Abstract

In this paper, I deploy an argument that I have developed in a number of recent papers in the service of three projects. First, I show that the most influential version of legal positivism – that associated with H.L.A. Hart – fails. The argument’s engine is a requirement that a constitutive account of legal facts must meet. According to this rational-relation requirement, it is not enough for a constitutive account of legal facts to specify non-legal facts that modally determine the legal facts. The constitutive determinants of legal facts must provide reasons for the obtaining of the legal facts (in a sense of “reason” that I develop). I show that the Hartian account is unable to meet this requirement. That officials accept a rule of recognition does not by itself constitute a reason why the standards specified in that rule are part of the law of the community. I argue that it is false that understanding the explanatory significance of officials’ acceptance of a rule is part of our reflective understanding of the nature of law. The second project of the paper is to respond to a family of objections that challenge me to explain why normative facts and descriptive facts together are better placed to provide reasons for legal facts than descriptive facts alone. A unifying theme of the objections is that explanations have to stop somewhere; descriptive facts, it is suggested, are no worse a stopping place than normative facts. Third, the paper spells out a consequence of the rational-relation requirement: if an account of what, at the most basic level, determines legal facts is true in any possible legal system, it is true in all possible legal systems. For example, if a Hartian account of legal facts is true in any possible legal system, it is true in all possible legal systems. I use this all-or-nothing result in my critique of a Hartian account, but the result is of interest in its own right.