Does it matter if you’re black or white? #CensusSA2011

As submitted to The Daily Maverick

On October 10, be ready to stand up and be counted. Well, you’ll need to stand up at least for as long as it takes to let the enumerator into your home, for that day marks the start of South Africa’s first census since 2001. There’s no question that conducting a full census costs plenty of money – R3 billion is the current estimate – but they also provide valuable information.

Based on details regarding education levels, expectations regarding future migration trends and fertility rates, we could know what the ideal location for a new school would be. In the event of some natural disaster, we could know just how much relief needs to go where. And questions related to fertility could give demographers valuable information related to child mortality and future population momentum.

One can of course question whether gathering these and other data are worth the expense, or whether there are more efficient – and more accurate – ways of getting those data. But some criticisms of the census aren’t as legitimate as these. Census data rarely results in any significant invasion of your privacy, and the gathering of that data won’t take up much of your time.

Thanks to the millions spent on posters advising you of who your local enumerators are (again a significant expense that could attract legitimate criticism), the fear of letting strangers enter your home is also partly alleviated. But amongst all the worthwhile data, and the uses that data could be put to, there’s at least one aspect of the census that merits critical reflection: Questions relating to race.

The 2011 census will follow the convention of the 2001 census in asking: “How would (name) describe him/herself in terms of population group?” Your options are Black African, Coloured, Indian or Asian, White, and Other. In an article forthcoming in the Journal of Ethnic and Racial Studies, Tom Moultrie and Rob Dorrington of the University of Cape Town trace the history of this self-reporting formulation of the race question, which first emerged in the 1996 census.

Given that the repeal of the Population Registration Act in 1991 stripped these classifications of any legal force, it stands to reason that allowing for the self-reporting of detail regarding one’s race is the only defensible way of asking the question. If the categories permitted by the form don’t correlate with any specific legal category, you can be whatever you want to be – or at least say that you are.

It wasn’t always the case that this choice was respected. The 1996 census was the first in which enumerators worked under strict instructions to not “correct” your answer if it didn’t seem to accurately describe the respondent. And even during that census, Moultrie and Dorrington observe that reclassification occurred during data-cleaning and processing. Some South Africans had successfully campaigned to have the classification “Griqua” recognised on the census form, but the database of results has those individuals listed under “Unspecified”.

Then, in both the 2001 census and the 2007 Community Survey, little respect seems to have been afforded to self-reported detail regarding race. While the 2001 census included the option of “Other: Specify”, responses offered here were removed during editing. In 2007 the option of an open category was removed. During the editing of both the 2001 census and the 2007 Community Survey, “a complex set of rules was devised to attribute one of the four population groups to people who did not respond to the question, thus removing all non-responses to this question.”

To some extent, then, it has historically been the case that it doesn’t really matter what you say you are, as some shoe-horning into what others think you should be has been inevitable. The complications arising from this are discussed at length in the paper, and I won’t attempt to do them justice here. Suffice it to say that to the extent that data on race is valuable, both self-reporting as well as the re-interpretation of self-reporting will systematically make any such data less meaningful as a way of determining the extent to which race correlates with other significant data.

Perversely, one could suggest that the only way to properly understand the intersections between race and factors such as poverty or health would be a return to some fixed classification – as the authors put it, a return to the “reification of race and racial difference”. But as things stand, the data become less meaningful over time, while nevertheless allowing for what end up being somewhat unprincipled claims regarding those intersections.

And while one has to acknowledge that even though the question of whether one is black or white – or whatever else – is still of enormous significance to people’s perceived realities (and their experience), some of us continue to hope for a future in which race is irrelevant. One possible starting point on the road to this future is to begin rejecting these classifications – especially when their assignation seems somewhat arbitrary.

The standard categories, like those to be offered to us as options in the forthcoming census, not only compromise our ability to properly self-classify (what does “Other” mean, without a text field to specify the nature of your otherness – and is it more or less accurate than “Black” or “White”?), but also further entrench the popular acceptance of the notion that we are defined by those categories.

In the Canadian census of 2006, 18.4% of respondents reported being “Canadian” in response to a question on ethnic origin, and a further 13.82% identified themselves as Canadian while also claiming some other identity. Something similar has been happening in New Zealand for quite some time. In 1986, roughly 20 000 people self-identified themselves as “New Zealanders”, rising to over 89 000 in 2001 – although these responses were also recoded to count towards the “European” category. The 2006 New Zealand census listed “New Zealander” as a dedicated option for the first time, and roughly 430 000 people (from 4.6 million responses) selected this option, making “New Zealander” the third-largest ethnic group in the country.

Even though our upcoming census asks you to fit in with the standard racial classifications – or to be an unspecified “Other”, we can nevertheless take this opportunity to protest both the arbitrary and unprincipled ways in which this data has been interpreted, as well as protest the idea of racial classification itself.

There’s no listed option of “South African” on the form the enumerator will present to you. However, there’s nothing stopping you from either refusing to answer the ethnicity question – including trying to ensure you aren’t simply classified as “Other” – or insisting that you are captured as being “South African”. If you think being South African is more important than being black or white – or if you think that being black or white is of no importance – it’s an option worth considering.

By Jacques Rousseau

Jacques Rousseau teaches critical thinking and ethics at the University of Cape Town, South Africa, and is the founder and director of the Free Society Institute, a non-profit organisation promoting secular humanism and scientific reasoning.