Real Essentialism

David S. Oderberg

• Publisher: Routledge
• Year: 2007
• Print ISBN: 9780415323642
• eBook ISBN: 9781134348848
• Edition: 1
• Topic: Philosophy
• Subtopic: Philosophical Metaphysics

Real Essentialism presents a comprehensive defence of neo-Aristotelian essentialism. Do objects have essences? Must they be the kinds of things they are in spite of the changes they undergo? Can we know what things are really like – can we define and classify reality? Many if not most philosophers doubt this, influenced by centuries of empiricism, and by the anti-essentialism of Wittgenstein, Quine, Popper, and other thinkers. Real Essentialism reinvigorates the tradition of realist, essentialist metaphysics, defending the reality and knowability of essence, the possibility of objective, immutable definition, and its relevance to contemporary scientific and metaphysical issues such as whether essence transcends physics and chemistry, the essence of life, the nature of biological species, and the nature of the person.

1
Contemporary essentialism and real essentialism

1.1 Against modalism: possible worlds

That there are at least some things in the world that have essences is a proposition to which more philosophers are prepared to subscribe than there once were. This is due almost exclusively to the growth of what might be called modal thinking – or modalism – in the light of the development of formal modal logic in the second half of the twentieth century.

The development of modal logic went hand in hand with the development of modal semantics, which it is standard to give in terms of the theory of possible worlds. The semantics naturally gave rise to speculation on just how we should understand possible worlds, with positions ranging from strongly anti-realist to strongly realist. Yet, whatever the position, most philosophers have come to believe that thinking about possible worlds can give us some insight into whether or not objects have essences.

The famous work of Saul Kripke and Hilary Putnam in the 1970s sparked a resurgence of essentialist thinking, and was based firmly on an understanding of meaning that relied heavily on the concepts of modality and possible worlds. In one way or another, Kripke–Putnam style reflection has supposedly allowed many to see that water is essentially H₂O; that tigers are essentially animals; that heat is essentially molecular motion; that material objects could not have been originally constituted differently from how they were in fact originally constituted; that maybe certain material objects could not have a wholly different constitution at any time from the one they actually have; that an animal could not have originated from a different sperm and egg to the ones it actually originated from; perhaps that the mind is not necessarily identical with the brain (and hence, according to Kripke’s well-known view, not actually identical with the brain); and so on.

Whether or not any of these propositions is true is not the present concern. (I will say more about such claims and many similar ones throughout this book.) Rather, what is of concern is that anyone should think that modalism can tell us anything – at least anything of any metaphysical significance – in the first place. There are serious problems with the very idea of appealing to possible worlds to tell us anything about essences. Some of them derive from the use of possible worlds itself, and I do not propose to add to the already voluminous discussion of which theory of possible worlds, if any, is correct.¹ We should, though, note a few issues that undermine appeal to the concept of possible worlds in trying to gain some metaphysical insight into modality in general and essence in particular. First there is the following worry: any realist theory of possible worlds will be circular in its attempt to illuminate modality, for there has to be some criterion of what counts as a possible world; there are by definition no impossible worlds. But then we have to have a prior conception of modality before we can use possible worlds to explain modality. Why, as Scott Shalkowski (1994) asks, are the pencils in my drawer not possible worlds either collectively or individually? David Lewis replies that they ‘bear no interesting relation to our common modal notions’ and do not have the ‘right constituents’ to serve as truthmakers for modal statements (Shalkowski 1994:679).² Yet this is to state the obvious. The question is why it is so, and the answer is that we already possess a prior grasp of modal notions sufficient to rule out pencils as possible worlds. But this means the realist theory of possible worlds is circular: it cannot be used to explain or analyse modality if we already have to understand modality (at least to some fairly robust degree) before we can even give the theory.

Secondly, even if the realist could get around the circularity problem, say by postulating possible worlds as primitive existents, as a modal given – rather than as entities for which we have to have modality-involving criteria – he would end up merely relocating the analysandum. Instead of having to understand the modal properties of objects within a world, we will have to come to terms with the modal properties of the worlds themselves. What is it for the worlds to have the modal properties they do (at the very least being possible, and perhaps also necessary)? We are still faced with unanalysed modality, only it has moved somewhere else. Now there may, as realists believe, be net theoretical benefits to be gained from explaining the modality of individuals within worlds in terms of the worlds themselves, but unless one is wedded, implausibly, to a cost–benefit approach to metaphysics, this will not be satisfactory. We want to know why objects have the modal properties they do. To answer that this is (at least in part) because worlds have the modal properties they do is only to push the problem from one place in the rug to another.

Thirdly, all possible worlds theories stare in the face the problem of irrelevance. Kripke famously stated the objection that when we say that Humphrey might have won the election, we are talking about Humphrey, not some counterpart (Kripke 1980:45). It is irrelevant to what Humphrey might have done that in some other world some other individual (albeit very similar to Humphrey) does something, no matter how similar what he does is to what Humphrey himself might have done. Lewis replies obscurely that the other world, with its Humphrey counterpart, ‘represents’ our Humphrey as winning – ‘[s]omehow, perhaps by containing suitable constituents or perhaps by magic’ (Lewis 1986:196). Yet this is beside the point: I could represent Humphrey as winning by painting a picture of him doing so, or writing a screenplay. The problem is how what is going on in another world – especially in a world causally isolated and spatio-temporally disjoint from ours, as it is in Lewis’s theory – could have any bearing whatsoever on what might have happened to Humphrey. Lewis is well known for attacking ‘magical ersatzism’, the view that possible worlds are simple abstract entities. Yet if any theory contains a hefty dose of magic, his does.

The problem of irrelevance undermines not only Lewis’s theory. Whether possible worlds are understood as abstract natures (Stalnaker 1976), possible states of affairs (Plantinga 1974), ‘world books’ (Adams 1974), or some other kind of real but abstract entity, the question arises as to how what is true of that kind of thing can have any bearing on the modal properties of a concrete material object such as a man, a mouse or a mountain. For example, according to Plantinga every possible world w is a maximal or ‘super’ state of affairs to which there corresponds one and only one ‘super-proposition’ S – the union of some set of propositions with the set of all of their consequences. Worlds are correlated with superpropositions (or ‘books’, as he also calls them) in the following way: w obtains if and only if every member of S is true (Plantinga 1970). Now it is true that Socrates is essentially not a number. For Plantinga this is true if and only if every world containing Socrateity contains the non-numberhood of the thing that possesses Socrateity, which in turn is true if and only if every book containing the proposition that Socrates exists entails the proposition that Socrates is not a number.

Now leaving aside for the moment the issue of individual essences or haecceities (about which I shall say a little in Chapter 5.4) and also the question-begging use of the modal concept of entailment, the problem is what bearing either states of affairs or books has on the essential non-numberhood of Socrates. Formulations involving either of them do not give the meaning of the statement that Socrates is essentially not a number, because the latter proposition is about Socrates, not about either non-actual states of affairs or books. But does either kind of formulation give the truth conditions of such a statement? It depends on what one means by ‘truth conditions’. In whatever sense possible worlds may be said to exist, of course ‘Socrates is essentially not a number’ is true if and only if every world (for Plantinga, maximal state of affairs) containing Socrates (or Socrateity) contains the non-numberhood of Socrates (or of the individual possessing Socrateity). It would be inconsistent to hold that Socrates was essentially a non-number but that there was some world in which Socrates was a number. In this sense the appeal to worlds (or books) provides truth conditions: it tells us what is the case if and only if Socrates is essentially a non-number. But it does not tell us what makes it true that Socrates is essentially not a number. Plantinga and other realists may think it does, but again they would be guilty of irrelevance, and of confusing the consequences of Socrates’s having the essence he does with the constituents of that essence, which belong to the individual person of Socrates. It is because of what Socrates is that, if there are possible worlds, then in every possible world in which he exists he is essentially not a number. But what he is by virtue of which this is true of him has no more to do with how things are in a possible world than his being snub-nosed has anything to do with what is going on in the Himalayas.

If the appeal to real possible worlds changes the subject from the possessors of essences to the situations in which those possessors exist, then the appeal to modal fictionalism or other anti-realist devices fares no better. No appeal to fictional discourse can explain why a real entity like Socrates has the essence he does – why he is essentially a man but essentially not a mouse. If talk about possible worlds is akin to talk about fictional characters in a novel but the modal truths they illustrate are still literally true, then the fictional discourse is a mere heuristic by-product of literal modal truths that remain to be explained. If the fictionalist takes modal truths themselves to be not literally true, then he has given up on real modality (and hence real essence) altogether, and should be classed together with other modal sceptics and anti-essentialists (on whom see Chapter 2). If the anti-realist takes appeal to possible worlds in the revelation of modal truth to be a akin to using a calculator or an abacus to uncover arithmetical truth, then one can question how useful the heuristic device may be (nowhere near as useful as a calculator, to be sure); but there will be no question of such a device having any explanatory or analytical force in giving flesh to the concept of essence. (For more on modal fictionalism, see Rosen 1990.)

Modalism is characterized in part by its reliance on possible worlds theory, or perhaps more accurately by its reliance on intuitions about possible worlds with a certain amount of theory to clothe the intuitions. It is also characterized by the modal approach to meaning, specifically via rigid designator theory. According to Gyula Klima, contemporary essentialism just is the thesis that ‘some common terms are rigid designators’ (Klima 2002:175). This approach was driven by the work of Kripke and Putnam (Putnam 1970, 1973, 1975a; Kripke 1980) and has had enormous influence on essentialist thinking ever since. The basic ideas are too well known to require restating here: what is relevant for our purposes is the central thought that one can approach essence by considering language, in particular whether a term functions as a designator of the same thing in all possible worlds in which it exists. It is via this consideration that Kripke, for instance, argues that heat is essentially molecular motion (Kripke 1980:131ff.). Scientists have discovered that heat is identical with a certain kind of molecular motion (more precisely, mean molecular kinetic energy). But since ‘heat’ and ‘molecular motion’ are rigid designators, their identity must be necessary if it obtains at all. Putnam approaches the matter in a similar way, but more in terms of indexicals such as ‘this’ (as in ‘this stuff’) guaranteeing sameness of extension for a term such as ‘water’ across possible worlds.

The problem with the rigid designator approach to essentialism is that it is shot through with essentialist assumptions from the beginning. First, the necessity of identity is built into the very conception of a rigid designator: if ‘a’ designates a in the actual world, then we know trivially that the conjunction ‘“a” designates a and “a” designates a’ is true, i.e. we know that ‘a = a’ is true, in other words that a = a in the actual world. But since a rigid designator designates the same object in every world (in which the object exists), we know that ‘a = a’ is true in every world, in other words that a = a in every world, i.e. that, necessarily, a is identical with itself. The same applies for distinct rigid designators ‘a’ and ‘b’ that designate the same object a (i.e. b) in the actual world. Hence the necessity of identity is part of the very concept of a rigid designator, and the necessity of identity is a fundamental – indeed the fundamental – truth of contemporary essentialism. (The point is a familiar one; see, e.g., Mellor 1977:75–6.) To the reply that it is a metaphysical truth but not an essentialist one, or that it is only ‘trivially’ essentialist in some innocuous sense, I claim that its apparent insubstantiality must not be confused with its real import. It is not simply that identity is the relation that everything necessarily bears to itself and nothing else,³ but that the necessity of identity carries the appearance of triviality. This is because it is in fact the eviscerated contemporary essentialist form of a foundational real essentialist truth to the effect that every object has its own nature – a matter to which I will return in Chapter 5.

Secondly, even if one were to argue that this objection confuses constitution with consequence – that the thesis of the necessity of identity is a consequence of rigid designator theory, albeit an immediate one, but not itself part of the theory – this will not help the contemporary essentialist. For even if one were to present the necessity of identity as an inference from necessary self-identity and Leibniz’s Law⁴ (as Kripke 1971 does), one would still have to presuppose Leibniz’s Law as a de re necessary truth, i.e. a necessary truth about objects, not a mere de dicto necessary proposition, and one would also have to assume the necessity of self-identity. Now, it is true that the necessity of identity is not really (as opposed to logically) distinct from the necessity of self-identity, in which case the presentation of the proof in this form is merely circular anyway.⁵ But, leaving that aside, Leibniz’s Law is also an essentialist truth: no object can, at the same time and in the same respect, lack the qualities it possesses. And this itself is but a species of the law of non-contradiction, viz. that nothing can both be and not-be at the same time and in the same respect.⁶ Note at this point that a retreat to the ‘formal mode’, reformulating the argument as one concerning substitutivity and referential transparency, is no more than a kind of anti-metaphysical escapism: if one takes refuge in language one will never get to any essentialist truths at all (except, perhaps, concerning language).

Thirdly, as Nathan Salmon has shown, Kripke’s own purported essentialist derivations presuppose even more substantial essentialist truths (if the preceding are not thought substantial enough). For example, in trying to prove the necessity of origin – specifically the necessity of original constitution – Kripke has to presuppose the sufficie...

Table of contents

• Routledge Studies in Contemporary Philosophy
• Contents
• List of illustrations
• Preface
• Acknowledgements
• 1 Contemporary essentialism and real essentialism
• 2 Some varieties of anti-essentialism
• 3 The reality and knowability of essence
• 4 The structure of essence
• 5 Essence and identity
• 6 Essence and existence
• 7 Aspects of essence
• 8 Life
• 9 Species, biological and metaphysical
• 10 The person
• Notes
• Bibliography
• Index