A Start | B Intro | C Staff functions | D Waiter function | E Word lists |
F Notes, Bibliog, App | Short waiter study

Dr Klaus Bung
68 Brantfell Road
Blackburn BB1-8DL
England
w: www.dynamic-language-learning-dr-bung.com

© 1973 and 2010 Klaus Bung

Klaus Bung:
The Foreign Language Needs
of Waiters and Hotel Staff 1
(aka "The long waiter study")
- Part B -

-------------------------------------------- page 11 -------------------------------------------

1 Introduction

This study was prepared at the instigation of a 'group of experts established by the Council of Europe to investigate the feasibility, and plan the introduction of, a unit-credit system in the field of adult language learning'.

The brief was 'to prepare an exemplary study of the structure of language unit (or units) in terms of the component modules, corresponding to the situation of a waiter with associated functions in a small hotel dealing with English-speaking tourists'.

In the discussion of the brief it was also made clear that the waiter should not be English and should work on the continent of Europe. The waiter was to carry out 'associated functions' as well, i.e. lend a helping hand wherever the situation and his knowledge of English required it.

The study was to be carried out and a final report to be submitted within four weeks. The report was not to exceed 30 pages.

The time-limit imposed severe restrictions on the work that could be done and on the methods that could be used. I spent one week in a small one-star hotel in Spain, which displays the foreign language notice reproduced on the title page of this study and with whose organisation, owners and staff I was well acquainted. I observed daily work, asked as many questions as I could without upsetting owners and staff, looked at the foreign language correspondence, and helped with any foreign language transactions that occurred.

Other sources of information were my general experience of restaurants and hotels, some individuals whom I asked about their experiences and the literature cited in the bibliography.

The limitation to 30 pages meant that I could fully present not even all the information obtained during the short period of four weeks. Since the empirical foundations of the study are, in any

-------------------------------------------- page 12 -------------------------------------------

case, weak (due to the time limit), the most important aspect of the study is not the 'results' taken at face-value but rather the methods and arguments used to obtain them. A later, more thorough, study of the same topic and of similar ones will be able to use these methods and arguments and may come to different results - using different data. It is therefore important for methods, arguments and side-remarks not to be lost.

They cannot be properly presented on 30 pages. I have therefore decided to submit two reports: the present one, which ignores the space-limit of 30 pages and contains the full argument, and a shorter one of less than 30 pages, which will contain some of the results without the full arguments and will satisfy the terms of the research contract. The shorter report will be entitled 'The foreign language needs of waiters and hotel staff 2'.

Both reports are a contribution to the development of the unit-credit system which the Council of Europe is now preparing (Trim 1971, Richterich 1972, van Ek 1972, Wilkins 1972)

2 The linguistic setting

As far as I know, no formal system for the specification of linguistic settings exists at present. I present here a simplified version of such a model (that will be discussed elsewhere in more detail) in order

1. to have an instrument for making my brief more precise
2. to generalise and extend its limits as far as permitted
3. to see all its implications in respect of the linguistic setting
4. to provide future studies and briefs with a means of specifying linguistic settings with any desired degree of generality or particularity.

The model uses the concept of vector (whose applications to education have been fully set out by Bung 1972, following earlier proposals by Ashby 1955 and Landa 1968). A vector is an ordered set of numbers, where each number can be used to describe a different aspect of an

-------------------------------------------- page 13 -------------------------------------------

object or process. In the following model, each component of the vector can assume the values 1, 2 or 3, but, for mnemonic reasons, these have been replaced by the letters M, O and T. M denotes the mother tongue of the learner, T denotes the target language, and O denotes the set of all other languages. The model will henceforth be referred to as the MOT-model.

It provides a means of specifying the properties of any given speech act and for investigating the interrelations of these properties. In the present simplified version of the model, I have selected seven properties (variables), each represented by a 'component' of the vector describing the speech act. The components are denoted by c0, c1, c2, ..., cn.

• c0 other languages; see c3, below
• c1 the learner's native language
• c2 the learner's target language
• c3 the native language of the partner in an act of communication; this may be either identical with c1, the learner's native language, or with c2, the learner's target language, or it may be some other language. In order to be able to express the last situation, too, as an identity of components, we introduce c0, denoting the set of all other languages. We can then say of certain acts of communication that c3 = c0, i.e. the partner is a native speaker of some language other than the learner's native language or target language.
• c4 language to which the learner is most frequently exposed; this is usually the language of the country in which he lives; it affects the speed at which he learns or forgets.
• c5 language of immediate environment for the speech act; this is the language of the majority of the people able to overhear a conversation without being directly addressed; this affects the choice of language, depending on whether or not the participants in an act of communication want to be overheard. Interpretations can be found for c4 and c5 which accommodate broadcasting, publication in print, correspondence, telephone conversation &c. The situation in which nobody overhears a speech act has been ignored in this version of the model.

-------------------------------------------- page 14 -------------------------------------------

• c6 language in which the speech act is performed

Since the number of vectors to be considered grows exponentially with each additional component, and since, in this study, we are only interested in speech acts in the target language, T, I have omitted c6 in this simplified version of the MOT-model.

To make reference to individual vectors easier, I have numbered them V1, V2, ..., V27, in the last column of Figure 1.

Concrete languages can be inserted for M and T. M and T are, by definition, distinct. Once M and T have been determined, membership of O is implicitly determined. Our brief states that T is to be English and that the speech acts under consideration take place in English.

If a speech act is to be more precisely defined than is possible with the 27 vectors listed, more components have to be added. The same applies if it is to be defined with an equal amount of precision but if different aspects are to be accounted for.

If a speech act is to be defined in more general terms (i.e. if we want to say that it belongs to one of a group of vectors among which we do not wish to distinguish), statements of any degree of generality or particularity can be made by stating for c3 whether it is identical with c0, c1 or c2, or by making such a statement for c4 or c5. If we make only one such statement, we narrow down the linguistic setting to the domain of 9 vectors. If we make two such statements, the linguistic setting is further narrowed down to the domain of 3 vectors. Three such statements would narrow the linguistic setting down to the domain of exactly one vector.

We illustrate this procedure of narrowing down the linguistic setting by a systematic attempt at finding the boundaries of the linguistic setting implicit in our brief: 'a non-English waiter in a continental country (i.e. a European country that is not Great Britain), serving English-speaking guests'.

-------------------------------------------- page 15 -------------------------------------------

Figure 1: General model of linguistic settings

Click here for larger image.

We reduce the possible implications of c4 by assuming (in this simplified version of the MOT-model) that any given country has only

-------------------------------------------- page 16 -------------------------------------------

one language. Our brief then excludes from consideration any vector of the type c4=c2 (henceforth briefly referred to as 'type 42'). This one statement excludes 9 vectors from consideration, namely V4-6, V13-15 and V22-24, i.e. those cases where the waiter works in an English-speaking country. Vectors of type 40 and 41 are permitted for consideration, namely the waiter may work in the country of, say, his birth (type 41) or in a country other than that of his birth (or one with an identical language) provided it is not English-speaking (type 40). However, we will give primary consideration to type 41 and exclude type 40 from systematic discussion in this study. (Type 40 is the object of Richterich 1972, p 46).

The desirability of a formal model for such specifications should by now be evident, if only from the clumsiness of the above prose paraphrases.

Our brief allows consideration of types 30, 31 and 32, i.e. the partner's native language is irrelevant. However, type 31 (when he uses the target language to communicate with a compatriot, either for reasons of concealment or in order to be overheard - depending on c5) is less urgent than the other types, and the investigation will therefore be confined to types 30 and 32. Both types are equally important for English. The importance of type 30 may be slightly less for French, German, Russian, Urdu, ... (when the world as a whole is regarded as the learner's domain) and may be negligible for Dutch, Swedish, Hungarian and other less widely spoken languages. However, the importance of Russian, Urdu, ..., in respect of type 30 can increase greatly as soon as only certain parts of the world are regarded as the learner's domain.

Types 50, 51 and 52 are all possible and important. Types 50 and 52 occur, for instance, when a hotel is entirely filled with foreigners of the same nationality, e.g. when thousands of English tourists are dumped in the hotels of Mallorca (type 52, V11). The waiter can then, for instance, rely on his native tongue to serve purposes of concealment as well as of communication with his colleagues (e.g. when he roars through the hall: 'Has olvidado de traer el helado de la vieja esa.' = 'You've forgotten the ice-cream for the old hag

-------------------------------------------- page 17 -------------------------------------------

over there!'). It might be argued that hotels of type 50 and 52 are most in need for a waiter who is highly competent in one or several foreign languages. On the other hand, hotels with mass invasions of foreigners may be of the cheapest type and have a very undemanding clientele, and may therefore be unable to afford the services of a linguist-waiter; or the clients may not demand them or be held in sufficient contempt to be refused them even though they would appreciate them.

However that may be, since we are here preoccupied not with the linguist-waiter but with a waiter with some very limited knowledge of a foreign language, our attention may be focussed on type 51, where the guests are a mixture of foreigners and natives.

We have excluded c5 as a factor that helps to delimit the linguistic setting in which we are interested. For component c3, we have retained an interest in types 30 and 32. For component c4, we have retained an interest in type 41 only.

This yields the following pairs of defining statements:
30 and 41
32 and 41

As we have said above, two such statements, jointly applicable, narrow the linguistic setting down to the domain of three vectors. Our brief therefore allows us to investigate linguistic needs occurring in six different linguistic settings, as shown in Figure 2.

Figure 2: Linguistic settings to be considered in this study

-------------------------------------------- page 18 -------------------------------------------

It will be shown later that different linguistic settings (different vectors) chosen from Figure 1 demand different skills (different subject matter specifications). Whether the reduced set of vectors listed in Figure 2 is unitary in respect of the subject matter specification it demands or whether different vectors in this set demand different skills and subject matters is a question which I could not answer in the limited time allowed for this report. However, it is certain that a module specification for the unit-credit system can be interpreted as a vector of at least five components:

L	linguistic setting (MOT-model; current report)
M	medium conversion (Bung 1973)
R	learner's need (Richterich 1972)
W	semantic notions (Wilkins 1972)
E	grammatical and lexical specification (van Ek 1972)

For heuristic reasons it is, at times, useful to focus attention on a more concrete situation than the general brief permits. Having shown how far the brief for the present study extends (this implies that all conclusions must be general or detailed enough to cover all the setttings listed), I will now concentrate attention on a more narrow field, namely vector V10. Since I have only been able to deal with very basic problems, this choice is not very significant. The results would presumably have been the same if I had chosen any other vector from Figure 2. More detailed analysis may, however, lead to differences for different situation-vectors.

-------------------------------------------- page 19 -------------------------------------------

Continue with Part C