Substance as Essence Substance and Definability
The Z 3 Alternatives Substance is spoken of if not in more senses, still at least in reference to four main objects; for both the essence and the universal and the genus are thought to be the substance of each thing, and fourthly the substratum. Now the substratum is that of which other things are predicated, while it is itself not predicated of anything else. And so we must first determine the nature of this; for that which underlies a thing primarily is thought to be in the truest sense substance. (Met. 10228b33-1029a2)
A Tidy Disjunctive Syllogism Depending on our answer to that question, we might understand Aristotle s procedure in Metaphysics Z as follows: 1. Substance is either: (i) essence; (ii) universal; (iii) genus; or (iv) substratum 2. Not (iv): Substance is not the substratum (Met. Z 3) 3. Not (i): Substance is not essence at least not where essence is understood in one sort of way (Met. Z 4-6, 10-11) 4. Not (ii) (and so not (iii)): Substance is not universal (and so not genus). So, if there is substance, it must be something other than (i)-(iv) at least as (i)-(iv) have been understood by their proponents. This is why Metaphysics Z 17 makes a fresh start.
Alternative Alternatives Aristotle s statement of his agenda is stunningly vague in many ways Lewis (2013, 16). Aristotle may be laying out: four alternatives, namely: (i) essence; (ii) universal; (iii) genus; or (iv) substratum or two alternatives, namely: (i) essence, as universal and genus; or (ii) substratum or just possibly, no alternatives, but rather: the essence, which into say, what is universal and generic, serving as the basic subject. One further question: is this list meant to be, from Aristotle s point of view, exhaustive of the possibilities?
A Disconcerting Triad? Or at least a heuristic triad for the wary? 1. No universal can be substance. (Met. 1038b8-9) 2. The form is a universal. (Met. 1034a5-8 (by implication); De Gen. An. 730b35; De Part. An. 644a24-25; cf. Met. 1038b11-12, 1036a28; Cat. 17a40- b1) 3. Form is substance. (Met. 1032b 1-2, 1033b 17, 1037a5*, 1037a27, 1041b 6, 1050b2)
Metaphysics Z 4-6 All concern essence. Some putative results: Z 4-5: Nothing everything has an essence, or, rather, not everything has an essence in the primary way. Only entities in the category of substance have an essence (in the primary way). Z 6: In such cases, the things having essences are somehow to regarded as being the essence they have.
Let s Speak Logikôs After discussing the subject (or substrate; τὸ ὑποκείμενον) in Z 3, Aristotle returns to essence (τὸ τί ἦν εἶναι) in the opening of Z 4, suggesting: Since we distinguished at the outset the number of ways we define ousia and since one of these seemed to be essence, it is necessary to make a study of it...let us first speak logikôs (λογικῶς) concerning it, seeing that the essence of a thing is that which is meant in its own right. (Met. 1029b11-14) ROT: And first let us say something about it in the abstract. Bostock: And first let us make some logical remarks about it. OOT: And first let us make some linguistic remarks about it.
Speaking λογικῶς Some questions: What sort of speaking is this? With what is it contrasted? How might we speak otherwise? After first speaking this way, do we begin to speak in some other (better, worse, merely different) way? Why should we speak first in this way? Is there some reason why we speak first in this way about essence (τὸ τί ἦν εἶναι)? We did not speak this way at all in Z 3 about the subject/substrate (τὸ ὑποκείμενον).
On Speaking λογικῶς Ross (vol. 2, 168): λογικῶς suggests plausibility rather than truth..., dialectic or sophistic as opposed to science..., a reference to the abstract considerations (λόγοι) rather than to the precise nature of the facts in question... Usually its sense is deprecatory... It probably always refers to linguistic inquiries or considerations... It is in 1030a28 that the real as opposed to the verbal inquiry begins. Ross (vol. 2, 171): λογικῶς, i.e. that which is cannot can said to be in the plain sense of that word, but speaking λογικῶς, with reference to linguistic usage (οἱ λόγοι cf. πῶς δεῖ λέγειν, l. 27), we may say that it is, since we can say that it is non-existent. Angioni (2012, 37): The logikos standpoint is a preliminary discussion of some logical and formal features of essences and has no metaphysical depth.
In Sum Plausible vs. True Dialectic/Sophistic vs. Scientific Abstract vs. Precise (?) Deprecatory vs. Laudable (as explanatorily fecund) Linguistic vs. Material Mode Merely verbal vs. Real Superficial vs. Metaphysically deep
Deprecatory (?) But if it is necessary to speak concisely about these matters, we say first that there being an idea not only of the the good but of just any thing whatever is said λογικῶς καὶ κενῶς (ROT: abstractly and idly ) this has been considered in many ways both in our popular and in our philosophic writings. (EE 1217b19-23; cf. DA 403a2 διαλεκτικῶς καὶ κενῶς)
Mildly and Slightly More Puzzling Mildly puzzling: Given that we are now turning our own attention to the question essence (τὸ τί ἦν εἶναι), why should we do what we have elsewhere castigated the Platonists for doing? Slightly more puzzling: However that may be, since we are doing this first, where does our λογικῶς speaking leave off? Three proposals: Ross: Z 4, 1029b13-1030a27 Woods: Z 4, 1029b13-22 Burnyeat: We are speaking λογικῶς for the entirety of Metaphysics Z 4-6
The Dominant Contrast The most repeated contrast is between arguing λογικῶς and other forms of argument is not intended as a contrast between (say) captious and legitimate, or verbal and substantive......but rather between what is is proper (or what belongs; οἰκεῖον) to a subject matter and what is not (APo. 83b35, 88a7-30, 93a15; Phys. 202a2, 204b4-10; Met. 1005b22, 1041a27-28; EN 1147a24. Does this mean that speaking or arguing λογικῶς is improper, in some sense? Proceeding λογικῶς involves our making meta-level observations about the characters of various deductions and then inferring a perfectly general, perfectly apt feature of deductions as a whole. Northing improper about all that. Rather, we prescind from the domain of discourse in order to draw (constraining) inferences about the behaviour of predicates in the domain
A Proposal S is speaking λογικῶς with respect to domain Δ with defining predicates φ 1... φ n iff : 1. S is speaking about items in Δ; 2. S prescinds from φ 1... φ n in the sense that S is not employing the predicates of Δ (as opposed to merely mentioning them); 3. S is none the less presenting arguments with ramifications for how one should or must deploy φ 1... φ n in Δ.
So, what is an essence? The essence of each thing is what it is said to be in virtue of itself. For being for you is not [the same as] being for something musical; for you are not musical in virtue of yourself. What, then, you are in virtue of yourself is your essence. (Met. 1029b13-16) So it seems: φ is an essence of x iff x is φ in virtue of itself Presumably, then, φ is an essence of x when φ is predicated per se of x in the first of the types of per se prediction of in the Posterior Analytics i 4 (APo. 73a34-b5): φ is predicated per se of x if (a) φ is predicated of x; and (b) φ must be mentioned in an [essence specifying] account of x.
Too Broad? Yes and No No: Met. 1030a17-27: Entities outside the category of substance can have essences. Yes: Met. 1030a27-b6, 1031a1, 1031a11-14: But they cannot have them in a primary way; only beings in the category of substance have essences in the primary way.
One Approach to this Distinction Extensionally: x is essentially φ in the primary way iff: (i) φ is predicated per se of x; and (ii) φ is a substance sortal x is essentially φ in the relaxed way iff: (i) φ is predicated per se of x; and (ii) either (a) φ is a substance sortal but x is not some this (tode ti), or (b) φ is not a substance sortal But this doesn t yield something non-trivial until we have a clear account of substance (ousia); for without that, we re at sea about which sortals are substance sortals.
First Approximations Sortals are those predicates which: provide counting criteria for instances of a kind provide criteria for both synchronic and diachronic identity for the instances of a kind Substance sortals then also do so in such a way that they: specify the MLI natures of those instances specify the MLI essences of those instances
Essences? Essence 1 : simple de re necessity. Essence 2 : binding de re necessity. Essence 3 : sortal necessity.
Essence 1 : simple de re φ is an essential 1 property of a = df not possibly a exists and is not φ So, for example, the number 6 is necessarily even. Or, Alcibiades is necessarily human. This cloak is a both pale and human?
Essence 2 : binding de re φ is an essential 2 property of a = df (i) a is essentially 1 φ; and (ii) anything which has φ has φ necessarily. So, e.g., according to essence 1, one might say, this human is essentially 1 φ (say, rational), but this android is contingently φ. On this second conception: being essential binds an entity to its kind. Further, it is not possible that we have any φ such that φa and φb such that a is essentially φ while b is contingently φ. Compare, then, being human and being red It seems that magenta is essentially (or at least necessarily) red; Olaf s barn is now red and now white. Perhaps, by contrast, whatever is a human is essentially a human; nothing is such that it is now a human and now a barn (or a fish or a doorknob...)
Essence 3 : sortal necessity φ is an essential 3 property of a = df (i) a is necessarily sorted up by φ; (ii) a is necessarily sorted down by φ; and (iii) a is not sorted across by φ. sorted up: a is φ, and is not a proper part of anything that is φ. sorted down: a is φ, and has no proper part which is φ. sorted across: possibly, some things are not-φ.
Aristotle on Sortal Necessity...for it is impossible for a thing still to remain the same if it is entirely transferred out of its species, just as the same animal could not at one time be, and at another not be, a man. (Topics 125b37-9) This implies (or near enough states), that nothing can switch from one essential kind to another.
An Astonishing (Sounding) Claim The opening of Metaphysics Z 6: We must inquire whether each thing and its essence are the same or different. This is of some use for the inquiry concerning substance; for each thing is thought to be not different from its substance, and the essence is said to be the substance of each thing. This is the argument: 1. Each thing is not different (= is the same as) its substance (ousia). 2. Essence is said to be (=is the same as) the substance (ousia) of each thing. 3. So, each thing is the same as its essence.