Monday, May 10, 2010

The Public Treasury, a Tale in Two Acts

Readers of this blog may have noticed that my posts have tended toward the negative as far as the Central African state is concerned. Today, I have positive news: in February, it became illegal for people to pay taxes to officials in their offices. Instead, good citizens must go to the bank, where there is now a special window marked “public treasury.” People deposit directly into various departments' coffers. In the first month with the new system, revenues more than tripled. The introduction of a basic technology thus eliminated a massive amount of pay-yourself-government (corruption, in common parlance). Of course, money deposited in bank accounts may still wind up building ministerial villas in the provinces, but at least this way there's a potential for more transparency.


This morning as I pulled into the market in Sibut, two hours' drive north of Bangui, I heard a familiar voice call out, “Louisa!” I know no one in Sibut, except maybe for the guys at Restaurant Destroy (the name shares sign space with those of the various humanitarian groups who have patronized the establishment), where I usually get an avocado salad and omelet. I opened the door and saw, to my surprise, a friend from Ndele. He works for some Sudanese merchants by accompanying their truck to DRC to buy coffee and then make the long trek north again to Am Dogon, Sudan to sell the beans. Along the way, they stop in Central African villages and peddle Chinese-made pots and tea sets, toothpaste, and dates. My friend's truck stopped in Sibut to await a money transfer of 200,000 CFA (more than $400) to pay for the rest of the voyage to Bangui. He estimated that in all he would pay 550,000 CFA in road barrier fees for the the Ndele – Bangui journey. Gendarmes, soldiers, police, water and forests ministry guards, and other entrepreneurial sorts set up barriers to extort fees on the roads. Because of the armed group insecurity in Ndele, they can now demand more from voyagers – especially “foreign” Muslims – as a kind of proof of loyalty. Pretty much the polar opposite of the efforts to de-personalize taxes described above.


I can't help but wonder, upon seeing the bedraggled, beaten-down trucks that ply this trade, how on earth they can make a profit from selling an occasional tupperware set and some mid-grade coffee beans if they must pay thousands of dollars in bribes for every round trip. And yet they keep doing it.

Experimental Ethnography

My apologies for the silence. I've been fine, mostly in Bangui due to the recent rebel attack in Ndele and a spate of highway bandit activity on the roads around there.


Partly, my no-post laziness has stemmed from plunging into reading in the wake of encouragement, thanks to a friend in Bangui , to explore the new world of economics – Nathan Nunn (who, incidentally, cites my adviser, Charles Piot, on multiple occasions), Edward Miguel, Stathis Kalyvas...(I welcome further recommendations!) I even caught myself thinking like an economist the other day, trying to draw little diagrams of potential causal relations and wondering how one could isolate the factors and determine the links.


As I approach the end of this chunk of fieldwork I'm filled with mixed feelings including, yes, some regret.


Though the idea of fieldwork as a rite of passage is oft-critiqued in anthropology, it nevertheless persists. My department, for instance, offers no methods course. Throw 'em in the deep end and see how they do. Most people swim, even if just dog-paddling. So on one level this rite of passage practice is fine. But I have started to notice myself developing a certain anxiety over the lack of structure in ethnographic research. This lack of structure was less apparent when anthropologists worked in small villages and immersed themselves in every detail of their social structure. The (artificial) geographic boundary of the village provided the contours for the study. But now as multi-sited ethnography has become common (my own research has been more multi-sited than I had envisioned, simply because the places I had hoped to do research have been intermittently unsafe, forcing me to find other options), there is a risk that the great strength of anthropology – its receptiveness and openness – will make for watered-down research unless one has years in which to do fieldwork, which for most of us is impractical. (The most common question I get from ex-pats these days is, “So, are you done with that dissertation?” Sigh. That's not quite the pace of these things.)


All of this is a rambling prelude to my main point: the methodology-envy that hit me upon reading about Elizabeth Levy Paluck's experimental ethnography.


Paluck sought to assess the impacts of encouraging listening to a conflict resolution-themed soap opera in the eastern Democratic Republic of the Congo (DRC). She randomly assigned villages to either receive the broadcasts or not and then designed an evaluation survey that used both quantitative and qualitative measures. At the end of each respondent's interview, the interviewer offered a bag of salt as a thank you. The interviewer then explained that a local humanitarian group had drawn attention to a needy group in the community and asked whether the respondent would share any salt. “Which group?” was the most usual reply. The interviewers asked whether there was a group the person would not feel comfortable giving to. Most people in this area plagued by recent years of violent conflict responded affirmatively. The interviewer then measured the quantity the person said s/he would give, depending on the group. As expected, people gave less salt to members of their rival group. Surprisingly, the results showed, tentatively, that those who had listened to the soap operas were less willing to donate than those who had not.


How fascinating to make social norms and relationships materially visible. I have tried, in my limited capacity as a solo researcher, to do similar things myself, but have bumped into frustration each time because, I think, of my failure to materialize the hypothetical examples I set out.


In my next project, whatever it might be, I would love to systematically integrate some experimental methods. But I have much to learn.