The politics of the census in consociational democracies

IDD teaching fellow Dr Laurence Cooley has recently been awarded ESRC Future Research Leaders funding for a project on the politics of the census in consociational democracies. The project will begin in February 2017. We sat down with Laurence to discuss his research plans.

Can you tell us a bit about what the focus of the project will be?

The main focus of the project will be on the relationship between the politics of the census and the design of political institutions in deeply divided societies. In particular, I will be investigating cases where national, ethnic, religious or linguistic divisions are accommodated through the use of power-sharing institutions – known in the academic literature as consociational democracies. The basic idea of consociational power sharing, which is probably now the most common institutional device employed in peace agreements and post-conflict constitutions worldwide, is to manage potential conflict between groups by guaranteeing each of them representation in parliament, government, the civil service, and sometimes the police and military, often in proportion to their shares in the population. A potential side effect of this means of managing conflict can be that the census, from which population shares are calculated, itself becomes a focus of political mobilisation, contestation and conflict. This contestation takes a variety of forms, but can include debates about whether questions on identity should be asked in the census, what the wording of such questions and tick-box answers should be, civil society campaigns to influence people’s answers to census questions, and debates about the interpretation of census results and what they mean for the design of power-sharing institutions. The aim of the project is to investigate how institutional design affects census politics, and vice versa.

How common are these debates about the census?

That’s hard to say, because they haven’t necessarily been documented in comparative research. One of the things that I’ll be doing early on in the project is to try to map out instances of contentious politics surrounding the census. My plan is to explore four cases in depth, though. These are Bosnia and Herzegovina, where a census originally planned for 2011 was delayed until 2013 and remains the subject of intense debates; Kenya, where the 2009 census followed closely on from the 2008 post-election violence; Lebanon, where no census has been conducted since 1932 due to sensitivities about the implications for the country’s power-sharing formula; and Northern Ireland, where successive censuses have been accompanied by speculation about the implications of shifts in demography for the future of the country. I’m most familiar with the Bosnian and Northern Ireland cases, so will likely start with conducting the fieldwork element of the research in those two countries. I haven’t conducted research in Kenya or Lebanon before, but I’m looking forward to learning more about those places and eventually to incorporating insights from my research there into my teaching. There’s plenty of scope for further research, too, with some other recent census controversies including those experienced by Sudan in 2008 and Myanmar in 2014, and in relation to Nigeria’s forthcoming but delayed census.

So is this something that has already been written about in academic literature?

Not really, or at least not systematically. There is some literature on the role of the census in the construction of identities – basically, documenting how the census was one of the mechanisms that states have used to build national identities amongst their populations. Benedict Anderson added a chapter entitled ‘Census, map, museum’ to the second edition of his seminal work, Imagined Communities, published in 1991. The anthropologist Bernard Cohn also wrote about the colonial censuses in South Asia – as has sociologist Charles Hirschman. These works have inspired more recent scholarship by sociologists and anthropologists, and in 2001, Cambridge University Press published an edited book called Census and Identity: The Politics of Race, Ethnicity, and Language in National Censuses. Since then, a number of political scientists have conducted comparative research about census politics, including Gëzim Visoka and Elvin Gjevori and Florian Bieber. I am interested in debates about role of the census in identity construction, and one of the things that excites me about this project is the chance to revisit some classic works of anthropology such as those by Anderson and Cohn, but the relationship between institutional design and census politics hasn’t been explored in much depth at all, and this is where I think my project can contribute something new.

Where did the idea for the project come from?

Bosnia was one of the case studies that I explored in my PhD research, and in the latter stages of that work it was difficult to avoid the controversy surrounding the census eventually held there in 2013. Having previously studied at Queen’s University Belfast, I also had a sense of the importance of the issue in Northern Ireland. The idea grew out of those experiences really, and I presented it at an IDD research awayday last summer. Shortly after that, I got an e-mail from our College Research Support Office about the call for applications to the ESRC Future Research Leaders scheme, and I worked up the initial idea into a full funding proposal. I’ve always had an interest in human geography dating back to school (and I’m pleased that as well as working with Stefan Wolff here at Birmingham, I will have as a secondary mentor the geographer Ian Shuttleworth, from Queen’s) and I suppose that explains my fascination with the census! Most citizens probably don’t think about the census that much, other than completing one every ten years, and while many social scientists use census data and are therefore concerned with its accuracy, they don’t tend to engage with questions about the process and politics behind the production of that census data. The census is quite a mundane exercise, but when one scratches the surface, there’s a lot for those of us interested in topics ranging from identity and conflict to state-building and constitutional design to engage with.

Isn’t the census a rather old-fashioned exercise, though? Are some states likely to abandon the census in favour of more regular surveys?

It’s true that there are ongoing debates about the utility of a decennial population count, given the expense of such exercises but also because a lot can change in the intervening periods – especially in contexts of conflict and high levels of migration. One option is to replace the census with a population register or with administrative data that the state already collects from its citizens. While this alternative has been pursued by some European countries, many developing and post-conflict states arguably lack the administrative capacity necessary to do this, and even amongst the advanced economies, many states appear to be committed to the future of the census. There also appears to be strong cultural attachment to the census in some societies, as the recent Canadian experience shows. Canada abolished its long-form census questionnaire in 2011, replacing it with a household survey, only to reintroduce it in 2016. Surveys can certainly provide more timely information based on a sample of the total population, which is clearly important in a post-conflict or development context, but a count of the entire population is necessary in order to construct those samples accurately. The census also provides important data that is required for the holding of elections, which have been prioritised in post-conflict state-building efforts. For this reason, donors are often keen to promote the holding of a census as soon as possible after the ending of a conflict. So, I think that while census techniques and technologies are bound to change, censuses themselves are likely to be with us for some time to come.

Thanks, and good luck with the project!