Laws in Social Science

The question of the possibility of social laws has divided social scientists along both disciplinary and epistemological lines. Macroeconomics has traditionally depended on laws, or at least lawlike statements (e.g., the “laws of supply and demand”), whereas most historians would be skeptical of the existence of historical laws. These positions (to which of course there are exceptions) reflect the epistemological foundations of the disciplines.

The epistemological grounds for the rejection of laws can be summarized as the view that there is [Page 559]too much variability deriving from individual consciousness and free will in the social world to permit lawlike regularities. Although the conclusion of this argument may be ultimately correct, it is also often premised on an oversimplified version of what a “law” is. The common notion of a scientific law is one that brooks no exceptions in the predicted behavior of phenomena. Classical laws of physics are often cited, such as gravity, thermodynamics, and so forth. However, most scientific laws have quite different characteristics, and even classical laws will exhibit certain local exceptions and, under certain circumstances (outside of human experience), may break down altogether (Nagel, 1979, chaps. 4, 5).

Laws can express fundamental theoretical principles (such as those in economics), they may be mathematical axioms or experimentally established invariances, or they may express statistical regularities. Although a theory is usually seen to be a more speculative statement about regularities, sometimes the difference between a law and a theory may be no more than a linguistic or social convenience. Thus, although they both enjoy similar degrees of corroboration, one speaks of Newton’s laws and Einstein’s theories (Williams, 2000, p. 33). If, then, by accepting the possibility of theoretical corroboration, one accepts there can be regularities in the social world, does this entail a commitment to laws that all have the property of claiming the existence of some nonaccidental occurrences? Perhaps regularities in economic and demographic behavior are as predictable, under given circumstances, as those of gases. Statistical laws, at least, would seem to be possible. However, to continue the gas analogy, although certain local conditions (such as temperature or pressure variations) will create statistical fluctuation in the distribution of gases, the characteristics of the exceptions to demographic or economic laws are much more dependent on unique cultural or historical circumstances (Kincaid, 1994). Thus, there will be regularity in the exceptions to physical laws, but demonstrating such regularity in social laws requires reference to other universalist principles such as rationality.

What is perhaps more important than whether there can be social laws is that good evidence suggests that there can be nonaccidental regularities in the social world, even though such regularities may vary in their cultural or historical applicability and extent.