Analogy in Religious Language

Aquinas' two forms of analogy provide a sophisticated framework that avoids the main pitfalls of univocal and equivocal language, and the analogy of attribution in particular captures how religious believers actually use language about God — but the analogy of proportion is weaker and its content remains indeterminate.

The analogy of attribution holds that a word like "good" is primarily applied to God and applied derivatively to creatures: God is good in the fullest, primary sense, and everything that is good in creation is good because it participates in God's goodness. When we say "God is good" and "this person is good," the word is not used identically (not univocal) but nor is it used with entirely different meanings (not equivocal): it is applied to the creature by reference to God, the primary instance. The analogy of proper proportion holds that words describe a relationship between a quality and a subject — "a faithful dog is faithful in proportion to its canine nature; God is faithful in proportion to the divine nature." Both analogies attempt to preserve the transcendence of God (his goodness is not merely human goodness writ large) while maintaining the meaningfulness of God-talk (we say something real when we say "God is good"). As the OCR examiner's report notes, stronger candidates understand that analogy "enables believers to use and defend the meaning of positive claims about God."

However, both analogies face challenges. The analogy of attribution seems to presuppose that we already know God is the primary instance of goodness — but how do we establish this without already having a concept of God's nature? It risks being circular: we explain what "God is good" means by reference to God, but we need to understand the analogy to know what God is in the first place. The analogy of proportion is arguably even weaker: saying "God is faithful in proportion to the divine nature" tells us very little unless we already know what the divine nature is. As Lauren's revision notes observe, analogy tells us "how we are using theological language, not what it means" — which raises the question of whether it communicates genuine content or merely describes a procedure for talking about God.

A defender can respond that the analogies are not offered as definitions but as linguistic clarifications: Aquinas is not claiming to give a complete account of God's nature through analogy but to show that meaningful, non-anthropomorphic speech about God is possible. The analogy of attribution works in a religious community where the attributes of God are already partially understood through revelation and tradition — it is a tool for refining and communicating, not for building understanding from scratch. Furthermore, as divinityphilosophy.net argues, analogy "treads the line between acknowledging the otherness of God and retaining the ability to say some meaningful things about God successfully" — which is the most that any finite language system can reasonably aspire to do about an infinite subject.

This response is largely persuasive: analogy does succeed in preserving a middle path between anthropomorphism and silence, and within a religious community it functions effectively as a tool for theological discussion. The proportion analogy is weaker because it offers so little determinate content, but attribution is genuinely useful in explaining how language can refer to God without reducing him to a super-human. The main remaining question is whether "partially meaningful" is sufficient for the demands of religious life — for prayer, worship and moral guidance — where believers typically want more than a gesture towards divine transcendence.

Aquinas' analogies are therefore effective in their primary goal of avoiding anthropomorphism and meaninglessness, but the analogy of proportion in particular is so indeterminate that its effectiveness for actual religious communication is limited.

Swinburne's challenge — that analogical language should simply be replaced by univocal language — and a comparison with the via negativa together reveal both the genuine advantages of analogy and the persistent question of whether it is religiously adequate.

Richard Swinburne argues that Aquinas' insistence on analogical language is unnecessary and confusing: religious language about God's goodness, knowledge and love can and should be understood univocally — with the same meaning as when applied to humans, since God shares properties with humans as a rational person. On this view, "God is good" means the same kind of thing as "this person is good," and no elaborate analogical qualification is needed. By contrast, the via negativa (Maimonides, Pseudo-Dionysius) argues that we can only say what God is not — God is not finite, not changeable, not comprehensible — since any positive attribution risks anthropomorphism. On this account, analogy is too positive and risks the same anthropomorphic errors as univocalism; silence or negation is more honest.

Swinburne's univocalism faces the classic objection: if "God is good" means exactly the same as "this person is good," then God's goodness is just a magnification of human goodness, which anthropomorphises God and reduces him to a super-human being. This seems incompatible with the classical theistic commitment to divine transcendence. The via negativa faces the opposite problem: Brian Davies argues it could lead to a mistaken or empty understanding of God, since if we can only negate, we cannot distinguish between different conceptions of God — "God is not finite" is equally consistent with pantheism, polytheism, or classical theism. As divinityphilosophy.net observes, the via negativa also prevents us from making the positive claims that religious believers actually want to make, such as "God loves us."

Analogy emerges from this comparison as a genuine middle path: it retains the positive content of univocal language (we say something real about God's goodness) while preserving the qualifications of the via negativa (God's goodness is not simply human goodness). For practical religious communication — sermons, prayers, theology — this middle path is more useful than either pure negation or pure univocalism, since it allows believers to affirm meaningful claims while acknowledging the limits of their language. Soskice's work on metaphor in religious language supports this: religious language often works metaphorically in a way that is analogical rather than literal, communicating genuine insight rather than precise description.

On balance, analogy is more effective than the alternatives for the purposes of religious communication, even if imperfect. Swinburne's univocalism overclaims — reducing God to a super-human — and the via negativa underclaims — preventing any positive affirmation. Analogy occupies the most defensible middle ground. However, Hick's question is fair: since analogy preserves so much indeterminacy, is there ultimately a meaningful difference between Tillich's symbolism and Aquinas' analogy? Both acknowledge that language about God points beyond itself without capturing divine reality fully — which suggests the effectiveness of analogy may be more about managing the limits of religious language than overcoming them.

Analogy is therefore effective relative to its alternatives, but its effectiveness is partial — it manages the problem of religious language rather than solving it, and for believers who require more precise content from God-talk, it may feel insufficient.