Page 193 - Encyclopedia of Philosophy of Language

between regulative rules, serving to regulate preexisting activities (traffic regulations), and so-called constitutive rules, which define insti- tutions and create new types of action (like checkmating).
(g) 'Rule' is tied to a notion of generality in a way which, for instance, 'instruction' is not and which is often taken to exclude overlap with the use of 'command.' Two kinds of generality are usually seen as characteristic of rules. On the one hand a rule concerns a type of action; it can be violated or complied with indefinitely many times. On the other hand it concerns agents generally, or agents in a type of situ- ation; it can be violated or complied with by an indefinite number of people. Although this characterization is not without problems it points at a feature which can be claimed to be essential to the concept.
(h) There is a use of 'rule,' as in 'strategic rule' and 'rule of thumb,' such that an item of this more factual kind (sometimes called technical) is sub-
ject to direct justification: does complyingwith it generally lead to desired results? It is essential here that what counts as a desired result iswell- determined; stating the purpose may be part of stating the rule (If you want to...) or else the purpose may be clear from the area of appli- cation of the rule (as in strategic rules of chess). This double usage is convenient, since some- thing may be called a rule whether it imposes or just registers a regularity (such as a gram- matical one).
In explicating the concept of a rule some writers are content to elaborate on features such as those already listed. Some proceed to analyze the structure of rules. Von Wright (1963), by using 'norm' as the most gen- eral term, distinguishes between 'character' (obli- gation, permission, prohibition), 'content' (type of action or activity; that which is obligatory, permitted, prohibited) and 'condition of application' (a condition that is met in a situation where someone can act in accordance with or against the rule). He also dis- tinguishes between categorical and conditional rules, and between positive, negative, and mixed rules, depending on whether the content is a type of doing something, forbearing to do something, or a mixed complex of these. This analysis is then fitted into a so- called deontic logic, that is, a logic of rules (norms, or better, norm-statements).
Explications or analyses more or less similar to that of von Wright have been given within ethics and phil- osophy of law. What is generally missing in such treat- ments are distinctions between kinds of correctness: an utterance may be semantically correct and yet a violation of etiquette. That is, the particular respect in which actions are evaluated with respect to a rule is not perceived as corresponding to an ingredient in the rule itself.
1.2 What Rules Are
The question of what rules really are has received much less attention. It is common to think of rules as abstract entities. Some, however, take them to be linguistic, while others take them to be nonlinguistic. Ross, for instance (1968), takes rules to be a species of directives, themselves intrinsically normative entities that are meanings of prescriptive sentences, like prop- ositions are of descriptive sentences. Although rules have even been thought to be particular inscriptions of rule-sentences, the concept of a rule is normally distinguished from that of a formulation of a rule, as described by Max Black (1962). Black, however, denies that rule-formulations designate, describe, or even express rules (as their meanings). Instead, to understand what a rule is we must look to the use of rule-formulations. This is in line with Wittgenstein's later philosophy.
To Wittgenstein (1958) the concept of a rule is a family resemblance concept: members of the family of rules have various features in common with other members, but it is misguided to look for any defining feature common to all members, an essence of rules. To understand the concept of a rule, consideration should be given not only to what is called a rule, but also to all that is involved in a rule-following practice, including training, explaining (how to proceed), jus- tification (and limits thereof), and evaluation of actions by means of reference to rules.
Many other theorists insist on social function or social acceptance as part of what it is to be a rule. Ross, for instance (1968), takes a rule to be a rule of some community, a (general) directive corresponding to social facts, being generally complied with in the community. Bartsch (1987) characterizes norms, social rules, as the social reality of correctness notions. This feature is particularly prominent in Shwayder's attempt at a truly informative explication (Shwayder 1965). Roughly, a rule (in the primary, communal sense) is a system of expectations in a community concerning behavior of its members, such that (a) members believe other members to have the same expectation, (b) the expectations of others constitute the reason for a member to act in accordance with them, and (c) members expect that other members conform for this reason. This idea has been developed and refined by David Lewis for the notion of a con- vention (regarded by Lewis as a kind of rule) and has, via Lewis, given rise to a whole tradition of varieties of the approach.
2. Linguistic Rules
The notion of a linguistic rule is perhaps most immedi- ately associated with very general rules of traditional school grammars; rules of spelling (e.g., nn never occurs before t), phonological rules (e.g., voiced end- ings turning voiceless in certain contexts), mor- phological rules (e.g., endings of regular verbs in
Rules
171