I’m currently working on a paper that examines whether expressivism understood as a theory of the metasemantics of normative discourse is compatible with normative realism. Here is how I came to be interested in this question.
Expressivism, at the most general level, is the idea that normative claims have the function of expressing desire-like mental states, like plans, attitudes of approval and disapproval, or states of norm acceptance. This view was traditionally pursued as a semantic project: its centerpiece was expected to be a psychologistic semantics that would match normative expressions with desire-like mental states and thereby account for the compositionality and logical behavior of normative language. This project was also seen as incompatible with normative realism: it was understood as an ontologically austere theory, on which normative discourse does not aim to describe normative facts.
Both pillars of this traditional expressivist project have come under pressure in recent times.
First, the contrast between expressivism and realism has become blurred. Expressivists like Simon Blackburn (1993, 1998) and Allan Gibbard (2003, 2012) now endorse core tenets of normative realism—e.g., that there are objective normative facts, or that our beliefs about these facts are by-and-large true—by adopting a deflationary approach to truth, fact, belief, and other related notions. On this deflationary account, realist commitments are treated as internal to normative discourse: for instance, to say that “Genocide is wrong” describes a fact is simply to rehearse the verdict that genocide is wrong. Nonetheless, Blackburn and Gibbard label their view “quasi-realism,” a term meant to indicate that, for all the realist-sounding claims, we are still dealing with an anti-realist view of normativity. This matters, they argue, because quasi-realism has important advantages over realism: it explains the connection between normative judgment and motivation, and makes good on commitments to truth and objectivity in the normative domain while avoiding the metaphysical and epistemological problems faced by realism. But it is contentious whether and how a stable divide can be drawn between this view and genuine realism.
Secondly, new versions of expressivism have been proposed which do not purport to offer a psychologistic semantics for normative discourse. These views restate expressivism as a metasemantic view: that is, as an account of why normative expressions have the meanings that they do, rather than as a theory of what those meanings are. (See, e.g., Ridge 2014 and Chrisman 2016.)
A key motivation for these proposals is the aim to preserve expressivist insights about the nature of normative meaning while avoiding the well-known difficulties faced by semantic expressivism in accounting for the meaning of normative terms in logically complex sentences—the cluster of problems known as the Frege-Geach problem. The shift to metasemantics allows expressivism to be compatible with an orthodox, well-understood truth-conditional semantics for normative terms, including truth-conditional accounts of validity, logical connectives, etc. Thus, Frege-Geach issues disappear, or can be addressed using tools available to cognitivists as well.
A second motivation for redefining expressivism in metasemantic terms is the aim to fit expressivism within a particular truth-conditional semantic model that has become paradigmatic in linguistics and the philosophy of language, namely a Kratzer-style contextualism for deontic and evaluative terms. (See, e.g., Kratzer 2012.) There is room for expressivism within this contextualist picture. For example, Ridge proposes that desire-like attitudes help determine the semantic content of deontic and evaluative terms in normative contexts of use, while Chrisman defines expressivism as a view about what we mean when we attribute context-sensitive meanings to normative claims.
I am sympathetic to this new understanding of expressivism as a metasemantic view, but I am also attracted to the quasi-realist project and have long been interested in whether there is any meaningful divide between quasi-realism and genuine realism. This is how I arrived at the main question of this paper: how should we think of the relation between expressivism and realism, if we accept some version of metasemantic expressivism but also use a deflationary account of truth, fact, etc. to make good on realist-sounding claims about normativity?
Someone might argue that it is precisely by articulating expressivism in metasemantic terms that we can draw a clear divide between quasi-realism and realism: while the two views might agree in their attributions of truth-conditional meanings to normative terms, they disagree about how to explain such semantic content. Realists explain meaning by invoking representational relations between normative language and the world, e.g. truth and reference relations, while expressivists appeal instead to desire-like attitudes in explaining why normative claims have the meanings that they do. Thus, the metasemantic conception of expressivism seems to dovetail with the most popular account of what separates quasi-realism and realism, namely the “explanation” explanation proposed by Dreier (2004) and others, e.g. Blackburn (1993), Fine (2001), Gibbard (2003), Dunaway (2016), Simpson (2018).
However, I want to argue that, if we adopt a deflationary approach to the relevant metaphysical and semantic notions (truth, fact, representation, etc.), there need be no explanatory divide between metasemantic expressivism and anything recognizable as a plausible notion of normative realism. In other words, that expressivists who adopt such a deflationary perspective should think of their view as vindicating realism pure and simple.
In my next blog post, I will give an overview of my arguments. (I’m still working on them!)