Cursus Philosophicus Thomisticus, Secunda Pars Artis Logicae

This difficulty depends on two other general difficulties: specification from extrinsic principles, for example from objects, and the individuation of accidents by subjects; these are to be treated elsewhere.¹

As to the first, for now WE ASSUME the more common opinion: that powers and acts can be specified by their objects, as St. Thomas proves in 1. p. q. 77. art. 3. They are specified by objects taken not materially but formally, that is, insofar as they pertain to the formality and order of the thing being specified. We will speak about this at greater length in the books On the Soul.²

Accordingly, since a relation depends on its foundation and term as on a principle and an end—the reasons for which objects are said to specify, as St. Thomas teaches in 1. 2. q. 1. art. 3.—we must consider how it derives its species from them. Here we do not take the term materially or the foundation remotely. For it is clear that several relations terminate materially at the same term: in one man, for example, the relation of father, the relation of likeness, and relations of cause or effect, among hundreds of others, terminate; likewise, several relations can be founded in the same subject, or remote foundation.

WE INQUIRE, then, into two matters concerning specific unity and diversity: First, whether they are taken from the foundation alone or also from the term; second, if they are taken from both, how each contributes to specification. As for numerical unity, the only question is whether it is taken from the unity or plurality of terms, or from the unity of the subject, even if the terms are numerically many.

On the first question, it seems on one side that a relative cannot be specified by its term. First, correlatives are simultaneous in nature; therefore one does not stand as prior to the other, and hence neither as specifying it, since the specifying factor is prior to what it specifies, as its cause. Second, the term is opposed to the relation; but a species is not taken from its opposite, since that would yield an opposite species. Third, the term is a pure terminus with respect to the relation, and so does not specify it: what is a pure terminus exercises no causality but is simply regarded, whereas specifying is a kind of causing. Finally, in a son there is only one relation to his father and mother, as St. Thomas teaches in 3. p. q. 35. art. 5.; for there is only one birth in him, and therefore one filiation, which follows that birth. Yet it terminates at terms that differ in species, namely paternity and maternity, which are different relations, just as the foundations for generation are altogether different. Therefore the specific unity of a relation is compatible with terms that differ in species, and thus is not taken from them.

Conversely, that the term alone specifies is proved as follows. First, by a general consideration: everything that is toward another is specified by that toward which it is, as are habits and acts. A relation, whose whole being is toward another, should all the more be specified by it, and hence so should its whole quiddity and species. Second, the same foundation can serve different relations: the same whiteness is the foundation of likeness to snow and unlikeness to a crow; the same quality in an inferior cause is the foundation of its relation to its effect and to a superior cause; and the same sign provides the foundation of its relation to a cognitive power and to what is signified. Therefore relations are distinguished by their terms, not by their foundations.

NEVERTHELESS, IT MUST BE SAID that both contribute to the specification of a relation, and neither excludes the other. The relation receives specification from its foundation as from a cause and specifying principle, but from its term not as from a cause of specification, rather as from that which completes and terminates the account of specification.

Hence Albert rightly said,³ as Master Soto reports in q. 1. On Relation, ad 8., that one relative is not defined by another but with reference to another: “by” indicates a cause, while “to” indicates an order. St. Thomas generally takes the specification of a relation from its foundations, because he says that a relation is real by virtue of its foundation. This is clear in 1. dist. 26. q. 2. art. 2. ad 4.,⁴ where he says: “Since a relation has natural being only because it has a foundation in a thing, it is placed in a genus on that account. Hence the essential differences of relations are taken according to the differences of other beings, as is clear from the Philosopher in 5. Metaph., where he says that some are founded on quantity, others on action, and so on. Accordingly, the order of relations is also according to the order of the things in which they are founded.” This, then, is why in many places St. Thomas takes the specific account of relations from their foundations: in 3. p. q. 35. art. 5., Quodlib. 1. art. 2.,⁵ and Quodlib. 9. art. 4.,⁶ Opusc. 2. cap. 212.,⁷ and in 3. dist. 8. q. 1. art. 5.,⁸ where, in reply to objection 4, he also distinguishes motion from relation: “Every motion is something in reality, but not every relation is. Therefore, although the relations of a relation are multiplied by its terms, it does not follow that relations are multiplied in reality, as motions are multiplied in reality by the diversity of their terms.” In that passage the Angelic Doctor always traces the reality of relations back to their foundations, and, given the foundation, to their terms, though not to the terms alone, as in the case of motion. The same can be drawn from St. Thomas’s general rule in 1. 2. q. 54. art. 2., where he says “that all things said relatively are distinguished according to the distinction of the things to which they are said.” Therefore relations too derive their specification from their terms. He states this also in 1. p. q. 32. art. 2., where he says “that two relations are not specifically different if one relation corresponds to them from the opposite side”; we will examine that passage below.

The reason for these positions follows from what has been said: the whole reality of a relation comes from the foundation according to its order toward the term, since, as its definition says, the whole being of a relation is toward another. Since it essentially requires both, foundation and term, its specification should not be taken from one alone, to the exclusion of the other.

THE SECOND PART of the conclusion, concerning how these two contribute to specification, should not be understood as their contributing only partially, with the foundation supplying one part of the specification and the term another. Rather, each supplies the whole specification in a different kind of cause. Some explain this by saying that the foundation contributes initiatively and the term completively. Others say that the foundation contributes in the genus of efficient cause and the term in the genus of extrinsic formal cause. Still others say that the foundation specifies by virtually containing beforehand the term to which it is proportioned, so that the diversity of foundations is referred back to a diversity in the formality of the terms.

Still, a distinction must be made. The term is either taken most formally as an opposed term, or fundamentally, from the side of the absolute thing that grounds this account of terminating. In the first way, the term contributes to specification purely by terminating, not by causing it, because in this way it is a pure term and is simultaneous in nature and cognition with the relation. Therefore, taken in this way, it is not a specifying cause, since a cause is not simultaneous in nature with its effect, but prior to it. If considered in the second way, it functions as an extrinsic formal cause and specifies in the manner of an object. In this way, a single account of specifying a relation arises from the foundation and the term, insofar as the foundation contains the term within itself in proportion and virtually. For a foundation is not toward such a term unless it is such a foundation, and conversely. Thus, insofar as they are proportioned to one another, they make up one account of specifying a relation, which requires both such a foundation and a term corresponding to it.

From this you can also gather what the formal term is in its role as a specifying factor. Although different relations specifically can terminate at the same material term, they cannot terminate at the same term formally. The formal specifying account in the term is taken according to its correspondence and adequate proportion to its foundation. For example, white constitutes the formal term of a relation of likeness insofar as it corresponds to agreement and identity; of unlikeness insofar as it corresponds to disagreement; and of paternity or filiation insofar as it corresponds to generation, and so in other cases. Thus, just as the foundation is considered under the ultimate account by which it grounds, so the term is considered under the proportion and correspondence by which it terminates.

We said “which corresponds completely or adequately,” because one total term can correspond to relations that differ specifically, as the relation of paternity and the relation of maternity correspond to filiation. For filiation is caused by those two modes of generation, paternal and maternal, but proceeds by one birth. St. Thomas teaches this in 3. p. q. 35. art. 5. ad 3. Yet he seems to say the contrary in 1. p. q. 32. art. 2., where he says “that two relations are not specifically different if one relation corresponds to them from the opposite side.” Those who think that both father and mother contribute actively do not escape this difficulty. For even if both contribute actively, they do so in different ways, sufficient to ground the different relations of paternity and maternity, just as the sun and a human being both contribute actively to generation, but in different ways that ground different relations of cause. Therefore the reply is that one total term, or correlative, can correspond to two partial relations, but not to two total relations. Father and mother, and several causes of the same order contributing to the same effect, are diverse in a quasi-partial way; together they constitute one complete cause and one total term, even though there are several partial ones. In the first passage, from 3. p., St. Thomas is speaking of several partial relations on one side, to which only one adequate relation can correspond on the other side. In the second passage, from 1. p., he is speaking of different relations that correspond adequately and completely.

TO THE ARGUMENTS FOR THE OPPOSITE OPINIONS: To the first, it is answered that although the term is simultaneous with its correlative, specification depends on it, not as on a principle that causes specification, but as on that which terminates and completes it. Thus all the priority of the specifying cause belongs to the foundation. The relation is not specified by the term, but with reference to it.

To the second, it is answered that specification can be taken from a term opposed by relative opposition, as from that which terminates and completes, not as from that which causes it. For relative opposition is the least of all oppositions: it does not remove the other extreme but posits or presupposes it, and thus the specification of the relation can be ordered toward it.

To the third, it is answered that a pure term, under the formality of a term, can specify by terminating and completing, not by causing; it specifies as that to which, not that by which.

The answer to the last argument is clear from what has been said.

TO THE ARGUMENTS FOR THE SECOND OPINION: The first has already been answered: terms specify by purely terminating, while foundations specify by causing. Thus the whole being of a relation is toward the term as that which purely terminates it.

To the second, it is said that the same foundation cannot serve proximately and formally for relations that differ in species. For example, whiteness, under the formality of agreement, grounds likeness; under the formality of disagreement, it grounds unlikeness. Likewise, in the other examples cited, if they are different relations (a question I am not disputing), there are also different proximate and formal foundations.

As to the second difficulty involved in this article, concerning the numerical distinction and unity of a relation, it depends on the famous question whether accidents are individuated by their subject, such that two accidents differing only numerically cannot be in the same subject. This need not be treated here. We need only explain whether one relation can extend to and concern several terms, each of which sufficiently terminates it, as one father has several sons. This seems impossible. First, the same effect cannot depend on several total causes of the same order and account. But each son is a sufficient and total term of paternity, since the relation is specified sufficiently and totally by him; therefore the same relation cannot depend on a subsequent son. Also, a new generation occurs with respect to a new son, and therefore there is a new relation, which is identified with the generation. Nor can the relation be extended anew, since this extension would be a new increase and addition to the relation, not part of it; but a relation has no parts, and so it would be a new relation. Finally, a relation is an indivisible and simple mode. Therefore, if one son dies while another remains, the relation to one persists and that to the other ceases; the same entity would thus persist and not persist, since it is simple.

THE ANGELIC DOCTOR HOLDS THE OPPOSITE OPINION in 3. p. q. 35. art. 5., Quodlib. 1. art. 2.,⁹ and Quodlib. 9. art. 4.¹⁰ AND ST. THOMAS PROVES IT not only from the general account of accidents, which cannot be multiplied numerically in the same subject, but also from the special case of relations: the cause or foundation is numerically the same. For example, in a relation of equality, the same quantity is related to all quantities of the same extension—a single cubit is equal in the same quantity to every cubit; likewise, the same generative power is related to all sons, though modified by different actions. Also, to avoid an infinite multiplication of relations: in one subject there are infinitely many proportions among parts, yet in one teacher teaching many there are not many acts of teaching, but one; and in one dominion and one king there are not as many dominions and relations as there are subjects, since many can be taught and ruled by the same action. Finally, a son is related to his father and mother by one relation, and several people drawing a ship are related to it by one relation. Therefore one relation can likewise reach all the terms of its species as inadequate to its extension, although each suffices for its existence; and so, although this relation numerically regards this particular term, it does not regard it adequately, but also regards all the others in which the formality of such a term is present.

TO THE OPPOSING ARGUMENTS it is answered: To the first, we say that one effect cannot depend formally on several total and adequate causes. If, however, the causes are several only materially, the effect can indeed depend on them, insofar as they agree in one mode of causing or terminating. Thus, if ten people suffice to pull a stone, twenty can do so as well; they are more numerous only materially, but formally agree in one account of causing. Likewise, the relation of a father is sufficiently terminated by one son; yet if there are several, it is also terminated by them, insofar as they agree in the same account of terminating one generative power modified by different actions. For this relation is founded on that power as thus modified, as we said from St. Thomas in article 3.

Finally, this is explained by different examples. We see that when one effect has been produced by one sufficient cause, and another sufficient cause comes along, it does not produce the effect anew, but engages with it in a new way. For example, alteration produces a quality, while intensification, which comes about anew, does not produce it but applies and reaches it in a new way, although it would be sufficient to produce the quality. Likewise, although transubstantiation in the Eucharist is sufficient to produce the substance of Christ’s body, because it finds that substance already produced, it does not produce it anew but applies it in a new way. So although one son sufficiently terminates the relation of a father, if a new son comes along after the relation has been produced, he does not terminate a new relation but extends the same relation as pertaining to him. For St. Thomas’s opinion that there is only one existence in the whole composite, a more fitting example is in 3. p. q. 17. art. 2.: if, when the whole exists, one part is missing—say, a hand or foot—and it comes along later, a new existence is not produced in it; rather, the existence of the whole is applied to it as something pertaining to the whole. We say the same about the simple habit of science extending to several conclusions without the addition of a new habit, and about an intensified quality taking root without the addition of a new quality.

To the second, it is said that although a new generation is posited with respect to a new son, because it takes place at a different time and from different matter (otherwise several sons could indeed be born by the same generation), nevertheless the relation is not identified with the action, nor founded on it, except insofar as it determines the generative power, as we said above; and this power is one. Thus, just as when several sons are born in the same delivery, the father’s relation can only be one, because it is founded on the same action, so too when they are produced by several actions. It is incidental that several actions or one action of the same power modify it toward a relation of the same account. This extension of the relation to different terms does not occur by adding a new relation, but by unfolding the pre-existing one, insofar as it of itself regards all terms proceeding from the same foundation. It is unfolded in act when an action either grounds or at least in fact posits a new term, as St. Thomas explains in 5. Phys. lect. 3.¹¹ concerning the resulting of a relation upon the term’s being posited from the root of its foundation, where it was already present beforehand. For a relation to be emitted anew from its foundation, or for one already emitted to be unfolded or applied to that new term as pertaining to it, is intrinsic to the relation itself: namely, to tend and be determined in act toward that to which it tended virtually of itself, and which was absent only because the term did not exist. This helps explain how the species of an angel can represent a new object without any change in its representation, an object that did not exist before and now begins to exist.

To the third, it is answered that a relation is simple in entity but multiple in its terminating, which is to be divisible relatively while indivisible entitatively. Therefore, when one of the terms materially present there perishes, the relation perishes as to its application and extension to that term, but not simply or in itself. Something similar is found in the simple habit of science, which extends to new conclusions and can lose one conclusion without failing as to the others or entitatively in itself; likewise in the existence of the whole communicated to its different parts, and in one angelic species reaching some objects that exist but not those that do not. All these matters will be examined in their proper places.