META-ETHICAL PLURALISM AND DISAGREEMENT.

• Author: van Gorkum, Stijn

1. Introduction

   The debate over how to understand moral concepts has gone on for a long, very long, time, and although its participants do seem to have moved a bit closer to each other in some respects (with cognitivists and noncognitivists, for example, both trying to incorporate the insights from each other's work, such as the importance of moral motivation and moral reasoning, respectively), arguably the broad battle lines have not moved much, and the same brute conflicts of intuitions are ever-present.

   In light of the persistence of such conceptual disagreement (among other reasons), some theorists in meta-ethics (mainly Francen Olinder 2010, 2012 and Gill 2008. 2009) have recently argued that we should abandon the idea that ordinary moral discourse is uniform, that it can vindicate one of the traditional accounts in meta-ethics. Instead, they argue, different parts of ordinary discourse--utterances made by different speakers, or in different contexts--may involve different meta-ethical commitments, and the traditional accounts may be right about some of these parts, but at the expense of ignoring others. In order to get the full picture, we should, they maintain, instead adopt meta-ethical pluralism, the idea that moral discourse involves several concepts of morality.

   But if such an account is correct, some critics (e.g. Johansson and Olson 2015 and Sinnott-Armstrong 2009) have objected, it becomes a mystery how many disputes in both ordinary moral discourse and meta-ethics could involve genuine disagreement. Starting with the former, if different speakers were to employ different moral concepts, then that would seem to imply that they are talking about entirely different things--compare two speakers talking about a financial bank and a river bank. For example, if one speaker, saying of an action that it is morally wrong, means that it is objectively wrong (i.e. its moral status is independent of anyone's actual views on the matter), but another speaker, claiming instead that it is morally right, means merely that it is conventionally right (i.e. according to the moral norms of his society), then they do not really seem to disagree: instead, their statements seem to be perfectly consistent, and the speakers seem to be talking past each other.

   And, the critics go on, pluralism would also call into question disagreement in meta-ethics: if proponents of different theories are not offering competing analyses of the same concept but are rather targeting different concepts of morality, then their dispute again does not seem genuine--to use the bank example again, imagine two philosophers arguing over the correct analysis of the word 'bank' while focusing on financial banks and river banks, respectively. If competing theorists are not analyzing one shared concept of morality but are rather focusing on different concepts, they, too, would seem to be talking past each other, and that is at least highly counterintuitive to many.

   As it happens, however, both objections rely on assumptions that can be, and have been, questioned. To begin with first-order moral disagreement, it turns out that explaining it is not only a problem for pluralism, but for certain other theories in meta-ethics as well. For example, if moral claims are relativized to societies (cultural relativism), or if their content varies by context (moral contextualism), then disagreement between societies, or across contexts, again seems (critics argue) to have been lost. And the solutions offered to help those theories, I will argue, can be taken up by pluralists as well. In a nutshell, the argument is that, even if speakers will sometimes fail to disagree about the literally expressed (that is, semantic) content of what they are saying, they may still have a disagreement about content that is communicated differently, as a matter of pragmatics (why they are saying it): for example, speakers may each use their different moral concepts to express (pragmatically) where they stand with respect to an action, leading to a so-called 'disagreement in attitude'. And such disagreements are perfectly consistent with the semantics of their claims not conflicting.

   And likewise, the idea that conceptual analysis in meta-ethics proceeds in the 'top-down' manner (analyzing a single, clearly delineated concept) hypothesized by critics of pluralism has been challenged: it may instead operate in a more 'bottom-up' kind of way, where the focus is on trying to discover what, if anything, particular uses of certain terms in a certain area of discourse have in common. And if that is how things work, meta-ethicists who operate with different concepts could still be having a genuine disagreement--namely, about how a certain area of discourse is to be conceptualized. And meta-ethicists could also still dispute which moral concepts are best for various purposes, which ones we should adopt.

   Of course, such strategies may well fail, and if they do, pluralism will fall with them (or will at least have to look elsewhere for a solution)--and there are many other objections unique to it (not discussed here) as well (e.g. the 'challenge from unity' described by Francen Olinder (2012), the objection that moral terms fail standard tests of ambiguity (Sinnott-Armstrong 2009), and the argument that pluralism does not go far enough, because it still hopes to find commonalities at lower levels of generality (De Mesel 2016)). However, provided the strategies do not fail in any decisive manner, the simple inference from the presence of disagreement to the falsity of pluralism is undermined, and that is at least a small victory for the account.

   I will make my case over four sections. In the first section, I will explain in some more detail what meta-ethical pluralists are committed to, exactly, and will quickly go over some arguments for their position. In the second section, I will then explain why many have thought that accounts like relativism and contextualism ('content-relativism' for short) are unable to account for many seemingly legitimate cases of first-order moral disagreement, and how pluralism faces much the same problem, but in a more radical form. In the third section, I will go on to discuss how content-relativists have tried to overcome this challenge, and will argue that pluralists can help themselves to similar resources to explain the cases of lost disagreement. Finally, in the fourth section, I will discuss the other disagreement-based objection mentioned, concerning disagreement not in ordinary discourse but in meta-ethics, and will argue that plausible rescue strategies are again available.

2. Defining meta-ethical pluralism

   Before discussing the details of meta-ethical pluralism, I should first explain what it is supposed to be a theory of. Pluralism, as I will understand it here, is a theory in moral semantics: in other words, it is an account of what our moral terms mean. And more specifically, it is an account of what they ordinarily mean--not of what they mean in a highly specialized context like a certain area of philosophical discourse, for example. In other words, pluralism offers to give an analysis of ordinary uses of moral terms, which concept(s) they are associated with. How does it propose to do so?

   Although he does not make use of the term 'pluralism' to describe his position, the view I have in mind is most clearly stated in the work of Michael Gill (2008, 2009). Meta-ethics, he argues, has traditionally proceeded on the assumption that ordinary moral discourse is uniform (the 'uniformity assumption'), that all ordinary uses of moral terms have particular commitments in common that give us conditions that are individually necessary and jointly sufficient for their application. (1) Some have argued, for example, that all moral judgments are about the norms of a speaker's society (cultural relativism); others claim that they concern universal standards that bind everyone regardless of what standards they actually hold (moral objectivism); and so on and so forth. Pluralists argue, instead, that ordinary uses of moral terms have no universal features in common.

   However, it is important to note that the rejection of the uniformity assumption is compatible with several different views in meta-ethics, and what I mean by pluralism is only one of them. Following Gill (2008, 2009), we may call any view that rejects the uniformity assumption--that is, any view that maintains that no classical analysis (in terms of necessary and sufficient conditions) can be given of moral discourse as a whole (because it lacks the uniformity required for that: there are no substantive commitments that all moral claims have in common)--a form of 'variabilism'. Pluralism, as I will understand it here, goes beyond variabilism by contending, in addition, that classical definitions, corresponding to several different concepts of morality, can be given for more specific parts of moral discourse. Or as Gill puts it: "while some parts of ordinary moral discourse are most accurately analyzed as involving a certain meta-ethical commitment, other parts of ordinary discourse are most accurately analyzed as involving the meta-ethical commitment that has traditionally been taken to be its meta-ethical competitor" (Gill 2009: 216). (Other, non-pluralistic variabilist theories include family resemblance theories (e.g. De Mesel 2016), Loeb's (2008a, 2008b) moral incoherentism, Polzler and Wright's (2020) anti-realist pluralism, and possibly also Sinnott-Armstrong and Wheatley's (2012) view that moral judgment is 'disunified'; and Finlay (2019) argues for ambiguity in metanormative theory, about concepts of normativity, rather than in meta-ethics narrowly construed.) For example, to continue with the above example, perhaps some moral claims are best understood in a relativist way (when two speakers have contrary opinions about a controversial issue but do not think either of them is in...