Ethnography Research Paper

By ‘anthropologist’ is meant here those anthropologists—sometimes called ‘ethnologists’—who work with living peoples to explore and document their culture. Their research involved long-term involvement in a community called ‘participant observation,’ discussions of topics of interest with community members called ‘informal interviews,’ and a record of the experience called ‘field notes.’ Ethnography also labeled the book-length report that an anthropologist wrote, using intuitive methods of analysis called ‘immersion in the data.’ The ethnography, the book, was also called a description of the culture of the community, although the culture focus was more characteristic of work in the USA than in Europe.

This straightforward definition of ethnography must be understood in its historical context. Anthropological research, as practiced in the early to middle twentieth century, usually involved an Anglo-American or European working in a small isolated community in a remote area of the world, with a group of people who often lived without modern amenities, had little or no formal education, and existed at a modest if not impoverished economic level. The anthropologist took up residence in the community, and set out to learn who the people in the community were and how they lived. Depending on their interests, anthropologists investigated topics as diverse as economic and household tasks, social organization and religious beliefs, family life and local markets— virtually anything that made up daily life. Since they also lived in the community, anthropologists visited with people, attended communal events, and worked together with community members on common projects, including the ethnography itself.

As one learned the community’s way of living, one said one was ‘doing ethnography.’ Ethnography, then, named the process of learning what for the anthropologist was a new and different way of talking, thinking, and acting.

When the work was done, a monograph was usually written—a report of this exploration into the community’s way of life. One could point to that book and say it was an ‘ethnography’ as well. Further, one could also say that the book was a description of the group’s culture.

Traditionally, then, ethnography named both a research process characteristic of anthropology and a research product, often a book-length description, but at any rate a representation of the culture of the community in which the research had been done. (Agar 1996 introduces this traditional view and contextualizes it in light of more recent discussions.)

Because anthropologists set out to learn ways of thinking, talking, and acting whose nature they could not specify before they started their research, ethnography did not look much like traditional science. In fact, the scientific status of ethnography has been an issue since its beginning.

Traditional science sought universal laws; ethnography went after local particulars. Traditional science emphasized control of the research process; ethnography featured adaptability as the nature of community emerged. Traditional science evaluated a sample by size and strategy; ethnography, by competence, social position, and relationship to subject. Traditional science preserved initial concepts through the course of the research; ethnography developed new ones. Traditional science rested on linear models; ethnography, on models more systemic and processual. Traditional science represented data with numbers; ethnography more often used words.

Over the last twenty years it has become clear that ethnography cannot be adequately modeled or evaluated by traditional science. Some still argue that it should be. Some argue that it is a different kind of science. Some argue that it should be classed with the humanities. The issue is treated in Sect. 2 of this research paper.

A second problem arose with the nature of ethnography as a product, a description of a culture that resulted from the research. Anthropologists assumed that small communities represented a single culture, one that was internally consistent and traditional, passed from generation to generation in essentially unaltered form. The ‘culture’ concept served as description, explanation, and generalization. The fact that a person did something was data for the description of that culture. What was done could be explained by membership in that culture, and it could be generalized to the single culture assumed to be relevant.

The culture concept that guided earlier ethnographies is now in a shambles. The earlier assumptions about culture, such as its traditional and consistent nature, have been shown to have been in error from the beginning (Wolf 1982). Further, contemporary life makes the assumption that a person is a member of only one culture look hopelessly naive. While the nature of ethnography as a research process endures, it is no longer clear how to describe, explain, or generalize what is learned, nor is it clear what form a report—the ethnographic product—should take. The old concept of culture is no longer adequate to the task. Sect. 3 of this research paper deals with the problem.

Until this point ethnography has been discussed in the context of anthropology. Because of its longstanding concern with the exploration of previously unknown ways of living, anthropology is the academic discipline with the most elaborate history of ethnographic research. Ethnographic work outside anthropology also exists. For example, predisciplinary efforts such as Tacitus’ description of the Germans in Roman times are often cited as a precursor. And ethnographic work in the discipline of sociology also has a long tradition, beginning with work based on Weber’s theory of social action, continuing with the symbolic interactionism of the Chicago school and with the development of grounded theory and ethnomethodology in the 1960s.

By now the historical link between anthropology and ethnography has turned complicated in at least two ways. First of all, ethnography is used in a variety of different disciplines—ranging from history to speech communication to organization and management to public health—as a new approach, as well as in kindred social sciences such as political science and psychology. Second, the concept now coexists with such related terms as qualitative research, grounded theory, and phenomenology, terms that are also used widely in a number of different research contexts. These multiple uses of ‘ethnography’ and related terms have further blurred the edges of how ethnography should be defined, a problem dealt with in Sect. 4 of this research paper.

In traditional cultural anthropology, then, ethnography had a clear meaning—the process of learning a community’s way of life and reporting the results of that learning under the label ‘culture.’ Beginning in the late twentieth century, many of the assumptions underlying traditional ethnography were called into question—questions amplified by the changing nature of the world we all live in now. Finally, the increasing popularity of ethnography and related nontraditional approaches to research in numerous academic and organizational settings further complicates any attempt to define the term.

This research paper, then, is built less on a definition than it is on a sample of key issues around the question of what a definition might look like. ‘Ethnography’ is now a cover term for several debates around the nature of human understanding; debates for which, at the time of writing, no end is in sight. In fact, this might be the major contribution of the term to current social research, a contribution with more historical importance than its use as a label for a discipline internal activity. If nothing else, this research paper alerts the readers new to ethnography and culture that they are stepping into a series of issues rather than a recipe for doing research. Many new to ethnography import outdated notions of ethnography and culture from traditional anthropology into their work, a move that replicates historical mistakes and distorts their results.

2. Science Or Humanity?

Traditional anthropology has long lived with the contradiction of being both a science and a humanity. In fact, ethnography makes this contradiction particularly visible. On the methodological level, ethnography seeks new patterns in ‘found’ data, displaying its kinship with philosophy, history, journalism, and second-language learning, fields typically found in the ‘humanities’ branch of a university. On the level of theory, ethnography usually deals with issues related to sociology, psychology, economics, and political science, disciplines categorized as ‘social sciences.’

The idea of a ‘science of society,’ as first articulated by Spencer and Comte, linked ‘social science’ explicitly with the Western Enlightenment tradition. In the classic formulation of this approach, one deduces hypotheses that should be true if a particular theory is true, where a hypothesis is a statement of a relationship among two or more variables. One then operationalizes the variables by specifying what observations one might make or what questions one might ask in order to measure them—i.e., assign a numerical value where differences in the number represent relevant differences in the phenomena of interest. One selects a sample of a population on some principled basis, the point being to say at the end how confidently one can generalize from the sample to that population. Finally, one specifies a decision procedure so that one can say at the end whether the hypothesis is accepted or rejected. Often a decision procedure involves a statistical test with a cutoff point of statistical significance, significance technically meaning, ‘there is no more than a 5 percent (or 1 percent, or some other figure) chance that I’m wrong when I say that the variables in the hypothesis are related in ways beyond what I’d expect by chance.’

Numerous studies in anthropology have revealed the power and value of traditional science in the context of ethnography. (For a good example of methodology from this angle of vision, see Bernard 1988.) Further, other social sciences recognize the importance of testing hypotheses ‘cross-culturally,’ linking those fields to anthropology. And finally, in a metaphorical sense, much of anthropology serves the purpose of challenging assumptions by comparison among different societies, a ‘hypothesis test’ of a rather fundamental sort. In fact, one can say that social research done only in one society that generalizes to all of humanity is bad science.

However, much of ethnography does not fit the tradition of social science. Consider two example problems. The first—scientific research is objective and replicable. By this is meant that well-designed research, done in the same way in any place by any trained person, will yield the same results. In contrast, ethnographers argue that, for some group X and some ethnographer Y, both the ethnographic process and the ethnographic product will vary as a function of Y as well as a function of X. In other words, an ethnography can take different shapes depending on the biographical and cultural background of the ethnographer. The ethnographer is, in a nutshell, part of the data. It is now a cutting-edge methodological issue—more than one ethnography of X is possible, but not all ethnographies of X are acceptable. The problem is to learn how to specify the difference between the two.

A second problem—the scientific emphasis on control. In traditional science, control is key, from theory through data collection and analysis to decision criteria on whether to accept or reject a hypothesis. When the variable of interest, the dependent variable, changes in value, it must be due to the independent variables and nothing else. Ethnography has a defined starting point, but as it progresses a study is emergent. New concepts are learned that suggest new questions; new knowledge of communication styles suggests new ways of asking them; something someone does opens new avenues of inquiry; new knowledge of social types within the group alters sampling designs; analysis consists of putting new pieces together into patterns rather than breaking them down into measurable variables. The research process is nonlinear, dialectical rather than sequential. An emphasis on control from the outset and throughout the research would eliminate these possibilities.

Many anthropologists now reject the scientific tradition and consider ethnography a humanity rather than a social science (see Geertz 1973 for a classic statement of this position). But the humanities offer no straightforward model for ethnography, either. To offer just a few examples, a look at journalism reveals genre constraints whose a priori structural requirements may be as limiting to emergent research as is traditional science. Models of second language learning deal awkwardly, if at all, with the massive amounts of background knowledge implicated in syntax and lexicon, and it is just this knowledge that an ethnographer hopes to learn and represent. Narrative approaches from both history and literature are promising, since much ethnographic data is narrative in form. However, narrative is built on chronological sequence, and ethnography abstracts and generalizes from such sequences.

Ethnography, then, is caught between agreement with general scientific goals and a poor fit with traditional social scientific practice, and it is caught on a research process akin to the humanities that calls for systematic approaches to modeling and validation that those fields do not provide.

What is needed is an epistemology that fits ethnography, a clear sense of how knowledge is obtained and what it looks like in the end. Epistemologies borrowed from either traditional science or traditional humanities are suggestive and helpful, but at the same time inadequate. Three conclusions are clear. First, a proper epistemology will not come from a simple classification—i.e., ethnography is a kind of social science or ethnography is a kind of humanity. Second, the proper epistemology is, as of now, far from established. And third, the development of such an epistemology will be a contentious issue well into the future.

3. Culture

The concept is relevant here because ‘culture’ traditionally has been used to characterize the results of ethnographic research. In the words of Roy D’Andrade, studying culture today is like studying snow in the middle of an avalanche (cited in Carrithers 1992, an introduction to various views of anthropological research). Traditionally, anthropologists assumed that the small communities they studied constituted a single culture. What people said or did could be described, explained, and generalized by attribution to that single culture, bounded in space and replicated over time.

Any contemporary ethnography now shows that such assumptions no longer hold. Indeed, numerous ‘cultures’—global to national to regional to local— might or might not be relevant to the description, explanation, and generalization of a particular ethnographic moment. There are cultures of generation and gender, of nationality and religion, of ethnicity and region, of occupation and leisure-time pursuit, media inspired and driven by tradition, and in the process of creation. Some cultures span the globe, some distribute unevenly across it, and some take shape only in a particular location. Some people enact a particular culture; some resist it; some ignore it. Some ‘craft a self’ that is a blend of different cultures (the phrase is from Kondo 1990, an excellent example of narrative ethnography). Popular phrases now tag almost anything with ‘culture’: the ‘culture of violence’, or the ‘culture of capitalism’, or the ‘culture of poverty.’

Culture can no longer be used as it was in traditional anthropology. Indeed, its use in traditional anthropology was inaccurate even then. The work of Eric Wolf (1982) showed that traditional assumptions were incorrect, even in the nineteenth century when anthropology first appeared as an institutional discipline. Differences among people still lend themselves to ethnographic investigation. But the multiple and simultaneous relevance of different ‘cultures,’ and the multiple positions of actors vis-a-vis those cultures at different points in time, have turned description, explanation, and generalization into a most difficult problem. Certainly it is now the rare case when any overview of group life can be modeled under a single ‘culture’ label (Hannerz 1992, Kuper 1999).

The culture problem took another twist with a book entitled Writing Culture: The Politics and Poetics of Ethnography (Clifford and Marcus 1986). Writing Culture asked why ethnographies were written the way that they were. What genre constraints did students unconsciously absorb as they read the works of their predecessors? And how did these constraints set limits on the images of the ‘other,’ the world in which they lived, and the role of the ethnographer who did the work?

The book inspired a series of debates that continue to this day. A few key issues can be mentioned here. First, what were the ‘politics’ of the ethnographer community relationship as it developed over time? Any ethnography builds on collaborative relation- ships, more so than traditional images of social science allow. Why doesn’t the writing reflect that collaborative process, with control of the final product distributed across all the participants?

Second, any community lives in a world of layered contexts, from local realities up to the global political economy. Traditional ethnographies tend to feature the local situation as if it hangs isolated in space and time. Clearly it doesn’t. Regional, national, and global histories help explain how the local community came about, as well as the systems of power that hold it in place. Why don’t traditional ethnographies include such material? And what is the proper way to obtain and analyze it, since some crucial data lie outside the awareness of community members who are interviewed and observed?

Third, who is this ethnographer? He or she has a biography, social identities, a history of training, a set of ideological biases, and a vested interest in doing the work and satisfying whoever pays for it. Traditional writing minimized the presence of an ethnographer, if it recognized it at all. In part, the writing reflected notions of traditional science, as discussed earlier. In an objective and replicable study, the identity of the researcher has no influence on the outcome. Now the question is, how and how much of the ethnographer to include in the writing of the results?

These and other issues around the ‘politics’ of the research implicate the ‘poetics’ of the writing, as suggested by the subtitle of Writing Culture. Traditional ethnographic writing does not handle issues of politics and broader national and global contexts and ethnographer presence. What kind of writing does? Narrative? Legal briefs? Postmodern collage? Or, never mind writing, how might an ethnographic product be represented in a film or in a museum exhibit? Here, as with previous topics discussed in this research paper, the questions around ‘writing culture’ have opened up a variety of new possibilities that simply did not exist in traditional anthropology.

The relevance of multiple ‘cultures’ and the call to include previously neglected ‘political’ issues have converted the old concept of culture into a list of issues that, for the near future, will remain under debate.

4. The Growing Ethnographic Family

Until recently, ethnography remained the specialty of anthropologists and sociologists. Now interest in ethnographic approaches to research has increased in a number of fields. Anthropologists often react with dismay to these developments, arguing that critical aspects of ethnography and the culture concept get lost in the transfer. More often than not they are right to do so. At the same time, outside anthropology various models of research are being developed that have a ‘family resemblance’ to ethnography, to use Wittgenstein’s terminology. The existence of this new ‘family,’ together with the indeterminate nature of ethnography within anthropology already discussed in this research paper, make for an even more complicated situation.

Consider the term ‘qualitative,’ the most widely used term to label this family of research. Strictly speaking, ‘qualitative’ labels only a type of data, propositional rather than numeric in form. Yet ‘qualitative’ may or may not also imply differences in methods, analysis, or epistemology. At one extreme, a traditional social scientist might add open-ended questions to a survey, code them with a few concepts derived from prior theory, and announce that ‘qualitative research’ had been done.

An ethnographer, who would also be called a ‘qualitative’ researcher, would criticize these data on several grounds. The open-ended questions are too narrow and the researcher invented them. They represent only interview data without any participant observation. No other contextual data are available to constrain interpretations. Codes are taken from prior theory rather than being developed dialectically with the texts themselves. For an ethnographer, these ‘qualitative’ data are simply alternative measures of prior variables characteristic of traditional science, measures that have little or nothing to do with ethnography.

‘Quantitative,’ on the other hand, supposedly labels traditional scientific research, but the same ambiguities operate with this term as well. An ethnographer, for example, might use census data as part of their investigation of a community. They might also use prior survey results to link local detail with larger populations. Or they might use a ‘natural experiment’ approach to organize ethnographic data to test co-variation among elements of a discovered pattern. An ethnographer uses quantitative data following a different epistemology when compared to traditional social science, as previously discussed. But ‘qualitative’ versus ‘quantitative’ is not a distinction adequate to capture such differences in research.

A second strategy to clarify the nature of the new ‘family’ is to divide it into types. Introductory text- books on qualitative research lay out subdivisions such as ethnography, grounded theory, and phenomenology. Such distinctions confuse historical sources of the family with the family itself. Phenomenology, and later developments such as hermeneutics, helps clarify the philosophical foundations of the family. Grounded theory isolates and makes explicit a part of the research process—the development of summary categories from uncontrolled data. Ethnography, as shown earlier, takes from anthropological traditions the notion that long-term intense involvement— participant observation—is a precondition of exploring a community. But these three historical sources are contributions to this new family of research, not separate ways of doing it. Dividing the family into its historical sources does not clarify what it is and what it is becoming, and ‘qualitative’ won’t serve as an adequate label for the new family of research, either.

For the author of this research paper, the new family is, in fact, an adoption of basic ethnographic epistemology. However, the scope and depth of data and the method of analysis vary widely across different disciplines and problem orientations. Many, probably most, approaches, do not feature participant observation as a sine qua non of the research.

For ethnographers, this new family poses a problem. Ethnography, as it developed in anthropology, is clearly a historical source and a current member of the family. Most family members are recognizable as instances of ethnographic logic in action. At the same time, for anthropologists, the ethnographic tradition defines the prototype for the family, and departures from this tradition count as a decline in quality. Narrow focus and lack of participant observation are two frequent criticisms of such departures. For many ethnographers, then, the growth of the new family is viewed negatively, as the appropriation of a logic without the disciplinary tradition that developed it. This negative view is complicated because, within contemporary anthropology, traditional ethnography is now viewed as inadequate on several grounds as well (Gupta and Ferguson 1997). The paradox generates much debate among anthropologists.

5. Conclusion

Fifty years ago this research paper would have been much simpler but much less interesting to write. ‘Ethnography’ labeled both the way that anthropologists did their research and the report that an anthropologist wrote, a report also called a description of the culture of a community. That definition, by the late twentieth century, had mutated into numerous unresolved questions about the nature of one human setting out to understand some others in our diverse, rapidly changing world. These same questions are now of interest in a variety of fields—other social sciences and humanities, artificial intelligence, biology, business management, and public health, to name a few examples. Ethnography, as practiced in anthropology, has the most elaborate history, the greatest amount of experience with such questions, and its insistence on participant observation remains unique. But what it will become, separately and as a part of the new research family described in the preceding section, remains to be seen.

Bibliography:

1. Agar M H 1996 The Professional Stranger: An Informal Introduction to Ethnography, 2nd edn. Academic Press, San Diego, CA
2. Bernard H R 1988 Research Methods in Cultural Anthropology. Sage, Newbury Park, CA
3. Carrithers M 1992 Why Humans Ha e Cultures: Explaining Anthropology and Social Diversity. Oxford University Press, Oxford, UK
4. Clifford J, Marcus G E (eds.) 1986 Writing Culture: The Poetics and Politics of Ethnography. University of California Press, Berkeley, CA
5. Geertz C 1973 The Interpretation of Cultures. Basic Books, New York
6. Gupta A, Ferguson J (eds.) 1997 Anthropological Locations: Boundaries and Grounds of a Field Science. University of California Press, Berkeley, CA
7. Hannerz U 1992 Cultural Complexity: Studies in the Social Organization of Meaning. Columbia University Press, New York
8. Kondo D K 1990 Crafting Selves: Power, Gender and Discourses of Identity in a Japanese Workplace. University of Chicago Press, Chicago
9. Kuper A 1999 Culture: The Anthropologist’s Account. Harvard University Press, Cambridge, MA
10. Wolf E R 1982 Europe and the Peoples Without History. University of California Press, Berkeley, CA