Portal de Periódicos da CAPES

Social Conventions: From Language to Law

2014; Duke University Press; Volume: 123; Issue: 2 Linguagem: Inglês

10.1215/00318108-2400652

1558-1470

Bruno Verbeek,

Law, Economics, and Judicial Systems

This little book brings together some of Marmor's recent thinking on conventions and offers a new application of his theory. In clear, straightforward prose, Marmor introduces an analysis of what conventions are and makes some theoretical distinctions (chapters 1–3), which he then applies to language (chapters 4–5) and law and morality (chapter 6–7).Marmor's theory is best regarded as a response to David Lewis's (1969) seminal Convention. According to Marmor, a social rule R is a convention if and only if:There is a population P that normally follows R in circumstances C.There is a reason A for members of P to follow R in C.There is at least one other potential rule, S, that if members of P had actually followed in C, then A would have been sufficient reason for members of P to follow S instead of R in C, and at least partly because S is the rule generally followed instead of R. The rules R and S are such that it is impossible (or pointless) to comply with both of them concomitantly in C (2).Several things are noteworthy about this definition. First, just like Lewis, Marmor stipulates that conventional rules are arbitrary in that other rules would be possible as well. Second, the reason for an agent to follow the convention is that most or enough others follow it as well. The reasons for an agent to comply, then, are compliance dependent (8). In the remainder of the book, Marmor often equates the two features of arbitrariness (in this specific sense) and compliance dependency (e.g., 42). In most cases, this is unproblematic, since rules that are arbitrary are also rules that are compliance dependent because for most rules that have possible alternatives, it will make sense to follow them only if others do so as well and vice versa. However, there are conventionalists who discuss rules that are compliance dependent, but not, in this sense, arbitrary.1 The final thing to note about Marmor's definition—and this is crucial—is that, unlike Lewis, he does not assume that all conventions are solutions to coordination problems (22). Chess, for instance, though conventional, is not a possible solution to some underlying coordination problem.This insight motivates two further distinctions between kinds of conventional rules that Marmor makes. First, inspired by Searle (1995), Marmor makes a distinction between “constitutive conventions” and “regulative conventions.” Examples of constitutive conventional rules are the rules of chess, whereas an example of a regulative rule would be the rule that requires one to drive on the right. After criticizing Searle's distinction between the two, Marmor—perhaps wisely—does not venture to give an explicit definition of these two types of conventional rules. Instead, he provides general characterizations and examples.The other distinction is that between “deep” and “surface” conventions (58–59). An example of such deep conventions are the various traditions in art (61). Rococo pictorial art, for example, is representational, celebrates the body, and aims to overwhelm. Impressionist art, on the other hand, aims to represent our perceptions and celebrates ordinary events and situations. Within the tradition of Rococo pictorial art, there are further surface conventions about the use of color, perspective, and pose. Here as well, Marmor refrains from offering a precise criterion or definition of what makes a convention deep and relies on an intuitive understanding that I at times found hard to follow.The second part of the book (chapters 4–5) discusses the question, what aspects of language and language use are conventional? Surprisingly “little” is the short answer. Apart from what Marmor calls “notational rules”—the rules that determine the sound-sense relations and their relations to script—important aspects of semantics and pragmatics are not conventional according to Marmor. Thus, the “literal meaning” of a word is not conventional (where the “literal meaning” just is that which a word refers to). There are a few exceptions to this claim, according to Marmor (97). For instance, words that refer to conventional rules have a conventional meaning, but these are the exception to the rule.Turning to pragmatics, Marmor discusses two questions. First, are there conventional implicatures and, second, do performative speech acts presuppose conventions. Marmor is mainly in discussion with Grice (1989) and Searle (1979). Against the former, he argues that there is no such thing as conventional implicature (implicature determined by conventions) in addition to literal meaning and conversational implicature (108). Against the latter, he argues that a detailed analysis of the various speech acts shows that only some of these presuppose background conventions in place (118–30). It goes too far to discuss Marmor's conclusions here in detail, but the general message is one of caution against broad-sweeping conclusions about the conventional nature of language.In the third and final part, Marmor turns his attention to morality (chapter 6) and law (chapter 7). In chapter 6, Marmor assumes without much argument that basic moral norms, for example, the norm against torture, are not conventional at all since one's reason to comply does not depend on whether others comply. It would seem, therefore, that there is nothing conventional about morality. However, there are conventional elements surrounding our moral norms. Sometimes there are moral reasons to follow a convention if it answers a moral need; sometimes (deep) conventions are morally significant. An example of such significance is the value of charity. Marmor argues, convincingly, that the norms that constitute and regulate charitable acts are conventional, but nevertheless, moral (150).The last chapter is about law, and here Marmor is at his best. Marmor gives a meticulous defense of the role of conventions in law that both builds on H. L. A. Hart's (1961) theory and amends it. All the theoretical work of the first part of the book is put to good use to defend the idea that the institutional rules of law are founded upon two types of deep conventions. On the one hand, there are deep conventions that determine the way law is organized (e.g., common law versus continental law) that are instantiated by (usually informal and conventional) norms about recognizing, interpreting, and changing law. On the other hand, there are deep conventions that are specific to particular legal systems (e.g., United States law versus British law). Many interpreters have argued that Hart thought that the rule of recognition, that is, the most basic norm of a legal system, is a coordination convention. Marmor argues that law rests on deep constitutive conventions that are not coordination solutions at all.Social Conventions is a nice, concise discussion of the various conventions that shape our everyday lives. While I would have preferred a bit more theoretical discussion about issues such as the emergence and stability of conventions and the reasons for complying with conventions and some more detailed discussion of other conventionalist theories, it makes a valuable contribution to current thinking about conventions, if only by showing what is not conventional about language, morality, and law and by avoiding blanket statements about these.