malcolm argues that if God's existence is possible at all, then God exists necessarily; the only alternative is that the concept is self-contradictory. He also separates 'God exists' from ordinary existence claims, treating it as a claim about necessary existence rather than about one more item in the world. Evaluation should test whether conceivability really tracks possibility, and whether the argument begs the question by assuming God's existence is possible.

Gaunilo's objection is a reply to Anselm's ontological argument. Gaunilo asks us to imagine a perfect island, greater than any other island, possessing every conceivable perfection. If Anselm's reasoning worked, then by the same logic this perfect island would have to exist, since an existing perfect island would be greater than a merely imagined one. But clearly we cannot bring an island into existence simply by defining it as perfect. Therefore, Gaunilo argues, the same flaw undermines the argument for God: you cannot move from a definition or idea to existence. The objection is a reductio ad absurdum, showing that the form of reasoning leads to an absurd conclusion when applied elsewhere.

Empiricists hold that substantive knowledge of reality comes from experience, so they challenge arguments that claim to establish God's existence by pure reason alone. The ontological argument is the clearest target: it moves from a concept of God to God's actual existence without any observational input. An empiricist objects that a concept tells us only how we define a word, not what exists; we cannot conjure a being into reality by analysing its definition. Hume's fork sharpens this: a claim is either a relation of ideas (true by definition, telling us nothing about the world) or a matter of fact (needing experience). A priori proofs of existence seem to be neither, so they fail to establish any real being. Learners should be able to explain why the empiricist rejects the inference from definition to existence and evaluate whether this objection is decisive.

Kant attacks the ontological argument by denying that existence is a real predicate that adds to a concept. A genuine predicate tells us something about the subject, as 'red' adds a property to 'apple'. Saying a thing exists adds nothing to its description; it only posits the thing as existing. Kant's example: a hundred real thalers contain no more than a hundred possible thalers in terms of their concept, though they differ in his actual finances. So if the ontological argument defines God as the most perfect being and claims existence is a perfection, Kant replies that existence cannot be a perfection because it adds no content to the concept. Denying that God exists does not contradict the concept of God; it denies that the concept is instantiated. Learners should explain the predicate distinction and assess whether the objection defeats the ontological argument.

Hume presents the design argument through Cleanthes, who reasons by analogy: the universe resembles a machine, and machines are produced by intelligent design, so the universe likely has an intelligent designer. The argument moves from observed resemblance between natural order and human artefacts to an inferred cause. Hume also has Philo expose the analogy's weakness: the resemblance is remote, we have experience only of minds producing limited effects, and the inference to a single, infinite, perfect designer exceeds the evidence. Learn to identify the analogy's structure, its inductive form, and the specific criticisms Hume raises against it, distinguishing the argument as presented from its evaluation.

Paley argues from spatial order: if you found a stone you might not wonder how it came to be, but if you found a watch you would infer a watchmaker, because its parts are arranged and adjusted for a purpose. The intricate arrangement of means to ends implies an intelligent designer. Paley then extends this: the universe displays similar spatial order and purpose, so it too implies a designer, powerful and intelligent. Learn to state the watch analogy, explain the inference from apparent purpose to design, and note that Paley's version emphasises the arrangement of parts rather than mere regularity.

Swinburne's design argument rests on temporal order: the world shows regularities that persist through time, such as the same laws of nature holding yesterday, today and tomorrow, so that water boils at the same temperature and planets keep to predictable orbits. For Swinburne this pervasive regularity is far more probable if a designing mind intended it than if it arose by chance, and he treats the order as evidence raising the likelihood of theism. Learn to distinguish temporal order (regularity over time) from spatial order (arrangement of parts, as in an eye), to state the probabilistic form of the argument, and to explain why Swinburne thinks regularity supports a designer. Be ready to identify the conclusion and the role of probability rather than treating the argument as a simple analogy.

Hume attacks the design argument where it reasons by analogy from human artefacts to the universe. Learn his main objections: the analogy is weak because the universe is unlike a machine made by a human craftsman; we have no experience of universe-making, so we cannot infer a cause by analogy; the evidence of imperfection and evil in the world counts against a perfect designer; and even if a designer were inferred, the argument cannot establish the traditional God of infinite power, knowledge and goodness. Hume also notes that the world may resemble a living organism more than a machine, and that multiple or limited designers are equally compatible with the evidence. Be ready to identify each objection and explain why it undermines the analogical reasoning rather than merely asserting disagreement.

Spatial disorder concerns how the arrangement of parts in space bears on design arguments. Hume, in the Dialogues, has Philo press that the universe's spatial layout shows no clear order: matter could fall into apparent regularity by chance, and apparent defects and irregularities weaken the analogy with a human artefact. Paley, in Natural Theology, poses the same problem himself and answers it: apparent disorder or imperfection in parts does not defeat the inference, because the contrivance of means to ends remains evident, and we judge a machine by its overall adapted structure rather than by isolated flaws. Learn both the challenge and Paley's reply, and how each uses spatial arrangement as evidence.

Hume argues that the design argument reasons from a single unique case, the universe, to a cause, a designer. Inductive and analogical reasoning normally work by comparing many similar cases, so we can check a conclusion against repeated experience. Because we have only one universe and no other worlds to compare, we cannot establish the analogy with human artefacts reliably. The learner should explain what a unique case is, why uniqueness undermines analogical inference, and how far this objection succeeds against Paley, who infers a designer from apparent order. Evaluation should consider whether the objection is decisive or whether the inference can still be reasonable.

This criterion asks you to evaluate the claim that God is the best explanation, or the only explanation, for some feature of the world — typically the existence and order of the universe. Teach the two distinct questions. 'Best explanation' is comparative: among rival hypotheses (theism, a brute-fact universe, a multiverse, an infinite regress), which has the greatest explanatory power, scope, simplicity and fit with evidence? 'Only explanation' is exclusive: are there genuinely no alternatives? A strong answer separates these, states the feature being explained, then weighs theism against at least one named rival. For example, if the feature is fine-tuning, assess whether a designer explains it better than chance plus observer selection. Reach a reasoned judgement rather than merely listing views.

The Kalām argument reasons from temporal causation: whatever begins to exist has a cause, the universe began to exist, so the universe has a cause, which is identified with God. Its distinctively temporal character lies in the claim that the universe had a beginning in time, so the causal series is not merely a hierarchy of simultaneous dependence but a first event requiring a cause. Defenders support the second premise with arguments against an actual infinite past, such as the impossibility of traversing an actual infinite, and with modern cosmological considerations. Critics question whether everything that begins must have a cause, whether the universe did begin, and whether any cause must be personal, intelligent or God. A strong answer separates the premises and assesses each.

Aquinas offers three a posteriori arguments from features of the world. The First Way begins from motion: things change, and whatever changes is moved by another; an infinite chain of moved movers cannot explain change, so there must be a first unmoved mover, which Aquinas identifies with God. The Second Way begins from atemporal causation: every effect has a cause, but causes cannot form an endless series of essentially ordered causes, so there must be a first cause that depends on nothing else. The Third Way begins from contingency: things come into and go out of existence, so they are possible but not necessary; not everything can be merely possible, so a necessary being must exist, and its necessity is grounded in God. Learn each starting point, the regress it rejects and the conclusion drawn.

Descartes argues from his own continuing existence to God as its cause. He notes that he exists now, but his existence at this moment is not explained by his existing a moment ago, since time is divisible and conserving his existence at each instant requires a cause. He cannot be the cause of his own continued existence, because if he could give himself existence he would give himself every perfection, which he lacks. His parents do not explain his continued existence now. Therefore a cause sufficient to sustain him at every moment must exist, and this cause must have the perfections he lacks, which Descartes identifies with God. The argument is a causal argument from conservation, not merely from his coming into being.

Leibniz argues from the principle of sufficient reason (PSR): nothing is the case without a reason why it is so rather than otherwise. Contingent things — things that could have failed to exist — require a reason for their existence, and that reason cannot lie wholly in other contingent things, since that merely postpones the question. An infinite series of contingent things would still contain no reason internal to itself for existing at all, so a sufficient reason must lie outside the series in a necessary being, God. Learn to reconstruct the steps, explain why Leibniz thinks an infinite series fails to supply the reason, and evaluate: is the PSR defensible, and does a necessary being have to be God?

Cosmological arguments ask whether the chain of causes or dependents can run back forever. An infinite series here is a regress with no first member: each item depends on a prior one without an originating term. Crucially, the series can still have a last member, such as the present effect or the thing now being explained. The question is whether such a beginningless regress can adequately explain why anything exists at all, or whether it merely postpones explanation. Learners should distinguish 'no first member' from 'no last member', and assess whether an infinite series removes the need for a first cause or merely relocates the demand for explanation.

Hume's objection targets the claim that whatever begins to exist must have a cause, the 'causal principle' used in the cosmological argument. Hume argues we never observe a necessary connection between cause and effect; we only experience constant conjunction and come to expect one event to follow another. So the principle is not a self-evident truth of reason, nor a demonstrable one: we can conceive of something beginning to exist without a cause without contradiction. It is at best a habit of expectation, not a necessary law. A learner should explain this, then assess it: does the objection really show the principle is unjustified, or only that it is not rationally demonstrable? Consider whether the cosmological arguer needs necessity or only a well-supported generalisation, and whether conceivability is a reliable guide to real possibility.

The fallacy of composition is inferring that a whole must have a property because each of its parts has that property. Russell's objection targets the cosmological argument's move from 'every event or thing in the universe has a cause' to 'the universe as a whole has a cause'. That inference is invalid unless independently justified, because wholes can lack properties their parts possess. A learner should state the fallacy precisely, apply it to the specific premise-to-conclusion step, and then assess the reply: defenders may argue the universe is not merely an aggregate, or that the principle of sufficient reason applies to the whole on separate grounds. Judge whether Russell's charge succeeds against the argument as actually stated.

This topic asks you to assess the claim that there cannot be a necessary being, using Hume and Russell. A necessary being is one whose non-existence is impossible; a contingent being could fail to exist. Hume's empiricist point is that we can only meaningfully assert what is conceivable or evidenced, and we can always conceive of any being not existing, so no being is demonstrably necessary. Russell argues that the word 'necessary' properly applies to propositions (logical truths), not to things: saying 'this being exists necessarily' confuses logical necessity with a feature of an object. To perform well, reconstruct each argument in premises and conclusion, explain the reasoning, then evaluate: does conceivability really show non-existence is possible? Does Russell's distinction between propositional and thing-necessity beg the question against the theist? Reach a supported judgement.

This topic asks you to assess whether the existence of evil can be reconciled with God's attributes, especially omnipotence, omniscience and omnibenevolence. The logical problem claims that an all-powerful, all-knowing, all-good God would prevent evil, yet evil exists, so such a God cannot exist. The evidential problem claims that the amount and distribution of evil make belief in God unreasonable rather than strictly contradictory. To perform well, state God's attributes precisely, set out the inconsistent-triad or deductive argument, distinguish logical from evidential versions, then evaluate responses such as the free-will defence, soul-making and the distinction between moral and natural evil. Reach a supported judgement on whether reconciliation succeeds.

Moral evil is suffering and wrongdoing brought about by the free choices of moral agents: a person deliberately deceiving, torturing or exploiting another. Natural evil is suffering arising from processes and events that are not the choices of moral agents: earthquakes, disease, drought and the resulting famine. The distinction turns on agency and causation, not on how bad the suffering feels. A useful test is to ask whether a responsible agent's choice is the proximate cause. If a builder cuts corners and a bridge collapses, the collapse is a moral evil traced to choice; if an unforeseeable earthquake destroys the bridge, the harm is natural. Some cases are mixed: a famine may be natural in origin yet worsened by corrupt distribution decisions.

The problem of evil comes in two forms. The logical form claims that the existence of evil is logically inconsistent with a God who is omnipotent, omniscient and omnibenevolent, so the set of beliefs is contradictory and at least one must be given up. It is deductive: if the contradiction holds, no amount of evil is needed, and a single instance would suffice. The evidential form is inductive or probabilistic: the amount, distribution and apparently pointless character of suffering makes God's existence unlikely, even if not strictly impossible. It concedes logical consistency and argues that the evidence counts against belief.

The Free Will Defence answers the problem of evil by arguing that a world containing free creatures who may choose good is better than a world of automata, and that God could not actualise a world of free creatures who always freely do right, since free choice cannot be causally guaranteed. Plantinga develops this: God permits evil because the actualisation of a world containing significantly free creatures who always choose the good may not be within God's power. Teach learners to state the logical problem of evil, then show how the defence targets it, and to weigh responses: critics say a good God could have made a world with less natural evil, or that the defence only addresses moral evil; defenders reply that natural evil may be a consequence of free choices or a necessary environment for soul-making.

Soul-making is a theodicy, associated with John Hick, arguing that God permits evil because it is necessary for the development of human souls towards a greater good. Hick draws on Irenaeus: humans are created at an epistemic distance from God, as immature beings who must freely develop into the likeness of God through responding to challenges. Evil and suffering provide the environment in which virtues such as courage, compassion and perseverance can be formed; a world of ease would produce no such growth. Teach learners to explain the two-stage creation (image to likeness), the role of epistemic distance, and the distinction between moral and natural evil as soul-making conditions. Then evaluate: critics argue the quantity of suffering is disproportionate, that not all suffering produces growth, and that the argument risks making God the author of evil.