30 Pragmatic Markers: Synchronic and Diachronic
Thus, there are indeed differences between the development of pragmatic
markers and prototypical cases of grammaticalization, in respect to increased
scope, syntactic fi xedness, and lack of bondedness.^32 At the same time, how-
ever, it has been shown that no cases of grammaticalization exhibit all of the
characteristic features.
1.5.2 Lexicalization
Discomfort with the idea that pragmatic markers result from grammatical-
ization has led scholars to suggest that they might be the result of lexical-
ization. While the defi nition of lexicalization is far from unproblematic (see
Brinton and Traugott 2005 ), it is often understood as a process by which a
free syntactic construction fossilizes, fuses, and coalesces and thus comes to
be treated as an unanalyzable lexical item. A complex unit becomes irregular,
opaque, and often monomorphemic, e.g., nuts- and- bolts ‘essentials’ (not bolts-
and- nuts ), out- of- pocket ‘paid out in cash.’ Fusion may occur, e.g., OE n ē ah-
geb ū r > neighbor. Thus, the concept of lexicalization captures the univerbation
that many pragmatic markers (especial phrasal and clausal ones) undergo. For
example, we see some degree of fusion and coalescence in pragmatic markers
such as you know > y’know , I dare say > I dessey , and look ye/ thee > lookee
and occasional ellipsis of phonological content in pragmatic markers such as
I guess > guess , as you say > you say , say to me/ us > say , look you > look
(see Brinton 2008 ). As a result, the complex unit becomes irregular and often
opaque. Further characteristics of the development of pragmatic markers that
are compatible with lexicalization include loss of semantic compositionality
(i.e., idiomaticization), syntactic independence, and “holistic” treatment of
the form as a lexical item. Understanding lexicalization as a complex phrase
becoming a monomorphemic word, Krug ( 1998 ) argues that is it not? > in’t
it > innit becomes invariant, inseparable, and morphologically opaque. There
is desemanticization of it , loss of phonological substance, and acquisition of
pragmatic function. Wischer ( 2000 ) sees methinks as undergoing lexicaliza-
tion: It is fossilized, unproductive, stored as a whole in the lexicon, and classi-
fi ed as an “adverb.” Discussing I think - type comment clauses, Fischer ( 2007a )
observes that the processes of decategorialization and divergence undergone
by the forms are common to both lexicalization and grammaticalization, but
suggests that the bonding argues for lexicalization; furthermore, she sees them
as retaining much of their lexical meaning.
32 Heine ’s assertion (2013: 33) that “neither scope extension nor increase in pragmatic-
metatextual functions or syntactic independence are in any way characteristic” of grammati-
calization seems an overstatement. On scope extension in grammaticalization, see Tabor and
Traugott ( 1998 ).