Why Are So Many African Borders Straight Lines? The Scramble for Africa and Its Long Shadow
Short answer: because many of them were drawn by European governments in the late nineteenth century, in European capitals, at a time when Europeans had barely settled in Africa. During the "Scramble for Africa", which began with the Berlin Conference of 1884–1885 and was largely complete by about 1900, European powers divided the continent into colonies, protectorates and spheres of influence. The lines frequently cut through the homelands of existing peoples, and the colonial borders outlived the colonial era. Economists have since measured what that partition has meant for conflict and public services.
This article explains how the borders were drawn, why they stayed, and what two well-known studies found about their effects, including where the evidence is limited.
How the lines were drawn
Economists Stelios Michalopoulos and Elias Papaioannou, in a study first released as a National Bureau of Economic Research (NBER) working paper in 2011 and later published in the American Economic Review, summarise the historical record. The event that came to stand for the partition of Africa was the conference that Otto von Bismarck organised in Berlin from November 1884 to February 1885.
The conference itself formally dealt with the boundaries of Central Africa (the Congo Free State). Its wider importance, the authors explain, is that it set out the principles Europeans would use to divide the rest of the continent. The key consideration was to preserve the status quo and prevent conflict among the European powers themselves, at a time when memories of Europe's wars of the eighteenth and nineteenth centuries were still alive.
Three features of the process stand out in their account:
- African rulers had no say. They quote the historian A. I. Asiwaju: the Berlin conference "was essentially a European affair: there was no African representation, and African concerns were, if they mattered at all, completely marginal" to the interests of the negotiating powers.
- Little local knowledge. The borders were designed in European capitals at a time when Europeans had barely settled in Africa and had limited knowledge of local conditions. In many cases, the authors write, European leaders did not wait for information from explorers, geographers and missionaries.
- Bargaining between powers. Territory was traded between European states. The study gives the example of Katanga, which became the richest province of the Congo Free State: King Leopold obtained it in exchange for the Niari-Kwilu area that France insisted on keeping.
Why straight lines?
Alberto Alesina, William Easterly and Janina Matuszeski made straightness a measure in their paper "Artificial States" (NBER working paper 2006; published in the Journal of the European Economic Association in 2011). They proposed two measures of how artificial a country's borders are, computed for every country in the world:
- how much borders split ethnic groups into two separate, neighbouring countries; and
- how straight a country's land borders are, on the assumption that straight land borders are more likely to be artificial rather than following natural features or long-standing boundaries between peoples.
They reported that both measures appear highly correlated with several measures of political and economic success: countries with more artificial borders tended to do worse.
Who was split
The Michalopoulos and Papaioannou paper gathers earlier estimates of how widespread partition was:
| Estimate | Figure | Source cited in the paper |
|---|---|---|
| Partitioned ethnic groups identified | 177 | Asiwaju (1985) |
| Average share of population in partitioned groups | 40% | Englebert, Tarango and Carter (2002) |
| Countries where the split-group share exceeds 80% | Several | Alesina, Easterly and Matuszeski (2011) |
Source: Michalopoulos and Papaioannou, "The Long-Run Effects of the Scramble for Africa," NBER Working Paper 17620 (2011), introduction and footnote 1.
A partitioned group is one whose historical homeland falls on both sides of a modern border, so that members of the same people live in two or more countries.
Were the borders really arbitrary?
Historians have long argued that they were. Michalopoulos and Papaioannou tested the claim statistically by comparing the historical homelands of groups that were split with those that were not.
Apart from two features, the land area of the homeland and the presence of lakes or other water bodies, they found no significant differences across a wide range of geographic and ecological characteristics. Nor did they find systematic differences in pre-colonial features such as settlement size, type of economy, or proxies for pre-colonial conflict. In other words, which peoples ended up split looks largely like chance with respect to their characteristics, which is what one would expect if the lines were drawn with little regard for local political geography. The authors present this as support for the long-standing view in African historiography, "at least with respect to ethnic partitioning".
What partition has meant: the measured effects
Because the split was close to random, the authors treat it as a quasi-natural experiment and compare outcomes in split and non-split homelands. Using georeferenced data on political violence for 1997 to 2013, they report:
- More conflict. Conflict intensity is about 40% higher, and the likelihood of conflict about 8% higher, in areas where partitioned groups live, compared with homelands of groups not divided by borders. The results are similar when they look only at homelands near national borders.
- Interventions from neighbours. Military interventions from adjacent countries are more common in the homelands of partitioned groups, in line with historians' argument that split groups are used by neighbouring states to stage proxy wars.
- Political exclusion. The likelihood that split groups face political discrimination from the national government is about 7 percentage points higher than for non-split groups, and they are more likely to take part in ethnic civil wars.
- Fewer public goods. Using household surveys covering more than 85,000 households in 20 African countries, people identifying with split groups have lower access to public goods and worse educational outcomes.
The pattern is specific. Partition matters most for fighting between government forces and rebel groups, and less for violence against civilians. The study finds no link between partition and riots or protests, which it describes as mainly a capital-city phenomenon, and no link with conflict between non-state groups.
Why the borders survived independence
The Michalopoulos and Papaioannou paper quotes the political scientist Jeffrey Herbst on why colonial borders mattered so much: at independence, "for the first time in Africa's history, territorial boundaries acquired salience… The boundaries were, in many ways, the most consequential part of the colonial state." The paper notes that the colonial borders outlived the colonial era, so that in many African countries today a significant fraction of the population belongs to groups split between states.
What the evidence does not show
- Not the only cause. The studies measure differences between split and non-split areas. They do not claim that borders alone explain conflict or poverty in any country.
- Averages, not destinies. The figures are averages across many groups and places; individual cases vary.
- Versions differ. The figures quoted here are from the 2011 working paper of Michalopoulos and Papaioannou. The final version, published in the American Economic Review in 2016, went through peer review, and readers wanting exact final estimates should consult it.
- Straightness is a proxy. Alesina and colleagues assume that straight borders are more likely to be artificial. That is a measurement choice, not proof about any single border.
Common questions
Did the Berlin Conference draw all of Africa's borders?
No. According to Michalopoulos and Papaioannou, the conference formally dealt with the boundaries of Central Africa, the Congo Free State. It came to symbolise partition because it laid down the principles Europeans then used to divide the continent over the following years.
Were any Africans present at Berlin?
No. The paper quotes the historian Asiwaju, who describes the conference as essentially a European affair with no African representation.
How many ethnic groups were divided?
Asiwaju (1985) identified 177 partitioned ethnic groups, as cited in the Michalopoulos and Papaioannou study.
Are straight borders always a problem?
Not necessarily. Alesina, Easterly and Matuszeski use straightness as a sign that a border may be artificial and report that, on average, more artificial borders go with weaker political and economic outcomes. That is a cross-country association, not a rule for every border.
Why weren't the borders redrawn after independence?
The studies cited here document that colonial borders outlived the colonial era and still shape where groups are split today. Their focus is the consequences of the partition rather than the decisions taken at independence.
Sources and how to check
- Michalopoulos S., Papaioannou E. "The Long-Run Effects of the Scramble for Africa." NBER Working Paper 17620 (2011). DOI: 10.3386/w17620 · nber.org/papers/w17620. Published version: American Economic Review 106(7), 1802–1848 (2016).
- Alesina A., Easterly W., Matuszeski J. "Artificial States." NBER Working Paper 12328 (2006). DOI: 10.3386/w12328 · nber.org/papers/w12328. Published version: Journal of the European Economic Association 9(2), 246–277 (2011).
- Works quoted within Michalopoulos and Papaioannou and cited there as Asiwaju (1985), Herbst (2000), Wesseling (1996) and Englebert, Tarango and Carter (2002).
- Paper pages and working-paper text checked on 24 October 2026.