A further clue to their origin was supplied by Lachlan Mackenzie (p.c.),
who relates this story: “I asked Simon why there were so few references to
other literature in the pragmatic functions chapter in Dik (1978). He said
that he had finished most of the rest of the book, but needed to add a vital
chapter on pragmatic functions. To do so, he needed to shut himself off
from the world for a few days, and therefore booked a hotel (in London, I
think it was). He took no or a few (I can’t remember) books with him, and
composed and completed the chapter in a few days. What references there
are were added later, back at his desk so to speak. This must all have been
around 1976/1977”.
In this structure ‘type’ is the lexical category (verbal, nominal, adjectival),
xn are terms, ‘s.r.’ are selection restrictions on the term, and SF is the se-
mantic function of the term in the state of affairs (SoA) designated by the
(nuclear) predication. It is hard to know where Dik derives the idea of selec-
tion restrictions from, since the only clue is an allusion to Weinrich (1966).
The notion was regularly used in generative writings in the 1960s (Katz and
Postal 1964: 15; Chomsky 1965: 95).
For recent critical evaluations of predicate frames in Functional Grammar
see Mairal Usón and Van Valin (2001) and García Velasco and Hengeveld
(2002).
He also dislikes Dik’s chapter on pragmatic functions: “It is deplorable that
Dik’s FG only sporadically and clumsily stumbles towards the findings of
the Prague school, rather than taking its starting point from there. ... Regret-
fully, I must conclude that this chapter of Dik’s book lags behind the other
ones in scientific importance” (Piťha 1980: 268).
This reconstruction is purely hypothetical and other Functional Grammari-
ans undoubtedly view the situation differently. The point is that it is only by
understanding if and how Functional Grammar accounts for syntactic struc-
ture that the theory can be critically evaluated.
‘Syntactic’ in FG is perspectival and not related to constituent structure.
Thus, the symbols ‘:’, ‘,’ and and – that is, restriction, apposition, and coor-
dination – represent micro-functional semantic relations inside the term
structure, which in turn enters into macro-functional relations with other
terms and predicates.
As evidence, consider Dik’s (1981a) article on ‘displacement’. It can also
be observed that between FG 1 (1978) and Bakker and Siewierska (2000)
few publications discuss autonomous constituent structure.
The hierarchy is simply a succinct way of expressing the oft-quoted position
of Dik (1987b: 7): “... pragmatics is seen as the all-encompassing frame-
work within which semantics and syntax must be studied. Semantics is
regarded as instrumental with respect to pragmatics, and syntax as instru-
mental with respect to semantics”.
