Twenty Years to Undo Four Centuries

In 1945 the European colonial empires covered most of Africa and much of Asia. By 1965 most of that was gone. It is one of the fastest transformations of the political map in history, and it happened to empires that had been, on paper, stronger than ever thirty years earlier.

Three things did it, and the war did all three.

The first was the destruction of a presumption. When Singapore surrendered to a smaller Japanese force in February 1942, the effect on the populations being governed by European powers was not primarily military. It was the visible demonstration that the presumption of invincibility, on which colonial rule with tiny European staffs actually rested, was false. Rule by a handful of administrators over millions works only while nobody seriously tests it, and Japan tested it across Southeast Asia and won.

The second was money. Britain ended the war effectively bankrupt, dependent on American loans, and rationing bread in 1946 in a way it had not during the war. France had been occupied. Neither could afford a long colonial campaign, and both discovered this by attempting one.

The third was the argument. Having fought a war justified in terms of freedom, self-determination and the rights of small nations (with the Atlantic Charter saying so explicitly) and having recruited millions of colonial troops to fight it, the case for continued colonial rule became extremely difficult to state out loud. Indian soldiers who fought in North Africa and Italy came home to a government explaining why self-determination was for Europeans.

But independence was not one process, and the variation is where the interesting history is. Some transitions were negotiated and largely peaceful; others produced wars lasting a decade and killing hundreds of thousands.

The best single predictor is the presence of a substantial settler population.

Where Europeans had come in numbers and taken the land (Algeria with around a million pieds-noirs, Kenya's White Highlands, Southern Rhodesia, South Africa) decolonisation threatened people who had nowhere to go back to and who had disproportionate political weight in the metropole. Those are the cases that produced the Algerian war, the Mau Mau emergency and its detention camps, and a Rhodesian rebellion against Britain to prevent majority rule.

Where the colonial presence was administrative and commercial rather than residential, a handover to a local elite was possible without anyone's home being at stake, and it generally happened with more speed and less blood.

The new states then faced a question that they answered, almost unanimously, in a way that has been criticised ever since and was probably right.

The borders were arbitrary. They had been drawn in European capitals by people who in some cases had never visited the territories, following rivers and lines of latitude, cutting through peoples and lumping others together. Every argument for redrawing them was strong in the particular case.

And in 1964 the Organisation of African Unity adopted the principle of uti possidetis: the colonial borders stand. The reasoning was that any process of revision requires agreeing whose claims count, in a continent where hundreds of groups had claims and many were incompatible, and that such a process would mean war everywhere at once rather than in a few places. Biafra and Katanga were the test cases, and the precedent held.

It is a genuinely hard trade-off, honestly made: bad borders judged less dangerous than the process of fixing them.

What independence did not deliver, in most cases, was economic independence. Trade flowed to the same buyers, mines and plantations kept the same owners, several states remained in currency unions pegged to the franc, and the technical and administrative staff were trained in the same places. Nkrumah's word for this was neo-colonialism, and whatever one makes of the framing as a theory, the empirical claim about continuity of economic structure after a political handover is well supported.

And the state that was handed over had been built for a particular purpose. Colonial administrations were designed to be cheap: rule through chiefs, authority concentrated in the executive, courts subordinate, emergency powers permanent, and almost no investment in education. The Belgian Congo at independence had a very small number of university graduates for a population in the tens of millions. Inheriting a machine built to extract and control, and being asked to run it for development, is an extremely demanding transition.

It also carries a specific political risk, which materialised almost everywhere. Whoever holds that apparatus on independence day holds the police, the emergency powers, the censorship and the patronage over every job and contract. A party in that position has enormous advantages over any challenger, and the temptation to treat opposition as subversion is considerable. One-party states appeared across Africa and Asia within a decade of independence, and this is a substantial part of the mechanism, not a fact about the people who led them, but about what they were given the keys to.