Teleological (design) arguments

Design qua purpose and design qua regularity#

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

Two strands of design

Fig. 1Design arguments draw on contrivance (means adjusted to ends) and on the lawful regularity of nature.

Teleological arguments (from the Greek telos, 'end' or 'purpose') reason from apparent purposiveness or order in the world to an intelligent designer. Unlike the ontological argument, they are a posteriori (starting from observed features of the world) and inductive (concluding to a designer as the best or most probable explanation, not as a logically guaranteed one). This character is crucial for evaluation: a design argument is not refuted by showing it is not a deductive proof (it never claimed to be), and it is to be assessed as we assess any inference to the best explanation — by whether the designer hypothesis explains the data better than its rivals.

It is useful to distinguish two strands the arguments draw on. Design qua PURPOSE points to CONTRIVANCE — the way parts of natural things are adjusted to serve ends, as the parts of a machine are fitted to a function. The human eye, whose lens, retina and muscles are arranged so as to produce sight, is the stock example: such an arrangement of means to an end looks like the work of purpose, not chance. Design qua REGULARITY points to the ORDER of nature as a whole — the lawful, mathematically describable regularity of the cosmos (the orbits of the planets, the constancy of natural laws) — which seems to call for an ordering intelligence. Paley emphasises the first; the argument from the order of the heavens emphasises the second.

The arguments have ancient and scholastic roots. Aquinas' Fifth Way (design qua regularity) observes that natural bodies lacking intelligence nonetheless act towards an end, achieving their goal regularly and 'for the best'; but a thing without awareness can be directed to an end only by something with awareness, as an arrow is directed by an archer; so there is an intelligent being directing all natural things to their end, and this we call God. Note that Aquinas' argument is not about contrivance (machines) but about the goal-directedness of unconscious nature — a point often confused with Paley's watch.

The design argument's enduring appeal is that it engages a powerful and natural intuition: complex, functional order cries out for explanation, and intelligent design is the kind of explanation we routinely and reasonably infer in daily life (we infer a watchmaker from a watch, an author from a book). The philosophical question is whether that inference, so compelling for human artefacts, legitimately transfers to the natural world and the cosmos as a whole — which is precisely where Hume attacks the analogy and Darwin offers a rival explanation of biological order. The topic is therefore a case study in the strengths and limits of inference to the best explanation.

Worked example

Classifying design considerations

Classify each as design qua purpose or design qua regularity: (a) the eye is structured for sight; (b) the physical constants permit a stable, law-governed universe.

Assess (a)
The eye's parts are adjusted to an end (sight) — contrivance — so this is design qua purpose (Paley).
Assess (b)
The constants underwrite the lawful order of the cosmos — regularity — so this is design qua regularity (and the fine-tuning version).
Note the inductive form
Each infers a designer as the best explanation of the feature, not as a logical necessity.
Anticipate the attack
Purpose-order in biology faces Darwin; regularity/fine-tuning faces the multiverse — different objections for different strands.

Result: (a) is design qua purpose, (b) is design qua regularity; separating them matters because Darwin targets biological purpose while the multiverse targets cosmic fine-tuning.

Typical mistakes

Attributing the watch analogy to Aquinas; the watch is Paley's, and Aquinas' Fifth Way is about unconscious things acting towards ends, not about contrivance.
Presenting the design argument as a deductive proof; it is an inductive inference to the best explanation, and treating it as a proof invites easy refutation.

Active revision

Explain the difference between design qua purpose and design qua regularity, with an example of each. (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 argument from analogy and Paley's watch#

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

Paley's argument from analogy

Fig. 2From the contrivance of the watch to a watchmaker, and by analogy from the world's contrivance to a designer.

In Hume's Dialogues Concerning Natural Religion, the character Cleanthes states the design argument as an argument from analogy: the world is like a vast machine, composed of intricate sub-machines whose parts are adjusted to one another with an accuracy that 'ravishes into admiration all men who have ever contemplated them'; machines are produced by intelligent design; like effects have like causes; so the world, resembling a machine, must have a cause resembling a designing intelligence, proportioned to the grandeur of the work — a mind, that is, like the human mind but far greater. Cleanthes' version is explicitly analogical and a posteriori, and Hume puts it forward precisely so that Philo can attack it.

William Paley, in Natural Theology (1802), gives the argument its most famous form with the watch analogy. Suppose, crossing a heath, I find a stone; I might think it had always lain there. But if I find a watch, and examine its intricate mechanism — the spring, the wheels, the hands, all adjusted and combined to produce regular motion and tell the time — I must infer that it had a maker who designed it for that purpose. Crucially, Paley insists this inference holds EVEN IF I have never seen a watch made, EVEN IF the watch sometimes goes wrong, and EVEN IF there are parts whose function I do not understand: the marks of contrivance are enough to establish a designer. Paley then argues that the natural world exhibits contrivance of exactly this kind, only far greater — the eye is a more exquisite contrivance for seeing than a telescope is — so, by the same inference, the natural world must have a designer, whom he identifies as God.

Paley's argument is carefully constructed to pre-empt objections. The 'even if it sometimes goes wrong' clause anticipates the problem of imperfect or dysfunctional adaptations; the 'even if I have never seen one made' clause anticipates the objection that we have no experience of worlds being made; the 'even if I do not understand every part' clause anticipates gaps in our knowledge. Paley's point is that the INTERNAL evidence of contrivance — parts adjusted to a purpose — is sufficient for the inference to a designer, independently of these considerations. This makes his version more robust than a crude 'the world is complex, so God made it'.

Reconstructed formally, the argument runs: (P1) the natural world (and its parts, e.g. the eye) exhibits complex order in which means are precisely adjusted to ends. (P2) In every case whose origin we know (the watch, the telescope), such adjustment of means to ends is the product of intelligent design. (P3) Like effects have like causes (similar features warrant inferring similar causes). Therefore (C) the order of the natural world is probably the product of an intelligent designer, namely God. The argument is a posteriori and inductive; its whole strength depends on how good the analogy between natural objects and human artefacts really is — the natural world resembling a machine closely enough to warrant inferring a machine-maker. It is exactly this analogy that Hume's Philo dismantles, and exactly this biological contrivance that Darwin explains without a designer, as the next two sections show.

Worked example

Reconstructing Paley's argument

Set out Paley's design argument as premises and conclusion, and identify the premise the objections target.

P1
Natural things (the eye) show means precisely adjusted to ends — contrivance.
P2
In every case we know the origin of (the watch), such contrivance is the product of design.
P3 and C
Like effects have like causes; so the natural world's contrivance is probably the product of a designer, God.
Target premise
The analogy in P2/P3 — that natural contrivance is relevantly LIKE artefactual contrivance — is what Hume attacks and Darwin undercuts.

Result: The argument is a valid inductive inference whose force rests on the world-machine analogy; its evaluation is therefore the evaluation of that analogy, which Hume weakens and Darwin explains away for the biological case.

Typical mistakes

Presenting Paley as arguing merely 'the world is complex, so it has a designer'; his argument turns on CONTRIVANCE — means adjusted to ends — not mere complexity.
Forgetting the 'like effects, like causes' premise; the analogy is the load-bearing element, and the objections target its strength.

Active revision

Outline Paley's design argument and explain the analogy it relies on. (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)

Hume's objections#

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

Hume's objections mapped

Fig. 3Hume attacks the analogy, the inference to the God of theism, and the assumption that order needs design.

In the Dialogues (published 1779), Hume's character Philo mounts a battery of objections that remain the classic critique, and, strikingly, they anticipate Darwin by eighty years. The first is that the analogy between the world and a machine is WEAK. The inference 'like effects, like causes' is only as strong as the likeness, and the universe is not much like a human artefact: it resembles an organism (which grows) at least as much as a machine (which is built), and if we pressed the organic analogy we would infer that the world was generated or grew, not manufactured by a mind. Since the analogy can be run in incompatible directions, it does not reliably support a designer.

Second, Philo argues that we are not entitled to infer the God of THEISM even if we grant a designer. The principle 'the cause must be proportioned to the effect' licenses inferring only a cause adequate to THIS world — no more. So we cannot infer an INFINITE designer (a finite world needs only a finite cause), nor a PERFECT designer (the world contains flaws and waste), nor a SINGLE designer (a great ship or house is built by many workers, so many gods might have made the world), nor an incorporeal or eternal one, nor even a still-living one (the designer may have died, or the world be the botched first attempt of an infant deity). The design argument, even if it works, underdetermines its conclusion drastically: it gets us at most 'a cause or causes of order in the universe that probably bear some remote analogy to human intelligence', not the God of religion.

Third, Philo presses the EPICUREAN hypothesis (the ancestor of the Darwinian point): order can arise WITHOUT design, by chance over vast time. In an eternal universe with a finite number of particles endlessly rearranging, order-producing configurations will eventually and inevitably occur and, once they occur, persist (unstable arrangements collapse; stable, self-perpetuating ones remain), giving the APPEARANCE of design without a designer. So apparent design is not decisive evidence FOR a designer, since a chance-plus-selection process could produce it. Fourth, Philo notes we have no experience of the ORIGIN of universes: causal inference (Hume's own theory) works only where we have observed constant conjunctions of cause and effect, but the universe is unique and its origin unobserved, so we cannot reliably infer its cause at all — the design argument extends causal reasoning beyond the domain in which it is legitimate.

Evaluation weighs these against the theist's replies. The weak-analogy objection is answered by insisting the machine analogy is apt for the specific feature of means-adjusted-to-ends (contrivance), which organisms also display and which cries out for explanation. The underdetermination objection is powerful, but the theist may reply that the design argument was only ever meant to establish A designer, with the FULL concept of God supplied by the other arguments and by revelation (the cumulative case); design does not have to prove theism single-handed. The Epicurean/chance objection is the deepest, and it is precisely what Darwin vindicated for biology — but the theist replies that chance alone is wildly improbable, and (crucially) that the Epicurean point does not touch the FINE-TUNING of the laws and constants themselves, which is the modern refuge of the argument (next sections). Hume's 'no experience of universes' objection is strong against a confident inference but arguably too strong, since it would forbid all cosmology and all reasoning about unique events. A balanced verdict is that Hume decisively weakens the argument as a PROOF and as an inference to the specific God of theism — the underdetermination and Epicurean objections are serious and largely stand — but that he does not show apparent order is worthless as evidence, leaving room for a probabilistic version (Swinburne) focused on the order Darwin cannot explain.

Worked example

The underdetermination objection

Explain Hume's objection that the design argument cannot establish the God of classical theism, and one reply.

The principle
We may infer only a cause PROPORTIONED to the effect — no more than the evidence supports.
Apply it
A finite, flawed world warrants at most a finite, imperfect designer; multiple builders suggest multiple gods; the designer might be dead or an apprentice.
The conclusion
So the argument yields at most 'some cause bearing remote analogy to intelligence', not the infinite, perfect, single God of theism.
A theistic reply
Design need not prove theism alone; it establishes A designer, and the other arguments plus revelation fill in the attributes (the cumulative case).

Result: The underdetermination objection stands as a limit on what design alone can prove — not the full God of theism — but a cumulative-case theist can accept this and treat design as one strand among several.

Typical mistakes

Attributing all the objections to 'Hume' as author; they are voiced by the character Philo in the Dialogues, and Cleanthes states the argument — cite the dialogue form.
Treating the underdetermination objection as showing there is NO designer; it shows only that the argument cannot establish the SPECIFIC God of theism (infinite, perfect, one), not that design fails entirely.

Active revision

'Hume's objections destroy the design argument.' 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)

The Darwinian objection#

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

Design versus natural selection

Fig. 4Natural selection explains biological contrivance without a designer, but not the laws or constants themselves.

Charles Darwin's theory of evolution by natural selection (On the Origin of Species, 1859) provides what many regard as the decisive objection to the design argument's central case — biological contrivance. Paley's strongest evidence was the exquisite adaptation of organisms: the eye adjusted for sight, the wing for flight, the hand for grasping. Natural selection explains exactly this apparent design WITHOUT any designer: organisms vary heritably; more offspring are produced than can survive; those variations that happen to confer a survival or reproductive advantage are preserved and passed on, while disadvantageous ones are eliminated; over vast time this blind, cumulative process produces the intricate adaptations that Paley took to be marks of a designing mind. The eye was not designed; it evolved, step by advantageous step.

The force of the objection is that it supplies a RIVAL explanation of the very data the design argument relies on, and a better one by the ordinary standards of inference. Natural selection is a naturalistic, testable, well-confirmed mechanism that explains adaptation, its imperfections (the blind spot, the awkward routing of the recurrent laryngeal nerve — 'unintelligent design'), its gradual history in the fossil record, and the tree of common descent, none of which a designer hypothesis predicts or explains as economically. Where Paley inferred a designer because he could imagine no other cause of contrivance, Darwin identified the other cause; so the inference 'contrivance, therefore design' is broken — contrivance can arise by selection. This is the vindication of Hume's Epicurean hypothesis: order without a mind, produced by chance variation plus non-random selection.

The theist has several replies, of varying strength. The weakest is to reject evolution, which is not a serious philosophical option given the evidence. A better reply distinguishes the domains: Darwin explains BIOLOGICAL adaptation, but not the origin of the LAWS that make evolution possible, nor the fine-tuning of the physical constants, nor the existence of a law-governed, life-permitting universe in the first place — so design retreats from biology to cosmology (the next section). A third reply is THEISTIC EVOLUTION: God designed the universe and its laws so that life would evolve by natural selection — evolution is the MEANS by which God's design is realised, not a rival to it. On this view Darwin explains the mechanism, not the ultimate purpose, and God is the author of the whole process, including the laws of variation and selection.

Evaluation weighs whether these replies preserve a viable design argument. Against them, the theistic-evolution reply concedes that biological adaptation is no longer EVIDENCE for a designer (it is fully explained by selection), so it saves theism's compatibility with the data but abandons the design INFERENCE from biology; and Richard Dawkins presses that natural selection makes a designer explanatorily redundant for the biological world, and that a designer complex enough to design the universe would himself require explanation ('who designed the designer?'). For the theist, the honest upshot is that Darwin defeats the design argument in its historically central, biological form — Paley's eye is explained without God — but does NOT touch the fine-tuning of the laws and constants, which is why the modern design argument relocates to cosmology. A defensible A-Level judgement is that the biological design argument fails after Darwin (natural selection is the better explanation of adaptation), so the argument survives, if at all, only as a cosmic fine-tuning argument — which faces its own rival, the multiverse.

Worked example

Natural selection as the better explanation

Explain why natural selection is a better explanation of the eye than design, and state the theist's best reply.

The data
The eye's parts are exquisitely adjusted for sight — Paley's prime evidence of contrivance.
The rival mechanism
Heritable variation plus differential survival gradually accumulates advantageous changes, building the eye step by step, no mind required.
Why it is better
It is testable, well-confirmed, and explains imperfections, gradual history and common descent, which design does not economically predict.
The theist's reply
Theistic evolution: God authored the laws of variation and selection, so evolution is the MEANS of design — but this concedes biology no longer EVIDENCES a designer.

Result: Natural selection out-explains design for biological adaptation, so the biological design argument fails; the theist can retreat to theistic evolution (compatibility, not evidence) or to cosmic fine-tuning, which Darwin does not touch.

Typical mistakes

Saying evolution disproves God; it removes the biological EVIDENCE for a designer, but is compatible with theistic evolution — the point is explanatory redundancy, not disproof.
Treating Darwin as answering the fine-tuning argument; natural selection explains biological adaptation, not the physical constants or the laws themselves.

Active revision

'Darwin's theory of evolution defeats the design argument.' 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)

Fine-tuning and Swinburne#

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

Fine-tuning: designer or multiverse?

Fig. 5A life-permitting universe is explained either by a designer setting the constants or by selection within many universes.

The modern design argument relocates from biological contrivance (defeated by Darwin) to the fine-tuning of the fundamental physical constants and laws. Physics reveals that the constants — the strength of gravity, the cosmological constant, the mass ratios of particles, the strength of the nuclear forces — appear to be 'fine-tuned': had they differed even slightly, the universe would not have permitted stars, chemistry, or life. The improbability that a randomly configured universe would be life-permitting is, on some estimates, extraordinary. The fine-tuning argument infers that this life-permitting arrangement is better explained by a designer who set the constants so as to permit life than by chance, since a designer would WANT a life-permitting universe whereas chance would almost certainly produce a lifeless one.

Richard Swinburne casts this as a probabilistic (Bayesian) inference to the best explanation, not a proof. He argues that theism is a SIMPLE hypothesis — it posits one being of infinite power, knowledge and goodness — and that a universe which is orderly (law-governed) and fine-tuned for conscious life is far more probable given theism than given the atheistic hypothesis of blind chance. A good explanation raises the probability of the evidence; theism makes the existence of a law-governed, life-permitting, consciousness-bearing universe much more likely than it would otherwise be, so that evidence CONFIRMS theism. Swinburne does not claim design proves God, only that it raises the probability of God, contributing to a cumulative case (with the cosmological argument, religious experience, etc.) that makes theism, on the total evidence, more probable than not.

The principal objection is the MULTIVERSE (or 'many universes') hypothesis: if there exist vastly many universes with randomly varying constants, then it is unsurprising that at least one is life-permitting — and, by an observer-selection (anthropic) effect, we necessarily find ourselves in a life-permitting one, since only such a universe could contain observers. On this view fine-tuning needs no designer: it is a selection effect within a large ensemble, exactly parallel to how Darwin explained biological adaptation by selection within a population. Just as the eye no longer needs a designer given natural selection, the fine-tuned universe no longer needs one given many universes plus the anthropic principle.

Evaluation weighs the designer hypothesis against the multiverse. In the theist's favour: the multiverse posits an enormous, unobservable ensemble of universes, which some argue is a less simple and less economical hypothesis than a single God (a violation of Ockham's razor), and which still leaves the meta-question of why there is a universe-generating mechanism fine-tuned to produce life-permitting universes at all. There is also the objection that the anthropic principle explains why WE observe a life-permitting universe but not why one EXISTS to be observed (the 'firing squad' analogy: surviving a hundred rifles all missing does explain why you are alive to wonder, but still cries out for a further explanation). In the multiverse's favour: it is a serious hypothesis with independent motivations in physics (inflationary cosmology, string theory), it is naturalistic, and it avoids the 'who designed the designer?' regress; and critics question whether the probabilities the fine-tuning argument invokes are even well-defined (we have one universe and no sample). A balanced A-Level judgement is that the fine-tuning argument is the strongest surviving form of the design argument — it concerns exactly the cosmic order Darwin cannot explain, and it is honestly framed by Swinburne as probabilistic rather than a proof — but that it is not decisive: the multiverse is a genuine rival explanation, the relevant probabilities are contested, and Hume's underdetermination objection (why the God of theism rather than some lesser designer?) still applies. The considered verdict is that design arguments do not PROVE God, and that Darwin defeated their biological form, but that the fine-tuning version remains a real, if inconclusive, probabilistic consideration in favour of theism, whose weight depends on how one assesses the multiverse and the intelligibility of the probabilities involved — a judgement that links directly to the cumulative-case strategy and to the cosmological argument.

Worked example

Model 25-mark plan: does fine-tuning support theism?

Plan an essay defending a judgement on whether fine-tuning supports the existence of God.

Thesis
Fine-tuning gives genuine but inconclusive probabilistic support to theism; it is the strongest design argument but not decisive against the multiverse.
AO1
Explain fine-tuning, Swinburne's Bayesian framing (theism simple, raises the probability of a life-permitting universe), and its contrast with Paley/Darwin.
For
Design better explains the specific, improbable life-permitting order Darwin cannot touch; the multiverse is a vast, unobservable, arguably less simple posit.
Against
The multiverse plus the anthropic principle explains fine-tuning naturalistically; the probabilities may be ill-defined; Hume's underdetermination still applies.
Judgement
Conclude that fine-tuning raises theism's probability within a cumulative case but does not prove God, since the multiverse is a serious rival and the argument underdetermines the designer.

Result: A defensible judgement: after Darwin the design argument survives only as fine-tuning, which gives real but inconclusive probabilistic support to theism — its force turning on the multiverse's plausibility and the intelligibility of the probabilities.

Typical mistakes

Treating fine-tuning as a deductive proof; Swinburne explicitly offers a probabilistic inference to the best explanation within a cumulative case.
Assuming the multiverse obviously wins or obviously loses; the debate turns on contested points (simplicity/Ockham, whether the anthropic principle explains existence, the definability of the probabilities) that must be weighed.

Active revision

'The fine-tuning of the universe is best explained by a designer.' 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

Classifying design considerations

Key points

Reconstructing Paley's argument

Key points

The underdetermination objection

Key points

Natural selection as the better explanation

Key points

Model 25-mark plan: does fine-tuning support theism?

Sources

Teleological (design) arguments

Key points

Classifying design considerations

Key points

Reconstructing Paley's argument

Key points

The underdetermination objection

Key points

Natural selection as the better explanation

Key points

Model 25-mark plan: does fine-tuning support theism?

Sources