The line that divided a subcontinent

Five weeks, two commissions, and consequences that have outlived everyone involved

A The line dividing British India was drawn by a lawyer who had never previously visited the subcontinent, in five weeks, using outdated maps and census returns of contested reliability. He was appointed chairman of two boundary commissions in July 1947, was instructed to complete the work before the transfer of power in mid-August, and had no cartographic or administrative experience of the territory. His unfamiliarity was presented as a qualification: a man who knew nobody could not favour anybody.

B The commissions themselves were unable to function as intended. Each comprised an equal number of Muslim and non-Muslim judges, who divided predictably on every substantive question, leaving the chairman to decide alone. The criteria he was given were contradictory: to divide the provinces on the basis of contiguous majority areas, while also taking into account "other factors" that were never specified. Any line satisfying the first requirement would leave substantial minorities on both sides, and the second gave him a discretion he had no basis for exercising.

C The completed awards were delivered before independence and published two days after it. The delay was a deliberate decision by the departing administration, taken on the reasoning that the outgoing power should not be present when the consequences arrived. Its practical effect was that for two days millions of people did not know which country they were living in, and that the administrative machinery which might have anticipated movement — police, troops, transport — was not positioned to do so. Districts celebrated independence without knowing whether they were celebrating in India or in Pakistan.

D Some idea of the arbitrariness involved can be got from the treatment of particular districts. A single district assigned to one country rather than the other gave that country a river headworks, a road, a rail junction; the criteria offered no principled way of choosing, and the recipients on each side subsequently found evidence of favouritism in every allocation that went against them. What followed was the largest and fastest movement of people in recorded history. Estimates of the numbers displaced across the new borders in the following months range between ten and eighteen million; estimates of the dead range from several hundred thousand to two million, and the range itself is a fact about the event, since no authority capable of counting existed in the districts where most of the killing happened. Both major provinces divided by the line, in the west and the east, suffered massacres, and columns of refugees moving in opposite directions passed each other on the same roads. The violence was not spontaneous everywhere. In several districts it was organised by groups that had been arming and drilling for months, and the pattern of attacks on trains in particular indicates planning rather than panic.

E Assigning responsibility has occupied historians ever since, and the arguments fall into three broad positions. The first holds the departing power principally responsible, on the grounds that it advanced the date of withdrawal by ten months, refused to deploy sufficient troops, and delegated the boundary to a man given an impossible brief and an impossible schedule. The second distributes blame among the political leaderships, whose positions had hardened to the point where partition was the only settlement any of them would accept, and who each expected the violence to be somebody else's problem. The third takes a longer view, tracing the outcome to decades of administrative practice — separate electorates, communal classification in the census, recruitment by community — which made religious identity the organising category of public life long before anyone proposed dividing the country along it.

F The three explanations are not mutually exclusive and the disagreement between their advocates is largely about weighting. What is not in dispute is that the speed made everything worse. A boundary announced six months earlier, with troops in position and time to organise movement, would still have divided communities and produced hardship; the specific catastrophe of that autumn — the columns attacked on the road, the trains arriving with nobody alive on them — was a consequence of doing in ten weeks what could not be done in ten weeks.

G The lawyer returned to England, declined his fee, and destroyed his working papers, saying that he expected to be disliked in both countries and had no wish to be quoted in either. The consequences of the line he drew have not settled. One princely state whose ruler and population differed in religion was left unresolved and remains the subject of the dispute between the two states; the eastern portion of the settlement dissolved in a further war within twenty-five years; and the demographic pattern the partition created — minorities in each country whose loyalty is periodically questioned — is a live political fact in both. It is among the clearest demonstrations available that the consequences of an administrative decision can outlast every person who took it.