Choosing what there is
In my previous note on Borges and Popper, I considered two presentations of a common sense argument for realism. Both begin from the thought that there are things (e.g., Bach’s music for Popper and a line from Victor Hugo for Borges) that I know1 I could not have created. For me, this line of argument is a satisfying reason to adopt a realist stance. But, as Popper freely admitted, it is not a conclusive one. Metaphysical and ontological realism, remain positions we adopt and then work from rather than a finding we can establish empirically.
Two papers I read last year2 help, I think, to draw out what follows from that point. The first, David Ludwig’s Ontological Choices and the Value-Free Ideal, complicates the common division encountered between questions of fact and questions of value; the so called value-free ideal. Ontology is often presented as if it were something settled before science begins: first decide what exists, then gather evidence and discover facts about it. Ludwig’s point is that scientists’ ontological choices are inherently value laden.
This challenges the value-free ideal once we take a moment to ask why one scientific ontology is chosen rather than another. The relevant considerations are not always exclusively epistemic ones particularly when they stem from our metaphysical commitment3. A result of this is an impact on the truth value of scientific statements as they are always made relative to a chosen ontology typically influenced by a chosen metaphysics.
But, whilst many take metaphysical and ontological realism to come hand in hand in science, the second paper, Alberto Corti’s Scientific Realism Without Reality? What Happens When Metaphysics is Left Out, separates these two issues that are often bundled together. Metaphysical realism is, roughly, the claim that there is a structured, mind-independent external world. Scientific realism concerns the ontological position regarding whether we should take scientific theories as real at face value, including their claims about unobservable entities.
Scientific realists commonly accept metaphysical realism as well, but Corti argues that scientific realism does not logically require it. It is possible to adopt scientific realism yet metaphysical agnosticism. I particularly found the thought experiment of Alice the solipsist, and Bob, to demonstrate this very nicely:
Let us suppose that Alice is a solipsist, i.e. she holds the view that she is the only real entity, and external reality is entirely the product of her mental activity. Alice notices that her reality presents many regularities. Moreover, she notices that some of her ideas (called ‘scientists’) study these regularities and claim that they are due to the existence of unobservable entities. Alice believes the scientists’ claims and thinks that (at least some of) these unobservable entities exist, in the sense that they are as real as her other ideas (like her idea of tables, chairs and the scientists themselves). Therefore, she has a strong disagreement with Bob, who think that scientific theories allow us only to make accurate predictions. According to Bob, indeed, we are rationally justified in believing only in those entities that human beings can observe directly.
Alice is a metaphysical anti-realist but, in this deliberately strange sense, also a scientific realist. The thought experiment is not intended to argue for solipsism. It merely shows that the two forms of realism are not inherently incompatible. Even someone who denies a mind-independent world can distinguish between taking the entities posited by science seriously and treating them as useful fictions. In essence, treating the imaginings of their mind as the reality that science is intended to investigate and explain.
This is a useful continuation of the Borges and Popper discussion. Their arguments offer an intuitive reason to reject solipsism and accept metaphysical realism, but they cannot prove it. Corti shows that this limitation need not force us into solipsism nor a kind of scientific instrumentalism. We can remain serious about the entities described by our best theories while being modest about the metaphysical foundations of that commitment.
Taken together, the two papers suggest a position of metaphysical humility without paralysis by analysis; a way to take a stance and move on to “getting shit done”4. I personally remain a metaphysical realist, and ontological realist in so far as I take the scientific task, for the common sense reasons described previously, but it is helpful to see more clearly which parts of scientific inquiry actually depend upon that commitment, and which do not. It’s also useful to see that scientists, both metaphysical realists and metaphysical solipsists, could set the question of metaphysics aside as, in many ways, irrelevant to the scientific task. Whilst they may not agree exactly what it is that exists and what it means in their ontologies in any metaphysical sense, they can still agree (or not) regarding what does exist ontologically speaking and upon methods for how to go about investigating and understanding it.
Footnotes
Or at the very least, suspect, I could not have created. The solipsist might counter the strong “I know I could not have created” argument by claiming, as Popper notes “I may perhaps underrate myself (and at the same time overrate myself) in my dream.”↩︎
Yeah… I’m plumbing some of the exported data from my deleted social media accounts to start as the basis for some content. Given that both my recent reading/re-reading of Borges and Popper had brought this topic back to my attention I remembered having read the papers referred to in this note, and having posted about them on Bluesky. So I went back to remind myself and write a short note expanding on the posts for the site to try and keep some momentum going for my writing here… though, despite capitalising on thoughts as they come, who knows if I’ll keep up this frequency of posting going forwards.↩︎
For example, I might say “I adopt metaphysical realism and therefore the things that make up my ontology are also real mind-independent things”. Of course, the choice of exactly what we think makes up our ontology is also a source of possible disagreement. The majority of Ludwig’s paper tackles this and provides interesting examples of the impact of choices in this regard.↩︎
To quote friend and colleague Dr Chris Matthews.↩︎