Jacques Maritain Center : The Logic of Analogy / by Ralph McInerny


We often read, in the writings of St. Thomas, of univocal as well as of non-univocal causes. Such an application of properties of signification to causes must be correctly understood, for predication implies universality and causes can be said to be universal in two widely different senses.{1} For example, when assigning the efficient cause of a pair of shoes, we might say the shoemaker made them or that an artisan did. By proceeding in the direction of greater predicable universality (for candlestick makers are also artisans), we can be said to assign prior or more universal causes of the effect. It is the same man who is designated shoemaker and artisan in the example of the shoes, and to say that "artisan" is a prior or more universal cause means only that the cause is being denominated in a more universal, common and vague way. Causes, however, are sometimes said to be universal in quite another sense, this time with respect to a community of causality and not a community of predicability. In this second sense, unlike the first, the more universal cause is numerically different from the less universal cause. For example, the farmer is the less universal, the sun the more universal, cause of the crop.{2}

    What is meant by a univocal or non-univocal cause? Such designations imply predication, and yet it is not likely that what is meant is that the cause is predicated univocally or non-univocally of its effect. Causes are never predicated of their effects.{3} Rather such qualifications of causes indicate the way in which the cause and its effect can be denominated. The univocal cause is that which has in common with its effect a name signifying exactly the same ratio  or definition. In non-univocal causes, this is not the case.{4} Of course, it is not because an agent is temporally prior to its effect that univocity is impeded. Thus, to speak of a cause as univocal or non-univocal is to say something of it in terms of signification, in terms of predicable universality, but there is no question of the cause being predicated of its effect, nor is the cause being looked at as in the first way of speaking of prior or universal cause mentioned above. We are not concerned with how the cause alone can be denominated, but how the cause and effect can receive a common name. And, as having a name in common, cause and effect will be named either univocally or non-univocally, something to be determined by appeal to the ratio signified by the name. Univocal names do not as such imply a cause/effect relationship between what is named, but cause and effect can have a name which is common to them either univocally or non-univocally.


Presupposed by the discussion of this chapter is the view that the agent and its effect are in some way similar, a view expressed in the dictum: omne agens agit sibi simile. To speak of univocal and non-univocal causes therefore, is to say something about the kind of similitude that exists between effect and cause. A perfect similarity enables us to name them univocally;{5} when perfect similarity is absent, a univocal name is impossible. It is in this connection that St Thomas will say that a diverse mode of existing on the part of that in which cause and effect are similar impedes univocation.{6} In order to arrive at an understanding of that phrase, which has sometimes been interpreted as meaning that esse  or existence is the source of analogy (rendering analogy fundamentally metaphysical, it is thought), we shall first examine statements on the gradation of similarity between effect and cause, a gradation at issue in designating causes as univocal or non-univocal.

    Although different statements of this hierarchy can be found, we take as basic a text referred to in a previous chapter.{7} The context of the passage to be considered is this: why is it that some things can be products both of art and nature while others are products of art alone? For example, health can be brought about either by natural causes or by the art of medicine, whereas houses are produced by art alone.{8} In either case, the argument runs, there must be similitude of cause to effect. Only per se causality implies this similitude, of course; the agent to which an accidental effect is attributed is not such that its effect is similar to it. But that which generates or causes per se is cause of its effect as such and there must be some kind of similtude of agent with effect.

Sed doc contingit tripliciter: (a) Uno mod quando forma generati praecedit in generante secundum eumdem modum essendi et simili materia. Sicut cum ignis generat ignem, vel homo generat hominem. Et haec est generatio totaliter univoca. (b) Alio modo, quando forma generati praecedit in generante, non quidem secundum eumdem modum essendi, nec in substantia eiusdem rationis; sicut forma domus praecedit in artifice, non secundum esse materiale, sed secundum esse immateriale, quod habet in mente artificis, non in lapidibus et lignis. Et haec generatio est partim ex univoco, quantum ad formam, partim ex aequivovo quantum ad esse formae in subiecto. (c) Tertio modo quando ipsa tota forma generati non praecedit in generante, sed aliqua pars eius, aut aliqua pars partis; sicut in medicina calida praecedit calor qui est pars sanitatis, aut aliquid ducens ad partem sanitatis. Et haec generatio nullo modo est univoca.{9}
As we have seen in the previous chapter, (a) and (b) here differ as do the ratio concreta and the ratio abstracta. If we do not take into account the fact that the form is realized in different matter, or in one case in matter and in the other not in matter, we can achieve a remote, abstract or logical univocity. A determinate statement of the of realization, the mode of existing, results in equivocity. Thus the sun and the terrestial bodies it wams will not be called hot univocally, i.e. "secundum eamdem rationem speciei, cum non communicent in materia."{10} In both cases, however, the form has esse naturale.{11} And just as a logical genus common to material things and angels can be formed if we ignore that the ratio substantiae is founded with matter in the former and without matter in the latter,{12} so the artisan who conjures up in his mind what he will effect in matter can be called a univocal agent insofar as the form in his mind and the form he effects are the same.{13} When we consider their different modes of existence, however, we can deny univocity; the form of the house in the carpenter's mind and as realized in bricks and lumber are not house in the same sense.{14} Such an agent, however, is not wholly equivocal. Let us look more closely at the second member of this division.

     St Thomas has exemplified the agent which is partly univocal and partly equivocal by the house in the mind of the artisan and the completed house. He uses this example in another text when he makes the remark which heads this section.

Item patet quod, etsi una sit ratio formae existentis in agente et in effectu, diversus tamen modus existendi impedit univocam praedicationem; licet enim eadem sit ratio domus quae sit in maeria et domus quae est in mente artificis - quia unum est ratio alterius - non tamen domus univoc de utraque praedicatur, propter hoc quod species domus in materia habet esse materiale, in mente vero artificis immateriale.{15}
It will be noticed that ratio is used in two ways in this text: as what the term means and, in the parenthetical remark that the artisan's idea is the ratio of the house, as cause. What is meant here by diversus modus existendi: is esse or existence being assigned the role of that which bases or causes analogy? In the first place, we must realize that the reference is not simply to the numerical diversity of acts of existence since this, far from impeding univocation, is a requisite for it. Unless Peter and Paul are diverse in existence, they would not be several things for what-it-is-to-be-a-man to be common to. "Hujus ratio est quia cum in re duo sit considerari: scilicet naturam vel quidditatem rei, et esse suum, oportet quod in omnibus univocis sit communitas secundum rationem naturae, et non secundum esse, quia unum esse non est nisi in una re..."{16} Since numerical diversity of existents (and consequently of acts of existence){17} does not destroy univocation, the meaning of the phrase in question must be sought elsewhere. That is, the meaning must be sought, not in existence, but in what diversifies existence. For the form to be is, in the one case, for it to be in the mind, in the other to be in matter, and this is a difference which is expressed in the concrete notion of each. Existence of itself causes no difference, since this is what all things have in common. There is, then, no basis for an "existential" interpretation of the phrase, "diversus modus existendi impedit univocam praedicationem," as if existence were the cause of non-univocal signification. Diversity must be sought in something other than existence.{18}

    It may be well to point out here that the diverse mode of existence of the form in the mind and in matter which impedes univocity is not the diversity of the concept and that of which it is the concept. If this is what were meant, we might think that the very notion of univocity is called into question by the example. That is, we might ask how "man" can univocally signify the concept of man and such individuals as Peter and Paul. But of course the concept is that through which Peter and Paul are named man univocally. The form in the mind that St Thomas is speaking of is not the concept, but the idea.{19}

    We have spoken thus far only of equivocal causes which are in some way univocal. What then of causes which are in no wise univocal? God is called an analogous cause of creatures.{20} "...non dicitur esse similitudo creaturae ad Deum propter communicantiam  in forma secundum eandem rationem generis et speciei: sed secundum analogiam tantum; prout scilicet Deus et ens per essentiam, et alia per participationem."{21} St Thomas points out that even if, per impossibile, a form had the same ratio in God and creature, the name signifying the form would not be univocally common to them because of diverse modes of existence.{22} But of course the form cannot be of the same ratio, for then there would be some univociaty between god and creature.


The designation of causes as univocal or non-univocal seems to involve little more than the application to cause and effect of what we have already seen of univocation and analogy. Nevertheless, because it is to causes that types of predication ae here applied, we must notice the difference between reduction of causes and reduction of predicates.

Dicendum quod licet in praedicationibus oporteat aequivoca ad univoca reduci, tamenin actionibus agens non univocum ex necessitate praecedit agens univocum. Agens enim non univocum est causa universalis totius speciei, ut sol est causa generationis omnium hominum. Agens vero univocum non est causa agens universalis totius speciei, alioquin esset causa sui ipsius, cum sub specie contineatur, sed est causa particularis huius individui, quod in participatione speciei constituit. Causa autem universalis totius speciei non est agens univocum. Causa autem universalis est prior particulari. - Hoc autem agens universale, licet non sit univocum, non tamen est omnino aequivocum, quia sic non faceret sibi simile; sed potest dici agens analogicum; sicut in praedicationibus omnia univoca reducuntur ad unum primum, non univocum, sed analogicum, quod est ens.{23}
When the analogical cause is called universal, its community is not one of predication. The sun is not predicated of all generable things, although its causality extends to them all. It is on the basis of the difference between these two kinds of community that St Thomas argues for a different reduction of causes and predicates. Causes are such that univocal agents are reducible to a first equivocal or analogical cause; things said equivocally, on the other hand, are reduced to the univocal. Or are they? In the passage quoted, St Thomas seems to contradict himself: at the outset, he says the equivocal is reduced to the univocal predicate; at the end, he says that the univocal predicate is reduced to an analogical one. There are two ways of explaining this shift.

    First, we can understand the reduction to univocity to apply to analogous names. The analogous name is first of all univocal, having like any name its ratio propria. So long as its proper notion is all it has, it can  only be used metaphorically of things which do not save the proper notion.{24} It is only when its signification is extended, when it receives a ratio communis, that it becomes analogous. For example, "healthy" first of all signifies what has a proper proportion among its humors, and only animals save this notion; anything else is called healthy metaphorically. However, when usage sanctions the extension of the meaning of the term, urine, food and medicine can be called healthy properly, if less so than animal. The extension of meaning whereby a univocal term becomes analogous does not eradicate its ratio propria, however; as we have seen, its extended meanings involve a reference to what saves the ratio propria. In any absolute reduction of names, St Thomas suggests, we are going to get back to "being" which is analogically common. Moreover, insofar as the first cause is named being, we are faced with a name common to God and creature which is incorrigibly analogous. Though creatures are named analogically, it is always possible to have a name which is univocally common to them, if only in terms of a logical genus. Between God and creature, however, no univocal name is possible and, since God must always be named from creatures, any name applied to him will be, if proper, analogous - even the name "God."{25}

    A second way of resolving the seeming contradiction is suggested by the fact that the objection is stated in terms of pure equivocation.{26} A name which, when applied to different things signifies unrelated rationes, can be reduced to univocation by restricting it to one ratio. Thus, while a star and barking animals are called dog equivocally, barking animals are so named univocally.

    In another text as well, St Thomas contrasts what is most common in predication and what is most common in causality. After setting down the difference between these two kinds of community, he writes:

Omnium autem entium sunt principia communia non solum secundum primum modum, quod appellat. Philosophus in XI Metaphysicorum, omnia habere eadem principia secundum analogiam, sed etiam secundum modum secundum, ut sint quaedam res eaedam numero existentes omnium rerum principia, prout scinicet principia accidentium reducuntur in principia substantiae, et principia substantiarum corruptibilium reducuntur in substantias incorruptibiles, et sic quodam gradu et ordine in quaedam principia omnia entia reducuntur.{27}
The analogical cause is not the reification of a more common predicate; rather this cause is unum numero and is designated universal from the multitude and diversity of its effects. And, having a name in common with its effects, a name which is in no wise univocal, the cause is designated analogical. but what is first in the order of causality need not be first in the order of the signification of the name. To use the familiar example, although from the point of view of the community of the name, animal is first denominated healthy, from the point of view of causality, medicine is prior. When medicine is said to be an analogical cause, the order of the signification is not thereby changed: we have seen that medicine may be able to cause health because it has part, or a part of a part, of what health consists in.{28}

    At this point, it may occur to one that the examples St Thomas gives when he is speaking of the analogy of names usually involve a cause/effect relationship and it may appear that it is impossible to speak of the former without appeal to the latter.{29} In such texts, St Thomas speaks of the primary analogate, the per prius of the name, in terms of the various kinds of cause. That the analogy of names is not inextricably linked with causality is a point which bears repetition.

    A diligent perusal of the texts in question indicates two things. First, that the doctrine of the analogy of names does not require any explicit mention of a cause/effect relation between what is named analogically; second, the foundation in things for analogical signification is often but not necessarily, such a relation. The proximate foundation of a logical intention is things as known and when the things named analogically are related as cause and effect, the order of the signification of the name need not reflect the real order of prior and posterior.{30} Moreover, when things are said to be proportioned to a cause, this is not necessarily their cause. This is obvious in the example usually given of an efficient cause as the per prius  of an analogous name.{31} The things which are said to receive a common name with reference to a cause are not necessarily its effects.

    What is relevant to the analogy of names is that there be an order among the things as known. Whatever be the foundation in reality for their similarity - and it need not be cause/effect - it is unimportant for the statement of what it means for things to be named analogically. Indeed, we would be hard pressed in the case of many analogous names - e.g. "virtue"{32} - to find a relation of cause and effect between the things named analogically. One reason for believing that things named analogically must be such that one is cause, the other effect, is that St Thomas' most explicit statements on analogy are found in treatments of the divine names. And yet, St Thomas will insist that God is not named good, wise, etc. only causaliter{33} - as we shall see in Chapter IX.


In speaking of God as cause, St Thomas calls him an analogical cause. Moreover, he is in every way an analogical cause so that in no way can a name signify God and creature univocally. Not even a logical genus can comprise God and creature. Despite the unequivocal nature of such statements, St Thomas sometimes speaks as if God were in the same genus as the creature. The most striking instance of this is found in the quarta via, a proof of God's existence drawn from the hierarchy in reality. Things are more and less good, true, noble; but "more and less" implies an approximation to the "most," as the warmer approaches the warmest. Thus there must be something which is truest, best and most noble, and, consequently, maxime ens.

Quod autem dicitur maxime tale in aliquo genere, est causa omnium quae sunt illius generis; sicut ignis, qui ext maxime calidus, est causa omnium calidorum (...) Ergo est aliquod omnibus entibus est causa esse, et bonitatis et cuius libet perfectionis: et hoc dicimus Deum.{34}
God is here spoken of as the maximum in the genus of being, something which seemingly involves two things elsewhere emphatically rejected by St Thomas: that God is in a genus and that being is a genus. That which is the maximum in any genus is the measure of everything else in that genus. This recalls the discussion from the Metaphysics where white is said to be first in the genus of color.{35} White is the chief color and its opposite, black, is its privation, that which is at the furthest remove from it. All other colors are spoken of as approaching more or less to white as to their measure.{36} Of the many things that could be said of this, let us single out the following: the genus in question is a physical genus.{37} The same thing must be said of the example of fire in the genus of warm things: the univocity involved is based on a physical genus.{38} But of the genus of warm things we can assign a maximum which is outside it, namely the sun which is accordingly said to be something more than the hottest thing.{39} The sun and terrestial hot things, though not in the same genus because of the supposed difference in their matter, are in the same genus logice loquendo. How can St Thomas apply such considerations to God's causality? Since the sun is considered to be a maximum outside the physical genus and cause of heat in the genus, the example provides St Thomas with a stepping stone to the view that God is a maximum outside the genera of his effects, whether genus be understood logice or physice: if God is first in the genus of being, "genus" must be understood largo modo.
Ita etiam in genere substantiae, illud quod habet esse perfectissimum et simplicissimum dicitur mensura omnium substantiarum, sicut Deus. Unde non oportet quod sit in genere substantiae sicut contentum, sed solum sicut principium, habens in se omnem perfectionem generis sicut unitas in numeris, sed tamen diversimode quia unitate non mensurantur nisi numeri, sed Deus est mensura non tantum substantialium perfectionum, sed omnium quae sunt in omnibus generibus, sicut sapientiae, virtutis et hujusmodi. Et ideo quamvis unitas contineatur in uno genere determinato sicut principium, non tamen Deus.{40}
God is not in a genus as the principle of one determinate genus. He can, however, be said to be in a genus largo modo.{41} What is this genus in a wide sense? It is precisely the commune analogicum which is opposed to the genus univocum.(42} Thus, God and creature are not in any genus which could give rise to univocity, any more than being can be a genus univocum. Nevertheless, St Thomas will sometimes say that being is a genus.{43} "Genus" must then be taken in an extended sense, i.e. analogously, and so long as we do not think "being" fulfills the ratio propria of "genus," we are not likely to attribute to St Thomas contradictory views. So too when God is said to be maximum in genere entis: this is in no way suggests that God is a univocal cause of creatures.


{1} In II Physic., lect. 6, n. 3: "Advertendum est autem quod causa universalis et propria, vel prior et posterior, potest accipi aut secundum communitatem praedicationis, secundum exempla hic posita de medico et artifice; vel secundum communitatem causalitatis, ut si dicamus solem esse causam universalem calefactionis, ignem vero causam propriam: et haec duo sibi invicem correspondent."

{2} Cf. In V Metaphys., lect. 3, n. 785.

{3} In X Metaphys., lect 3. n. 1964.

{4} In VIII Physic.,  lect. 10, n. 4: "Et similiter est in omnibus aliis, in quibus movens est univocum, idest conveniens in nomine et ratione cum moto; sicut cum calidum facit calidum, et homo generat hominem. Et hic ideo dicit, quia sunt quaedam agentia non univoca, quae scilicet non conveniunt in nomine et ratione cum suis effectibus, sicut sol generat hominem. In quibus tamen agentibus, etsi non sit species effectus secundum eandem rationem, est tamen quodammodo altiori et universaliori." Cf. In II Physic.,  lect. 11, n. 2.

{5} In II Metaphys.. lect. 2, n. 293: "Facit autem mentionem de univocatione, quia quandoque contingit quod effectus non pervenit ad similitudinem causae secundum eamdem rationem speciei, propter excellentiam ipsius causae. Sicut sol est causa caloris in istis inferioribus: non tamen ; inferiora corpora possunt recipere impressionem solis aut aliorum caelestium corporum secundum eamdem raationem speciei, cum non communicant in materia. Et propter hoc non dicimus solem esse calidissimum sicut ignem, sed dicimus solem esse aliquid amplius quam calidissimum."

{6} Q.D. de pot.,q. 7, a. 7.

{7} In VII Metaphys., lect. 8, nn. 1443-6; cf. Q.D. de ver.,  q. 27, a. 7; I Sent. d.8, q. 1, a. 2.

{8} Another example: although men have a natural ability to move themselves about, they need an art to be able to dance. Cf. In VII Metaphys., lect. 8, n. 1439.

{9} Ibid. nn. 1444-6.

{10} In II Metaphys., lect. 2, n. 293.

{11} Q.D. de ver.,  q. 27, a. 7; cf. In II de anima, lect. 5, nn. 282-6.

{12} In Boethii de trin.,  q. 4, a. 2; ibid., q. 6, a. 3.

{13} In VII Metaphys., lect. 8, n. 1447: "Potest enim dici quod generatio fit vel ex forma, sive parte formae, vel ex habente formam, vel partem formae. Sed ex habente quidem sicut es generante; ex forma sive parte formae, sicut ex quo generans generat. Nam forma non generat nec agit, sed habens formam per eam."

{14} Cf. Q.D. de ver., q. 27, a. 7, "tertio modo." Man, in reproducing himself, is not acting as artisan, but as agens naturale. Cf. Ia,  q. 15, a. 1: Q. D. de pot. q. 7, a. 1, ad 8. For a discussion of father and semen as causes of the child, cf. In VII Metaphys., lect 8. nn. 1451-3.

{15} Q.D. de pot., q. 7, a. 7.

{16} I Sent., d. 35, q. 1, a. 4; cf. ibid.,  d. 8, q. 4, a. 2; Q.D. de ver.,  q. 2, a. 11.

{17} Cf. Q.D. de pot., q. 7, a. 2, ad 5.

{18} Q.D. de pot., q. 7, a. 7: "Et praeterea ens non dicitur univoce de substantia et accidente, propter hoc quod substantia est tamquam per se habens esseaccidens vero tamquam cuius esse est inesse. Ex quo patet quod diversa habitudo ad esse impedit univocam praedicationem entis." The difference between esse substantiale and  esse accidentale is founded, not in esse, but in essence. It is difficult to see what it would mean "to put the accent on esse" in this matter.

{19} Q.D. de ver. , q. 3, a. 1.

{20} I Sent.,  d. 8, q. 1, a. 2.

{21} Ia, q. 4, a. 3, ad 3.

{22} Q.D. de pot., q. 7, a. 7. On the problem of the creature's similarity to God and to God's ideas, cf. ibid., q. 7, a. 1, ad 8.

{23} Ia, q. 13, a. 5, ad 1. Cf. In boethii de trin., (ed. Calcaterra), proemium, q. 1, a. 4, ad 4.

{24} See Chapter VIII, section 4.

{25} Ia,  q. 13, a. 10.

{26} Ia, q. 13, a. 5, obj. 1: "Omne enim aequivocum reducitur ad univocum, sicut multa ad unum. Nam si hoc nomen canis aequivoce dicitur de latrabile et marino, oportet quod de aliquibus univoce dicatur, scilicet de omnibus latrabilibus; aliter enim esset procedere in infinitum. Inveniuntur autem quaedam agentia univoca, quae conveniunt cum suis effectibus in nomine et definitione, ut homo general hominem; quaedam vero agentia aequivoca, sicut sol causat calidum, cum tamen ipse non sit calidus nisi aequivoce. Videtur igitur quod primum agens, ad quod omnia agentia reducuntur sit agens univocum. Et ita quae de Deo et creaturis dicuntur, univoce praedicantur."

{27} In Boethii de trin., q. 5, a. 4.

{28} In VII Metaphys., lect. 8, n. 1446.

{29} Cf. De principiis naturae,cap. 6; In IV Metaphys, lect. 1, nn. 537-9.

{30} I Contra Gentiles, cap. 34; Q.D. de ver.,  q. 1, a. 2.

{31} In IV Metaphys., lect. 1, n. 538.

{32} Cf.  Q. D. de virt. in com., a. 7.

{33} Ia q. 13, a. 2.

{34} Ia, q. 2, a. 3.

{35} In X Metaphys., lect. 5, n. 2023, 2027-0. St Thomas explicitly reclls this discussion. Cf. I Sent., d. 8, q. 4, a. 2, ad 3: "Exinde transumptum est nomen mensurae ad omnia genera, ut illud quod est primum in quolibet genere et simplicissimum et perfectissimum dicatur mensura omnium quae sunt in genere illo eo quod unumquodque cognoscitur habere fe veritate generis plus et minus, secundum quod magis accedit ad ipsum vel recedit, ut album in genere colorum." Cf.  I Sent., d. 24, q. 1, a. 1.

{36} In X Metaphys. lect. 5, n. 2025,

{37} Ibid., n. 2024.

{38} In II Metaphys.,  lect. 2, n. 292.

{39} Ibid., n. 293.

{40} I Sent., d. 8, q. 4, a. 2, ad 3; cf. Q.D. de pot.,  q. 7, a. 7, ad 4.

{41} "Vel dicendum quod veritas prima est quodammodo de genere animae largo modo accipiendo genus, secundum quod omnia intelligibilia vel incorporalia unius generis esse dicuntur." - Q.D. de ver.,  q. 1, a. 4, ad 8 in contr.

{42} Q.D. de malo, q. 7, a. 1, ad 1.

{43} Q.D. de malo, q. 1, a. 1, ad 11: "...prout genus dici potest id quod genera transcendit, sicut ens et unum."  De ente et essentia, cap. 7: "...substantia, quae est principium in genere entis..." In IV Metaphys.,  lect. 4, n. 583: "...inducunt in unum et ens tramquam in genera; sed ratione suae communitatis quandam similitudinem generum habent." In IV Metaphys., lect. 2, n. 563: "...omnes partes habent pro genere unum et ens." In X Metaphys., lect. 2, n. 563: "...omnes partes habent pro genere unum et ens." In X Metaphys., lect. 8, n. 2092: "Sed est quasi genus, quia haabet aliquid de ratione generis, inquantum est communis."

© 2012 by the Estate of Ralph McInerny.
All rights reserved including the right to translate
or reproduce this book or parts thereof in any form.

<< ======= >>