A SYNOPSIS OF ROLE AND REFERENCE GRAMMAR 157- It appears that this clause-initial focus position in questions is the PCS; definitive
proof of this requires an analysis of Russian clause structure which is beyond the
scope of this discussion. - Normally, the answer would be simply "Maksim" and the topical elements would
not be repeated. However, they are included here in order to better illustrate the
focus structure constraints on word order. - See Comrie (1979, 1984) for detailed discussion of these and other constructions.
The full range of Russian focus constructions and word order possibilities will not
be explored here; the constructions in (78) are sufficient for illustrating the
interaction of focus structure and linking in a "free-word-order" language. - Ne cannot, however, realize the quantified single argument of attributive/identifi-
cational essere "be", which otherwise behaves like an "unaccusative" verb, and
this follows from the RRG analysis; see Van Valin (1990a), Schwartz (this vol
ume). - It was noted in fn. 53 that ne does not occur with attributive/identificational essere
"be" (see Schwartz, this volume), an otherwise "unaccusative" verb, and there are
clear pragmatic reasons for it: attributive/identificational "be" does not function
as a presentational verb, in Lambrecht's sense (1987: 374). But, as Schwartz
shows, locative and existential uses of essere do allow rce-cliticization, and this is
predicted by the semantic analysis in Van Valin (1990a) and by this pragmatic
analysis. This reinforces the semantic explanation for the non-occurrence of ne
presented in Van Valin (1990a). - In FVV, clausal junctures were referred to as "peripheral junctures".
- For detailed discussion of these points, see FVV, chapters 5 and 6 (especially pp.
271-2), Van Valin (1984), Foley & Olson (1985), Silverstein (1976), Givón (1980).
It should be noted that the semantic relations at the top end of the hierarchy are
often lexicalized and not realized by a complex syntactic construction, e.g. causal
ity as in Japanese or Turkish. In this situation, the tightest syntactic linkage will
not instantiate the top semantic relations, but this is not a counterexample to the
claims of the IRH. Indeed, the fact that it is the strongest semantic relations that
are grammaticalized into morphological constructions, replacing the nuclear
junctures, follows the basic claim of the IRH: the stronger the semantic relation,
the tighter the morphosyntactic bond between the units, and the evolution from a
tightly bound syntactic construction to an even more tightly bound morphological
construction represents a natural extension of the iconic relation between form
and meaning expressed in the IRH. - Of course, illocutionary force can never be independently specified in a subordi
nate clause. Cf. section 6.6. - For arguments as to why the complement clause in Figure 35 is not syntactically a
core argument, despite being an argument in the verb's LS, see FVV: 251-255.