Reading Ethnographic Research

Martyn Hammersley

• Publisher: Routledge
• Year: 2016
• eBook ISBN: 9781134962310
• Topic: Social Sciences
• Subtopic: Social Science Research & Methodology
• Index: Social Sciences

Provides a practical guide to the critical reading of ethnographic studies: discussing in detail how to identify the main arguments and what is involved in making an assessment of such studies.

Chapter 1
The nature of ethnography

The term 'ethnography' is not clearly defined in common usage, and there is often disagreement about what count and do not count as examples of it. Furthermore, the meaning of the term overlaps with that of several others – such as 'qualitative method', 'interpretative research', 'case study', 'participant observation', 'life history method', and so on. Nor are these terms used in very precisely defined ways.¹ In part, this diversity and looseness of terminology reflects some disagreement, even on fundamental issues, among advocates of these approaches. It also results, perhaps, from a certain vagueness in thinking about methodological matters that arises from widespread emphasis among ethnographers on the primacy of research practice over 'theory' about how to do it. Sometimes this amounts to an anti-theoretical or anti-methodological prejudice. Ethnographers often tend to distrust general formulations, whether about human social life or about how to do research, in favour of a concern with particulars. In some ways, this is healthy: methodology cannot tell us what to do, it can only provide guidelines and cautions. But, however much one distrusts methodology, one cannot escape it. As we shall see, the practice of ethnography is surrounded by a host of methodological and philosophical ideas. And, while we must recognise the limits of abstract reasoning, we should press it as far as it will go in providing illumination, whether as regards doing research or reading and assessing its products.

I will examine some of the philosophical ideas associated with ethnography later in this chapter. First, though, I want to look briefly at the forms of research design, data collection and analysis that are characteristic of it, and to sketch something of its history.

Ethnography as method

In terms of method, generally speaking, ethnographic research has most of the following features:

1. People's behaviour is studied in everyday contexts, rather than under conditions created by the researcher, such as in experiments.
2. Data are gathered from a range of sources, but observation and/or relatively informal conversations are usually the main ones.
3. The approach to data collection is 'unstructured', in the sense that it does not involve following through a detailed plan set up at the beginning, nor are the categories used for interpreting what people say and do entirely pre-given or fixed. This does not mean that the research is unsystematic; simply that initially the data are collected in as raw a form, and on as wide a front, as is feasible.
4. The focus is usually a small number of cases, perhaps a single setting or group of people, of relatively small scale. Indeed, in life history research the focus may even be a single individual.
5. The analysis of the data involves interpretation of the meanings and functions of human actions and mainly takes the form of verbal descriptions and explanations, with quantification and statistical analysis playing a subordinate role at most.

As a set of methods, ethnography is not far removed from the sort of approach that we all use in everyday life to make sense of our surroundings. It is less specialised and technical in character than approaches like the experiment or the social survey; though all social research methods have their historical origins in the ways in which human beings have always gained information about their world. The more specific origins of ethnography lie in the writings of travellers concerned to inform their fellows about other societies, whether it is Herodotus exploring the western provinces of the Persian Empire (Rowe 1965) or Hans Stade reporting on his captivity by the 'wild tribes of Eastern Brazil' (Pratt 1986). These other societies were not always geographically distant either. Thus, in the nineteenth century, we have Friedrich Engels, the son of a German industrialist, investigating the lives of Lancashire factory workers, Tocqueville reporting on Democracy in America, and Henry Mayhew writing newspaper articles about working class life in London (Engels 1892, Tocqueville 1966, Mayhew 1861-2 and 1971).

The writings of such people form the early history of the disciplines now called social or cultural anthropology and sociology.² However, the data on which these reports were based were often unsystematic and sometimes misleading; and the analysis was frequently speculative and evaluative (though sometimes valuable for all that). One of the most important features of the development of social science disciplines in the nineteenth and early twentieth centuries, under the influence of the model of the natural sciences, was the attempt to overcome the methodological deficiencies of earlier accounts. In the nineteenth century anthropologists tended to rely on the reports of travellers and missionaries for their data, but from around the turn of the twentieth century it became widely accepted that it was necessary for them to collect their own data, and to do so in a systematic and rigorous manner. Initially, this tended to take the form of expeditions that were primarily concerned with collecting artifacts, both material and non-material (such as myths). Later, particularly through the influence of Bronislaw Malinowski, it came to be required that the anthropologist live among the people being studied, learn their language, and observe their lives firsthand.³ In addition, anthropologists became much more concerned with avoiding the speculative excesses of their predecessors, and with adopting a scientific approach to the interpretation of evidence. Since the early decades of the century, ethnography has been the staple research method employed by social and cultural anthropologists, though they have sometimes combined it with social survey work and even with the use of psychological tests.

Sociologists have also employed ethnographic methods since early in the twentieth century, though it has not occupied as central a place in their discipline as it has in anthropology. One of the earliest developments here arose in what is sometimes referred to as the 'Chicago School of Sociology'. Chicago was among the first universities in the United States to establish a sociology department, and that department was very influential in the 1920s and 1930s. Sociologists working there became preoccupied with the character of their city. It had experienced a tremendous growth in size in the 20 years each side of the turn of the century, and many of the new inhabitants were European immigrants. Given this, Chicago sociologists came to see the city as a kind of natural laboratory in which the diversity and change characteristic of human behaviour (and particularly of modern social life) could be studied. While investigation of the city had begun before his arrival, Robert Park, an ex-newspaper reporter, was the main force behind the series of studies produced by students at Chicago in the 1920s, 1930s and 1940s. Some of the work carried out was statistical in character, but much of it took the form of what was called 'case study research'. This involved detailed investigation of a single case, or a small number of cases, these being areas of the city (as with Zorbaugh's 1929 exploration of the different parts of North Side Chicago and their inhabitants), city organisations (for example, Cressey's 1932 study of the taxidance halls where men paid the dancers a fare for each dance), or even individual people (as in the case of Shaw's 1930 The Jack Roller, a life history of what we would now refer to as a 'mugger'). These studies used methods that approximated the ethnography of today, though they tended to place greater emphasis on written documents.⁴

At around the same time as the Chicagoans were engaged in studying their city, another important source of ethnographic method in sociology and anthropology was emerging: what came to be called the 'community study'. The inspiration for this was in large part the application of anthropological methods to the study of Western societies. One of the first and most important community studies was the investigation of 'Middletown' (the small town of Muncie, Indiana) by Robert and Helen Lynd. On the basis of participation and observation in the community and interviews with community members, they described key aspects of life in Middletown: "getting a living', 'making a home', 'training the young', 'using leisure', 'engaging in religious practices', and 'engaging in community activities'. This specification of aspects of community life derived from a scheme developed by anthropologists for studying non-Western societies. The Lynds had originally been concerned with religious beliefs and practices, but they realised that these could only be understood in the context of community life as a whole (Lynd and Lynd 1929). Following this, many other communities, usually but not always small towns, were studied; and sometimes they were restudied. The Lynds themselves returned to Middletown some ten years after their first investigation, to document the changes that had occurred, and especially the effects of the Depression (Lynd and Lynd 1937). Later, a similar tradition of studying small communities was established in Britain, producing, for example, Williams' book about the village of Gosforth (Williams 1956), Emmett's account of Llanfrothen (Emmett 1964), and Stacey's study and restudy of Banbury (Stacey 1960, Stacey et al. 1975). In addition, a number of investigations were mounted by the Institute for Community Study in the 1950s and 1960s, most notably of working-class areas in London (see Piatt 1971). This community study tradition has witnessed a decline in the past 20 years, in both the United States and Britain; though there have been some recent studies of aspects of community life in the West by anthropologists (see, for example, Cohen 1982 and 1986).⁵

As I noted, a crucial element of the emergence of specialised social sciences was the drive to make them systematic and rigorous, on the model of natural science. However, this took a variety of forms, depending on what were believed to be the essential features of science, and ideas about the extent to which and ways in which human social life differed from the subject matter of the physical sciences. As we saw. within anthropology, it resulted in an emphasis on more direct observation by the anthropologist and attempts to make the descriptions produced as objective as possible, being neither deceived by the historical myths of the natives nor blinded by one's own cultural assumptions and prejudices. An analogous emphasis on documented and first-hand information was the initial effect of the drive for scientific status in sociology, as represented by the early Chicago School. However, from at least the 1930s onward, at Chicago and elsewhere in the United States, social science came increasingly to be associated with the statistical method; and by the 1940s and 1950s survey research was the dominant method within US sociology. This is not to suggest that case study and community study work ceased. Research in anthropology continued much as before. Even in sociology, ethnographic studies still appeared, notably William Foote Whyte's (1993) Street Corner Society (an investigation of the cultures to be found among young males in an Italian American community) first published in 1943, and the work of Everett Hughes and his students on various occupations and on medical and educational institutions. In addition, partly stemming from the dominance of quantitative method, there were increasing efforts to enhance the rigour of ethnographic research.⁶

However, in the 1960s the overwhelming dominance of survey methodology in US sociology and elsewhere started to wane, and ethnographic research became more widely advocated and practised (though survey research has remained very influential in the United States, more so than in Britain). The growth in ethnography at this time was especially notable in the study of deviant groups, but also in the sociology of medicine and other areas. In Britain there was a relatively sudden increase of interest in ethnographic research towards the end of the 1960s, largely as a result of the diversification of theoretical approaches to include symbolic interactionism, phenomenology and ethnomethodology. A considerable body of British ethnographic research has been produced since then, especially in the sociologies of deviance, medicine and education.⁷ And this resurgence of interest in ethnography in the 1960s, 1970s and 1980s, following a period in which survey research had become almost identified with sociological method, was accompanied by considerable debate about methodological issues. This occurred not just between advocates of survey and ethnographic research, but also among ethnographers themselves. In the next section I want to look briefly at these discussions.

Methodological debates about ethnography

Debates about ethnography can be broadly divided into two categories, with the first preoccupying discussions in the 1960s and 1970s, the second becoming more salient in the 1980s and 1990s:

1. Those centring on criticisms of ethnography for not meeting the criteria assumed to be characteristic of science.
2. Those concerned with arguments that ethnography has not broken sharply enough with, or moved far enough away from, quantitative research and the model of natural science.

I will look at each of these two sorts of debate in turn

Criticism of ethnography for not being scientific

Within sociology, competition with quantitative approaches, and especially with survey research, has long shaped the way in which ethnographers have thought about their work, raising the question: is ethnography scientific? By the 1960s, quantitative research in the form of experimentalism and social surveys had become highly developed, and sophisticated methodological literatures had grown up around them which took it for granted that they were implementing scientific method. While there was dissensus about some points, there was considerable agreement on the basics, and in large part this was framed in terms of the positivist outlook that had dominated the philosophy of science in the 1940s and early 1950s. Central here was the idea that what was required in empirical research was clear and operational specification of hypotheses; the selection of a research design that allowed those hypotheses to be tested, either by the physical manipulation of variables (as in experiments) or through statistical analysis of large samples of cases (as in survey research); along with the assessment of measurement error by means of reliability (and, more rarely, validity) tests.

Ethnographers responded in several ways to this influential view of the nature of social science methodology:

1. Some claimed that ethnography is more scientific (that is, closer in character to natural science) than quantitative research. In the early decades of the twentieth century, this was argued on the grounds that case study can produce universal laws, not just the probabilistic findings characteristic of statistical method (Znaniecki 1934). This argument is rarely used today, both because of the statistical character of some parts of more recent work in physics (notably quantum theory), and because most ethnographers have lost faith in the possibility of discovering sociological laws. Later, it was more common to find the scientific character of ethnography being justified on the grounds that it is more suited than are experimental and survey research to the nature of human behaviour, in particular its processual and meaning-laden character (Bruyn 1966, Blumer 1969, Harré and Secord 1973).
2. A second sort of response involved broadly accepting the view of sociological method characteristic of quantitative research, but treating ethnography as distinctive in its suitability to particular phases of the research process or to particular research problems. For example, it was often regarded as useful in the pilot stages of social surveys, or in the debriefing phases of experiments; and/or it might be seen as well-suited to the study of particular types of people, for example deviant groups, whose small numbers and/or inaccessibility make them difficult to survey. On this view, ethnography could be usefully combined with other methods, which are assumed to have complementary strengths and weaknesses.⁸
3. A third position was that ethnography represents a different kind of science to that characteristic of the natural sciences, and quantitative method was criticised for aping those sciences. Often associated with this view was the claim that ethnography is idiographic rather than nomothetic (that it focuses on the unique as much as the general), and/or interpretative rather than observational (that the understanding of human behaviour always involves interpretation, not mere physical description).⁹

In their methodological writings ethnographers frequently combined these three positions in various ways, and not always with great consistency. The first and third were typically given the most emph...

Table of contents

• Cover
• Half Title
• Series Title
• Title
• Copyright
• Contents
• Dedication
• Acknowledgements
• Preface
• Chapter 1 The nature of ethnography
• Chapter 2 Understanding ethnographic accounts
• Chapter 3 Standards for assessing ethnographic research
• Chapter 4 Making an assessment: validity
• Chapter 5 Making an assessment: relevance
• Chapter 6 An example
• Chapter 7 Conclusion: varieties of assessment
• Appendix: Staffroom news
• References
• Index