aristotle on focal meaning and the unity of science

Aristotle on Focal Meaning and the Unity of ScienceAuthor(s): Michael T. FerejohnSource: Phronesis, Vol. 25, No. 2, Aristotle Number (1980), pp. 117-128Published by: BRILLStable URL: http://www.jstor.org/stable/4182088 .

Accessed: 03/09/2013 15:07

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 129.68.65.223 on Tue, 3 Sep 2013 15:07:29 PMAll use subject to JSTOR Terms and Conditions

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

http://www.jstor.org/stable/4182088?origin=JSTOR-pdf

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


Aristotle on Focal Meaning and the Unity of Science

MICHAEL T. FEREJOHN

This study owes its impetus to a group of suggestions put forward by G. E. L. Owen in his 1960 article, "Logic and Metaphysics in Some Earlier Works of Aristotle".1 In that work, Owen gives predominantly textual arguments for the chronological scheme represented by the following three theses.

(I) In the earliest works of the Corpus (including those treatises which comprise the Organon), Aristotle was convinced that the verb "to be" [Fivasl and its cognates were ambiguous expressions, and that this ambiguity foredoomed the Platonic and Academic program of constructing a single, unified, and all-encompassing science of Being.

(I1) During an intermediate stage of Aristotle's development (represented by the seventh book of the Eudemian Ethics), he regularly noticed and worked with the possibility of systematic, or non-accidental, ambiguity, though he did not view this possibility as having any bearing on his negative claim in (I).

(111) In Metaphysics r2, which reports his mature views on this subject, Aristotle finally struck upon the discovery that the verb "to be" itself was a case of systematic ambiguity (of a sort which Owen labels focal meaning ambiguity). Moreover, this discovery influenced him to reverse his previous estimation of the prospects for a universal science, and even to resurrect in the Metaphysics the Platonic program that he himself had dismissed as futile in his earlier works.

Most of what I have to say concerns (III). What will be at issue, however, is not its truth. That matter is put beyond serious dispute by the relatively plain structure of Metaphysics F2. The chapter opens with the observation that the expression ov, which I shall translate here as "existent", is "said in many ways" [roXXxCas XiyCTaLJ, but that this ambiguity is not the sort Aristotle calls simple "homonymy" (wherein two or more wholly unrelated meanings come to be associated with a single expression as a result of etymological happenstance). Rather, the various applications2 of ov are said to be interrelated in that they all "make reference to (or "point toward") a one" 11Tp4Os gvJ (1003 a 33), which remark Owen paraphrases as saying that the term hasfocal meaning. Then, after a few lines of sketchy argumentation (which will be scrutinized below), Aristotle extracts from this the "obvious" conclusion (cf. 8Xov oivv at 1003 b 15) that the study of beings, qua beings, falls within the scope of a single science.

117



But even though (III) is securely grounded in the text of Metaphysics r2, it nonetheless presents something of an impediment to full and precise understanding of Aristotle's doctrines. This is because he never explains in adequate detail just how the different applications of ov are all supposed to "Spoint towards a one." Nor do his essentially programmatic remarks at 1003 a 33 - b 15 more than hint at the line of reasoning by which he moves from the missing focal meaning analysis of ov to his "'obvious" conclusion (which is anything but obvious) that there can or must be an all-encompassing field of inquiry which studies the totality of existent things. Owen's (III) therefore sets before us two exegetical projects, the completion of one of which will be necessary for the other. We have first to reconstruct the focal meaning analysis of 6v which is presupposed but not stated in Metaphysics 172. From this reconstruction will emerge materials, in the form of premises, which can then be used in combination with sup- plementary principles drawn from Aristotle's theory of demonstrative science to develop a plausible interpretation of the argument of Metaphysics 172.

The Elements of Focal Meaning Analysis

What, first of all, is the sort of the things that are supposed to "point towards a one" in the case of a rirpos ?v term? Owen infuses a distinctly intensional character into this condition when he paraphrases Aristotle as saying that all the senses of a Tpos 'Ev term "have one focus, one common element."3 But this is to irmport an alien ontology into a philosophical system where it has no place. Aristotle's usual manner of theorizing about language proceeds without reference to such intensional entities as meanings and senses. Instead, his style of analysis characteristically makes do with a relatively lean ontology containing nothing more than pieces of language (words and phrases) and the extra-linguistic entities they signify [an%uaLvwJ (i.e. standfor, or denote4). Hence, an explication of the nrrp6s ?V

condition which is framed in this limited ontology will be considerably truer to Aristotle than the intensional reading given to it by Owen.

A first step towards this more restrained interpretation will be to develop a general understanding of ambiguity along these lines. We may begin by noting that correspc nding to every unambiguous term there is a very special phrase, called its logos, which not only signifies precisely the same thing or things signified by the term, but also stands with it in a quite intimate relation which approximates what we now think of as synonymy (De Interpretatione 21 a 30-4, Prior Analytics 49 b 5, Topics 101 b 39 - 102 a 1,

118



130 a 39, 142 b 2-6, 147 b 13-5, 149 a 1-2, b 3-5). Now what distinguishes terms that are "said in many ways" is that they do not have such unique logoi. There is instead for an ambiguous term a plurality of logoi each having a distinct signification.

This way of dealing with ambiguity is clearly evidenced in the statement of the focal meaning condition given at Nicomachean Ethics 1096 b 23-9:

But in what sense are different things called good?... Is it because they are all derivative of a single thing [&q' vbs e'LvaL], or [i.e. equivalently] because they are all directed towards a single thing [-rp6s Ev OVVTEXEVJ . . .?

The proposal being considered here is that the very things that are called by the name "good" (that is, all of the various significata of this ambiguous term) are so called because they themselves are all Trrpos Ev (or acp' ev&s5).

Thus, what qualifies a term as having focal meaning is that even though it has a plurality of logoi with distinct significations, at the same time all of these significata "make reference to (or "point towards") a one".

Of course, there is generally much more to know about a relation than the sort of relata it takes, and this discrepancy is nowhere more evident than in the passage just quoted, where the entire burden f making focal meaning ambiguity intelligible is made to rest on the cryptic phrase 'pos 'E'v. Of the two reasonably literal ways of translating this key piece of Aristotelian terminology, one ("makes reference to a one") is as hopelessly vague as the other ("points towards a one") is intolerably metaphorical. It is therefore necessary to look elsewhere in the Corpus to ascertain precisely what relation is being asserted of the significata of a Tpos Ev term.

Perhaps Aristotle's fullest exposition of this relation is found at Eudemian Ethics H2, 1236 a 15-23. His avowed purpose there is to analyze an ambiguity in the term "friendship", but the passage also contains a substantial amount of information concerning the ambiguity of "medical" (atpiXOV], one of his favorite examples of focal meaning ambiguity. After first noting that "friendship" is neither univocal [xota' '?v] nor strictly homonymous (a 16-17), Aristotle goes on to specify the sort of systematic ambiguity it has in common with "medical".

... [the various kinds of friendship] all make reference to a certain one, and that one is primary [Trrp,TovJ, just as in the case of "medical": we call a soul medical, and also a body, and an instrument, and an operation, but the term is applied properly [xvpLAs] to that Imedicall which is primary.

Two critical pieces of new information are divulged in these lines. The first is that the "'one"9 towards which the various significata of a urpos iev term all point is not something external to them, but is drawn from their own

119



numbers. So, in Aristotle's example, the "one" towards which all the various kinds of "medicals" point is itself a certain kind of "medical" (identified a few lines later at a 21 as the kind doctor [6 Larpos], or alternatively as the class of doctors, which constitute a unity in the sense that they are all "one in species" or "one in logos"6). The second point, which matches this in importance, is that what distinguishes the significatum which serves as the "focus" of a 'rrpos ?v term is its priority over the remaining parts of the term's overall extension.

Eud. Ethics 1236 a 17-20 therefore obviates further exegetical reliance on the problematic phrase, 'Ipbos 9v; but its own explanatory value is at once called into question with the observation that the terms "prior" [lTpo'reposJ

and "primary" [rp&wros] are themselves expressions that Aristotle regularly characterizes as " said in many ways."7 Fortunately, Aristotle himself disarms this potential confusion in the very next lines (a 20-1) by specifying that the sort of primacy he has in mind is that which he elsewhere refers to as "logical primacy" [X4 Xoy; 1Tp&rosJ,8 and which he explicates in terms of logos-inclusion:

The primary is that whose logos is contained in [the logoil of all [the rest]. [a 20-2 1]

With this specification, we may now formulate an Aristotelian definition of 'IpOs Ev ambiguity:

(FM) A term T has focal meaning iff (i) T is "said in many ways", and (ii) one of T's many logoi is non-reciprocally contained in T's remaining logoi (i.e. its significata are logically prior to theirs).

Now in the case of i'aTpLx6v, (FM) dictates that the things most properly called "'medical" (i.e. doctors) are logically prior to every other kind of "medical", and in fact the remainder of the passage under study (a 22-3) is aimed at showing just that.

During this procedure, Aristotle's analysis develops a slight wrinkle which will become important later because the device he uses to smooth it over will be crucial to the parallel analysis of "existent" developed in the next section. By relying on the general intersubstitutivity between term (i.e. "name" [5vopa]) and logos, Aristotle actually undertakes to show at a 22-3 that the name rather than the logos of the logically primary "medicals" is contained in the non-primary logoi of the term. But this shift presents an immediate difficulty. Since kaTpLxOv is after all an ambiguous term, it would appear that all of its significata share a common name, and that is LavrpLxv itself. But if this is so, it would seem to be impossible, on Aristot- le's strategy, to isolate any logically primary "medicals" because any logos

120



would either contain the name of all "medicals" or else the name of none. Such a result would leave the focal meaning analysis of "medical" in shambles.

Aristotle's reaction to this difficulty is to introduce as the "name" of the logically primary "medicals" the genuine noun-expression "doctor" [LxTpos], which he then treats as synonymous with the "proper" application of LaTpLX6V, but not with its degenerate fellows. This may be viewed as bringing something more than grammar into the identity conditions for names because it implies both (i) that the same name can take on different grammatical appearances (e.g. LaTpos and the proper application of LaTpLxov), and (ii) that a single grammatical form (e.g. atrpLx6v) can function as different names.9 In any case, this identification must be understood if Aristotle's treatment of "medical" is to hold together, for it is larpos and not the other which is shown to be implicit in the non-primary logoi of "medical" at a 22-3.

For example, a medical instrument is an instrument a doctor would use, whereas the logos of [medical] instrument is not contained in that of doctor.

The Focal Meaning Analysis of "Existent"

As I am understanding the use of ov in Metaphysics 1r2 (e.g. at 1003 a 33), this participle functions as an adjective falling within the existential role of the verb "to be". It therefore carries a meaning very close to that of the English adjective, "existent". Construed in this way, the term has an extension encompassing not only what are now roughly termed "objects" (of various sorts), but the entire range of Aristotelian "entities" inventoried in Categories 4 (1 b 25-8). Now, in keeping with the framework of analysis established in the last section, the present order of business is to identify the various sorts of "existents" into which the all-inclusive extension of this ambiguous term resolves, and then show that one of these sorts is logically prior to the rest.

The opening lines of Metaphysics Z settle the first of these issues, for at 1028 a 12-3 the distinct signification-ranges of bv are unmistakably identified as the ultimate ontological divisions which play the title role in Categories. 10 Moreover, Aristotle's explicit declaration just a few lines later (a 15) leaves no doubt as to which of the categories functions as the "one" towards which all the others point; it is that which stands at the head of virtually every list of categories given in the Corpus, and which is referred to alternatively as "Substance" [oviu.o] or "What-is-it" ['r( iarLv]. All that is needed, then, to show that ov satisfies (FM), is a demonstration that

121



Aristotelian substances are logically prior to the contents of the rest of the categories.

There is no great difficulty in identifying the "name" of the logically primary "existents", for Aristotle uses precisely the same grammatical ploy to obtain it that we saw him use in his focal meaning analysis of iaTpLxOv in Eudemian Ethics. Just as there he introduces the noun LaTpos as the "name" of the logically primary "medicals", so in Metaphysics Z and r he takes the noun ",substance" [ov'xa] as synonymous with "'existent" in its proper application (cf. 1028 a 15, and esp. 1003 b 5-19).

Identifying the non-primary logoi of ov which are supposed to contain this name proves to be a much more difficult matter. To my knowledge, there are no explicit references to such expressions anywhere in the Corpus. And yet Aristotle's positive insistence that ov has focal meaning entails that there must be such logoi. The best course, in the face of this puzzling textual hiatus, is to work backwards by first asking what these logoi must be like, and then determining if there are suitably Aristotelian phrases satisfying this characterization. What we are looking for, therefore, is a single phrase for each non-substantial category which (i) contains the "name" okvo'L, and which (ii) is interchangeable with ov in all of its applications within that category.

It will be useful at this point to return briefly to Owen's chronology. In order to defend his claim (Thesis I1) that the Eudemian Ethics signals no recognition that 'ov is a focal meaning term, Owen finds it necessary to explain the fact that at 1218 a 1-15 Aristotle already seems quite aware of the priority of Substance over the other categories. He quite correctly discounts this passage on the grounds that the sort of priority alluded to there is not logical priority, but another, "older" sort which is given the name "natural priority" [TC qPVUCL 'nrpoTepos] at Categories 14 b 13, and which is just the inverse of ontological dependence:11

(NP) x is naturally prior to y iff x can exist without y, but y cannot exist without x.

Hence, according to Owen, what led Aristotle to reverse himself on the question of the unity of science was his discovery that substances are not just naturally prior to other rvro, but are logically prior to them as well.

I believe that this account is substantially correct as far as it goes, but that it leaves out a crucial connection. On the view I propose, the two sorts of priority that substances enjoy over non-substances are not independent of one another, but actually represent two divergent attempts by Aristotle to express a single ontological intuition. More specifically, I aim to show that the same basic anti-Platonic insight which motivates Aristotle's claim for

122



the natural primacy of substances also supplies the substitutional machinery needed to demonstrate their logical primacy.

In view of the fact that the Aristotelian system doesn't countenance the possible existence of "naked" substances, it is doubtful that the asymmetric relation of natural priority is capable of fulfilling its intended function in Categories (I a 24-6), which is to ground the distinction between substances and the attributes which inhere in them. But this question aside, it is pretty clear what distinction Aristotle is reaching for: whereas it makes sense to speak of (primary) substances as existing simpliciter and not as modes or features of anything else, items in the other categories must always be conceived of as existing as qualities, quantities, etc. of substances. If this much is accurate, then the point Aristotle is trying to make about the whole range of non-substantial ̀v'a in Categories is just a generalization of the specific point he makes about health at Metaphysics A3, 1070 a 22, where he observes that "it is when the man is healthy that health exists." By employing the paraphrase technique given in Categories 2 for eliminating adjectival (or verbal) references to non-substances in simple affirmations in favor of nominal forms (whereby "a is F" becomes "Fness inheres in a"), we may represent the general point (which plays the pivotal role in Aristotle's attack on the Platonic characterization of the Forms as "separ- able" attributes) as a universal principle which equates the existence of non-substantial 6vTro with their inherence in substances. Where "Fness" denotes any non-substance,

Fness (is an) existent iff Fness inheres in some substance(s).

Now this schematic principle certainly sanctions a pattern of reductive translationl2 wherein every non-substantial application of "existent" can in favor of nominal forms (whereby "a is F" becomes "Fness inheres in a"), be replaced by the phrase "inheres in some substance(s)" (which establishes the logical primacy of Substance), but there is another respect in which it is defective. As it stands, the principle implies that there is just one meaning of "existent" applicable to all non-substances. But since the irreducible ambiguity of 6v is precisely what defines the ultimate status of the categories, this would reduce their number to two. It may be that such a streamlined categorial theory is not philosophically objectionable (cf. e.g. Leibniz' Substance and Attribute, and Frege's Concept and Object), but our texts show that it is certainly not Aristotle's (Metaphysics 1028 a 12-3).

What is missing from the principle given above, therefore, is some form of recognition that there is a distinct use of "existent" appropriate to each of the non-substantial categories. In addition, if these different uses of the

123



term are not to collapse into synonymy (by virtue of sharing a common logos), the correct formulation of this principle will need to incorporate the idea (reflected at Prior Analytics 48 b 24, 49 a 6-9, and Metaphysics A7, 1017 a 23-7) that each sort of non-substantial existence is equivalent (and reducible) to a distinct species of inherence-relation. All of this may be accomplished by means of correlated subscripts: if "Ci" is taken as in- dicating the non-substantial category into which Fness falls, it becomes possible to accurately represent Aristotle's anti-Platonic thesis by means of a schematic principle which entails the logical primacy of Substance:

(AP) Fness (is an) existentci iff Fness inheresci in some substance(s)

This principle establishes that 6v satisfies (FM) because it permits the replacement of any non-substantial occurrence, "existentci", with a discrete phrase ("inheresci in some substance(s)") containing the name "substance".

In the final section, it will be shown how (AP) combines with principles drawn from the Aristotelian theory of demonstrative science set out in the A nalytics to form the premises of the argument for the unity of science in Metaphysics F2.

The A rgumentfor the Unity of Science

Despite some fairly recent groundbreaking attempts in this direction,13 there has yet to be produced an account of the theory of demonstration which even attempts to place everything Aristotle says on this topic in a unified and intelligible analytical framework. Nonetheless, it will suit our rather limited present purposes here to work within an admittedly crude conception of Aristotelian science which contains two elements. First, a demonstrative science will be conceived of here as an axiomatic system of statements or propositions that proceeds from a relatively small set of "first principles" [&pQXa; which are assumed in that science, to a larger set of "theorems" which are proved in that science by means of (first figure) syllogistic inference. Among the &pxat of a given science are some, such as the Laws of Non-Contradiction and Excluded Middle (An. Post. 77 a 10-25), that are "common" [xoLvail to all sciences (but cf. 76 a 38-40), and others (the &M(L &pxao) which are peculiar to the science in question. Secondly, each science is uniquely determined by (i) a distinctive set of &pxaL', (ii) a distinctive set of entities it studies (i.e. its genus), and (iii) a distinctive set of attributes proper to these entities (henceforth its attribute-set).

Now there are numerous passages which indicate that for a science to be "one", in the sense of being autonomous (as opposed to being merely a

124



subordinate branch of some other science), none of its &pxxi can be subject to proof in any other science (An. Post. 87 b 1, and see also 75 b 16, 76 a 10). So, for example, it is intimated at 76 a 10 that harmonics is subordinate to arithmetic because some of its first principles are provable in that wider discipline. This may be represented as a definition of subor- dination

(S) Science S is subordinate to science S' iff any of the &apxa of S are provable in S'

On the present interpretation, the argument of Metaphysics F2 is aimed at showing that any science whose objects of concern are drawn from a non-substantial category must be construed as subordinate to a science of Substance. If such an argument could be made general for all the non- substantial categories, then since the categories represent an exhaustive classificatory system, that would elevate the science of Substance to a position where every science distinct from it could legitimately be viewed as subordinate to it. Therefore, let's arbitrarily select one of the non-substantial categories (Ci) and a science (Si) whose genus is subsumed by that category, and see how it might be argued on the basis of (AP) and (S) that Si is subordinate to the science of Substance.

The sole remaining premise of the argument will be extracted from An. Post. 76 b 3-7:

Also special to each science are those subjects whose existence it assumes li.e. its genus], and whoseperse attributes lattribute-set] it studies ... Of the elements in the genus both the existence and the meaning are assumed, but of their attributes only the meaning is assumed [while the existence must be proved].

This passage enunciates the following meta-scientific principle, which places general limitations on what can be demonstrated and what can be assumed within an Aristotelian science.

(L) Among the &pxai of a given science S are (i) statements which assert the existence of (some or all of) the elements of the genus of S, (ii) statements which define the genus of S, and (iii) statements which def'ine (all or part of) the attribute-set of S; however, (iv) statements which assert the existence of (some or all of) the elements of the attribute-set of S (i.e. that (some or all of) these attributes belong to (some or all of) the members of S's genus) are not &pxot of S (but are instead subject to proof in S).

With the acquisition of this principle, we have all the materials needed to reconstruct Aristotle's argument. On the assumption that the genus of Si is included in category C,, (L) necessitates that among the apxaL of Si are some which assert the existenceci of the members of C, belonging to that genus. However, since (by hypothesis) Ci is non-substantial, (AP) entails

125



that these existential statements are all reducible to statements which assert the inherenceci of those non-substances in substances. Now if these latter statements are provable at all, (L) requires that they be provable in a science (S*) whose genus is composed of substances, and whose attribute- set is subsumed under Ci. But since we now have the result that some of the &p-Xa of Si are equivalent to statements subject to proof in S*,(S) permits the conclusion that Si is subordinate to S*.

I should like to close with a disclaimer. It cannot and should not be expected that we can read off even the broadest features of the science of being qua being from the structure of the above argument. It would not, for example, be correct to infer that the science of being will necessarily have an immediately subordinate branch corresponding to each of the non- substantial categories, nor even that a science of substance would be capable of establishing the existence of all non-substantial oVTa. Such expectations emerge out of a mistaken idea of what the argument is all about. It is not intended to stand as a conclusive showing that there must be, or even that there can be, a science of being qua being. Rather, it is designed to meet and remove but a single a priori objection formulated in the Eudemian Ethics (1217 b 25-35) against the program of constructing such a science. The objection is that the theory of categories, and its corollary that there is no single genus containing 'roVTX Tar OVTXO, blocks the possibility of there being a science which studies all existents. Aristotle's response, as I have interpreted it, is to argue that the theory of categories is consistent with the possibility of a universal science of being, if by that is meant a science of being to which all other sciences of beingci are subordinate.

But Aristotle is fully aware that the success of this argument would not by itself close the issue in his favor. This can be seen from the fact that Metaphysics B 1 ackowledges at least twelve other prima facie objections that would have to be overcome before the possibility of a universal science could be guaranteed.

The critical point is this: by choosing in Metaphysics F2 to meet this objection on theoretical grounds, he opts against the alternative strategy (which is employed in Metaphysics Z-9) of revealing enough of the science of being qua being to allow his audience to see how it circumvents his earlier objections. In making this choice, he avoids (for the time being, anyway) the need to commit himself on the questions of how precisely the science of being qua being is structured, and whether this structure qualifies the study in all respects as a legitimate demonstrative science.

126



Indeed, it is entirely consistent with what has been said here that Aristotle was not yet in possession of answers to these questions when he composed Metaphysics F2.

Washington State University

NOTES

G. E. L. Owen, "Logic and Metaphysics in Some Earlier Works of Aristotle", in 1. Ditring and G. E. L. Owen (edd.) Aristotle and Plato in the Mid-Fourth Century. Gdteburg, 1960. 2 I am using the term "application" here and below as a theoretically neutral expression which should not be uncritically equated in meaning with such metaphysically "loaded" terms as "sense" or "meaning". I argue below (p. 118) for a specific view as to how the application of an Aristotelian term should be characterized. 3 Owen, "Logic and Metaphysics", p. 167. 4 The sorts of things that can serve as significata include (I) primary substances (concrete individuals), (2) non-substantial particulars (individual qualities, quantities, times, places, etc). and (3) the genera and species which contain things of these sorts. (1 make no claims here as to whether these genera and species are intensional entities, or simply classes of entities of sorts (I) and (2).) Also included are even more bizarre "entities" such as (4) differentiae (e.g. two-footed) and (5) "compounds" 1[auv8va6pEval such as white man. s I am persuaded by the argument in J. Owens, The Doctrine of Being in the A ristotelian Metaphysics, Toronto 19632, p. 117-8, that irp6s Ev and &p' kv6s are meant to designate the same condition. Unfortunately, they are also equally obscure. 6 Metaphysics A6, 1016 a 33 - 1016 b 7. 7 Categories 10, 14 a 25 - b 24; Metaphysics All, 1018 b 9 - 1019 a 14. 8 Cf., e.g., Physics 227 a 19. 9 Aristotle's procedure here isprimafacie open to an alternative interpretation according to which he uses Bvolua to stand only for genuine nouns (as opposed to noun-phrases like To6 LQTpLx6v), and consequently sees xarp&s as the only name which applies to "medicals" of any sort. On this view, being signified by an 6vopa is a necessary condition for service as the "focus" of a %Tp6s Ev term, so that the kind doctor acquires this position in the case of laTpLx6v by default. This interpretation is suggested (if not entailed) by Ross' general

view (Aristotle's Metaphysics, vol. I, Oxford 1924, p. 256) that the primacy involved in focal meaning ambiguity amounts in the end to nothing more than the grammatical or etymological priority alluded to in the explication of paronymy given at Categories 1, I a 12-15.

To go along with J. Owens' general argument against this way of understanding focal meaning ambiguity (The Doctrine of Being, p. 111), there are three special reasons for resisting its application in this instance: (1) it makes all of the logos-inclusion talk at a 22-3 idle: if the Ross view were correct, it would be hard to see why Aristotle doesn't simply define the primary "medicals" as those to which the name 6vo%a applies; (2) taking such a restrictive view of bv6paira leaves Aristotle with no grammatical category to cover such expressions as Tr 'LTpLxOv; (3) most importantly, the Ross view presents a rather

127



unflattering picture of Aristotle attempting to derive significant metaphysical and meta- scientific conclusions from observations that he must have seen were due to the accidents of etymology. This last point becomes especially clear in the case of "existent", the focal meaning analysis from which he draws his most important conclusions. It is utterly inconceivable that Aristotle could have placed the entire weight of his argument for the unity of science on a perceived grammatical priority of the noun ovaia over the adjectival form 6v, failing all the while to see the obvious fact they are both participial derivatives of the verb EvaL. 10 But cf. Metaphysics H2, 1042 b 26, where Aristotle says that E'VatL (and, by implication, ov) means something different when applied to thresholds than it does when applied to ice. This suggests that there are distinct uses of 5v for different sorts of substances, and perhaps for different sorts of attributes within a given category as well. Owen explains this discrepancy by suggesting that the different kinds of existents within a category could 'e construed by Aristotle as subspecies of a single kind, but that this would still leave him with an irreducible and insubordinate species of existence for each of his categories ("Aristotle on the Snares of Ontology", in R. Bambrough, New Essays on Plato and Aristotle, New York 1965, p. 73, n. 2). 11 Owen, "Logic and Metaphysics", p. 172. 12 This pattern of reduction is hinted at (though certainly not developed) at Posterior Analytics B2, 90 a 17, where Aristotle seemingly equates the questions, "Why is there an eclipse?" and "Why is the moon eclipsed?". 13 J. Barnes, "Aristotle's Theory of Demonstration" in J. Barnes et al. (edd.) Articles on Aristotle, vol. 1: Science, London 1975, pp. 65-87; J. Hintikka, "On the Ingredients of an Aristotelian Science" Nous VI, 1972, pp. 55-69.

128



aristotle on focal meaning and the unity of science

Documents

sort aristotle

earlier works of aristotle

jstor terms

jstor archive

aristotle number

case of systematic ambiguity

terms conditions of

pmall use subject