From:
C. Roberts et al. (2001) Language learners as ethnographers. Clevedon (U.K.): Multilingua Matters, pp. 142-150. Part II, Chapter 7: Developing an ethnographic method

Developing Principles for an Ethnography Course

Students are given a step-by-step guide to interviewing based on SpradIey's (1979) The Ethnographic Interview. They then have an opportunity to practise and observe others trying out this new approach in their first or dominant language. At this stage, students tend to raise a number of questions about ethnographic conversations and interviewing:

• How do you reach a stage when you can ask people if you can talk to them in a more focused way?
• How to 'capture' the information? Should they take notes, taperecord or simply try to remember?
• How should they cope with such interviews in the foreign language, given that they will be non-native speakers?

0.

After a discussion on the relationship of Participant Observation (PO) and interviewing, on the advantages and disadvantages of tape-recording, but leaving the fears about their own language competence until they have attempted an ethnographic interview in a foreign language, other more complex issues have to be raised about the nature of the data they are collecting. These are raised by Briggs in his study of interviews as a research method in anthropology.

2.

He argues that the way interviewing is used acts as a hidden filter and blocks our ability to hear the data as a piece of socially constructed material:
 

lf I were to put my finger on the single most serious shortcoming relating to the use of interviews in the social sciences, it would certainly be the common sense unreflexive way in which most analyses of interview data are conducted. ( ... ) Questions and answers are presumed to have obvious significance. (Briggs, 1986: 102)


Briggs is arguing here that the conclusions drawn from data collected in interviews are not unproblematic facts. The questions are asked in particular ways and construct and constrain the answers. A different question would produce a different response and so different data. So any interview data is jointly produced and is as much the product of the interviewers' social and cultural world as it is of the informants'. This idea of reality as socially constructed (Berger & Luckmann, 1966) is a theme that runs throughout the course and is fundamental to the process of cultural learning (see Chapter 3). Students have to learn that data from informants are partly produced by them, the researchers, because they ask certain questions, follow up on certain themes, foreground certain concerns in the way they ask questions and so on. They also need to appreciate that any item of data is situated in a particular context. By asking an informant, for example, about their life as a blind student in Marburg, the category of 'blind student' is the one that the informant is being asked to identify with and represent their views on. At other times and in other contexts, that same category may be deemphasised and the informant might present quite a different aspect of themselves. For example, they might be part of the student community which is on strike because of govern-ment underfunding or they might be single parents or jazz musicians. In any of these cases, they would present themselves in different ways depending upon the particular group identity that was in focus.

Working within this reflexive frame of mind, the ethnographer can resist the easy generalisations and stereotyping which are a commonplace reaction when people are confronted with cultural differences. It also helps them to think about their own cultural worlds.

3.

They ask a particular question because it seems significant to them - is part of their way of believing, acting and valuing and so on. So, every interview can deepen the language learners' understanding of themselves as cultural beings as well as teaching them about others. This approach may also, paradoxically, help them not to stress the ethnic and cultural differences in their informants - not necessarily to see particular behaviours or opinions as French or German. Instead, they need to realise that ethnic and cultural differences (and identities) may be unstressed and what they are observing is not so much French, for example, with a capital F but to do with particular personalities, social positions and biographies rather than Frenchness. Discussing these issues at this data collection stage also raises another common theme in the course, the interdependence of data collection and analysis which we shall consider below.

Some of the most common difficulties experienced by students when they first attempt ethnographic interviewing are given in Figure 10.
 
 
4.

• Difficulty in getting started.
• Using closed or leading or overly direct questions, e.g. 'Doesn't it give you a bit of a buzz if your horoscope says you are going to ... ? or e.g. 'Are you doing French commerce?'
• Using questions which are not obviously leading questions but where assumptions are built into the question by the ethnographer, e.g. 'What did you achieve in doing your degree?' Where the commonplace assumption that there is achievement enters into the way the question is asked.
• A tendency to start a new topic of conversation, once the informant has given one response, so the ethnographic conversation becomes more like a formal interview.
• A tendency to ask direct questions rather then find less interventionist ways of continuing the interview, such as nods, uh huhs or even silence.
• A tendency to feed back to the informant an interpretive summary of what they said rather than working with the informants' own words.
• A difficulty in picking up on the informant's style - for example her use of metaphors to describe a key event.

Figure 10 Common difficulties

Throughout the process of leaming to interview, thinking reflexively about the data collected and reflecting on the cognitive and interpersonal demands of doing ethnographic conversations and interviews, students are continually reminded of the attention to language which such a technique involves. In some ways it is like the most demanding kind of listening comprehension that they might be familiar with or the multiple skills required in simultaneous interpreting. Their informant will, if all goes well, be relaxed and talking freely and informally in a local vernacular which may still be relatively unfamiliar to the student. The capacity to listen in a highly focused and intense way, attempting to record, either mentally or on tape, the exact words of the informant; to follow up wherever appropriate with questions; to keep open a tracking channel so that, periodically, a quick scan can be made of the check list to decide what areas this interview might still try to cover -and to do all this in a foreign language - emphasises yet again the inseparability of language and cultural learning in the field and the skills and intellectual demands placed on students.

Once they have worked on ethnographic interviews in their first language, they are given the task of doing one in the foreign language. Interestingly, the fears about their own competence in interviewing in the foreign language are quickly laid to rest.

1.

They find that ethnographic interviewing requires relatively little productive competence because the whole point is to give the informant control of the interview and because questions so often use the informant's own language.

Students were asked to transcribe a part of their interview and comment reflexively both on the way in which they had framed the informant's responses and on the themes that those responses suggested. The opening few turns from an interview between a male Spanish student learning German and a female German student in Britain for her period of residence abroad, shown in Figure 11, illustrate the very first steps in a reflexive process in which data collection and analysis are combined:
 



5.


Original German + English

Ich móchIe herausfinden, ob es viele kórperliche Kontakte innerhalb der deutschen Familie gibt Zum Beispiel kùßt du deine Mutter?
[I should like to find out if there is much physical contact in the German family. For example, do you kiss your mother? ]
Comments about interviewer: Perhaps I shouldn't have asked this Question so early

Nee, gar nicht  [No, not at all]
Comments about interviewer: I was a bit troubled by her reaction - short answer

Und warum ist das? [And why is that?]

Weil ich es als Kínd gehaflt habe  [Because I hated it as a child.]

Ja? [Yes?]

Ja, also mein Vater wolIte mich immer kússen und das habe ich gehasst. Ich konnte das nicht leiden, weíl meine Oma mich immer kùssen wollte. Tatsáchlich! [Yes, my father always wanted to kiss me and I hated that. I couldn't stand it because my granny always wanted to kiss me. Really!]
Comments about informant: She was justifying herself but she krìew my opinion was completely different

Tatsdchlich? [Really?]
Comments about interviewer: I give a short reaction to show interest

Figure 11

The example in Figure 11 shows a student who is just beginning to think about the way in which he could elicit data from an informant. He noted from this first attempt at an ethnographic interview that his opening question produced only a short answer and he also commented on the value of simply repeating back the other's words to encourage expansion. However, although he is beginning to learn some of the possible techniques, he still opens the interview as if it were a sociological survey rather than an ethnographic encounter and comments on her utterances from a psychological perspective (she was trying to justify herself rather than asking questions like: 'What does it mean to say you "cannot stand" to be kissed by your family?' Discussion of this kind of self-reporting in the sessions begins the process of moving students towards a cultural and social orientation and helps them to see how data collection and analysis are inextricably linked together.
 

Recording and analysing data

6.

As we have suggested, the process of collecting data and analysing it is an interconnected one. However, all the stages of an ethnographic study from an initial 'foreshadowed problem' (Malinowski, 1922 ) to the writing up are all interdependent:
 

The process of theory construction, hypothesis formulation, categorisation, data production, coding and interpreting are inextricably bound-up with one another. (Ellen, 1984: 285)


This is well illustrated in Shirley Jordan's narrative account of her first ethnographic venture, discussed in Chapter 6, on a group of cleaners at the university. At first she planned to undertake a 'topic-based' study on cleanliness (Zaharlick & Green, 1991 discuss the difference between a topic-based and a 'compreliensive ethnography') but once she had started lurking and soaking' she found she had to rethink her research problem according to the informants' concerns.

7.

This discussion of student interviews (in Units 9 and 10) highlights a number of the issues around recording data, specifically note-taking. After two sessions on ethnographic interviews and conversations, students are introduced to the discipline of note-taking. In addition to developing a systematic approach to note-taking, for example acquiring the habit of labelling each note with date, time and place so that it can be easily indexed and referred to, students are introduced to the different kinds of notes which they may accumulate - from 'head notes' which are those memories, impressions and perceptions which never get written down but influence the analysis, through 'scratch notes' which are the rough, often seemingly chaotic field notes that are a first attempt at documenting observations, to more analytic notes, as the raw data is continuously reread and interpreted (Sanjek, 1992).

As the notes are gradually processed and themes begin to emerge, so the ethnographer moves from initial recording to consolidation and analysis. As further data are collected, the preliminary analysis is then reviewed and adjustments made so that the ethnographer constantly toggies between recording and interpreting. This process of consolidation 'facilitates a constant cumulative dialogue with your material, (Ellen, 1984: 283). Ellen suggests writing up a 'cumulative contents page' in which key descriptive themes are brought together. In the process of doing this, connections and patterns can also begin to emerge and crossreferencing may be helpful. For example, in a home ethnography on despatch riders by Chris, leather clothing was an important element in the way in which the riders designed their image for the public and managed the tension between their very controlled working practices and the free wheeling 'Easy Rider' image which they conveyed to the world outside. As part of his field notes, he started to bring together a list of events and themes (see Figure 12).
 

Events:
Radio traffic, road traffic problems (constant), breaks, pick-ups, deliveries, problems, petrol, lunch, meetings, work, maps, the file.

Themes:
Rush, hassle, stress, boredom, frustration, excitement, exhilaration, self-containedness, capable, durable, knowledge of streets, amount of inoney, amount of work around, sense of sharing problems, distribution of work, watching people, remarking on different customers and staff.

Figure 12

Undoubtedly, for most students the most difficult aspect of data analysis is moving from initial themes to more conceptual categories and relating these categories to some of the concepts introduced during the course. At this stage, students are shown the scatter sheet from Ana Barro's ethnographic study of the porters at the university, 'lt's not my job' (see Chapter 6). Students are also given an example from Agar's (1986b) study of self-employed long-distance truck-drivers in America, Independents Declared. He was interested in finding out the extent to which the trucker lived up to the image of the free-wheeling independent cowboy mythologised in American narratives. There are obvious links here with the student ethnography on despatch riders. The study was carried out over nine months and the data collected during 40 days on the road, attending meetings, hundreds of conversations and 17 extended career history interviews. The interviews came to be the core data and it was to this data that Agar made himself accountable, drawing on the rest of the data to illuminate and enrich the interview data.

His conclusions centre around the tension that the truckers live with: the tension between imagined freedom and the very real dependency which is their 'lived experience'. They are dependent on the carriers who give them work, the mechanics who keep them on the road and the regulatory officials who keep them on a bureaucratic treadmill. Given the cowboy image that they project publicly, Agar asks why their accounts of trucking life are, in the main, so negative. He concludes that they represent worst case scenarios. In other words, these events happen but the frequency and awfulness of their experiences are overstated. This is, Agar argues, a rhetorical device in which the informant overdramatises the consequences of the situation in order to make it a good story for the ethnographer.

In coming to his conclusions, Agar discusses his methods and the ways in which the concepts he is working with are 'grounded' in the data (Claser & Strauss, 1967). The descriptive themes at the early stages of analysis are interpreted in terms of concepts which are then grouped together in categories. Through a continuous process of comparing different descriptions and different concepts, broad categories are decided on. For example, in his study, Agar has a chapter called 'Getting Underway'which consists of a number of categories: 'Load and Unload', 'Lumping', 'Manoeuvring' and 'The Driver's Log'.

The terms Agar uses for categorising the data are a combination of terms drawn from the literature, for example 'dependence'; terms used by the informants themselves, for example 'lumping'; and his own terms, for example 'road life'. Yet none of these concepts are brought in from outside and ìmposed on the data, nor does Agar simply present the voices of the truckers or describe their lives as if the events they described could be presented in an unproblematic way.

8.

Agar's study can be used to sum up some of the key concerns of this part of the course:


At the end of the 1993/94 course, one student wrote: 'Thanks again for helping me to develop skills by which to open my eyes and mind, and experience many weird and wonderful aspects of life I might otherwise have overlooked.' This brief quotation illustrates many of the principles upon which the Introduction to Ethnography course is based: an intellectual openness, specific skills of data collection and analysis, the making strange (and 'wonderful') what had seemed ordinary and uneventful, and a sense of process, of things not being finite but in a constant state of development. To sum up, therefore, the course aims to give students experience of:


Underlying much of these experiences, was the pedagogy of the course, the assumptions about learning and teaching and the roles of student and teacher. For many students, the course did not follow an accustomed pattern and indeed the experience of taking the course was something of a culture shock. In the following chapter, we describe some aspects of the ethnography classroom.

***