Realism and Natural Science

Observable Phenomena, Metaphors and Mechanisms

In Chapters 2 and 3 we noticed the problem faced by empiricism in accounting for the apparently important role played in scientific theory by reference to whole classes of entities and processes of which we would otherwise be completely unaware: atoms, molecules, subatomic particles, fields of force, viruses, genes, quasars, black holes and so on.

Strict versions of empiricism have difficulty in accepting this as the core of scientific knowledge, since so many of these theoretical entities are not accessible to direct observation. However, realists argue that the great intellectual achievement of science is to have discovered that the world is so much more complex in its structure than common-sense understanding could have imagined. The task of philosophy of science is to try to understand the forms of investigation and reasoning that enabled scientists to do this. Philosophers such as Rom Harre and Mary Hesse emphasized the role of analogy and metaphor in scientific theory-building. The synthesis of proteins by the molecules of DNA is likened to the interpretation of a sign system, when biologists speak of the genetic code. The concept of an electric current contains an analogy with the flow of a liquid, and Darwin’s concept of natural selection contains an analogy with the selective breeding of domesticated animals and plants.

We can think of theory-building through the use of metaphors as a three-stage process. First, there is the collection of evidence about patterns of observable phenomena.

This does not, of course, settle the issue of when scientists are justified in moving from the tentative use of a metaphor to the claim that the hypothesized mechanism really does exist. It is important at this stage, however, to recognize that there is a difference between (a) being able to prove the existence of some particular class of entities or mechanisms claimed by science and (b) being able to justify the realist account of science as an attempt to discover and study such mechanisms. Realist theories of science are attempts to do (b) not (a). However, the historical evidence of the simultaneous discovery of mechanisms by independent groups of scientists (for example, the independent invention of the concept of natural selection to explain the formation of new species by Charles Darwin and Alfred Russel Wallace), and the various ways in which different sciences converge around common beliefs about underlying mechanisms (for example, the theory that matter is made up of atoms, and that these combine in various ways to form molecules, is the basic ontology of physics, chemistry and modern biology) are hard to explain unless we assume that there really are independently existing things and processes which are more or less adequately grasped through the theories which scientists invent. However, critical realism offers apparently much stronger philosophical arguments for this view of the nature of science.

Transcendental Arguments

These arguments begin with what is taken to be an uncontroversial description of some phenomenon,p. The question is then asked, ‘What must be the case for p to be possible?’ Let us suppose that some condition, or state of affairs, c, can be identified as a necessary condition for p. Since we have already accepted that p is actual, then it must be possible, and so the conditions which make it possible must be satisfied. So, c must be the case. An example from social life might be: ‘Jane is a student.’ For anyone to be a student there must be teachers, bodies of knowledge to be taught and learned, educational institutions to define the roles of teacher and student and so on. Since Jane is a student, the necessary conditions of possibility for someone to be a student must be satisfied, and so, for example, we can conclude that educational institutions exist. Another well-used example is: ‘Jay cashed a cheque.’ For anyone to be able to cash a cheque there must be a banking system, a money economy and so on. So, we can conclude that there is a money economy. In these rather obvious cases, the transcendental argument has conclusions which are not very surprising or interesting. However, in Roy Bhaskar’s realist theory of natural science, arguments of this type have led to some very interesting and important conclusions.

Science: Transitive and Intransitive Dimensions

Bhaskar constructs transcendental arguments on the basis of(presumably) Uncontroversial descriptions of scientific practices such as experimentation, scientific disputes and the application of scientific knowledge in technology. If we ask the question, ‘What must be the case for (for example) scientific experiments to be possible?, the answers fall into two groups. These will be, on the one hand, statements about what the world must be like for experiments to be possible, and, on the other, statements about what scientific investigators must be like for them to be able to conduct experiments. Bhaskar uses the term ‘intransitive dimension' to characterize the referents of the first set of statements, and the term ‘transitive dimension' for the second. For the moment, we will focus on the intransitive dimension - what must the world be like in order for distinctively scientific practices such as experiments to be possible? (Note: it is arguable that the same form of reasoning could be applied to any other human social practice, with comparable results.)