An A-Cultural, Scientific Supertheory of World Society?

Abstract

All the contributions to this Forum raise incisive questions about Mathias Albert's (2007) attempt to inject new life into “globalization theory” with the help of Luhmann's systems theory of world society. While I agree with some of the criticisms of this particular appropriation of Luhmann's theory, I would surmise that these do not point to inherent limitations of a Luhmannian approach but partly reflect a number of misconceptions about what such an approach entails. Among these misconceptions (some of them perhaps invited by Albert's IPS piece, though much less by his other work involving modern systems theory) three seem particularly noteworthy: 1. that Luhmann's theory necessarily figures as a kind of supertheory for IR; 2. that a Luhmannian approach seeks to bestow a “scientific” aura on international political sociology; and 3. that it is unable to grapple with questions of culture or discourse. Let us consider each of these misconceptions in turn. Leander and Hindess directly take Albert to task for proposing Luhmann's systems theory of world society as a supertheory for globalization, and the specter of systems-theoretical imperialism also seems to lurk behind Thomas's and Robertson's calls for a broader theoretical agenda in globalization studies. Albert's seemingly supertheoretical ambitions with respect to “globalization theory” notwithstanding, he has generally favored a more selective and pragmatic engagement with Luhmann's theory as “a source and… strategy for concept-formulation” (Albert 1999:242) and as in fact open to revision by IR theories (including theories of globalization) (Albert 2002, 2004b). More generally, as a grand theory Luhmann's systems theory has always been a somewhat self-subverting enterprise given its (epistemologically) constructivist commitment to the polycontextuality of observations (that is, the ever-present possibility to observe observations and to observe otherwise). Once we bear in mind this commitment and consider Luhmann's theory more as a “toolbox” (Albert 2004a:3) than a …