Overlapping Reasons

Since at least 1930, theorists working in moral philosophy and normativity more generally have recognized the importance of accounting for trade-offs in normative theories of what we ought to do.1 The most common way to do this is to appeal to contributory or pro tanto reasons. The thought is that the reasons for and against the options open to the agent interact and compete—in highly complex ways—to support a conclusion about what the agent ought to do, all things considered. On most views, the reasons compete by contributing normative weight for or against the relevant option.2 This thought already raises many interesting questions, and this paper will interact directly with these issues. One goal of the paper, as we’ll see, is to bring out an underappreciated complication for this picture. But our main goal concerns what is largely an orthogonal issue, about the nature of reasons themselves, rather than how they interact to support conclusions about what we ought to do. We argue for what we call a ‘bottom up’ theory of normative reasons, according to which reasons are grounded in some underlying property or state. The most popular bottom up theories are desire-based theories (e.g., Schroeder (2007), Manne (2016)) and value-based theories (e.g., Wedgwood (2009), Maguire (2016)); we also include duty-based theories such as Ross (1930). These theories share a common structure (or at least certain structural features); since our argument is primarily about the structure of reasons, we remain neutral about which bottom up theory is correct. The main goal here is to argue in favor of bottom up theories, as compared to Reasons Fundamentalism, as defended by Scanlon (2014), Parfit (2011), and others. What’s distinctive of Reasons Fundamentalism, for our purposes, is the claim that reasons are not grounded in anything at all, but are rather fundamental. (More carefully, reason relations are fundamental.) The argument turns on a class of phenomena that we group under the heading overlapping reasons. As we’ll see, we group several different kinds of cases under this heading. What they have in common is that they are cases in which the reasons—that is, the facts which are reasons—come apart in some way from what we call the normative contributors—the things that contribute normative weight for or against an option.