by John Keay
The mutual suspicion was made worse by the terms of Radcliffe’s actual award. Dividing erstwhile India into its Muslim- and non-Muslim-majority provinces had been comparatively straightforward, but the lesser territorial units to be parcelled out when dividing up the Punjab and Bengal posed a trickier challenge. These lesser elements had been specified merely as ‘areas’; they might be districts, sub-districts or even smaller units. And although the twin principles of partition – division on the basis of the religious majority plus contiguity to ‘areas’ of a like complexion – were generally paramount, ‘other factors’ (like local traditions, irrigation networks and strategic necessity) might be taken into account. There was thus scope for exceptions, and still greater scope for suspicions about exceptions. Well-founded rumours would circulate that Radcliffe had indeed been ‘influenced’. India’s expectations in respect of the Punjab border, especially where it afforded an access route to Kashmir, seem to have found favour with him. So did Indian demands in respect of a northern corridor, or ‘chicken-neck’, linking West Bengal and Assam; concessions to Pakistan in the Chittagong and Khulna areas of East Bengal were supposedly made in return.
In Karachi and New Delhi these matters were warmly debated. But to the toiling masses for whom the border’s various ‘corridors’, ‘salients’, ‘irrigation headworks’ and ‘enclaves’ were home – and had been since time immemorial – the announcement of the new border was positively incendiary. Being ‘awarded’ to what was considered a hostile state, or excluded from what was considered a supportive one, amounted to an existential threat. As Indian Muslims seeking Pakistan, and Pakistani Sikhs and Hindus fleeing from it, began pouring across the border, Punjabis on either side of the delimited but still undemarcated line were swept along by the tide.
Whole villages, clans, sub-castes and kinship groups upped sticks, sometimes literally as they detached the roof joists of their homes to cart them away in the hope of re-using them. In early September, Penderel Moon, back in Bahawalpur after curtailing his holiday, recorded the arrival there of a dishevelled and unwashed gentleman called Bagh Ali. ‘He arrived on foot … along with 5,000 members of the Sakhera tribe, many of whom were his tenants’; after a week on the road ‘one could hardly imagine that he was a wealthy Muslim landowner and a MLA [Member of the Punjab Legislative Assembly]’, recalled Moon. Bagh Ali and his people hailed from Ferozepur, a place expected to go to Pakistan but which had in fact been awarded to India. But what most distressed Moon was the news that this throng, along with their bullocks, carts and farm implements, had been officially ordered to migrate. It was not the feared Sikh paramilitaries who had forced them out, but a government directive from Ferozepur’s Sub-Divisional Officer. Unbeknown to Moon, Delhi and Karachi had just agreed on an exchange of population between the two halves of the partitioned Punjab. The arrangement was intended to reduce the violence, which both governments roundly condemned. But forcible migration was a different matter. In the Punjab it was state-sponsored.
Over the long border between western India and the western wing of Pakistan some ten to twenty million people are thought to have crossed, some going east, others west, during the months of August, September and October. Additionally, anything between 200,000 and one million were massacred – in their homes, in their fields, on the road, in the trains – or left to die by the wayside. In a sandy tract near Fazilpur in Bahawalpur, Penderel Moon spied what he thought were some piles of manure. Closer inspection revealed them as heaps of bodies.
In two and threes and sixes and tens, more and more came into view as we rounded the curve of the village … till they lay ‘Thick as autumnal leaves that strew the vale of Vallambrosa’. Men, women and children, there they all were jumbled up together, their arms and legs akimbo in all sorts of attitudes and postures, some of them so life-like that one could hardly believe that they were really dead.16
It reminded Moon of pictures he remembered from his childhood of the Napoleonic battlefields. Three hundred and fifty Hindus had been mown down by Pathan rifle fire in this one incident.
Hundreds of thousands more were plundered of their chattels, a term that was taken to include their womenfolk and children. Girls and young mothers were perceived as embodiments of all that the other community held most sacred and were picked off accordingly. Abducted, exposed, traded, raped, mutilated or forcibly appropriated, most would never know justice and many would prefer suicide. Those who would later be ‘recovered’ and repatriated fared little better. Dishonoured, they might find themselves unwanted by their former loved ones; traumatised or not, they might be locked away by them.
The horror lay as much in the obscenity of the atrocities as the scale; and to these atrocities, as to all the other massacres and burnings, there was often a pattern. Though characterised as ‘lunacy’, the mayhem was a madness with method. On both sides the perpetrators were invariably male, well armed and often ex-soldiers or paramilitaries. Incitement came in the form of pamphlets, partisan press reports and pronouncements from political and religious leaders; premeditation was evident in both the planning and the execution of the attacks; and guns as much as knives were the weapons of choice.
This was not haphazard, frantic killing but, at its worst, routine, timetabled and systematic ethnic cleansing. Large groups of men, with their own codes of honour and often with a sense of warlike righteousness, set out day after day in August and September to eliminate the other.17
Of the few things that disqualified the conflict as ‘war’, the near absence of battles was the most obvious; for the aggressors, instead of engaging one another – something which respect for the border largely precluded – directed their attacks exclusively at the innocent and the defenceless. Conversion was occasionally an option for the victims, mere surrender rarely so. For the assailants, the objectives were simply expropriation and maximum slaughter.
Most refugees travelled on foot, with or without livestock and sometimes accompanied by wagons bearing their possessions. The caravans stretched as far as the eye could see where they converged at river crossings. An airborne Nehru following the line of a cross-border road in east Punjab would recall overflying the same massed column for all of sixteen kilometres. He put its human component at over 100,000 souls. Another caravan, tracked in west Punjab, was thought to number 400,000. In September Penderel Moon recorded an influx into Bahawalpur of 40,000–50,000 Muslims from Rohtak and Hissar (west of Delhi); they were so severely undernourished that ‘some two thousand of them died within a few days of their arrival’.18 As late as November an official from the British High Commission in Delhi, while driving through Mewat, encountered a ten-mile column of Meos still on the move.19
Exposure, debilitation, dehydration, starvation, disease and drownings (the monsoon had returned with a fury in September) may have claimed as many fatalities as the knife and the bullet. Yet the subsequent figures would seldom distinguish deaths from natural causes, nor would they attempt to define what causes might be considered ‘natural’. All that can be said with confidence is that the scale of the tragedy was such as to frustrate accurate assessment at the time – and ever since.
‘Estimates of casualties are largely a matter of guesswork,’ noted Moon, who nevertheless gave his own calculation of the number killed: it was ‘unlikely to have been more than two hundred thousand’, and was probably rather less. This was based on ‘fairly precise figures for about half the districts of West Punjab and … intelligent guesses regarding the remainder’; in this total, on the basis of reports from across the border, he had included twice as many fatalities for India’s east Punjab, plus much fewer for the neighbouring states of Rajasthan, Sind and Balochistan.20 Moon was writing only of the border between India and West Pakistan; he did not include fatalities in Bengal or elsewhere in India, nor apparently those in the princely state of Jammu and Kashmir. But his ‘guesswork’ deserves some respect. As a one-time member of the British Indian Civil Service, a current member of the Pakistan Adm
inistrative Service (Bahawalpur formally joined Pakistan in early October) and a soon-to-be member of the Indian Administrative Service, he straddled the divide and had no particular axe to grind.
Nor, having witnessed some of the attacks and collected descriptions of many more – indeed having accepted responsibility for not having prevented some of them – could Moon be accused of generalising about them. It has been suggested that any consensus around higher estimates of half a million to a million, or even two million, fatalities may be a means of ‘distancing ourselves from the specificity and details of those killings even as we seek to underline their enormity and consequence’.21 This recourse to rounded-up figures is thought to be especially common practice in respect of the atrocities suffered by those classed as ‘others’ rather than ‘ours’; ‘their’ losses could be approximately quantified, ‘ours’ tended to be recorded in gruesome and specific detail.
Into this error falls the account produced by Gopal Das Khosla in 1989. An avuncular figure, Cambridge-educated, Justice Khosla was much respected in Indian government circles as ‘a safe pair of hands’, and would head several government-sponsored investigations. By the 1980s he was semi-retired and often in Manali (Himachal Pradesh), there with walking stick to pace the hill paths and write his Stern Reckoning. Using the records of a 1948 Government of India ‘Fact Finding Organisation’, he came up with a total for non-Muslim fatalities of ‘between 200,000 and 250,000’, to which he ‘believed’ that an equal number of Muslims who ‘perished in the riots in India’ might be added. Hence the ‘half a million’, a figure which more than doubled that given by Moon. Khosla further ignored Moon’s careful calculation of the fatalities in Pakistan’s west Punjab (Moon had given 60,000 instead of Khosla’s 200,000–250,000); and he contradicted Moon’s contention that the killings in India’s east Punjab might be twice as many. Yet by combining these two assessments – Khosla’s 200,000–250,000 in the west and Moon’s ‘twice as many’ in the east – the total could be, and was, further conflated to three-quarters of a million.
To substantiate his findings, Khosla compiled a tabulated appendix listing over five hundred places where mass killings, conversions and conflagrations had taken place. Each entry included a note on the nature of the atrocities (‘Murder, arson, mass conversion and loot’, ‘Murder, rape, loot and abduction’ etc.) together with an estimate of the numbers killed, injured, forcibly converted or expelled. Yet on examination, all his listed incidents occurred in Pakistan, the victims being Sikhs and Hindus, as were Khosla’s informants. Of the Muslims who died in the massacres in the new India – or ‘the riots’ as he preferred to call them in this case – there is no listing at all. Nor does it appear that the figures given for any of the listed incidents were corroborated by Pakistani witnesses. Yet this was crucial, as a relief worker at the time discovered. In the Sialkot district of Pakistan, Richard Symonds was informed by the Indian Liaison Officer that in a recent assault ‘1,500 were killed’; yet ‘the Pakistan account said only thirty’. Or again, two weeks after an attack at Mianwali, ‘estimates of the number of Hindus killed varied between 400 and 2,000’.22 In the face of such flagrant misrepresentation, probably by both parties, extreme caution is in order. Without it, ‘otherising’ becomes just as partial as the blatant propaganda that has marred – indeed ‘dis-figured’ – nearly all such later calculations.23
A further explanation for the wildly divergent assessments of Partition’s casualties lies in the uncertainty over the figures for the other province to be partitioned, namely Bengal. While some calculations, Moon’s and Khosla’s for example, ignore Bengal altogether, a few go to the opposite extreme and infer a casualty rate comparable with that in the Punjab. This is absurdly pessimistic, and the ‘guesswork’ here is even more conjectural. Much depends on how ‘Partition’, a flimsy term when stripped of its more horrific associations, is defined and on what is taken to be its timeframe.
With over sixty million inhabitants, Bengal had been easily British India’s most populous province (pre-Partition Punjab had about twenty-eight million). It was also its most volatile. The potential for sectarian strife had already been demonstrated in the Calcutta Killings of August 1946 and in the subsequent massacres in Noakhali and Tripura (Tipperah). Violence like that which seems to have taken so many by surprise in the Punjab was here expected. In anticipation of it, Gandhi had already re-established himself in Noakhali, from where he transferred to Calcutta two days ahead of Independence. He needed to be at the likely epicentre when the seismic shift of 15 August occurred.
Now frailer and seemingly smaller than ever, the Mahatma was trundled round the city in an ancient Chevrolet. As he toured the trouble spots and drew massive crowds to his evening prayer meetings, his reputation transcended the religious divide. He talked up a spirit of mutual regard and inspired a sense of brotherly achievement in maintaining the peace. Mountbatten called him his ‘One Man Boundary Force’. For three critical weeks he remained there, preaching communal harmony, praying for it and fasting to exact pledges of it. He also promoted it by example, cohabiting with Husayn Shaheed Suhrawardy, the bon-viveur barrister ‘with a nimble brain but an irritating habit’ who led the Muslim League in Bengal. Lately Chief Minister of the province, it was Suhrawardy whose incitement had been widely blamed for the earlier killings.24
Notwithstanding their incompatibility, such was the influence of the two men – the stick-like Mahatma and the ‘rotund’ Muslim Leaguer – and such the military presence prompted by fear of another bloodbath, that the tactic worked. Observing the near absence of sectarian massacres in the subcontinent’s greatest metropolis, first Gandhi and then the press dubbed it ‘the miracle of Calcutta’. Optimists noted ‘a spectacle of friendship and fraternity between Hindus and Muslims’; Communists detected a comradeship born of working-class solidarity; and intellectuals rejoiced in what they took to be evidence of the Bengalis’ cultural superiority. The normally dyspeptic general who headed Eastern Command went further. ‘The love in Calcutta was impressive above all other places,’ he recalled. But he ascribed it less to Gandhi’s non-violence than to a combination of the Muslim community’s ‘depression’, the non-Muslim community’s exultation and his own increased troop levels.25
The euphoria in Calcutta lasted throughout the crisis months immediately after Independence, and dissolved only when the city reverted to its usual levels of industrial strife, social upheaval and chronic politicisation in 1948. Overall, when compared to Lahore and the Punjab, Calcutta and Bengal seemed to have got off lightly. The death toll could almost be described as bearable, while the atrocities were largely localised. On the other hand, the population transfer was here more destabilising than in the Punjab, much more protracted and ultimately perhaps greater.
Dispersal being a lesser evil than death, this raises the question of why the Partition experience in Bengal differed so from that in the Punjab, and whether the precautions taken in Bengal could have proved equally effective in the Punjab. The answer to the last is probably no. In the Punjab there were more guns, for one thing. There, and in the neighbouring North-West Frontier Province, society prided itself on its decidedly military ethos. The north-west had long been the British Indian army’s main recruiting ground, and accounted for around half its intake; service families, military colonies and paramilitary fraternities abounded. Come the end of the war, many thousands of Punjabi Sikh, Muslim, Hindu (Dogra and Jat) and Pathan servicemen had been demobilised; but not all surrendered their arms, and of those who did, many were emboldened to reacquire them or obtain equivalents of local manufacture. In championing the anxieties of their co-religionists and avenging the massacres reported from across the border, Punjabi ex-servicemen of every persuasion found employment in a cause that was lucrative, congenial to their traditions and applauded by their kinsmen.
This was not the case in Bengal. Generally Bengalis, whether Hindu or Muslim, were supposed to disdain the military arts. The province was thus under-re
presented in the army’s ranks and almost devoid of officers. When he arrived in Dhaka as East Bengal’s first General Officer Commanding in late 1947, the then Brigadier Ayub Khan, Pakistan’s future ruler, found ‘there was no army’, just two half-battalions, and ‘no office, no table, no chair, no stationery – virtually nothing at all’.26 Firepower had played little part in the earlier ‘riots’ in Bengal, and there had been even less evidence of tactical planning. The killing sprees had often seemed spontaneous and unpredictable; and in West Bengal the heavy, and usually heavy-handed, presence of the largely Muslim police had already been depleted by migration. In short, Gandhian pressure, plus greater official awareness here stood a chance. Conversely, against the professionals orchestrating the carnage in the Punjab such intervention would probably have failed.
Other factors were also important. Given the deltaic terrain, communications in Bengal were notoriously slow and depended more on waterborne transport than on roads and railways. In the monsoon conditions of August and September whole districts were temporarily submerged, so distracting the inhabitants from mutual hostilities and severely restricting their mobility. In addition, the governments of India and Pakistan, though in the Punjab officially sponsoring an exchange of population, here actively discouraged it. It was supposed that mass migrations might destabilise the delicate political arithmetic on which both the Congress in West Bengal and the League in East Bengal based their prospects of retaining power. If conducted on any scale, migration could easily deplete one half of the province while overwhelming the other; and both Prime Ministers, Nehru and Liaquat Ali Khan, were dead against it. ‘I have been quite certain, right from the beginning,’ Nehru wrote, ‘that everything should be done to prevent Hindus in East Bengal from migrating to West Bengal … even if there is a war.’27 Throughout the period 1949–52, when a further two million Hindus from East Bengal joined the million or so who had migrated in 1947–48, Nehru remained firm. But twenty years later Indira Gandhi, when faced with precisely the war scenario that her father had envisaged, would take a very different line. East Pakistan’s Bengalis, now calling themselves Bangladeshis, would be admitted to India whatever their religion, so furnishing the justification for another Partition, this time of Pakistan.
