George P. Simmonds
Oxford Brookes University
Scientific realism (SR) asserts that science describes an objective and real world, a world about which its theories may claim some sort of truth. Putnam describes (SR) as ‘the only philosophy that does not make the success of science a miracle,’since it explains its success in terms of its cohesion with a truthful reality. Though this has since been disputed, this quote nicely encapsulates the appeal of (SR)—it allows for an encouraging viewpoint of science. This is not to say that it is rashly optimistic; rather, it facilitates a straightforward metaphysical, semantic and epistemological theory which enables simple commitment to science’s proposed models. (SR) appears to be the most suitable theory for modern attitudes to science.
(SR) by no means enjoys universal popularity, however: a number of anti-realist ideas have risen to confront it, including those of American philosopher van Fraassen, whose constructive empiricism (CE) will be the focus of this essay.
Though (SR) and van Fraassen’s (CE) agree on semantic and metaphysical terms, they differ in their epistemological conceptions of science, namely its allusions to truth. In this van Fraassen challenges (SR)’s central cornerstone, the no-miracle argument (NMA), which serves to reinforce the theory’s epistemic principles. The (NMA) is a vital defence of (SR), and its survival against van Fraassen’s refutations is crucial to the position’s legitimacy. If (CE) is able to deny the (NMA) it would strengthen the case for (CE)’s supremacy as a philosophy of science.
This essay will aim to support van Fraassen in his (CE), proffering it as a favourable alternative to scientific realism. Sections (1) and (2) will aim to clear the decks in clarifying the (SR) and (CE) positions, while section (3) will discuss van Fraassen’s objections to the (NMA).
(1) Scientific Realism
(2) Constructive Empiricism
(3) The No-Miracles Argument
2. Scientific Realism
Van Fraassen claims that ‘truth must play an important role in the formulation of the basic realist position.’In this he was correct: an underpinning feature of (SR) in all its forms is a fundamental commitment to science as something which lays a claim to truth. That is that science reflects the reality of things, as they are and in themselves. (SR) has an important ‘dual character’ whereby ‘on one hand it is a metaphysical doctrine, claiming the independent existence of certain entities’ and ‘on the other hand… an epistemological doctrine asserting that we can know what individuals exist and that we can find out the truth of the theories or laws that govern them.’This can be condensed to the following:
(SR) holds that…
(1) Scientific theories describe an objective, mind-independent reality and should be interpreted literally as descriptions of real things.
(2) Scientific theories aim to provide a true account of what the world is like, and the acceptance of such theories identifies with a belief that they are true.
A full understanding of (SR) demands an elaboration of these two commitments. (1) is both a metaphysical and semantic claim. By the realist conception, the world—as investigated by science—is mind-independent and in no way subject to that which the human mind brings to it. Since the realist makes no effort to abstract the world from the investigation of science, it follows that they should interpret the discourse of scientific theories literally. This means that when a theory describes an entity or an event, it is so reflected in the objective world. A realist, then, holds that ‘what makes [a scientific theory] true or false is something external—that is to say, it is not our sense date, actual or potential, or the structure of our minds, or our language, etc.’
In addition to (1) a scientific realist must also uphold (2). The first aspect of (2) is normative, and speaks only of the aims of science, and not of its achievements or limitations. It is thus considered too weak to characterise the epistemological commitments of (SR) on its own.[ The second aspect, however, encapsulates the scientific-realist position rather nicely. The acceptance of a theory, to the realist, identifies with the belief that it is true or approximately true, entailing a positive approach to the epistemic achievements of science. This is not to suggest that the realist holds all our current theories to be true, only that they see truth as the marker of a successful scientific theory. This loose commitment to truth allows for the scientific realist to commit to theories ‘tentatively,’ or in virtue of its approximate rather than absolute truth. There is a distinct pledge in (SR) to ‘the idea that our best theories have a certain epistemic status… [that] they yield knowledge of aspects of the world.’
An anti-realist theory, then, must deny one or more of these positions. Van Fraassen, while agreeing with the metaphysical and semantic aspects of (SR), contends with its conceptions of science’s aims and what it means to accept a scientific theory. This is to say he maintains (1) while rejecting (2). (1) is to be hereon assumed with regard to (CE).
3. Constructive Empiricism
According to van Fraassen, and all constructive empiricists, ‘science aims to give us theories which are empirically adequate; and acceptance of a theory involves a belief only that it is empirically adequate.’This formulation of the normative and epistemological aspects of (CE) has earned van Fraassen the title of the ‘rehabilitator of anti-realism.’
(CE) claims that we are unable to properly account for unobservable phenomena, and that laying claim to truth in our theories regarding them is going beyond the enterprise of science. A constructive empiricist would insist that theories describing unobservable phenomena should be instead considered empirically adequate, since we are able to account for the unobservable phenomena only in terms of their explaining (or ‘saving’) the observable phenomena. This contrasts the ideas of (SR), which holds that a theory’s accounting for the phenomena entails its truth and thereby the truth of the objects it describes. Recall that when a scientific realist accepts a theory, they commit to more than just its empirical adequacy; they commit to its observable truth. (CE) does not do this, and allows for an agnostic approach to the existence of unobservables.
This can be illustrated in the example of gold. By observable means we are able to determine certain qualities possessed by gold, such as its melting point, its hardness, or its colouring. (CE) would accept theories positing the truth of these qualities, since if they were present to us under appropriate circumstances we would be able to observe them. The unobservable qualities of gold, however, such as its atomic number and molecular composition, cannot be observed in this way. (CE) would, with regard to these qualities, see the theories relevant to them as simply accounting for, or making sense of, the observable phenomena. (CE) would not agree that these theories deliver any sort of truth, in the strict sense, regarding gold’s unobservable properties.
Many believe this adoption of empirical adequacy over truth to be the more prudent, less epistemically audacious, choice.The caution of (CE), though restraining our scientific theories, allows its subscribers to remain truly faithful to the ‘spirit of empiricism.’
4. The No-Miracles Argument
As stated in my Introduction, the (NMA) is a vital cornerstone to scientific-realist thought. It was famously introduced by Putnam in 1975 (though it can be traced back to Boyd’s unpublished work) and has since been discussed somewhat widely. The (NMA) is built upon the observation of science’s extraordinary success. From our best theories we are able not only to make sense of current phenomena, but accurately predict future phenomena based on the models they present. The accuracy of our best theories can be so formidable that they become laws in themselves, to which we expect all conceivable phenomena to adhere. This is what is meant when the (SR) describes the ‘success’ of science. This does, however, raise the question of what can explain this success.
The (NMA) indicates the solution to this problem for the scientific realist. Recall this essay’s opening quote: (SR) is ‘the only philosophy that doesn’t make the success of science a miracle.’ This is because exponents of (SR) regard the observable success of science as evidence of our best theories’ truth. This idea is based on the principle that if our approved theories possessed no element of truth, their empirical success would be miraculous, since one would expect such theories to yield only false results. As far as (SR) is concerned, this observation leaves us with only two choices: to accept a theory’s congruence with the current phenomena as a miracle, or as an indication of its truth. A scientific realist would choose the latter.
To the constructive empiricist, then, (SR) would argue that we are able to secure the truth of our scientific theories—and thereby the objects they describe—on the basis that they provide the best explanation for the current phenomena. Returning to the gold example: the scientific realist would aver that because the theories surrounding gold’s unobservable properties provide the best explanation for the observable phenomena, we are justified in committing to their truth, and thereby the truth of the properties they describe. This is where the (NMA) makes reference to Inference to the Best Explanation (IBE), a notion which underpins the spirit of both the (NMA) and (SR). According to (SR), following the logical structure of (IBE) is a part of everyday life. When we experience a set of phenomena we routinely regard its best explanation as the one which is true, or at least likely to be true. The (IBE) is essential to understanding the (NMA).
The (NMA) appears to cast doubt upon (CE)’s allowing of an agnostic approach to unobservables and the theories surrounding them. Surely, the scientific realist would argue, the success of a scientific theory is a clear indication of its truth, and the truth of that which it describes.
Though a number of anti-realist authors have opposed the (NMA) this essay will for depth’s sake focus on van Fraassen’s response. He says: ‘I claim that the success of current scientific theories is no miracle. It is not even surprising to the scientific (Darwinist) mind. For any scientific theory is born into a life of fierce competition, a jungle red in tooth and claw. Only the successful theories survive — the ones which in fact latched on to actual regularities in nature.’In this van Fraassen is suggesting that our best theories appear to possess truth because those that do not simply fail to ‘survive’ the environment of scientific inquiry. He uses the analogy of natural selection to illustrate how only the theories which sport the guise of truth (those that are empirically adequate) endure the scientific process of elimination. Our ‘best’ theories, then, are simply those which best match the current phenomena, viz. those that are empirically adequate. His point here is that the success of a scientific theory need not entail its truth, since it can be empirically adequatewithout being remotely true. According to this view it would be no miracle for a scientific theory to be simultaneously successful, empirically adequate, and completely and utterly false.
This rejoinder and dismissal of (SR)’s claim to truth defends the (CE) right to be agnostic regarding the existence of unobservables and the theories which describe them.
The ramifications for (SR) as a result of this rejoinder are arguably grave. If the extraordinary success of science is unable to provide reason to believe that science is able to touch on some element of truth, then there is no way for (SR) to advance its conceptions beyond those of (CE). If our best theories are able to remain both successful and empirically adequate without referring to any sort of truth, then it becomes difficult for the scientific realist to construct any real means of deciphering where truth lies and where it does not. Here they might encounter further problems, such as that of underdeterminism, which poses that for any scientific theory there is set of empirically equivalent rival theories that are equally believable, and that belief in the truth of any one of these is ipso facto irrational. From here the scientific realist would feel himself inclined to accept van Fraassen’s supposition that ‘rationality is only bridled irrationality,’and perhaps abandon (SR) entirely.
Van Fraassen asserts that the (NMA) does not establish that the success of scientific theories is able to ratify their truth value, and it follows forthwith that (SR) is unable to support its fixed belief in unobservables from within the enterprise of science.
This leaves (SR) with the task of developing a means by which its dogmatism can be justified. Van Fraassen, meanwhile, would not only regard such measures as futile but also unnecessary, since to conceive of a successful theory as empirically adequate as opposed to true in no way inhibits the spirit of science, only absolves it of having to associate with excessive inflationary metaphysics. For these reasons, and those listed above, (CE) may be argued to be a favourable alternative to (SR).
Boyd, R.N. (1983). What Realism Implies and What it Does Not. Erkenntnis.
Boyd, R.N. (1971) Realism and Scientific Epistemology. Unpublished typescript.
Cacioppo, J.T. et al. (2004). Realism, Instrumentalism, and Scientific Symbiosis: Psychological Theory as a Search for Truth and the Discovery of Solutions. American Psychologist.
Chakravartty, A. (2013). Scientific Realism. Available: plato.stanford.edu/archives/sum2013/entries/scientific-realism/. Last accessed 12/03/14.
Chibeni, S.S. (2005). A Humean analysis of scientific realism. In: Guimarães, L. Ensaios sobre Hume. Belo Horizonte: Segrac Editora. 89-108.
Cummiskey, D. (1992). Reference Failure and Scientific Realism: A Response to the Meta-Induction. The British Journal for the Philosophy of Science. 43 (1), 21-40.
Duhem, P. M. M. (1954). The Aim and Structure of Physical Theory. Princeton: Princeton University Press.
Fine, A. (2000). Concise Routledge Encyclopedia of Philosophy . London: Routledge.
Frost-Arnold, G. (2010). The No-Miracles Argument for Realism: Inference to an Unacceptable Explanation. Philosophy of Science.
Iranzo, V. (2007). Abduction and Inference to the Best Explanation. Theoria.
Kitcher, P. (1993). The Advancement of Science: Science Without Legend, Objectivity without Illusions. Oxford: Oxford University Press
Kuhn, T.S. (1962). The Structure of Scientific Revolutions. Chicago: University of Chicago Press.
Lipton, P (2004). Inference to the Best Explanation. 2nd ed. London: Routledge.
Long Luong. 2013. 7 – 4 – Part Four – Scientific Realism and the No Miracles Argument (0331). [Online]. [14/03/14]. Available from: http://www.youtube.com/watch?v=OV-JiSAPqfQ.
Long Luong. 2013. 7 – 5 – Part Five – Scientific Anti-Realism Constructive Empiricism (0726). [Online]. [14/03/14]. Available from: http://www.youtube.com/watch?v=wLUnwOo-ygE.
Monton, B. & Mohler, C. (2014). Constructive Empiricism. Available: plato.stanford.edu/archives/spr2014/entries/constructive-empiricism/. Last accessed 12/03/14.
Musgrave, A. (1988). The Ultimate Argument for Scientific Realism. In: Nola, R. Relativism and Realism in Sciences. Dordrecht: Kluwer.
Putnam, H. (1975). Mathematics, Matter and Method. Cambridge: Cambridge University Press.
Richardson, A.W. (1998). Carnap’s Construction of the World. Cambridge: Cambridge University Press.
Rosen, G. (1994). What is Constructive Empiricism?. Philosophical Studies.
Sellars, W. (1962). Science, Perception and Reality. New York: Humanities Press.
Van Fraassen, B. (2001). Constructive Empiricism Now. Philosophical Studies.
Van Fraassen, B. (1989). Laws and Symmetry. Oxford: Oxford University Press.
Van Fraassen, B. (1980). The Scientific Image. Oxford: Clarendon Press.