It is a prevalent habit of philosophers to periodically proclaim the death of realism. Whether it is Ludwig Wittgenstein’s notion of truth as being immanent to particular forms of life, Richard Rorty’s neo-pragmatism which shatters the mirror of nature, or Jean Baudrillard’s apocalyptic narrative of the death of reality in late Capitalism; realism has been pronounced dead many times over. And yet, realism lives on. A cynic might suggest that this repeated ritual is an attempt to affect the very thing it is trying to proclaim – a declaration repeated like a magical formula aimed at exorcising or conjuring away realism by declaring the specter dead in order to put it to death thereby reassuring us that what is dead is indeed dead. And yet realism continues to haunt philosophy in spite of the (un)holy alliance which aims to cast it out.

Over the past few years a growing number of thinkers from different disciplines have been arguing for the need for a return to realism. “Speculative realism”, drawing on the work of Quentin Meillassoux, has mounted a challenge against what it has called correlationism – a philosophy, or rather an anthropology, which is only concerned with how things appear and are understood by human subjects (phenomenology). Similarly, the “new materialisms” have argued for the need to do ontology and move beyond a purely anthropocentric account of the world. In the past month Hubert Dreyfus and Charles Taylor have released their new book Retrieving Realism. On its own account it offers “a radical critique of the Cartesian epistemic picture that has captivated philosophy for too long and restores a realist view affirming our direct access to the everyday world and to the physical universe” (see Retrieving Realism). For Dreyfus and Taylor knowledge of the world is more than representation – instead we can be said to gain knowledge of the world through our bodily engagement with it ― by being in contact with things, handling things, responding to them – in short, by “being in the world”. Even Deleuzians have been proclaiming the need for realism and, on more than one occasion, citing the work of Bhaskar. The long and short of it is that realism of all forms is back in.

Much of this resurgence in popularity is based in a desire to do justice to the knowledge we have of the world, and in particular, there is a growing concern to give science its due. It should also be no surprise that within the philosophy of science the growing consensus for some time has been towards a realist approach. Realists such like Sellars, Putnam, Harré, Boyd and Bhaskar are very much in fashion, as are contemporary realists like Cartwright, Psillos and the neo-Aristotelians. Should we be surprised at any of this? I think not. The argument for scientific realism is relatively simple and extremely effective. As Putnam suggested, realism is the only philosophy of science that does not make the success of science a miracle. This is often called the “no miracles argument” and is occasionally called the “ultimate argument for realism”. The fact is that the success of scientific theories and their ability to issue in novel predictions or practical applications lends support to the idea that scientific theories should be interpreted realistically. Success in science is to be expected, and this expectation is confirmed in experience and in practical application. This is not to suggest that science is always reliable or guaranteed to come to the truth of things, only that the success of science must be accounted for in a way that its success is not always a giant fluke. Science may at times be contested. It may be wrong. It may be unclear. It may not have a guaranteed method for success. Nevertheless we do still come to knowledge of the world. All this is to say, the simplest and clearest answer we can give for the success of science is that it comes to fallible and revisable, but truthful knowledge of the world it seeks to understand.

How does it do this? The fallible but reliable quality of scientific theory and practice arises from its capacity to make approximately true or adequate observations, predictions, descriptions and interpretations of the structures and powers of things which can then be measured or applied in different contexts. This includes the instrumental reliability and practical applicability of scientific theories which underpin the technology we all see and experience in our everyday lives, from water boiling in an electric kettle to the thousands of planes currently in the air. Science comes to knowledge about things, how things work, how things operate in different contexts, and why things work and operate in such ways by coming to an understanding of their natures (of what they are and what they do). This success is not simply pragmatic or instrumental, but rests on a realist assertion that there is something true about our knowledge of things which finds some confirmation or corroboration in instrumental and practical applications. The reason science has this dual quality of both fallibility and reliability is because it not based on a methodology which guarantees certainty, but on the broader notion of abduction – inference to the best explanation. Against a strict empiricism or pragmatism (a la Rorty for example), the realist argues that only by accepting the (relative) reality and (approximate) truthfulness of theoretical knowledge can we begin to understand the (uncontested) practical reliability and success of scientific knowledge and scientific methodologies. In short, pragmatism and instrumentalism require realism.

But why do we explicitly need to identify with realism? Why can’t we just get on with things and not worry about theory? An important aspect of the turn towards realism represents an acknowledgement that theoretical considerations play an important part in actual scientific practice, including the background theories which invariably inform (constrain and enable) scientific innovation and discovery. The theories we adopt inform our practice (and possible practice), and visa-versa. The realist account of the theoretical fallibility, revisablity and yet success of science, rests upon an idea of scientific research as being broadly cumulative by successive (but not necessarily convergent) approximations to the truth based on the principle and practice of inference to the best explanation (see Boyd). Approximately true theories provide approximately true and reliable measurements, methods and applications for the theoretical entities and powers they suggest account for the behavior of things. This approximate truth then serves as the basis for a cumulative and ongoing development between a current theory and its improvement (including its revision). The approximate truth of current theories explains why our existing measurements and applications are relatively reliable while being revisable, which in turn also helps to explain why our experimental or observational investigations and applications are successful in uncovering new theoretical knowledge, which, in turn, may produce improvements in measurement techniques, instrumental applications, etc. Such a recognition of the successful development and application of science only holds if our theories are approximately true and fallible, but still relatively reliable. Accordingly, it is simply not possible to explain the instrumental and practical success of science without adopting a realist conception of scientific knowledge. Likewise, it is impossible to justify the use of scientific methods and their success, even for instrumental applications, if one rejects realism.

The short answer to all this is that science is realistic about what realism entails, including the reliable but revisable status of its theories. This is not because science is based upon a certain and sure methodology which guarantees success, but because science is based upon inference to the best explanation, and therefore finds its accounts to be approximately true. To account for both the growth and revisability of our knowledge it must be granted that there is something in the real world which we, however partially, get a hold of. This something, this contact with reality, allows us to refine, develop and deepen our understanding of things and how they work giving science traction while remaining both revisable and yet reliable. Given that science is often held to be the gold standard for all knowledge, this realistic account of science has necessarily broader implications for other disciplines, perhaps above all social science which often attempts to model itself directly on natural science. In short, all of this means that realism is back in because it is the only theory which does not make our knowledge of the world, including its practical application, a miracle.

For more see:

Boyd, R (1989) ‘What Realism Implies and What it Does Not’. Dialectica Vol. 43. No 1-2, pp. 5-29

Cartwright, N. (1999) The Dappled World: A Study of Boundaries of Science. Cambridge: Cambridge University Press

Groff, R. Greco, J. (2012) Powers and Capacities in Philosophy: The New Aristotelianism. London: Routledge