LESSON 13

Rejection of Universals as Substances

  Chapter 13: 1038b 1-1039a 23

             650. But since our investigation has to do with substance, let us return to it. And just as the subject and the essence and the composite of these are called substance, so also is the universal. Two of these, then, have been discussed already, namely, the essence (576-597; 622-649) and the subject (568-575); and it has been stated that a thing is a subject in two ways: either as this particular thing (as an animal is the subject of its attributes), or as matter is the subject of actuality. But according to some thinkers the universal also seems to be in the fullest sense a cause and principle. Therefore let us treat of this.

             651. For it seems impossible that any of those things which are predicated universally should be substance. For, first, the substance of each thing is the substance which is proper to it and belongs to nothing else, whereas the universal is common; for that is said to be universal which is suited by its nature to be found in many things. Of what particular thing, then, will it be the substance? For it is either the substance of all or of one. But it cannot be the substance of all. And if it is the substance of one, all things will also be that one; for those things whose substance is one have one essence and are themselves one.

             652. Furthermore, substance means what is not predicated of a subject, whereas a universal is always predicated of some subject.

             653. But while a universal cannot be a substance in the way in which the essence of a thing is, it is found in this in the way in which animal is found in man and in horse. Therefore it is evident that it has some kind of intelligible expression. However, it makes no difference if there is no definitive expression of all those things which are present in substance; for none the less this will be the substance of something, as man is the substance of the particular man in whom it is present. Hence the same thing will happen again, for substance will be the substance of that thing, as animal will be the substance of that in which it is present as its proper form.

             654. Furthermore, it is both impossible and absurd that this particular thing, or substance, if it is composed of certain parts, should not be composed of substances or of a particular thing but of quality; for that which is not substance, i.e., quality, will then be prior both to substance and to the particular thing itself. But this is impossible; for accidental attributes cannot be prior to substance either in intelligibility or in time or in the process of generation; for they would then be separable from it.

             655. Furthermore, Socrates will have a substance in his substance, and therefore it will be the substance of two things. And in general it follows, if man and all terms used in this way are substance, that no one of the parts in the intelligible expression is the substance of anything, nor does it exist apart from the species or in anything else. And I mean that there is no animal existing apart from particular ones, and the same is true of everything contained in the intelligible expressions of things. From these considerations it is evident to those who study the matter that no universal is a substance, and that none of the categories signify particular things but things of such and such a kind.

             656. And if this is not the case, many absurdities will follow, among them the third man (107).

             657. Furthermore, it is also evident in this way that a substance cannot be composed of substances which are actually present in it, for what is actually two can never be actually one; but if something is potentially two, it will be actually one; for example, the whole line consists of two halves existing potentially. For actuality separates. Hence, if substance is one it will not consist of substances present in it. And in this sense Democritus is right; for he says that it is impossible for one thing to be produced from two, or two from one; because he makes indivisible continuous quantities substances. It is evident, then, that the same thing will also be true of numbers if a number is a composite of units as some say, because either the number two is not one or the unit is actually present in it.

             658. But the result involves a difficulty; for if no single substance can consist of universals (because a universal signifies such and such a thing but not a particular thing), and if no single substance can be composed of actual substances, then every substance will lack composition. Hence no substance will have an intelligible expression. But it appears to all, and this has already been stated (587), that it is either substance alone or chiefly substance that is defined. But now it seems that not even this kind of substance is defined. Hence there will be no definition of anything, or in one sense there will be and in another there will not. The meaning of this will become clearer from what follows (669-676; 733-741).

COMMENTARY

             1566. Having settled the issue about substance in the sense of quiddity, the Philosopher now comes to certain conclusions about substance insofar as the universal is considered by some thinkers to be a substance; and in regard to this he does two things. First (650:C 1566), he links up this discussion with the preceding one. Second (651:C 1569), he carries out his plan ("For it seems").

             He therefore says, first (650), that since this science is chiefly concerned with the study of substance, we must return again to the division of substance in order to see what has been said and what remains to be said. Now it is clear from the preceding discussion that substance has the following meanings. First, it means what has the nature "of a subject," namely, matter, which is related to substantial form in the same way as a subject, which is a complete substance, is related to accidental form; second, it means the essence of a thing, which refers to its form; third, it means "the composite of these," i.e., the composite of matter and form; and fourth, it means the universal, according to some thinkers.

             1567. Now the division of substance given here is the same as that given at the beginning of Book VII (568:C 1270), although it seems to differ; for there he gave four senses of substance: the subject, the essence, the universal and the genus. And he divided subject into three meanings: matter, form, and the composite. And since it has already been made clear that essence derives from form, he puts essence in place of form; and again since a common genus is said to be substance on the same grounds as a universal is, as will be shown, he concludes that both belong in the same class; and thus there remain only the four senses in which substance is spoken of here.

             1568. Two of these, then, have been discussed already; for essence has been treated (576:C 1299) and also the subject (568:C 1270), which is taken in two senses. For, first, it means a particular thing and an actual being, as animal is the subject of its predicates, and as any particular substance is the subject of its accidents. Second, it means primary matter, which is "the subject of actuality," i.e., of substantial form. These things were discussed where it was shown (629:C 1501) how the parts of matter pertain to the form and to the individual. But since not only the matter and the quiddity seem to be causes, but also the universal, because "according to some thinkers," i.e., the Platonists, this seems to be in the fullest sense a cause and principle, we will therefore treat "of this," i.e., the universal, in this same seventh book. And in Book VIII (691:C 1681) we will treat of composite and sensible substances, to which the things treated in this seventh book are related as principles.

             1569. For it seems (651).

             Here he begins to investigate whether universals are substances, and this is divided into two parts. In the first (651) he shows that universals are not substances, as some thinkers claimed. In the second (681:C 1642) he shows to what extent the statements of those making this claim are true and to what extent they are false ("But those who").

             In regard to the first he does two things. First, he shows in a general way that universals are not substances. Second (678:C 1637), he shows this in a special way with regard to being and unity, which were assumed to be the substances of things in the highest degree ("And since").

             The first is divided into two parts. In the first he shows that universals are not substances; and in the second (659:C 1592), he shows that they are not separate entities ("And from these").

             In regard to the first he does two things. First, he shows that universals cannot be substances on the grounds that they are predicated of many things; and second (654:C 1579), on the grounds that species are composed of universals as parts of their definition ("Furthermore, it is"). For he had said above, in Book V (524:C 1119), that in one sense a genus is a whole inasmuch as it is predicated of several things, and in another sense it is a part inasmuch as a species is composed of a genus and a difference.

             In regard to the first he does two things. First, he shows that a universal is not a substance on the grounds that it is predicated of many things. Second (653:C 1577), he rejects a captious answer ("But while a universal").

             1570. For the clarification of this chapter it must be noted that the term universal can be taken in two senses. First, it can be taken to mean the nature of the thing to which the intellect attributes the aspect of universality, and in this sense universals such as genera and species signify the substances of things inasmuch as they are predicated quidditatively; for animal signifies the substance of the thing of which it is predicated, and so also does man. Second, a universal can be taken insofar as it is universal, and insofar as the nature predicated of a thing falls under the aspect of universality, i.e., insofar as animal or man is considered as a one-in-many. And in this sense the Platonists claimed that animal and man in their universal aspect constitute substances.

             1571. This is what Aristotle aims to disprove in this chapter by showing that animal in general or man in general is not a substance in reality, but that the form animal or man takes on this generality insofar as it exists in the mind, which understands one form as common to many inasmuch as it abstracts it from all individuating principles. Hence in support of his thesis he gives two arguments.

             1572. Concerning the first of these (651) he says that in the light of the succeeding arguments it seems impossible that any one of those attributes which are predicated universally should be a substance, i.e., insofar as it is taken in its universality. This is proved, first, by the fact that while the substance of each thing is proper to each and does not belong to something else, a universal is common to many; for that is said to be universal which belongs by nature to many things and is predicated of many. Hence, if a universal is substance it must be the substance of some thing. Of what thing, then, will it be the substance? For it must either be the substance of all the things to which it belongs or of one. But it is impossible for it to be the substance of all things, because one thing cannot be the substance of many, since those things are many whose substances are many and distinct.

             1573. But if it is held to be the substance of one of the things in which it is found, it follows that all other things in which it is found, and of which it is held to be the substance, are that one thing; because it must also be their substance for the same reason, since it is found in all in the same way. Now those things of which the substance and essence are one must also be one themselves. Hence, since a universal cannot be the substance of all the things of which it is predicated or of any one of them, it follows that it is not the substance of anything.

             1574. Now it should be noted that he describes a universal as what is naturally disposed to exist in many, and not as what exists in many; because there are some universals which contain under themselves only one singular thing, for example, sun and moon. But this is not to be understood in the sense that the very nature of the species, considered in itself, is not naturally disposed to exist in many things; but there is something else which prevents this, as the fact that all the matter of the species is included in one individual, and the fact that it is not necessary that a species which can last forever in a single individual should be numerically many.

             1575. Furthermore, substance (652).

             Here he gives his second reason. He says that substance refers to something which is not predicated of a subject. But a universal is something which is always predicated of some subject. Therefore a universal is not a substance. But this argument seems not to be cogent, for it is said in the Categories that it belongs to the notion of substance not to exist in a subject. But to be predicated of a subject is not opposed to the notion of substance. Hence in that place second substances are posited, and these are predicated of a subject.

             1576. But it must be said that in the Categories the Philosopher is speaking from the viewpoint of logic. Now a logician considers things insofar as they exist in the mind, and therefore he considers substances insofar as they take on the character of universality from the way in which the intellect understands them. Hence in reference to predicating, which is an act of reason, he says that substance is predicated "of a subject," i.e., of a substance subsisting outside of the mind. But the first philosopher considers things insofar as they are beings, and therefore in his view of the matter there is no difference between existing in a subject and being predicated of a subject. For he takes something to be predicated of a subject which is something in itself and belongs to some actually existing subject. And it is impossible that this be a substance, for then it would have to exist in a subject. But this is contrary to the notion of substance, as is also stated in the Categories.

             1577. But while a universal (653).

             Here he rejects the captious answer by which someone might oppose his first argument, in which he had said that all things are one whose substance and quiddity are one. For someone might say that a universal is not a substance in the sense of the essence of a thing, which is proper to one thing. Therefore with a view to rejecting this the Philosopher says "But while" it might be said, in opposition to the first argument introduced, that it is impossible for a universal to be a substance in the way in which an essence is, it is substance only as something existing in these particular things, as animal exists in man and in horse. For the nature of animal is not found in man in such a way that it is proper to him, because it is also found in horse--as if to say that the argument cannot be answered in this way.

             1578. For if animal in common is a substance, it follows that there is an intelligible expression of this substance. And it makes no difference to his thesis if there is no definitive expression of all those things "which are present in substance," i.e., which are given in the definition, lest there be an infinite regress in definitions, but all parts of any definition must be further defined. For this substance must be the substance of something, even though it does not have a definition, no less than if it has. Thus we might say that, although man in common does not have a definition, it must nevertheless be the substance of the man in whom it is present, namely, of man in common. Hence the same conclusion follows as before, because, even though this common substance is not held to be proper to any one of its inferiors, it must still be proper to that common substance in which it is first found. For example, if animal in common is a substance, animal will be predicated primarily of that common substance and will signify its proper substance, whether it be definable or not. Hence, since this substance is proper to one thing, it will be impossible for it to be predicated of many things.

             1579. Furthermore, it is (654).

             He now shows that the universal is not a substance by basing his arguments on the grounds that the universal is part of the definition and essence. In regard to this he does two things. First (654:C 1579), he gives the arguments in support of his thesis. Second (658:C 1590), he disposes of a difficulty ("But the result").

             In regard to the first part he gives four arguments. First, he says that it is both impossible and untenable that a particular thing and a substance should not be composed of substances or particular things but of those things which signify quality--if it is composed of anything (which he adds to allow for simple substances). For since those parts of which a thing is composed are prior to it, it follows that what is not substance but quality is prior both to substance and to this particular thing. But this is impossible, because it is impossible for modifications and qualities and accidents to be prior to substance either in intelligibility or in time or in generation.

             1580. For it has been shown above (563:C 1253) that they are not prior in intelligibility, because substance is given in the definition of accidents, and not the reverse. And from this it has also been proved above (563:C 1257) that they are not prior in time. From this in turn he further proves here that it would follow that attributes would be capable of existing apart from substances; and this is impossible. And priority in generation comes under priority in time, although the reverse is not true. For even though things which are not related to the generation of something are prior in time, they are still not prior in generation; for example, a horse is not prior in generation to a lion which exists at this moment, even though it is prior to it in time. However, the parts of which a thing is composed are prior in the process of generation and therefore in time, and sometimes also in intelligibility, as was shown above (570:C 1278). Hence it is impossible that substances should be composed of things which are not substances. But universals do not signify particular things, but of what sort things are, as was said about second substances in the Categories. It is evident, then, that singular things, which are particulars, cannot be composed of universals if these are some kind of things which exist apart from singulars.

             1581. But it seems that this argument is not a satisfactory one; for even though second substances, which are genera and species in the genus of substance, do not signify particular things but of what sort things are, nevertheless they do not signify of what sort things are in the same way in which attributes that signify accidental quality do, but they signify substantial quality. However, he argues here as if they signified accidental quality.

             1582. But it must be said that if universals are things, as the Platonists claimed, we shall have to say that they signify not only substantial quality but also accidental quality; for every quality which is distinct from the thing of which it is the quality, is accidental. For example, whiteness differs from the body of which it is a quality, and it inheres in the body of which it is the quality as its subject; and therefore it is an accident. Hence, if universals as universals are things, they must be distinct from singulars, which are not universals. Therefore, if they signify the quality of those things, they must inhere in them as in substances and thus must signify accidental quality.

             1583. However, for those who claim that genera and species are not things or natures distinct from singulars but are the singular things themselves (for example, that there is no man who is not this man), it does not follow that second substance signifies an accident or modification.

             1584. Furthermore, Socrates (655).

             He gives the second argument. He says that if universals are substances, it follows that Socrates will have a substance in his substance; for if all universals are substances, then just as man is the substance of Socrates, in a similar fashion animal will be the substance of man; and thus these two substances, one of which is man and the other animal, will exist in Socrates. His conclusion is "and therefore it will be the substance of two things," i.e., it therefore follows that animal is the substance not only of man but also of Socrates. Hence one substance will belong to two things. Yet it has been shown above that one thing has only one substance.

             1585. And the result mentioned applies not only in the case of Socrates but universally in all cases. For if man and the other things which are called species in this way are substances, it also follows that no one of the parts in the intelligible structure of a species is substance, and that it cannot exist without the species in whose definitions it is given or exist in anything else; just as there is no animal "apart from particular animals," i.e., apart from the species of animal. And the same thing applies to all other predicates which are given in definitions, whether they are genera or differences. And this is true because, if those parts which are given in the definitions of species are substances, then since species are substances there will be many substances in singular things, and many things will have one substance; as was said about Socrates. From what has been said, then, it is evident that no universal is a substance, and that common predicates do not signify a particular thing but of what sort a thing is.

             1586. And if this (656).

             Then he gives the third argument. He says that, if the preceding conclusion is not admitted, many absurdities will follow, and one of these will be the need to posit a third man. This can be explained in two ways. First, it can mean that besides the two singular men, Socrates and Plato, there is a third man, who is common to both. This is not absurd according to those who posit Ideas, although it seems absurd from the viewpoint of right reason.

             1587. Second, it can be explained as meaning that there is posited a third man besides a singular man and man in common, since they have a common name and intelligible expression, just as do two singular men in addition to whom a third common man is posited; and the reason is that they have a common name and definition.

             1588. Furthermore, it is (657).

             He gives the fourth argument. He says that universals are not substances for this reason that it is impossible that a substance should be composed of many substances actually present in it; for two actual things are never one actual thing, but two which are in potentiality are one actually, as is clear of the parts of a continuous quantity. The two halves of one line, for instance, exist potentially in the whole line, which is one actually. And this is because actuality has the power of separating and distinguishing; for one thing is distinguished from another by its proper form. Hence in order that many things may become one actual thing, it is necessary that all should be included under one form, and that each one should not have its own form by which it would exist in act. Hence it is evident that if a particular substance is one, it will not be composed of substances actually present in it; and thus if it is composed of universals, universals will not be substances.

             1589. And in this sense Democritus is right when he says that it is impossible for one thing to be produced from two, and two from one; for it must be borne in mind that two actual existents never make one. But in failing to distinguish between the potential and the actual, he claimed that indivisible continuous quantities are substances; for he thought that, just as one thing does not contain many things actually, neither does it contain them potentially; and thus any continuous quantity is indivisible. Or this might be explained differently. I mean that Democritus was right if we assume his own position to be true, in which he claimed that indivisible quantities are the substances of things and thus are always actual, and in this way no one thing is produced from them. And just as this is true in the case of continuous quantities, in a similar way it is true in the case of numbers, if number is composed of units, as some thinkers claimed. For either the number two (or any other number) is not one thing, or the unit is not actually present in it. Thus the number two will not be two units, but something composed of units; otherwise a number would not be a unity, essentially and properly, but only accidentally, like a heap.

             1590. But the result (658).

             He poses a difficulty about the above answer. He says that the result of the foregoing discussion gives rise to a difficulty; for first (as was said), a substance cannot be composed of universals, because a universal does not signify a particular thing but of what sort a thing is; and second, a substance cannot be composed of actual substances; and thus it seems to follow that substances cannot be composed or made up of substances. It follows, then, that all substances lack composition. And thus, since no definitions are given of substances which lack composition (and this is clear from the fact that the definition is an intelligible expression having parts, as was shown above [622:C 1460]), it follows that no substance has a definition. But it seems to everyone, as was shown above (582:C 1331), that a definition is either of substance alone or chiefly of substance, and it has now been concluded that there is no definition of substance; hence it follows that there is no definition of anything.

             1591. Now the answer to the above difficulty is that in one sense substance is composed of substances and in another it is not. But this will become clearer from the following discussions in this book (669:C 1606) and in Book VIII; for substance is composed of potential substances, not of actual ones.