Aristotle’s account of definition

On the one hand, Aristotle is clear that his account of definition is only appropriate for expressing the essence of physical beings (and mathematical things). On the other hand, the only natural thing that we can define perfectly is the human animal, who, oddly enough, is formally definable only through a power that Aristotle repeatedly proves is non-physical. Reason, the one positive, formal specific difference in a physical substance is not physical. What to make of this?


1.) We could see the genus/ species structure not as biological but as anthropological. Man alone is defined because definition as such is only appropriate to the ontological structure of human persons. We have only found one definition because the criteria of termination for the thing can be satisfied only at the limit of nature and not at the subordinate grades.

2.)   We could see definition as requiring introspection. Only what is present to itself can define itself. Perhaps each choir of angels can define itself, the lower choirs, and human persons. By this account, persons can define only themselves and what they themselves give formal being to: artifacts and (perhaps) mathematicals.

2b.) We could see all causes as known only by introspection, and that definition as a sort of cause. This is my idiosyncratic reading of Hume, who denies the possibility of knowing causes in the first book of the Treatise (which seems to focus on our knowledge of nature) and then speaks about causes ad nauseam in the second book (which deals with the sort of moral facts one can know by introspection.) I’ll continue to believe this crazy idea until Brandon refutes it.

3.) We could see man as a paradigm for what is definable. Banez seems to understand man in this way when, in his commentary on the Fourth Way, he objects that the premises of the argument would mean that man is the cause of all animals, but then proceeds to simply say that man is such a thing.

4.) We could just deny that we lack definitions of other natural things. True, we have no good definition of any animal or plant, but we have a pretty good definition of malaria, diabetes, acceleration, etc. But this seems just wrong. We don’t have a good definition of a malaria parasite, only an account of how it affects us and some general essential facts that fall short of a species making difference. For acceleration, see #2.



  1. May 30, 2014 at 9:21 pm

    It seems as reasonable a hypothesis for the difference between Book 1 and Books 2&3 as anything else I’ve seen — there’s a sense in which Hume holds that we can have knowledge of causes in Book 1, but it’s in something very like the introspective sense.

  2. Paul said,

    May 31, 2014 at 6:38 am

    Sir, the difficulty that you are addressing here is not apparent to me. We start with a substance, a “this,” and obtain the definition by separating its predicaments, so that at the hypothetical perfection of this process we have just “substance” and a very long list of attributes, some of which are essential attributes or properties, others of which are accidents. If we take as our “this” a particular physical being, then the first act of the intellect will be to abstract some universal out of it, and the “level” (or degree) and type of abstraction we perform will determine which attributes are taken as essential and which as accidental. (Clearly I am distinguishing definition from scientific demonstration — but not in a way, I think, which is inconsistent with Aristotle.) The limits of the powers of our senses to determine its sensible predicaments, and of our intellect to discern its form, relations, etc., will limit the perfection of our definition.

    If I were somehow to perform absolutely no abstraction, to skip the first act and proceed directly to the second — which is not something I think possible for the limited human mind, as supported by our limited senses — then every single (perhaps numerically infinite?) predicament would be of a property “essential” to that particular thing at that time in that place, etc.; that is, all of those predications are required to define perfectly what makes that single substance “what it is,” right now, in every detail. Or alternately, considering that the affirmation or denial of genuses (sp?) or species could be conceived of as being “attributes” in some sense, I may have to include all of these in my perfectly individual definition — again, not something we can do. Maybe the divine mind can operate in this manner, from the maximum level of abstraction all the way down to its total absence, making all true predications in one intellectual act.

    But probably the fault is my own; you did start out by saying that “Aristotle is clear that his account of definition is only appropriate for expressing the essence of physical beings (and mathematical things),” and while I don’t have the passage at my fingertips, I must bow to your greater erudition. Still, as I read/recall relevant portions of On Interpretation and the Posterior Analytics, while the examples are drawn from physical and mathematical things, I’m not sure that Aristotle limits the applicability of his thought to these. Again, given your scholarship, I’m sure that you have good reasons for thinking that he does, but then again what Aristotle thought is one thing, and the truth that we can see in it another. Perhaps I am wrong on both counts.

    • May 31, 2014 at 9:04 am

      The problem is that, among natural substances, which is exactly what definition is made to describe, the only known definition is for the human animal. So why did we define “man” right away, and even “animal” right away, but, after many thousands of years of association, we can’t define anything else (say, a cow or a dog or a mosquito)?

      • thenyssan said,

        May 31, 2014 at 9:14 am

        What, you can’t give me a one-word specific difference for bear? Why not? 🙂

  3. thenyssan said,

    May 31, 2014 at 7:20 am

    I need to understand the Banez claim more. I remember thinking this the last time you brought it up–it’s a staggering claim that pushes natural theology way beyond a limit I would have guessed. Incredible harmony with the Genesis account.

  4. davidus said,

    June 2, 2014 at 1:54 am

    What’s the uncertainty, e.g., with defining hydrogen as possessing genus of atom and specific difference of possessing one proton? (David Oderberg uses this example and suggests that, while this definition is of course falsifiable, we currently have no good reason to think that doesn’t give us the essence of hydrogen). It doesn’t seem to fall into the category of #2 either.

    • June 2, 2014 at 12:14 pm

      Whether elements are substances is an open question, but it seems reasonable to take this as an account of the essence of the thing. That said, the sheer numbering of particles is more material than formal – orbitals might be a better account of the thing. But then the contingent character of the orbitals makes it harder to speak of this as formal.

  5. D.S. Thorne said,

    June 2, 2014 at 10:45 am

    *That* there are essences in general seems fairly straight-forward to me. *What* the essence of any particular thing is strikes me as the hard, if not impossible part.

    Can you cite the passage in Aristotle where he says that man is the only natural thing we can define?


    • June 2, 2014 at 12:07 pm

      Re: that and what. Even if you allow for essences, it is another thing to say that definition is capable of getting hold of it. Definition is one account of how to attain to essence.

      Aristotle never said that we only knew the definition of one substance. Those are just the facts on the ground. A. didn’t even favor giving “rational animal” as a definition, preferring instead to speak of “bipedal” as a difference, even though he certainly knew it was not an adequate difference.

      • thenyssan said,

        June 2, 2014 at 12:43 pm

        At the very least, we don’t have a definition of any other living thing. See my joke above: what’s the specific difference of a bear? Or consider the angels, whether we follow St. Thomas on them or not. What’s the specific difference of serpahim (or, for STA, each angel)?

  6. Maureen said,

    June 3, 2014 at 2:26 pm

    An old book on my parents bookshelf, (iirc, ‘Readings in Ethics’ St. Thomas Aquinas, postulates,virtues, highest good; references to Adler, Hume, etc) led me eventually to getting a ruler and paper to exercising primitive renderings of perfect lines and circles in somewhat of a way of what I was reading. I have no formal higher education in physics. Going back to reading some of the passages, now years later, definitely tells me that the substance that moves things is (still) of lengthy debate these days. It brings back the passages of unmoved movers, heavenly bodies and all that the book has frozen in time my knowledge bank. Never knew until the last year or so the relationship to physics! One thing I note is that Aristotle as quoted in this book, never used the word symmetry or asymmetry; Or dimensions, per se. He didn’t use 5th elements as much as what I read now was quintessence. Guess I am a time capsule of an old book 🙂 Interesting!

%d bloggers like this: