A microhistory of violence

The brutal nature and scale of the two riots may have caused much anxiety and alarm at both the levels of Centre and Province at the time, but they have managed only a cameo appearance in the work of most historians so far, taxing little imagination or analysis. The riots are seen as primarily communal, between Bengali and Bihari workers. One set of reasons put forward is that the riots were instigated by the Centre to undermine the new Provincial government.Footnote ²⁰ Another explanation suggests the riots were an extension of the enthusiasm generated by the victory of the United Front and the promises of greater social, political, and economic power and control for the Bengali-speaking populace of the East Wing.Footnote ²¹

The violence at Adamjee and Karnaphuli unfolded in the heartlands of the postcolonial industrial project of Pakistan. However, unlike other communal riots during this period, which were more geographically contagious, the Adamjee and Karnaphuli events remained confined to the mill areas. Although some scholars have alluded to the question of labour, the riots have been subsumed into narratives of high politics.Footnote ²² This has emptied the events of historical agency, content, and significance, and retrospectively justified them to ‘fit’ a particular perspective. Using rare and unseen documents, such as the intelligence documents of the Special Branch for both riots and court cases against Bengali workers in the Karnaphuli riot, this article uses a combination of microhistory, subaltern studies, and labour history in order to counter conventional narratives around the riots. I do this in order to examine other paths and possibilities of the event and centre workers’ agency in narratives where they were the primary protagonists. More broadly, I suggest that a closer reading of the riots will help us to understand and interpret the relationship between labour and politics in the postcolonial period.

Ravi Ahuja describes the late-colonial period up to the 1980s in the subcontinent as the ‘age of labour’—a time when the working class constituted a ‘source of political energy to reckon with, like it or not’.Footnote ²³ He argues that industrial labour was central to the development of political identities, mass movements, and conflicts. In state questions of social policy, labour was also central. Yet, in the context of East Pakistan, we know very little about what the ‘age of labour’ looked like, in terms of labour relations, processes, and politics in new mill spaces.Footnote ²⁴ This means that discussions of labour militancy in Pakistan have been confined to the ‘narrow chronological window of 1968–1972’, with a focus on events in the West Wing.Footnote ²⁵

A microhistory of the Adamjee and Karnaphuli riots offers an avenue for thinking about labour politics, and especially labour militancy in an earlier period of Pakistan. Sailer, in her work on the 1929 Bengal Jute Mill Strike, demonstrates how a microhistorical perspective allows the exploration of different possibilities, paths, contradictions, and tensions of an event, without it being determined or overwhelmed by larger developments.Footnote ²⁶ Similarly, I show how a close-up analysis of the riots as they unfolded, spatially and temporally, offers multiple fine-grained images of workers in action. This dynamic narrative makes the worker agency, politics, and structures, obscured by the narrative of high politics, visible. My close-up analysis is complemented with ‘extreme long shots’ in order to situate the riots in a larger context. Why were these ‘enemy agents’, the Bengali workers, proclaiming ‘Pakistan Zindabad’ as they rioted? How did the workers see their relationship with the state? The article looks at what connects and disrupts our understanding of the riot events to the political drama at the national level. According to Carlo Ginzburg, the ‘constant back and forth between macrohistory and microhistory, between close-ups and extreme long-shots’ demonstrates that ‘reality is fundamentally is discontinuous and heterogeneous’.Footnote ²⁷ A microhistory of the riots thus offers multiple readings of the event and Pakistan itself.

I argue that the riots, though ostensibly between the Bengali and non-Bengalis of East Pakistan, sprang from a more salient tension inherent within the labour situation: a severe crisis in labour–management relations. I will show how the mills of East Pakistan produced differing postcolonial imaginations, labour formations, and solidarities, and different registers and practices of authority. These differences led to an estrangement between workers and employers over who were the real masters of the machines, bodies, and postcolonial state of Pakistan. I focus on the mobilization of Bengali workers during the riots, the complex solidarities they formed, and the demands they made. I conclude by looking at the state responses to the riots and the implications they had for labour politics in the postcolonial phase. This article will attempt to examine both riots, but explores the Karnaphuli riots in more detail, due to the availability of material.

Mills and industrialization

The work of scholars like Akhil Gupta and others has shown how postcolonial nation states were constituted, actively experienced, and imagined by the populace in multiple ways through different mechanisms, modalities, and practices. Srirupa Roy, for example, shows how the discursive and material practices of the state in the form of Republic Day parades, Film Divisions documentaries and newsreels, and State Development Plans gave form to the Indian nation state.Footnote ²⁸ East Bengal, a hinterland to the bustling metropolis of Calcutta during the colonial period, became the principal space in which Pakistan performed industrial and state power.

The spoils of partition had been humiliatingly minor for Pakistan. The eastern wing received only 12 per cent of industrial establishments of United Bengal, but no jute mills, iron and steel plants, paper mills, chemical works, coal mines, or established hydroelectric projects.Footnote ²⁹ The response to this injured pride was the industrialization of East Pakistan at an impressive and rapid scale. The 1948 ‘Statement of Industrial Policy’ proclaimed that ‘free play will be given to private enterprise and individual initiative’ and clarified that certain industries were to be owned and operated by the state, while others were to be made the subject of Central Planning.Footnote ³⁰ To this effect, the government set up the Pakistan Industrial Development Corporation (PIDC) and entrusted it with the task of developing the major industries for jute, paper, sugar, textiles, shipbuilding, cement, iron, and steel, as well others. These industries were to be financed entirely by the state or in partnership with private industrialists. Papanek has shown how Pakistan's pro-private enterprise position in the early years enabled the growth of a new class of businessmen and industrialists.Footnote ³¹ In the first couple of years, PIDC directed almost 70 per cent of their initial investment towards the making of the Adamjee Jute Mill and Karnaphuli Paper Mill—the crowning jewels of Pakistan's postcolonial ambitions.Footnote ³²

The construction of the Adamjee Jute Mill, situated in Siddhirganj, about four miles north of the bustling river port town of Narayanganj, began around 1950. By 1954, it was the largest jute mill in the world. The glossy brochure of the PIDC in 1955 stated that the three mills of 1,000 looms with their ‘modern machinery, ready access to all qualities of jute, [and] plentiful supply of labour capable of being trained to a high standard of efficiency [,] stood in a strong position capable of successfully competing with any country in the world’.Footnote ³³ The construction of Karnaphuli Paper Mill, situated on the borders of Chittagong Hill Tracts and Chittagong District, was equally massive. By 1954, the mill was equipped with the latest machines in papermaking and included facilities such as its own powerhouse, water filtration, and chemical plants. The PIDC boasted that the electricity and water consumption of the factory exceeded that of Karachi, Pakistan's capital.Footnote ³⁴ By 1954–55, Adamjee Jute Mill and Karnaphuli Paper Mill were two of the largest employers of an industrial working-class population in East Pakistan.Footnote ³⁵

The constant reference to Karachi—the seat of national political power—in the PIDC brochures was not an incidental comparison of scale, but underlined the high-risk political gamble being made by the state. The Pakistani state had intimately crafted these industrial projects as a vision of its own spirit, power, and legitimacy. They were the physical and symbolical stamps of Pakistan's immortality and permanence in the postcolonial era. Travelling through the riverine delta of Bengal, these massive sprawling industrial establishments, with thick, grey smoke billowing out from their chimney towers and thousands of bodies at work, were the visual spectacle desired in place of rolling paddy fields. The industrial grandeur of Adamjee Jute Mill and Karnaphuli Mills reflected the dizzying heights of postcolonial ambition and, most importantly, the growing might of the newly emerging free postcolonial nation state of Pakistan.

However, industrialization was not simply a project of the possible postcolonial state futures, but also an attempt to deal with the pressing material needs of the province, which emerged as a result of extraordinary population pressures on the land, a largely unemployed and urban refugee population, and a locked-in trade situation with India.

The 1951 Pakistan Census recorded approximately 700,000 immigrants coming into East Pakistan from West Bengal, Assam, Orissa, Tripura, and Bihar.Footnote ³⁶ Dina Siddiqi describes the refugees as a heterogeneous group in terms of class and ethnicity but united by their linguistic difference from the Bengali-speaking majority in East Pakistan.Footnote ³⁷ The catch-all term ‘Bihari’ designating the refugee population was a marker of that difference.Footnote ³⁸ However, this was not the only thing that the refugees shared; they were in many ways also a working-class population. The majority of Biharis who migrated to East Pakistan were a lot poorer than their counterparts in the West Wing. Fleeing from bloody and horrific riots around partition and after, these refugees had crammed their bodies onto trains and carts, leaving behind their homes and meagre possessions, to arrive in East Pakistan empty-handed.Footnote ³⁹

Although East Bengal saw smaller inflows of refugees than West Pakistan and a greater outflow of minority communities, it had less land and fewer economic opportunities to offer these muhajirs (refugees). With no material attachments to the land of East Bengal, the refugees constituted an additional pressure point on the state of Pakistan to industrialize as quickly as possible in order to resolve unemployment problems both in rural Bengal and amongst the refugees. The mills, thus, acquired pride of place in the state's programme of refugee rehabilitation, providing them with jobs, housing, and a sense of belonging. The refugees, in turn, must have forged a specific conception of, and devotion to, the mills and the state.Footnote ⁴⁰

Thus, in 1954, when these symbols of national pride turned into sites of carnage and destruction, the Central government could not help but react forcefully against those who threatened to paralyse industrial progress in Pakistan. However, if massive and spectacular industrial projects represented state power, perhaps the rioters’ occupation of mill space and violence can be interpreted as a counter-performance of power and belonging by postcolonial citizens. The cryptic comment of Bengali mechanic Amiruddin Chowdhury to his Hyderabadi colleague Abdur Razzaq appears to support this interpretation. On 12 March 1954, at the Karnaphuli Paper Mill, a week or so before the riot, Amiruddin Chowdhury, making conversation with Abdur Razzak, stated: ‘If one big officer is killed, everything will become O.K.’Footnote ⁴¹

Complex solidarities

Why did these mills become sites of industrial carnage? In 1954, a general restlessness spread among the population of both wings of Pakistan. In the East Wing, the police recorded a rise in riots. Over 2,225 cases were reported in the first six months of 1954 compared to the 1,754 cases reported in the first five years of independence.Footnote ⁴² The most serious disturbances occurred in industrial workplaces between Bengali and Bihari workers. But this bad blood had not always been there. In 1952, Bengali and Bihari peons, orderlies, and sweepers jointly defied their government employers by wearing ‘hunger’ badges and facing suspension together.Footnote ⁴³ The police attributed the changing relationship between the groups to the elections, which had stirred feelings of Bengali provincialism.Footnote ⁴⁴ At face value, the connection between the riots and the broader political context seems evident. The Karnaphuli and Adamjee riots appear to be a communal riot, with Bengali workers attacking and killing Bihari workers and vice versa. However, on closer inspection, this picture starts to disintegrate. R. K. Bose, the co-ordinating engineer of the Karnaphuli Mill, who captured the riots on film, was one of the few people able to offer multiple and diachronic shots of the riot. The intelligence reports stated:

As I will argue, the close-up shots of the Karnaphuli riot provide us with alternative pictures to that of a mob whipped up by communal frenzy and hatred. The ethnicity of workers was not immutable, but rather an unstable and fluctuating category of identity; workers actively constituted and reconstituted identities through discussions and bodily interventions during the riots.

The workers’ negotiation of identities can be demonstrated through several specific instances. Syed Badaruddin, the personal assistant to the operative director, Khurshid Ali, was rescued from his attacker, a ‘bearded man with a stout stick’, by Bengali workers on account of the fact that he was a ‘good man’. The senile and poor-sighted Abdur Rahman Khan, the Pathan chowkidar, was rescued by the shouts of ‘bas karo, mat maro’ (stop it, don't hit him) by another Bengali worker, at which point his attackers abandoned their assault. The assailants of Abul Kasem Zaidi, an assistant electrical engineer at the Power House, escorted him to a safe place after they had identified him as a ‘worker’.Footnote ⁴⁶ Other stories emerged of Bengali rioters who were both assailants and, at other times, protectors of non-Bengalis. It appears that, where the Bihari person was identified as a worker or someone who appeared to possess moral qualities that workplace relations and labour processes failed to suppress, their ethnic identity was no longer their primary identification and their life was spared. The compartmentalization of labour processes within the mill can help to explain the inconsistencies and contradictions of the behaviour of the rioters. Their role as assailants, spectators, or protectors was based on workplace relations, which were not the same for every one of the rioters or those they attacked.

These identities were produced through not simply the mill environment, but also other commingled spaces such as the neighbourhood. In the case of Hatom Serang, accused of incitement and distributing sticks to the Karnaphuli rioters, the judge was puzzled at his benevolence towards Yusuf Khan, the cousin of the murdered Pathan durwans: ‘If he were a wicked man, he would deal with Yusuf Khan, P.W. 90 in the manner which other rioters had dealt with Mudassar Khan [head durwan]. I hold that Sk. Hatom has established his innocence.’ Hatom Serang's behaviour becomes less enigmatic when the judge comes to let slip in his summary that both Yusuf Khan and Hatom Serang were ‘close neighbours’.Footnote ⁴⁷

The ‘Bihari’ and ‘Bengali’ identities were more pronounced during the riot for certain sections than others. Anu Miyan, a ‘helper’ at the Saw Mill, aware of a commotion happening at the mill, had remained at his house in Dobaashi Bazaar, cleaning. It was only when Saiyedur Rahman, the secretary of the Chandraghona Paper Mill Union, and a contingent of 50–60 men arrived at his doorstep and shouted, ‘How is it that you are sitting idle, don't you work at the mill?’ that Anu Miyan left his house to take part in the riot. Anu Miyan's identity as a ‘Bengali’ was defined by his relationship to the mill.Footnote ⁴⁸ His participation in the riot mattered because he was a Bengali worker, who was being asked to demonstrate solidarity because of a common experience and knowledge of labour within the mill. However, this demand of Bengali solidarity was not made on all the Bengalis at the mill. Mujibur Rahman, the Bengali administrative officer, was intentionally set upon by workers during the riots. He was charged with having done little to redress their grievances despite being their ‘Bengali brother’.Footnote ⁴⁹ These bonds of fraternity, then, were not assumed or natural. They were based on a certain set of expectations based on their relationship within the mill-scape and, if not met, that relationship was easily broken. Mujibur Rahman's association with Khurshid Ali undermined his claims to Bengali-ness in the mill context. Conversely, the Bihari identity was pronounced for some because of their relationship to authority. As the riot unfolded at Karnaphuli, it was clearly a conflict between labourer and mill management, who happened to be non-Bengali as well. In these specific industrial circumstances, to be ‘Bihari’ meant to be marked as an authority figure and open to attack.

The question, then, turns to why had the death of Khurshid Ali, the operative director of Karnaphuli, not been enough? Why did papermakers, labour officers, peons, clerks, and durwans have to die—some of whom were on the bottom rung in the workplace, much like their Bengali counterparts? It was because these men had come to be seen as ‘Khurshid Ali's men’.Footnote ⁵⁰ Their bodies were seen as extensions of Khurshid Ali's authority, particularly so in the case of labour officers and durwans, who were often used to discipline workers, and were therefore the visible, and more brutal, signs of domination and subordination of the workers.Footnote ⁵¹ The death of Mudassar Khan, the head durwan, had not come as a surprise to the police, who had described him as a ‘marked man’ since 1951, when he had been tried for murder and acquitted in a previous riot.Footnote ⁵² The workers meted out their own form of ‘justice’ as they struck against these men for the crimes they had committed in their position of authority.

The Adamjee Jute Mill riot narrates a similar tension between management and workers. On the morning of 15 May, Bihari workers gathered to pay respect to the murdered Bihari durwan, and the mill managers and administrative stuff—the majority of whom were also non-Bengalis—joined them.Footnote ⁵³ The mourners wore black badges, unfurled and hoisted black flags, and declared the day as ‘martyrs day’. As Bengali workers went to collect their wages from the Pay Office, they were confronted by Bihari workers, armed with guns from the durwan store, and shouts of ‘rokter badla rokto cai’ (we want blood for blood).Footnote ⁵⁴ The riot started off as coordinated managerial violence unleashed on Bengali workers; however, it assumed a different dynamic and an up-ending of those relationships of power as it spread out to the working-class neighbourhoods, where it was now the turn of non-Bengalis to be killed.Footnote ⁵⁵ These spaces after all housed workers from various industrial establishments in Narayanganj, such as those employed in the dockyards, jute presses, textile mills, and glass factories. For these men, on the day of the riot, the Bihari labourers of Adamjee Jute Mill had become coterminous with an authority that had turned guns on its workers.

Labour historians have long debated the question of communal violence between workers in South Asia and taken several different approaches. Dipesh Chakrabarty wrote about an enduring communal consciousness amongst workers, reinforced through the sardari (jobber) mode of recruitment and control, which was based on rural, caste, and religious connections, and prevented workers from forming a united front along the lines of class.Footnote ⁵⁶ Chandavarkar, arguing against Chakrabarty's cultural determinism, suggested that disunity amongst workers was instead a rational response to business strategies, employer policies, and the mechanisms of the labour market.Footnote ⁵⁷ However, Sumit Sarkar is correct to observe that both Chakrabarty's and Chandavarkar's deterministic models omit ‘moments of labour self-activity, militancy and autonomy’.Footnote ⁵⁸ Nandini Gooptu and Chitra Joshi develop a more fluid, dynamic, and complex understanding of worker activity, identity, and solidarity, situated in their everyday working-class life and events of militancy.Footnote ⁵⁹ Chitra Joshi writes: ‘Workers bear the mark of multiple identities, a multiplicity not captured in the neatness of homogenous categories.’Footnote ⁶⁰ Workers had a combination of identities—caste, class, religion, nation—and these were constantly shifting, contested, and reconfigured, depending on the specific context. Class struggles were at times refracted through caste, religious, and national lens.

Following Gooptu and Joshi, I offer a picture of complex solidarities and unities that emerged during the riots. Ties were formed between different types and forms of labour, between workers of different regions, and between the workplace and neighbourhood. These connections were fundamentally based on the life of labour in and outside the industrial establishments. I suggest that these were complex because these ties were not in any sense assumed or natural, nor were they exceptional or anomalous to the context. These solidarities offer a different explanation to the presumed Bengali–Bihari axis of the riot. However, they were complex because there were very precise and real ways in which the Bengali–Bihari axis was being utilized by the management.

The new borders between India and Pakistan in 1947 became a means of labour discipline and capital accumulation. Historically, Bihari men have been ‘one of the largest and most enduring labor diasporas’ from the eighteenth century onwards, working as mercenaries in early modern India and later as earthworkers, railway porters, jail wardens, and millhands in the jute mills of Hooghly and Calcutta.Footnote ⁶¹ After partition, workers previously described as having ‘one foot in the rural world and other in the urban industrial complex’ had little opportunity to return to their homes. They were now truly a proletariat class—landless and urban. They were left dependent on the management for employment, housing, and a sense of community. These workers, previously perceived as unruly and troublesome, were now vulnerable and precarious. A report on the human and social impact of industrialization in East Pakistan stated:

The management used this weakness to discipline workers in two ways. First, the threat of Bihari workers organizing against the management was removed as a consequence of their economic dependency.Footnote ⁶³ Second, the management used this ethnic division to discipline Bengali workers. Because the Biharis were seen as tough, they were appointed as security guards, supervisors, contractors, and labour officers at the mills. They were paid better and given free accommodation. A University of Dhaka report on industrial labour showed workers from outside the province were earning on average Rs 67.5 per month as compared to Rs 60 for other workers.Footnote ⁶⁴ Thus, Bihari labour power came from them being able to discipline and undermine Bengali labourers. G. P. Hampshire, the UK Deputy High Commissioner, noted, in his report on the Adamjee riots, how one of the strategies of the management to prevent labour organization was to ‘import and place at strategic points in the Mills trusted employees from Calcutta (more Biharis) who were in effect management spies and goondas [gangsters]. The effect of this was not unnaturally a marked increase in tension’.Footnote ⁶⁵ The management used the Bihari worker as a buffer between themselves and an unfamiliar, and consequently unstable, workforce. The contradiction between the image of the powerful Bihari workers as the strongmen for mill management and their actual vulnerability as refugees led to their deaths at the hands of their Bengali co-workers.

The Bengali workers

The Bengali workers at the Karnaphuli Paper Mill were the main protagonists in the riot, but what do we know about them? Like the Biharis, there is a vulnerability that becomes apparent as the labour-recruitment process is more closely examined. There were 2,000 workers on the ‘regular pay roll’ of the Karnaphuli Paper Mill for the three shifts.Footnote ⁶⁶ It is not clear as to whether all these workers were part of a permanent labour force or drawn from a pool of badli (regular substitute) or casual labourers.Footnote ⁶⁷ Anu Miyan, one of the accused, had only been hired back as a ‘helper’ after having been retrenched from his earlier job as ‘stone crusher’ at the mill.Footnote ⁶⁸

The practice of hiring temporary labour, daily-pay workers, and contract workers remained common in most industrial establishments, with just a small number of workers assigned with permanent status.Footnote ⁶⁹ Arjan de Haan notes how, in the pre-partition period, the badli system was used by the mill management as an ‘adaptation to the migratory pattern of workers’, so that mills continued to function when workers returned to their villages to attend marriages, emergencies, harvests, and for other family reasons.Footnote ⁷⁰ The badli workers often constituted a vast majority of the jute-mill workforce, recruited on a daily basis at the factory gates.Footnote ⁷¹ The badli system continued to persist in spaces like the Karnaphuli Paper Mill and the Adamjee Jute Mill for the Bengali workers, who continued to have rural attachments and obligations, unlike the Bihari workers, who were most likely permanent.

The International Labour Organisation (ILO) Mission of 1952–53, commissioned by the Pakistani government to identify labour problems, highlighted the labour inefficiency in industries, with employers hiring twice the number of workers they needed. The report attributed this to the abundant supply of cheap labour and managerial resistance to new work methods as well as to repairing expensive machine parts. Interestingly, the ILO Mission also noted that labour inefficiency was a deliberate worker tactic in East Pakistan, which had large numbers of employed and underemployed persons.Footnote ⁷² Hence, many workers found themselves bought in as ‘helpers’ and apprentices for months by relatives and friends, living on less than a subsistence wage, before they could be assured of some regularity of employment, even as part of a reserve force of casual labourers. The University of Dhaka report recorded the case of a 22-year-old married man at the Adamjee Jute Mill who had ‘smuggled’ his way in to learn work in the spinning section for two months pretending that he had relatives at the mill when he had none. After two months, when work at Mill No. 2 started, he walked up to the manager for a job, was given a test and found suitable, and therefore given a more permanent position.Footnote ⁷³

The working-class neighbourhood, in particular the bazaar, emerged as an important space for solidarities to form between temporary, irregular, and contract workers. Dobaashi Bazaar in Karnaphuli was the waiting area of labour. The day after the riots, the superintendent of police of Chittagong Hill Tract prepared to undertake a raid on the bazaar to round up suspects on the information that the large group of mill employees, as well as ex-employees and those looking to be employed, had congregated there.Footnote ⁷⁴ This gathering was not related to the riot itself, but indicates the mill's regular practice of hiring and discharging workers daily and the blurring of the distinction between permanent and casual labour.Footnote ⁷⁵ Chandavarkar has demonstrated the importance of the neighbourhood in mediating a worker's relationship to the workplace. He describes it as a ‘distinct arena of social and political action’.Footnote ⁷⁶ As Bengali workers gathered and waited anxiously in Dobaashi Bazaar for different contractors to come and employ them for earthwork and work in the brickfields, bamboo yard, and mills, they drank tea, gossiped, and exchanged information on jobs, daily wages, and jobbers.Footnote ⁷⁷ The bazaar and neighbourhood were spaces were solidarities and dependencies were formed between Bengali workers on the basis of their shared experience of the impermanency of labour at the mill and their impatience with their precarious conditions.

The Noakhali connection

Although the original charge sheet of 63 Bengali workers in the Karnaphuli riot revealed a mix of workers from various districts of East Pakistan, the majority of the rioters had been from Noakhali (see Table 1).

Table 1. The native district and subdistricts/police stations of Bengali workers charged in the Karnaphuli case

These data have been compiled from the charge sheet in Government of East Bengal (1957), Chandraghona Files. Source: Chandraghona File, 1957.

This pattern appears again in the Adamjee riots, where Noakhali workers played a prominent role.Footnote ⁷⁸ Noakhalias have long been an industrial and footloose community due to their proximity to places like Chittagong and Narayanganj and their mobility, driven by severe landlessness, continual floods, soil erosion, and severe food shortages. Prior to partition, men from Noakhali were working in the ports, docks, and factories of Calcutta.Footnote ⁷⁹ Although an itinerant community, many left for East Bengal after the 1946 Calcutta killings and some were directly involved in the Noakhali riots that occurred later on in the year.Footnote ⁸⁰ After the birth of Pakistan, Noakhalis continued to be mobile, travelling to Dhaka, Narayanganj, Chittagong, and other industrial spaces for work.

However, there were no natural lines of unity between the Noakhalis and other Bengal workers. Richard Kriegel, in his report on Labour in East Pakistan, 1947–1959, stated: ‘There is also what might be called “normal tension” between the people of various districts or grouping of districts.’Footnote ⁸¹ Noakhalias were regarded as ‘ruffians’ and resented by Bengali workers of other districts because they were able to obtain more secure and better-paid work due to their industrial experience.Footnote ⁸² What, then, explains the solidarity between Noakhali workers and other Bengalis during the riots? What role did the Noakhali workers play in these riots?

Noakhali labourers were found to be influential in supporting the establishment of workers’ unions at Adamjee Jute Mill and Karnaphuli Paper Mill. The Chandraghona Paper Mill Union meeting, which had taken place the evening before the Karnaphuli riot, had been presided over by a Noakhali, who worked as a subcontractor at the mill, and two of the speakers, a foreman and a labourer, were from Noakhali as well.Footnote ⁸³ These were the unions that fell out of the ambit of managerial favour, often described as ‘red-flag unions’, and came to rival the officially backed unions.

Though Pakistan had signed up to International Labour Organization conventions immediately after partition, the government and industries were often unfavourably disposed to the existence of trade unions.Footnote ⁸⁴ Around 1952–53, employers had recognized only 71 of the 321 registered unions, and there were many that continued to operate unregistered.Footnote ⁸⁵ Employers did not have to officially enter into negotiation with any unions that were not recognized. Workers who were suspected of organizing were frequently dismissed by employers under false charges of ‘unsatisfactory work’, ‘neglect of duty’, ‘agitation for strike’, ‘creation of disturbances amongst workers’, ‘insubordination’, ‘demonstrating against authorities’, ‘delivering of instigatory [sic] speeches’ and ‘shouting of slogans’.Footnote ⁸⁶ Employers appointed work committees and labour officers to counteract any form of self-activity amongst workers.

After partition, the East Pakistan Federation of Labour emerged as the representative body for workers, with Faiz Ahmed as the principal labour leader. His close relationship with Aftab Malik, the first labour minister of Pakistan, who initially proposed the idea of a central labour organization, meant that he was someone that the government and industrial owners were willing to bargain with.Footnote ⁸⁷ This use of trade unions to control and undermine self-organizing amongst workers can be traced back to the partition period. Subho Basu's work demonstrates how, during the 1937 general strike in the jute-mill industry, the Muslim League set up the White Union to disrupt the activities of ‘red-flag’ union and the radical factory committees during the 1937 general strike in the jute-mill industry in Calcutta. The White Union, supported by the state and the British and native industrial lobby and funded by the mills, was able to effectively diminish the threat of labour militancy.Footnote ⁸⁸

Over the 1950s, Faiz's position as the sole representative of workers grew weaker among the competing demands and counterclaims of representation.Footnote ⁸⁹ It was in this somewhat democratic atmosphere of unionization that the Adamjee Jute Mill Mazdoor Union and Chandraghona Paper Mills Workers Union emerged and solidarities were forged more strongly between Bengali workers of different regions. Though the unions may not have been established by workers, but by political outsiders, their location in the bazaars of the working-class neighbourhoods and their claims as an alternative to the officially backed union suggested a more familiar relationship with the life of labour. This enabled these unions to engage with workers in new ways and adopt a more uncompromising stance with management. S. A. Priddle, the labour adviser to the British High Commission in Karachi, noted in his report a week before the Adamjee riots the rapid growth of ‘communist’ influence in unions. Priddle described a depressed Abdul Currim Karawadia, general manager of the Adamjee Jute Mill, lamenting the strength of the new Adamjee Jute Mill Mazdoor Union, which he had been forced to recognize, and that he had not been able to get ‘a decent day's work’ out of Adamjee workers since.Footnote ⁹⁰

Men in black

As the courtroom drama of the Karnaphuli riots unfolded, it became apparent that the main protagonists of the riot were not just the 49 Bengali labourers in the dock, but also 2,000–3,000 ‘unknown persons’ involved in the disturbance. In the First Information Report (FIR) filed by the police, the deputy inspector general of police mentioned the difficulties in identifying the perpetrators: ‘in continuation of the previous report, I am to state that 75% of the rioters were not workers at the mill.’Footnote ⁹¹ That statement, however, was not entirely accurate. Under cross-examination, Prosecution Witness 25 disclosed that the 500–600 men who had started off the hulla (riot) had been dressed in black genjis and lungis (vests and sarongs).Footnote ⁹² These men in black were part of the 2,000-odd contract labour force working and living in the northern hinterland of the factory, in an area known as Baragonia.Footnote ⁹³ In conversation with the British trade commissioner, M. A. H. Ispahani, member of the Jute Board, PIDC, and one of the leading industrial families in East Pakistan, also hinted at the participation of the contract labour force in the riot as they faced unemployment after having recently completed their assignment.Footnote ⁹⁴ These were men who had been employed by majhis (petty subcontractors) and worked under various thikadars (contractors) and were hired primarily as earth-diggers for road building, as well as in brick manufacturing.

The ILO Mission noted that industrial employers preferred to use contractors for the execution of projects where time was limited and a huge labour force required.Footnote ⁹⁵ Ahmed Sofa's short essay, ‘Karnafulir Dhare’ (‘The Banks of Karnaphuli’), offers a rare and detailed insight into the practice of the contract labour regime in the postcolonial space and, more importantly, at the Karnaphuli Paper Mill. Various thikadars were employed for the building of roads into the hills. These thikadars, who were usually not local, would enter into agreements with local majhis for labour recruitment. The majhi would receive an advance of money and a promise of a percentage of the daily wages of the workers.Footnote ⁹⁶

Traditionally, the majhis have been Santhal chiefs—members of one of the oldest tribal communities in north-east India. Since the colonial period, the Santhals have been involved in circulating labour, moving from one area to another doing railway work, earthwork, and mining.Footnote ⁹⁷ Majhis operated within kinship and caste structures to recruit and supervise workers. According to Ian Kerr, these earthworking communities were ‘marginalised, even criminalised by the authorities and local inhabitants in the areas through which they moved. They became the quintessential social outsiders, the feared, despaired or degraded “others” despite their importance to local economies’.Footnote ⁹⁸

In the postcolonial period, we know little about the background of the majhis and the contract labour force. It is possible that Santhals and other tribal or lower-caste communities continued to engage in contract labour. However, the increasing availability of circulatory labour in postcolonial Pakistan meant that it was no longer necessary for the petty contractor to retain their deep and familiar previous connections with labourers. The majhi now simply rented out eight to ten shacks in working-class neighbourhoods, laid out coarse mats, came to an arrangement with the local shopkeeper for the supply of food and other goods, and then set out to recruit workers. Ahmed Sofa, in his description of this recruitment process, stated:

Having convinced the men that their new employer was ‘a devout person, prays 5 times daily and doesn't squander anyone's lawful earnings’, they would be taken to the shacks. Once the majhi had his required number, they were handed over to the thikadars to work long hours, in unpleasant working conditions, for low wages.

The unity between the millworkers and earth-diggers during the riot appears to arise for several reasons. First, the continuing segmentation and dominance of the labour force along religious and ethnic lines, with Biharis in better-paid, supervisory, and piece-work jobs and the Bengali, tribal, and low-caste Hindu workers in inferior, poorly paid jobs with limited mobility, united these different groups of workers.Footnote ¹⁰⁰ Second, the overlap of the informal with the formal sector, demonstrated through the presence of work gangs employed by subcontractors in the mill space, highlighted the precarity of all the workers who were not permanent, organized, and protected by labour legislation.Footnote ¹⁰¹ Finally, what united the workers was the new set of expectations and hopes that the postcolonial context generated. On the morning of the Karnaphuli riot, news that the contractor Amir Bin Rashid (later killed in the riot) was at the Pay Office receiving his bill was relayed to the earth-diggers. Over 300–400 earth-diggers laid down their tools, picked up bamboo sticks and spare pieces of rod, and moved towards the heartland of the mill. As they approached Gate No. 1, they shouted, ‘hum lok ka bichar karlo’ (give us justice); their appeals were directed not at the contractor, but at the senior manager, Khurshid Ali. However, their cries fell on deaf ears, as Khurshid Ali left the office in his jeep whilst labourers continued to shout ‘humara bichar karo’ (do us justice). When Mudassar Khan and the other Pathan durwans sought to disperse the crowd, the rejected workers turned on them and the Pay Office instead.Footnote ¹⁰²

The majhis who had been so critical to the labour-recruitment process fled the area after the riots and could not be traced by the investigators.Footnote ¹⁰³ While very little is known of the involvement of the majhis during the riots themselves, their subsequent disappearance raises important questions. Had the majhis fled because they were afraid of the earth-diggers or was it because they, too, had participated in the riots? Their possible participation narrates a complex solidarity between two groups, given that this relationship was fundamentally based on exploitation. In many ways, the majhis and thikadars functioned like the sardars, mukadam, maistry, ghat serangs, and arkattis that appear in the labour historiography of colonial South Asia as intermediaries between labour and management, responsible for the recruitment, control, and supervision of workers. The jobbers were instrumental for workers in the provision of employment, credit, and information, but their power was also dependent on how well they served the workers.Footnote ¹⁰⁴ Thus, during the course of the Karnaphuli riot, a majhi who did not demonstrate loyalty to the workers might have been considered a Bihari and thus liable to attack.

The neighbourhood

During the riots, the working-class neighbourhood emerged as a space that challenged the authority of the mill. There were two men in the Karnaphuli court case who had stood out amongst the accused: Aminullah Patwari of Noakhali and Gunu Mia Sawdagar of Chandraghona. Whilst the rest of the accused had been workers at the mill, these two men had been traders of Dobaashi Bazaar. Aminullah Patwari, the owner of Haji Stores, a tea and groceries shop, was accused of playing a major role during the riots. He was charged with having ‘repeatedly instigated and exhorted’ the earth-diggers and labourers of Baragonia, and supplying them with sticks during the riots. Aminullah Patwari was heard shouting that he would ‘consume 4 tins of kerosene oil to set fire to the houses of outsiders’, which had gestured to the fact that he was a man of means and of his power and status amongst workers.Footnote ¹⁰⁵

The work of Chandavarkar, Basu, and Gooptu demonstrates the importance of working-class neighbourhoods in sustaining the material, social, and political life of workers: the energy, buzz, activities, and relationships in the neighbourhood shaped by workers’ needs for housing, credit, employment, sex, and leisure. Chandavarkar writes: ‘The material needs of the working class were not the only factor in the constitution of their neighbourhood connections. Leisure and political activities also contributed to the development of street and neighbourhood as a social area.’Footnote ¹⁰⁶ The worker built relations in the narrow alleys that led to their houses, the messes (shared housing) they lived in, the brothels they visited, and the tea stalls they stopped at for a chat. At times, the neighbourhoods were an extension of the workplace and, at others, they were the location of power and patronage that competed with the mill space.

The life of labour was all-important in the working-class neighbourhoods of Adamjee and Karanphuli Paper Mills. The union meetings, located in the bazaar, provided for the public airing of grievances by labourers against the authorities. The undercover agent of the Special Branch, who had been attending the Chandraghona Paper Mill Workers Union meetings at Dobaashi Bazaar, remarked that, though 90 per cent of the attendees were workers, the rest were members of the public.Footnote ¹⁰⁷ Thus, on the eve of the Karnaphuli riot, more than 100 attendees out of the 1,200–1,500-strong audience were members of the public. These meetings presented a forum for discussion not simply on issues specific to the mill, but also on life in the neighbourhood, whose fate was so closely intertwined to that of labourers in the mill. Abdur Rahman, employee in the chipper section of the mill, was a speaker at the meeting on 21 March, where he disclosed that the mill authority had closed down the premises of ‘poor shopkeepers’ and demanded that the mill help reopen the shops. This was not, however, a charitable gesture that was being made on behalf of ‘poor’ shopkeepers: these men had their own resources and played the critical role of providing credit at times of deprivation, and therefore posed a challenge to the mill's monopoly of local authority. The neighbourhood may have forged solidarities between different workers and groups leading to new relationships of power and authority emerging to challenge the hold that the mill management had over its workers.Footnote ¹⁰⁸

Men like Aminullah Patwari came to acquire a powerful position in the neighbourhood; he was not simply the teashop manager, but also a credit-lender, landlord, contractor, and a financial patron of the union as well.Footnote ¹⁰⁹ Aminullah Patwari was a lifeline for many workers, particularly for those confronted with job insecurity or, even worse, unemployment. His interest in their labour meant that Aminullah was able to wield influence and set the agenda for workers in ways that the management could not. The influence of Aminullah Patwari could be discerned in the cries of labourers as they darted in and out of the Dobaashi Bazaar, saying, ‘Amir Bin Rashid ko mari felaitechi’ (we are killing Amir Bin Rashid).Footnote ¹¹⁰ Amir Bin Rashid, a rival Bihari contractor, provided competition to Aminuallah Patwari that needed to be eliminated. There was none better to help him in this than his debtors, tenants, and chai-drinking friends—the aggrieved Bengali workers.

‘No to race’, ‘no to retrenchment’, ‘no to prostitution’

What turned this seething underbelly of discontent into a full-blown riot? This section examines more fully the immediate provocations of the Karnaphuli riot, in particular the dismissal of one Md. Ekhlas, a Bengali chemist, and the slogans that were shouted by workers before and after the riot.

At the time of the riot, Md. Ekhlas, the most senior of Bengali officers in the mill, had only just returned to work at the Soda Boiler Recovery plant of the paper mill, having been sent to the United States of America by the PIDC for further training and education. His return was marked by a strained relationship with the operative director, Khurshid Ali. The chief accusation against Md. Ekhlas was his allegedly extravagant ambition to be the ‘No. 1 in management’.Footnote ¹¹¹ Other complaints, no less serious, were to do with his friendship with the prominent labour leader, Zahur Ahmed Choudhury of Chittagong Town, and attendance of union meetings in the bazaar.Footnote ¹¹² On the morning of the riot, an already aggrieved Khurshid Ali entered into an altercation with Ekhlas, which ended up with him firing Ekhlas on the spot.

It was after the dismissal of Md. Ekhlas that the riot broke out. Amidst the din of riots, shouts were heard of ‘Ekhlas Sahib Zindabad’ and ‘Khurshid Ali Murdabad’ (death to Khurshid Ali). These shouts were a continuation of the slogans chanted by workers the night before after the union meeting in the bazaar. Little is known about Md. Ekhlas's relationship with trade unions or the workers, but the significance of these slogans can be understood in terms of the possibilities that his position offered to labourers. Three weeks prior to the riot, a vacancy had arisen in the boiler room. Ekhlas had introduced a man from his district to Mr Young, the mechanical engineer, for the vacant position. This person was later found to be participating in the riot.Footnote ¹¹³ Ekhlas, therefore, was not just any employee, but a modern-day jobber, who employed his seniority and influence as a salaried officer in the mill to secure jobs for the precarious Bengali worker. His personal intervention on behalf of the Bengali workers produced an alternative vision of authority—one that was responsive to their problems and accessible to workers in ways that the mill management was not.

Khurshid Ali was a man said to be obsessed with paper, but also equally paranoid about sabotage.Footnote ¹¹⁴ In 1953, the inspector general of police had been informed of mounting labour unrest in the mill, but little had been done by Khurshid Ali to resolve it. In this constant state of insecurity, he feared that any entertainment of demands made by workers would be seen as a weakness in the management armour. Bengali workers perceived the instant dismissal of Md. Ekhlas as proof of the brutal and autocratic nature of managerial power. The altercation, which was overheard and transmitted in various ways by different labourers, emphasized Khurshid Ali's excessive authority. Badaruddin, the personal assistant to Khurshid Ali, in his version of what transpired, stated:

Khurshid Ali's authority was not only excessive, but now transgressed the boundaries of legitimate authority. His supersession of the powers of government was viewed as having removed that fundamental layer of protection between workers and their employers, and a sign of his undisguised oppression and brutality. In effect, Khurshid Ali's power had become illegitimate. This, then, explains why workers invoked what they would have regarded as the supreme authority in matters of what was legitimate and illegitimate: God and the state. As workers moved from one section of the mill to another and attacked Khurshid Ali and his men, the slogans ‘Naraya Takbir’ (Say, God is the Greatest), ‘Allahu Akbar’ (God is Greatest), and ‘Pakistan Zindabad’ provided the rhythm and lyrics of the riot.Footnote ¹¹⁶

The delegitimization of mill authority did not simply happen as a result of the sacking of Md. Ekhlas. The evening before the riot, as the union meeting ended, a procession of 200 or more workers left the bazaar to march around the cricket fields of the mill. Various slogans were shouted, especially ‘race chaina’ (no to race), ‘retrenchment chaina’ (no to retrenchment), and ‘randibaji chaina’ (no to prostitution).Footnote ¹¹⁷ ‘Race chaina’ was not against non-Bengali workers, but against their differential treatment—a demand for equality of labour. The difference between Bulu Meah, the Bengali durwan accused of murdering the head durwan Mudassar Khan, and his Pathan colleagues was not just on ethnic grounds, but on the value that management placed on his labour.Footnote ¹¹⁸ Despite sharing the same disciplinary responsibilities, the Bengali durwans were on a lower pay scale than their Pathan counterparts, who were paid Rs 80 compared with their Rs 60. At the union meeting, these differences were described as part of the ‘step-motherly treatment’ by the mill management, who had made ‘race’ matter.Footnote ¹¹⁹ Analogous to the stepchild in South Asian fables, the Bengali worker felt a deep sense of rejection and denied of his rightful share, though not through any fault of his own.

‘Retrenchment chaina’ was an expected demand made by Bengali workers, against the frequent use of casual labour by the mill, but ‘randibaji chaina’ was a more complex demand, raising the question of whether workers were referring to prostitution in a literal or metaphorical sense. Was this a demand against labouring women at the mill or in the neighbourhood? Samita Sen's work has highlighted the moralizing discourse around working women in the mills and working-class areas of colonial Calcutta in the late nineteenth century, and how the diverse marital, sexual, living, and work arrangements of these women were equated with prostitution; lodging housekeepers, washerwomen, barbers, sweepers, and midwives were reported as prostitutes.Footnote ¹²⁰ Little is known about the composition of the workforce in terms of age or sex at the Adamjee and Karnaphuli Mills or the presence and role of women in these working-class neighbourhoods. However, the University of Dhaka report found that many of the young men working in semi-urban or urban areas made use of the readily available drinking dens and brothels.Footnote ¹²¹ Thus, the demand could have emanated from conservative reactions to the increasing visibility of women in public spaces as a result of urbanization and industrialization, as well as perhaps to their fear of Biharis ‘using’ and ‘dishonouring’ local women.Footnote ¹²²

On the other hand, it may have been a call of solidarity for workers who fell prey to the more rapacious demands of those in power. The women and young boys hired as contract workers by the majhis were often subjected to sexual violence at the hands of thikadars and police:

It is also possible that the ‘randibaji chaina’ referred more specifically to how Bengali workers felt about the worth of their labour and consequent emasculation of their bodies. The Bengali durwan, for example, grew physically weaker as a result of low pay and poor housing and sanitary conditions. This was used to justify the higher pay of the Pathan durwans, who were paid more for their heavily built physique. A management that differentiated between the bodies of workers was seen as illegitimate in the Pakistan that the workers envisioned.

The aftermath

As the news of the Karnaphuli and Adamjee riots spread, the business community and government grew more panicked and fearful about the possible flight of capital. However, for all the talk of sabotage, production was resumed at both mills quite soon after the riots. By the first week of April 1954, most of the European technicians had returned to Karnaphuli, workers had been screened and issued passes, machines were cleaned, and work had started. The Adamjee Jute Mill also began operations again within two weeks of the riots. During the riots, as workers in both mills had weaved in and out of the buildings, wielding weapons of various sorts, smashing windows, turning over furniture, burning down homes, and injuring and killing people, not a single loom or papermaking machine had been touched.Footnote ¹²⁴

That there was no wrecking of machinery, even by accident, implies that this was a more selective kind of sabotage—one that allowed the workers to continue working once their demands were met. The fact that machines were the only things left intact during the riots appears to be a pointed statement, made by all those involved, that machines were not the problem. George Lamb, who was employed by the contractor Mr Hanson at the Karnaphuli Mill, described how earth-diggers had come into the stores, carefully laid down their earth-cutting equipment, picked up spare pieces of iron, and headed towards the mill.Footnote ¹²⁵ The machines were seen as an integral part of their labour and of their future as labourers. As the rioters shouted ‘Allahu Akbar’ and ‘Naraya Takbir’, their actions were attempts to recover sovereignty over other elements of their labouring existence: in particular, over the restrictions imposed on them by the management, which the workers understood as weakening their new position in Pakistan—that of strong workers labouring over strong machines. As the populace of the East Wing decided on the Pakistan they wanted via the ballot box, workers were also choosing their own fate.

The Pakistani state, which had made a significant investment in the industrialization of East Pakistan, as an exhibition of their power and progress in the club of independent, postcolonial, and free nations, had not thought enough about the bodies that toiled within these mills and on those machines. The response of Central government to the riots was to humiliate those involved. The Communist Party was banned in both wings of the country, trade-union organizations were disbanded, and hundreds of trade-union leaders, communists, and other left-leaning political leaders were jailed. In East Pakistan, the East Pakistan Rifles moved into the industrial areas straight after the riots and Karnaphuli Paper Mill and Adamjee Jute Mill were listed in the Dhaka Gazette from 1 July as ‘protected areas’.Footnote ¹²⁶ Workers at the Adamjee Jute Mill and Karnaphuli Paper Mill found themselves incarcerated in open cages and gazed upon by the public as ‘criminals’ as they awaited their release or transfer to closed prisons.Footnote ¹²⁷ Screening boards came into operation to weed out ‘undesirable’ workers, leading to hundreds of workers being retrenched.Footnote ¹²⁸ The local working-class neighbourhoods and bazaars lost the buzz and activity of labour.

In September 1957, as the Adamjee Jute Mill court case concluded, with many of the workers on trial being exonerated, a rather unnoticed report was also published: the Dacca-Narayanganj Police Committee Report. The report contained the initial proposals for a permanent industrial police force and intelligence agency to ensure prompt and effective action for the ‘maintenance of law and order’ and vigilance towards ‘communist infiltration into the labour force’.Footnote ¹²⁹ The restoration of parliamentary rule by 1957 did not lead to any corresponding increase in faith in workers as important citizens of Pakistan.