On 9 April 2015, a movement of student activists at the University of Cape Town called “Rhodes Must Fall” won a decisive victory with the removal of Cecil John Rhodes’ statue from their campus.

This spurred renewed interrogations of Rhodes’ legacy both in South Africa and beyond. Oriel College at the University of Oxford had to revisit this question when faced with similar calls for the removal of a statue of its former student and benefactor, and the Rhodes Trust, home of the international Rhodes Scholarship, has likewise engaged in constructive retrospection.

One of Rhodes’ less-understood legacies was his use, at the end of the 19th century, of ostensibly legal voter suppression techniques to disenfranchise black and coloured voters whose right to vote was legally protected by the Cape Constitution of 1853.

Voter suppression techniques are now widely used in modern democracies by opportunistic politicians seeking to disenfranchise certain segments of the electorate — often but not always black voters. 

Rhodes was an “early adopter” of these tools. Beginning in 1854 with the first election of the Cape Parliament, the Cape Qualified Franchise allowed all men, regardless of race, aged 21 or older to vote if they occupied property of a certain value, or earned a salary of a certain value.

For almost a century the Cape’s non-racial franchise was a point of sustained conflict for racist and opportunistic politicians. In a newly released working paper, we provide the first full account of the extent of racial disenfranchisement that occurred specifically under Rhodes’ leadership. 

Acts of destruction 

While Rhodes was not the only politician engaged in voter suppression — reforms passed in 1887 under Prime Minister Gordon Sprigg were perhaps the first clear efforts — his legacy looms large.

Following the eastward expansion of the Colony into the modern-day Eastern Cape, Rhodes, as Prime Minister, was responsible for drafting and passing the 1892 Cape Franchise and Ballot Act which increased the occupancy qualification and introduced a literacy test. In 1894 he conceived the Glen Grey Act which devised a spatially targeted pattern of landholding for the black population that would exclude certain types of property from the occupancy qualification. 

The 1892 Act leveraged the racial logic of the Cape’s economy to deliver racial disenfranchisement: because black and coloured people were systematically paid less than white people, they rarely qualified on the basis of salary, relying instead on the occupancy qualification. By raising the occupancy qualification (and introducing a literacy test), black and coloured voters were more likely to be denied the vote. 

The 1894 act, which Rhodes viewed as a “native act for Africa”, leveraged the spatial geography of different race groups to deliver disenfranchisement. This became the backbone of the 1913 Land Act, and the various efforts to create and codify the Bantustans.

These acts were deliberately intended to suppress specifically black voters. Socioeconomic and spatial levers became common tools for successive Cape and South African governments, until the apartheid government finally extinguished voting rights for all but white people in the mid-1950s. 

Despite the importance of the 1892 and 1894 laws in South Africa’s history, little is known about their precise impact on the electorate. In collaboration with the Laboratory for the Economics of Africa’s Past (Leap), we digitised the entire 1903 voter roll of the Cape of Good Hope, comprising 135,457 individuals who were registered to vote in that year.

This vintage of the roll was the first that included every voters’ official “race distinction”. It also provides information on who was already registered in 1891, prior to the reforms initiated by Rhodes.

These and other unique features allow us to produce a detailed data-driven assessment of the composition of the electorate prior to and after the passage of this legislation. Our methodology — which we set out in detail in our working paper — calculates how many additional voters of colour would have had to be on the roll in 1903 for them to have the same share of the electorate that year as in 1891.

We show that Rhodes’ franchise reforms increased political inequality between population groups in favour of white men. The magnitude of these effects is striking — we estimate that, all told, somewhere in the region of 10,000-15,000 mostly black and mixed-race voters were disenfranchised between 1891 and 1903.

Without suppression, the entire electorate in 1903 would have been roughly 7.5 to 11.5% larger, with the number of voters of colour roughly 50 to 75% higher.

The 1892 Act accounted for the majority of disenfranchisement over this period because it affected the entirety of the Cape. By contrast, the 1894 Act, which was geographically targeted at the modern-day Eastern Cape, was extraordinarily effective at disenfranchising voters, but its effects were more narrowly focused. 

Far-reaching impact

The broader impact of these reforms on South Africa’s political history is hard to overstate. We show that in the 1904 election (for which the 1903 roll was prepared) the number of disenfranchised voters of colour was likely large enough to have been pivotal in almost 42% of the electoral races that year.

By potentially altering the outcome of the 1904 election, these laws strengthened the electoral prospects of nationalist forces. More broadly, the reforms laid the groundwork for the formal establishment of a whites-only franchise over the course of the 20th century. 

At the same time, the laws contributed to growing political activism among people of colour. The early experience of political participation and of targeted disenfranchisement contributed to black political mobilisation in a unified South Africa.

Indeed, three founders and inaugural principal office holders of the South African Native National Congress (what would become the ANC) — John Dube (its President), Walter Rubusana (Deputy President) and Sol Plaatje (Secretary-General) — can be found on the 1903 roll.

The struggle over the right to vote in this period triggered a response that has shaped the history and politics of modern South Africa. 

Sadly, voter suppression efforts — for instance through voter ID requirements — remain ongoing in several countries today. We hope our work demonstrates the harm such manipulation can do to even constitutionally protected voting rights. DM 

Daniel de Kadt is Assistant Professor of Quantitative Research Methods and Joachim Wehner is Associate Professor in Public Policy at the London School of Economics and Political Science. Their full analysis is available here.