The transposition of substantive and adjective makes no difference in the meaning of the phrase; Est albus homo is equivalent to Est homo albus. If it were not equivalent, there would be two negations corresponding to the same affirmation; but we have shown that there can be only one negation corresponding to one affirmation, so as to make up an Antiphasis.[31]
[31] Ibid. b. 1-12. That ἐστὶ λευκὸς ἄνθρωπος, and ἐστὶν ἄνθρωπος λευκός, mean exactly the same, neither more nor less — we might have supposed that Aristotle would have asserted without any proof; that he would have been content ἀπὸ τῶν πραγμάτων πιστοῦσθαι (to use the phrase of Ammonius in a portion of the Scholia, p. 121, a. 27). But he prefers to deduce it as a corollary from a general doctrine much less evident than the statement itself; and after all, his deduction is not conclusive, as Waitz has already remarked (ad Organ. I. p. 351).
In one and the same proposition, it is indispensable that the subject be one and the predicate one; if not, the proposition will not be one, but two or more. Both the subject and the predicate indeed may consist of several words; but in each case the several words must coalesce to make one total unity; otherwise the proposition will not be one. Thus, we may predicate of man — animal, bipes, mansuetum; but these three coalesce into one, so that the proposition will be a single one. On the other hand the three terms homo, albus, ambulans, do not coalesce into one; and therefore, if we predicate all respecting the same subject, or if we affirm the same predicate respecting all three, expressing them all by one word, the proposition will not be one, but several.[32]
[32] Aristot. De Interpr. p. 20, b. 13-22.
Aristotle follows this up by a remark interesting to note, because we see how much his generalities were intended to bear upon the actual practice of his day, in regard to dialectical disputation. In dialectic exercise, the respondent undertook to defend a thesis, so as to avoid inconsistency between one answer and another, against any questions which might be put by the opponent. Both the form of the questions, and the form of the answers, were determined beforehand. No question was admissible which tended to elicit information or a positive declaration from the respondent. A proposition was tendered to him, and he was required to announce whether he affirmed or denied it. The question might be put in either one of two ways: either by the affirmative alone, or by putting both the affirmative and the negative; either in the form, Is Rhetoric estimable? or in the form, Is Rhetoric estimable or not? To the first form the respondent answered Yes or No: to the second form, he replied by repeating either the affirmative or the negative, as he preferred. But it was not allowable to ask him, What is Rhetoric? so as to put him under the necessity of enunciating an explanation of his own.[33]
[33] See the Scholia of Ammonius, p. 127, Br.
Under these canons of dialectic debate, each question was required to be really and truly one, so as to admit of a definite answer in one word. The questioner was either unfair or unskilful, if he wrapped up two questions really distinct in the same word, and thus compelled the respondent either to admit them both, or to deny them both, at once. Against this inconvenience Aristotle seeks to guard, by explaining what are the conditions under which one and the same word does in fact include more than one question. He had before brought to view the case of an equivocal term, which involves such duplication: if himation means both horse and man, it will often happen that questions respecting himation cannot be truly answered either by Yes or No. He now brings to view a different case in which the like ambiguity is involved. To constitute one proposition, it is essential both that the subject should be one, and that the predicate should be one; either of them indeed may be called by two or three names, but these names must coalesce into one. Thus, animal, bipes, mansuetum, coalesce into homo, and may be employed either as one subject or as one predicate; but homo, albus, ambulans, do not coalesce into one; so that if we say, Kallias est homo, albus, ambulans, the proposition is not one but three.[34] Accordingly, the respondent cannot make one answer to a question thus complicated. We thus find Aristotle laying down principles — and probably no one had ever attempted to do so before him — for the correct management of that dialectical debate which he analyses so copiously in the Topica.
[34] Aristot. De Interpret. p. 20, b. 2. seq.; Ammonius, Schol. pp. 127-128, a. 21, Br. Compare De Sophist. Elench. p. 169, a. 6-15.
There are cases (he proceeds to state) in which two predicates may be truly affirmed, taken separately, respecting a given subject, but in which they cannot be truly affirmed, taken together.[35] Kallias is a currier, Kallias is good — both these propositions may be true; yet the proposition, Kallias is a good currier, may not be true. The two predicates are both of them accidental co-inhering in the same individual; but do not fuse themselves into one. So, too, we may truly say, Homer is a poet; but we cannot truly say, Homer is.[36] We see by this last remark,[37] how distinctly Aristotle assigned a double meaning to est: first, per se, as meaning existence; next, relatively, as performing the function of copula in predication. He tells us, in reply either to Plato or to some other contemporaries, that though we may truly say, Non-Ens est opinabile, we cannot truly say Non-Ens est, because the real meaning of the first of these propositions is, Non-Ens est opinabile non esse.[38]
[35] Aristot. De Interpr. p. 21, a. 7, seq.