The Evolution of Pragmatic Markers in English Pathways of Change

(Tina Meador) #1

146 Epistemic Parentheticals


I have ignored some initial and fi nal adverbials (such as why or thus ) or inter-
jections (such as yis or nay ). In addition, there are a number of independent
uses of parentheticals, which I call “absolute.”
Sentence- initial parentheticals include forms occurring at the beginning of a
sentence as well as clause initially following a coordinating conjunction ( and ,
but , for ), conjunctive adverb, or vocative. However, the treatment of clause- initial
fi rst- person know - constructions is problematic: when the that - complementizer
following a subject– verb sequence is omitted, the structure is indeterminate
between a matrix clause and a parenthetical (see Section 1.4.1.2 ):^16


(6) a. “ I woot right wel I nam but deed,” quod she. (1387– 1400 Chaucer, CT A.Mil. 3296)
‘ “I know very well I am not dead,” she said’
b. “ Me thynketh they been lyk Jovinyan ...” (1387– 1400 Chaucer, CT
D.Sum. 1929)
‘It seems to me they are like Jovinianus ...’


In Present- day English, comment clauses and matrix clauses are potentially
distinguishable by intonation in spoken English (Quirk et  al. 1985 :  1113)  or
by a tag question test (e.g., cos I believe you can get it on the National Health ,
{ can’t you? *don’t I } [Coates 1987 : 117]). However, in Middle English, where
such evidence is lacking, we must depend upon the uncertain evidence of punc-
tuation and the decisions of modern editors. Therefore, in the “initial” category
I have included only those forms punctuated as parenthetical. If other sentence-
initial forms had been included in my count of fi rst- person parentheticals, they
would have increased number by forty in The Canterbury Tales (to fi fty- fi ve)
and twenty in Troilus and Criseyde (to 26).


16 Thompson and Mulac ( 1991 :  317)  exclude all sentence- initial forms from the class of epis-
temic parentheticals.


Table 5.2 Sentential position of fi rst- person epistemic parentheticals in The Canterbury tales and
Troilus and Criseyde


The Canterbury tales Troilus and Criseyde Total

Initial 15
17%


6
12%

21
15%
Medial 37
43%


31
61%

68
49%
Final 33
38%


14
27%

47
34%
Absolute  2
2%


 0
0%

 2
1 %

Source: adapted from Laurel J. Brinton, Pragmatic markers in English: Grammaticalization and
discourse functions. Berlin and New York: Mouton de Gruyter, 1996, p. 217; with permission.

Free download pdf