1.5 Processes of Change 35discourse- pragmatic dimension, and many aspects of pragmatics, such as topic
and focus, have a “grammatical” (syntactic) dimension (cf. Traugott 1995a : 5).
Some elements of grammar (such as disjunct adverbials ) are also asyntactic
(stand outside the clause) and carry scope over discourse. Thus, there seem to
be good reasons for viewing grammar more broadly to encompass discourse
functions. As Traugott argues:
Grammar encompasses phonology, morphosyntax, and truth- functional semantics, and
is rich enough to license interaction with the general cognitive abilities such as are
involved in the speaker- addressee negotiation ... These include information processing,
discourse management, and other abilities central to the linguistic pragmatics of focus-
ing, topicalization, deixis, and discourse coherence. (Traugott 2003a : 626)
Diewald goes further to argue that pragmatic functions are “genuine grammat-
ical functions which are indispensable for the organization and structuring” of
discourse and “the fundamental features of grammar itself are rooted in prag-
matics” (2006: 405, 2011: 451). Degand and Evers- Vermeul ( 2015 : 74) agree
that “pragmatic functions are genuinely grammatical functions which are
indispensable for the organization of spoken dialogic discourse, as well as for
the coherence of ... texts.”^39 Moreover, pragmatic elements could be seen as
having “communicative obligatoriness” ( Diewald 2010 : 25) – i.e., not obliga-
tory by language- internal criteria but by the “communicative intentions of the
speaker.” In this view, then, the development of pragmatic markers falls nat-
urally under the rubric of grammaticalization, and pragmaticalization can be
dispensed with altogether (position a).
In other works I have argued for Degand and Evers- Vermeul ’s position
(b), that pragmaticalization is a “subspecies” or subtype of grammatical-
ization.^40 Barth- Weingarten and Couper- Kuhlen ( 2002 : 357) describe the
relationship better as one of a “family resemblance” between grammatical-
ization and pragmaticalization, a view, they argue, that removes “the neces-
sity of making a binary decision as to whether a particular case is to be
included in the category of grammaticalization or not [and] ... allow[s] us
to focus on the similarities with prototypical grammaticalization rather than
on the differences.” An additional advantage of uniting grammaticalization
and pragmaticalization is cross- linguistic; given the variety of grammatical
systems, e.g., where a topic marker might be fully grammaticalized in one
language and not so in another, a unifi ed view avoids the diffi culty of the
development of a topic marker, for example, counting as grammaticalization
39 Degand and Evers- Vermeul ( 2015 : 66n.) point out that even proponents of this view of gram-
mar admit some differences between core and peripheral domains of grammar.
40 Degand and Evers- Vermeul (2015) note that Wischer’s ( 2000 : 365) conception of two types of
grammaticalization – one leading to new grammatical items on the “propositional level” and
one leading to such an item “on the textual or interpersonal level” – is consistent with this view.