Making Sense of the Census: Classifying and Counting Ethnicity in Oceania, 1965-2011

As the flagship government effort to count and classify its population, censuses are a key site for rendering and making visible group boundaries. Despite claims to objective rationality, however, census taking is a political and inherently subjective exercise. Censuses help shape the very categories they claim to capture: censuses do more than reflect social reality, they also participate in the social construction of this reality (Kertzer and Arel, 2002b, p. 2). While ethnicity – as a social construct – is imagined, its effects are far from imaginary, and census categorisations may have significant material consequences for the lives of citizens. Although an increasing number of studies have examined how and why governments in particular times or places count their populations by ethnicity, studies that are both cross-national and longitudinal are rare. Attempting to in part bridge this gap, this thesis studies census questionnaires from 1965 to 2011 for 24 countries in Oceania. In doing so, it explores three general questions: 1) how ethnicity is conceptualised and categorised in Oceanic censuses over time; 2) the relationship between ethnic counting in territories to that of their metropoles; and 3) Oceanic approaches towards multiple ethnic identities. Spread over an area of thirty million square kilometres of the Pacific Ocean, Oceania provides an interesting context to study ethnic counting. The countries and territories which make up the region present an enormous diversity in physical geography and culture, languages and social organization, size and resource endowment. As the last region in the world to decolonise, Oceania includes a mix of dependencies and sovereign states. The study finds that engagement with ethnic classification and counting is near-ubiquitous across the time period, with most countries having done so in all five cross-sectional census rounds. In general terms, in ethnic census questions ‘racial’ terminology of race and ancestry has been displaced over the focal period by ‘ethnic’ terminology of ethnicity and ethnic origin. Overall, the concept of ethnic origins predominates, although interestingly it is paired with race in the US territories, reflecting the ongoing social and political salience of race in the metropole. With respect to ethnic categories provided on census forms (and thus imbued with the legitimacy of explicit state recognition) the study finds a shift away from the imagined and flawed Melanesian/Micronesian/ Polynesian racial typology and other colonial impositions to more localised and self-identified Pacific identities. It is theorised that these shifts are emblematic of broader global changes in the impetuses for ethnic counting, from colonially-influenced ‘top down’ counting serving exclusionary ends to more inclusive, ‘bottom up’ approaches motivated by concerns for minority rights and inclusive policy-making