Arguments from Consciousness and Reason

A cluster of more recent and more theoretical theistic arguments turns not on the existence or order of the physical world but on the existence of minds within it — on the fact that the universe contains consciousness, and that it contains reason reliable enough to grasp truth. The thought is that subjective experience and rational cognition are precisely the features of reality that a purely physical, undirected process is least equipped to produce or to vindicate, and that they are far more at home in a theistic universe, where mind is fundamental, than in a naturalistic one, where it is a late and accidental by-product. These arguments thus engage directly the philosophy of mind and the standing of naturalism.

This page sets out the argument from consciousness (Swinburne), the argument from reason (Lewis), and its sharpest modern descendant, Plantinga's evolutionary argument against naturalism (EAAN), together with the principal naturalist replies.

The argument from consciousness

Richard Swinburne offers consciousness as a strand in his cumulative case. The argument exploits the notorious hard problem of consciousness — the apparent unbridgeable gap between physical facts (neurons, electrochemistry) and phenomenal facts (what it is like to see red, to feel pain). In outline:

1. Conscious mental states — qualia, subjective experience — exist and are not (the argument claims) reducible to or entailed by physical facts: a complete physical description of the brain leaves out what the experience is like.
2. Their correlation with brain states is therefore not something physical science can explain; the psychophysical laws linking matter to mind are not derivable from physics and are, from a naturalistic standpoint, brute, contingent, and extraordinarily specific.
3. A personal explanation is available and better: a God who values conscious agents has reason to bring about, and to connect by reliable laws, brains and the rich mental lives they support.
4. So the existence of consciousness, and the orderly mind–body correlation, is more probable given theism than given naturalism, and confirms theism.

The argument is strongest if some form of dualism or property dualism is correct — if consciousness is genuinely non-physical. The naturalist replies along familiar lines from the philosophy of mind: physicalism (consciousness is a physical/functional property, the hard problem a confusion), emergence (consciousness arises naturally from sufficiently organised matter, no God required), or panpsychism (experience is a fundamental feature of matter itself — increasingly taken seriously, and a naturalistic way to make mind fundamental without God). If any of these succeeds, premise 1 or 2 fails. The dispute thus inherits the unresolved state of the mind–body problem: the theist needs the hard problem to be intractable for naturalism, which is contested.

The argument from reason

A distinct and in some ways deeper argument concerns not experience but rationality — and it is a self-undermining argument against naturalism rather than a direct argument for God. C.S. Lewis (Miracles, 1947) gave the popular form; its ancestry runs through Balfour and Haldane:

If naturalism is true, then every belief — including the belief in naturalism — is the product of non-rational causes (ultimately, blind physics and chemistry in the brain). But a belief produced wholly by non-rational causes, with no regard for whether it is true, cannot be rationally held or count as knowledge. So if naturalism is true, no belief is rationally grounded — including naturalism itself. Naturalism is therefore self-refuting.

The key move is the distinction between the cause of a belief (the physical process that produces it) and its ground (the reasons that justify it). Reasoning requires that one belief be caused by another in virtue of a logical relation between their contents — that we believe a conclusion because it follows. But on naturalism, mental states have causal efficacy (if any) only through their physical properties, not their semantic content; the meaning of a thought seems causally inert. If so, no belief is ever held because of the reasons for it, and rational inference — the very thing the naturalist relies on to argue for naturalism — is an illusion.

Plantinga's EAAN

Alvin Plantinga sharpened the argument from reason into a rigorous evolutionary argument against naturalism (EAAN). The target is the conjunction of naturalism (N) and evolution (E), and the claim is that this conjunction is self-defeating. The argument concerns the probability that our cognitive faculties are reliable (R) — that they generally produce true beliefs — given N and E:

$$P(R\mid N\&E)\text{is low, or inscrutable.}$$

The reasoning: natural selection rewards adaptive behaviour, not true belief. Selection "sees" only what an organism does, not what it believes; belief influences fitness only via its effect on behaviour. But behaviour is a product of belief together with desire, and indefinitely many belief–desire pairs yield the same adaptive behaviour. Plantinga's example: a creature that flees tigers may do so because it believes the tiger dangerous and wants to live, or because it believes the tiger is a cuddly pet it wants to pet but believes the best way to pet it is to run away — the behaviour (fleeing) is identical and equally fitness-enhancing, while the beliefs are wildly false. Since adaptiveness underdetermines truth, there is no reason to expect selection to have tuned our belief-forming faculties for truth rather than mere fitness; so $P(R\mid N\&E)$ is low or cannot be assessed.

The argument then turns reflexive — this is the "self-defeat":

1. If $P(R\mid N\&E)$ is low, then anyone who accepts $N\&E$ has a defeater for $R$ — a reason to doubt that their own faculties are reliable.
2. But a defeater for $R$ is a defeater for every belief those faculties produce — including belief in $N\&E$ itself.
3. So $N\&E$ cannot be rationally accepted: it saws off the branch it sits on.

Crucially, the theist faces no parallel defeater: a God who creates us in his image, for knowing, gives reason to expect reliable faculties, so $P(R\mid \text{theism})$ is high. The EAAN does not directly prove God; it argues that naturalism is irrational to hold, leaving theism as the worldview that can underwrite the trust in reason that everyone, including the naturalist, must presuppose.

Naturalist replies

The EAAN has generated a large critical literature; the main lines:

• Truth-tracking is itself adaptive. Against the underdetermination claim: in cognitively sophisticated, flexible creatures, generally true beliefs are by far the most reliable route to adaptive behaviour across the open-ended range of situations evolution must handle. The "cuddly tiger" stories work for one behaviour but multiply into incoherence across a whole web of beliefs and novel circumstances; a piecemeal-false belief system would mislead constantly. So selection does, indirectly but robustly, favour reliable faculties — $P(R\mid N\&E)$ is high. (Plantinga's rejoinder: this assumes content is causally relevant to behaviour, which on naturalism — especially given worries about the causal efficacy of semantic content — is exactly what is in question.)
• Local reliability suffices, and is all we have anyway. Even granting global scepticism is possible, our faculties are demonstrably reliable enough in the domains we test (science works, predictions succeed); the bare probability on $N\&E$ is outweighed by this first-order evidence. (Plantinga: using the faculties to vindicate the faculties is the very circularity the defeater blocks.)
• The defeater is self-defeating or over-general (Fitelson–Sober and others): the argument's own probabilistic premises must be grasped by the faculties it impugns; and a parallel argument could be mounted against any metaphysics, including theism (could a deceiving or indifferent process not equally afflict a theist's faculties?). The conditional-probability claims, critics argue, are far more inscrutable than Plantinga allows — which, if so, blocks the inference to a low value.
• Companions in guilt. The argument proves too much: if it works, it threatens to make all empirical reasoning rest on a prior, un-argued trust in our faculties — which is just the familiar problem of the criterion / epistemic circularity, not a special defect of naturalism.

Where these arguments sit

The arguments from consciousness and reason represent the theoretical wing of natural theology: where the cosmological and teleological arguments point to the world's existence and order, these point inward, to the existence of minds and the trustworthiness of thought, and contend that mind is more intelligible if it is fundamental (theism) than if it is derivative (naturalism). The argument from consciousness is hostage to the mind–body problem: it is forceful only if physicalism, emergence, and panpsychism all fail to explain experience, which is unsettled. Plantinga's EAAN is the most discussed and most ingenious — a genuine reductio attempt that, if sound, would make naturalism self-refuting and so do more than any cumulative strand could — but it stands or falls on the contested claim that natural selection would not favour reliable belief, and on the prior worry about whether semantic content can be causally efficacious at all. Neither argument is a proof; both function, like the others, as strands in a cumulative case — and both turn, in the end, on questions in the philosophy of mind as much as in the philosophy of religion.