International Journal of Corpus Linguistics - Volume 13, Issue 3, 2008

This paper proposes a concept of ‘semantic sequences’, that is, series of meaning elements that can be demonstrated to occur regularly either in a general corpus or in one comprising specific types of text. Semantic sequences may be identified by examining the co-text of a lexical item such as make sure, but are also recognisable if grammatical words such as prepositions are treated as the object of investigation. It is argued that semantic sequences are useful for characterising discourses of particular kinds.

The role played by mitigation in academic discourse has been the subject of intense scholarly interest over the last two decades, but interest in the role played by intensifying textual elements expressing evaluation and stance — emphasizers — is a more recent turn. This paper presents a preliminary attempt at capturing the uses of the adverb really across spoken and written academic registers. The adverb really is examined with an eye to how its frequencies, meanings and uses vary across spoken and written academic discourse. The findings will also be interpreted in terms of variation across genres and disciplines. A quantitative analysis of this adverb reveals significant distributional trends across both academic genres and disciplines, and a qualitative analysis of concordance lines confirms that these trends are motivated by genre-specific purposes and disciplinary-specific practices, respectively.

This paper focuses on phraseology used within the domain of politics, both in written and spoken discourse. We concentrate on the lemma TERROR and on the recurrent sequences in which it is embedded, reflecting how native speakers, both American and British, tend to use it in preferred environments making routinized blocks of language. The data come from two corpora: the spoken corpus includes speeches of George W. Bush and Tony Blair, and the written corpus is made up of articles from The Wall Street Journal and The Economist. Since text is nothing but phraseology of one kind or another (Sinclair 2008), our attempt here is to uncover which of the two varieties lends itself more willingly to creating phrases that are handled like single units. The two pieces of software used to retrieve such units (n-grams and concgrams) are WordSmith Tools (Scott 2004), and ConcGram (Greaves 2005).

This paper investigates the idiom principle realized as four-word phrases (4-grams) headed by prepositions in specialized corpora in English and Italian. Concentrating on at the end of, it reports that the collocates of at the end of regard time, and that apparently synonymic 4-grams are not used in the same contexts. It then explores realizations of at the end of in a specialized comparable corpus of Italian. Two findings emerge: firstly, that the most obvious equivalent, alla fine d*, occurs more frequently than in the English corpus; secondly, this n-gram is frequently used, but has weaker collocational relations, and several synonymic 3-grams share its collocates. This invites contrastive research on lexical variation and repetition and on the strength of collocations of multi-word units in English and Italian. Lastly, the paper recounts an experiment with students who gained awareness of language by concentrating on phraseology in comparable corpora.

This paper aims to describe the phraseology associated with adjectives in the language of tourism. The adjectives appearing in the word list of a British farmhouse holidays corpus were grouped into three different semantic fields (‘description of rooms’, ‘description of surroundings’, and ‘description of food’) and from which their collocational profiles were then identified. In order to compare and contrast the Italian and the British languages of tourism, we searched an Italian comparable agriturismi corpus for items that are used to describe rooms, food, and surroundings. The results are discussed with reference to Sinclair’s theories on the influence of context and register on language choices (Sinclair 1991), and Hall’s theory of high vs. low context cultures (Hall 1976, 1989; Katan 2003). This analysis shows that the language of tourism is highly phraseological. It also gives insights into some differences and similarities between English and Italian in terms of ‘language systems’ and cultural orientations.