38 Matthew P. Anstey
ent(s) of the term in question.” This suggests that neither type of UR pre-
sented above represents the pragmatic significance of the use/non-use of the
article.^23 Once we recognize from the above examples that ‘intrinsic’ and
‘generative’ can be understood as ‘semantic’ and ‘syntactic’ respectively, the
problem of the URs (PR2) can be understood as reconciling the tripartite
functional hierarchy of influence with a monolinear formal notation.^24
5.4.3. Extreme Functionalism
Strangely enough, in spite of the absence of constituent structure in FG 1 , in
1983 Dik is still willing to admit its theoretical existence. He writes
(1983b: 75): “Does this mean that every structural property must some-
where find a functional explanation? No: as with all human instruments,
the functional requirements put on a language leave quite a bit of leeway
for alternative specifications of non-functional properties; historical devel-
opments unavoidably create rudimentary properties ...; and conflicts
between different functional requirements may even create dysfunctional-
ity in given areas of linguistic organization”.
Hence one could imagine that either FG 1 would make room for such
structure or Dik’s views would change. Three years later the latter occurs.
Dik (1986b: 11) writes: “Saying that a certain feature of linguistic design
or change cannot be functionally explained is tantamount to saying that we
have not yet been able to find a functional explanation for that feature”.
Autonomous, non-functional structure is completely excised.
5.4.4. The problem of psychological adequacy
Nuyts (1983; cf. also 1985, 1992) is the first to attempt to graft FG into a
broader cognitive model of grammar and to take seriously the all-
encompassing framework of pragmatics. Nuyts concludes, “It is obvious
that Dik’s theory is not a complete functional language theory: (a) it does
not deal with cognitive structures; (b) the pragmatic rules are well-nigh ab-
sent (apart from the pragmatic functions); (c) a discourse component is
missing. ... Thus FG is intended to be a grammar in the restricted sense of
the word (a formal description of only the verbal aspects of language), and
not in its wider sense” (1983: 384).
From 1983 onwards it is clear that Dik is likewise increasingly con-
cerned with the psychological adequacy of FG 1. We can chart the