I read David S. Oderberg’s book Real Essentialism (2007) (of which David kindly sent me a copy) already some time ago and I’ve been meaning to write a review. The book, as the title suggests, is a study and defence of neo-Aristotelian essentialism — a topic which overlaps with my own research interests. David is also one of the contributors in my forthcoming Contemporary Aristotelian Metaphysics volume. More news about the volume will follow soon…

Anyway, on to the book. Real Essentialism is a rich book and I cannot hope to do justice to it in a short review, so I will focus on one or two issues which are of specific interest to me. These are covered mostly in the first half of the book, where the role of essentialism in contemporary metaphysics is analysed, different anti-essentialist views are discussed, and the epistemology and structure of essence are examined. Oderberg goes on to discuss applications of essence concerning identity and existence, which are fairly straight-forward, but also life, species and the person. The latter form a topic of their own and I found the discussion of the species concept especially interesting, but the conclusions drawn about the essence of life and personhood are certainly among the more controversial ones. I will however omit a detailed discussion of them and focus especially on the epistemology of essence, which is a topic that I’ve been thinking about recently.

An important clarification concerning the epistemology of essence, which Oderberg makes in the very beginning of the book, concerns the baggage from the work of Kripke and Putnam and especially the reductionism that it would seem to entail: essences concern the ‘internal structure’ of, say, chemical substances, and they can only be discovered a posteriori, typically by empirical research conducted by scientists, i.e. ‘experts’. Oderberg notes, correctly in my opinion, that:

Scientists play an indispensable role in helping to explain the real essences of things — and for some kinds of entity, the ones proper to those fields of science requiring more or less elaborate technical devices or measurement and experiment, their role may be exclusive. But it is incorrect to hold that the job of the real essentialist just is the job of the scientist. It is also, and primarily, the job of the metaphysician informed by science, and additionally, for many kinds of entity, the job of everyone, expert or not. (p. 13)

David Oderberg's Real Essentialism

This passage emphasizes a point which I’ve also made repeatedly: while metaphysicians need to be scientifically informed, expert knowledge of science is not necessary to be able to grasp essences. Similarly, while training in metaphysics may help to acquire more detailed knowledge about essences, everyone who is capable of rational thought is also capable of grasping essences. Indeed, this is a precondition for rational thought. Oderberg discusses the implications of this in some length, especially in the case of chemical substances. I agree with the general line that he takes, but the details require more work. The case of chemical substances in particular is problematic, and I believe that the traditional examples (water, gold etc.) familiar from the work of Kripke and Putnam are still often taken at face value, despite the devastating criticism already familiar from van Brakel’s 1986 paper, ‘The Chemistry of Substances and the Philosophy of Mass Terms’ (Synthese 69: 291–324). Well, this is a topic which I discuss in detail in a draft, which I’ve just revised (partly with the help of David’s comments): ‘On the Metaphysical Status of Modal Statements’.

Later on, Oderberg makes another specification in passing, which I also support: our epistemological access to essence does not rely on any ‘special insight’ or ‘intellectual intuition’, but is rather a combination of perceptual information and intellectual abstraction from that information (via a consideration of form, according to Oderberg) (p. 31). On related note, this makes essentialism a fallibilist position — something that is often overlooked in anti-essentialist accounts.

Regarding the structure of essence, Oderberg defends the view according to which essences are ‘not mere bundles of essential features’, that is to say that we cannot reduce the essence of an entity to a bunch of essential properties of that entity. I understand that this line comes from Aristotle and I can see why one might want to avoid reducing essences to bundles of properties, however, I would still like to be able to list the essential properties of an entity with the intention that, although the list may be incomplete, it will be sufficient to explicate the essence of the entity in question. Now, this does not have to mean that the essence itself is just a bundle of such essential properties, but my worry is that if more is required to explicate essence, then it will start to look as if the essence is an entity of some sort. That essence is an entity itself is a result to be avoided, as E. J. Lowe has pointed out time and time again: it leads to an infinite regress. In any case, Oderberg does offer an argument in support of his view:

Having a capacity for humour is an essential property [...] of human beings, and in this sense we can say it flows from the essence of human beings to have a capacity for humour. But the essence of being human is to be a rational animal, and humans have a capacity for humour only because they are rational animals. (p. 49)

This seems correct insofar as we need to distinguish between the essence of an entity and what that essence may entail. But the term ‘flows’ is unfortunate. It is a historical notion which is being adopted into contemporary discussion for want of a better notion; I’ve heard Lowe use it as well. But it sounds very mysterious. Does it simply suggest that the essential properties of an entity are entailed by the essence of that entity? What exactly is the relationship between essence and essential properties? My concern is that it will be impossible for us to distinguish, epistemically, between the essence itself and an essential property that ‘flows’ from the essence. For instance, if we are looking for the essence of water, we will presumably be able to point out a number of essential properties of the H20 molecule, perhaps, say, its molecular weight, but which essence, exactly, will these properties be related to? The essence of the natural kind ‘water’, if it indeed is a natural kind? Or the combination of the essences of hydrogen and oxygen atoms? The distinction between essence and property is further discussed in chapter 7 of Oderberg’s book, and I will return to this shortly.

The problem that I envisage is that we must already have grasped the essence that we are looking for before we can identify the essential properties that ‘flow’ from it. But it seems to me that our epistemic access to essences is often piecemeal: we wonder if a certain kind of entity, say, the Higgs boson, is possible, and we determine this by considering the essential properties that the Higgs boson would have, its mass and so on. But since we acquire knowledge of real essences by a combination of intellectual abstraction and empirical information, it would appear that we can grasp a real essence only after we have acquired sufficient knowledge about the essential properties associated with that essence. This is an epistemic rather than ontological worry, since it may still be the case that real essences are something over and above the essential properties associated with them (although they better not be entities in their own right, as noted above), but it seems like a difficult problem to overcome without relying on some more or less mysterious cognitive capacity which enables epistemic access to real essences. Oderberg does not appear to be committed to anything of the sort, so I’m not sure if I see how the epistemic story is supposed to go in his account.

Something related to this is discussed in section 3.4 of the book, ‘Coming to know essence’. Oderberg for instance notes that knowing the essence of something does not require knowing all of it, or knowing it in detail, rather, knowing a part of the essence may be sufficient, and indeed often this is enough for the purposes of distinguishing one entity from another. However, it seems to me that this is precisely the story we might tell about the essential properties of an entity — knowing some of the essential properties of an entity is sufficient for identifying it, while the full essence which encompasses all the essential properties of the entity may remain elusive. In fact, on p. 56 Oderberg says that ‘In general, it is true to say that we mostly identify and come to know the essences of material objects indirectly via their properties and accidents’. The case of abstract objects such as mathematical entities may be somewhat different, but all this seems to concern our knowledge of the essential properties of entities rather than the elusive essences themselves.

Well, my reading of the matter may not do full justice to Oderberg, as he does go on to explain, in considerable detail, how this view of essence fits in with the Aristotelian theory of hylemorphism, according to which essences are a mixture of form (actuality) and matter (potentiality) (p. 65). These notions are part of a long Aristotelian tradition and they certainly add considerable depth to the story about the structure of essence, but I’m not entirely sure that they help to address the epistemological worry which I raised above. Oderberg is quite aware of this issue and there are some good passages where he defends the importance of the distinction between essence and property, such as the following, using mammals as an example:

To be sure, there are mammalian properties — having fur, lactating, and so on — but being a mammal and having mammalian properties are not the same thing. A mammal has mammalian properties because it is a mammal; these properties point to its essence. But isn’t it a mammal because it has these properties? Isn’t the real essentialist order of explanation upside down? To reverse the order of explanation, however, is ultimately to do away with essence, not to explain it. More accurately, it does away with real essence and replaces it with a surrogate bundle theory of essence as a collection of properties. And the problems with such a conception resurface.

Now, like I said, I’m inclined to accept the idea that we need to avoid such a bundle theory of essence; the ontology seems sound to me, if in need of some further work (such as an explication of the notion of ‘flow’). But if the epistemological story is driven by how we come to know essential properties, then I’m concerned that an epistemic gap remains between essential properties and the essence itself.

There are a number of details concerning the structure of essence and its applications that I found extremely interesting in Oderberg’s book, but this little review is already getting rather long, so I will not discuss these here. I will just conclude by noting that Real Essentialim is essential reading for anyone interested in essence, which should include all metaphysicians, and it serves both as an interesting survey of the notion as well as an original account of some its applications.