Notes · Religious StudiesUK · A-Levels

Religious language

Is language about God meaningful? This topic examines the challenge that religious statements are neither true nor false but literally meaningless, through the verification (Ayer) and falsification (Flew) debates, and the cognitivist and non-cognitivist responses — analogy (Aquinas), symbol (Tillich), the via negativa, and Wittgenstein's language games — before evaluating whether religious language succeeds in saying anything.

5 sections~23 min reading time2 competenciesLevel Foundation 1 · Standard 2 · Advanced 2

T·0555 / 16

Exam profile

AO1 · Demonstrate knowledge and understanding of cognitive and non-cognitive theories of religious language and the verification and falsification debatesAO2 · Analyse and evaluate whether religious language is meaningful, reaching a justified judgement

Operators:explainanalyseassessevaluate'...' Discussto what extent

basic level

At AS-Level, explain the verification principle and one cognitivist response such as analogy.

higher level

At full A-Level, evaluate the verification and falsification challenges against the full range of cognitivist and non-cognitivist responses and judge whether religious language is meaningful.

Depth

Text

Contents · 5 sections

Religious language

The problem: cognitive and non-cognitive language#

FoundationAQA 7062/1 Section A: Religious languageDfE GCE Religious Studies subject content

Theories of religious language

Fig. 1Theories divide into cognitivist (fact-asserting) and non-cognitivist (expressive) families.

The problem of religious language is not whether statements about God are true, but the prior and more radical question of whether they are meaningful at all — whether 'God is love' or 'God created the world' succeed in making any assertion that could be true or false, or whether they are, in the logical positivists' phrase, literally nonsense. The problem arises because God is held to be transcendent, infinite and beyond sense experience, yet is spoken of in words whose meaning was learned by applying them to finite, observable things; when we call God 'good' or 'wise' or say he 'acts', it is unclear whether these words retain any content or have been stretched until they mean nothing.

A first framework for the problem distinguishes three ways a word can be used across two subjects. A word is used univocally when it means exactly the same in both cases ('the dog is loyal' and 'the soldier is loyal' — the same loyalty); equivocally when it means something entirely different ('the bank of the river' and 'the money bank' — merely a shared sound); and analogically when it means something related but not identical (a 'healthy' body and a 'healthy' diet — related by causation). The dilemma for religious language is that univocal talk seems to reduce God to the human scale (anthropomorphism), while equivocal talk empties religious statements of content (agnosticism); analogy is offered as the middle way, and much of the topic is the contest between this cognitivist middle way and non-cognitivist alternatives.

The deepest divide, however, is between cognitive and non-cognitive theories of religious language. Cognitivism holds that religious statements are cognitive — they assert facts, make truth-claims, and are true or false ('God exists' says something that is either the case or not). Non-cognitivism holds that religious statements are non-cognitive — they do not assert facts but do something else: express and evoke attitudes, commitments, ways of life, or moral resolves, and so are not the sort of thing that could be true or false in the first place. The verification and falsification challenges assume cognitivism (they test religious statements for factual meaning and find them wanting); the non-cognitive responses (symbol, bliks, language games) sidestep the challenges by denying that religious language was ever in the business of stating facts.

This basic map determines how every position in the topic is to be read. A cognitivist who accepts the verification principle will conclude religious language is meaningless (Ayer); a cognitivist who rejects the principle will defend the factual meaningfulness of religious language by other means (Hick's eschatological verification, Swinburne, analogy). A non-cognitivist will regard the whole verification debate as a category mistake — testing for factual meaning something that never claimed to be factual (Wittgenstein, Phillips, R. M. Hare). Keeping straight whether a thinker is cognitivist or non-cognitivist, and whether they accept or reject verificationism, is the discipline that turns a descriptive answer into an analytical one.

Worked example

Placing a thinker on the map

Where do Ayer, Aquinas and Wittgenstein each stand on the cognitive/non-cognitive divide, and why does it matter?

Ayer
Cognitivist about meaning, applies the verification principle, and concludes religious statements are meaningless — factual talk that fails the test.
Aquinas
Cognitivist: religious statements assert truths about God, made meaningful through analogy rather than univocal or equivocal use.
Wittgenstein
Non-cognitivist: religious language is a distinct language game expressing a form of life, not fact-stating, so verificationism does not apply.
Why it matters
The verification challenge only threatens cognitivist religious language; non-cognitivists escape it by denying the fact-stating premise.

Result: Ayer and Aquinas are cognitivists who disagree about whether religious language has factual meaning; Wittgenstein is a non-cognitivist for whom the challenge misfires — so the divide dictates which objections are relevant.

Typical mistakes

Treating the debate as about whether religious statements are true; the positivist challenge is that they are neither true nor false but meaningless.
Confusing cognitive and non-cognitive positions, e.g. presenting Wittgenstein as trying to prove religious statements true — he denies they are fact-stating at all.

Active revision

Explain the difference between cognitive and non-cognitive views of religious language. (Structured AO1 response)

Active recall

Recall the key points — then reveal.

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

Verificationism (Ayer, Hick)#

StandardAQA 7062/1 Section A: Religious languageDfE GCE Religious Studies subject content

The verification principle

Fig. 2A statement is meaningful only if analytic or empirically verifiable; religious statements are held to be neither.

The verification principle was the weapon of the logical positivists of the Vienna Circle in the 1920s and 30s, and was brought into English philosophy by A. J. Ayer in Language, Truth and Logic (1936). The principle states that a statement is meaningful (factually significant) only if it is either analytic — true or false by definition, like the truths of mathematics and logic — or empirically verifiable, that is, capable of being confirmed or disconfirmed by sense experience. Statements that are neither, Ayer argued, are literally meaningless: they may express emotion or evoke a mood, but they assert nothing. Since 'God exists', 'God loves us' and 'there is life after death' are neither analytic nor open to empirical verification, Ayer concluded that religious statements — along with metaphysics, aesthetics and much of ethics — are literally nonsense.

Ayer distinguished strong from weak verification. Strong verification requires that a statement be conclusively established by experience — a standard so demanding that even universal scientific laws ('all metals expand when heated') and statements about the past fail it, since they can never be conclusively verified. Recognising this, Ayer adopted weak verification: a statement is meaningful if experience can render it probable, if some observations would count for or against it. This saves science and history, but Ayer maintained that religious statements fail even the weak test, because no possible observation would count for or against 'God exists' understood as a claim about a transcendent being. Ayer thus held that the believer and the atheist are not really disagreeing about a fact at all.

The verification principle faced decisive objections, several of them turning it on itself. The most famous is that the principle is self-refuting: 'a statement is meaningful only if analytic or empirically verifiable' is itself neither analytic (not true by definition) nor empirically verifiable (no observation confirms it), so by its own standard it is meaningless. Ayer could only reply that it was a definition or recommendation, which weakens its authority to rule out religion. Further, the principle was criticised as arbitrarily narrow — it excludes not only religion but ethics, history's unrepeatable events, and other minds — and as failing to capture how religious language actually functions for believers.

John Hick offered the most important reply on behalf of cognitive religious language: eschatological verification. In his parable of the Celestial City, two travellers walk the same road, one believing it leads to a city, the other not; nothing on the journey settles their dispute, yet it is a genuine factual dispute, for at journey's end one will be proved right. Similarly, Hick argues, 'there is life after death' and, ultimately, 'God exists' are verifiable in principle — after death, if the theist is right, the believer will experience the fulfilment that confirms the claim (though the atheist, if right, will experience nothing). Religious statements are therefore not unverifiable in principle, only unverifiable now, and so are meaningful factual claims. Critics object that the verification is one-sided (only confirmable, never falsifiable, since if there is no afterlife no one will be there to note it) and depends on the contested coherence of post-mortem existence.

Worked example

The self-refutation objection

Explain why the verification principle is often said to be self-refuting, and how this bears on religious language.

State the principle
Meaningful iff analytic or empirically verifiable.
Apply it to itself
Is the principle analytic? No — it is not true by definition. Is it empirically verifiable? No observation confirms it.
Draw the consequence
By its own standard the principle is meaningless, so it cannot authoritatively rule anything out — including religion.
Ayer's fallback
Ayer calls it a definition/recommendation, but a mere recommendation has no power to prove religious language nonsense.
Bearing on religion
If the principle collapses, the strongest positivist case against religious language collapses with it.

Result: The verification principle fails its own test and so cannot demonstrate that religious language is meaningless; this self-refutation is the main reason the positivist challenge is no longer regarded as decisive.

Typical mistakes

Muddling verification (meaningful if confirmable) with falsification (meaningful if disconfirmable) — Ayer's is a verification principle.
Forgetting the self-refutation objection, which is the standard reason the principle is now widely rejected.

Active revision

'The verification principle successfully shows that religious language is meaningless.' Assess this view. (Extended essay)

Active recall

Recall the key points — then reveal.

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

Falsification (Flew, Hare, Mitchell)#

StandardAQA 7062/1 Section A: Religious languageDfE GCE Religious Studies subject content

The falsification challenge and replies

Fig. 3Flew charges that religious claims exclude nothing; Hare and Mitchell reply non-cognitively and cognitively.

Antony Flew reframed the challenge in terms of falsification, drawing on Karl Popper's insight that what makes a statement genuinely factual is not that it can be verified but that it can be falsified — that the speaker can say what would count against it. A factual assertion, Flew argued, necessarily excludes some states of affairs; to assert 'the cat is on the mat' is to deny that the mat is bare. His challenge to the believer is: what would have to happen for you to admit that 'God loves us' or 'God exists' is false? If nothing could ever count against it, then it asserts nothing.

Flew illustrated this with John Wisdom's parable of the gardener, retold: two explorers find a clearing with both flowers and weeds; one says a gardener tends it, the other denies it. They watch, set up barbed wire, bloodhounds and electric fences, but never detect any gardener; yet the believer keeps qualifying his claim — the gardener is invisible, intangible, insensible to shocks, undetectable. At last the sceptic asks how this 'invisible, intangible, eternally elusive gardener' differs from no gardener at all, or from an imaginary one. Flew's charge is that religious believers do exactly this: faced with evils and unanswered prayers, they endlessly qualify 'God loves us' (a father's love, a mysterious love, a love beyond understanding) until the assertion 'dies the death of a thousand qualifications' and no longer denies anything, and so no longer asserts anything.

Two influential replies appeared alongside Flew's in the 'Theology and Falsification' symposium (1955). R. M. Hare conceded that religious statements are not falsifiable factual claims but denied that they are therefore meaningless; they express what he called a blik — an unfalsifiable but meaningful way of seeing the world that profoundly affects how one lives. His parable is of a lunatic convinced all dons want to murder him: no amount of kindly dons dissuades him, because his blik is unfalsifiable — yet the difference between his insane blik and a sane one is real and important. Religious belief, on this non-cognitive reading, is a blik: not a factual claim open to falsification, but a meaningful orientation. The weakness is that it seems to make religious belief arbitrary — how do we tell a sane blik from an insane one? — and reduces faith to a way of seeing rather than a set of truths believers take themselves to hold.

Basil Mitchell defended a cognitivist middle way with his parable of the partisan and the stranger. In wartime, a member of the resistance meets a stranger who deeply impresses him and claims to be on the resistance's side; thereafter, even when the stranger appears to help the enemy, the partisan keeps faith, trusting him on the strength of their meeting, while acknowledging that the ambiguous behaviour really does count against his trust. Mitchell's point is that religious believers do allow that evil and suffering count against 'God is good' — they feel the force of the problem of evil — yet do not treat it as decisively falsifying, because their trust rests on other grounds (revelation, experience, relationship). Religious statements are thus genuine factual claims, provisionally held against contrary evidence, not vacuous ones immune to all evidence. Mitchell's reply preserves cognitivism while accepting that faith is not naive.

Worked example

Mitchell's answer to Flew

How does Mitchell's partisan parable answer Flew, and does it succeed?

Flew's demand
State what would count against 'God is good'; if nothing does, it asserts nothing.
Mitchell's concession
The partisan admits the stranger's ambiguous acts really do count against his trust — evil really does count against God's goodness.
Mitchell's move
But the trust is not abandoned, because it rests on the founding encounter (revelation/experience); the claim stays factual yet is held against contrary evidence.
Assess
This preserves cognitivism honestly (unlike endless qualification), but a critic asks how much contrary evidence is too much before trust becomes irrational.
Judgement
Mitchell meets Flew's challenge better than Hare, because he keeps religious statements factual and lets evidence count, without conceding falsification.

Result: Mitchell answers Flew by keeping religious statements factual and allowing evidence to count against them while not treating it as decisive, which is more faithful to how believers actually reason than either Flew's demand or Hare's non-cognitive retreat.

Typical mistakes

Confusing the three symposiasts: Flew attacks, Hare concedes non-cognitively (bliks), Mitchell defends cognitively (partisan).
Presenting Hare's blik as a factual claim — the whole point is that a blik is unfalsifiable and non-cognitive.

Active revision

'Religious language dies the death of a thousand qualifications.' Discuss with reference to Flew, Hare and Mitchell. (Extended essay)

Active recall

Recall the key points — then reveal.

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

Analogy, symbol and the via negativa#

AdvancedAQA 7062/1 Section A: Religious languageDfE GCE Religious Studies subject content

Analogy and symbol compared

Fig. 4Both make religious language meaningful without univocity, but analogy claims positive knowledge by proportion while symbol claims participation.

Thomas Aquinas developed the theory of analogy to solve the univocal/equivocal dilemma while keeping religious language cognitive. If words applied to God are univocal, God is dragged down to the creaturely level; if equivocal, we speak in the dark. Analogy is the middle way: when we call God 'good', the word is used in a sense related to, but proportionate to, its creaturely use, because creatures reflect their creator. Aquinas distinguished the analogy of attribution — a quality is attributed to God as its source, as 'healthy' is attributed to food because it causes health in the body, so 'good' is attributed to God as the cause of goodness in creatures — from the analogy of proportionality, in which a quality belongs to each thing in proportion to its nature: a dog has 'life' proper to a dog, a human 'life' proper to a human, and God 'life' proper to God, infinitely. Analogy lets us say something true and positive about God without claiming to comprehend him.

Paul Tillich argued that religious language is essentially symbolic, and drew a sharp distinction between a sign and a symbol. A sign (a red traffic light) is arbitrary and points to something beyond itself; a symbol participates in the reality to which it points and opens up levels of reality and of the soul otherwise closed to us, as a national flag participates in the nation's power and dignity. Religious statements, for Tillich, are symbols: to say 'God is our Father' or that God is 'the ground of being' is not to make a literal factual claim but to point to and participate in the ultimate. Only one statement about God is non-symbolic, that God is being-itself; everything else is symbol. This makes religious language meaningful in a non-literal, participatory way, but critics (including Hick) ask how we can know a symbol is appropriate, or whether a symbol that participates in nothing verifiable really points to anything — the objection that Tillich's account, like Flew's target, cannot be checked.

The via negativa (the apophatic way), associated with Pseudo-Dionysius and Maimonides, takes the opposite tack: since God infinitely exceeds our concepts, we cannot say what God is, only what God is not. To call God 'good' is really to deny that God is evil or limited; positive language inevitably falsifies by importing creaturely limitation, so the safest and truest religious speech is negation ('God is not finite, not corporeal, not mutable'). The via negativa protects divine transcendence and mystery, and takes seriously the failure of literal language. Its weakness is that pure negation seems to leave us with an empty concept — a being of whom we can say nothing positive is hard to distinguish from nothing at all, and it does not obviously match the rich, positive language of worship and scripture ('the Lord is my shepherd'), which does seem to intend positive content.

These three cognitivist-leaning strategies (analogy, symbol, negation) can be read as a graded response to the same problem — the inadequacy of univocal literal language for a transcendent God. Analogy says the most, claiming genuine positive knowledge by proportion; symbol says less, claiming participation without literal reference; the via negativa says least, claiming only knowledge of what God is not. A candidate can evaluate them as a spectrum from cataphatic (positive) to apophatic (negative) theology, weighing the trade-off each makes between saying enough to be worth saying and not saying so much as to reduce God to a creature. The best answers note that the strategies are complementary rather than exclusive: the tradition uses analogy and symbol to speak, and the via negativa to keep that speech humble.

Worked example

The apophatic-cataphatic spectrum

Arrange analogy, symbol and the via negativa on a spectrum and state the trade-off.

Most positive
Analogy: claims genuine positive knowledge of God by proportion (cataphatic).
Middle
Symbol: points to and participates in the ultimate without literal reference.
Most negative
Via negativa: says only what God is not (apophatic).
The trade-off
The more positive the theory, the richer the content but the greater the risk of anthropomorphism; the more negative, the safer but the emptier.
Complementarity
The tradition uses positive analogy and symbol to speak, tempered by negation to preserve mystery.

Result: Analogy, symbol and via negativa form a spectrum from saying much (risking anthropomorphism) to saying little (risking emptiness); used together they let theology speak meaningfully while preserving divine transcendence.

Typical mistakes

Confusing Aquinas' two analogies, or presenting analogy as a non-cognitive theory — analogy defends cognitive, positive God-talk.
Treating Tillich's symbol as merely a sign; the participation of the symbol in its referent is the crux of his view.

Active revision

'Analogy is the most successful way of talking meaningfully about God.' Assess this view. (Extended essay)

Active recall

Recall the key points — then reveal.

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

Language games and evaluation#

AdvancedAQA 7062/1 Section A: Religious languageDfE GCE Religious Studies subject content

Cognitive and non-cognitive language compared

Fig. 5Cognitivism keeps religious statements factual and vulnerable to verification; non-cognitivism makes them expressive and immune but non-factual.

The later Wittgenstein offers the most radical non-cognitive response. In the Philosophical Investigations he argued that the meaning of a word is its use in the practices of a community — 'the meaning of a word is its use in the language' — and that language consists of many distinct 'language games', each with its own rules, embedded in a 'form of life' (science, law, poetry, joking, religion). To ask whether a religious statement is 'true' or 'verifiable' by the standards of the scientific language game is a category mistake, like asking whether a chess bishop is legal in draughts. Religious language is meaningful within the religious form of life, where it does its own work (worship, confession, consolation, commitment); the positivist and falsificationist demands import alien rules and so simply miss what religious language is. D. Z. Phillips developed this: religious statements are not factual hypotheses but expressions of a way of living, so 'God exists' is not a claim that could be shown probable or improbable but a feature of the grammar of faith.

The language-games view has real strengths: it explains why believers are unmoved by verificationist attacks (the attacks misdescribe their language), it respects the integrity of religious practice, and it dissolves rather than answers the meaninglessness charge. But it is widely criticised for cutting religion off from reality. If religious language is only meaningful 'within the game', it appears to become immune to external criticism and unable to make the objective claims believers actually intend: most believers mean that God really exists, not merely that 'God exists' is a move in their game, and Phillips's reading seems to reduce robust theism to a form of expressivism (fideism), emptying it of the factual content for which martyrs died. The view also struggles to explain how someone can rationally join or leave a religious form of life, or how inter-faith and faith-vs-atheism disagreements can be genuine.

Standing back, the evaluation turns on the cognitive/non-cognitive question that opened the topic. The positivist challenges (verification, falsification) succeed only against cognitive religious language, and even there they are blunted: the verification principle is self-refuting, and Mitchell shows that cognitivist religious statements can let evidence count without being falsified. The non-cognitive responses (Hare, Tillich, Wittgenstein, Phillips) escape the challenges entirely, but at the price of conceding that religious language does not state facts — a price most believers will not pay, since it seems to change the subject rather than defend it. The cognitivist strategies (analogy, symbol read realistically, eschatological verification) try to keep religious language both meaningful and factual, and are the most faithful to religious intention, though none is wholly free of difficulty.

The most defensible judgement is that religious language is meaningful, but that its meaningfulness has been more clearly vindicated than its factual reference has been secured. The meaninglessness charge fails: the verification principle collapses under its own weight, and religious language plainly functions richly in worship, ethics and life, which no adequate theory of meaning should rule out as nonsense. Whether that meaningful language succeeds in referring to a real, transcendent God — whether it is cognitively true and not merely expressively meaningful — is not settled by the philosophy of language alone; it depends on the arguments and experiences examined in the earlier topics. A strong essay therefore concludes that religious language is meaningful (against Ayer and Flew), that analogy and symbol give the best account of how it can say something about a transcendent God, but that its truth is a further question the theory of meaning cannot decide.

Worked example

Model evaluation plan: is religious language meaningful?

Plan an essay assessing whether religious language is meaningful.

Thesis
Religious language is meaningful — the meaninglessness charge fails — but the philosophy of language secures its meaning, not its truth.
Challenges (AO1)
Verification (Ayer) and falsification (Flew) allege it is nonsense or empty.
Defeat the challenges
The verification principle is self-refuting; Mitchell keeps cognitive claims factual without being falsified; non-cognitivists (Wittgenstein) dissolve the charge.
Weigh the cost
Non-cognitivism escapes the challenges but concedes religious language is not factual (fideism), which most believers reject; analogy/symbol keep it meaningful and (in Aquinas) factual.
Judgement
Conclude meaningful (against Ayer/Flew); analogy and symbol best explain how it speaks of God; its truth is a further, separate question.

Result: Religious language is meaningful — the positivist challenges fail — and analogy and symbol best explain how finite words can speak of a transcendent God; whether that meaningful language is also true is not decidable by the theory of meaning alone.

Typical mistakes

Presenting language games as proving religious statements true — the view denies they are fact-stating hypotheses at all.
Ending with 'religious language is meaningful, therefore God exists' — meaningfulness does not entail truth or reference.

Active revision

'Religious language is meaningful.' Evaluate this claim with reference to the verification and falsification debates. (Extended essay)

Active recall

Recall the key points — then reveal.

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

References & sources

Sources

Department for Education

GCE AS and A level subject content for religious studies

Religious language

5 sections~23 min reading time2 competenciesLevel Foundation 1 · Standard 2 · Advanced 2

Religious language

Key points

Placing a thinker on the map

Key points

The self-refutation objection

Key points

Mitchell's answer to Flew

Key points

The apophatic-cataphatic spectrum

Key points

Model evaluation plan: is religious language meaningful?

Sources

Religious language

Key points

Placing a thinker on the map

Key points

The self-refutation objection

Key points

Mitchell's answer to Flew

Key points

The apophatic-cataphatic spectrum

Key points

Model evaluation plan: is religious language meaningful?

Sources