Abstract

A central question in constructivist studies of science is how the analyst should deal with the material objects handled by scientific practitioners in laboratories. Representatives of 'radical constructivism' such as Knorr-Cetina and Latour have gone furthest in exploring the role of these 'non-humans', but have also maneuvered themselves in untenable positions due to a fatal conflation of different meanings of the term 'construction'. The epistemological and ontological commitments of 'moderate constructivism', especially of the Strong Program defended by Barnes and Bloor, are more suitable for dealing with the task at hand. While radical constructivists treat the domains of nature and human society as largely coterminous, an alternative ontology stresses that natural reality is never fully absorbed into the world of culture but only interacts with the latter at localizable interfaces such as practices and artifacts. This perspective promises a more relaxed relationship with current forms of scientific realism.