52 On the Items in the Category of Relation
On the Items in the Category of Relations for many. For one thing, the books have been translated incorrectly. For another, Greek expressions are not perspicuous when translated into Latin; and this makes for difficult reading and sometimes even invites incorrect interpretation. Finally, the remarks of the authors are frequently employed incorrectly for although what the authors say is true, those who incorrectly employ those remarks confound the points being made.
The second source consists in the tendency to multiply entities according to the multiplicity of terms, so that for every term there is a thing. This is a wrong-headed approach, and more than any other it leads one from the truth. For one should not ask in the case of every term what the relevant thing is. In the case of many terms the question is what the term means.
Examples include all relative terms and some other terms which are equivalent in signification to longer expressions. Therefore, propositions in which they play a role have to be analyzed and broken down replacing a name sometimes by a description, for words and concepts by themselves can deceive.
52: On the Items in the Category of Relation Now that we have outlined the view which Aristotle seems to hold, we shall consider the things which are subsumed on that view under the genus of relation.
First, it should be noted that for Aristotle nothing is to be placed in the genus of relation except those mental, vocal and written names to which, when functioning in a proposition, it is possible to add an expression in an oblique case.
There are, however, two different kinds of names to which an expression in an oblique case can be added. When some names are truly predicable of something, the only expression that can be added correctly is the corresponding abstract name. (I am using the term 'abstract' in the first of the senses mentioned earlier.) Expressions like 'white', 'hot', 'sweet', and 'besouled' are examples. Nothing is white except by whiteness, nor is anything hot except by heat, and similarly for the other examples. But while expressions like these are connotative, they are not said to be relatives.
But there are other names which cannot be truly predicated of anything unless it is possible to add to them names which are not their abstract forms;
On the Items in the Category of Relations the added names are in one of the oblique cases, but they are not the relevant abstract forms. Examples are names like 'master' and 'slave' and 'father' and 'son'; for no one is a father unless he is someone's father, nor is anything similar unless it is similar to something. These expressions are called relatives according to being, since they cannot hold true of something unless it is possible to add to them an expression in one of the oblique cases. When relatives according to being are contingently predicated of something and contingently hold true of that of which they are predicated (provided the subject remains the same), then it is impossible for a person to know that the term holds true of something unless he knows determinately that with respect to which the term is predicated. Thus, it is impossible to know that someone is a slave without knowing whose slave he is. Likewise, it is impossible to know that something is similar to something else without knowing that to which it is similar. The Philosopher is referring to relatives of this sort when he says in the Categories that "he who knows determinately one of the relatives also knows the other determinately."! But if a relative according to being is truly predicated of something in such a way that it cannot fail to hold true of that thing (when the subject remains the same), then he who knows determinately one of the relatives need not necessarily know the other.
There are, on the other hand, names to which it is sometimes possible to add an expression in an oblique case and sometimes not. Thus, sometimes it is possible to say truly that the donkey is Socrates', so that it is the donkey of some one. But sometimes it happens that although 'That donkey is a donkey' is true, 'That donkey is someone's' is false. Expressions like this can be called relatives according to speech. Sometimes they can be said of other things with the addition of some expression in an oblique case, but this addition is not a necessary condition of their being predicated of a thing. Consequently, these expressions do not belong per se in the genus of relation.
Names like 'hand' and 'head' are examples. Thus, sometimes that hand is a man's hand; but sometimes it is not any man's hand as when it has been cut off. However, if in that case it were not called a hand, then the term 'hand' would be in the genus of relation. The same holds in other cases. Thus, every name which cannot, when significatively taken, be predicated of anything unless it is possible to add to it the oblique form of some expression other than its abstract counterpart, is in the genus of relation. All such expressions are called relatives according to being. Some expressions, on the other hand, can be called relatives according to speech since sometimes one can and sometimes one cannot add to them an expression in an oblique case.
Nonetheless, the Philosopher does not use the expressions 'relative according to being' and 'relative according to speech', nor does he use another set
On the Items in the Category of Relations of expressions which teachers of philosophy nowadays employ frequently— 'real relation' and 'relation of reason'. No such distinction is to be found in the Philosopher's writings. To him names like 'cause' and 'lord' are relatives in the same way as all the other expressions. He always, or at least frequently, uses the name 'master' as his example of relatives; for no one can be a master unless he is the master of some slave. Therefore, for Aristotle there is no distinction between real relations and relations of the reason. He does distinguish several forms of relatives in the fifth book of the Metaphysics," but I will not deal with that distinction at this time. It suffices to know that for Aristotle every name or expression having the force of a name (participles then are included) which cannot, when significatively taken, be truly predicated of anything unless it is possible to add to it the oblique form of an expression other than its abstract counterpart, is really a relative and falls under the category of relation. It is irrelevant whether the term signifies just substance or quality or both. It is likewise irrelevant whether it connotes something else, whether in act or potency, whether affirmatively or negatively, or in some other way.
Thus, all of the following expressions are to be placed in the category of relation—'master' and 'slave', 'sign' and 'significatum', 'cause' and 'effect',
'producer' and 'product', 'agent' and 'patient', 'what is able to heat', and 'what is able to be heated'.
But it follows that in Aristotle's view, while a verb may be in the category of action or passion or some other category, the participle or verbal name corresponding to it may be in the category of relation. He implies this view in the fifth book of the Metaphysics in the chapter on Relation while enumerating the forms of relatives, "Other things are said to be relative—as that which can heat to that which can be heated and that which can cut to that which can be cut and, in general, the active to the passive." He says later, "Things are active and passive according to an active and passive potency and the actualization of such a potency—that which is capable of heating is related to that which is heatable because it can heat. Again, that which heats is related to that which it heats and that which cuts to that which is cut as agent to patient.'* But note, that which heats is an example of the second type of relatives. Thus, I think that for Aristotle all names and participles which cannot truly be predicable of something unless one can add to them an expression in an oblique case and which correspond in signification to active and passive verbs, being active and passive not only in form but also in signification (in that they signify that something really acts oris acted on or that something does something or has something done to it), are relatives of the second sort. Examples are 'agent' and 'patient', 'cause' and 'effect', 'active' and 'passive', etc.
On the Property of Relatives Furthermore, that Aristotle subsumes verbal names or participles under the category of relation and their corresponding verbs in other categories is clear from his remarks in the chapter on Relatives in the Categories. He says,
"Likewise, lying, standing, and sitting are also positions, but position is a relative. To lie, to stand, and to sit are not themselves positions, but take their names from the positions enumerated."* It is clear from this remark that 'to stand' is not in the genus of relation but that 'standing' is; for if there be a standing it is the standing of someone. Nevertheless, in the final analysis it is necessary to resolve a proposition embodying this sort of expression into a proposition of the following form: 'The parts of this body are separated from each other in this or that way'. Here, a name ora participle of the relevant sort (i.e., one that cannot be predicated of something unless it is possible to add some other expression in an oblique case) will make itself explicit.
53: On the Properties of Relatives Now that we have examined Aristotle's view on relations, we shall consider the properties he attributed to them.
In the case of some relatives there is contrariety. This property, however, does not hold in the case of all relatives. But it is clear that contrariety is exhibited by some relatives; for 'virtue' and 'vice' are relatives, and they are contraries. It should be noted that those relatives are called contraries which cannot simultaneously apply to one and the same thing. Thus, if relatives are contraries, then only one of a pair can apply toa particular individual at a given time; the other can apply only at a different time. In this sense
'father' and 'son' are not contraries, for one and the same person can be simultaneously both father and son. He cannot of course be both with respect to the same person. Likewise, nothing is the contrary of 'triple' because to anything that 'triple' applies some other relative can also apply at the same time. If it cannot apply at the same time, it never can.
The second property of relatives is that they admit of more or less; but this does not apply in the case of all relatives. Some relatives can be predicated of something, first with the adverb 'more' and, later with the adverb
'less'. Thus, something is first more similar and later less similar and vice 56> 11-14, On the Property of Relatives versa. But, this cannot happen with 'triple', 'double', 'equal', or a number of other relatives.
The third property is that all relatives are said with respect to some correlative; that is, to every relative one can add an oblique form of its own correlative. It may happen that we lack the requisite name. In that case we can invent one; and then the oblique case of the existing name can be added to the nominative case of the invented name and, likewise, the oblique case of the invented name to the nominative case of the existing name. Thus, if some one is a master it is necessary that he be the master of a slave, and if something is similar that it be similar to what is similar. Likewise, if 'wing' is a relative, so that it is impossible for something to be a wing without being the wing of something, then we can invent a name to correspond to it, to function as its correlative. Thus, we might say that a wing is the wing of the winged and that the winged is winged by way of a wing. The same holds in all other cases.
But it should be noted that sometimes one and the same name is placed in both the oblique and nominative cases. These are called relatives of similar names or equivalent relations. Thus, 'Every similar thing is similar to the similar' and 'Every equal thing is equal to the equal' and similarly in other cases where the same name is used in both the nominative and an oblique case. But sometimes a different name is placed in the nominative and oblique case. Thus, if someone is a father, he is the father of a son; and it is not necessary that he be the father of someone who is also a father. Likewise, if someone is a slave, he is the slave of a master; and it is not necessary that he be the slave of a slave. These are called relatives of different names or nonequivalent relatives. The general rule, however, is that it is always possible to add an oblique form to the nominative case of a relative, as Damascene indicates in chapter 34 of his Logic."
The fourth property of relatives is that they are simultaneous by nature.
This should be understood to mean that existence cannot be truly predicated of one relative unless, when the relevant proposition is formed it is also truly predicated of the other. Thus. if 'The double is' is true, then 'The half is' is also true. If 'The father is' is true, then 'The son is' is also true and vice versa. The claim that relatives are destroyed together should be understood in the same way. If the negative proposition in which existence is denied of one relative is true, the corresponding negative proposition in which it is denied of the other is also true. Thus, if 'The father is not' is true, then
'The son is not' is also true and vice versa.
But it should be noted that according to the Philosopher in the Categories, these rules are not absolutely general. Thus, he shows that this particular property does not apply in the case of all relatives.? He proves this in the ' Dialectica, chapter 34, 3-18.