Notes on empiricism and autonomy

Euro-American common sense tends to the reflex that it is important to stipulate the conditions under which science can be properly carried out. This is because scientific inquiry needs to be protected from the distortions that might come from outside. The idea that science needs to be protected in this way is often (though not always) linked to ‘empiricism’ and to ‘positivism’. Empiricism is a family of traditions in the philosophy of science which argue that scientific truths grow out of, and are properly generalised from, appropriate empirical observations. Positivism is another, closely related, set of traditions which argue that scientific truths are rigorous sets of logical relations or laws that describe the relations between (rigorous) empirical descriptions.

In the social sciences, empiricism and especially positivism are now usually seen negatively. Raymond Williams comments that positivism is a ‘swear word by which nobody is swearing’ (1988, 239). No doubt this is right. However their basic intuitions are widespread in Euro-American common-sense thinking about science and social science. It is commonly assumed that observations should be unbiased and representative, and that theories should be logical and consistent both with one another, and with observation.

The sociology of science, which was invented by Robert K. Merton (1973a; 1973b) started out on this assumption. There were good reasons for Merton’s intuitions. He was writing at the time of Nazi racial science, and Stalinist Lysenkoism (which argued that plants could transmit and inherit acquired char- acteristics). He argued that these lethal lapses from proper scientific standards were a consequence of the failure to insulate science from political agendas in totalitarian societies. Scientists’ capacities for unbiased observation and logical thinking were being eroded by these agendas. Instead science should, he said, be protected by a ‘scientific ethos’. First, it should be universalist, testing its ideas in terms of: ‘ preestablished impersonal criteria: consonant with observation and with previously confirmed knowledge’ (Merton 1973a, 270). This meant that the race, gender, politics, or national origins of the scientist were not relevant to truth. Second, it should be disinterested. Scientific claims should be assessed independently of local social, economic, political, and personal interests. Third, it should be sceptical. Scientists should not take things on trust. (Merton talked of organised scepticism.) And finally, it should be communal. By this Merton meant that scientists should always publish their results: that science would best advance if it published its findings.

Merton’s vision of science throws up some problems. (It is, for instance, difficult to see how scientists are consistently sceptical: in practice if they are to be effective they have to take a lot on trust.) And there are problems, too, with empiricism and positivism (we will encounter some of these below). But this is a convenient place to start because Merton is very clear that anything that interferes with ‘empirically confirmed and logically consistent statements of regularities’ (1973a, 270) is illegitimate because it detracts from the proper empirical and

Interlude 17

logical basis of truth. Merton’s theory, then, is that research needs to be disentangled from the social and the psychological, and entangled solely with logic, with facts, and with methods for determining the facts.14

This is a language to which we will return. Different visions of science propose that it should be (or it is) entangled and disentangled with the world in different ways. Empiricism offers one recipe for this. It tells us that science (and social science too) have to be autonomous if they are to work properly. They should be disentangled from the social.

Дата добавления: 2015-11-14