Recent accounts of the relation between law and other social spheres have emphasized law’s “relative autonomy. ” The intui- tion behind the “relative autonomy” formula is that law is neither wholly independent of, nor entirely reducible to, political, eco- nomic and other social processes. Sensible as this intuition is, however, the idea of “relativeautonomy” by itself remainspurely negative. It excludes two unpalatableextremes-pure formalism and pure instrumentalism-butit does not by itself characterize, in positive theoreticalterms, the relation between law and other social discourses or practices. This Article examines an attempt in recent German social thought to specify theoretically the relation between law and other social spheres. The theory examined-Niklas Luhmann’s theory of “autopoiesis”-is,though familiar to Continentalread- ers, not yet well-known to American legal academics. This Arti- cle presents autopoietic theory to the American legal audience, with particularattention to the way in which Luhmann reformu- lates the “relative autonomy” problematic. Throughout, the Ar- ticle focuses on the connections between autopoietic theory and issues in American law and contemporaryAmerican legal theory. The Article’s strategy is to criticize those aspects of autopoietic theory that deserve criticism, but at the same time, to show how he theory might operate as a productive stimulus for American legal theorists.

Excerpt: Niklas Luhmann’s theory of modern society’s differentiation into autopoietically operating subsystems has attained a most prominent position in contemporary theoretical debates. However, the theory is widely accepted on false grounds. Luhmann himself, and most of his consenting and critical commentators, conflate the analytical and the empirical differentiation of societal subsystems. Whereas analytical differentiation can be conceived of as self-referential, as described by Luhmann/ empirical differentiation cannot be conceived of in terms of autopoiesis, because the autonomy of the societal subsystems operating in the real world is permanently being produced and reproduced by a multiplicity of action elements which are both inside and outside an analytically defined subsystem. Luhmann’s introduction of the concept of structural coupling is an answer to the mounting criticism in this direction.