Phronesis Volume 24 Issue 3 1979 [Doi 10.2307%2F4182075] William Jacobs -- Aristotle and Nonreferring Subjects

Aristotle and Nonreferring SubjectsAuthor(s): William JacobsSource: Phronesis, Vol. 24, No. 3 (1979), pp. 282-300Published by: BRILLStable URL: http://www.jstor.org/stable/4182075 .

Accessed: 01/09/2013 20:55

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

BRILL is collaborating with JSTOR to digitize, preserve and extend access to Phronesis.

http://www.jstor.org

This content downloaded from 128.135.12.127 on Sun, 1 Sep 2013 20:55:55 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=bap

http://www.jstor.org/stable/4182075?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


Aristotle and Nonreferring Subjects

WILLIAM JACOBS

It is a widely accepted view amongst scholars that Aristotle believed that the subject of an assertion might fail to refer. Two texts, De Interpretatione xi 21 a 25-28 and Categories x 13 b 12-35, are generally cited as evidence for this belief. In this paper I will argue that both passages have previously been misunderstood and that Aristotle did not accept the possible referential failure of the subject of an assertion. In Section I, after first discussing the standard interpretations of both texts, I note the difficulties which result from these accounts. In Section II I offer a brief general argument showing that Aristotle's own account of what an assertion is implies that it is impossible for the subject of an assertion to fail to refer. In Section III I present my own analysis of each passage and show that when properly understood neither is in any way concerned with the problem of referential failure.

I

The two passages with which we are presently concerned come from two unrelated discussions. The first, De Interpretatione xi 21 a 25-28, is a tangential comment made after a discussion of certain types of predications. After worrying first about determining the complexity of an assertion, Aristotle then shifts the discussion to the circumstances under which it is permissible to move from asserting two predicates separately of a subject to asserting them together of the subject. Near the end of this discussion, in 21 a 25-28, Aristotle remarks (according to John Ackrill's translation):

For example, Homer is something (say, a poet). Does it follow that he is? No, for the 'is' is predicated accidentally of Homer; for it is because he is a poet, not in its own right, that the 'is' is predicated of Homer.'

When rendered in this way, it is easy to see why the standard interpretation of this passage is in terms of Aristotle's querying whether or not "Homer exists" follows from "Homer is a poet."

The second passage with which we are concerned, Categories x 13 b 12-35, forms the concluding section of a discussion of the types of

282



opposition.2 After briefly listing the four types of opposites, Categories x proceeds to discuss each successively: the opposition of relatives (e.g. double and half), the opposition of contraries (e.g. odd and even or white and black), the opposition of possession and privation (e.g. blindness and sight), and finally the opposition of things that are opposed as affirmation and denial (e.g. "Socrates is seated" and "Socrates is not seated"). The distinguishing feature of this last sort of opposition is that only in this kind of opposition must one opposite be true and the other be false. Though, in part, this criterion is trivially true since only this fourth sort of opposition involves combined terms, Aristotle goes further and adds that not even when contraries or possessions and their privations are combined with subjects in order to form assertions are these resultant assertions so opposed that always one is true and the other is false.

Our concern is with Aristotle's explanation of this concluding point. For a typical translation of these remarks, let me again cite Ackrill's rendering:

It might, indeed, very well seem that the same sort of thing does occur in the case of contraries said with combination, 'Socrates is well' being contrary to 'Socrates is sick'. Yet not even with these is it necessary always for one to be true and the other false. For if Socrates exists one will be true and one false, but if he does not both will be false; neither 'Socrates is sick' nor 'Socrates is well' will be true if Socrates himself does not exist at all. As for possession and privation, if he does not exist at all neither is true, while not always one or the other is true if he does. For 'Socrates has sight' is opposed to 'Socrates is blind' as possession to privation; and if he exists it is not necessary for one or the other to be true or false (since until the time when it is natural for him to have it both are false), while if Socrates does not exist at all then again both are false, both 'he has sight' and 'he is blind'. But with an affirmation and negation one will always be false and the other true whether he exists or not. For take 'Socrates is sick' and 'Socrates is not sick': if he exists it is clear that one or the other of them will be true or false, and equally if he does not; for if he does not exist 'he is sick' is false but 'he is not sick' true. Thus it would be distinctive of these alone - opposed affirmations and negations - that always one or the other is true or false.3

The commonly accepted interpretation of Aristotle's remarks is given in terms of nonreferring subjects.4 Thus, when he writes in 13 b 12-19 that if A and B are contraries, predicating A and B of some individual (e.g. Socrates) need not result in two assertions one of which is true and the other of which is false, Aristotle's explanation is said to be that should Socrates not exist then both assertions - "Socrates is A" and "Socrates is B" - are false due to the subject's failure to refer. Similarly, when Aristotle notes in 13 b 20 -27 that if C and D are a possession and its privation, then predicating C and D of some individual (e.g. Socrates) need not result in one true and one

283



false assertion, Aristotle's point is taken to be that should Socrates not exist, then both assertions are false. Only in the case of contradictories which are opposed as affirmation to denial - "Socrates is E" and "Socrates is not E" - need always one be true and the other be false. On the standard account this is because (1) if Socrates exists, then "Socrates is E" and "Socrates is not E" will be true or false depending upon whether Socrates is or is not E while (2) if Socrates does not exist, then since that which is not is not anything, Socrates is not A and .. . and Socrates is not E and ... and therefore being E cannot and not being E can be truly aflirmed of Socrates.5

The proponents of these two interpretations have themselves recognized the fatal difficulty with which their readings are confronted - namely that on their understanding the two texts apparently contradict one another, for whereas Categories x 13 b 12-35 is taken to imply that if Socrates does not exist, then "Socrates is a poet" would be false, De Interpretatione xi 21 a 25-28 seems to deny this implication. To the best of my knowledge, no adequate resolution of this problem has been proposed.6 However, in addition to this difficulty, there are other objections to the traditional interpretation of these texts. In the case of De Interpretatione xi 21 a 25-28, the standard analysis of Aristotle's remark appears to force us to understand the two crucial terms of the passage in a way that conflicts with their normal senses.7 In the case of Categories x 13 b 12-35, it is unclear why Aristotle should suddenly introduce the notion of referential failure without even the briefest explanation. Even for Aristotle it is a bit peculiar to introduce something so problematic without prior explanation.

All of these considerations lead us to suspect the adequacy of the standard account of these two remarks and cause us to seek alternative ex- planations which do not involve the notion of referential failure. However, before offering such analysis, I first wish to present one final, and perhaps the most important, argument against reading either De Interpretatione xi 21 a 25-28 and Categories x 13 b 12-35 or any other Aristotelian text in terms of nonreferring subjects - namely that the adoption of such a position would contradict Aristotle's account of what it is for an assertion to be capable of being true or false.

II

While previous interpreters have quite readily noted how on their reading Aristotle's remarks in Categories x conflict with his comments in De Inter- pretatione xi, they have overlooked how Aristotle could not accept the

284



possible referential failure of the subject of an assertion without contradicting his account of affirmation and denial in De Interpretatione v, vi, and x.

According to Aristotle, those things which are true or false are not propositions or statements (where these two terms have their modern technical senses) but rather sentences. On his account, whereas every sentence is meaningful, only certain sentences are either true or false. Many, such as those in prayers, are neither true nor false and hence are fit objects of rhetorical or poetic study only. Other sentences, by virtue of their being either true or false, make assertions about the world. Aristotle calls such literal fact-asserting sentences TOqX(OVTLXOL Xo-yoi ("assertions"; liter- ally: "assertive sentences").

Aristotle's discussion of assertions makes clear the existential import which he finds in them:

But of assertions, some such as those affirming something of something or denying something of something [Tir ZTt1 TLVos ii Ti &rr6 T1V6o] are simple assertions, others such as a composite assertion are compounded out of these.8

Similarly, since an assertion must be either an affirmation or a denial, Aristotle's remarks on these relationships also are revealing:

An affirmation is an assertion affirming something of something [TLV6s xoxr& TLV6S].

A denial is an assertion denying something of something [TLVOS &iT6 rLVOSJ.9

But since an affirmation means something is affirmed about something, the subject is either a name or a 'non-name' and what is affirmed must be one thing about one thing. . . Thus every affirmation and denial is either by means of a name and verb or by means of an indefinite name and verb. Unless there is a verb there is no affirmation and denial.10

Each of these passages reveals that for Aristotle, affirmation and denial indicate relationships holding between things. As his analysis of hom- onymy and synonymy,11 his discussions of the respective roles of words and of things in our reasoning,12 and his view that ov andTL are coextensive show, Aristotle's understanding of the relation of language to the world is such that in the above texts 'Ti? must signify an actual thing.13 Though every affirmation and denial is by means of a noun or indefinite noun and a verb or indefinite verb, what is being asserted in the affirmation or denial is not solely or even primarily a linguistic relationship. Rather, Aristotle's concern is with what this linguistic vehicle indicates about the relation of things in the world, namely that in an affirmation one thing is being

285



asserted to belong to one thing while in a denial one thing is being asserted not to belong to one thing.

It is this ontological aspect of affirmation and denial that precludes Aristotle's believing in the referential failure of the subject of an assertion. Since an assertion, in order for it to be an assertion, must assert either that one thing belongs to one thing or that one thing does not belong to one thing, should the subject fail to refer, then there will not be anything to which we can assert the predicate belongs. In other words, should the subject of a sentence fail to refer, the sentence will be neither true nor false and hence will not be an assertion. Thus, to use a standard example, for Aristotle, since there is no present King of France, we can neither affirm nor deny anything of him and hence no sentence in which "the present King of France" is the subject, such as "The present King of France is bald," will be an assertion.

The argument that I have just offered shows why Aristotle's analysis of affirmation and denial prevents his believing that the subject of an assertion might fail to refer. To do so would simply contradict De Interpretatione v, vi, and x.

III

Turning first to De Interpretatione xi 21 a 25-28, I will argue that rather than asking whether "Homer exists" follows from "Homer is a poet," Aristotle is really merely pointing out that since being a poet is not part of Homer's essence, "Homer is a poet" is a contingent truth. However, in order to recognize that this is Aristotle's intention, we must not neglect the context of his remark. Whereas other commentators have tended to treat 21 a 25-28 in isolation, we can properly interpret these lines only by understanding the issues with which the whole chapter is concerned.

De Interpretatione xi opens with a discussion of the problem of determining the complexity of an assertion and then shifts to the question of the permissibility of compounding predicates, i.e. moving from saying "X is Y" and "X is Z" to saying "X is YZ." Though on certain occasions this is allowable, on others it is not. Aristotle offers us two such cases. The first (20 b 35-36) concerns the use of an adjective that can both be predicated of the subject of a sentence and be used as a modifier of a direct object (e.g. "The man is good"' and "The man is a cobbler" do not imply that "The man is a good cobbler"). The second (20 b 37-21 a 3) is that unrestricted combination of predicates will lead to redundant predications (e.g. "The man is white"' and "The man is a white man" are acceptable, while "The

286



man is a white white man" is not allowable). In order to avoid these problems Aristotle concludes (21 a 5-7) that unrestricted combination of predicates is unacceptable.

But when is such combination permissible? Not when either an accidental predicate is asserted of a subject (e.g. moving from asserting "Man is musical" and "Man is white" to asserting "Man is musical white") nor when two accidental predicates of a given subject are asserted of one another (e.g. moving from asserting "The white is musical" to "The white is musical white"). Nor, Aristotle adds, with as many predicates as inhere in the essense of a subject (e.g. "Man is a man animal" or "Man is a two-legged man").

However, at 21 a 18-20 Aristotle recognizes that this last blanket statement needs qualification since such combination is acceptable when speaking of a single individual (e.g. "This particular man is a two-legged man"). Yet before allowing such combinations in all cases where we speak of individuals singly, Aristotle has second thoughts and argues that we cannot always do this.

There are two cases to be considered. First are those where the added term implies a contradiction. Whenever a contradiction follows because the added term contradicts that which inheres in the subject's essence, the assertion is not true but false. In an example reminiscent of the Phaedo Aristotle notes that it is simply false to say that a dead man is a man. Since to be a man is to function in a certain way and that which is dead cannot function in this manner, a dead man is only a man homonymously. Second are those cases where, because the added term does not contradict something which already inheres in the subject's essence, we will not have a contradiction and the assertion will be true. In this second case, Aristotle is thinking of assertions such as "The white man is a man." Since being white is not essential to that which is a man, this assertion is true.

Now come the crucial lines immediately preceding the text with which we are concerned. At 21 a 24 Aristotle repeats what he has just said:

In other words, whenever on the one hand one of the opposites should inhere [ivvuT&p-x-], the assertion is always not true; whenever on the other hand neither of the opposites should inhere [Ivut.6ppxJn, the assertion is not always true.14

If we see Aristotle's point as a repetition of what he has just said, then his remark is quite straightforward. Since for Aristotle kvvudLpxO has the technical sense of "to inhere in the essence of a thing," the first alternative simply notes that when we have a pair of opposites such as being dead and being alive, one of which inheres in the subject (e.g. being alive in man),

287



then should we predicate its opposite of the subject (e.g. "The man is dead"), the assertion will always be false. The second alternative is the case where, since neither opposite should inhere in the subject (e.g. being white and being non-white), then predicating either opposite of the subject (e.g. "The man is white") will not always be false. In other words, on the second alternative, since we are merely predicating an accident of the subject, then the assertion will sometimes be true and sometimes be false.

The reason why 21 a 24-25 is another way of saying what is said in the preceding lines is that the point of the previous sentence was to note how the adjective modifying the subject (e.g. "dead" in "The dead man is a man" and "white" in "The white man is a man") was related to the subject. In the first case, we had the adjective contradicting the essence of what it modified, in the second it merely denotes an accident of the thing modified. Thus Aristotle realizes that the key issue about which he should be concerned is predication of the essence and predication of an accident.

That 21 a 25-28 should be understood as an example relating to 21 a 24-25 and, in particular, as an explication of the second alternative, is clearly implied by Aristotle's beginning his comment by WLarreop, thereby showing that he simply is illustrating what he has just said. Therefore the aim of 2 Ia 25-28 merely is to provide us with an example of an assertion which involves something which does not inhere in the subject, i.e. an accident, and hence is "not always true." Thus the full passage reads:

In other words, whenever on the one hand one of the opposites should inhere [Evu'n&PXB], the assertion is always not true; whenever on the other hand neither of the opposites should inhere [ll ivvu1T&px-ql, the assertion is not always true. As for instance Homer is something, e.g. a poet. Well now, is he or isn't he? For of what is accidental is "is" predicated of Homer, for it is the case that he is a poet, but not of to what is essential is "is" predicated of Homer.15

Aristotle's worry is the following: Assert of Homer something which does not belong to him essentially, for example his being a poet. Because being a poet does not inhere in Homer's essence, i.e. since Homer is essentially a man and since he would continue to be a man even if he ceased to be a poet and became a shoemaker or a shipbuilder, being a poet is only an accident of Homer. Hence "Homer is a poet" need not be true but might be false. This reveals that the real point of the question "Well now, is he or isn't he?" is not a query about whether Homer does or does not exist. Rather it is simply a question as to the truth of "Homer is a poet," namely that since being a poet is accidental to Homer, is it true that Homer is a poet or isn't it?

It is in this context that 21 a 26-28 is to be understood. Because in these lines Aristotle is clearly talking about "is" and since he intends these lines

288



to be an explanation of his preceding remarks, 21 a 26-28 must refer back to whether or not "Homer is a poet" is true or false. Therefore, since (1) Aristotle is concerned with "is," (2) being a poet is an accident of a man, not part of his essence, (3) Aristotle is concerned with the sentence "Homer is a poet," and (4) Aristotle now says that in the sentence "Homer is a poet" "is" is being used of what is accidental, not of what is essential, then it follows that 21 a 26-28 simply points out that "is" is being used to predicate something which is an accident of Homer rather than being used to predicate either the whole or a part of his essence, i. e. something which inheres in Homer. Hence rather than making any reference whatever to existence, predicating "is" of what is essential refers to what "is" in the sentence "Homer is a poet" does not do - namely predicate of Homer either his essence or a part of his essence.16

Such a reading as I have proposed would be perfectly consistent with the conclusion which Aristotle offers to his discussion of those cases where the predications are said singly of a particular instance, namely that if we take any sentence which (1) involves a predication which does not contradict what is inherent in the subject and (2) should be a predication of what is essential, not of what is accidental, then the replacement of nouns by their definitions will mean that what we will be saying of the particular instance will be true. Nor does the afterthought with which Aristotle closes De Interpretatione xi - namely that just because that which is not is an object of opinion, it is not true to say that it is something - in any way conflict with my interpretation of 21 a 25-28.17

Thus we see that the traditional interpretation of Aristotle's remarks is unjustified. Rather than asking whether "Homer exists" follows from "Homer is a poet," this passage merely notes that because being a poet is an accident of Homer, "Homer is a poet" may be true or false.

Turning now to the second text, Categories x 13 b 12-35, I will argue that in this passage Aristotle really explains the opposition of contradictories in terms of the applicability or inapplicability of certain predicates to the subject of an assertion. My proposal is that in 13 b 12-19 Aristotle notes that only when a given pair of contraries are predicated of a subject to which they are applicable will necessarily one assertion be true and the other be false, since should both contraries be inapplicable to the subject of which they are predicated, then both assertions will be false. In 13 b 20-27 he makes a similar observation with respect to possessions and their corre- sponding privations. Should a possession and its privation both be inapplicable to a given subject, or even if they are applicable but the appropriate time for the acquisition of either one or the other has not yet arrived,

289



then it will be false to predicate either the possession or its privation of the given subject. Only when the possession and its privation are applicable to a given subject and the time is appropriate for their acquisition, need one of the two assertions be true and the other be false. In 13 b 27-35 Aristotle concludes that the case of affirmation and denial is different simply because, whether or not the predicate is applicable to the subject, either affirming this predicate of this subject or denying this predicate of this subject will be true. Thus in none of these three cases does Aristotle make any mention of the nonexistence of that to which the subject purportedly refers.

In order to appreciate this analysis of 13 b 12-35 fully we first must review the vital role which the notion of applicability plays in Aristotle's earlier discussions of contrariety and of possession and privation. Rather than being of only slight importance, applicability in fact is fundamental to the explanation of both notions. Aristotle's discussions of contrariety, both here in Categories x and elsewhere, repeatedly stress that the basis of the notion of contrariety is not just any sort of difference but rather difference with respect to some thing.18 Thus even if A and B are not opposed as relations, it does not follow from the fact that A is different from B that A and B are contraries. Rather only when individuals of some kind Z are either capable of receiving A or capable of receiving B but not both A and B at the same time are A and B said to be contraries.

Let us consider how this applies to several of Aristotle's examples of contraries such as odd and even, well and sick, white and black, and just and unjust. Each of these pairs of contraries properly belongs only to things of a certain kind. Thus being odd and being even properly belong only to numbers, being well and being sick to living things, being white and being black to physical surfaces, and being just and being unjust to human beings. Aristotle divides contraries into those which do not have intermediates (e.g. odd and even) and those which do (e.g. white and black). In the former case, everything which is capable of being odd or capable of being even must be either odd or even. However in the latter case, a physical surface need not be either white or black. Rather it also can be some intermediate color such as grey or brown. Sometimes, as in the case of colors, these intermediates have names; sometimes, as in the case of being just or being unjust, the intermediate does not. In this event we only can designate this state by saying that it is neither one extreme nor the other.

Examine the first two pairs of contraries, both of which lack intermediates. Even though every integer must either be odd or be even and every living thing either be well or be sick, it does not follow that every

290



living thing must either be odd or be even or that every integer must either be well or be sick. In fact, except in an accidental way, it is impossible for an integer to be well or to be sick and it is impossible for a living thing to be odd or to be even. The reason is clear: except in an accidental way living things simply are not capable of being odd or of being even and integers simply are not capable of being well or of being sick. Hence saying that the number five neither is well nor is sick does not show that being well and being sick are not contraries since the number five simply is incapable of receiving either of these contraries except in an accidental way. The same remark applies mutatis mutandis for some living thing such as Socrates and the contrary qualities being odd and being even.

In the case of contraries which do have intermediates, there is no necessity that that which is capable of receiving either contrary possess one contrary or the other simply because that which is capable of receiving either contrary might instead possess some intermediate. However, the necessity that the recipient either be one of the extremes or be one of the intermediates remains. Thus a physical surface need not either be white or be black since it also might be brown or be gray or be some other color. Nevertheless every physical surface must be some color or other. Just as in the previous case, contraries are contraries only with respect to that which is capable of receiving them. Hence no physical surface either is well or is sick; nor is any physical surface just, unjust, or the intermediate state neither just nor unjust. Therefore, saying that this physical surface neither is well nor is sick does not show that being well and being sick are not contraries but rather reveals that physical surfaces simply are not capable of receiving these qualities except in an accidental way. Aristotle writes that

With contraries between which there is nothing intermediate it is necessary for one or the other of them always to belong to the things they naturally occur in or are predicated of. For there was nothing intermediate in just those cases where it was necessary for one or the other to belong to a thing capable of receiving them, as with sickness and health and odd and even. But where there is something intermediate it is never necessary for one or the other to belong to everything; it is not necessary for everything to be white or black that is capable of receiving them ... , since something intermediate between these may perfectly well be present.19

Turning to the opposition of possession and privation, it is clear that considerations of applicability again determine whether it is necessary for one or the other to belong. Thus, as Aristotle explains the terms "possession" and "privation," if C is something which individuals of some kind Z can possess and D is the privation of C in these individuals, then (I) nothing can have both C and D at the same time, (2) neither C nor D can be

291



essential to being a Z in the sense that if an individual Z ceases to have either C or D, then it ceases to be, (3) at some time t, it becomes necessary for an individual Z either to have the possession C or its privation D, (4) if at some time t an individual Z loses C, then at no later time can the individual Z reacquire C, and (5) if at some time t an individual Z has the privation D, then at all later times the individual Z will have the privation D. Applying these considerations to Aristotle's standard example of a possession and its privation, namely being sighted and being blind (where "X is blind" means "X does not have the vision that X naturally would have when X naturally would have it"20), we see that: (1) Nothing is ever both sighted and blind; (2) Being sighted and being blind are not essential to any of those animals which possess them (after all, blind men are still men); (3) Though in the case of human beings we are born either sighted or blind, other species such as dogs are neither sighted nor blind at birth. However there does come a point in time when puppies' eyes open and at that moment the puppies must either be sighted or be blind; (4) If an animal possessing sight loses its sight, then it can never regain it; and (5) Blind animals never regain their sight.2'

More clearly even than in the case of contraries, the opposition of a possession and its privation depends upon the nature of that to which the possession and its privation belong. Not only will the failure of everything either to be sighted or to be blind be due to those cases where both simply do not belong at all, but there are other cases where both the possession and its privation fail to belong simply because the recipient has not yet reached the stage where it would either acquire the possession or suffer the privation. Thus the fact that this tree or that newborn puppy neither is blind nor is sighted does not show that being blind and being sighted are not opposed as a possession to its privation. Rather what it shows-is that a thing must have either the possession sight or its privation blindness only when a thing both is capable of receiving the possession and its privation and has reached the proper stage of its development.

For it is necessary for one or the other of them li.e. the possession and its privation] always to belong to a thing capable of receiving them, since if it is not yet natural for something to have sight it is not said either to be blind or to have sight ... [However] once it is natural for something to have sight then it will be said either to be blind or to have sight . . 22

Returning to 13 b 12-35, it is clear that the phenomena that Aristotle there describes could be explained by means of the notion of applicability. Thus, for example, one of the pair of contraries being odd and being even need only belong to that which is a number; if being odd and being even

292



are predicated of that which is not a number, then both assertions will be false. Or again, the possession sight and its privation blindness need only belong to those things capable of receiving sight and even then only when they have arrived at the point in life when it is necessary for the thing to have received either the possession or its privation. Should the thing be a stone, a plant, or even a newborn puppy whose eyes have not yet opened, it is false to say that the thing either is sighted or is blind. Finally, in the case of the opposition of contradictories, if, on the one hand, the predicate is capable of belonging to the subject, then according to whether it does or does not belong, either the affirmation of this will be true and the denial will be false or vice versa (e.g. since being wise is capable of belonging to Socrates, whether or not Socrates is wise will determine which of the pair of assertions "Socrates is wise" and "Socrates is not wise" is true and which is false). If, on the other hand, the predicate is incapable of belonging to the subject, then it will be false to affirm and true to deny this predicate of this subject (e.g. since being even is incapable of belonging to anything that is not a number, "Socrates is not even" is true and "Socrates is even" is false).

However can such an analysis fit the text of 13 b 12-35? The answer is clearly yes. The crux of any explanation of these lines will be the translation and interpretation of a number of occurrences of the genitive singular participial form of Ea'L, often accompanied by the genitive E2xp6vrouV.23 Whereas previous translators have understood all of these phrases in terms of Socrates' existence, it also is possible to read these clauses as genitive absolutes that elliptically refer to Socrates' being, i.e. to what Socrates essentially is. Thus, instead of rendering such phrases as OVTOs yap M:xp0aLTovs as "For if Socrates exists," it also is perfectly acceptable to treat the remark as "For of Socrates' being," where this expression is understood as referring to his being what he is, i.e. his essence. Since Socrates' being is to be a sort of living thing or, more specifically, to be a certain kind of animal, understanding these clauses in the manner that I am suggesting allows us to read these phrases as referring to the genera in virtue of which Socrates necessarily is either well or sick and either sighted or blind.24 Read in this way, in 13 b 12-19 Aristotle would be noting, on the one hand, that of Socrates' being, i.e. of his being a living thing, then, since every living thing must either be well or be sick, necessarily one of the pair of assertions "Socrates is well" and "Socrates is sick" will be true and the other will be false. On the other hand, of Socrates' not being, i.e. of his not being a living thing, then Socrates is not capable of receiving either contrary and hence both assertions are false. In 13 b 20-27 Aristotle's point would be that of Socrates' not being, i.e. of his not being an animal which normally would

293



possess sight, then neither "Socrates has sight" nor "Socrates is blind" will be true. Even were Socrates an animal of the appropriate sort, both assertions still might be false because he might not yet have reached the appropriate stage of his development to acquire either the possession or its privation. Only if Socrates is a sufficiently mature animal of a certain kind will it be necessary for one of the pair of assertions "Socrates has sight" and "Socrates is blind" to be true while the other is false. Finally, in 13 b 27-35 Aristotle would be noting that regardless of Socrates' being a living thing or not being a living thing, either the affirmation "Socrates is sick" or the denial "Socrates is not sick" will be true and the other will be false.

With these points in mind, I propose to translate 13 b 12-35 as:

It might, indeed, very well seem that such happens [i.e. necessarily it will always be the case that one assertion will be true and the other assertion will be falsel in the case of those contraries said with combination - "Socrates is well" being contrary to "Socrates is sick" - but not even as concerns these is it necessary always for one to be true and the other to be false. For, on the one hand, of Socrates' being a living thing, one will be true and one will be false, while, on the other hand, of Socrates' not being a living thing, both will be false. For of Socrates himself not being a living thing at all, neither "Socrates is sick" nor "Socrates is well" will be true.

As for privation ,and possession, of Socrates' not being an animal which normally is sighted at all, neither are true. Even of Socrates' being an animal which normally is sighted, not always will one or the other be true. For "Socrates has sight" is opposed to "Socrates is blind" as privation and possession. Even of Socrates' being an animal which normally is sighted, it is not necessary for one or the other to be true or false, since until the time when it is natural to have the possession or its privation, both are false. For of Socrates' not being an animal which normally is sighted at all, then again both are false, both "he has sight" and "he is blind,"

As for affirmation and denial, always, if Socrates should be a living thing or if Socrates should not be a living thing, one will be false and the other will be true. For consider "Socrates is sick" and 'Socrates is not sick." Of his being a living thing it is evident that one of them will be true while the other will be false and of his not being a living thing "he is sick" is false while "he is not sick" is true. Thus only of these is it distinctive that always one of these is true while the other will be false -just as many as are opposed as affirmation and denial.

Since so much of Aristotle's Categories x discussion of the opposition of contraries and of possessions and privations concerns their applicability to the recipient, it is only natural to try to see his discussion of contradictories in these terms. When we further recognize that it is possible to read 13 b 12-35 in terms of this relatively clear notion, it becomes obviously prefer- able to interpret these lines as concerning applicability rather than as dealing with referential failure.

294



Hence, to summarize this section: Neither Categories x 13 b 12-35 nor De Interpretatione xi 21 a 25-28, the two texts that generally are cited as evidence for contending that Aristotle held that the subject of an assertion might not refer, in any way justify ascribing such a view to him.

Conclusion

In this paper I have argued that the traditional view that Aristotle believed that the subject of an assertion might fail to refer is false by showing (1) that Aristotle cannot accept this position without contradicting his account of affirmation and denial and (2) that the two texts which usually are adduced as evidence for ascribing to him such a belief previously have been misin- terpreted. Obviously the doctrines which I have discussed in this paper have significant implications upon Aristotle's account of existential assertion, implications which I hope to draw in later papers.25'26

Appendix

Subsequent to the above article's being accepted, Michael Wedin has published a piece on the same problem ("Aristotle on the Existential Import of Singular Sentences," Phronesis 23 (1978) 179-196). For the sake of clarifying the issue, the editor kindly has permitted me briefly to indicate my differences with Professor Wedin.

1. To a large extent Wedin's interpretation of De Interpretatione 21 a 25-28 rests on an indefensible reading of Metaphysics A. vii. Pace Wedin, Aristotle never uses the expressions xaO' ov'o / XaT& aV,uEPXOS To ov (and its cognates) to denote anything other than the difference between using "is" (and its cognates) to assert what is essential and what is accidental (cf. notes 6 and 7 above; given that Thorp's article appeared in this journal, it is disturbing that Wedin fails to mention it).

2. Even aside from I1, Wedin still would err in reading 21 a 25-28 in terms of the distinction between what is accidentally a predication and what is essentially a predication (e.g. "That white is wood" vs. "That wood is white") rather than the distinction between predication of what is accidental and predication of what is essential (e.g. "The man is white" vs. "The man is an animal"). Given Aristotle's example in 21 a 23 and his use of the technical term kVV1rPX.ELV in 21 a 24-25, Aristotle's concern must be with essences and accidents. Hence in 21 a 25-28 he must intend the latter, not the former, contrast.

3. At best Wedin's reading would resolve the apparent contradiction

295



between De Interpretatione 21 a 25-28 and Categories 13 b 12-35. It would not resolve the contradiction between his reading of these texts and Aris- totle's theory of predication (cf. Section II above).

4. Finally Wedin's analysis of Prior Analytics I.xlvi involves several serious difficulties, the most noticeable being his inadequate basis for rejecting the traditional reading of this text (cf. his paper, p. 195 n. 19). Standard Greek usage of course permits a definite article + adjective construction to refer either to an attribute or to the individual to which the attribute belongs (cf. Herbert Weir Smyth, Greek Grammar, Cambridge, Mass.: Harvard University Press, 1966, ? 1153a). Just because Aristotle uses To O1VLGJOV in the first way in one place does not imply that he cannot use it in the second way in another place. Thus, while Wedin is quite correct in noting that in Categories vi 6 a 26-35 To &Lvloov concerns the quantity itself, this provides him with no basis whatsoever for denying that, as the traditional reading holds, in Prior A nalytics I.xlvi 5 1 b 27 To6 XVLCoOV refers to the quantified individuals.

Virginia Commonwealth University

1 Categories and De Interpretatione, Oxford 1963 p. 59. On the crucial points, the Oxford translation by E. M. Edghill (in The Basic Works of Aristotle ed. by Richard McKeon. New York, 1941) is basically the same as John Ackrill's. On the other hand, H. P. Cooke's Loeb edition, Cambridge, Mass. and London, 1967 is somewhat confusing. Initially he adopts the same reading as Edghill and Ackrill, i.e. that the question in 21 a 26 concerns the existence of Homer:

For example, take 'Homer is something'- 'a poet' will do for our purpose. But can we say also 'he is'? Or will that be incorrectly inferred?

However, then Cooke appears to adopt what I will argue is the correct reading of 21 a 26-28:

'Is' was used incidentally here. For our statement was 'he is a poet,' and 'is' was not predicated of him in the substantive sense of the word.

So as to leave no doubt concerning the meaning of these lines, Cooke adds a footnote claiming that by the substantive sense of "is" Aristotle "Otherwise [means] the sense of existence. For the word 'is' expresses exists in addition to being the copula" (p. 154 n. a). If "otherwise" here means "in other words," then contrary to appearances, Cooke's view is really the same as those which I have already cited. However, if "otherwise" means "in other places" then Cooke (1) correctly understood the real point of lines 21 a 26-28 but (2) misunderstood the X1Y0'iv,T6 ("'essential") use of "is" since nowhere, despite what various commentators have alleged, does the xoa'c,-r6 use of "is" have the meaning of "exists". On this last point, cf. below n. 7.

296



2 Because it is a part of the Postpraedicamenta, Categories x through xv, doubt may be cast on the authenticity of Categories x. Though I am inclined to accept the authenticity of the Postpraedicamenta, the argument of this paper in no way depends upon such an assumption. Since my sole reason for treating Categories x 13 b 12-35 is to examine the evidence which it may provide that Aristotle believed that the subject of an assertion might fail ro refer, should Categories x prove spurious then it will of course provide no support for the ascription to him of such a position. On the other hand, should Categories x prove to be authentic, then my analysis of 13 b 12-35 will show why this text does not provide any evidence for believing that Aristotle thought that the subject of an assertion might not refer. 3 Op. cit., p. 37-38. On the crucial points, the translations of E. M. Edghill, op. cit., and H. P. Cooke, op. cit., are the same as John Ackrill's, all three seeing Aristotle's explanation as involving nonreferring subjects. 4 It should be noted that Aristotle does not even mention the opposition of relatives in his explanation of the opposition of contradictories in 13 b 12-35. The reason is simply that since pairs of relatives will clearly always be true or false together, no one would think that they might be similar to the opposition of things opposed as affirmation and denial. 5 Amongst the commentators who accept this analysis are: John Ackrill, op. cit., p. 1 10-1 1 1; Nicholas White, "Origins of Aristotle's Essentialism, " Review of Metaphysics 26 (1973) 62; Eugene Babin, The Theory of Opposition in Aristotle, Notre Dame, Indiana, 1940 p. 81; and Manley Thompson, "On Aristotle's Square of Opposition," The Philo- sophical Review 62 (1953) [Reprinted in Aristotle ed. by J. M. E. Moravcsik, Garden City, New York, 1967 p. 55-57. Subsequent references to Thompson's article are to this edition.]. 6 Both Manley Thompson (ibid., p. 55-57) and R. M. Dancy (Sense and Contradiction: A Study in Aristotle, Dordrecht 1975, Appendix II) have attempted to resolve the apparent contradiction by reinterpreting De Interpretatione xi 21 a 25-28. Thompson's proposal was to see the xaO'aVOr6 use of "is" in 21 a 25-28 as being ". . . the substantive sense of the word . . . [meaning] the same as 'is a substance"' (Thompson, op. cit., p. 56). Thus Aristotle would not be baldly contradicting himself since in 21 a 25-28 he would not be denying that "Homer is a poet" implies "Homer exists." Rather, according to Thompson, Aris- totle's point is to deny that "Homer is a substance" follows from "Homer is a poet."

Though this interpretation may resolve the apparent contradiction between Categories x and De Interpretatione xi, it isprimafacie mistaken. First, there is the obvious point that "Homer is a poet" does imply that "Homer is a substance," just as much as "Homer is a man" implies that "Homer is a substance," since being a poet is being a certain sort of man. Second, and more deeply, Thompson's claim that the xoHo',x{r6 use of "is" is ". . . clearly . . . 'is' in the substantive sense of the word and means the same as 'is a substance"' contradicts Metaphysics A. vii 1017 a 22-27, where Aristotle says that the verb "to be" can be said xa0'avr6 in each of the categories, i.e. not merely "is a substance" but also "is a quantity," "is a quality," etc. Perhaps Thompson's reading of TO6 Ov xa0'nv{T6 as the substantive sense of"is" may be due to his erroneously conflating the sense of"is" where "is" is used to make an essential predication (e.g. "Man is an animal" or "Blue is a color") with the way in which, for Aristotle, all secondary uses of"is" must ultimately relate to a primary use of"is", i.e. ". . . is a substance," because everything else which is, is only due to its in some way being related to that which is a substance. These two points are quite different and should not be equated. Regarding the correct interpretation of the xn0'aVOr6 sense of"is," again cf. below, n. 7.

297



Dancy's proposal is to understand Aristotle as asking whether "Homer is a man" follows from "Homer is a poet." However, since to be a poet is to be a certain sort of man, Dancy would have us read Aristotle as rather implausibly asking whether being a certain sort of man implies being a man. Furthermore Dancy's suggestion rests on a misinter- pretation of De Interpretatione xi. Contra Dancy, Aristotle is not concerned with whether assertions of the form "X is YZ" entail those of the form "X is Y," but rather the reverse, i.e. whether we can compound assertions of the forms "X is Y" and "X is Z" together to yield "X is YZ. " On this point, cf. below, p. 286-7. 7 The use of EIV4tL xVTr& (v R0EPqx6OS and xa0`avT6 elsewhere means using the verb "to be" or one of its cognates to assert of a subject what is accidental or essential to it (cf. Metaphysics A.vii 1017 a 7-30 and VI.ii). However reading De Interpretatione 21 a 25-28 in the traditional manner forces us to find the xaxo'T6 use of ELvat in the existential use of "to be" (e.g. "Homer is") and to relegate all copulative uses of ELVIL to being the xaT&

GVu10E033x6s use of rIvaL. For a defense of reading the xa-r av[upEJx6s and xOt`aVTO uses of ElVaL as the use of "to be" to predicate what is accidental and what is essential, cf. J. W. Thorp, "Aristotle's Use of Categories," Phronesis 19 (1974) 238-256 and my thesis Ervat and Existence in Aristotle (unpublished Ph. D. dissertation, Ohio State University, 1974) Chapter Four ("'Being' in the Dictionary: An Analysis of Metaphysics A.vii"). 8 De Interpretatione v 17 a 20-22. Unless otherwise noted, all remaining translations are my own. The text used was L. Minio-Paluello's O.C.T. edition. 9 De Interpretatione vi 17 a 25-26. "0 De Interpretatione x 19 b 5-12. 1 For example, cf. Categories i 1 a 1-12 and Acrill, op. cit., p. 71. 12 E. g., Topics I.xviii 108 a 18-26, Sophistici Elenchi i 165 a 6-9, and Metaphysics r.iv 1006 b 20-22. 13 Though it might be thought that for Aristotle,just as for the Stoics, TL not only refers to actual entities but also to merely possible entities or to fictional entities, such an interpretation clearly is unacceptable. For example, cf. De Interpretatione xi 21 a 32-33 ("But just because that which is not is an object of opinion, it is not true to say that it is something. For opinion of it is not that it is, but that it is not."), Prior A nalytics Lxxxviii 49 a 36, Sophistici Elenchi xxv 180 a 32-38, and Metaphysics A.xxix 1024 b 21-24. Unlike the Stoics, Aristotle does not believe that TL has wider scope than "ov."

Also compare J. M. Rist's discussion of the Stoic that view 6ov and ri are not coextensive since there are things which are not (cf. his Stoic Philosophy, Cambridge. U.K., Cambridge University Press, 1969 p. 153-155). While citing the differences between Alexander of Aphrodisias (who, in his commentary on Aristotle's Topics, preserves the Aristotelian conception of the relation of ov and Tv (cf. M. Wallies (ed.), In A ristotelis Topicorum Libros Octo Commentaria (Berlin, 1891) p. 301, 19 ff. and p. 359,12 ff.)) and the orthodox Stoics, Rist seems not to appreciate that this divergence is not due to Alexander's ". . . ignoring problems ... about fictional or otherwise non-existent name- ables" (op. cit., p. 154) but rather stems from real philosophical disagreement between Peripatetics and Stoics on the relation of language to the world. 14 De Interpretatione xi 21 a 24-25. 15 De Interpretatione xi 21 a 24-28. 16 In his edition of De Interpretatione Ackrill considers the very treatment which I have proposed, only to reject it:

It is clear that the accidental predication of which Aristotle speaks in this paragraph [21 a 18-33] is 'accidental' in the second sense of the two senses distinguished above;

298



it is incidental or indirect predication. Aristotle's example is not a happy one. But when he says that in 'Homer is a poet' the 'is' is predicated accidentally of Homer ('because he is a poet, not in its own right') his point evidently is not that 'is' gives an accidental as opposed to essential property of Homer, but that it attaches to Homer only indirectly, qualifying him only qua poet. (p. 148)

In other words, Ackrill is conceding that his interpretation of 21 a 25-28 forces him to be dissatisfied with Aristotle's example. However, rather than accept the interpretation I have proposed for these lines, Ackrill advocates a theory whereby the text claims that "is" is being linked to "Homer" only via the direct object "poet" and not unqualifiedly as in "Homer is."

Aside from the intrinsic awkwardness of such an interpretation - one which does not even fully please its author - the crucial question which we must consider is why Aristotle's ". . . point evidently is not that 'is' gives an accidental as opposed to essential property of Homer . . . " i.e. what basis Ackrill has for rejecting the much simpler and more straightforward explanation. Unfortunately, we are left ignorant as to what are these evident grounds. Are they that such an interpretation cannot be reconciled with the rest of chapter xi and, in particular, the lines immediately preceding 21 a 25? If so, my proposal that 21 a 25-28 simply is meant to be an illustration of 21 a 24-25 not only dissolves this difficulty but in fact shows that the interpretation which Ackrill rejects is correct. However, if this is not the alleged ground for Ackrill's dismissal, then the basis of his rejection is not the least bit evident. In short, Ackrill's rejection of an analysis of 21 a 25-28 in terms of "is" giving ". .. an accidental as opposed to an essential property of Homer .. ." appears to be inadequate. 17 Manley Thompson read 21 a 32-33 as evidence in favor of his understanding the xc0'eavf'r meaning of E'vast as "is a substance":

[T]he fact that a non-existent Homer may be the object of opinion, as he would be if we were to construct a myth about him, does not mean that it is true to say Homer is something. The assertions "Homer is merely an object of opinion" and "Homer is a mythical being" are about Homer only in the sense of denying that he is in fact anything. While we might say that "Homer is a poet" is true in fiction, what is true in this case is true of the myth and not of Homer. And the myth does exist, even though not per se as a substance does. (Op. cit., p. 57)

However, surely Aristotle's point is not merely that myths are not substances. Of course, to use Thompson's example, that Homer is a part of a myth is in and of itself sufficient indication that the mythical Homer is not the same as the flesh and blood man; simply because it is a myth, it is about something which is not. Nevertheless, contra Thompson, this does not in any way imply that the focus of Aristotle's closing remark is that to think about what is contrary-to-fact, such as a myth, is to think about what is not a substance. Rather, the point simply is that what is not is not by virtue of our thinking about it, something which is (cf. above, n. 13). 18 E.g., Metaphysics A.x and liv and Topics Il.ii. 19 Categories x 12 b 27-35 (Translated by Ackrill, op. cit., p. 34-35). 20 Cf. Topics Vl.vi 143 b 34-35 and Metaphysics E.iii 1047 a 8-9. 21 Though modern medicine obviously would force us to modify Aristotle's example, the point which he is trying to explain still is clear and could be defended via suitable modifications.

299



22 Categories x 13 a 4-10 (Translated by Acrill, op. cit., p. 35-36). 23 Since the two occurrences of the present subjunctive of eLvcL in 13 a 28 (Qiv T?E EOLV TE

j) clearly take the same sense as the previous occurrences of the genitive participle of dLvoL, it is these prior uses of eLVoXL that are crucial to the understanding of these later senses. 24 1 would like to thank Nicholas Smith for suggesting this reading to me. 25 For several of these implications, cf. my "The Existential Presuppositions of Aristotle's Logic" (forthcoming in Philosophical Studies). 26 1 am indebted to Marvin Fox and Robert Turnbull for their many helpful criticisms and suggestions on an earlier version of this paper. Any errors are, of course, my own responsibility.

300



Phronesis Volume 24 Issue 3 1979 [Doi 10.2307%2F4182075] William Jacobs -- Aristotle and Nonreferring Subjects

Documents

jstor terms

pmall use subject

jstor archive

terms conditions of

aristotle remarks

concluding section

terms of aristotles

types of opposites