THREE LAWS OF ROBOTICS AS ETHICAL “CODE” AND NARRATIVE MACHINE IN “THE CAVES OF STEEL” BY ASIMOV

Abstract

Isaac Asimov’s Three Laws of Robotics have been repeatedly read as an early “code of ethics” for artificial agents and, simultaneously, as a generative narrative constraint that produces plots built from rule-conflict. This article reviews
and synthesizes scholarly approaches to the Laws with a focused re-reading of “The Caves of Steel” (1953) as a text in which ethical codification functions less as solution than as engine: it manufactures dilemmas (conflicts of duty), paradoxes
(inconsistent imperatives under uncertainty), and interpretive disputes (what counts as “harm,” “human,” and “obedience” in a socially stratified world). Drawing on machine-ethics critique that stresses the Laws’ under-specification and their
dependence on capacities real machines do not possess (e.g., context mastery, reliable perception, and accountable agency) , and on literary criticism that frames the Laws as “literary machines” that generate narrative effects , I argue that “The Caves of Steel” uses the Laws to stage a double drama: (1) the procedural drama of a techno-detective investigation in which a humanoid robot’s law-bound behavior is both clue and obstacle; and (2) the sociopolitical drama of “C/Fe” cohabitation, where legalistic moral programming collides with human prejudice, institutional pressure, and geopolitical negotiation. The novel thereby anticipates contemporary debates in AI ethics about rule-based governance, interpretability, and the political construction of “safety,” while insisting that ethical “codes” do not eliminate moral responsibility but redistribute it among designers, institutions, and publics