So, searching for J. L. Mackies "argument from queerness", I stumbled upon this blog entry.

Now, Mr. Feser seems to be a quite... controversial figure, to say the least, but please let's resist the temptation to simply dismiss him.

He writes:

[...] a Euclidean triangle drawn slowly and carefully with a ruler is a good one while a Euclidean triangle drawn sloppily is a bad one; and so forth. The core idea is that of a good or bad specimen of a kind of thing, of something which more or less adequately instantiates what is of the essence or nature of the kind.

I find this usage of "good" quite weird and I am very suspicious of the claim that it is possible to get from this kind of "good" to moral goodness. But let's leave all that aside.

The fundamental question is how this "essentialism" (Aristotelian, Thomist) can ever make sense:

  • How to find out what the "essence" of a particular thing is?

  • How to find out if a particular thing approximates its essence well? It may simply have a different essence...

I've drawn the following example, which hopefully makes the problem clear:

enter image description here

Now what is the green figure? What is the gray figure? Does the green figure instantiate its essence, which is "being a Reuleaux-triangle", near perfectly? Or is it instantiating its essence, which is "being a circle", extremely badly? Similarly for the gray figure.

I can't be the first person who sees this very obvious and in my opinion very serious problem with postulating "essences".

There are probably counterarguments to it. Does somebody know any?

3 Answers 3


One could argue for essences along the following line: Consider a well-designed hammer and screwdriver. Each of them is perfectly fit for its own respective job. Next, join the two tools into one: a hammer-screwdriver. This new "tool" is a kind of misfit. It does not perfectly fit any job.

What is the essence of the hammer-screwdriver? Is it a hammer? A screwdriver? Something else? No. It has no essence, because it does not fit. It is a misfit, an essence-less lump, rather than a full-fledged individual entity, a substance.

Aristotle seems to have argued in a similar way. For example, at the beginning of the Politics:

Now nature has distinguished between the female and the slave. For she is not niggardly, like the smith who fashions the Delphian knife for many uses; she makes each thing for a single use, and every instrument is best made when intended for one and not for many uses.

  • So we see what Aristotle would have replied. But for a modern reader this serves more as a reductio ad absurdum of his metaphysics. Quite a pompous ontology where "femaleness" is a fundamental part of reality, or worse "slaveness"... which is completely crazy.
    – Ystar
    Commented May 17, 2015 at 9:22
  • 1
    @Ystar Right, essentialism is rightfully disparaged in modern philosophy (and culture). Still, Platonic and Aristotelian philosophies were held by many intelligent and decent persons, so it better, imho, to try and treat them with respect. Commented May 17, 2015 at 10:45
  • essentialism is not "rightfully disparaged in modern philosophy." Cf. Brian Ellis, "Scientific Essentialism" Cambridge University Press, 2007 or any of the ~450 papers cited here: philpapers.org/browse/essentialism
    – user5172
    Commented May 18, 2015 at 15:20
  • @shane Hi. Did you -1 my answer? I see that you didn't like my comment, but the comment is independent of the answer Commented May 18, 2015 at 16:45
  • Probably it would have been better to write "Aristotelian/Thomist essentialism", Brian Ellis writes: "It is not a reversion to Aristotelianism, or an attempt to resuscitate medieval views about the nature of reality." Because its one thing to believe in essential properties on the most fundamental level, where quantum mechanics gives us "discrete distinctions" (e.g. one might say that spin 1/2 is one of the essential properties of an electron), and another thing to believe in "femaleness" and "slaveness".
    – Ystar
    Commented May 19, 2015 at 14:45

In both the Aristotelian tradition and the tradition of contemporary metaphysics "essences" are discovered empirically. An essence is the thing expressed by a definition of a kind of object. Empirical science is in the business of discovering such definitions. (I'm bracketing here some difficult questions about biology--it's not clear ALL sciences are about finding essences, just that some sciences are concerned with this.)

Here's an example of the kind of thing that I have in mind:

"Gold is the element with atomic number 79." "Water is H2O."

These are genuine statements of the essence of a kind, in the sense that these definitions give necessary and sufficient conditions for a particular individual being a member of that kind. If an atom doesn't have 79 protons in its nucleus, it's not a gold atom. If a molecule doesn't have exactly two atoms of hydrogen and one of oxygen bound together in the right configuration, then it's not a water molecule.

If this is right, then there's nothing particularly difficult about getting to know some essences, since none of those sentences are controversial as items of empirical knowledge.

  • 1
    The question about approximation you ask is also important. The aristotelian/thomist line is going to be that essences are connected to functions. This, incidentally, explains the connection between essence and goodness in that tradition--x is a good F if and only if it fulfills the function of Fs excellently. Aristotle will say that rationality is part of the human essence, and therefore to be human is to act in accordance with reason, and so there is a clear sense in which somebody who acts reasonably is "good" and someone who doesn't is "bad."
    – user5172
    Commented May 15, 2015 at 16:36
  • I think your answer completely dodges the question. First, gold is a general term, but the question was about particular things. Then, how do we empirically find the essence of something? It is true that science has refined many terms which were already in use long before, but it is not clear that this gives us an objective essence. If we would say that heavy water is essentially different from normal water, would that be really wrong? After all, heavy water is toxic.
    – Ystar
    Commented May 15, 2015 at 18:37
  • "gold" is like "circle": it's a general term that has particular instances. There isn't a general answer to the question: "How do we know x is an F?" because there are different methods for different kinds of things. We know that circles are sets of points equidistant from a center because that's a stipulation. But obviously, that procedure, which is appropriate in math, is not appropriate in chemistry. There won't be a simple (i.e. SE length) answer to such a question.
    – user5172
    Commented May 15, 2015 at 18:38
  • 1
    Regarding functions, no simple figures won't have them. But with living things, especially there are going to be many clear cases. "Canine teeth are for ripping and tearing food." "The function of the cellular membrane is to keep enzymes concentrated in a higher quantity inside the cell than outside." and so on. Now it's another leap to get to Aristotle's claim that the function of a human is to act with reason. (That's a separate subject of its own.) But, hopefully it is clearer what the Aristotelian tradition is talking about--the toy cases of shapes aren't really the best examples.
    – user5172
    Commented May 15, 2015 at 18:42
  • "gold" is not like "circle", one never says "a gold". Also you introduce a further complication with the mathematical definition of "circle". If you use this definition, there simply are no real circles in the physical world. But surely gold, defined as the element with atomic number 79, exists in the physical world. In short, you simply confuse questions about the definition of a property with questions about when a particular thing exhibits a certain property.
    – Ystar
    Commented May 19, 2015 at 14:09

This is, in a strong sense, the basic issue in Platonic philosophy. And the basic answer is the concept of 'participation', there are Forms of important Ideas and the remainder of the variation and concrete instantiation is a compromise with matter and perception.

The Platonic argument against the kind of thing you are talking about is that humans are not necessarily very good at knowing what kinds of things have Forms. Socrates jokes with Parmenides about whether there is a Form of Matted Hair, if so there would be a perfect immaterial sort of matted hair despite that to some degree matting is a completely material effect, and a flaw of hair. (Rastafarians notwithstanding.)

Clearly with the basic notion of equivalence classes and degrees of equivalence, 'essences' can make sense. Informally, mathematics clings to the notion of essence quite strongly. We know what a triangle is, and we have an idealised notion that captures the behavior of ideal triangles. Drawn triangles capture that essence to the degree they allow us to correctly imagine and reason about triangles. And we do so, so there is some element of reality to the essence of a triangle, and even to the essence of one with a given set of side-lengths and angles. From that, it seems to be reaching a bit far to insist this basic approach cannot be meaningful or productive.

I some sense, to the degree physics abandoned this notion completely when it accepted atomism, psychoanalysis brought it full circle with the notion of signification and the projection of the master signifier. Meaning, in that model is constructed out of combinations and interactions between basic socially or developmentally derived references that we assume have a shared content for all of us. So we have again, a notion of basic essential references, and everything else is seen as an interaction between them and a compromise with the material instantiation and perception. But we presume a somewhat smaller set of ideals, and a lot more construction. We are still not very good at narrowing down what should and what should not be fixed, shared, basic signifiers.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .