Historical Epistemology and Structural Marxism

However, a quite different approach to understanding the nature of the sciences was emerging in France. In the empiricist tradition, debate about science tended to be carried on as if some abstract criteria for distinguishing science from non-science could be established independently of historical or social context.

Despite these differences, the French tradition of ‘historical epistemology' (see Lecourt 1975; Gutting 1989) shared with the other approaches so far mentioned a strong commitment to the distinction between science and non-scientific patterns ofbelief. Their historical studies focused on the processes of formation of scientific disciplines through a struggle against the networks of error and delusion which precede them. Common­sense and pre-scientific thought is not mere ignorance, or absence of knowledge, but, rather a tenacious and powerful source of resistance to genuine science. Even after the establishment of a science, this tissue of errors continues to exist and threatens to invade the tentative achievements of a new science in the shape of ‘epistemological obstacles'.

As the notion of a tissue of errors suggests, this approach recognizes the ways in which particular concepts are bound together through mutual definition into a systematic network. Scientific ideas are not invented one by one, but coexist in theoretical ‘problematics'.

Another consequence of the concept of the problematic is that scientific change must necessarily be discontinuous, and non-cumulative. For a new science to be born, a whole interconnected mesh of falsehood has to be overthrown. This moment is termed an ‘epistemological break’. Similarly, subsequent scientific change has to take place, not through the steady accumulation of facts (as in the empiricist picture) but by way of wholesale restructuring of the problematic. In this process, previous assumptions are called into question, and science advances not so much by adding to its past achievements as by overthrowing and replacing them.

Key features of this approach were carried forward in France by Louis Althusser and structural Marxism, and by the post-structuralist Michel Foucault (see Chapter 10 for more on Foucault). In a series of articles written during the 1960s (collected in English translation as Althusser 1969), Althusser made use of the concepts of problematic and epistemological break in an attempt to relaunch Marxism as a creative scientific research programme, after decades of orthodox subservience to the policy requirements of Communist Party leaders (see Benton 1984; Elliott 1994). He and his collaborators provided close textual readings designed to show the emergence of a new ‘scientific’ problematic in the study of history in Marx's works after 1845.

But Althusser and the structural Marxists faced a particularly difficult problem. Like their predecessors, the historical epistemologists, they emphasized the gap between ordinary, common-sense thought, or ‘ideology’, and genuine science. The radically egalitarian student movement of the 1960s was quick to point out the potentially elitist and undemocratic implications of this insistence on the distance between ‘science’ - the possession of a small coterie of self-styled intellectuals on the left - and ‘ideology’ - the allegedly inadequate and defective beliefs of the ‘masses’ whose benefit the science was supposed to serve.

Though he radically revised his views, Althusser never resolved this problem in a satisfactory way. However, his earlier writings were significant for his distinctive way of thinking about science as a social practice of a quite special kind. For Althusser, science was a practice of social production of knowledge, in which the key role was played by the theoretical problematic of the science concerned. On this model of science, the theoretical concepts making up the problematic are employed as ‘means of production’ of new theoretical concepts by transforming conceptual ‘raw materials’ (which might be ideas produced by prior ‘theoretical practice’, or drawn from the wider culture, and so on). This model opens up the possibility of sociological explanation of the content of scientific knowledge, but without giving up the notion of scientific objectivity (unlike most other ‘constructionist’ sociological approaches - see Glossary, p.

In effect, Althusser’s approach welds the insights of the historical epistemologists together with the realist and materialist legacy of the Marxist tradition. In his terms, theoretical practice produces knowledge in the form of ‘thought-objects’, which, in the case of science, correspond to ‘real objects’, whose existence is independent of and external to the process of producing knowledge of them. However, Althusser never succeeded in solving the problem of how, in science, it is ensured that this correspondence really does hold. Unless this could be done, the crucial distinction between science and ideology could not be sustained.

Post-Althusserian sociology of knowledge and culture tended to accept that this problem was insoluble and as a consequence moved towards analyses of language and cultural processes in abstraction from epistemological questions of truth, falsity or reference to a real world independent of people’s beliefs. As we will see (Chapter 10), much of the work of Michel Foucault takes this form, as does the work of ‘post-Marxists’ such as Ernesto Laclau and Chantal Mouffe (1985). However, there were some who attempted to hold onto Althusser’s project: to combine together the model of science as a social practice of production of knowledge with a realist understanding of knowledge as about something independent of itself.

The critical realist approach, first developed in the early 1970s, tried to do just this, and to find ways of solving the problem Althusser had powerfully posed. Central to the critical realist case was the use of ‘transcendental arguments’ in relation to a range of scientific practices. A transcendental argument is one which takes as its premiss an uncontroversial or undisputed description of something actual (for example, scientific experiment), and then poses the question, ‘What must be true for this to be possible?’ In the case of scientific experiments, Roy Bhaskar, the most influential critical realist philosopher of science (see Bhaskar 1975, 1997 and 1979, 1998), demonstrated that only if the external world existed and had certain properties, and only if human agents had certain abilities to intervene in that world and monitor the consequences of doing so, could there be anything recognizable as experimentation.

These arguments offer a convincing way of combining together an understanding of science as a historically changing and socially situated human practice with an acknowledgement of it as, distinctively, a social practice whose aim is the production of knowledge about objects, relations, processes and so on which exist and act independently of our knowledge of them. However, Althusser’s own search for guarantees that our knowledge corresponds to its external objects is decisively abandoned: science is displayed as rational enquiry into the nature of an independently existing reality, but also as one whose products (currently accepted scientific knowledge-claims) are always provisional and subject to modification or rejection in the face of future evidence and argument. It is only, realists argue, if we recognize science as making knowledge­claims about independently existing realities that we can make any sense of what it is for a belief to be shown to be wrong. While appearing to be more modest in its claims, relativism turns out to be a way of insulating beliefs from refutation. Some of the most basic intuitions underlying realist approaches are derived from the experience of getting things wrong, of making mistakes and being - sometimes painfully - reminded of this by reality. We’ll return to a discussion of this approach and its implications for the practice of social science in Chapter 8.