The lexical rules characterize possible lexical items of the language, and the phrasal rules characterize their
combinatorial possibilities. What makes elements of the languagerulesrather than basic elements is that they contain
typed variables (i.e. open places)—that is, they describepatternsof linguistic elements.
It is an open question how rules of grammar are to be incorporated into a model of performance, and hence into a
theory of neural instantiation. Here are three plausible options.
- Rules are (in some sense) explicit in long-term memory within the f-mind, and the language processor
explicitly refers to them in constructing and comprehending utterances. Following the computer metaphor,
rules are like data structures in a computer. - Rules are partialdescriptions of theoperations ofthe processor itself.Inthecomputer metaphor,therules as
we write the mare high-level descriptions of parts of the processor's progra m. - Rules are implicit in the f-mind; they describe emergent regularities (perhaps statistical regularities) among
more basic elements, but are not themselves implemented in any direct way.
As suggested in section 3.2, the traditional formulation of phrase structure rules like (2) and of derivational rules like
(8) is conducive to viewing the rules as like a progra mfor constructing sentences. The connotations of the ter m
“generate” in“generative grammar” reinforce such a view. However, as already mentioned, students are always
cautioned to resist this interpretation. In particular they are exhorted to view derivational movement as metaphorical:
“We are, after all, describing competence, not performance.”The upshot is that the status of such rules vis-à-vis
performance models is left unspecified. We return to this issue in Chapter 7.
Feature composition rules and constraints lend themselves better to a“data structure”interpretation, in the sense that
they delimit a space of possibilities within which linguistic entities can be located. In the approach to be worked out
here, particularly in Chapter 6, a considerable proportionof standard rules of grammar willbe interpreted in this light.
On the other hand, we will also conclude in Chapter 6 that certain classes of lexical rules fall into the“implicit”
category. Such rules are nowhere present in the f-mind in the form we write them. Rather, these rulesareindeed just
descriptions of regularities in the organization of linguistic memory. I will maintain, however, that such rules still shed
light on the nature of linguistic structure, so they are worth studying with traditional linguistic techniques.