Jonathan Lowe: "There are (Probably) No Relations"

Jonathan Lowe: "There are (Probably) No Relations". Listen to the talk here.


A logico-metaphysical distinction needs to be drawn between material and formal predication, which applies both to monadic and to relational predicative propositions. Where formal predication is concerned, no real property or relation is predicated of the subject(s) of predication – that is to say, no genuine monadic or relational universal is predicated of the subject(s). Paradigmatic examples of formal predication are, in the monadic case, predications of existence and, in the relational case, predications of identity. Very plausibly, neither ‘existence’ nor ‘identity’ denotes a universal, nor are there existence or identity ‘tropes’ or ‘modes’.

In the case of relational propositions (propositions containing relational predicates), a distinction is often drawn between ‘internal’ and ‘external’ relations, with the former (but not the latter) sometimes being said to ‘supervene’ upon their relata, and hence to be ‘no addition of being’, or being ‘nothing over and above’ their relata. This way of talking is to be discouraged, because it suggests that ‘internal’ relations do exist, and yet somehow in a bloodless way which makes them only second-class constituents of reality. It is preferable to say that the only relations which could be ‘real’ at all would be so-called external relations. But it is probably better still to use the terminology of ‘material’ and ‘formal’ predication, and say that real relations, if they exist, would be entities materially predicable of their relata (their subjects), and that the sign that a true relational proposition involves material predication is that its subjects do not, on their own, suffice to constitute its truthmakers.

So, are there any ‘real relations’ in this sense? This is not to be confused with the question of whether there are really any (irreducible) relational truths, since there fairly evidently are — for example, truths of identity. The question, rather, is whether any true relational propositions involve material predication. Amongst the most plausible candidates are spatiotemporal and causal relational propositions, the implication being that there are, at least, real spatiotemporal and causal relations. But a case can be made for saying that this is false, and that all relational truths concerning spacetime and causality involve formal predication, being made true solely by their subjects of predication. Indeed, a case can be made for saying that the very notion of a ‘real relation’ – an entity (whether a universal or a trope/mode) somehow capable of characterizing more than one subject — is ultimately an incoherent one, compelling us to adopt either a monadistic or a monistic ontology at ‘the fundamental level’. That this is indeed so will be the tentative conclusion of this paper.