FG from its inception 39problematic course that Dik sought to navigate in this regard. The problems
begin when Dik makes a crucial redefinition of psychological adequacy. In
FG 1 psychological adequacy is defined weakly as “[a grammar] should not
be incompatible with strongly validated psychological hypotheses about
language processing” (Dik 1978a: 7 [emphasis mine]). But in 1983 he
writes, “... FG would like to relate as closely as possible to psychological
models of linguistic competence and linguistic behaviour. ... [W]hy should
a grammar be neutral as between producing and comprehending linguistic
expressions, when producing and comprehending such expressions is just
what the grammar is there for?” (1983b: 76 [emphasis mine]).
Dik (1986a) attempts to integrate FG 1 into a theory of verbal interac-
tion, or more precisely, “an integrated model of linguistic interpretation”.
Psychological adequacy reaches its most extreme (1986a: 6): “psychologi-
cal adequacy implies that a grammar developed according to FG
specifications should be a good candidate for incorporation into operational
models of natural language users” (cf. Dik 1989b: 13, 1997a: 13).
But clearly, for such a model to be accurate, it must represent humans
and their participation in verbal interaction. Dik adopts a simple but haz-
ardous strategy (1986a: 2): “the form of ... knowledge representation
structures can be derived from the theory of Functional Grammar.” In
1989, this equivalence between URs and human cognition is taken to its
logical conclusion, where he proposes the following equation (Dik 1989a:
100):
(8) Lcog =Lur = Lkr = Llog
Lcog = general cognitive representation language
Lur = language of underlying representations
Lkr = language of knowledge representation
Llog = logical language for logical reasoning.
Not surprisingly, a thorough critique was forthcoming. Hesp (1990) ap-
praised Dik’s computational Functional Grammar model, raising serious
doubts about the possibility of Dik’s interpretation of psychological ade-
quacy. Due to the subsequent inattention to this issue, the role and
formalization of conceptual representations in Functional Grammar re-
mains unresolved.
Finally, a double irony in the history of Functional Grammar’s relation-
ship to psychology should be noted. One the one hand, since its inception it
has been a theory that explicitly seeks psychological adequacy, and yet it is
formal linguistic theories such as Chomsky’s that have been the most in-