Abstract:
The Realism-Antirealism debate in the philosophy of science has taken a major turn after the rise of the philosophy of quantum mechanics and especially after the introduction of a new form of structural realism by Worrall, Ladyman, French and others to address the problem of what is real. In this paper we take a close look at the realism-antirealism debates focusing on various types of realism, along with an analysis of van Frassen's Constructive Empiricism (CE) and Structural realism (SR) as viable alternatives to scientific realism. We formulate the EP-DP postulate to explain the problems of scientific realism and discuss how CE and SR address aspects of this postulate.
The structure of this paper will be as follows we will first examine realism and anti-realism and then move on to Constructive empiricism and Structural realism, discussing the EP-DP argument along with the other issues.
Psillos (1999) distinguishes three stances or approaches taken by realists the semantic, the metaphysical and the epistemic - the metaphysical version implies that the world has a natural mind-independent existence, the semantic stance takes scientific theories as 'descriptions' of the observable and unobservable and implies that theories represent objects in the world, the epistemic stance considers scientific theories as approximately ?true? of the real world. Thus scientific realism would in its entirety mean 1. Objects/ Entities exist (metaphysical claim) 2. Scientific theories are descriptions of objects (semantic claim) 3. Scientific theories are approximately true representations of the world (epistemic claim). Apart from the metaphysical, semantic and epistemic claims, direct realism is a more na�ve version of realism which implies that there are external objects existing independent of our minds, which we directly perceive with our senses. Causal realism is a step further because it claims that the external objects exist independent of our minds and yet 'cause' our indirect perception of them via the senses .
This does not differ too much from an understanding of scientific realism which claims that objects exist; this existence is independent of our knowledge, and some of their properties can be measured. In so far as theories are 'descriptions' of objects, the objects would be independent of our knowledge or of these descriptions and in so far as theories are 'approximately 'pictures of our world, the implications are that, certain parts of reality are 'accessible'.
Yet, the problem of existence is a complicated one. As also cited by Ladyman (2002), Eddington's two tables aptly explicate this complication. The distinction between 'appearance and reality' is described in terms of our commonsense world and the scientific world where a table is either an object as we perceive it or as Eddington wrote, 'mostly emptiness' as science describes it. Eddington described the two table concepts - table1 as having extension and colour and substance and table2 as having electric charges and atoms not describable in terms of the bulk that we perceive. The scientific nature of objects is different from the commonsense view of the object and so the question arises - which is real and how are these two related? This necessitates a proper discussion on scientific realism.
Standard Scientific Realism (SSR)
The central tenet of scientific realism is the NMA or the no miracles argument (termed by Putnam), which aims to show that the best and successful scientific theories are reasonably true, so NMA actually defends the realist claim that successful theories should be nearly appropriate descriptions of the world both in its observable and unobservable aspects. The realist explanation for the empirical success of scientific theories rests on the fact that scientific theories actually describe the true nature of the world. The success of science is thus not miraculous, science describes the way the world is. As Psillos writes, [I]t best explains why the observable phenomena are as they are predicted to be by those theories. (1999, p.71) So the no miracles argument is an instance of inference to the best explanation or IBE. The NMA offers the best explanation for the success of science by defending that scientific theories are approximately true. NMA thus leans on the reliability of IBE for its own validity, a method criticised by the anti-realists.
Yet, the success of science and SSR depends on the knowledge of theoretical entities and this knowledge is mainly by observation, so an important and contentious question in the study of scientific realism is what is observable and how is it different from the unobservable?
Grover Maxwell (1962) suggests that in the sense of observable- there are no a priori or philosophical criteria for separating the observable from the unobservable. Yet in science certain theoretical entities like electrons, or ultraviolet radiation, are considered as 'unobservables'. So, is observability a limitation in principle? Positivists would think so. Heisenberg's uncertainty principle seems to suggest a limitation in principle. Yet pre-historic animals are also unobserved but they may not be unobservable in principle. Maxwell (1962) suggests that a change in our 'language' from the term 'observable' to 'observed' would solve at least a part of the problem, yet observational or descriptive language must refer to what has been observed. We differ here as we argue that in case of theoretical entities 'observational language'; do not necessarily have to be on what has been observed, it can be based on that which has never been observed.
Observability seems to be both a limitation in principle and a limitation in practice. Electrons are not observable in principle; dinosaurs are not observable in practice. In case of theoretical entities, observability is thus a limitation in principle. In case of material entities, the observability seems to be a limitation only in practice.
In so far as scientific realism characterizes scientific theory as a story of what there really is and as scientific realism is seen as an attempt to 'discover rather than invent' anything (van Frassen, 1980), the issues on the unobservable remain contentious. As Wilfrid Sellars (1963) puts it 'to have good reason for holding a theory is ipso facto to have good reason for holding that the entities postulated by the theory exist'. van Frassen quotes Brian Ellis, a scientific entity realist, 'I understand scientific realism to be the view that the theoretical statements of science are, or purport to be, true generalized descriptions of reality' . Thus scientific realism would claim that the entities postulated by the scientific theories, whether observable or unobservable, also exist. van Frassen considers this as some kind of a realist blunder, for observables according to him might be putative, i.e. they may or may not exist, so observations and existence have no logical relation.
The two strong claims of scientific realism ? theoretical entities exist (we call this the Existence Postulate, EP) and this 'existence' might include all theoretical entities and not just objects, and that scientific theories describe the world as it is the no miracle argument and inference to the best explanation (we call this the Description Postulate, DP) seem to be both contentious. DP is contentious because EP is contentious; we might not be able to say that science really describes the world unless we are able to say that the entities that science describes really exist. Yet, we might not be able to postulate the existence of theoretical entities until we have a scientific theory to describe them. The relation of the two postulates is thus reciprocal. If we can describe theoretical entities, there must be corresponding entities in the world and if there are theoretical entities in the world, they must be adequately described by our theories. We call this the EP-DP reciprocity postulate.
Constructive Empiricism (CE)
Scientific realism thus entails an existence of theoretical entities (metaphysical commitment), a description of these entities as real (semantic commitment) and a reciprocal relation of this through our knowledge (epistemic commitment). According to Ladyman, anti-realism denies any or all of these commitments and so there are different kinds of anti-realists . van Frassen?s Constructive empiricism (a kind of anti-realism) denies that scientific theories can describe real objects via our knowledge. He also denies that the existence and description of objects have a necessary correspondence leading to our knowledge, thus denying the EP-DP relation, at least to an extent. He is agnostic about EP, that all theoretical entities exist. In his own words, scientific realism is guided by the fact that 'Science aims to give us, in its theories, a literally true story of what the world is like; and acceptance of a scientific theory involves the belief that it is true (van Frassen, 1980, p.8).' Yet, according to van Frassen we can get around the problems of providing a 'true story of the world' by establishing constructive empiricism according to which, 'Science aims to give us theories which are empirically adequate and acceptance of a theory involves as belief only that it is empirically adequate'(1980, p.1069). To van Frassen, a theory is empirically adequate if what it says about the observables in the world is true. This approach although apparently close to the description postulate in scientific realism, is not completely similar. Whereas the description postulate is based on the fact that scientific theories adequately describe the world as it is, CE is based on the fact that only those observations/descriptions of the world that 'fit in' with the scientific theory can give empirical adequacy and subsequent acceptance of that theory. CE differs from the other anti-realist positions like positivism and instrumentalism, as in positivism, theoretical entities have meaning only through their connection with the observable so the implication of existence is not completely removed. CE removes this philosophical baggage by emphasizing only on empirical adequacy. As this is a pragmatic dimension rather than an epistemic dimension, an overt disagreement as in realism-antirealism debates is not obvious between realism and CE (van Frassen, 1980, p.1070). Realists and anti-realists differ in the epistemic dimension that our knowledge of the EP-DP relations are true. Anti-realists deny our knowledge of the fact that our descriptions of the world correspond with its existence, thus denying completely the EP-DP reciprocity relation, whereas CE seems to be suggesting that some of these descriptions do fit in to give us the theory so EP-DP may not be a reciprocal but a unidirectional relation.
Constructive Empiricism tries to solve the problem of the existence of the unobservable and observable theoretical entities and the problem of NMA by postulating that there is no need to believe that good theories are true (denying DP) and that the entities they postulate are real (denying EP). Although Ladyman claims , van Frassen accepts the semantic and the metaphysical claims of realism, on the face of it; things do seem more complicated than that. As we have seen CE has to entail the unidirectional relation from DP to EP, because some descriptions have to fit in to give empirical adequacy to a scientific theory, so constructive empiricism seems to be based on the very premises that it seems to deny! CE has also been criticised because it seems to have given ontological significance to the observable-unobservable distinction by emphasizing on empirical adequacy. van Frassen also denies the positivistic stance of a clear demarcation between the unobservable and the observable terms. Although relying on underdetermination seems to be the only way out for CE, the data itself can underdetermine which theory is empirically adequate. The case of other types of anti-realism is simpler because they deny the EP-DP reciprocity completely without necessarily denying EP and DP, which seems a tenable but controversial approach.
Structural Realism (SR)
The ontological discontinuity of theoretical entities, across theory change forms the basis of pessimistic meta-induction and is seen as the most important criticism against scientific realism . Theory change leading to changing concepts of theoretical entities signifies a break in the EP-DP relations, so the existence of objects and its corresponding descriptions as representing its true existence faces an explanatory problem. Theory change does not undermine either EP or DP, yet with the theory change, there might be a change of DP, so the EP-DP relations can be broken down here, justifying the antirealist stance. Structural realism is an attempt to solve this problem of the epistemic claim of SSR. Structural realism argues that it is the 'structures' rather than the objects that should be the area of concern. Thus coming back to Ladyman's point , we argue that if realism is about existence of objects (metaphysical claim), then by denying objects, SR justifies anti-realism, as according to Ladyman, denying any one of the three realist claims implies a denial of realism. Yet if we define realism as establishing the existence of theoretical entities which are ?structures?, (as claimed by structuralists) rather than objects, the problem seems apparently diluted.
French and Ladyman quotes Worrall to explain shifts in the ontology of entities, [t]here was a continuity or accumulation in the shift, but the continuity is one of form or structure, not of content (1996) . Ladyman (1998) has also suggested a semantic or model-theoretic approach to explain 'partial relations' to capture the continuity of partial structures through theory change . French and Ladyman explain Worrall's position by elaborating on the nature of structures and the structure of the content. Worrall used the theory of light to show that the ontology of light - from waves to corpuscles to wave-particle duality, has gone through several changes, yet the mathematical equations have remained unchanged getting incorporated into Maxwell's electromagnetism and then to quantum electrodynamics. Yet, as French & Ladyman claims, mathematical equations represent one aspect of the structure and thus structures can be non-mathematical. Although Worrall?s main concern was to account for theory change and pessimistic meta-induction, French, Ladyman and other structuralists have differed in their aims and are also concerned with the metaphysical problems of scientific realism especially concerns of a descriptive nature, i.e. whether and how do theories represent the world? Thus although Worrall was concerned only with addressing the problem of the continuity of EP-DP relations, other structuralists like French and Ladyman are also concerned with the adequacy of the description postulate, DP. In fact SR is also concerned with EP postulating the realism of theoretical entities. Yet, the only difference of SR and SSR is that SR believes that these entities are structures rather than objects. As long as SR does not deny the existence postulate (EP), a form of realism is maintained.
To address the question on the ontological content of things, Worrall draws on Poincar� who wrote, that theoretical terms are ?merely names of the images we substituted for the real objects which Nature will hide forever from our eyes. The true relations between these real objects are the only reality we can ever obtain? (Poincar�, 1905) . Here we come across the phrase, real objects, possibly used in a Kantian sense, but this is again controversial. Psillos distinguishes nature and structure and argues that structural retention cannot be the basis of a realist position (Psillos, 1999). He claims that the nature and structure of an entity form a continuum and that the nature of an entity, process or physical mechanism is no less knowable than structures (Psillos, 1999, p.155); structures are equally unknowable, so using structures instead of natures of objects does not solve any problem. Psillos claims that natures of objects are embedded in the equations that describe them, so natures are not over and above structures, yet knowing all the laws that an entity follows does not exhaust knowing what this entity is (1999, p.156). So it seems, Psillos talks of entities which exist in itself, and has features beyond its structural features that cannot be specified in terms of mathematical formalisms. The approach looks similar to Penrose?s claims on consciousness and the implications of Godel's theorem. Do structures or natures exhaust the description of theoretical entities? This question should not concern the structural realist because as long as they maintain the existence postulate (EP) that structures exist, and as long as they explain the EP-DP relations, they might not have to concern themselves with EP itself. Instead of emphasizing on denial of objects, SR seems to be an acceptance of EP, and as long as SR doesn't deny EP or DP (unlike what we think constructive empiricism seems to imply), an explanation of realism is achieved by SR in terms of the EP-DP reciprocity.
Conclusion
After the advent of quantum mechanics, and the associated philosophical problems of 'entities', Scientific Realism might have to be considered as a philosophy of accepting, yet not necessarily justifying, the existence postulate (EP) and the description postulate (DP), but has to be seen as a means to justify the EP-DP reciprocal relations between the two. Structural realism justifies the EP-DP relations, denying neither the existence postulate nor the description postulate. Constructive empiricism seems to deny both EP and DP and yet tries to accept a part of the EP-DP reciprocity and hence has more problems than SR. Anti-realism denies EP-DP reciprocity completely yet this does not necessarily imply a denial of either the existence postulate or the description postulate and that is problematic in antirealism. Standard scientific realism can only justify its claims by replacing its metaphysical commitment with the existence postulate which is broader in its range of theoretical 'entities' and can also save itself from existing antirealist criticisms by defining itself as only representing or justifying the EP-DP relation.
Bibliography
Images of science : essays on realism and empiricism, with a reply from Bas C. van Fraassen / edited by Paul M. Churchland and Clifford A. Hooker.
Chicago ; London : University of Chicago Press, 1985.
Devitt, Michael
Realism and truth / Michael Devitt.
Oxford : Blackwell, 1984.
Feigl, Herbert and Maxwell, Grover (eds).
Scientific Explanation, Space, and Time vol.3
Minnesota Studies in the Philosophy of Science (Minneapolis University of Minnesota Press, 1962)
French, Steven and Ladyman, James
Remodelling Structural Realism: Quantum Physics and the Metaphysics of Structure
Synthese
136(1): 31-56; Jul 2003
Steven French; James Ladyman
French, Steven
Scribbling on the Blank Sheet: Eddington vs. Braithwaite and the Structuralist Conception of Objects', Studies in History and Philosophy of Modern Physics 34 (2003), 227?59.
Kukla, Andre
Studies in scientific realism / Andr� Kukla.
New York : Oxford University Press, 1998.
Ladyman, James
Understanding philosophy of science / James Ladyman.
London : Routledge, 2002.
Ladyman, James
What is structural realism? / James Ladyman.
Studies in history and philosophy of science; v.29, 3, 1998, pp. 409-424
1998.
Ladyman, James
What's Really Wrong with Constructive Empiricism? Van Fraassen and the Metaphysics of Modality
The British Journal for the Philosophy of Science, Vol.51, Issue 4, December 2000
pp. 837-856
McMullin, Ernan
Van Fraassen's Unappreciated Realism
Philosophy of Science, 70 (July 2003) pp. 455-478
Papineau, David
The philosophy of science / edited by David Papineau.
Oxford ; New York : Oxford University Press, 1996.
Psillos, Stathis
Scientific realism : how science tracks truth / Stathis Psillos.
London : Routledge, 1999.
Sellars, Wilfrid
Science, perception and reality / by Wilfrid S. Sellars.
London : Routledge & K. Paul, 1963.
van Fraassen, Bas C.
On McMullin's Appreciation of Realism Concerning the Sciences
Philosophy of Science, 70 (July 2003) pp. 479-492
van Fraseen, Bas
The scientific image / Bas C. Van Fraassen.
Oxford : Clarendon, 1980.
Worrall, John
Structural Realism: the Best of Both Worlds?
Dialectica, 43/1-2 (1989):99-124