178 Jarle Ebeling, Signe Oksefjell Ebeling and Hilde Hasselgård
Johansson in exploring new avenues in cross-linguistic research by using a bal-
anced bidirectional parallel corpus, viz. the ENPC, in phraseologically oriented
contrastive studies. If it is the case, as many linguists have suggested, that meaning
is often expressed by multi-word combinations, should not cross-linguistic studies
of meaning take multi-word units as their point of departure?
Our aims are to explore a bootstrapping method (extraction technique) on
bilingual data and to study how recurrent word-combinations work cross-lin-
guistically in order to contribute to the discussion of what is general and what is
language specific, and how source texts differ from translated texts.
To bootstrap our investigation, we will produce n-gram lists of recurrent word
combinations in English and Norwegian original and translated fiction texts. The
hypothesis is that such lists can point to cross-linguistic differences that might
elude corpus investigations that take specific lexicogrammatical constructions as
their starting point.
Some of the differences revealed by the lists reflect morphological differences
between the languages: for example, the most frequent English 3-gram I do n’t is
due to DO-periphrasis and thus cannot have a corresponding word combination
in Norwegian, where negation is expressed simply by the negative adverb ikke
(‘not’). A potentially more interesting feature is the more frequent use of three-
word combinations including a light-subject construction with det (‘it’ / ‘there’)
in Norwegian texts, both original and translations, as compared to English texts.
These constructions will be explored in the first case study in Section 5; three-
word combinations beginning with all the will be investigated in Section 6, while
temporal expressions form the basis of the third and final case study in Section 7.- From recurrent word-combinations to phraseology
Based on the assumption that language use is heavily dependent on multi-
word combinations as carriers of meaning, this study focuses on the use of such
combinations.^1
Our object of study has only been loosely defined above as “recurrent word-
combinations”. However, the word-combinations we have selected for our case-
studies have more characteristic features than just being three-word combinations
that occur with a certain frequency in our material. They are part of what some
linguists would call a language’s phraseology.- Sinclair (2008: 409) even claims that “multi-word units of meaning seem to be the norm”.