To analyze why partitioning has a bad name I did some study of the partition of India in 1947. In that event some 12.5 million people were displaced. Estimates of the number of death vary from a few hundred thousand to a million. No one had expected it: everyone expected just another peaceful change of an administrative border. Since then partitioning has a bad name. I think this is not really justified as there were other factors that made the partition of India so deadly.
Relations between Muslims and Hindus had become worse in the previous decades. Some like to blame British policies that made a distinction between the groups, such as the 1905 partition of Bengal and reserved seats in elections. Others blame the Muslims for demanding secession and threatening with violence. But I think the real reasons are elsewhere.
The partition of Bengal in 1905 deserves some special consideration as it was the first time Muslims and Hindu's stood on opposite sides of an issue. Bengal at that time was a very big province that included Assam and parts of Orissa and contained a quarter of India's population. The colonial administration considered it too big and wanted it partitioned. But instead of following linguistic lines - what before and after that was the usual habit in India - he cut the province in a Eastern and a Western half. In the Eastern half the Muslims had the majority (about 60%). The official explanation was that there were natural barriers between the East and the West side and few connections and so it made sense to place the border there. But many suspected (and there is good reason to believe) that the British also wanted to diminish the power of the Bengals who had become a major opposition against the colonial rule. The motor of the opposition were Hindu landlords who didn't want to live under Muslim rule. The Congress Party organized major opposition against the "division of the Bengal nation" and it is considered one of the highlights in Indian resistance against British rule. In 1911 the partition was withdrawn and a partition along linguistic lines was introduced instead.
In the decades before the war the British had a policy of gradually handing more power to the Indians. Initially this was to rather small elite groups. This basis was gradually widened but even in the last elections before independence only 10% of the population was allowed to vote. Universal suffrage only arrived after independence.
One consequence of this policy was that each change might affect the power distribution in a region. Where one day the Muslims controlled the local government the next day the Hindus might receive control - and opposite. Such power changes offered the opportunity to set things your way and to retaliate for past injustices. But the lack of stability made it difficult for people to find permanent solutions.
Another factor was that the British allowed local governments to appoint the people on the boards that settled tenancy disputes. This mixed up justice and politics and the effect was predictable: in Muslim ruled areas Hindu landlords had a hard time to collect rents from Muslim tenants (and opposite). As it are usually the rich people who finance the political parties this had a major effect on politics.
A third factor were the youth movements. In the 1930s these were very popular worldwide. Boy Scouts, Komsomol, Hitler Jugend, Young Socialists and others offered all a culture of discipline, physical fitness and exploration of the world while teaching their values. Muslim and Hindu organizations started similar youth movements. Soon this developed into "self-defense" groups and militia. The British didn't do anything to stop this as they didn't see it as a threat to their rule.
The next factor was the war (World War II). The Congress - the main political party - tried to exploit the war to force the British to concessions. As a consequence the main Congress leaders were imprisoned. This allowed smaller - and often more extremist - political parties to flourish during the war.
Over two million Indians had fought on the side of the British in World War II. Now that they came back many lent their expertise to the militia or joined them.
The next factor was the lack of clarity. The call for an independent Pakistan only arose in the 1930s and it was unclear what it meant. Many Muslims expected that a much larger part of India would go to Pakistan. Some even believed that Pakistan would mean Muslim rule over all of India. As a consequence there were very few Muslims who worried what would happen to them if they would end up at the wrong side of the border after a partition. The British or the Congress Party could have created more clarity but they didn't. Muslim leaders deliberately kept it unclear and created an atmosphere where you were traitor if as a Muslim you were against the creation of a separate Muslim state.
For a long time there were only occasionally clashes between village militias. These were usually about religious issues: building permits for mosques, the right of (noisy) Hindu processions to pass mosques where people were praying, Muslims slaughtering (sacred) cows in the village, etc. Occasionally someone died in those clashes but these stayed local issues.
Violence reached a new level with the Calcutta Killings in 1946. Jennah, the Muslim leader, had called for a National Action Day in their struggle for Pakistan. The local Muslim leaders interpreted it so that they sent a large mob of Muslims pilfering and murdering Hindus. Police didn't do anything (Bengal was ruled by Muslims) and as a retaliation the Hindu militias immediate held a counterpogrom among the city's Muslims. Some 5000 people died that day. Local governments often supported "their" people and the British resorted to collective punishment: a fine for all Muslims or Hindus in a region. This only helped to unite the local population behind the extremists.
One effect of the violence was rumors. Just like in Yugoslavia rumors about violence by the others - and its use as propaganda - often served to justify violence or extremist actions. Information was hard to find and in the information that reached the people was biased and usually painted their own group as a victim.
To understand the British one has to consider the era. They knew independence was coming and as a consequence the British government was making little new investment in the colony. It meant also that there were few new recruits for the colonial service and that many colonial officials just bided their time until it was over. It didn't help that for many the last time they had been to England was before the war. On the other side the Indians became less obedient as they had nothing to fear from the English administrators who would be gone in a couple of months. Unfortunately this diminishing of British power was not counterbalanced by an increasing power of Indian institutions, so there was a power vacuum. In fact many Indian institutions were falling apart as they underwent their own partition.
The British announced the decision for a partition on 3 june 1947. 2 1/2 month later, on 15 august, India became officially independent. So there was a very short time to arrange the partition. The definitive border was only decided on 12 august and only published on 17 august. All borders had been determined by two British commissions within 6 weeks so they didn't have the time to visit the border regions themselves. The inevitable inconsistencies would never be corrected. Rumors claimed - as did some politicians - that those staying behind might lose their political rights or worse (in fact there were decent minority rights). Some politicians encouraged "their" people to flee from areas dominated by the other group.
The government of the new Pakistan decided to settle their new capital in Karachi. This was the biggest city and a center of the British administration. But it was in the province Sindh that had its own character. As a capital it would attract many refugees of other ethnic groups in the population exchange and that gives up to the present day trouble. It would have been a much better idea to have the new capital in the Punjab - as many refugees came from the eastern Punjab.
Partition also concerned the institutions. The army was slowly split. Government employees got the choice whether they wanted to stay where they were or to go to the area of their group. As the situation escalated police and army found it increasingly difficult to stay neutral.
For a long time both new countries more or less ignored the refugees. As a consequence as long as they were on "wrong" territory they were frequently robbed or even murdered. Even trains with refugees were regularly ambushed. At the end of august 1947 the governments finally took a more active position in regulating the refugee stream. But that created its own problems as some local officials took this as a license to tell their minority community that they had to leave.
Partition meant literally partition for the provinces Punjab and Bengal. These were also the provinces where most of the violence happened. Between East and West Punjab there was a nearly complete population exchange (remaining populations < 2%). In Bengal the cleansing was less complete: Bangla Desh still has 9% Hindus and West Bengal has 25% Muslims.
Could it have been different?
There has been a lot of speculation about how things could have been different. Much blame has gone to the Muslims for their demands and the extremist language and means they used to achieve it. Blame has also heaped on Hindu extremist organizations like the RSS for their violent militias. But I like to take the extremists for granted and to look how the British or the Congress Party could have reacted differently.
Would have helped not to partition Punjab and Bengal in which case they would have gone as a whole to Pakistan? I doubt it. It might have worked. But it might also have resulted in the cleansing of all the Hindus in the whole Punjab and resulting revenge cleansings of Muslims all over India. Hindus might also have reacted by not accepting the solution in which case you would have had open war over an undetermined border.
Maintaining provincial borders was for some time popular after the dissolution of the Soviet Union where it seemed to lead to a peaceful transition. But it didn't work in Yugoslavia. And the gradual cleansing of the Russians from the Central-Asian republics, the large scale cleansing in the Caucasus and the recent events in Kyrgyzstan suggest that it didn't work in the Soviet Union as well. In fact it is hard to find ethnic relations that have not deteriorated due to the dissolution of the Soviet Union. The initial quietness was just a consequence of the surprise effect. Georgia's cleansing of over a third of its minorities before it ran into real resistance in South Ossetia is a good example: it hardly reached the international news.
On the other hand the partition of India shows that just having a shotgun agreement is not enough. The Congress Party, the Muslim League and the British had all agreed to the partition plan and it had been agreed that there would be decent minority rights. But those minority rights had been badly communicated and the dynamics of violent militias creating "facts on the ground" proved hard to stop.
In my opinion it is the suddenness of the change that causes most problems. At the side of the "winners" it brings nationalist extremists in power who deal harshly with the new minorities. At the side of the "losers" it creates distrust and suspicion. According to this vision the best way to deal with ethnic conflicts is to make changes gradual.
Another issue is militia's. This was the era of fascism where radical parties had militias. This applied to both the Muslim League and the RSS. I believe militia's are very harmful on the political process. For example I never understood the lenient attitude of the Americans towards Al Sadr in Iraq.
In other ways too the strategies of Jennah were classical fascist. The way he managed to be seen as the sole representative of India's Muslims and the way he sidelined all other Muslim voices were typical.