Metaphysics as a Tool for the Realist

As I noted in the introduction, the relationship between metaphysics and science has recently come under scrutiny, with a number of commentators declaring the former not fit for purpose, given recent developments with regard to the latter.

Now some caveats are in order here. With regard to the first point, it has to be said that not all metaphysicians are ignorant of developments in science. Paul and Schaffer, for example, have both appealed to features of quantum mechanics in support of their different positions (a one-category ontology and monism respec­tively; see Paul 2013; Schaffer 2013). And when it comes to the second, this ‘ruling out' is not always definitive (French and McKenzie 2015). Take Leibniz's Principle of Identity of Indiscernibles for example. Following French and Redhead (1988) it has long been held to be violated by quantum physics yet a ‘Quinean' version has recently been constructed that is compatible with the physics (Muller and Saunders 2008; but for criticism, see Bigaj and Ladyman 2010). However, it might be felt that these are exceptions and that in general the dismissal of much of current metaphysics by philosophers of science is well justified on the grounds that it is simply out of touch with modern science (think, for example, of the way the notion of intrinsicality is usually understood in terms of ‘lonely objects' and how this is discussed in the absence of any consideration as to whether physics can even accommodate a model in which there is a lone particle in the universe).

Now if one is a realist, exemplar based or otherwise, seeking to articulate a locally delineated view of how the world is, what are one's options given the above?

One, of course, is simply to eschew metaphysics entirely and in answer to the question ‘what is the world like according to theory T?”, to simply point to T, set out in all its glory on a whiteboard, say, and to declare ‘That! It is like that!'. Now, that setting out will typically be—and certainly so in the case of physical theories— in terms of the relevant mathematics but only a radical Platonist will leave it at that. Physical realists will of course insist that the relevant terms must be interpreted, and those in eschewal mode will further insist that this interpretation will be ‘purely' physics based. Now of course, purity is a slippery notion but one can imagine our eschewalling realist declaiming ‘That! The world is like that! Where this term refers to the electron and that to the electro-magnetic field and refusing to say

anything more. But of course the door to metaphysics has already been opened via this interpretation, since it invites the further question ‘Yes, but what is the elec­tron? Is it a particle? Is it a wave? Is it even an object?' Of course, one could simply refuse to answer such questions, insisting that to do so would take us beyond what can be legitimately grounded in the relevant physics. But I suspect that many would feel that in so refusing the realist hasn't really lived up to the name and that our understanding of how the world is remains thin and impoverished.

And of course, even appealing to the ‘pure' interpretation of T invites com­parison to similar interpretations of both related theories and its predecessor. The term ‘electron' for example is freighted with certain connotations associated with its deployment in, say, classical mechanics. There the electron is regarded as a particle and, further, as an object that possesses certain properties and, further still, as an individual object, assemblies of which can be statistically considered in certain ways which are dependent on permutations of those objects being counted.

An alternative is to eschew metaphysics as it is currently formulated and adopt some form of ‘bespoke' metaphysics constructed to directly clothe the relevant features of modern science. We have been here before of course. One example is that of Whitehead, who drew on the early (or ‘old') quantum theory and its apparent ‘vibratory' features to motivate his process philosophy (see for example Whitehead 1926; for a recent consideration of this motivation see Epperson 2004). Another would be Eddington, who took the above feature of quantum statistics in particular (that is, its permutation invariance) to motivate a form of structuralism according to which objects are not prior to but on a par with the relevant relations and subse­quently went on to articulate this structuralist metaphysics in the context of what can be considered to be an early form of quantum gravity (Eddington 1946). The obvious problem with such a move—which is evident in the later works of both Eddington and Whitehead—is that such a bespoke metaphysical framework must be elaborated via bespoke terms, concepts, principles and categories and runs the risk of descending into incomprehensibility.

Fortunately, there is a third and, I would argue, more reasonable alternative: treat current metaphysics as a kind of toolbox that although it may contain some devices that are not ‘fit for purpose' may still contain others that the realist can use (French 2014; French and McKenzie 2012, 2015). So, although we might conclude that notions such as intrinsicality or principles such as Leibniz's are ruled out by modern physics, there may be others that we can adapt to fit. Let me expand on an example from (French 2014) and express it in the context of exemplar based realism.

4