10.5 Envoi 295manner/ degree adverbial ( admittedly ≠ ‘in an admitted manner’). Therefore, an
alternative course of development from (verb) > adjective > disjunct adverbial
seems more plausible here, as the use of admitted as a pure adjective predates
its use as a disjunct adverbial. But there is also historical evidence that the It/
that be Adj/ Part that construction contributes to the rise of the disjunct adver-
bial, as the sentential structures antedate the appearance of the disjunct adver-
bial and likely undergo some process of reduction to produce the adverbial
form (though the details of this process are not clearly understood).
As noted briefl y in Chapter 5 (see Section 5.6.1.1 and 5.6.1.2 ), Old English
possesses a small set of epistemic (disjunct) adverbials such as soðlice , witod-
lice , sweotollice , gewislice , openlice , and cuðlice. But it also has a number
of periphrastic expressions of the form It/ that be Adj/ Part that , such as hit is
sweotol þæt ‘it is clear that,’ hit is soð þæt ‘it is true that,’ and þæt cuð is þæt
‘it is quite clear that.’ We fi nd a fairly substantial class of epistemic adverbi-
als in Middle English, including certes , trewly , forsoothe , soothly , verily , sik-
erly , peraventure , nedes , iwis , certeynly , plainly , witerliche , dowteles , clerly ,
faythfully , in dede , evidently , dredeles , redyly , surely , clierliche , possibly ,
actually , aperteliche , patently. While some of these disjuncts are direct bor-
rowing from French , some undoubtedly develop from adjectives (borrowed
or native), by the process described above, from higher structures It/ that
be Adj/ Part that , while other may derive from adjunct adverbials directly
into disjunct adverbials. Detailed examination of each of the separate forms
would be necessary in order to determine which path of development the
disjunct followed.
10.5 Envoi
In Brinton (2008: 254– 256), I suggested that a possible new approach to prag-
matic markers was provided by Construction Grammar. The Construction
Grammar approach might have advantages over the grammaticalization
approach because, as Hilpert (2013) observes, “each construction has its
own history” (3) and “constructional change rarely proceeds just along a sin-
gle dimension. Rather, many formal and functional aspects of a construction
change simultaneously. Not all changes will maintain the same direction over
time” (6).
Although the non- propositional material of spoken discourse (e.g., pragmatic
markers) may not initially seem to be encompassed by Construction Grammar,
Fried and Östman ( 2005 ) argue that pragmatic markers appear in recurring
patterns, serve conventionalized functions, are pragmatically constrained,
and form a system; hence, they cannot be dismissed or seen as not belong-
ing to grammar. They suggest that in order to account for pragmatic markers
(and other features of spoken language) within Construction Grammar, three