Cosmological arguments

The structure of cosmological arguments#

FoundationAQA 7172 3.3 Metaphysics of God — Cosmological argumentsDfE GCE Philosophy subject content

The structure of cosmological arguments

Fig. 1A starting premise plus a causal principle, minus infinite regress, yields a first cause or necessary being.

Cosmological arguments are a posteriori arguments that begin from some very general feature of the universe — that things move or change, that things are caused, that things are contingent, or simply that the universe exists at all — and reason to a first cause, unmoved mover, or necessary being, identified with God. Unlike the ontological argument (a priori, from the concept of God) and the design argument (from the ORDER of the world), the cosmological argument needs only the bare EXISTENCE (or changeability, or contingency) of the world, not its detailed order; so it is not touched by Darwin, and it survives even a chaotic universe. This makes it, for many, the most robust of the theistic arguments.

All versions share a common three-part structure. First, a starting premise about the world: it contains motion/change (Aquinas' First Way), or efficient causes (Second Way), or contingent beings (Third Way), or simply exists (Leibniz, Kalam). Second, a causal or explanatory principle: everything moved is moved by another; every effect has a cause; every contingent being depends on something for its existence; everything that exists has a sufficient reason. Third, the rejection of an infinite regress or of a brute unexplained totality: the chain of movers/causes/dependent beings cannot go back infinitely (or the totality cannot be unexplained), so there must be a first, uncaused, necessary term — God.

It is essential to distinguish two kinds of causal series the arguments invoke. A series 'ordered per accidens' (accidentally) is a temporal chain in which each member causes the next and then may cease to exist — a father begets a son who begets a grandson; the earlier members need not still exist for the later to operate. The Kalam argument concerns this kind, arguing the temporal chain has a beginning. A series 'ordered per se' (essentially) is one in which each member depends SIMULTANEOUSLY and continuously on a sustaining cause here and now — a hand moving a stick moving a stone; if the first member stopped, all would stop at once. Aquinas' Ways concern this kind, arguing the sustaining chain requires a first sustaining cause NOW, regardless of whether the universe had a beginning in time. Confusing the two is the commonest error in the topic.

The arguments' overall strength is that they respond to a deep and natural demand for explanation: it seems that the existence of the universe — a vast, contingent, changing thing — cannot simply be a brute, unexplained fact, but must have a reason or cause outside itself. Their vulnerability is at each of the three joints: the starting premise may be questioned, the causal/explanatory principle may not extend to the universe as a whole (Hume, Russell), and the rejection of infinite regress or of a brute totality may be unjustified (perhaps an infinite regress is possible, or the universe is a brute fact). The evaluation of any cosmological argument is the evaluation of these three moves, and the identification of the resulting first cause with the God of theism is a further, contestable step.

Worked example

Identifying the three joints

For the claim 'everything contingent depends on something else, and this cannot go back forever, so there is a necessary being', identify the three structural parts.

Starting premise
'There are contingent beings' (things that could have not existed).
Explanatory principle
'Every contingent being depends on something else for its existence.'
Regress rejection
'The chain of dependent beings cannot regress infinitely (or be a brute totality).'
Conclusion and its joint
'So there is a necessary being' — and the further step of identifying it with God is separately contestable.

Result: The argument decomposes cleanly into starting premise, principle and regress-rejection; each is a distinct point of attack, and the God-identification is an additional step beyond the bare 'necessary being'.

Typical mistakes

Confusing the temporal (per accidens) and sustaining (per se) series; Aquinas' Ways do not claim the universe began in time — that is the Kalam claim.
Stating the argument self-defeatingly as 'everything has a cause, so the universe has a cause, so God has a cause'; the principle exempts the necessary/uncaused first term.

Active revision

Explain the general structure of a cosmological argument and the difference between a per se and a per accidens causal series. (5-mark AO1 response)

Active recall

Recall the key points — then reveal.

Sources: GCE AS and A level subject content for philosophy (Department for Education)

The Kalam argument#

StandardAQA 7172 3.3 Metaphysics of God — Cosmological argumentsDfE GCE Philosophy subject content

The Kalam argument

Fig. 2A valid syllogism from the causal principle and the universe's beginning to a first cause.

The Kalam cosmological argument, of medieval Islamic origin (al-Ghazali) and revived by William Lane Craig, is the sharpest and simplest version. It is a temporal argument concerning a per accidens series, and runs as a valid deductive syllogism: (P1) whatever begins to exist has a cause; (P2) the universe began to exist; therefore (C) the universe has a cause. The form is modus ponens-like: if something begins to exist (

B

) then it has a cause (

C

), i.e.

B \rightarrow C

; the universe began to exist (

B

); so the universe has a cause (

C

). The argument is plainly valid, so all evaluative weight falls on the two premises and on the further identification of the cause with God.

P1 ('whatever begins to exist has a cause') is defended as a self-evident metaphysical principle: things do not pop into existence uncaused, out of nothing; 'out of nothing, nothing comes' (ex nihilo nihil fit). Craig stresses that P1 concerns things that BEGIN to exist, so it does not apply to God (who did not begin to exist), blocking the self-defeating 'so what caused God?' objection. Critics reply that our confidence in P1 comes from experience of causation WITHIN the universe and in time, and may not extend to the beginning of the universe itself, where there is no 'before' and no prior time in which a cause could operate; and that quantum mechanics may exhibit genuinely uncaused events (virtual particles), suggesting P1 is not exceptionless.

P2 ('the universe began to exist') is defended on two fronts. Philosophically, Craig argues that an actual infinite series of past events is impossible: a beginningless past would be a completed actual infinite, which generates paradoxes (Hilbert's Hotel), and one cannot 'traverse' an actual infinite to reach the present by successive addition; so the past must be finite, and the universe began. Scientifically, he appeals to Big Bang cosmology (the universe expanding from an initial singularity ~13.8 billion years ago) and to the second law of thermodynamics (the universe running down towards equilibrium, which an infinitely old universe should already have reached). Critics respond that the impossibility of an actual infinite is contested (mathematicians work with actual infinities), that Big Bang cosmology describes the early universe but is compatible with various models (including some past-eternal or cyclic ones), and that the initial singularity may mark a boundary rather than a caused beginning.

Even granting the conclusion that the universe has a cause, Craig argues (and critics dispute) that the cause must be uncaused, beginningless, changeless, timeless, spaceless, immaterial, enormously powerful, and — crucially — PERSONAL, because only a free personal agent could bring about a temporal effect (a beginning) from a timeless, changeless state (an impersonal timeless cause would either always or never produce its effect, but a free will could choose to create). This last step, from 'a cause' to 'a personal creator', is the most contested: the cause could be some impersonal physical or quasi-physical condition, and the inference to a PERSON imports much of the concept of God that the bare argument does not establish. A balanced verdict is that the Kalam is valid and its premises are seriously defensible (especially P1 and the philosophical case for a finite past), so it is a strong argument for a first cause; but that each premise is contestable at the frontier (causation at t=0, the possibility of an actual infinite, the interpretation of cosmology), and the leap to a PERSONAL God is a further inductive step — so it establishes, at most, a probable transcendent cause rather than the full God of theism.

Worked example

Testing the Kalam's premises

Reconstruct the Kalam argument and assess whether it is sound.

State it
P1: whatever begins to exist has a cause. P2: the universe began to exist. C: the universe has a cause. Valid.
Test P1
Intuitive and grounded in 'nothing comes from nothing', but our confidence derives from causation WITHIN time; it may not apply at the beginning, and quantum events may be uncaused.
Test P2
Supported by the impossibility of an actual infinite past and by Big Bang cosmology; critics dispute the actual-infinite claim and cite past-eternal/cyclic models.
Test the God-step
Even granting a cause, the inference to a timeless, PERSONAL creator is a further, contestable step; the cause might be impersonal.

Result: The Kalam is valid and its premises seriously defensible, so it supports a first cause; but each premise is contestable at the cosmic frontier and the leap to a personal God is additional, so it establishes at most a probable transcendent cause.

Typical mistakes

Stating P1 as 'everything has a cause' (which would need a cause of God); Craig's premise is 'whatever BEGINS to exist has a cause', exempting the beginningless God.
Confusing the Kalam (temporal: the universe began) with Aquinas' Ways (sustaining: the universe depends now); the Kalam needs the past to be finite, Aquinas does not.

Active revision

'The Kalam cosmological argument proves that the universe has a cause.' Assess this claim. (25-mark essay)

Active recall

Recall the key points — then reveal.

Sources: GCE AS and A level subject content for philosophy (Department for Education)

Aquinas' first three Ways#

StandardAQA 7172 3.3 Metaphysics of God — Cosmological argumentsDfE GCE Philosophy subject content

Aquinas' Third Way

Fig. 3From the contingency of things and the impossibility that all be contingent, Aquinas infers a necessary being.

Aquinas' first three of his Five Ways (Summa Theologica) are cosmological, and — unlike the Kalam — they concern per se (sustaining) causal series here and now, not a temporal beginning; Aquinas actually thought the universe's temporal beginning was known only by revelation, not provable by reason. The First Way (from motion) observes that things in the world are in motion (understood broadly as change, the actualisation of a potential); whatever is moved is moved by another (nothing actualises its own potential); this series of movers cannot regress infinitely (in a per se series, without a first mover there would be no motion at all, here and now); so there must be a first, unmoved mover, which everyone understands to be God.

The Second Way (from efficient causation) runs the same structure for causes rather than motion: in the world we find an order of efficient causes; nothing can be the efficient cause of itself (it would have to exist prior to itself); the series of sustaining efficient causes cannot regress infinitely (without a first cause, there would be no intermediate or ultimate causes now); so there must be a first efficient cause, which we call God. Both Ways rest on two claims: that a per se series of sustaining causes requires a first member (remove it and the whole present series collapses), and that such a series cannot be infinite (an infinite per se series would have no first sustaining cause, so nothing would be sustained).

The Third Way (from contingency/possibility and necessity) is the most philosophically interesting. It observes that things in the world are contingent — they come to be and pass away, so it is possible for each of them not to exist. Aquinas argues: if everything were contingent, then (since for each contingent thing there is a time it does not exist) there would have been a time when nothing at all existed; but if there had ever been nothing, nothing could have come into existence (from nothing, nothing comes), so there would be nothing now — which is false. Therefore not everything can be contingent: there must exist at least one NECESSARY being (whose existence does not depend on another). Aquinas then argues this necessary being does not receive its necessity from another (else another regress), so there is a being necessary of itself — God.

Evaluation weighs the Ways against their objections. The rejection of infinite regress is defensible for PER SE series (an infinite chain of sustainers with no first sustainer arguably could sustain nothing), though critics ask why there must be a first rather than an infinite sustaining structure, or why the universe itself could not be the necessary sustaining ground. The Third Way faces a famous logical objection (often attributed to the argument's own wording): it seems to commit a quantifier-shift fallacy in moving from 'each contingent thing at some time does not exist' to 'at some time all contingent things do not exist' (from 'for each thing there is a time it fails to exist' to 'there is a time when everything fails to exist') — the latter does not follow from the former, since the non-existence-times might not coincide. Defenders reply that Aquinas can be read charitably, or that a Leibnizian version (next section) avoids the fallacy by appealing to sufficient reason rather than to a shared time of universal non-existence. The identification of the first mover/cause/necessary being with the God of theism is, again, a further step. A balanced verdict is that the Ways establish, if sound, a first sustaining cause or necessary being — a real metaphysical result — but that the rejection of regress and (for the Third Way) the contingency reasoning are contestable, and that the argument delivers a necessary ground of the world rather than, by itself, the personal God of religion.

Worked example

The quantifier-shift objection to the Third Way

Explain the logical objection that the Third Way commits a quantifier-shift fallacy.

The premise
For each contingent thing, there is some time at which it does not exist.
The alleged inference
The argument seems to move to: there is some time at which NO contingent thing exists (a universal blank).
Why it is a fallacy
'For each x there is a time t(x) of non-existence' does not entail 'there is one time t at which all x fail to exist' — the times need not coincide (compare: everyone has a birthday, but not everyone shares one).
A charitable reply
The argument may be reconstructed without this step (e.g. Leibniz's sufficient-reason version), or read as concerning the contingency of the whole, avoiding the shift.

Result: On its face the Third Way commits a quantifier-shift fallacy in reaching a universal time of non-existence; the argument survives only if reformulated (e.g. via sufficient reason), which is why the Leibnizian version is often preferred.

Typical mistakes

Claiming Aquinas argues the universe 'began to exist'; his Ways are about dependency here and now, and he held the beginning was known only by revelation.
Treating the rejection of infinite regress as obviously true for all series; it is defensible for per se series but not for per accidens (temporal) ones, which Aquinas grants could be infinite.

Active revision

Outline Aquinas' Third Way and explain one objection to it. (12-mark AO1 response)

Active recall

Recall the key points — then reveal.

Sources: GCE AS and A level subject content for philosophy (Department for Education)

Leibniz and the principle of sufficient reason#

AdvancedAQA 7172 3.3 Metaphysics of God — Cosmological argumentsDfE GCE Philosophy subject content

Leibniz's argument from sufficient reason

Fig. 4Even an infinite contingent series is a contingent whole, so by the PSR its reason lies in a necessary being.

Leibniz gives a cosmological argument that avoids reliance on the impossibility of an infinite regress and instead rests on the Principle of Sufficient Reason (PSR): everything that is the case must have a sufficient reason why it is so and not otherwise — nothing is without a reason. Applied to existence: for everything that exists, there must be a sufficient reason (an explanation) for its existence. Leibniz then argues that even if the universe (the series of all contingent things) were infinitely old, with each state explained by a prior state, the whole SERIES would still be contingent and would lack a sufficient reason for its existence, because explaining each member by an earlier member never explains why there is the series AT ALL rather than nothing. The reason for the existence of the whole contingent series must therefore lie OUTSIDE it, in a being that exists necessarily and contains its own sufficient reason — God.

The argument's cleverness is that it sidesteps the regress debate. Grant, for the sake of argument, an infinite past: even so, an infinite chain of contingent things is itself a contingent fact (the whole chain could have not existed, or been different), and by the PSR this contingent fact needs a sufficient reason. That reason cannot be another contingent thing (which would itself need a reason, continuing within the series), so it must be a NECESSARY being whose existence is self-explanatory — a being that exists by its own nature, so that the question 'why does IT exist?' is answered by its very essence. This connects the cosmological argument to the ontological argument's necessary being and to Aquinas' 'necessary of itself'.

The argument's strength is that the PSR expresses a deep and widely shared intuition — that things do not exist for literally no reason, that the existence of a contingent universe cries out for explanation — and that it does not depend on the contested claim that an actual infinite is impossible. Its central vulnerability is the PSR itself. Is the PSR true? It cannot be established a priori without controversy, and applying it to the TOTALITY of contingent things is exactly what critics resist: perhaps the universe as a whole simply has no explanation — it is a 'brute fact'. Hume and Russell (next section) press this directly.

A further, subtle objection is that the PSR may prove too much or lead to paradox. If EVERYTHING must have a sufficient reason, including God, then either God has a reason (and is not the terminus) or God is an exception to the PSR (and then why not make the universe the exception?). Leibniz answers that God is not an exception: God's sufficient reason lies in God's own necessary nature (God is self-explanatory), whereas the universe, being contingent, cannot be self-explanatory. But critics reply that 'self-explanatory necessary being' is obscure, and that if a necessary being can be self-explanatory, perhaps the universe itself is the necessary being (a 'necessary universe' hypothesis), removing the need for God. A balanced verdict is that Leibniz's argument is the most sophisticated cosmological argument — it avoids the regress and quantifier-shift problems and isolates the real issue, the demand for an explanation of the contingent universe as a whole — but that it stands or falls with the PSR applied to the totality, which the theist takes as rational and the critic rejects as unwarranted (the universe may be a brute fact); the argument therefore sharpens the debate to a single, deep question about explanation, without settling it against a determined critic.

Worked example

The brute-fact objection to the PSR

Assess the objection that the universe may be a brute fact requiring no explanation.

Leibniz's claim
The PSR requires a sufficient reason for the existence of the contingent universe as a whole, which must be a necessary being.
The objection
Deny the PSR for the totality: perhaps the universe simply exists with no explanation — a brute fact — as Russell says, 'the universe is just there, and that's all'.
The stakes
If the PSR does not apply to the whole, the argument collapses; if it does, a necessary being follows — so everything turns on the PSR's scope.
Weigh
The PSR is intuitively compelling and underlies science's search for explanations, but cannot be proved and may not extend to the totality; the theist finds a brute universe arbitrary, the critic finds a necessary being no less so.

Result: The argument reduces to a stand-off over the PSR applied to the universe as a whole: neither a brute-fact universe nor a self-explanatory necessary being is obviously more acceptable, so Leibniz sharpens the question without decisively settling it.

Typical mistakes

Treating Leibniz's argument as relying on a finite past; its point is that even an INFINITE series is a contingent whole requiring a sufficient reason.
Assuming the PSR is obviously true; whether it applies to the TOTALITY of contingent things is exactly what is in dispute (the brute-fact reply).

Active revision

'The existence of the universe requires an explanation beyond itself.' Discuss. (25-mark essay)

Active recall

Recall the key points — then reveal.

Sources: GCE AS and A level subject content for philosophy (Department for Education)

Objections: Hume and Russell#

AdvancedAQA 7172 3.3 Metaphysics of God — Cosmological argumentsDfE GCE Philosophy subject content

The fallacy of composition objection

Fig. 5That each part has a cause does not entail the whole has one; the theist replies the contingent whole needs a reason.

Hume and Russell mount the classic objections to cosmological arguments, targeting the causal principle, the coherence of a necessary being, and the demand for an explanation of the whole. The first is the FALLACY OF COMPOSITION objection: the cosmological argument infers that because each PART of the universe has a cause (or reason), the WHOLE universe has a cause (or reason). But this is fallacious — what is true of the parts need not be true of the whole (each brick is small, but the wall is not small; each player has a mother, but the team does not have a mother). So even if every contingent thing IN the universe has a cause, it does not follow that the universe AS A WHOLE has a cause; the totality may need no cause at all. Russell put the point in his 1948 radio debate with Copleston: 'every man who exists has a mother, but that does not mean the human race has a mother' — the move from parts to whole is illegitimate.

The theist replies that the argument need not commit the fallacy: Leibniz's version does not reason from parts to whole by composition, but applies the PSR DIRECTLY to the contingent totality (a contingent whole needs a reason in its own right), and Copleston argued that a series of contingent beings is itself contingent and so requires an explanation independently of any inference about parts. Whether this escapes the fallacy is disputed: the critic responds that applying the PSR to the totality simply ASSUMES the totality needs an explanation, which is the very point at issue. So the fallacy-of-composition objection and the brute-fact reply are two faces of one dispute about whether the universe as a whole requires or admits explanation.

The second objection concerns the NECESSARY BEING. Hume argues that the notion of a necessarily existing being is incoherent: 'whatever we can conceive as existent, we can also conceive as non-existent', so there is no being whose non-existence is inconceivable — necessary existence, if it means logically necessary existence, cannot apply to any being, since existential claims are always deniable without contradiction. If Hume is right, the cosmological argument's conclusion (a necessary being) is impossible, and the argument fails. The theist replies (with Aquinas and Leibniz) that the relevant necessity is METAPHYSICAL, not logical — a being that does not depend on anything else and cannot fail to exist given its nature — which is not obviously incoherent; but critics maintain that metaphysical necessity is obscure and that Hume's point about the deniability of existence still bites.

The third is Russell's BRUTE-FACT reply, the culmination of the debate: the universe may simply be a brute fact — 'the universe is just there, and that's all' — with no cause and no sufficient reason, so the whole demand that drives the cosmological argument (that the universe's existence must be explained) is unwarranted. Russell also charged that 'necessary' can only properly apply to analytic propositions, not to beings, reinforcing Hume. Evaluation weighs these against the theist's case. The fallacy-of-composition objection is answered by Leibnizian versions that apply the PSR to the whole directly, but this only relocates the dispute to whether the PSR governs the totality. The necessary-being objection is met by distinguishing metaphysical from logical necessity, but at the cost of relying on a notion critics find obscure. The brute-fact reply is the deepest: it does not refute the argument but denies its motivating premise, so the argument and the objection reach a stand-off over the intelligibility and explicability of the universe as a whole. A defensible A-Level judgement is that the cosmological argument is more robust than the design and ontological arguments (it is untouched by Darwin and by Kant's 'existence is not a predicate'), and that its best form (Leibniz) isolates a genuine and profound question — why is there something rather than nothing? — but that it does not amount to a proof: it relies on the PSR applied to the totality, which the brute-fact reply rationally resists, and it reaches at most a necessary first cause, whose identification with the personal God of theism is a further step. The argument therefore keeps theism a live rational option and poses a challenge naturalism must answer, without compelling assent — a verdict that feeds directly into the cumulative-case strategy and, by contrast of standards, into the problem of evil.

Worked example

Model 25-mark plan: does the cosmological argument succeed?

Plan an essay defending a judgement on whether the cosmological argument succeeds.

Thesis
The cosmological argument does not prove God, but its Leibnizian form poses a genuine challenge; it fails as a proof because the brute-fact reply rationally resists the PSR.
AO1
Reconstruct the Kalam, Aquinas' Third Way, and Leibniz's PSR argument, distinguishing temporal from sustaining series.
Objections
Fallacy of composition (Russell), the incoherence of a necessary being (Hume), and the brute-fact reply.
Replies
Leibniz applies the PSR to the whole (dodging composition); metaphysical (not logical) necessity answers Hume; but each reply relocates rather than removes the dispute.
Judgement
Conclude that the argument keeps theism rational and poses 'why is there something rather than nothing?', but does not compel assent (the PSR for the totality is deniable) and reaches only a necessary first cause, not the personal God.

Result: A defensible judgement: the cosmological argument is the most robust theistic argument (untouched by Darwin and Kant) and isolates a profound question, but it fails as a proof because it rests on the PSR applied to the totality, which the brute-fact reply rationally resists.

Typical mistakes

Assuming every cosmological argument commits the fallacy of composition; Leibniz's PSR version applies to the totality directly, so the objection must be pressed as 'why does the whole need a reason?'.
Confusing logical and metaphysical necessity; Hume's objection targets LOGICAL necessity, and the theist's necessary being is usually METAPHYSICALLY necessary.

Active revision

'The cosmological argument fails because it commits the fallacy of composition.' Is this claim true? Defend your answer. (25-mark essay)

Active recall

Recall the key points — then reveal.

Sources: GCE AS and A level subject content for philosophy (Department for Education)

Key points

Identifying the three joints

Key points

Testing the Kalam's premises

Key points

The quantifier-shift objection to the Third Way

Key points

The brute-fact objection to the PSR

Key points

Model 25-mark plan: does the cosmological argument succeed?

Sources

Cosmological arguments

Key points

Identifying the three joints

Key points

Testing the Kalam's premises

Key points

The quantifier-shift objection to the Third Way

Key points

The brute-fact objection to the PSR

Key points

Model 25-mark plan: does the cosmological argument succeed?

Sources