Qualitative methods II: On the presentation of ‘geographical ethnography’

I Introduction

This is the second in a series of three reviews that examine current conventions of doing and describing qualitative research in human geography. Drawing on a sample of 200 papers that were taken to represent this field, our first review focused on what was found to be far and away the most popular qualitative method – the interview. In this review, we turn to the second most common approach – ‘ethnographic’ research. In doing so, we have drawn on any paper that either explicitly claimed to be employing ‘ethnographic’ data or evidently sought to surpass a decontextualised analysis of respondent talk by seeing what further insights might come from a fuller engagement with social action in situ. Rather than engaging in wrangles over what can properly be called ‘ethnographic’, this choice partly stems from how the ‘geographical ethnographers’ in our sample themselves used the term.

Madden observed in 2010 that ethnography was no longer the ‘jealously guarded possession’ (2010: 1) of anthropologists. Many contemporary geographers have been keen to prove his point. Two out of five (80) of our papers contained ‘ethnographic’ elements as we heard from researchers questioning UK austerity policies by waiting for discounted supermarket goods (Kelsey et al., 2019), risking student disdain in North Korean classrooms to understand the local co-existence of socialist principles and capitalist practices (Wainwright et al., 2018), or scrambling ashore after arduous canoe journeys to see what petrochemical wealth has really done for indigenous Amazonians (Wilson, and Bayón, 2018). We found some at Spanish dinner parties listening to gay men discuss the surrogates they hoped would soon bear them children (Schurr and Militz, 2018) and others taking tea with UK prisoners (Richardson and Thieme, 2018). Yet others have endured rat-infested housing in East Africa (Caretta and Jokinen, 2017) or found common cause with locals living under the eye of a less than benevolent Cambodian state (Schoenberger and Beban, 2018). These gutsy researchers will have tackled all sorts of challenges compared to those who cling to the procedural clarity and personal distance typical of other techniques.

So, whilst Herbert (2000) worried about the limited geographical interest in ethnography, and Crang (2002) foresaw even further decline as funders push geographers towards methods more suited to pre-defined questions, at first blush our sample leaves us with a very different impression. Many of today’s geographers are evidently quite taken by the immersive promise of the ethnographic approach. This raises the question: what is ‘geographical ethnography’? If Madden (2010) was right to claim that ethnography has broken free from its association with anthropology, how has it mutated in the process of becoming a human geography commonplace? This question provides the focus for our second review in which we particularly consider the implications of how this work is currently presented.

II Two routes to authority

As discussed in our last review (Hitchings and Latham, 2019), qualitative geographers do not enjoy a stable, universally accepted set of rules for deciding whether a given empirical account is to be trusted. In reflexive social science, within which practitioners are well aware that each research site presents its own unique challenges, the onus is rather placed upon the individual researcher to make the case for why they organised their studies as they have. How then do our ‘geographical ethnographers’ establish their authority to write about those with whom they have spent time in the field? Here we found a divide between those who had travelled to more distant places and those who stayed closer to home.

For the former group, total time in the field commonly provided their route to authority, although this was rarely an openly made argument. Generally, this is expressed in months and years and, of those papers that invoked duration in this way, total time in the field never dipped below six months. The average was around an impressive two years (though some were cagey with their readers about whether this actually represented the start and end of a series of trips). Examples include Grant’s (2018) work involving 17 months within a Tibetan enclave in China, Clark et al. (2017), who spent two years exploring how Nepali villagers live with water, Naylor’s (2018) three-year study of Mexican coffee co-operatives and ‘fair-trade’ products, and Cook’s (2018) 15 months with Jordanian olive oil producers. Some were also seemingly there for much longer than that as part of a commitment to an ongoing relationship with their chosen field site (for example, Karrar and Mostowlansky, 2018). Though they will have surely needed some respite, the presumption here is that these researchers were doing whatever was needed there to understand their topic – literally thinking on their feet in the field.

Those who did not travel so far were more inclined to describe specific data collection events. For example, Taylor and Carter (2018) specify how they undertook daily observation and photography of in-water sessions, alongside attending clinic sessions and consultations, to understand dolphin therapy; Roelofsen and Minca (2018) took part in 50 Airbnb events – either welcoming guests or being welcomed by others – to understand how these social interactions played out; Bos (2018) observed five military videogaming sessions, along with interviewing people as they played, to explore this form of everyday geopolitics. Meetings were also popular amongst the second group, with Boyce (2018) attending meetings between activists and officials to question border policing practices in Detroit, Cockayne (2018) being present at over 100 conferencing and networking events in San Francisco to explore gender performance in the digital economy, and Kay (2018) observing industry meetings and conferences to see how American conservation finance companies handle forests. We found only three cases of total fieldwork duration being invoked in ethnographic work closer to home, and then this was done more precisely than in the studies further away: Hall (2016) undertook 200 hours of ethnography in various situations to see how difficult economic times are impacting upon Manchester families, Price-Robertson and Duff (2018) did ethnography for 155 hours in sites including mental health support groups and courtrooms to understand the Australian ‘family assemblage’, and Shin (2018) spent 170 hours in the ‘main street area’ of a London suburb to question how Korean enclaves work.

Presumably these events involved a more concentrated kind of data collection than that undertaken by those who ventured further afield. Either way, authorial authority came through two distinct routes in our sample. Those who travelled significant distances to their field sites present data collection in a comparatively loose way by invoking overall fieldwork duration based on the assumption that cultural understanding comes through a slow process of engagement. For those who did not journey so far, specified activities and events were the common currency such that readers are better placed to imagine the social scenes involved and to evaluate the presented findings accordingly. Notwithstanding how this division perpetuates certain notions of ‘exotic’ and ‘familiar’ field sites that some would challenge, this leaves us wondering whether these two ways of researching and reporting could themselves be more mobile. Might we more often take a more immersive approach in our own countries? And, turning to our next section, might some of the travelers say more about which strategies worked best for them?

III Magical moments and how to get them

When ethnographic field material featured in our papers, the vignette was predominant. We found instances of researchers reflecting on public reactions to them breastfeeding (Mathews, 2018), Turkish officials boarding buses to examine the documents of those inside (Isleyen, 2018), students responding to assessments in North Korea (Wainwright et al., 2018), an activist railing against the treatment of suspected poachers in India (Baroba, 2017), and a Javanese procession that might initially appear straightforwardly traditional but could also represent the enactment of a more hopeful future for those who live with flood risk (Bunnell et al., 2018). This is just a selection of examples – more than half of our ethnographic papers featured vignettes. Given this popularity, it is worth thinking about how this presentational device, often found at the start of an empirical section, is being used by geographers. Certainly, we enjoyed reading these vignettes and, if their purpose is to draw the reader into the account, our ‘geographical ethnographers’ are often doing an excellent job. But it also feels churlish to challenge the analysis that follows (the reader rather feels they should submit to it since the author has done their best to write engagingly). And that could be a problem if it stops us from considering other possible interpretations. Finally, it is worth noting how these vignettes often effectively served as ‘magical ethnographic moments’ in which the key themes of the paper were perfectly realised in a field encounter. And that did not often pair very well with a reflexive account of the practical research that preceded them (though see Kelsey et al., 2019, and Wainwright et al., 2018, for two exceptions). Instead, the aim was often to jump directly from vignette to analysis such that delving into the backstory rather feels like an annoying brake on the pace of the paper. Vignettes can, of course, do many things for us. They can convey a sense of how fieldwork proceeded. They can illustrate turning points in the data collection process when working assumptions are suddenly called into question. And, to be clear, we are not arguing against the value of vignettes here. Our observation is rather that, intentionally or not, currently popular ways of using them may be discouraging us from saying more about the nitty gritty field practices through which the possibility of observing such exemplary scenes was achieved.

This is not to say that we learnt nothing from our colleagues about how to implement ethnographic techniques. Evans (2018) says you can develop rapport through your failed attempts to wear culturally appropriate dress. Simcik Arese (2018) reassures us that, though you may be considered a comical outsider at the start, that could eventually help you become a kind of external confidante. Dyson (2018) suggests spending time outside buildings, if you can, because doing so will naturally draw you into the more public conversations that take place there. Hall (2016) describes how she became ‘a familiar face’ though various strategies and, if respondents invite you to stay over, you can benefit from accepting the offer (Basnet et al., 2018). Drawing on shared worries might also break down perceived social barriers (Schoenberger and Beban, 2018), and walking together with people can create a productive interactional dynamic (McLean et al., 2018). Perhaps unsurprisingly, ‘geographical ethnographers’ also suggest we should pay particular attention to how patterns of talk vary according to context – how children speak of sexuality differently depending on where they are at school (Hall, 2018); how certain forms of self-presentation are found in the workplace (Cockayne, 2018); and how the specificity of the sales pitch can convey a particular affective vision for cities (Ernwein and Matthey, 2018). Other studies provide some relatively prosaic but no less useful pointers: if you want to understand social interaction inside shops, go when the shops are busy (Kaplan and Recoquillon, 2016); if you want to understand how weather influences the countryside experience, go in varied conditions (Adams-Hutcheson, 2019); biking across field sites can help you appreciate them more fully (Brock and Dunlap, 2018).

So, on sifting through our sample, we extracted some nuggets of useful ethnographic advice (in which attending to how talk patterns are tied to particular places seems especially promising). But there were not that many overall. Perhaps this is unsurprising and goes beyond how vignettes are currently used. The ethnographer, probably more than any other type of social researcher, cannot say everything about their data collection procedures or, perhaps more accurately, the story of their field research when the point is often for that to evolve. However, our ‘geographical ethnographers’ still often seemed to be slipping relatively effortlessly into the social scenes that interest them. Returning briefly to some of the above examples, we can see this happening on arrival at an Airbnb, at the back of a meeting room, or perhaps even when people are already engrossed in dinner party conversation. But it is unlikely to be the case everywhere. And especially not for many of the geographers who are tackling sensitive topics in daunting contexts. We found exceptions in which the authors provided a valuable sense of how they handled such challenges (such as Caretta and Jokinen, 2017; Basnet et al., 2018; Simcik Arese, 2018; Hall, 2016). But this is not currently part of how we routinely report upon our ethnographic practice in human geography papers. And so, whilst anthropologists have long agonised over ethnographic writing strategies (Clifford and Marcus 1986), in a less introspective vein this leaves us pondering what other ways of reporting on ethnographic fieldwork have been squeezed out of our papers by the geographical ascendance of a particular kind of vignette. For a method whose authority, as we have seen, is often linked to temporal immersion, it is not common to speak of how insights gradually emerged from a range of practical endeavours. We almost never see scratch notes that authors work through with us or reflections on early thoughts and accounts of where the analysis went afterwards. In current work we are reading what Van Maanen (2011) calls ‘realist tales’ depicting the apparent truth of life in the field more often than ‘confessional’ ones that speak of the researcher’s journey into understanding it.

IV Adding some ‘ethnographic richness’

Even more common than the vignette, in terms of the empirical material that we saw in our sample of ethnographic papers, were quotations. Almost all of our ethnographers drew on interviews in some way, as detailed in our last review. That too is perhaps unsurprising. Hammersley has noted (2006: 9) that ethnography has always ‘relied very heavily or even entirely on interviews’. But what, for current ‘geographical ethnographers’, is the relationship between interviews and observation? Haugen (2018) tells of how observations provided detail that her respondents might not recall in Nigeria. Cockayne (2018) suggests that paying close attention to workplace presentation can attune you to how they handle the interview itself. Counter (2018) argues something similar in suggesting that researchers should look for recurring themes that surface in both formal interviews and casual conversation. Munoz (2016) talked of how her repeated interviews with a key respondent built a sense of trust that opened up unexpected avenues for her street vending research. Richardson and Thieme (2018) say their interviews were enriched by an ‘ethnographic sensibility’ that came from spending time within the prisons that they studied. Other authors, however, say less about how the two were combined. Graf (2018) tells us only that accompanied trips ‘provided further insights’ into the views of the Eritrean diaspora. In an otherwise full account of methods, Motzer (2018) simply states that participant observation ‘improved’ data collection in her study of US regional food hubs. Wilson et al. (2018) say only that observing university-industry seminars ‘gave depth’ to their analysis. Clark et al. (2017) tease us with passing talk of ‘the invaluable insights’ that were provided by field notes in their Nepalese water project and, more widely, we saw not infrequent mention of the appealing extra ‘richness’ that ethnographic elements apparently add (see, for example, Bunnell et al., 2018, or Shin, 2018).

So, though we are left with some sense of how they were combined, more generally the difference between an interview and an ethnographic study was often unclear. At times, it rather seemed that, if you’ve gone somewhere to do your study and you’ve noticed something outside of a formal interview, you should probably frame your work as ethnographic. But are observations explicitly drawn on to encourage interview respondents to talk about topics they would otherwise avoid? Are they used to press them on matters they may not otherwise notice? Is our presence about building trust with those we plan to interview later? Or is it simply to root our questions more fully in the contexts of their lives – to be sure that we know what to ask and how to ask it? Perhaps the process worked the other way? Have, for example, our interviewees tipped us off to social actions or scenes that deserve more systematic observation? In line with how the ethnographic journey towards understanding is often obscured by how vignettes are used, it is currently rare, for example, for researchers to reveal the sequencing of their activities (though see Bos, 2018, for an exception). Of course, we know that many ethnographers take their interview chances when and where they can. But they will have still had an idealised plan at the start. So, when Russell Bernard (1995) argued there has always been a ‘mystique’ about the practice of ethnography, in current geographical work, this particularly seems so when thinking about the ‘depth’ or ‘richness’ that ethnography apparently adds to interviews. In saying this we are not arguing against adding these elements. Rather we think that saying more about their role in the recipe could help ensure that more of our studies end up with these appealing attributes.

V Doing away with ethnography

By this point, readers may be hankering after an overview of the practical data collection techniques adopted by the ‘geographical ethnographers’ in our sample. Surely this should be central to any evaluation of what is happening with this work. We have left this topic until now, focusing largely on the final presentation of research, because we often came to the end of papers without a strong sense of these aspects. What were these researchers writing in their field notebooks, if anything? Where and when were they doing it? How did these approaches develop? Admittedly, one reason why this material is not often presented is because so many activities can be undertaken under the justificatory banner of ethnography. If ethnography is, probably more than any other method, defined by an ethos of practical experimentation – of testing out what helps us to identify and answer our research questions (Becker, 2009) – then many things might be done in the field, and so perhaps authors cannot be expected to itemise them all. Nonetheless, whilst ‘geographical ethnography’ is currently typified by ‘observing’ social action in situ and interviews (we saw this more often than studies involving ‘participation’, for example), a good number of papers do just straightforwardly state that they did ‘ethnography’.

Self-evident as such statements may seem, we also note how they cast a shroud over our research practices, spiriting them away from the presented analysis, and allowing them to escape any scrutiny from outside. Furthermore, this situation does rather tempt researchers to grab at some field examples that fit with the ideas and concepts they have chosen to apply before returning to resume the debate within more comfortably familiar environments back home. This is not to suggest that the geographers who found their way into our sample were doing this (many had undertaken impressive amounts of fieldwork). Rather our observation is that this reporting practice opens the door for less empirically inclined researchers to ‘play the ethnography card’ – baldly deploying the term and daring reviewers to ask for more on what actually happened in the field. One final feature that we noted in our second sample of papers was how some of those who said the most about how their empirical studies practically proceeded refrained from much mention of the overarching term of ‘ethnography’ despite clearly being quite interested in how people respond to various social and physical situations (see, for example, Adams-Hutcheson, 2019; Veal, 2018; Von Benzon, 2019). It was almost as if they had moved beyond it – that the broad-brush notion of ‘ethnography’ had become too unhelpfully baggy for them when they wanted to evaluate the potential of specific techniques. That leaves us wondering whether geographers more generally might do away with the term if, in practice and rather ironically for one of the most reflexive of research methods, it ends up discouraging us from saying a great deal about what we did, why we did it, and, perhaps most importantly of all, how that all went.

VI Conclusion: And now the elusive ethnographer

This review has considered how current ‘geographical ethnography’ is presented. Based on a survey of recent work in the discipline, we started by noting a healthy geographical interest in ethnographic methods and the diversity of contexts within which current geographers were proving their worth. With this in mind, we then subjected this work to further inspection. We found this to be an endeavour for which authorial authority is derived from either duration when researchers travel to more remote field sites or a more precise account of activities if they stay closer to home; we saw how field experiences are most often presented through engaging vignettes in ways that may encourage us to say less about the backstory of the study; we observed how their ethnographic work is closely connected to the interview, but in ways that might be more fully detailed; and we identified some pitfalls in straightforwardly presenting fieldwork as ‘ethnographic’. In this respect, though current ‘geographical ethnographers’ are no doubt quietly tackling all sorts of practical challenges as they embark on the process of entering and examining their respective field sites, we end our second review by asking whether more might yet be said in their papers about how various analytical strategies were combined along the way. Admittedly such matters may be discussed more fully in other outlets elsewhere (in books and in theses and on project websites) where the word count is less prescribed and limited. But the academic paper is still the mainstay of exchange between peers in human geography. And, on looking across our sample for this second review, it does rather seem that, like the ‘invisible interviewer’ who we introduced in the first, the geographical ethnographer can often prove quite an elusive figure in terms of leaving the reader with a strong sense of how specific strategies were adopted and refined as individual studies went on. Though, as was the case last time, this leaves us wondering about missed opportunities for collective methods learning, this situation is not necessarily a problem, depending on how we see the purpose of the papers that we write for one another. And that is a topic for our third and final review.