Johan Hietanen
Recently, I’ve had the pleasure of discussing with philosophers of science about the virus behind COVID-19. Among other topics, I’ve heard an argument along the following lines: “If you’re not, at your core, a realist about the virus, then it would make no sense for you to support any prevention or management of the virus either.” This argument seems to follow from quite an uncontroversial initial assumption: you have to believe something exists before any action pertaining to it becomes sensical. Yet, the statement above is inaccurate. How so?
Scientific realism, however defined by any given author, is a general positive attitude towards the statements made in science. Anti-realism is defined by Chakravartty (2017) to mean “any position that is opposed to realism along one or more of the [following] dimensions[:] […] the metaphysical commitment to the existence of a mind-independent reality; the semantic commitment to interpret theories literally or at face value; and the epistemological commitment to regard theories as furnishing knowledge of both observables and unobservables.”
This categorization might work for many purposes, but as is often the case with two-valued divisions, important detail is lost. The categorization is not a slight of Chakravartty, however, as this is more or less how stances against scientific realism are generally understood. As well as contention to binary divisions, this may have to do with the anti-realistic counterparts to realism’s “truth”, that is, concepts such as “empirical adequacy”, “warranted assertability”, and the like. The carousel then goes around: if you deny that theories, models, et cetera produced by scientific endeavour are true and state that they are only empirically adequate or something similar, you’re not taking science itself seriously and the success of science will be left as a mystery.
Variations of anti-realism differ, some being more and some less sensical than others. Be that as it may, it is important (as in many other contexts) to differentiate between non-commitment and denial. Anti-realism is best understood as a collection of stances that give a positive argument against the possibility or the confirmation of truth in science. This is to be contrasted with a position that simply refuses to commit to truth in science – such a position is not defined by a positive argument but a negative one. This is to say, if there is reason to doubt the soundness of scientific realism, that is reason enough to not commit to its statements – but that alone does not lead to anti-realism. We can call such positions non-realism.
It is important to stress that adopting non-realism does not require any positive argument against scientific truth (such as the pessimistic metainduction), but a simply an undercutting defeater of the realist argument. Thus one does not have to argue against truth in science, but rather against realism about science, which espouses the former.
Before going further, there is need to clarify what we are talking about when discussing realism. To return to the very general characterization given in the beginning of the text, a very weak version of scientific realism could be understood to just be the belief that science gets things right as an epistemic endeavour. To argue against this would be futile, as the position is obviously correct. However, this form is so weak that it can be questioned whether the noble title of “realism” can be attached to it at all. What is really being discussed in the context of scientific realism is the two-fold commitment with a metaphysical and epistemological branch. The metaphysical branch states that there exists a mind-independent reality and the epistemological branch states that knowledge of this reality is attained (by the means of science, at the very least). It’s this definition that we’ll apply our non-realism to.
What sort of undercutting defeaters are there for scientific realism, then? The most recognized argument for scientific realism is undoubtedly the no-miracles argument by Putnam (1975). The argument goes like this: if one does not accept scientific realism, then the success of science becomes a miracle. The crux of the argument is that, if no miracles are allowed, the overwhelming success of science necessitates that science is right about the mind-independent reality.
One counter-argument to this is to appeal to underdetermination, but a shorter path is available. One can simply ask whether or not the results of science could be exactly the same even if there were no ontological connection of the aforementioned kind. If such a possibility exists, then the success of science does not necessitate realism. Indeed, such a possibility can be sketched quite easily – just imagine observations about, say, A and B, and observations about their relationships, R, and you can start doing science with these observations. There really is nothing outside of the intuitions of some necessitating the miraculousness of this endeavour. (The no-miracles argument is actually, in regard to its epistemological content, just the track record argument for science. That, in turn, is actually quite sound when applied to educate people why scientific methods should be trusted above other forms of enquiry.)
Another common argument for realism is the corroboration argument, which goes roughly as follows: if two or more separate detection mechanisms observe the same entity in a crucially similar way, it would be very extraordinary if the entity did not exist. The corroboration argument is actually just a mirror image of the principle of triangulation, which is central to improving the reliability of scientific knowledge in countless fields of science. Yet, it does nothing to ascertain an ontological connection: as long as we agree that we are able to observe the same set of phenomena, then it should not be so extraordinary that we are able to observe the same phenomena in various ways; the fact that I can both see and feel the calluses in the palms of my hands is hardly a winner argument for scientific realism. (It is to be noted, though, that the corroboration argument is a very sound one when applied against naïve empiricism which takes all phenomena to be separate.)
Other arguments for realism and their defeaters could be considered, but that would soon enough start to repeat a pattern. The core of non-realism, as defined here, is the fact that scientific realism does not bring to science anything that wasn’t already there without it. This would, of course, be alright if realism was a necessary consequence of science – but it’s not. Non-realism is not a new idea: along these lines were, among others, Fine (1984) and Rorty (1993). It is central to understand that scientific realism (or anti-realism, for that matter) is not science – and not really fruitful philosophy of science either since it adds nothing of value.
To return to the statement presented in the beginning of this text, it is now easy enough to clarify where it errs. It states that supporting actions against the spread of the coronavirus requires one to accept, in terms of scientific realism, the mind-independent existence of the virus as described by the relevant scientific fields examining it. The problem here is that what we are able to infer about the virus is done in terms of observation of phenomena: we have contracted people with some symptoms, which are linked to some observations about biology, which in turn are observed to react to different environments and interventions in some ways, and so forth. To deflate realism here requires absolutely no rephrasing of how the virus is understood: it actually just requires phrasing the scientific understanding of the virus exactly as it is.
This still leaves us to answer the question: so what? If one is uncomfortable without the notion of ontological connection, why shouldn’t they pursue a philosophy that provides that to them? This theme could, of course, escalate into ruminations about the nature and purpose of philosophy itself, but I’ll try to steer clear of that topic this time around. However, I offer you this: if there is nothing to be gained from pursuing a field of study, that should at least prompt a consideration of moving on to other topics. Further, insisting on something that cannot be found runs the risk of building a house on top of an imaginary plinth. And that might, in the worst-case scenario, become a health hazard.
References
Chakravartty, Anjan (2017): “Scientific Realism”. Stanford Encyclopedia of Philosophy, Stanford.
Putnam, Hilary (1975): “What is Mathematical Truth?”. Mathematics, Matter and Method: 60—78. Cambridge University Press, Cambridge.
Fine, Arthur (1984): “The Natural Ontological Attitude”. Scientific Realism: 83—107. University of California Press, Berkeley.
Rorty, Richard (1993): “Holism, Intrinsicality, and the Ambition of Transcendence”. Dennett and his critics: demystifying mind: 184—202. Blackwell, Oxford and Massachusetts.
The Helsinki Circle 