Karachi is in Madras and Lahore is in Bengal; when will they be in Pakistan, and Pakistan in Southasia? During the late 1840s, the troops of the Bengal Presidency invaded the Sikh empire and conquered Punjab. Earlier in the same decade, a British expeditionary force had fought the Talpur Mirs at Miani, and annexed Sindh to the Bombay Presidency. But colonisation was not merely a matter of military defeat. Punjab and Sindh had seen many defeats and victories in their times. It was the establishment of a political system around the 'settlement' of land revenues that marked the real shift. By then, the landlord-based Permanent Settlement in Bengal was already a half-century old, and in disrepute among British colonial officers. The Madras Presidency experimented with the cultivator-based ryotwari system, which was eventually adopted by Bombay and implemented in Sindh. The Bengal Presidency drew a different lesson from the failure of the Permanent Settlement, and in its later conquests settled land revenues with 'village estates', on a mahalwari basis. Hence, Karachi is in Madras and Lahore is in Bengal.
This is not just a 'checking whether you're awake' historical vignette. Although much has changed, much has also stayed the same. The village records of the late Bengal Presidency today remain the administrative base of Punjab, the NWFP (or Pashtunkhwa) and Balochistan. The NWFP was, after all, carved out of Punjab in 1901, and Balochistan was infiltrated and annexed by colonial officials from Punjab province. The basic administrative unit in these provinces is the mouza, and the lowest government functionary is the patwari. Sindh has the nomenclature of Bombay, with its deh and tappaydar respectively. In essence, Sindh and Punjab (and Balochistan and Pashtunkhwa) are still in presidencies of the British Indian empire.
This retention of colonial structures has enormous ramifications. The Punjab village might have been a useful unit of land-revenue administration during that time. But it was also the repository of caste-based social hierarchy, which the British formalised. Revenue administration continues to classify people into 'cultivator' and 'non-cultivator' castes – the former even if they have never touched the soil are the gentry, the latter even if all they know is farming are the menials. The concept of 'village proprietors' – who might be joint owners of uncultivated land in the village estate, as well as of the residential land within the settlement – has survived successive waves of partition, land reform, political enfranchisement, Islam and devolution.
The government-school admission form in the Punjab village still insists not only on recording the caste of a child, but also whether the student is a 'cultivator'. It is as though the Bengal Presidency village in Punjab is to this day simply a device for humiliating people, and keeping them in their places. In Sindh, the village does not know where the people are. Rural folk live in small hamlets called goths, and the goth has its own solidarity based on kinship relations. The administrative village, or the Bombay deh, is an arbitrary agglomeration, one that has no identity save in the land-revenue records. There can be twenty large and small goths in a deh, each demanding attention and asserting itself as an authentic inhabitation whose needs must be met. The revenue-based system once tried to generate lists of these settlements, but soon gave up because it was not geared for the task. It is nightmarish to plan social-policy interventions when the state machinery has no coherent way of knowing where the people actually are.