"Decision usefulness" has been the organizing criterion for accounting policy and accounting scholarship for over forty years. Its authority however, was not gained over time through explicit theory development or argumentation, but was instead "born full grown" (Staubus, 1999, p. 163)as a conceptual "takeover in the dead of night" (p. 338). In essence, the concept of decision usefulness assumed a central place as the rationale for accounting choice by stealth rather than by careful argument or based on empirical evidence. The lack of an extended period of argumentation and debate over the meaning of decision usefulness meant its conceptual weaknesses were allowed to remain remarked upon but never resolved.In this paper we argue that as the central focus defining and guiding accounting policy, decision usefulness has not proven any more useful than earlier, allegedly normative theories of accounting did. The recent debates on IFRS versus GAAP provide vivid testament to the inadequacy of decision usefulness as a criterion for selecting among alternative accounting choices. We briefly examine the history of decision usefulness and its emergence as the chief criterion for selecting policy. We then argue that in light of recent developments in our understanding of human decision-making and the shortcomings of conventional economic understanding, decision usefulness cannot be defined in a manner that allows its application to policy choice at either the micro (individual decision-maker) level or the macro, economy-wide level. . We find that the current concept of decision usefulness is incoherent because policy makers and scholars have not seriously dealt with deeply-flawed ontological assumptions inherent in its definition and justification. We also describe how the decision usefulness criterion applied in the policy realm seriously undermines the current dominant methodology employed in accounting research to understand financial reporting phenomena. In our conclusions we argue for a root metaphor for accounting policy and research that is more consistent with the regulatory function of accounting in society, i.e., a more explicit return to accountability for understanding accounting and for making accounting policy.3