Analogy in Theology
ANALOGY IN THEOLOGY
Religious discourse has been under scrutiny since ancient Greece when Anaxagoras said if oxen and dogs could paint, they would depict the gods in their own likenesses. The Jewish, Christian, and Muslim scriptures depict the divine being in vivid humanlike traits while conveying the divine otherness, mystery, immateriality, and eternity. Thus there are religious currents of anthropomorphism, of transcendentalism, of metaphor and symbolism, and of literalism about the being and nature of God. The Greek philosophical ancestry of Western culture presents the divine as immaterial, immutable, everlasting, perfect, and incomprehensible. Both the Platonic and Aristotelian metaphysicians developed theories of analogical predication that were later extended to theology, the study of the revealed divinity.
Theologians used a theory of analogy that had three parts: analogy of being (of reality between God and world, and among created realities, too); analogy of meaning (of words and concepts); and analogical thinking (of conception by proportionalities). The aim was to explain how words that apply to sensible things also adapt in meaning to apply literally, not only metaphysically, to the transcendent deity known only by inference, revelation, or mystical experience. Words applied to God—"wise" and "good," for example—are neither entirely equivocal (such as bank /savings; bank /river), nor merely metaphorical (drop /an argument), but rather, they are analogous; that is, they adjust in ways explained below to the context, just as words generally adjust to contrasting contexts, say, as "knows"/the way differs from "knows"/arithmetic, and as "exist" does in "there exist /trees/species/numbers/shapes." Metaphysics articulates theoretical truth-conditions for such statements and for ordinary religious beliefs—conditions not accessible without such metaphysics—the way science states the molecular structure for water.
1. Secular Origin in Plato and Aristotle
The thesis that words fit in literal meaning to diverse verbal contexts that reflect differences of reality—the analogy theory—has its origin in secular philosophy. For Plato, things that share in the Forms are not said to exist in the same sense as the Forms (compare Sophist ; Parmenides ), and the Form "Human" is what -it-is-to-be-human, and thus is human, but not in the sense in which Callicles is human by participating in the Form. Further, Plato used the same names, such as the courageous man and the courageous act, just /state; just /man, for things related as cause to effect and sign to signified.
Aristotle used those distinctions, added more, and regarded real, entitative analogy, reflected in word-meaning, as central to his explanatory principles. (Metaphysics 1070a.31). Such predication is literal, as opposed to metaphorical (Poet 1457b)—for example, "the fields smile with the sunlight" (Aquinas called that improper proportionality [Summa Theologiae 1.13.3.ad 3]). Aristotle acknowledged analogy by attribution (relational naming: healthy /animal; healthy /diet), and by proper proportionality (e.g., genus is to species as body is to soul, namely, as potency is to act ). The explanatory terms—for example, "act/potency"—apply to diverse things analogically (Met 1048b, 5–8). Aristotle further reasoned that qualities, such as color and shape, and other accidentals, are said to be derivatively (pros hen ) to substances; and "analogically the same things are principles, i.e. actuality and potentiality; but they are not only different for different things, but also apply in different ways to them" (Met 1071a.5). Aristotle says the causes and principles of different things are analogous and are spoken of analogously (Met 1070a.31). Moreover, the contrast-dependent notions, "act/potency," "matter/form," "substance/accident," "cause/effect," are all analogical in meaning because the phenomena to which they apply are really, de re, analogous; for instance, body is matter for soul, and clay is matter for a statue.
2. Transition to Theology
The Arabic philosophers adopted Aristotle's views on analogy in their metaphysics and physics and in their discussion of the simplicity of God in the Qurʾan. That made the first connections of Aristotelian analogy-theory to scriptural theology. Islamic religious believers differed on how to interpret the physical descriptions of God's face, eyes, hands, speaking, sitting, and so on, in the Qurʾan, as well as the description of God's feelings—for example, wrath, satisfaction, and God's traits, such as cunning and patience—whether anthropomorphically, metaphorically, symbolically, and so on (compare Van Ess 1954). Al-Kindī (c. 850) thought a literal reading of the Qurʾan on creation is coherent with Aristotelian concepts. In his treatise "On the One True Agent" he holds God is literally the only agent bringing being from (absolute) nonbeing, whereas humans are only metaphorically (analogically?) agents, bringing being from potentiality. Al-Fārābī (c. 900) in chapter 1 of "On the Perfect State" says "existing," "having intellect," "knowing," "being wise," "real," "true," "living," and the like, are said of God in senses different from what we say of creatures because the divine being is simple, without composition or distinct traits. And Avicenna (980–1037) used Aristotle's views about analogy of meaning and of reality directly in his metaphysics and in his physics, where "motion," for instance, is said (as Aristotle also said) to apply analogously, to augmentation, alteration, and locomotion, and the analogy of "being" within the ten categories is acknowledged.
Avicenna reasoned that being and essence are really the same in God, and indicated that a creature's being is not explained by "what -it-is" as is the divine; Aquinas would adopt this. Avicenna also formulated the principle that God's knowledge is the cause of things (later used by Aquinas as cognito dei causa rerum, ST 1.14.8), whereas our knowledge is posterior to things known. The Arabic writers, including the Jewish Moses Maimonides, all hold that God is simple; he is not a body, without any plurality of attributes except by attribution from the divine effects, infinite and incomprehensible. It is from those Arabic, chiefly Islamic, sources, along with the corpus of Aristotle, that the analogy theory came into Latin theology, Avicenna being the most influential in metaphysics.
Maimonides (1135–1204) argued that the eternity of the world is not demonstrated, and that it is both created and has a temporal beginning. Like Avicenna, he affirmed the divine simplicity in a strong sense, so that: "either every attribute we predicate of Him is an attribute of action [and so named from the received effect], or, if the attribute is intended for the apprehension of His essence and not of his action, it signifies the negation of the privation of the attribute in question" (Guide for the Perplexed, 1, p. 58). Thus, saying, "God is all knowing" means "God is not unknowing of anything," and saying "God is simple" means "God is not composite," and saying "God is eternal" means "God is without beginning or end." That came to be known as "negative theology," with no positive ascriptions to God, except existence and creation and the metaphors provided by scripture.
Christians, from the earliest fathers of the church, developed explanatory analogies—that is, proportional comparisons, say, of the Trinity to the unity amidst distinction of the essence, power, and operation of the human soul, and an analogy of the relation of the Father to the Son as "light from light" (in Nicene Creed, and Augustine's De Trinitate ). Such explanatory analogies, not part of the theory described here, were devised throughout the predominantly neoplatonic first millennium of Christian thinking, for instance in Augustine's De Trinitiate (c. 410) and Boethius's De Trinitate (c. 510), the School of Chartres (twelfth century), and the School of St. Victor (twelfth century), and continued throughout the later history of theology (compare Chollet 1923–1967).
A neoplatonic writer historians call Pseudo-Dionysius (c. 500) was widely believed, but not by Aquinas, to have authority as a disciple of St. Paul. He proposed, in his Divine Names, that one first knows God by negation (via negationis ), "not a body," "not with parts," and so on, then by inadequate affirmation as "wise," "good" "loving," qualified by "but not in the way of creatures," and then in a third stage by superlatives, such as "infinitely knowing" and "good beyond excellence" (via eminentiae ). But in his Mystical Theology Pseudo-Dionysius is more restrictive, saying one starts via remotionis by denying of God the things most remote from him, such as "drunkenness and fury," then progressing by denial even through all the higher attributes of creatures until one reaches "the super-essential darkness," entering "the cloud of unknowing," mystically united to what is "wholly unknowable" (because of the limitations of the human mind). This work had a profound influence on the development of transcendentalism in medieval theology and even into the twenty-first century.
3. Aquinas (1225–1274)
Aquinas combined the influences of Avicenna, Maimonides, and Pseudo-Dionysius, along with mastery of Aristotle and Plato. He held that God infinitely transcends every true description achieved by human philosophical efforts, but that, nevertheless, a great deal can be known and positively established about God; in fact, Aquinas believed, there can be both a philosophical science of God from unaided reason, and a divine science whose first principles are given by revelation (ST, 1.q.1.a.2). Furthermore, he absorbed Aristotle's notion of analogy of "being" (pros hen ) for the ten Categories into his own wider theory of analogy between creatures and God by participation. Aquinas said "being can be essentially predicated of God alone, because to be divine is to be subsistent and absolute, whereas being is predicated of any creature by participation; for no creature is its own being, but is something having being," as the actuality (esse ) of its potentiality (its essence), because creatures do not exist on account of what they are, but on account of God (Quod.2, q.2, 1.1). Further, what God is, essentially, is not naturally knowable to humans, though it is disclosed to the blessed by divine gift (ST 1.12.1).
Thus, Aquinas reasoned that our knowledge is not limited to what we can attribute negatively or only by metaphor, or merely by the extrinsic attribution that would make "God is good" mean merely "God is the cause of creaturely goodness" (ST 13. a.6) in the way that a person is called "captain" because of what he does. Many writers, influenced by Philo Judaeus (c. 20 BCE–40 CE) whose work came to the West through Clement of Alexandria and Origen, held that God is named only with names of his effects. Aquinas, however, says we can know that pure perfections (unmixed with limits, such as "educated") apply intrinsically to God by explanatory priority because the divine perfections are the cause and exemplar of all perfections in creatures, such as being, life, knowledge, freedom, and love. This position is variously developed in Summa Theologiae (q.13, a.4–5), and Summa Contra Gentiles (1, chap. 34), and Q. D. De Veritate (q 2.a.1). Nevertheless, the names and concepts of pure perfections are acquired only through our perceptual experience with creatures (ST 1.q.13, a.6), even though their primary reality is in God. Thus the words "loves," "knows," "chooses," and so forth, used of God and of humans, have similar definitions but differing presuppositions that reflect the diverse manner of being of God and creatures, the perfections being prior and all the same as God's being, and finite, received, and really separable from one another in creatures.
So whatever is predicated positively of God is either by attribution, as God is called "creator" on account of what is made and "happy" because of his perfect enjoyment, or predicated by proportionality and priority, as God is said to be "knowing, loving, wise, excellent and beautiful," and so on, but in a manner explanatorily prior to the creature's imperfect and derived being and knowledge. Aquinas also acknowledged metaphorical predicates of God, too (ST 1.19.11), many sanctioned by scripture ("angry," "Prince of Peace"), and many useful negative ones ("not a body," "not in space," "not with parts or complexity," "not with a beginning or end").
The religiously and philosophically central attributes are predicated literally and intrinsically, with their presuppositions adjusted to religious discourse (e.g., "God chooses" but does not deliberate), and elaborated theoretically (e.g., God's attributes are all "really the same as the divine being, esse, differing from one another only in concepts"). They include "knowing," "loving," "good," "righteous," "just," "omnipotent," "omniscient," "immutable," and "present everywhere"—and every other unmixed perfection, too. They apply to God but are adjusted to the priority and perfection of divine being. Thus, God knows but does not find out; God loves but does not need. All creation participates in God's being, not as being divine in any way, but as being continuously from and on account of God, and thus, being said "to be" analogously. Created being is God's proper and continuous effect; the way setting-alight—igniting—is the proper effect of fire; and the illumination of the air is the continuous effect of the sun (ST 1, 8.1).
Aquinas thought the real analogy between divine subsistent being (ipsum esse subsistens ) and creaturely, participated being is an adequate basis for demonstrative knowledge of the existence and of the many attributes of God by reasoning that he displayed in Summa Contra Gentiles.
Nevertheless, Aquinas emphasizes that because what is received is received in the manner of the recipient (quidquid recipitur recipitur modo recipientis, ST 1.75.5), God is disclosed through nature only as far as nature is capable, with all creatures falling infinitely short of the divine reality. And he holds that the divine biblical revelation, though vastly exceeding anything humans could discover or even conceive on their own, is proportioned to what is fitting for humankind, thus leaving the infinite divine mystery "wrapped in a mist" (caligine abvoluta, Const. Dei Filius ch. 4, Vat. 1), with the essence of God beyond all natural understanding.
By the Reformation in the sixteenth century, a religious role for scholastic philosophy was largely rejected, and the reformers held the faith to be in no need of fragile and contested support from philosophy. Biblical authority was said to stand on its own, to be understood by the "analogy of Faith" (analogia fidei, based in Rom. 12.6, according to both Luther and Calvin). Thus the analogy discussions dried up, except among Catholic philosophers such as Cardinal Cajetan (1458–1564), Sylvester of Ferrara (1474–1528), and Francisco Suárez (1548–1617), and mostly stayed that way, apart from the historical scholarship that continues to the present.
David Hume (1711–1776) inaugurated modern noncognitivism, consigning metaphysics to the flames (Enquiry, 1748), asserting that all truths are grounded in sense impressions or relations of ideas, thus setting the framework for twentieth-century verificationism and the attack on the cognitive content of religious discourse.
4. Contemporary Context
In the twentieth century, positivist philosophers, seeking to be like scientists, questioned whether talk about God had any cognitive content at all. Alfred Jules Ayer argued that talk about God is without content because it is unverifiable. Some believers, such as Richard Bevan Braithwaite and Frank Plumpton, proposed empirical understandings of its content; others, such as John Hick, even propose eschatological verifiability. Philosophers such as D. Z. Phillips argued that religious discourse belongs to its own "language game"—a notion adopted from Ludwig Wittgenstein—with its own conditions for meaningfulness, and its own conditions of rational belief, analogous to mathematics and aesthetics. Mostly, however, the discussion of meaningfulness was unconnected to the historical positions on analogy in metaphysics and theology.
One twentieth-century adaptation of the classical accounts (Ross 1981) reasoned that analogy, as "fit of word-meaning to contrasting contexts," is a universal feature of natural languages within which the Aristotelian cases of relational and presuppositional adaptation are particular species, and that the cognitive content of utterances is a function of the family of statements and practices in which they are employed (and often craft-bound to specialized skills and tasks, such as medicine or sailing). Thus, analogy of meaning in religious contexts is a special case of the analogy phenomena found in all the neighborhoods of discourse, whether specialized or not. And Aquinas's metaphysical theory, say of participation and esse subsistens, was interpreted, on that account, as his articulating theoretical truth-conditions for the ordinary and analogous talk of divine existence, perfection, and action, the way a chemist might explain the atomic constitution of a commonly known metal such as lead.
Thus there are at least two additions to Aristotle's and Aquinas's work on analogy: first, that the linguistic phenomena involve differences of discourse commitment (e.g., "God decides," but does not deliberate), as well as the differing theoretical presuppositions articulated by metaphysicians, such as "all divine perfections are really de re the same"; and second, that analogous fit of meanings to diverse context is lawlike, universal, and dynamic in natural languages. But lexical meanings of words are not to be regarded as direct pairings of words to concepts (considered to be their meanings), but are relations of contrast-dependence within the language itself (compare Saussure 1915)—that is, relations of contrastive expressive capacity, so that meanings and the world are correlated in clouds or clusters of discourse, not simply item by item.
As Wittgenstein, Wilfrid Sellars, and others observed, the cognitive content of verbalized beliefs is a function of the community of social behavior in which they have a place in the giving of explanations, reasons, motivations for actions, and interpretations. Thus, although a lot of nonsense is easily formulated in religious talk—as in any other talk—expressed convictions that modify action and attitudes either reflect reality or fail to, and either do so poorly or well. They are thus suitable for epistemic attitudes such as belief and denial. Nevertheless, the truth or falsity of what is said by the religious may not be accessible from outside the practicing community, just as the truth of medical, musical, manufacturing, or scientific expert discourse is largely inaccessible from outside the community of expertise.
The late twentieth- and early twenty-first-century cognitivity issue for religion involved three challenges: (i) whether characteristic expressions (say "Jesus is my personal savior"; "There is one God in three Persons") have stable conditions of appropriate utterance, qualification, reasons, rejoinder, and so on, within a practicing (relatively narrow) religious community; (ii) whether the community practice is one of coherent stable conditions, positive or negative, for acceptable use and endorsement and reason-giving for such assertions; and, (iii) whether basic claims, say, about the existence and nature of God, or some of them (praeambula fidei ), can be rationally accepted or rejected, as well, from outside the confessing community. The common core of Judeo-Christian-Islamic monotheism meets the challenges affirmatively, and many find it externally well supported, even demonstrated in part, though other competent assessors disagree.
Some participants, such as the Reformers, thought external assessment carries no weight or utility for religious faith, though it may have some value in defense of the faith (apologetics). Note also that, in general, the false may sometimes be rationally accepted and the true rationally rejected, as the history of medicine and physics illustrates. Nothing requires a body of convictions to be decidable entirely, or even at its heart, from outside the practice in which conviction is arrived at and sustained. Otherwise the fabric of science would be subject to nonscientific rejection, rather than just parts of it. The same holds for religion. Still, Augustine and Aquinas held that the scripture cannot mean literally what science has demonstrated to be false (ST 1.68.1).
Some recent writers talk as if words, including temporal ones, apply to God not only literally but univocally; for instance, Richard Swinburne, in The Coherence of Theism (1977) said he applies "good" to God in the "perfectly ordinary" sense in which he would say his grandmother was good, though the conditions differ (p. 71). That contrasts with those philosophers such as D. Cupitt or Bishop Robinson who regard talk about God as merely metaphorical. Perhaps, like Duns Scotus, Swinburne and others consider the meaning of the words to be unaffected by differences in the mode of a thing's being. Charles Hartshorne, a Whiteheadean "process metaphysician," came closer to anthropomorphic literalism when he said that God, in process of self-surpassing, can suffer, change, and have other temporal predicates. Analogy theory is often mistaken for a theory of non literal predication, when it is just the opposite: an account of the literal but not anthropomorphic.
Some theologians such as Karl Barth say the meanings of "God loves," "forgives," "redeems," and "commands" are determined by the scriptural context as understood by the church (the community of believers): "Language about God has the proper content, when it conforms to the essence of the Church, i.e., to Jesus Christ.… according to the analogy of faith, (Rom 12.6)" (Church Dogmatics ). In accord with Luther and Calvin, he probably meant that nothing more than the analogia fidei, as understood by the Church, determines what a faithful Christian is to believe and mean. But to say there can be no further truth-conditions at all, say, for "Jesus is the Son of God," would conflict with simple logic. So, sciences might investigate such conditions. And whether extrascriptural theoretical content is sometimes required for faithful belief (say, Eucharistic consubstantiation vs. transubstantiation, vs. mystical presence) is a matter not settled by sola scriptura and analogia fidei, unless theological inquiry is included.
Thus the analogy theorists, historically and in the twenty-first century—like the Reformers—and Barth, the Evangelicals, and philosophers such as Swinburne and Alvin Plantinga, hold that talk about God is neither empty of intelligible content (noncognitivism), nor only metaphorical, poetic, or symbolic (Paul Tillich); nor only negative, except for God's existence (Maimonides); nor positive only in superlatives (Pseudo-Denis—via eminentiae ). And they reject the principle that what is not observationally verifiable or falsifiable is meaningless. They agree that the scripture is the norm for what is to be said about God as Revealed. But analogy theorists additionally maintain (i) that analogous predication is literal and perfectly common in discourse generally, and characteristic of discourse about God, and (ii) that the metaphysical exploration of the divine, even of what is revealed, discloses theoretical truth-conditions, not otherwise accessible, for claims that God exists and has the divine perfections, just as science discloses microconditions for water that are not contained on the surface of the ordinary vocabulary.
So it seems that analogy theory both as linguistic theory and as metaphysical account of being has more innings to play in the history of theology.
See also al-Fārābī; Anaxagoras of Clazomenae; Aristotelianism; Aristotle; Augustine, St.; Avicenna; Ayer, Alfred Jules; Barth, Karl; Boethius, Anicius Manlius Severinus; Braithwaite, Richard Bevan; Cajetan, Cardinal; Calvin, John; Clement of Alexandria; Creation and Conservation, Religious Doctrine of; Duns Scotus, John; Hume, David; Infinity in Theology and Metaphysics; Luther, Martin; Maimonides; Origen; Philo Judaeus; Philosophy of Religion, History of; Plantinga, Alvin; Plato; Platonism and the Platonic Tradition; Pseudo-Dionysius; Ramsey, Frank Plumpton; Reformation; Saint Victor, School of; Sellars, Wilfrid; Suárez, Francisco; Sylvester of Ferrara, Francis; Thomas Aquinas, St.; Tillich, Paul; Wittgenstein, Ludwig Josef Johann.
Aquinas, Thomas. Summa Theologiae. Vol. 1, edited by Thomas Gilby. Garden City, NY: Image Books, 1969.
Aquinas, Thomas. Summa Contra Gentiles. Translated by A. Pegis. Notre Dame, IN: University of Notre Dame Press, 1975. Both contain easily accessible and brief accounts of ideas developed in many other places of his work. See Wippel, below, for a comprehensive exposition of Aquinas.
Barth, Karl. Church Dogmatics. Vol. 2, edited by G. W. Bromiley and T. F. Torrance. Translated by G. W. Bromiley. Edinburgh: T. & T. Clark, 1957.
Cajetan, Thomas De-Vio. The Analogy of Names, and The Concept of Being. 1498. Translated by E. Bushinski and H. Koren. Pittsburgh: Duquesne University Press, 1953. An influential, brief, systematization and interpretation of Aquinas's positions, much disputed by later scholars, but still useful and the source of some classifications such as "analogy of proper proportionality."
Chollet, A. "Analogie." In Dictionnaire de Theologie Catholique, edited by A. Vacant et al. Paris: Librairie Letouzey, 1923–1967. A scholarly survey of analogy in Roman Catholic natural and dogmatic theology.
Klubertanz, George. St. Thomas Aquinas on Analogy: A Textual Analysis and Systematic Synthesis. Chicago: Loyola University Press, 1960. With an appendix of the passages (in Latin) in which Aquinas discussed analogy.
Lyttkens, H. The Analogy Between God and the World: An Investigation of its Background and Interpretation of its Use by Thomas of Aquino. Uppsala, Sweden: Lundequistska bokhandeln, 1953. A published doctoral dissertation with historical sweep and a comprehensive study of the primary sources.
McInerny, Ralph. Aquinas and Analogy. Washington, DC: Catholic University of America Press, 1996.
Ross, James F. Portraying Analogy. Cambridge, U.K.: Cambridge University Press, 1981. Expands Aristotle and Aquinas into a new account of analogy of meaning, its role in philosophy, and in the debate about the cognitive content of religious discourse. It does not address the analogia entis, the metaphysics (see Wippel).
Saussure, Ferdinand de. Cour de linguistique generale. Paris, 1915. Translated by W. Baskin in Course in General Linguistics. New York: Philosophical Library, 1959. By the "father" of modern linguistics, in which the "linguistic meaning is contrast of meaning," idea is developed along with the notion of paradigmatic contrasts that map meaning relationships (employed in Ross, 1981).
Van Ess, J. "Tashbih wa-Tanzih." In Encyclopaedia of Islam, edited by H. A. R. Gibb et al., vol. 10, 341–344. Leiden, Netherlands: Brill, 1960. Reference courtesy of Dr. Jon McGinnis.
Wippel, John, F. The Metaphysical Thought of Thomas Aquinas: From Finite Being to Uncreated Being. Washington, DC: Catholic University of America Press, 2000. The most comprehensive and up-to-date exposition of the whole of Aquinas's theory, its development, and its rationales, especially in chapters 3 and 13.
James F. Ross (2005)