1.4 Pathways for Development 231.4.1.4 From Clausal Construction to Pragmatic Marker: Nominal
Relative Comment Clauses. Pragmatic markers that are type (iii) comment
clauses ( Quirk et al. 1985 : 1117) or “non- restrictive relative clause” parentheti-
cals, both nominal (e.g., what is more interesting ) or sentential (e.g., which is
more interesting ) ( Kaltenböck 2007 : 29), also cannot be accounted for by the
matrix clause hypothesis. Like an adverbial comment clause, the relative begins
life as an adjunct with limited scope modifying a clausal element, but the rela-
tive element loses its precise anaphoric reference, the clause expands its scope
(often moving position), acquires clausal scope, and becomes a sentential rela-
tive. Here it undergoes semantic/ pragmatic changes and becomes a pragmatic
marker. For example, Koivisto- Alanko and Rissanen ( 2002 ; see also Visser
1969 : 1465, 1467) note that that is to wit begins as a relative clause introduc-
ing a new or important piece of information. But that gradually gives way to it
and the clause is reduced to to wit. This comes to function as an “emphasizing
discourse marker” meaning ‘truly, indeed’ as an introductory particle. That is to
say likewise arises as a relative clause modifying a specifi c element in the main
clause (as in 11a) but undergoes scope expansion and comes to have scope over
the entire sentence (11b) (examples from Brinton 2006 : 106– 107):
(11) a. Hie was fet of weste wunienge þar he funden was, scilicet in terra deserta
in loco ... uaste solitudinis, Ðat is to seien on weste londe (a1225(?a1200)
Homilies in Cambridge, Trinity College (Trin- C B.14.52) 161 [MED])
‘he was fed from a wasteland dwelling where he was found, that is in a
deserted land in a place ... of vast solitude, that is to say in a barren land’
b. nevere truage schal we gyve, That ys to seye , whiles that we lyve (1328
Rob.Brunne, Chron. (Zetsche) 4320) [Visser 1969 : 1466])
‘never shall we give payment, that is to say, whilst we live’
In this case, reduction leads to loss of the verb: that is to say > that is.
In Chapter 7 , the rise of what I’m saying (is) and all I’m saying (is) in the
nominal relative clauses of pseudo- cleft sentences is examined. Chapter 9 con-
siders the history of what is more / which is more , focusing on several possible
routes of development.
The sources and pathways discussed here do not exhaust the possibilities for
clausal pragmatic markers. For example, an alternative source is main verb +
phrasal element, which we see in the case of I mean (see Brinton 2006 : Ch. 5).
Section 7.2 investigates that (having been/ being) said , which belongs to what
Quirk et al. ( 1985 : 1113, 1118) call type (vi) comment clauses with ed parti-
ciples. Tag questions are also another important source of pragmatic markers,
as in the case of the pervasive invariant British tag, innit < isn’t it? (e.g., Krug
1998 ; Andersen 2001 : Ch. 4; Pichler 2016 : Ch. 6). Other tag sources include
see < do you see? , hear < do you hear? , and what else (Brinton 2006 : 152,
211– 218, 253).