Chapter 7

We call relatives (pros ti) all those things which are said to be just what they are of other things, or which are in some other way said in relation to something else. For instance, the greater is said to be what it is of something else, since it is said to be greater than something; and the double is said to be what it is of something else, since it is said to be double of something. And so likewise with all other things of this sort.

The following too are among the relatives: state (hexis), disposition, perception, knowledge, position. For all of the things just mentioned are said to be what they are of other things, and not anything else. For a state is said to be a state of something, and knowledge knowledge of something, and position position of something, and so with the rest likewise.

Relatives, then, are all those things which are said to be just what they are of other things, or which are in some other way said in relation to something else. Thus a mountain is called large in relation to something else, for the mountain is called large relative to something; and what is similar is said to be similar to something; and the other things of this sort are likewise said in relation to something.

Further, reclining and standing and sitting are particular positions, and position is one of the relatives. But to be reclining or to be standing or to be sitting are not themselves positions; rather, they take their names paronymously from the positions just mentioned.

Contrariety is found in relatives too. Virtue, for example, is contrary to vice, and each of the two is a relative; and knowledge is contrary to ignorance. Yet contrariety does not belong to all relatives, for there is nothing contrary to the double, nor to the triple, nor to anything else of that kind.

Relatives also seem to admit of the more and the less. For the similar and the dissimilar are said to be so more and less, and the equal and the unequal are said to be so more and less, each of them being a relative; for the similar is said to be similar to something, and the dissimilar dissimilar to something. But not all relatives admit of the more and the less, for the double is not said to be more or less double, nor is anything else of that kind.

All relatives are said in relation to correlatives that reciprocate. The slave, for example, is called slave of a master, and the master master of a slave; and the double double of a half, and the half half of a double; and the greater greater than a less, and the less less than a greater. And so likewise in the other cases, except that sometimes there will be a difference of case in the way it is expressed. Thus knowledge is called knowledge of the knowable, and the knowable knowable by knowledge; and perception perception of the perceptible, and the perceptible perceptible by perception.

Sometimes, however, they will not seem to reciprocate, if the thing in relation to which something is said has not been properly rendered, and the one rendering it has gone wrong. For example, if the wing is rendered as of a bird, then bird of a wing does not reciprocate. For in the first place it was not properly rendered as wing of a bird; for it is not qua bird that the wing is said to be of it, but qua winged, since there are many other things too that have wings and are not birds. Hence, if it is properly rendered, it does reciprocate: the wing is a wing of a winged thing, and the winged thing is winged by a wing.

Sometimes it is perhaps even necessary to coin a name, if there is no established name for that in relation to which the thing might properly be rendered. For example, if the rudder is rendered as of the ship, the rendering is not a proper one; for it is not qua ship that the rudder is said to be of it, since there are ships that have no rudders. That is why they do not reciprocate: the ship is not said to be a ship of a rudder. But perhaps the rendering would be more proper if it were rendered somewhat like this: the rudder is a rudder of a ruddered thing, or in some other such way, since no name is established for it. And it does indeed reciprocate, if it is properly rendered, for the ruddered thing is ruddered by a rudder.

And so likewise in the other cases. The head, for instance, would be more properly rendered as of a headed thing than rendered as of an animal, for it is not qua animal that a thing has a head, since many animals have no head. In this way one might most easily grasp the things for which no names are established, if from the primary things one were to assign names also to the correlatives that reciprocate with them, as in the cases just mentioned "winged" was assigned from "wing" and "ruddered" from "rudder".

All relatives, then, provided they are properly rendered, are said in relation to correlatives that reciprocate; since if they are rendered in relation to some chance thing, and not in relation to the very thing they are said in relation to, they do not reciprocate. I mean that not even in the case of things that are agreed to be said in relation to reciprocating correlatives, and for which names are established, does any of them reciprocate, if it is rendered in relation to one of the accidents and not in relation to the very thing it is said in relation to. For instance, if the slave is rendered not as of a master but as of a man, or of a biped, or of anything else of that kind, it does not reciprocate; for the rendering is not a proper one.

Again, if something has been properly rendered in relation to that which it is said in relation to, then, when all the other things that are accidental to it are stripped away and only that is left in relation to which it was properly rendered, it will always be said in relation to that. For example, if the slave is said in relation to a master, then, when all the other things that are accidental to the master are stripped away, such as being a biped, being receptive of knowledge, and being a man, and only being a master is left, the slave will always be said in relation to that; for the slave is called slave of a master.

But if it has not been properly rendered in relation to whatever it is said in relation to, then, when the other things are stripped away and only that is left in relation to which it was rendered, it will not be said in relation to that. For let the slave be rendered as of a man, and the wing as of a bird, and let being a master be stripped away from the man: the slave will no longer be said in relation to the man, for if there is no master neither is there a slave. Likewise, let being winged be stripped away from the bird: the wing will no longer be one of the relatives, for if there is nothing winged, neither will the wing be a wing of anything.

Hence one must render whatever it is that a thing is properly said in relation to. And if a name is already established, the rendering is easy; but if there is none, it is perhaps necessary to coin a name. When they are rendered in this way, it is evident that all relatives are said in relation to correlatives that reciprocate.

Relatives seem to be simultaneous by nature, and in most cases this is true. For the double and the half are simultaneous, and when there is a half there is a double; and when there is a master there is a slave, and when there is a slave there is a master; and so likewise with the rest. Moreover, they cancel one another out: for if there is no double there is no half, and if there is no half there is no double; and so likewise with all the other cases of this sort.

But being simultaneous by nature does not seem to be true of all relatives. For the knowable would seem to be prior to knowledge. For as a rule it is of things already existing beforehand that we acquire our knowledge; in few cases, if in any at all, could one see knowledge coming into being at the same time as the knowable. Again, the knowable, when it is destroyed, cancels knowledge along with it, but knowledge does not cancel the knowable along with it. For if there is no knowable there is no knowledge, since it would then be knowledge of nothing; whereas if there is no knowledge, nothing prevents there being a knowable. Take the squaring of the circle: if indeed it is knowable, there is as yet no knowledge of it, and yet it is itself knowable. Again, if animal is destroyed there will be no knowledge, yet many of the knowables may still exist.

Matters stand similarly in the case of perception. For the perceptible seems to be prior to perception. For the perceptible, when it is destroyed, cancels perception along with it, but perception does not cancel the perceptible along with it. For perceptions are concerned with body and are in a body, and if the perceptible is destroyed the body too is destroyed, since body is one of the perceptibles; and if there is no body, perception too is destroyed. Hence the perceptible cancels perception along with it. But perception does not cancel the perceptible along with it: for if animal is destroyed, perception is destroyed, yet there will still be a perceptible, such as body, hot, sweet, bitter, and all the other things that are perceptible. Again, perception comes into being at the same time as that which is capable of perceiving, for it comes into being at the same time as the animal; but the perceptible exists even before there is an animal or perception. For fire and water and things of that sort, out of which the animal is composed, exist even before there is any animal at all, or perception; so that the perceptible would seem to be prior to perception.

There is a puzzle whether no substance is said to be a relative, as it seems to be the case, or whether this is possible for some of the secondary substances. In the case of primary substances it is true: neither the wholes nor the parts are said to be relatives. For an individual man is not said to be an individual man of something, nor is an individual ox said to be an individual ox of something. And so likewise with the parts: an individual hand is not said to be an individual hand of something, but a hand of something, and an individual head is not said to be an individual head of something, but a head of something.

It is likewise with the secondary substances, at any rate with most of them. Man, for example, is not said to be man of something, nor ox ox of something, nor wood wood of something, though wood is said to be someone's property. In such cases, then, it is evident that they are not relatives. But with some of the secondary substances there is a dispute. The head, for instance, is said to be head of something, and the hand is said to be hand of something, and so with each thing of that sort; hence these would seem to be relatives.

Now if the definition of relatives has been adequately rendered, it is either exceedingly difficult or impossible to show that no substance is said to be a relative. But if it has not been adequately rendered, and rather the relatives are those things for which to be is the same as to be somehow related to something, then perhaps something may be said in reply. The earlier definition does indeed apply to all relatives, but for them to be relatives is not the same as their being said to be just what they are of other things.

From these considerations it is clear that if someone knows any relative definitely, he will know definitely also that in relation to which it is said. And this is evident even from the facts themselves. For if someone knows of some particular thing that it is one of the relatives, and for relatives to be is the same as to be somehow related to something, then he knows also that in relation to which this thing is somehow related; for if he does not know at all that in relation to which this thing is somehow related, neither will he know whether it is somehow related to something at all.

This is clear in particular cases too. For example, if someone knows definitely of some particular thing that it is double, he thereby knows definitely also what it is double of; for if he does not know it to be double of anything definite, neither does he know at all whether it is double. Likewise, if someone knows of some particular thing that it is more beautiful, then for these same reasons he must thereby know definitely also what it is more beautiful than. He will not know indefinitely that this is more beautiful than something inferior; for such a thing turns out to be supposition, not knowledge. For he will no longer know exactly that it is more beautiful than something inferior, since it may well turn out that there is nothing inferior to it. Hence it is evident that, whatever relative anyone knows definitely, he must know definitely also that in relation to which it is said.

But as for the head and the hand and each thing of that sort, which are substances, it is possible to know definitely the thing itself, just what it is, but it is not necessary to know that in relation to which it is said. For it is not possible to know definitely whose this head is, or whose the hand is. Hence these would not be relatives. And if these are not relatives, it would be true to say that no substance is a relative.

It is perhaps hard to make confident pronouncements on such matters without having examined them many times; still, to have gone through the difficulties on each of them is not without use.