viii Preface
FDG brings grammar closer to cognitive science in introducing a
Cognition module. In his chapter, Carlos Inchaurralde of the University of
Zaragoza (Spain), assesses the cognitive adequacy of FDG. While stressing
the impenetrability of cognitive processes, he points out various ways, for
example in a language like Japanese with morphosyntactic marking of
speakers’ awareness of social roles, that cognition impacts linguistic form
in a manner that can be modelled within FDG.
The following chapter, by John H. Connolly of the University of
Loughborough (United Kingdom), turns to another module of FDG, that
for the Communicative Context. He details the multiplicity of contextual
factors that can influence the form of a discourse, and proposes a notation
for representing contextual information. He distinguishes carefully between
information which belongs to the contextual module and that which is
more properly assigned to the interpersonal level. The chapter contains a
detailed application of Connolly’s proposals to an invented mini-discourse.
Discourse analysis is also central to the chapter by Francis Cornish, of
the University of Toulouse (France), but his data are all taken from attested
language use. His purpose is to show how the new FDG architecture pro-
vides a sound environment in which to account for phenomena of topic and
focus in discourse. He achieves this aim by confronting FDG with work
from the Columbia School of Linguistics (CS), which he finds to lack the
tools needed to deal with the flow of discourse; on the other hand, incorpo-
ration of CS’s concern with attention and accessibility would further
strengthen FDG treatments of language use.
Michael Fortescue, of the University of Copenhagen (Denmark), ob-
serves a certain contradiction in the fact that FDG mimics language
production while remaining a grammar. His position is that we need to ad-
dress both Process and Pattern in a complementary fashion. He illustrates
this thesis with data from Nootka, a language whose structures are heavily
controlled by considerations of discourse pragmatics. Fortescue concen-
trates on Focus, showing that in languages like Nootka it belongs to both
Process and Pattern; in a language where Focus is not encoded grammati-
cally, it will be absent from the account of Pattern. For FG, the conclusion
might be to retain the traditional architecture for Pattern and adopt the new
for Process.
The tension between static grammar and dynamic language production
also underlies the contribution by J.Lachlan Mackenzie, of the Vrije Uni-
versiteit Amsterdam (Netherlands). He proposes a variant on Hengeveld’s
proposals, dubbed Incremental Functional Grammar, in which the interper-
sonal and expression levels account for real-time processes. The