The Incompatibility of Causation, Constitution, and Persistence

Constitution is typically regarded as a non-causal phenomenon. So, for instance, there is no reference to anything causal in the entry ‘Material Constitution’ in The Stanford Encyclopedia of Philosophy (Wasserman 2018). One rare exception is Karen Bennett (2017). Bennett objects to the received view that what she calls ‘building relations’ are non-causal (such as grounding, composition, determination, realisation, and constitution). According to her ‘there are particular building relations that are partially defined in causal terms, and there are even building relations that only obtain diachronically, in virtue of causal facts’ (2017: 2). I agree with this claim, but I also think that she doesn’t specify the kind of causal account needed to substantiate that claim.

The idea that persistence might be causal occurs more frequently, but, again, usually in process ontology. Whitehead takes persistent entities to be processes, i.e. a sequence of temporally ordered stages, each of which undergoes ‘concrescence’—a transition from potentiality to actuality— and the concrescence of each stage causally triggers the concrescence of the next (Emmet 1992). Peter Simons observes that the persistence of things seems to be dependent on the processes taking place in the thing and suggests we should think of persistent entities (or continuants, as he calls them) as dependent on a base of occurrents that stand in causal relations to each other (Simons 2000). Simons intends his view to be an alternative to the endurance and perdurance views, which he thinks are flawed because neither addresses the explanatory gap between the permanence of simples and the constitution and continued existence of compounds. Now, while I agree with Simons about the flaws of the endurance and perdurance views, I don’t like the idea of process as a structured series of stages or occurrences either. As I have previously argued, the stage view of process reduces change either to variation between stages— a problem it shares with perdurantism (see Ingthorsson 2001, 2009)—or to concrescence of stages, i.e. transitions from potentiality to actuality, which I think inevitably violates the genetic principle. I will discuss the nature of processes in Chapter 7.

Salmon also thinks persistent objects are causal processes but denies that this commits him to an event ontology of process (1984: 139ff). However, Salmon really doesn’t give an account of what a process is, nor does it seem possible to apply his account of causation in terms of propagation and interaction to the continued existence of processes. It rather seems that persistence, i.e. the ability to propagate a mark, is constitutive of causation rather than vice versa.

The general lack of attempts to make something of the causal aspects of constitution and persistence—outside process ontology—can plausibly be explained by the fact that the mainstream definitions of the notions of causation, persistence, and constitution present them as prima facie incompatible. Causation is understood as a diachronic relation between distinct relata—one event being the cause of another—and thus as an external relation. Indeed, the concept of non-causal spontaneous change is typically understood as one that happens in the absence of an external cause, like the decay of unstable particles. Persistence is also understood as a diachronic relation, but one holding between an object and itself, and thus intrinsic to the persistent thing. This may be the main obstacle to a causal account of persistence. If we are looking for the cause for the persistence of an object, we are not looking for an external compulsion; if an object persists, it does so of its own accord. Finally, constitution is typically understood as a relation between a whole and its parts and is traditionally taken to be a relation that can hold at any given time, i.e. synchronically. Consequently, it is a relation that is intrinsic to the constituted object but still not a relation of identity. In sum, tradition dictates that causation is a diachronic relation between distinct relata, persistence a diachronic relation between an object and itself, and constitution a synchronic part-whole relation. Against this background, to offer a causal account of constitution and persistence will require us to think differently about one or more of these notions. Here I explore what happens if we accept the interaction view presented in Chapter 4.