240 Jennifer Herriman
and they have a light, short subject which postpones the main informational con-
tent to the Rheme.
Section 4 compares the extraposition vs. nonextraposition of the remain-
ing subject clauses in the English and Swedish samples, and the correspondence
between them. Sections 5 and 6 examine their noncongruent correspondences and
Section 7 discusses what conclusions can be drawn from these.- Extraposition vs. nonextraposition
There are 434 subject clauses in the Swedish sample (151 att-clauses and 283
infinitival clauses), and 327 in the English sample (104 that-clauses and 223 to-
infinitival clauses) (see Table 1). The proportion of subject clauses which are
extraposed is slightly higher in the English sample (93% vs. 86%). However, the
total number of subject clauses is greater in the Swedish sample, both when they
are extraposed (372 vs. 305 statistical significance p < .001) and when they are
nonextraposed (62 vs. 22, statistical significance p < .001). There are, in fact, three
times as many nonextraposed clausal subjects in the Swedish sample.Table 1. Subject clauses in Swedish and English samples
Swedish sample
(originals and translations)English sample
(originals and translations)
Att-cl. Infin. Tot. That-cl. To-infin. Tot.
Extrapositions 136 236 372 101 204 305
Nonextrapositions 15 47 62 3 19 22
Total 151 283 434 104 223 327The number of subject clauses is also greater in the English translations from
Swedish (ET) than in the English originals (EO) (202 vs. 125) (p > .001) (see
Table 2), which suggests that there may be a translation bias due to the influence
of a higher frequency of clausal subjects in the Swedish source texts (cf. Gellerstam
1996: 61).Table 2. Subject clauses in source texts and translations
Swedish sample English sample
SO ST EO ET
That/att-clause 83 68 39 65
Infinitive clause 157 126 86 137
Total 240 194 125 202