178
I/You Admit and Admittedly
(b) already independent clauses, e.g., I know, apples he likes.
These are infl uenced later by main clauses with deletion of that.
6.5.2 Disjunct Adverbials
Disjunct adverbials are typically assumed to arise from manner or degree adverbs
(see, e.g., Swan 1991 : 411). This development involves:
(a) scope expansion (from word modifying to sentence modifying);
(b) pragmatic reorientation (increased subjectivity , speaker focus);
(c) semantic change from concrete situation to discourse context; and
(d) syntactic shift to (disjunct) initial position, less often to terminal position (see
Swan 1988a , 1988b , 1991 ; Breivik and Swan 1994 : 19).
This shift has also been discussed by Traugott ( 1995a ) in regard to the develop-
ment of pragmatic markers from clause- internal adverbials via an intermediate
stage of sentence or IP adverbials, as discussed in Section 1.4.1.1.
As with epistemic parentheticals , a dual source for disjunct adverbials has
been proposed. Swan ( 1988a : 11) suggests that they result from blending of
manner/ intensifi er adverbs (word modifi ers), and (in the majority, but not all
cases) non- adverbial sentence modifi cations, “speaker comment” sentences,
often in the form It is Adj that. For example, the manner adverb strangely ‘in a
strange manner’ and the construction it is strange how/ that have both contrib-
uted to the rise of strangely as an “adverbialized speaker comment.”
The historical, underlying structures, then, of English sentence adverbials (or rather, sen-
tence adverbial- shifts) are ... a very varied and rich class of intensifying adverbial modi-
fi ers as well as It is ADJ structures, the one grafted onto the other, as it were, via blends, and
with what one might call syntactic consequences (e.g. the move towards initial position).
(Swan 1991 : 420; also 1988b : 531)
Similarly, Fischer argues ( 2007b : 280– 297, contra Traugott 1995a ) that word-
order evidence suggests that there is not a direct line from adjunct to disjunct
adverbial.^11 Rather, the form occurs in a separate, higher clause (generally occur-
ring sentence initially) which then undergoes ellipsis: “the wide- scope sentence
adverbial/ pragmatic marker must originally have been placed outside the main
clause, in the form of a prepositional phrase, a reduced clause or a predicative
clause followed by a ‘that’- complement” (2007b: 296). The change in scope
can only come about, Fischer explains, because the form occurs in a separate
predication. For example, þæt is sarlic þætte S ‘it is sad that’ > sarlice ‘sadly.’
11 In Old English, topicalized adverbs have Adv Subject V order (i.e., they do not cause verb-
second order). This suggests that they are clause external (see Swan 1988a : 7, 1988b: 227– 232;
Breivik and Swan 1994 : 15; Fischer 2007b : 275, 287– 288).