In my recent research I investigate the possibilities of the compilation of a pedagocically applicable tourism collocation list. This research is based on and is the extension of my previous studies and their findings, and the experience I gained in a collocational study. There is no available tourism collocation list, I intend to compile and share one in public while writing my dissertation, and this recent study will be part of it, and hopefully as a pilot study will provide guidelines to it. In this study I investigate whether a small corpus is suitable to the compilation of a two-word collocation list on the topic of beach holidays.
First I briefly talk about collocations, corpora and corpus methods, then I present some findings of my previous collocational study and outline my research.

The continuous extension of specialist lexicon is crucial in learning English for specific purposes (ESP), this way the acquisition of the core tourism vocabulary is the main objective of teaching English in tourism. The language of tourism is so pervasive in tourism texts, using lots of attributes, intensifiers and superlatives, that without it tourism would surely cease to exist, as Dann (1996) the first researcher of this peculiar language remarked. Vocabulary is often learned in lexical phrases comprising several words. Collocations, two or more words occurring together are easily retrievable ready-made patterns that facilitate fluency. In Kennedy’s view (1990) it is collocations, where grammar and vocabulary meet. Learners often have difficulties in fluency as their collocational competence is limited, and this competence has to be improved in ESP as well.

Corpus linguistics and corpus methods have many potentials to exploit in vocabulary teaching and learning. Corpora, large collection of texts in computerized format allow investigations on the authentic usage of words and lexical chunks, as the aim of corpora is to provide data for linguistic research. Sinclair (1997, p. 170) detected, drawing up the basic idea of corpus linguistics that a single word carries only meaning through several words in a sequence.
Different corpus processing tools can be applied for data analysis. The majority of corpus research is based on frequency counts using software to produce word lists and a widely used corpus tool is the concordancer that generally provides an alphabetical listing of the key-word called node or target item and its collocate. Vocabulary lists help to select the most important vocabulary items and with the help of concordancer learners can explore the authentic usage of words, their collocates and grammatical features. In my previous study I used Cobb’s Compleat Lexical Tutor, the CLT, and Antony’s freeware, AntConc.
ESP settings, including teaching English in Tourism, are one of the most obvious applications of corpus methods. However, only a few language teachers have ever built corpora or examined concordance lines as corpus compilation and analysis seem technically difficult. I would like to demonstrate that the compilation and analysis of a corpus are not complicated procedures and the findings are applicable in vocabulary teaching.

In my collocational study I investigated how small corpora can be utilized in teaching special vocabulary. Why do I need self-built corpora? In my experience students like word lists and with the help of teacher-built corpora we can provide them good ones, and these small corpora give better insight to the lexical patterns of a specific language than a large reference corpus.
I compiled a corpus comprising 30.000 tokens (words) selecting internet articles on the keywords city tours mostly from a handy website, Articlecity. To the collocational profile analysis I used Antconc as it saves the uploaded text files, this way can be used effectively for data processing.
In this collocational study I examined whether small corpora are adequate for the examination of lexical patterns that reflect the commercial, marketing-orientated, alluring aspect of tourism language. To investigate these features, I administered a frequency count of adjectives, nouns and action verbs and analysed the collocational patterns of adjectives related to some key nouns such as view and building, and examined the occurrence of superlatives and intensifiers as well.
The frequency and keyword list of nouns, adjectives and main verbs provided word lists characteristic to city tours, and I found a large proportion of adjectives, superlatives and intensifiers.
Let’s see how concordances work. How would you describe view?

A problematic key noun, view was generally described as beautiful and wonderful by my students, yet it appears with different connotations in tourism context.
You can see the attributive collocates of the key word view in these concordance lines.
Besides discovering synonymous adjectives that describe view with attributes that sound a bit sublime in general English, like picturesque, breath-taking, magnificent, and stunning, in the concordance lines students meet a word twice, skyline, that is generally left out of tourism course books yet it is characteristic to cities. The observation of concordance lines is an efficient way of teaching synonymous adjectives instead of some threadbare ones and learning new words as well.

To find references to the superb nature of tourism language I investigated the occurrence of superlative forms and intensifiers and found a large proportion of them. The total number of superlative forms was 187, and the most frequent superlative was the best with 55 occurrences.
You can see the concordance lines of the best superlative clusters in an Antconc window

These superlative attributes refer to prices, places, hotels, restaurants and beers of course and function as tourist traps to manipulate and convince tourists to buy the tourism product.
As a sum, this small specialized corpus was suitable for producing wordlists, examining frequent and problematic tourism collocations, it was a useful aid to teach synonymous adjectives and words characteristic to city tours.
Not only concordance lines but a wide range of corpus-based vocabulary teaching and learning activities can be applied in tourism classes, among my students the printouts of multiple-matching and gap filling activities were the most popular. Cobb’s Compleat Lexical Tutor offers various means of corpus-based vocabulary teaching and testing tasks and activities.

My recent study is a pilot study to the compilation of a larger high-frequency two-word tourism collocation list on the main topics of Tourism English and this small corpus called beach holiday corpus will be a subcorpus of a larger-scale corpus. I plan to include 50 000 tokens which I expect to be sufficient to the compilation of such list, however, I might have to increase the size of the corpus, or merge its data with my previous corpora. Hopefully, I will be able to compile an applicable collocation list, and this list will be where collocations, tourism and corpus linguistics meet.