Beyond the Baseline:Actor-Network Theory and Lateral Experiments with Robotic Agency

The issue of nonhuman agency has often been a contentious element of ANT-inspired approaches to STS. Critics have consistently maintained that ANT's principle of generalized symmetry flattens, erases, or deliberately ignores any fundamental differences between humans and nonhumans. The result, they argue, is an impoverished understanding of agency that, among other things, neglects issues of intentionality and subjectivity. In contrast, proponents argue this misunderstands how ANT proposes symmetry and flatness as baseline conditions of empirical inquiries rather than as substantial ontological claims. That is, by rejecting prior assumptions about essentialized differences of any kind, ANT refocuses our attention on what can be shown to make a difference in practice while avoiding speculation about differences we imagine should exist in principle. In this paper, I briefly rehearse the issues raised in the debates over more-than-human agency in ANT. With inspiration from Gad and Jensen’s (2016) work on lateral concepts, I subsequently discuss a series of experiments from my earlier fieldwork in Japanese robotics laboratories where agency, in different ways, emerges as a key matter of concern. I show how the configurations of concepts, methods, and technologies in these experiments enact forms of agency that simultaneously resemble and challenge classical tenets of ANT. Ultimately, I suggest that ANT can be a helpful companion in moving beyond the flat conceptual baseline of more-than-human agencies by engaging more directly with occasions where the ontological status of agency remains underdetermined, unsettled, and open to practical reconfigurations and experimental interventions of different kinds. <br/><br/>