76 On the Supposition of Relatives
On the Supposition of Relatives sive proposition; and because the subjects of the affirmative and negative exponents differ from the subject of the exclusive proposition, that subject can have one of the three forms of supposition. In 'Socrates begins to be literate', 'Socrates ceases to be white', and 'Socrates was twice black', however, one and the same term is subject in the exponential propositions and in the proposition whose exponents they are.
76: On the Supposition of Relatives Now that we have examined the supposition of absolute terms we shall examine the supposition of relatives. Here I am using the term 'relative' as it is used by grammarians rather than logicians. In this sense a relative is an expression which refers to an antecedent.
First it should be noted that as grammarians use the term there are relatives of substance and relatives of accident. 'He', 'that', and 'same' are relatives of substance. Those expressions are called relatives of accident which are in some way taken or derived from accidents. Examples are 'such' 'of such a kind' and 'so many'. Among relatives of substance there are relatives of identity and relatives of diversity. Of relatives of identity some are reciprocal and some are not.
Non-reciprocal relatives include expressions like 'he' and 'same'. A number of rules hold as regards these relatives. First, they always supposit for that for which their antecedent supposits, so that if they hold true of anything, they hold true of the same thing as their antecedent. This is clear in 'Socrates runs and he disputes'. For the truth of this conjunction it is required that the second part hold true of the same thing as the first part. Likewise in the case of 'Man is a species and it is predicated of many'.
It should be noted that a relative of this sort should never be placed in the same categorical proposition as its antecedent. Thus, in 'Socrates is he', the word 'he' is a demonstrative and not a relative pronoun.
Likewise, it should be noted that when the antecedent ofa relative is a common term with personal supposition, one can never, by substituting the antecedent for the relative, generate a proposition which is convertible with and equivalent to the original proposition. Thus, 'A man runs and he disputes' and 'A man runs and a man disputes' are not equivalent. But in other cases one can do this for 'Socrates runs and he disputes' and 'Socrates runs and Socrates disputes' are equivalent.
On the Supposition of Relatives Likewise, it should be noted that a negation never makes a relative stand confusedly and distributively. On the contrary a relative always supposits for just that thing of which its antecedent holds true or is asserted to hold true. Thus, while 'Some man is Plato and Socrates is not he' is true, one cannot infer from this that Socrates is not a man. All that is required is that 'Socrates is not Plato' be true. Thus, the following two propositions can, strictly speaking, hold true simultaneously: 'Some man runs and Socrates is not he' and 'Some man runs and Socrates is he', for if Socrates and Plato are both running both conjunctions are true.
Concerning reciprocal relatives of identity it should be noted that they differ from other relatives in that they can be placed indifferently in either different or the same categorical propositions as their antecedents. This is clear in the case of 'himself' and 'his', for 'Socrates disputes and sees himself? and 'Socrates sees himself' are both well formed. So are the following:
'Socrates sees his donkey' and 'Socrates runs and his donkey walks'.
It should be noted that sometimes a relative is a part of an extreme and sometimes it is an extreme. When it is an extreme, so that it immediately precedes or follows the verb, it supposits for that for which its antecedent supposits. Thus, 'Socrates sees himself? and 'Every man sees himself'. But when it is a part of an extreme, it does not supposit for that for which its antecedent supposits. It supposits, on the contrary, for something that is designated by that to which it is added. This is clear in 'Socrates disputes and his donkey runs'. Here the word 'his' does not supposit for Socrates; it supposits for Socrates' donkey rather than for some other donkey.
It should also be noted that a relative of this sort has the same kind of supposition and supposits for the same things as its antecedent. However, when its antecedent supposits either confusedly and distributively or determinately, it has a similar form of supposition but exhibits this singularly—by referring particulars to particulars. Therefore, it is not possible to descend either conjunctively or disjunctively or in any way other than with respect to something contained under the antecedent. For example, in 'Every man sees himself', the word 'himself supposits for every man by means of confused and distributive mobile supposition; but it does this singularly since it is not possible to descend without altering the other extreme. It does not follow that if every man sees himself every man sees Socrates. Nonetheless, it is possible here to descend to Socrates with respect to Socrates. Thus,
'Every man sees himself; therefore, Socrates sees Socrates'. This is not, however, the case with a proposition like 'Man is an animal'. Likewise, in 'A man sees himself? the word 'himself' supposits determinately yet singularly, for it is possible to make the descent only in the following way: a man sees himself; therefore, Socrates sees Socrates or Plato sees Plato or...' (and so on for all the relevant particulars). It is also possible to ascend, but not On the Supposition of Relatives in the following way: a man sees Plato; therefore a man sees himself. The ascent operates as follows: Socrates sees Socrates; therefore, a man sees himself.
It is clear then that when a relative of this sort mediately follows the sign of universality it has confused and distributive supposition, yet only singularly. Likewise, a relative in a categorical proposition (whether reciprocal or not) has confused and distributive supposition as a result of the addition of the sign of universality to its antecedent. In the same way a term of this sort has merely confused supposition, even if there is no sign of universality in the relevant categorical proposition, and provided that in the preceding categorical the sign of universality mediately precedes its antecedent. This is clear in 'Every man is an animal and every donkey sees him'.
Concerning relatives of diversity it should be noted that an expression is called a relative of diversity because it is not truly predicated of the same thing as its antecedent. This is clear in the following example where the reference is to two contradictory propositions: 'One of them is true and the other false'. Here 'other' is truly predicated of something of which its antecedent in 'One of them is true' is not truly predicable.
Relatives like 'so much', 'such', and 'how much' are not, in Aristotle's view, called relatives of accident because they supposit for accidents, but because they supposit for something while connoting something predicable in a way other than in quid. It should be noted that a relative of this sort neither supposits for nor holds true of that for which its antecedent holds true. It supposits for something other than but similar to or equal to that for which its antecedent supposits. This is clear in 'Socrates is white and Plato is such'. Here the word 'such' does not supposit for Socrates but for something similar to Socrates: likewise, in 'Socrates and Plato run and they dispute as much'. The word 'as much' does not supposit necessarily for Socrates and Plato nor does it supposit necessarily for the things for which 'run' supposits. It can supposit for other things. The same is true in the case of 'Socrates is six feet tall and Plato is as much.' Thus, it should be noted that while a relative of this sort can supposit for the same thing, this is not necessary. Likewise it should be noted that the antecedent of this sort of relative is frequently or always a name in the genus of quantity or quality or in some other accidental category.