JOHN F. MCCAULEY
▶ FIELDWORK LOCATION: BURKINA FASO
Dolo, the beer brewed from millet in villages across Burkina Faso, is different from the beer one might enjoy in Western contexts. Different in taste, of course, and different in presentation, but most importantly, different in social meaning. As the opaque, honey-colored brew served warm in large calabash gourds goes down by the liter, and village elders express their satisfaction in words you cannot understand, a process surprisingly central to conducting field research unfolds.
My familiarity with that beer is part of a story about obtaining authorization to field a population-based survey in the village of Boussoukoula. Quite apart from the tedium of completing forms from my computer at home, I learned in Boussoukoula that the authorization process can also require long days and nights drinking with village elders.
Researchers interested in gathering data from human subjects via surveys, for example, should be well acquainted with the institutional review board requirements at their home institutions. Those guidelines protect the subjects of our research from unethical treatment, in keeping with the 1979 “Belmont Report.” In addition, researchers working abroad must be aware of any additional expectations from in-country. In the West African countries where I do research, this often means making a good-faith effort to share research designs with a national ethics office. In some places, approval must also be obtained at a subnational, provincial level. Although our formal obligations as researchers often stop there, gathering data may never get off the ground without the blessing of authorities in the localities where we work. And, from my experience working in West Africa, it’s hard to say ahead of time what that might require.
As a graduate student, I conducted field research in four sites located on each side of the Burkina Faso–Côte d’Ivoire border, and this research was later published with my advisor (and fellow contributor to this book), Dan Posner. Using a cross-border research design that he and others popularized,1 I aimed to gauge individual-level attachments to religious versus national identities, via surveys, in different political contexts situated in very close proximity (to minimize potential confounding factors). Two of the locations—Niangoloko in Burkina and Ouangolodougou in Côte d’Ivoire—could be classified as towns, a short drive from the border on the main road running between the two countries. The other two—Boussoukoula on the Burkinabé side to the north and Kalamou on the Ivoirian side to the south—are nestled on the eastern part of the border fairly close to a third frontier with Ghana. Boussoukoula and Kalamou are very small, remote, rural villages with predominantly Lobi ethnic populations and are connected by a dirt path and separated by only about seven kilometers. Depending on the security context at any given time, one may not even notice the international border on the path between the two.
My fieldwork started in Boussoukoula where, as in many other small, rural villages in the region, customary traditions remain firmly entrenched. To give a sense of Boussoukoula’s insularity and attachment to traditional culture, the first churches and mosques—markers of social influence from outside—did not arrive until the 1990s. Given its size of only several dozen households, it is also quite impossible for outsiders to arrive and leave unnoticed. Of course, that is particularly true for Westerners.
I was accompanied to Boussoukoula by two research assistants and a research coordinator, my longtime friend Ollo—the son of a Lobi chief whose name indicates his place as the third son born to his mother. On arriving, we were met by the delegate to the chief of Boussoukoula. There must have been a local government official appointed to the village, but we were never introduced nor made aware that such introductions would be important or necessary. That is not where local power lies in Boussoukoula. The delegate took us to our accommodations and let us know that we’d be welcome for as long as we needed to stay.
This was the first indication I received that the fieldwork plans I had made would, shall we say, evolve. Our research team had planned to stay in the nearby town of Batié, where we had found paid accommodation knowing that no such accommodation existed in Boussoukoula. Our intention was simply to drive out in the mornings and return in the evenings until the data collection was complete. Instead, however, we were graciously given our own mud hut with a concrete floor, plastic mats for sleeping, and a kerosene lantern. As guests, we were deeply grateful, particularly because we were able to avoid the bumpy back-and-forth ride to Batié. As a researcher, I have to admit that I had other concerns. First, we spent our nights lighting and relighting mosquito coils because we were in the middle of the rainy season in a porous mud hut with a small window but obviously no screening. I was often awake to manage that task because the rolled up t-shirt that served as my pillow quickly lost its efficacy, and I never figured out a sleeping position suitable for a thin plastic mat on a concrete floor. Needless to say, I was not well rested during the week we spent in Boussoukoula. I had also planned to spend some time entering data as it came in daily from the research assistants who served as enumerators. That proved entirely impossible: there was no way to charge my computer’s battery, little light to review the data in the evenings, and just too many interpersonal expectations to make the solitary task of data entry feasible.
The conditions were not new to me. Prior to graduate school, I had been a Peace Corps volunteer in a different part of Burkina Faso, also in a small village with no electricity or running water, and I like to think I’m pretty comfortable in such environments. Yet in doing research and collecting data that I hoped would eventually be published in political science journals and contribute to our collective knowledge, I guess I thought that my workspace and working environment would need to be orderly and controlled. I should not have assumed this, and they were not.
Once we were settled, the delegate—a quiet, earnest man of middle-age who had the obvious respect of community members—showed us around the village. We passed the chief’s compound, the rudimentary health clinic, the primary school, and the main well around which women were gathered to fetch water, and he introduced us somewhat informally to families along the way. People greeted us warmly in Lobiri, the local language in which the enumerators would conduct the surveys, and children stared and often tagged along. Eventually, we wove our way back to the chief’s compound, which was a somewhat more elaborate set of huts, and took up seats on wooden benches under a large tree just outside. It was late morning at this point, and the sun was high in the sky; we were pleased to be sitting in the shade and vaguely aware that introductions were starting to become more formal.
The delegate waved to the brewing station a stone’s throw away, and two women brought over several liters of dolo, in jugs like the ones in which gasoline is sold on the informal market around small villages like this one. I am certainly not the first to note the central role of beer in conducting local research—in 1934, Monica Hunter described her attendance at “beer meets” as part of her observational data collection among the Bantu in South Africa2—but I hadn’t even begun the research yet. Our calabash gourds were filled, and other members of the chief’s cabinet and the council of village elders gradually joined us. We explained the very general contours of the research, talked about where we were from, and asked questions about Boussoukoula, pausing in between for translations between French and Lobiri. The elders described their roles and shared stories from local life in the past, which did not seem very different from the present. They filled their own calabash gourds, and the women brought more jugs of dolo. At some point, we were presented with trays of meat that I later learned included sheep brain, a local specialty. At a later point, the sun sunk beyond the horizon, and a few young dancers did an informal but lively performance for us, accompanied by a version of a xylophone. More dolo was served, and at a time of night that, in the absence of electricity, felt very late, the small group began to thin out and we stumbled back to our hut.
Any semblance of control I thought I might have over the agenda was thoroughly dispelled by the morning. To be honest, I was unsure if we had actually received permission from the elders to go forward with the data collection. So Ollo, the research assistants, and I just went with the flow when we were first brought a porridge breakfast and then met by the delegate for another walk around the village, figuring that at some point we would have the opportunity to map out our sampling frame and start working. The walk happened earlier in the day on that second day, but it led to the same place: sitting outside the chief’s compound, drinking bottomless gourds of dolo, conversing with the elders, and nodding along to the conversations that either went untranslated or were lost to intoxication. Finally, as night again fell, the chief came out to briefly greet us, and afterward the delegate explained that we were invited to carry out our research beginning the next day. Any assistance we needed would be readily forthcoming.
To recap, we spent over a day and a half drinking dolo, the significance of which only became clear at the end of that process. I ate very good food that I sometimes could not identify, and our conversations with elders ranged from the mundane to the profound, with plenty of quiet drinking or awkward silence in between. Mostly, we invested time with these keepers of tradition, culture, and respect in Boussoukoula. I felt perpetually drunk and had spent those two days not even touching my planned research, but the time on those wooden benches outside the chief’s compound was well spent. It was the required local authorization process that we needed to undergo to successfully carry out our work.
I took one concern and three important lessons away from this experience in Boussoukoula, which was repeated to some extent in Kalamou, although less so in the larger towns of Niangoloko and Ouangolodougou. My concern was that the time spent in Boussoukoula prior to beginning the data collection might somehow bias the outcome of our research. Community members were well familiar with the research team by the time the enumerators showed up at their homes to conduct the surveys. I personally attended few surveys beyond the pilot phase to avoid translation difficulties and potential enumerator effects (as the lone Western enumerator), but respondents all associated the research with me, the foreigner who had come to visit. If that familiarity had bred expectations about our survey outcomes, the time spent in Boussoukoula prior to the data collection could have exacerbated concerns regarding social desirability bias. In hindsight, the key outcomes of interest—individuals’ primary modes of self-identification—were not likely to have been affected by my presence, and other researchers have certainly spent far more time in localities prior to collecting quantitative data, even if those localities differ in size and insularity from Boussoukoula. So I have no reason to suspect the introduction of additional bias, but it is nevertheless worth keeping in mind how our presurvey engagement with communities might affect the outcomes we wish to measure. In a separate research project I conducted in northern Côte d’Ivoire during the period in which the rebel Forces Nouvelles controlled the territory, I was assigned a rebel soldier who accompanied me and the research assistants to households. In that case, I made a deal with the soldier to keep his distance from households when we entered so as not to influence respondents.
Regarding lessons learned from the research in Boussoukoula, the first is that the time we allot for our field activities while sitting in the comfortable confines of our home universities rarely corresponds to the unexpected twists that arise in the field. There are simply too many things that we cannot foresee. In my research in Boussoukoula, only two days were “lost,” although to even think of them as such is to overlook their importance to the research process. I have since dealt with hang-ups over authorization at the national level in Chad, imprisoned enumerators in Niger, delays in questionnaire translations in Burkina Faso, and other issues that have cost significantly more time. I now try to build in extra time that can serve as a buffer or, if all goes well, be used for additional soaking and poking, qualitative data collection, and building familiarity with my research context (which we should prioritize in any event to convey the expertise that is expected in talks and papers).
The second lesson learned is that the notion of consent can be complex and ambiguous in the field. Our IRB protocols emphasize the fair treatment of individual subjects and mandate that we obtain voluntary consent from each of them to go forward with a survey interview. In Boussoukoula, however, I learned that consent may be perceived as village-wide, and that traditional leaders have much to do with the consent that any one village member might extend. Lee Ann Fujii noted that it is reckless to attempt data collection in the developing world without authorization from local authorities, and their importance in the issuing of consent is a critical reason for this.3
Finally, we are inclined to think of ourselves as researchers and the participants in our research as “subjects.” But on that trip to Boussoukoula, I learned that the local people see us first as foreign visitors, with all the good and bad that that entails. Before we can expect to take, we must first develop a level of trust with those whom we expect to give. This can mean long sessions of drinking with elders, eating unfamiliar foods, and demonstrating sincere interest in local community members as partners, particularly in small or tightknit localities. I recognize that some researchers may be unwilling or unable to consume alcohol or meat, and that female researchers may face additional barriers when it comes to building these relationships. I think it would have been possible to avoid certain aspects of the “authorization process” had I not been comfortable, but in that case it would be doubly important to have a competent and culturally informed colleague like Ollo to manage expectations.
I have not been back to Boussoukoula since that project was completed, but the authorization process we went through there remains among my fondest memories as a researcher … at least what I can remember of it in between the dolo.
______
John F. McCauley is associate professor of government and politics at the University of Maryland.
PUBLICATIONS TO WHICH THIS FIELDWORK CONTRIBUTED:
• McCauley, John F., and Daniel N. Posner. “The Political Sources of Religious Identification: A Study on the Burkina Faso–Côte d’Ivoire Border,” British Journal of Political Science 49, no. 2 (2019): 421–41.
• ——. “African Borders as Sources of Natural Experiments: Promise and Pitfalls,” Political Science Research and Methods 3, no. 2 (2015): 409–18.
NOTES
1. Daniel N. Posner, “The Political Salience of Cultural Difference: Why Chewas and Tumbukas Are Allies in Zambia and Adversaries in Malawi,” American Political Science Review 98, no. 4 (November 2004): 529–45, https://doi.org/10.1017/S0003055404041334.