Ethics of Algorithm
Living in a world that is increasingly saturated with algorithmic processes has profound implications for ethics. The scholarly and public accounts of the ethical dimensions of algorithm have overwhelmingly placed the human being as the locus of ethical deliberation. Thus, philosophers and scientists search for ethical frameworks or guidelines for the human designers of software; legal scholars remind us that the protection of humans from “automated decisions made about them” by algorithms is enshrined in data protection law; and social scientists urge that the “black box” of algorithmic decisions be opened to critical scrutiny. Yet, the idea of an ethics that opens the algorithm to human scrutiny has important limits, not only because many algorithms are proprietary and secret, but more significantly because they can operate at a speeds and scales beyond the threshold of human perceptibility.
Louise’s Leverhulme Major Research Fellowship provides an account of the potential for an ethics of algorithm. In a tradition of ethics as the situated giving of an account of one’s actions at the limits of intelligibility – “I cannot give an account of myself without accounting for the conditions under which I emerge” (Judith Butler 2005) – the research excavates how the algorithm might give an account of itself and its emergence. The research develops six conceptual themes – to form the chapters of a book – opening up dimensions of the relation of algorithm to ethics: undecidability; doubt; cognition; perception; association; automation.