Recurrent word-combinations in contrast 197
This case study started from the observation that some recurrent time expressions
are more frequent in English translations than in originals. It turned out that they
all have Norwegian main correspondences that are more frequent in Norwegian
originals than in translations. These correspondences are not necessarily 3-grams,
which is why a comparison of similar Norwegian lists of 3-word combinations did
not reveal a similar tendency. What is most interesting, however, is that the 3-word
combinations in the English translations and originals may point to a real differ-
ence between the languages, namely that Norwegian expresses temporal relations
in the form of adverbials more often than English does. The greater frequency of
time adverbials in Norwegian than in English is also suggested by other investiga-
tions (Hasselgård 2009 and in prep.). However, the extent of this difference, as well
as the reason for it, needs to be examined further on a broader basis.
- Concluding remarks
The exploratory bootstrapping approach used in these case studies has proved to
be a good starting point for contrastive studies.
- It complements the more traditional approach of starting out from pre-
selected lexical items. - It reveals differences between the languages which would have been difficult
to spot otherwise. - It supports the idea of conducting contrastive studies on the basis of recurrent
word-combinations rather than on single lexical items.
The method is clearly enhanced by using a bidirectional parallel corpus in that
it “makes it possible to distinguish between language differences and translation
effects” (Johansson 2007: 12). Being able to compare frequency lists from com-
parable originals and translations in the same language has proved useful in the
study of preferred ways of putting things in the two languages compared.
As Johansson (2009b: 36) points out, “lexical items which go beyond the
individual word [...] seem to function as translation units. How the meaning
is expressed varies depending upon the individual language.” The lists used as a
starting point for the present study highlighted the fact that the idiom principle
works differently even in closely related languages such as English and Norwegian;
most trivially, the ‘same’ meaning may be expressed by a 3-word combination in
one language and a shorter or longer phrase in another. Less obviously, the lists
revealed some different constructional tendencies in the two languages (e.g. i det
minste – at least), which nevertheless represent word combinations with a simi-
lar degree of fixedness and similar conditions of use. Similar-looking expressions