Question Nine
[QUESTION NINE
Whether this Rule, "When One Thing is Predicated of Another,"
Etc., is True]
t is asked whether the first rule of Aristotle1 is true,
namely, "when one thing is predicated of another,
whatever is predicated of the predicate, is predicated of the subject."2
1. It seems that it is false:
Since "animal" is predicated of man, and "white" <is predicated> of animal, therefore "white" <is predicated> of man; therefore, according to this rule this implication would hold,
"an animal is white; therefore, a man is white," where it seems to be a fallacy of the consequent. Similarly, the antecedent can be true without the consequent <being true>.3
2. Second, according to this rule, it would follow that "sub1. Categories, Ch. 3 (1b 10–15).
2. In order to understand the significance of this question, it should be noted that the phrase "when one thing is predicated of another" implies that what is predicated has a necessary, essential relation to the subject that is not captured by the simple categorical structure "s is p."
3. This initial argument is a classic example of the syllogism known as Barbara, i.e., a universal affirmative syllogism in the first figure. In its strongest form, the argument would look like this: "All animals are white; all men are animals; therefore, all men are white." The dubious implication that Scotus refers to here, however, is a conditional argument in this form: "If a man is white, then an animal is white; but an animal is white; therefore, a man is white," which is an example of affirming the consequent, for it is possible that there are white animals without there being any white men.
QUESTION 9
stance is the highest genus; therefore, an animal is the highest genus."4
3. Third, if something is predicated of a predicate, it is predicated of the subject; but "predicate" is predicated of a predicate, since a predicate is a predicate; therefore, "predicate" is predicated of a subject; therefore, a subject is a predicate.
4. To the opposite is Aristotle.
[I. The Status Of The Question:] [A. The Opinion of Others] 5. It is said5 that the rule is true: (A) when three essentially ordered <terms> are taken in the same genus: whatever is predicated essentially (as "to be predicated" is taken here) of the predicate, is predicated essentially of the subject.—With this
<condition there is also a second qualification, namely> (B) the rule must also be understood when an intermediate predicate is not changed in comparison to its predicate and to the subject of which it is predicated.
6. According to the first condition, the first two arguments6 are solved: since "white" and "highest genus" are not predicated of animal and of substance essentially.—Similarly, the second
<argument>7 is solved according to the <second rule, i.e., since there is a> variation of the intermediate predicate, since by arguing in this way, as <it is argued> there,8 there is a fallacy of accident.
7. To the third <argument>,9 it is said10 that (C) the rule ought to be understood of "to be predicated" <as> signified in 4. The missing premise in this enthymeme is as follows: substance is predicated of animal, or an animal is a substance.
5. Cf. Albertus Magnus, Praedic., Tr. 1 c. 6 (ed. Borgnet, I 161b).
6. Cf. supra, nn. 1–2.
7. Cf. supra, n. 2.
8. Namely, "substance is the most general [genus]; animal is a substance; therefore, animal is the most general [genus]."
9. Cf. supra, n. 3.
10. The identity of those who held this position is uncertain.
JOHN DUNS SCOTUS
the <second> intentions; in the bases <i.e., in the first intentions, it ought to be understood> of "to be predicated" as it is exercised.11 It is argued, however, in the <second> intentions with <the linking verb> "is," through which the predication is effected; therefore, it does not hold.12
11. According to Nuchelmans, "the application of the distinction exercitus/ significatus to the act of predication" was more widely diffused by Scotus. "In that use it seems to have been made especially popular by John Duns Scotus, who connected the distinction between a praedicatio exercita—for instance Homo est animal—and a praedicatio significata—for example Genus praedicatur de specie— with the distinction between two levels of conceiving of which first-order and second-order concepts (intentiones primae and intentiones secundae) are characteristic"; Gabriel Nuchelmans, "The Distinction Actus Exercitus/Actus Significatus,"
74. See also I. Rosier, "La Distinction Entre Actus Exercitus et Actus Significatus
Dans Les Sophismes Grammaticaux du MS BN Lat. 16618 et Autres Textes Apparentes," in Sophisms in Medieval Logic and Grammar. Acts of the Ninth European
Symposium for Medieval Logic and Semantics, ed. S. Read (Boston: Kluwer Academic Publishers, 1993), and Scotus's Super Porphyrii, q. 8, 16. Ockham makes the following observation about this distinction: An effected act [actus exercitus] is one designated by "is" or some other expression which not only signifies that something is predicated of something else but actually effects the predication of one thing of another. Examples are: "Man is an animal," "Man disputes," and "Man runs." A signified act [actus significatus or signatus], on the other hand, is one that is designated by a verb like "to be predicated of," "to have as subject,"
"to be affirmed of," and "to belong to," all of which signify one and the same thing. For example, when one says, "'Animal' is predicated of 'man,'" animal is not actually being predicated of man; for in that proposition "animal" is functioning as subject and not predicate. The act, then, is signified. It is not the same thing to say, "'Animal' is predicated of 'man,'" and to say, "Man is an animal." The one is structurally more complex than the other. Likewise, it is not the same thing to say that a genus is predicated of the general term "man" and to say that the common term "man" is a genus. Further, saying that a genus is predicated of its species or that the word "animal" is predicated of the word
"man" is very different from saying, "Species is a genus," or, "The word 'man' is the word 'animal.'" The first two are true, and the second false. See William
Ockham, Ockham's Theory of Terms, trans. Loux, 194–95.
12. In the third argument, the questionable premise is that "a predicate is a predicate." According to the opinion of others, the problem with this premise is that the word "predicate" is a second intention and yet the sentence is a simple declarative sentence using the linking verb "is" to combine a subject and a predicate. According to the rule prescribed, however, whenever a second intention, such as the word "predicate," is used, it should be used in such a construction where the term "predicate" is not actually a predicate of a sentence but rather
QUESTION 9
[B. Against the Opinion of Others] 8. Against this <position>: according to all these conditions, this implication "an animal is a substance; therefore, a man is a substance," would be valid, since here three essentially ordered <terms> are taken in the same genus, and the intermediate predicate is not changed, and it is argued in the first intentions of an act of predication. But this implication is not valid;13 therefore, all the aforesaid conditions do not preserve the truth of the rule. Proof of the minor, since it follows "a donkey is a substance; therefore, an animal is a substance," and it does not follow that "therefore a man is a substance"; therefore, this antecedent "a man is a substance" is not convertible to this "an animal is a substance."14 Therefore, by arguing from the converse, there is a fallacy of the consequent from the position of the consequent, and from the proposition having more causes of truth to one of these.
9. Second, no implication holds except in virtue of a syllogism; but if this <rule> were to be reduced to a syllogism, since
"man" and "substance" comprise the conclusion, they will be the extreme <terms> and "animal" <will be> the middle <term>.
Therefore, by making the previous antecedent into a major <premise> now, it will be a syllogism in the first figure having a particular major <premise>, which is not valid with any terms; since the form of arguing in all these is the same, therefore neither was the prior implication valid.
10. Third, there is the same form of arguing in essential predicates and non-essential <predicates>, since an implication holds based upon the signification of the propositions, just as only signifies a predicate. As it is, then, the questionable premise violates this rule by using a second intention in a simple declarative sentence.
13. According to the subsequent paragraph, this enthymeme is not valid because if it were reduced to a syllogism, the proposition "an animal is a substance" would be a particular major premise in the first figure. All first-figure syllogisms, however, must have a universal major premise in order for the syllogism to be valid.
14. In other words, one could argue that the reason that an animal is a substance is because a donkey is a substance, in which case it would not be necessary to posit that a man is a substance.
JOHN DUNS SCOTUS
a syllogism <holds based upon the signification of the propositions>; however, according to this <proposition> "an animal is a substance," a greater identity of the extremes is not signified than through this <proposition>, "an animal is white"; therefore, based upon the signification of the sentence, one predicate is not more signified to be the same as man than another.
11. If it is said that <the linking verb> "to be" is a union of the extremes, and so it expresses such a union as is required by the extremes, thus also a greater identity is signified here <in the proposition> "animal is substance" than here <in the proposition> "animal is white":15
12. This does not seem to be true, since then every proposition of <something> inhering <in something> absolutely would signify that it is such, and every proposition of <something> inhering <in something> right now, would signify that it is of <something> inhering <in something> right now. And by expressing it in this way, it would be a vain repetition, as here
"man, as <considered> right now, is white"; and there would be a repugnance to the understanding <of this proposition> here
"man is white absolutely or necessarily"; which does not seem probable. Similarly, here there would be a vain repetition "man necessarily is an animal" or "<man> is always an animal"; and here <it is> a repugnance to the understanding "man, as <considered> right now, is an animal." And if this <is the case>, then the opposite one would diminish the other. And thus in the implication, "a man, as <considered> right now, is an animal; therefore, a man is an animal" there would be a fallacy in a certain respect and absolutely (fallacia secundum quid et simpliciter).
And similarly here, "an animal necessarily is white; therefore, an animal is white," and so from that <argument> regarding necessity there would not follow that regarding <something> inhering in all these terms. That the implication is not valid,
"an animal necessarily is white; therefore, an animal is white,"
15. The alleged reason why one proposition has a greater unity than the other is that the predicate "substance" is essential to an animal, while the predicate "white" is accidental to an animal.
QUESTION 9
is evident according to Aristotle in Bk. II of On Interpretation,16 since whenever an opposition is in an addition, it does not hold
<when one reasons> from the conjoined to the divided <terms>.
Similarly if it is valid, then the consequent does not posit an understanding opposite to the antecedent.
13. On account of <these> three principal reasons,17 one can say that it does not follow formally that "an animal is a substance; therefore, a man is a substance," just as it does not follow
<that> "a substance is white; therefore, an animal <is white>," since from the signification of speech the antecedent on both sides is related to the consequent in the same way.
[II. The Response To The Question] 14. But to the question, then, it must be said that this rule is true, since it is given essentially (per se) of second intentions, as the logician ought to be speaking, and by taking "to be predicated" properly, which is "to be said before another." Then only
<this much> and not more is signified through the rule:18 "what is prior to the prior in genus is prior to the posterior" or "what is superior to a superior is superior to the inferior," of whose truth there is no doubt. And in this way the rule is suitably posited, since its truth is known to all.19
15. If, however, it is said that Aristotle exemplifies <the rule>20 from things of first intention, then in what way must it be argued in <i.e., applied to> these <first intentions> according to this rule?
I say that Aristotle exemplifies <the rule> in this way: "just as 'man' is predicated of a certain man, and 'animal' <is predicated> of man; therefore, <animal is> also <predicated> of a cer16. On Interpretation, Ch. 11, 21a (20–24).
17. Cf. supra, nn. 1–3.
18. Cf. Categories, Ch. 12 (14a 26–29); cf. also Duns Scotus, Metaph., Bk. I, q. 6, n. 16, and Bk. V, q. 1, n. 5; Ordinatio, Bk. I d. 19 q. 2, n. 33 (V 281–282).
19. See Pini, Categories and Logic in Duns Scotus, 184.
20. Categories, Ch. 3 (1b 12–15).
JOHN DUNS SCOTUS
tain man," where animal is signified to be essentially predicated of man.21
16. If, however, such a predication has to be effected with this verb "is," it is necessary, for arguing according to the rule, to modify the composition through something <i.e., through some modal qualification> that indicates that the predicate is to be predicated of the subject, just as "to be predicated" is taken here. And then in this way it must be argued that "man essentially (per se) in the first mode is animal; a certain man essentially (per se) in the first mode or essentially (essentialiter) is man; therefore, a certain man essentially (per se) is an animal."
17. If, however, it is argued absolutely that "an animal is a substance; therefore, a man <is a substance>," nothing is expressed in the antecedent that indicates <or observes> the virtue of this rule, since through "is," absolutely <taken>, the predicate "to be predicated" is not signified, that is, "is" (esse) prior to the subject or superior to it.
18. If, however, it is said that Aristotle takes the composition absolutely,22 "for a certain man is both a man and an animal":
I say that he understands it "essentially" (per se),23 since he thus expresses the prior in the second intentions:24 "animal is predicated of man," whence the following <premise> must be understood according to the preceding <premise>.25
21. One should notice, especially in contrast to the following paragraph, that even though the argument is exemplified by terms of first intention, the structure of the argument is not that of a simple categorical form, but rather is in the form of "a is predicated of b," where the term "a" signifies a predicate but does not function as a predicate.
22. Categories, Ch. 3 (1b 15).
23. For the four modes of per se, cf. Posterior Analytics, Bk. I, Ch. 4 (73a 34– b 24). Cf. also q. 7, n. 13.
24. Categories, Ch. 3 (1b 12–14).
25. To clarify this argument, I have recast the initial syllogism as follows:
Animal is predicated of man (the prior or the preceding premise).
A certain man is a man (the following premise, which must be understood in the same way as the preceding).
Therefore, a certain man is an animal [or animal is predicated of a certain man].
QUESTION 9
[III. To The Principal Arguments] 19. To the <first> two arguments26 against the rule, <the solution> is evident.
20. To the third <argument>27 I say that in respect to a subject <i.e., taken as a subject of a sentence> a predicate is predicated only by a signified predication; but in respect to a predicate <i.e., as a predicate of a sentence> a predicate is predicated only by an actual <or effected> predication. The rule, however, must be understood for each predication taken uniformly.
21. If it is argued: it holds that "animal is predicated of man" where there is a signified predication, and "a certain man essentially (per se) is man" where there is an actual predication; therefore, "a certain man essentially (per se) is an animal":
I say that this <proposition>, "animal is predicated of man," is accidental (per accidens); nevertheless, "animal" is not predicated <of man> accidentally (secundum accidens); and this is true for a predication <which is> actual <or effected>. And thus on both sides there is truth, by taking the actual predication uniformly.
26. Cf. supra, nn. 1–2.
27. Cf. supra, n. 3.