Analogy: The Only Answer To The Problem Of Religious Language

By Norman L. Geisler

[Trinity Evangelical Divinity School, Deerfield, Illinois 60015]

Most contemporary religious language theory vascilates between equivocation and analogy in a desperate struggle to provide a meaningful vehicle by which God can be expressed. The root of the problem is twofold: Medieval Mysticism and Modem Empiricism. A brief background will help focus the problem.

I. The Background Of The Problem Of Religious Language

Following Plotinus, Medieval mystics from Proclus to Nicholas of Cusa stressed the ineffibility of God. All agreed on one point: nothing positive can be affirmed of God; God is named only by the via negativa.

A. Plotinian Mysticism: God Is Inexpressible

Plotinus is the fountainhead of negative theology. In him Greek Rationalism culminated in Neo-platonic Mysticism. And through his Neo-platonic followers, Proclus and Dionysius, the Middle Ages inherited a strong negative emphasis in language about God.

The entire Plotinian system is based on the notion of unity. Since all multiplicity is made up of unities, there must be a prior unity which is absolutely Simple (God). And since all thought and language involves multiplicity, God is beyond all reason and words. The only way God can be known is via ascending from the multiplicity of the sensible and intellectual worlds and becoming one with the One in a mystical union.

God “can neither be spoken of nor written of” (VI, 9, 4).[1] Even “this name, The One, contains really no more than the negation of plurality.” And, Plotinus adds, “If we are led to think positively of the One, name and thing, there would be more truth in silence.” For the very term “One is nothing but an aid to enquiry and “was never intended for more than a preliminary affirmation of absolute simplicity to be followed by the rejection of even that statement” (V, 5, 6).

The means by which these preliminary affirmations are made is via God’s offspring or emanations. Plotinus wrote, “We can and do state what it is not, while we are silent as to what it is: we are in fact speaking of it in the light of its sequels [offspring]” (V, 3, 14). That is, since we cannot know It we may only speak of It in terms of what comes from It.

However, what we speak of the One is not really found in it, because it bestows what itself does not possess” (VI, 7, 15). God is a “nobler principle than anything we know as Being; above reason, mind, and feeling; confering powers, not to be confounded with them” (V, 3, 14). The reason that God produces being but does not possess being is simple: “He had no need of being, who brought it to be” (VI, 8, 19). Hence, God is called Being by extrinsic attribution, not because He has being, but because He causes other things to have being.

Therefore, concludes Plotinus, “The One is in truth beyond all statement” (V, 3, 13). And “in our writing and telling we are but urging toward it: out of discussion we call to vision; to those desiring to see, we point the path” (VI, 9, 4). In other words, language about God consists of mere signs and pointers to stimulate one who seeks to have an intuitive vision of God. God Himself is so far beyond all reason and language that “only by a leap can we reach to this One…”(V. 5, 4). But anything this side of that intuitive vision which is attributed to God from His emanations must be understood only in strictly negative sense.

Plotinus was not totally unaware of the implications of a purely negativetheology. He wrote, “It is impossible to say, ‘Not that’ if one is utterly without experience or conception of the ‘That’; there will generally have been, even, some inkling of the good beyond Intellection” (VI, 7, 29). How positive “inklings” of God are possible when all we know of Him is through extrinsic attributions known only negatively is not clear. The implication is that there is a positive intuition of God which enables us to form negative concepts of Him. One thing is clear, there is nothing in the realm of concepts or language which is positively descriptive of God. God is beyond all words.

The history of Neo-plotinian thought after Plotinus, including Proclus, Dionysius, Maimonides and Nicholas of Cusa, is little more than a series of footnotes on Plotinus. God is known only mystically and expressed only negatively. We move, then, from Medieval Mysticism to Modern Empiricism to the other root of the contemporary problem of religious language.

B. Modern Empiricism: God-talk Is Nonsensical

Not very many non-mystics are convinced that a purely negative language really says anything about God. Every negation implies an affirmation. And a purely unknowable God is unworshipable. As Ludwig Feuerbach remarked, “The truly religions man can’t worship a purely negative being … Only when a man loses his taste for religion does the existence of God become one without qualities, an unknowable God.”[2] At any rate, there is in addition to Mysticism the problem of Empiricism for many modern religious analysts. The problem springs from David Hume.

In the last lines of his famous Enguiry[3] Hume insisted that there were only two kinds of meaningful statements which can be made, both of which would disqualify religious statements. He wrote,

If we take into our hands any volume; of divinity or school metaphysics, for instance; let us ask, Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reason concerning matter of fact and existence? No. Commit it then to the flames: for it contains nothing but sophistry and illusion.

In short, there are only two meaningful kinds of statements: those involving the mere relation of ideas, such as mathematical statements and those involving matters of fact, such as those known from sense experience. And if we entertain any suspicion as to a religious statement, “we need but inquire,” wrote Hume “from what impression is that supposed idea derived? And if it be impossible to assign any, this will serve to confirm our suspicion.” Now since statements about a transcendent God obviously cannot be reduced to sense impressions, Hume left no doubt what we should do with the books containing religious statements.

Ludwig Wittgenstein, the father of linguistic analysis, combined both Mysticism and Empiricism and created religious non-cognitivism. In the last lines of his Tratatus he wrote: “What we cannot speak about we must consign to silence.” Wittgenstein’s “silence,” to be sure, is less severe than Hume’s “flames,” but in either case language about God turns out to be nonsensical. Wittgenstein, like Plotinus, did not eliminate the experience of the mystical; he merely wished to point out it could not be expressed in any linguistically meaningful way. He wrote, “There are indeed, things that cannot be put into words … They are what is mystical.” Again, “It is not how things are in the world that is mystical, but that it exists.” For how things are in the world is a matter of indifference for what is higher. God does not reveal himself in the world.”[4] In short, the door is open for mystical experience, but it is closed for any linguistic expression of it.

A. J. Ayer was one of the more vocal members of Wittgenstein’s Vienna Circle of followers. Ayer carried through the implication of Hume’s Empiricism in an attempt at “The Elmination of Metaphysics,” to borrow the title of his first chapter. All meaning, said Ayer, is subject to the verification principle which asserts that only statements which are either true by definition (i.e., tautological) or else are empirically verifiable (i.e., known to be true from sense experience) are meaningful. All other statements are literally nonsensical. This means that “all metaphysical assertions are nonsensical,” and as for “the possibility of religious knowledge, we shall see,” said Ayer, “that this possibility has already been ruled out by our treatment of metaphysics.”[5] The problem with Ayer’s original verification principle is that it eliminated itself, since it was not intended as a tautology nor was it empirically verifiable. This led Ayer to revise the principle, but he did not give up the belief that religious statements would still turn out to be nonsensical, since he expected that no theologian would yield to any kind of empirical verification.[6]

Contrary to Ayer’s expectation some theologians did yield. Paul Van Buren, in fact, carried through the empirical tradition to semantical Atheism. He wrote, “The empiricism in us finds the heart of the difficulty not in what is said about God, but in the very talking about God at all.” For “we do not know ‘what’ God is, and we cannot understand how the word ‘God’ is being used.” In fact, “today, we cannot even understand the Nietzschian cry that ‘God is dead!’ for if it were so, how could we know? No, the problem now is that the word ‘God’ is dead.”[7] That is, the word “God” and all of its transcendent equivalents is devoid of empirical meaning. We are left in semantical Atheism.

In summary, there are two roots to the contemporary problem of religious language: Medieval Mysticism and Modern Empiricism. Both agree that God is literally inexpressible; both lead to linguistical Atheism. In Mysticism there is no way to speak positively about God, and in semantical Atheism there is meaningful way to speak about God whatsoever.

II. Contemporary Alternatives In The Problem Of Religious Language

Many contemporary religious analysists have simply capitulated to the non-cognitivism school. They engage in various kinds of non-descriptive discourse about God with no attempt to claim cognitive truth value for their statements. Ogden’s “myths,” Tillich’s “symbols,” Jasper’s “ciphers” and Crombie’s “parables” are all a concession to acognitivism. Among the contemporary religious analysists who hold out for the truth value of religious language there are two notable examples: Ian Ramsey and Frederick Ferre.

A. Contemporary Objections to the Non-cognitive School

Ferre emphatically resists the trend to relegate all religious statements to the truth-valueless category. He opposes what he calls the “non-cognitive autonomy” on three grounds: (1) Acognitivism is unnecessary, since attitudes and actions are just as subject to critical demands as are matters of fact; (2) The non-cognitive forsaking of the truth value of religious language is irresponsible, for religious commitments have consequences which demand critical examination as well as do scientific theories; (3) It is dangerous because one’s own life policies seriously affect the lives of others. In brief, “to throw critical reflection to the winds in just the most important aspects of life would be the ultimate example of penny-wisdom and pound-foolishness … I cannot conceive,” Ferre continues, “a more perfect definition of the unexamined life.”[8]

But if one rejects mystical negativism and empirical Atheism, then in which direction does one turn for an answer to the problem of religious language? Ferre is equally certain that the answer is not found in Scholasticism’s doctrine of analogy.

B. Contemporary Objections to Analogous Religious Language

There are two kinds of analogous relationships between God and the world distinguished in scholastic philosophy; extrinsic and intrinsic analogies. In an extrinsic analogy the cause is called by a given characteristic simply because it causes the effect to have this characteristic but the cause does not possess the characteristic itself. For instance, food is called “healthy” because it causes health in an organism, whereas “healthy” is not properly a characteristic of food but of organisms. In an intrinsic analogy, on the other hand, both the cause and the effect have the characteristic properly, as when a hot stove causes water to be hot. In this case both the stove and the water have the characteristic of heat.

Now Ferre objects to either kind of analogy as the basis for talking about God. He gives six reasons, the first two of which are directed at extrinsic analogy and the last four at intrinsic analogy.[9] (1) If there is only an extrinsic casual relation between God and the world, then we know nothing about the true characteristics of God; we know only the characteristics which He causes other things to have. (2) If the analogy is only extrinsic and God does not really possess any of the characteristics attributed to Him, then why select some and reject others? Why not apply everything He causes to God? (3) When words are disengaged from their finite mode of signification they become entirely vacuous and without meaning. (4) If analogy is held to be intrinsic, then it is based on the questionable metaphysical assumption that causes resemble their effects in some important way, i.e., that causality provides the basis for the similarity. (5) There is no univocal way to understand the meaning of an infinite, unconditioned “first cause” from the finite, conditioned relations between cause and effects in our experience. And to attempt to understand “first cause” in a non-univocal way leads to an infinite regress in an attempt to identify the ever elusive univocal element. (6) Likewise, there is no uni-vocal way to express the whole ontological similarity between God and the world, and non-univocal expressions lead to equivocation.

What direction, then, can one take to find a meaningful theological language? If negative language tells us nothing about God and analogical language is reducible to equivocation and if literal descriptions from finite experience cannot possibly express an infinite, then in what direction do we turn for an answer? Both Ferre and Ramsey feel that a satisfactory answer can be provided by the language of qualified models.

C. An Exposition of Qualified Model Language

Ian Ramsey, working out of empirical anchorage, builds what he calls disclosure models of God. Everyday experiences sometimes “come alive” and take on “depth” which is more than empirical. When the disclosure arising out of these empirical experiences evokes a total commitment, then the experience is said to be religious. Words too have disclosure power, especially verbal odd expressions. In fact, no straightforward descriptive language is appropriate currency for religious language. Only models and metaphors taken from everyday empirical experience which are appropriately qualified have religious value.

Ramsey develops three groups of qualified models. ( 1) The first group consists of negative attributes of God which do not arise out of specific experiences but grasp all aspects of experience in a unity. Hence, we speak of God as “immutable” meaning that our changing experiences must be qualified by an “im” or “not” as applied to God.[10] (2) The second class of qualified models are one word positive attributes of God such as “perfect.” Here we begin by assessing various imperfections in everyday experience in a decreasing order until the “light dawns” as to what “perfect” means. (3) The third group comprises the two-word positive attributes of God such as “infinitely good.” In this case the noun is the model and the adjective is the qualifier.

Now individual models may be constructed into a master-model by means of “family resemblances” and integrated into the term “God.” First-person language (“I-talk”) has an interesting parallel to God-talk, in that both are verbally odd and both integrate experiences into one term. The adequacy of the overall integration of all our experience into one macro-model can be judged, says Ramsey, by its “empirical fit,” and not by any scientific type verification. For a theological model is “judged by its stability over the widest possible range of phenomena, by its ability to incorporate the most cIiverse phenomena not inconsistently.” That is, “the theological model works more like the fitting of a boot or a shoe than the ‘yes’ or ‘no’ of a roll call.” So, the many models rooted in human experiences and appropriately qualified are incorporated into a macro-model which is expressed in the term “God.”[11] This master-model is adequate to the degree that it “fits” the whole spectrum of our experience. It becomes more adequate as it consistently incorporates more experience the way a polygon becomes more like a circle by adding more sides to it.

Frederick Ferre offers a more comprehensive way of testing the truth value of the overall metaphysical model based on qualified religious models.[12] (1) First, Ferre says, the overall synthesis must be consistent or free from contradiction in its key statements. (2) Second, it must be coherent, extending in a unified way to all bodies of knowledge. (3) Third, the mastermodel must have applicability to individual experiences. (4) Fourth, it must have adequacy, i.e., be relatable to all domains of feeling and perception, (5) Finally, the macro-model must be effective as a tool for coping with the total environment of human experience.

Ferre sometimes summarizes these truth-tests for the metaphysical model as three: consistency. coherence and adequacy. In brief, the metaphysical synthesis built on religious models is adequate only if it is able “to put all experience into a pattern that is whole, that is pervasive, and that is adequate.”[13]

D. An Evaluation of Qualified Model Language

The difficult in evaluating qualified models is that they vascilates between equivocation and analogy. Roughly speaking both Ferre’s and Ramsey’s views can be classed as analogous God-talk. Ramsey admits this and Ferre implies it. Ramsey wrote, “To talk of God as ‘heavenly father’ or ‘divine creator’ is to speak analogically … That is one is endeavoring to suggest a qualified similarity … .”[14] One of Ramsey’s serious commentators wrote of “the similarity between Ramsey’s view and that of Thomas Aquinas. Little, if any, distortion results from roughly classifying Ramsey’s position within the tradition which maintains the analogical nature of talk about God.”[15]

Ferre does not explicitly admits to analogy but there are some indications that his religious language is analogous. First, the influence of Dorothy Emmet’s metaphysical analogy is indicated by the following quote: “metaphysics starts from the articulations of relationships, which are judged to be constituitive of an experience or experiences in a significant way … A conceptual expression of such a relationship is then extended analogously as a coordinating idea, in terms of which further ranges of experience may be interpreted.”[16] The similarity of this kind of analogous metaphysical thinking is Ferre’s own procedure is obvious. Secondly, Ferre would not admit that religious language is either completely equivocal (for it has truth value) or totally univocal (since it is not literally descriptive of God). This leaves only one alternative, analogy.

The dilemma, then, is this: according to Ferre’s own admission, analogies must be either intrinsic or extrinsic. That is, they either inform us of God’s characteristics or they do not. But if the analogy is extrinsie, by his own admission, Ferre is left in skepticism about God and may as well join ths semantical atheist he opposed. On the other hand, if the analogy is intrinsic, then Ferre may as well join the Thomists. Maybe Ferre would accept Thomistic analogy if he could see this way out of the objections he had to intrinsic analogy. At any rate, we should at least take another serious look at Thomistic analogy. For if today’s congnitive religious analysist are playing in the ballpark of analogy then we ought to at least watch the game.

III. A Reevaluation Of Analogous Religious Language

A. Scotus: The Insistance on Univocal Concepts

The central problem with scholastic analogy has been corrected by recent studies. The conflict centered on the nature of analogous concepts. Scotus insisted that unless concepts are univocal then they are ultimately equivocal and we are left in skepticism. Arguing against Henry of Ghent’s “analogous concept of being,” Scotus insisted on two points: (1) First, if God and creatures are distinguished only by negation, then there is no distinction at all. For “every denial is intelligible only in terms of some affirmation.” (2) Second, Scotus notes that an analogous concept is really two different concepts, one to account for the similarity and one to explain the difference. If so, then the concept which is the same is univocal and not really analogous. If one concept is not univocal, then the whole concept is really equivocal since there is no common meaning there at all.

The positive side of Scotus’ argument for univocal concepts may be summarized by one central argument: without univocity there is no certainty that we have any knowledge whatsoever. We do know something about God. Hence, there must be somewhere at the basis of all of our concepts at least one univocal notion in terms of which we can commonly predicate something of both God and creatures. In Scotus’ own words,[17] “one and the same concept cannot be both certain and dubious. Therefore, either there is another concept [which is univocal], or there is no concept at all, and consequently no certitude about any concept.” In brief, so-called analogous concepts must have a univocal element in them. Without this univocal element there is no unity in the concept and no certainty that one really has a concept.· And with this univocal element, one has a univocal concept which makes analogous concepts unnecessary. The only alternative to univocity is complete uncertainty. And since Scotus rejected complete skepticism, he concludes simply: “I say that God is conceived ·.. in some concept univocal to Himself and to a creature.”

Our evaluation of Scotus’ argument is brief. Scotus is right: univocal concepts are indespensible to cognitively meaningful religious discourse. Without a univocal element one’s concepts are equivocal. And by the use of equivocal concepts no knowledge of God is gained.

B. Aquinas: The Necessity of Analogical Predication

Granting to Scotus the need for univocal concepts, Aquinas is concerned with an entirely different area, the necessity of analogical predication. For the univocally conceived perfections, drawn from the finite world of our experience, cannot be applied to an infinite God in entirely the same way they are applied to creatures. Aquinas offers six reasons for his contention that “it is impossible for anything to be predicated univocally of God and a creature.”[18] (1) First, “the forms of the things God has made do not measure up to the specific likeness of the divine power. (2) Even if they did, still “they do not receive it [the form of likeness] according to the same mode of being.” (3) Thirdly, because “nothing is predicated of God as a genus or a difference; and thus nothing is predicated as a definition … “(4) Fourthly, “whatever is predicated of many things univocally is simplier than both of them, at least in concept. Now there can be nothing simplier than God either in reality or in concept.” (5) Fifthly, “everything that is predicated univocally of many things belongs through participation to each of the things … But God does not participate in anything, all things participate in God.” (6) Finally, “nothing is predicated of God and creatures as though they were in the same order, but, rather, according to priority and posteriority. For all things are predicated of God essentially.”

Elsewhere Thomas rests his whole case against univocal predication on one central argument:[19] “all perfections existing in creatures dividedly and multiplied pre-exist in God unitedly … Hence it is evident that this term wise is not applied in the same way to God and man. The same applies to other terms.” There is also an implied argument in this same article: “God is more distant from creatures than any creatures are from each other. But the distance of some creatures [from each other] makes any univocal predication of them impossible, as in the case of those things which are not in the same genus. Therefore, much less can anything be predicated univocally of God and creatures.” In brief, no finite perfection can be applied in entirely the same manner to an infinite God. Or, in contemporary language, all models taken from human experience must be qualified before they are appropriately attributed to God.

Aquinas is just as emphatic in his rejection of equivocal predications of God as he is of univocal predications. He offers the following arguments:[20] (1) First, in equivocals “it is entirely accidental that one and the same name is applied to diverse things [e.g., “bark” of a tree or a dog]. …But this is not the situation with names of God and creatures, since we note in the community of such names the order of cause and effect…” (2) Secondly, “where there is pure equivocation, there is no likeness in things themselves; there is only a unity of a name. But… there is a certain likeness of things to God…since it belongs to the nahrre of action that an agent produce .its like … “(3) Thirdly, “when one name is predicated of several things in a purely eqnivocal way, we cannot from one of them be led to the knowledge of another … Now, from what we find in other things, we do arrive at a knowledge of divine things …” (4) Fourthly, “if nothing was said of God and creatures except in a purely equivocal way, no reasoning proceeding from creatures to God would take place. But the contrary is evident from all those who have spoken about God.” (5) Fifthly, “it is also a fact that a name is predicated of some being uselessly unless through that name we understand something of the being.” (6) Even if the names of God were negative so that God was called living only because He is not lifeless, “it will at least have to be the case that living said of God and creatures agrees in the denial of the lifeless. Thus it will not be said in a purely equivocal way.”

Elsewhere Aquinas supports the case against equivocal predication with one argument:[21] “This name God…is taken neither univocally nor equivocally, but analogically … For we can name God only from creatures. Hence, whatever is said of God and creatures is said according as there is some relation of the creature to God as to its principal cause, wherein all the perfections pre-exist excellently.” That is, God cannot be totally diverse from the perfections found in creatures since creation received those perfections from Him. God can not give what He does not have to give.

Our evaluation of Aquinas is also brief: He is right. Terms taken from the finite world of our experience cannot be applied to God in exactly the same way they are applied to creatures. An infinite cannot be conceived in a conceptually limited way. Predicating limited concepts of an Unlimited being would be sheer anthropomorphism or conceptual idolatry. And idolatry is idolatry whether the image is metal or mental. On the other hand, purely eqnivocal predication leaves us in total ignorance of God’s essential attributes. The only remaining alternative is analogical predications.

C. Reconciling Scotus and Aquinas

The positions of Scotus and Aqttinas are not irreconcilable. Scotus argued justifiably for a univocal conception of the perfection which is to be attributed to God; Aquinas demonstrated that this perfection could not be predicated of God and creatures in a univocal way. In contemporary language all models drawn from the finite world must be properly qualified before they are applied to an infinite God. For example, the concept of “good” taken from human experience may yield the definition of “that which is desired for its own sake.” Now this same definition certainly applies to both God and man, but it does not apply in exactly the same way. An infinite good should be desired for its own sake infinitely; a finite good should not be desired as an infinite value. In fact, to do so is the very essence of what idolatry means. In short, analogous language takes characteristics which are understood from the world God has created (and, hence, not totally diverse from their Cause) and attributes them to a higher Being in a higher way. The finite perfection has the same definition, but it does not have the same application.

D. Defending Intrinsic Analogy Against Contemporary Criticisms

We are now in a better position to evaluate Ferre’s criticisms of intrinsic analogy. Five of his criticisms are applicable to intrinsic analogy.[22]

(1) Why select only some perfections from the world to apply to God? Because not every characteristic of finite creation actually applies to an infinite Being. Some things defy qualification without destroying their very concept. An infinite “rock” is a contradiction. Hence, “rock” may be said of God only metaphorically and not metaphysically. Only those terms whose univocal conception can be appropriately qualified, without being totally negated, for an application to the infinite can be actually predicated of God’s essence. Everything else which flows from God can at best be applied to Him only symbolically.

(2) Words divorced from their finite conditions lose all meaning. This objection is simply not true. Every item which actually applies to God retains its univocaIly conceived meaning. And the univocal meaning of some terms is not totally negated by predicating it of an infinite Being. There is nothing, e.g., in the definition of love, goodness, truth, beauty, unity and consciousness which necessarily implies limitations. Granted, all of these terms are conceived within finite conditions and sometimes with limited connotations. But this is why the finite model must be qualified before it is attributed to God. We conclude, therefore, that whenever adding the qualifier “infinite” totally negates the univocal concept of a perfection, then it tells us nothing about God’s essence but at best something about His activity. But whenever a univocally conceived perfection caused by God can be qualified of its finite conditions and connotations without negating its univocal meaning, then it is truly informative about God.

(3) Analogy is based on an alleged ontological similarity between God and the world. Theism pleads guilty of this charge and offers two lines of evidence. First, the Bible teaches that creation does reflect and reveal its Creator (Ps. 19:1; Rom. 1:19–20; Acts 17:28, 29). Man was made in God’s “image and likeness” (Gen. 1:27). Secondly, an analogy of being is based .in the casual connection between the Creator and His creatures. God’s work, like that of other artists, resembles the Artist. However creation resembles the Creator only insofar as it flows from his principal and efficient and intrinsic causality. (a) The painting does not resemble the artist brush but only the artist’s mind. Likewise, creation resembles God only insofar as it comes directly from His principal causality but not from instruments he uses to cause things. (b) Further, hot water makes an egg both hot and hard. The hotness is caused by efficient causality but the hardness from the material conditions of the egg (heat softens some things). In like manner, only that which flows from God’s efficient causal ity can resemble Him, not those aspects of the world which are due to the material conditions of finitude. (c) Finally, only what flows from God’s intrinsic causality resembIes Him. Musicians generate non-musicians, but humans generate humans. So too, God can and does cause extrinsically things which do not resemble Him. But whatever flows from the principal (direct), efficient and intrinsic causality of God must resemble Him. What is this? Being and all of the perfections of being which do not necessarily imply finitude. God is infinite Being and it is finite being which He creates.

The reason for the similarity between cause and effect was given by Aquinas. An efficient cause of being must be before It can cause others to be. The Cause of all existence must Himself exist. And, further, every characteristic of existence removed from limiting conditions must resemble the Cause of all existence which communicated those characteristics to the world. This much is certain, without this created likeness of the world to God there is no basis in reality for attributing perfections from our experience to God’s essence.

(4) There is no univocal way to express the “first cause.” The answer to this objection lies in the distinction between a univoeal concept and an analogical predication. All of the concepts in an “uncaused cause of being” are univocally understood from our experience of them. “Cause” means necessary and sufficient condition for something. “Being” means that which is or exists, and “uncaused” or “unlimited” means that it is not-limited in its being. It is the later term which serves as the negative qualifier and sets up the analogy which modifies the meaning of the terms “cause” and “being” as applied to God, God is an unlimited kind of limiter; He is an unconditioned kind of conditioner, and an uncaused kind of cause. In short, the definition of the terms which describe God are univocally understood but they are analogically applied to God. All of the concepts applied to God are univocally expressed in thought but they are not univocally applied to things, certainly not to things as diverse as finite and infinite.

(5) The alleged ontological similarity is not univocally expressible. The answer to this objection follows from what has just been said, viz., we have a univocal conception of being, but this concept is predicated analogously of different kinds of beings. Being is defined the same way for God and creatures, viz., as “that which is or exists.” God exists and creatures exist; herein is the similarity. But God exists infinitely and independently, whereas creatures exist finitely and dependently; herein is the difference. Therefore, the fact of existence is the same for both God and creatures but the mode of existence is different. That is, the concept of being is univocal but being itself is analogous.

In fact, an analogy of being is the only alternative to Monism. For if being is entirely the same (i.e., univocal) wherever it is found, then there can be only one being. For to differ in being there would have to be some difference in the very being or things which is precisely what there cannot be if being always means entirely the same thing wherever it is found.[23]

On the other hand, if being means something entirely diverse (i.e., equivocal), then there can be only one being in the universe at most. For once one thing has been identified as being, then the only thing which can be totally diverse from being is non-being. Hence, every other supposed “being” is really non-being, if it is really totally different from being. Therefore, only if there is some similarity without total diversity (viz., analogy) can one avoid Monism. And since the Christian is committed to the belief that there are more beings than one in the universe he is thereby also committed to a belief in the analogical similiarity of beings.

In summation. From Scotus we learn that meaningful religious language must employ univocal concepts or end in skepticism. From Plotinus we learn that all limitations and plurality must be negated from concepts before they are applied to a transcendent God. For contemporary language analysts we discover that this can be accomplished by qualified models. But from Aquinas we learn that unless there is an analogy of being based in the created similarity of God and the world, then we cannot avoid either Monism or equivocation in talking about God’s essence. In brief, equivocal predication leads to skepticism; univocal predication leads to Monism, but only analogical predication leads to God.

Notes

1. Quotations from Plotinus are taken from his Enneads, translated by Stephen MacKenna, London: Faber and Limited, 1956.
2. Ludwig Feuerbach, The Essence of Christianity, translated by George Eliot, New York: Harper Torchbooks, 1957, p. 15.
3. David Hume, An Enquiry Concerning Human Understanding, New York: Bobbs-Merrill Company, Inc., 1962.
4. Ludwig Wittgenstein, Tractatus Logico-Philosophicus, translated by D. F. Pears and B. F. McGuiness, New York: Humanities Press, N. D., 6.522; 6.44; 6.432.
5. A. J. Ayer, Language, Truth and Logic, New York: Dover Publications Inc., 1946, pp. 41, 114.
6. Ibid., pp. 10, 16.
7. Paul van Buren, The Secular Meaning of the Gospel, New York: The Macmillan Company, 1963, pp. 100, 83-84.
8. Frederick Ferre, “Science and the Death of God,” in Science and Religion: New Perspective on Dialogue, edited by Ian Barbour, New York: Harper and Row, 1968, pp. 150-51.
9. Ferre, “Analogy in Theology,” The Encyclopedia of Philosophy, edited by Paul Edwards, New York: The Macmillan Company, Vol. I (1967), 94–97.
10. Ian Ramsey, Religious Language, New York: The Macmillan Company, 1956, p. 56 f.
11. Ramsey, Models and Mystery, London: Oxford University Press, 1964, p. 16.
12. Ferre (Editor), Exploring the Logic of Faith, New York: Association Press, 1962, p. 166 f.
13. Ferre, “Science and the Death of God,” op. cit., p. 147.
14. Quoted by Jerry Gill in “The Meaning of Religious Language,” Christianity Today, IX:8; Jan. 15, 1965, p. 19.
15. Jerry Gilt, “Ian Ramsey’s Interpretation of Christian Language,” an unpublished doctorar dissertation (in University of Michigan Microfilms, Ann Arbor), p. 42.
16. Dorothy Emmet, The Nature of Metaphysical Thinking, London: Macmillan and Go., 1946, p. v.
17. John Duns Scotus,Philosophical Writings, translated by Allan Wolters, Now York: The Bobbs-Merrill Company, Inc., 1964, pp. 89-101.
18. Thomas Aquinas, Summa Contra Gentiles, I, 32, translated by A. C. Pegis.
19. Aquinas, Summa Theologica, I, 13, 5.
20. Aquinas, Summa Contra Gentiles, I, 33.
21. Aquinas, Summa Theologica, I, 13, 5.
22. The discussion here is follows the classic work on Thomistic analogy by Battista Mondin, The Principle of Analogy in Protestant and Catholic Theology, The Hague: Martinus Nijhoff, 1963. Other good recent works on anology are by Ralph MCInerny, The Logic of Analogy (The Hague: Martinus, 1961) and George Klubertanz, St. Thomas on Analogy, Chicago: Loyola University Press, 1960.
23. See Parmenides, “Proem” in The Presocratic Philosophers, translated by G. S. Kirk and J. E. Raven, London: Cambridge University Press, 1964, p. 266 f.