Anderson, L  (2006) 'Analytic Autoethnography', in Journal of Contemporary Ethnography, 35 (4): 373 - 95.

Autoethnography emerges as result of the number of background interests: blurred genre, self reflexivity, an increased interest in emotion, and 'post-modern scepticism regarding generalization of knowledge claims are' (373). However, most of the work so far has been in  'what Ellis  (1997, 2004) refers to as "evocative or emotional autoethnography"' (374). An analytic variant is also possible, however, given certain conditions. In particular, it will be necessary that  'the researcher is (1) a full member in the research group or setting,  (2) visible as such a member in the researcher's published texts, and  of bracket three) committed to an analytic research agenda focused on improving theoretical understandings of broader social phenomena' (375).

Ethnographers have always drawn on personal experience and biographical insight, from the early days of the Chicago School  [lots of examples are given 375 - 7]. However, despite the occasional confession, most of those ethnographers were not self conscious about the method or their personal involvement. There have been very detailed self observational accounts of particular actions, such as learning to play piano jazz, but these have lacked a broader ethnographic concern. One of the earliest advocates of autoethnography proper was Hayano (1979), suggesting that the end of the colonial phase would involve more observation of familiar settings in which ethnographers were also members.

However, autoethnography soon came to break with analytic and realist conventions, and thus to question conventional sociology itself. Autoethnography openly converged with novel-writing and even poetry and performance, in order to  'create an emotional resonance with the reader' (377). There was scepticism towards any attempt to accurately represent others or to produce generalizable discourse. Such evocative autoethnography also tended to focus on emotional experiences  'such as illness, death, victimization and divorce' (377). Epistemological claims have been much discussed since. However, analytic autoethnography is equally possible.  [spelling out the conditions listed above...].

If the research is a complete member of the group being studied, they can clearly become full participants as well as observers. Membership maybe opportunistic or as a result of conversion  [Wacquant's work on boxers is an example of the latter]. However, there are still tensions in that participants are also still social scientists, still have to record events, and still need to balance some kind of objectivity with full participation. In Anderson's case, while his fellow skydivers were rehearsing jumps on the way up, he was trying  'to consciously observe and it should conversations and events deeply enough in my mind that there will be able to recall and recall them' (380). As an ethnographer, he was acutely aware of the variation existing among members in similar positions. Overall, instead of claiming some immediate and direct knowledge as a result of being a member, it is better to see that knowledge emerges from  'engaged dialogue' (382).

This leads to the point about reflexivity, which includes 'an awareness of reciprocal influence between ethnographers and their settings and informants' (382). This has not always been clarified by conventional ethnographies, but is central to autoethnography, since full participants can reflect upon the effects of settings on themselves and their cultural understanding. Sometimes, such reflection 'transforms the researcher's own beliefs, actions, and sense of self' (383).

In traditional ethnography, the researcher is often invisible yet omniscient, although this is not always the case. However, in autoethnography, the researcher is highly visible, and personal experiences become part of the data. This was often the case with field notes, but not in conventional published texts. This textual visibility demonstrates personal engagement, and can illuminate issues of membership and participation, and the construction of meaning and understanding. However, it is not enough to offer mere self absorbed description, and detail of the emotional lives of researchers One important requirement is to extend theoretical understanding of social processes. It is also essential to describe other participants and to grasps something of their otherness. Even reflexivity should be seen as  'relational' (386), a dialogue between researcher and other participants. Thus introspection needs to be based on data and the views of others.

The analytic agenda is crucial if the intention is to gain insight and not just reveal personal feelings. Empirical evidence is still needed  'to formulate and refine theoretical understandings of social processes' (387), as a self conscious commitment. It is not enough to describe what is going on, and this is what makes ethnographies different from various other first-person narratives. The aim is not to provide completely valid knowledge, but rather to  'contribute to a spiralling refinement, elaboration, extension, and revision of theoretical understanding' (388).

Analytic autoethnography is simply a branch of ethnography rather than some completely new discipline. It offers methodological advantages, including greater commitment to particular social groups. At the same time, participants must not disappear into the field. Autoethnography gives an unusual access to insiders, but it must not remain just with insider interpretations. Personal experiences can illuminate interesting aspects of participation -- such as the thoughts going through the head of a skydiver  [which Anderson is]. Above all, autoethnography offers  'grounded opportunities to pursue the connections between biography and social structure' (390) and their mutual interlinking. It provides a kind of sociological self understanding about how the personal is connected to the socio-cultural. Nevertheless, it is obviously not always available unless we belong to particular groups.

Overall, progress does not always follow from substantial epistemological breaks with existing research traditions. Oddly, autoethnographers seemed not to be particularly  concerned with autoethnography itself, especially analytical autoethnographers. This has left the field to the more evocative kinds.

back to key concepts