Region, religion, and locality: revisiting Punjab politics and the Unionist Party, 1923–1947

Abstract

This article revisits a series of articles published in the early 1980s on Punjab politics in the pre-partition decade that grew out of my thesis ‘The Growth of the Muslim League in the Punjab’ which was supervised by Francis Robinson. It initially locates the publications in the existing literature, before assessing the extent to which subsequent research has enriched academic understanding. The article concludes with some brief reflections on the conceptualisation of the Unionist Party's power-sharing as a form of consociationalism.

During the early 1980s, I published a series of articles on Punjab politics in the pre-partition decade which grew out of the thesis Francis Robinson had supervised, entitled ‘The Growth of the Muslim League in the Punjab’.¹ Earlier, my unpublished research had made a small contribution to his celebrated debate with Paul Brass on social mobilisation and Muslim separatist politics.² My thesis coincided with the switch from concern with the Muslim League development in its heartland in the United Provinces (UP) to the future Pakistan regions. Jinnah had indeed termed Punjab the ‘cornerstone’ of Pakistan, yet its regional politics institutionalised in the cross-community approach of the Unionist Party problematised the Muslim League's Two Nation theory.

This article reassesses these early studies in the light of the rich scholarship that has been subsequently produced relating to religion, region, and identity at the end of the British Indian empire. The opening sections highlight the main themes of my 1980s work after locating it in the pre-existing literature. The analysis then turns to subsequent publications that have deepened understandings of political identity and religious nationalism, and the extent to which Unionist rule shared features of consociational power-sharing arrangements such as segmented autonomy and religious proportionality in government employment. We turn first, however, to a brief description of the Unionist Party's foundation and development.

The foundation of the Unionist Party

The Unionist Party emerged as a formal grouping in the Punjab Legislative Council in 1923. From the outset it cut across ‘community’ interests. It drew support from large Muslim landowners in what became Pakistan Punjab and Hindu peasant proprietors in what is now the Indian state of Haryana. The Hindu Jat leader Chhotu Ram, who came from a peasant farming background, played a leading role in the party's development. Early in January 1926, he set out the agenda of the Unionist party as follows:

The Party (Unionist) seeks to unite Hindus, Mohammadans and Sikhs on the basis of their common secular and economic interests which are the only true foundations on which the edifice of lasting political parties can be raised. The Party offers a common platform to all who desire to see the end of communal strife in the province. We are the friends of the poor, the supporters of the weak, helpers of the down-trodden, the advocates of the voiceless, and the champions of the exploited classes of the Punjab.³

From a rather uncertain beginning, when in 1926 the Punjab governor Hailey removed Chhotu Ram as minister of education in favour of the urban Hindu leader Manohar Lal, the Unionist Party emerged as the dominant force in Punjab, locking both Congress and the Muslim League out of power until the eve of the British departure. The Unionist Party's foundational objectives were: a) to attain dominion status within the British Commonwealth by constitutional means; b) to demonstrate that Indians are capable of assuming increasing responsibilities of self-government; c) to provide equal opportunities for all, and special government assistance to backward classes and rural areas; d) to secure a fair distribution of taxes between agricultural and urban areas; e) to check the exploitation of economically backward classes by economically dominant classes; f) to provide support to indigenous industry; and g) to encourage and undertake social measures such as literacy, education, and suppression of corruption.⁴

The Unionists’ main platform in the depression-hit 1930s was to close loopholes in the Alienation of Land Act legislation which protected indebted agriculturalists from exploitation by moneylenders. The beginning of provincial autonomy in 1937 paved the way for the so-called ‘Golden’ agrarian acts.⁵ Prime Minister Sikander Hayat Khan and other ministers, including his future successor Khizr Tiwana, arrived by train in Lyallpur on the evening of 3 September 1938 to attend a huge meeting to celebrate the Golden acts. The crowded station was decorated with Union Jacks and bunting, as befitted the Unionist Party's close ties with the colonial administration.

The Unionist ascendency seemed unchallengeable because of its association with agricultural interests. Congress could not risk alienating its base among the commercial classes by supporting Punjab's farmers. The Muslim League was an urban party which had just one assembly member, Malik Barkat Ali, following the 1937 provincial elections. For much of the 1920s and 1930s, it had competed with other parties such as the Ahrars and Khaksars for the support of lower and middle-class Muslim town dwellers.⁶ The Ahrars lost support in the wake of the 1935 Shahidgunj Mosque dispute.⁷ Nonetheless, the Punjab Muslim League had to await the outbreak of the Second World before its fortunes improved.

The Unionist Party, Punjab politics, and partition in the early literature

Before the beginning of the 1980s, little had been written about Punjab politics in the pre-partition decade. The tussle for power between the Unionist Party and the Muslim League was confined to a footnote of history. Pakistan's creation was explored primarily through works on All-India developments, and on the roles played by leading national figures.⁸ Indian official historiography maintained that British ‘divide and rule’ policies culminated in partition.⁹ Pakistani writing focused on Jinnah's leadership role and on the Two Nation theory. The fulfilment of a separate state was a ‘natural’ expression of the distinctive Indian Muslim social order. Western scholars such as Peter Hardy did look beyond All-India developments, but their work lacked a fine-grained analysis. Hardy suggested that the Muslim League succeeded in defeating the Unionist Party in the key Punjab region by making an appeal over the heads of the professional politicians, but he did not elaborate on this.¹⁰

Francis Robinson expertly explored Muslim political developments in the UP, which provided the leadership with separatist demands.¹¹ His research extended the approach of the Cambridge School of Indian History, signalled in the seminal collection by John Gallagher, Gordon Johnson, and Anil Seal, Locality, Province and Nation.¹² The Cambridge School saw elite self-interest, rather than ideas, as driving nationalism. It also focused on the colonial state's role in framing politics. Francis Robinson modified this approach by pointing out the deep cultural and religious roots of Muslim separatism in his debate with Paul Brass.¹³

Academic attention shifted from UP to the Muslim majority areas which were vital for the successful achievement of separatist demands. My work on Punjab was thus part of a turn to the region and locality in understanding the creation of Pakistan.¹⁴ The emphasis in Pakistani historiography on All-India developments meant that little attention had been devoted to the province. The existing literature comprised primarily some memoirs, collections of speeches, and statements of leading Punjabi Muslim politicians, along with readings on the history of Punjab.¹⁵ The Unionist Party's history was absent, apart from Imran Ali's little-known work, Punjab Politics in the Decade before Partition.¹⁶ This pioneering study introduced such important political developments as the pact in 1937 between Jinnah and the Unionist Party prime minister Sikander Hayat Khan.¹⁷ Two other important early texts were, first, Azim Husain's 1946 biography of his father Mian Fazl-i-Husain who had founded the Unionist Party in 1923. Fazl-i-Husain was a lawyer connected with both the Congress until 1920 and the Muslim League.¹⁸ Secondly, in 1974 Craig Baxter published an important piece on the continuing influence of Punjabi landed families in the electoral politics of contemporary Pakistan.¹⁹

Little had been written about the Punjab's political economy at the time of the Unionist Party's foundation.²⁰ Yet the colonial state's agrarian policies and establishment of the Indian Army's headquarters in the province formed a vital backdrop. Gerald Barrier's insights, however, provided a fruitful avenue for future research. He argued that colonial policy in Punjab, despite dissenting voices, abandoned economic laissez faire policies to protect the interests of key local intermediaries.²¹ The 1900 Punjab Alienation of Land Act symbolised this policy.

It is now well established that the Punjab Alienation of Land Act formed the lynchpin of the Unionist political system that emerged after the 1918 Government of India Act. The Alienation of Land Act divided the population between designated ‘agriculturalist’ and ‘non-agriculturalist’ tribes. This cut across political identity around the basis of religion. The legislation halted the growing expropriation of cultivators caught in a rising tide of indebtedness. The Unionist Party became a vehicle for the ‘agriculturalists’ which, as we have noted, included not only large Muslim landowners but also small-scale Hindu peasant proprietors. There was also a significant Sikh landowning class in the central Punjab districts. Sikh landownership also increased in the largely Muslim southwestern Punjab through grants to ex-servicemen and Jat cultivators in the canal colonies. Irrigation in this area transformed over six million acres of wasteland populated by nomadic herders into the richest farming area in India.²² However, the rural Sikh community was never as fully integrated into the Unionist system as the landholding Muslims or Hindus. This resulted from the twin legacies of the Ghadr movement and the Gurdwara Reform Movement.²³

The limited material on colonial governance and its framework for Unionist politics was matched by the paucity of work on religious nationalism, and the latter's growing challenge to the Unionist Party's mediatory system of politics. Within the Sikh community this came from the Akali Dal. In 1974, Kailash Chander Gulati had published a collection of essays on the Akalis.²⁴ Mohinder Singh's more systematic work appeared four years later.²⁵ The conservative Chief Khalsa Diwan, which worked in coalition with the Unionist Party, had to wait until 1981 before it was studied in detail.²⁶ Accounts of Punjabi Hinduism and politics were especially limited. The notable exception was K. W. Jones’s work Arya Dharm: Hindu Consciousness in Nineteenth Century Punjab.²⁷ It linked Hindu religious reform with the growing influence of the commercial castes who profited from the economic opportunities brought by colonial rule. Members of the reformist Arya Samaj were influential in both the Punjab Congress and the Hindu Mahasabha. Jones also examined Arya-Sikh relations in the context of Kahan Singh's celebrated tract Ham Hindu Nahin (We are not Hindus).²⁸ There were also studies of the Brahmo Samaj, which predated the Arya Samaj, but never outlived its Bengali base, the Punjab Congress, and some biographies of the Hindu Jat leader Chhotu Ram who co-founded the Unionist Party.²⁹

Contributions by Barrier, Baxter, Heeger, and Oren pointed to further avenues of research.³⁰ Oren's work proved especially useful with respect to Sikh-Unionist Party relations. He revealed the backdrop to the 1942 Pact between the Sikh politician Baldev Singh and the Unionist premier Sikander. The Akali Dal had previously opposed the Unionist Party because of its loyalty to the Raj and preferential treatment of Muslims in terms of educational access and administrative appointments. However, it saw the opportunity for increased wartime military recruitment as providing leverage for the community in future All-India constitutional negotiations. At this stage the Akalis also saw power-sharing in a downsized Punjab province as crucial for safeguarding Sikh interests threatened by the Pakistan demand.³¹ The twists and turns in policies—swerving from what might be termed ‘voluntary consociationalism’ to the eventual demand for Khalistan—have become a key focus in recent historiography relating to Sikh politics at the end of empire.³²

Themes in ‘The 1946 Punjab elections’ and ‘Deserted collaborators’

My early publications examined how Punjab politics operated in the late colonial era. The Unionist Party's success in the Muslim West Punjab areas in 1937 had been based on the support of the leading landowners and pirs.³³ By the time of the 1946 polls, which the Muslim League had presented as a referendum on Pakistan, many of the most influential rural powerholders had joined its ranks.³⁴ Landholders, clan heads, and pirs mobilised their followers to vote for the League.³⁵

Access to the Muslim League records in the Archives of the Freedom Movement at Karachi University and to interviews and newspaper sources underpinned the research. The correspondence of the Punjab Muslim League with Jinnah during the years 1937–1946 contained in the Shamsul Hasan Collection, held by his son Khalid Shamsul Hasan in Karachi, also provided a vital source. In 2009, Waheed Ahmad published a selection of these records, which have now been donated to the Cabinet Division, Government of Pakistan, covering the years 1940–1947.³⁶ The resulting understanding of the importance of the leading Punjabi pirs’ switching of allegiance from the Unionist Party to the Muslim League emerged independently of David Gilmartin's early research on the role of Sufi networks in mobilising support for the Muslim League.³⁷

My work linked the pirs’ desertion of the Unionist Party with the Second World War's wider impact on the Party's rural constituency.³⁸ The War resulted in unpopularity for the government, with its inflation, grain requisitioning, and heavy military recruitment demands. The Punjab Muslim League gave the situation a communal twist by highlighting the profiteering of Hindu contractors. Six resolutions were passed complaining about cloth distribution in Lahore following a series of meetings in its leading mosques.³⁹ The urban Muslims’ long-established resentment at Hindu economic domination gained traction with their co-religionist cultivators.⁴⁰ The Unionist Party's position was also weakened by the factional realignments that followed the deaths of its leaders Sikander Hayat in December 1942 and Chhotu Ram in January 1945.⁴¹ Khizr Tiwana, who succeeded Sikander as Punjab's prime minister, faced growing challenges to his leadership as Jinnah put greater pressure on his cross-communal approach to politics.

With each passing year, the Punjabi landowners’ traditional loyalism to the Raj appeared to be outliving its usefulness. The July 1945 Simla Conference was a key turning point. Jinnah blocked Muslim Unionist representation on an extended Viceroy's Executive Council. Two months later the collapse of the Khizr-Noon Talks ended the hopes of a Unionist-League rapprochement in advance of provincial elections.⁴² Unionist members defected in droves. In the two districts of Jhang and Sheikhupura, for example, all seven of the Muslim Unionist members of the Legislative Assembly crossed over to the Muslim League.⁴³ The party's leaders were left to reflect bitterly on how the viceroy had let them down in his handling of Jinnah at Simla.⁴⁴

While the thrust of my research was on the tussle between the Muslim League and the Muslim Unionists, it also reflected on the erosion of Hindu Jat support for the party. For over two decades, the Congress's lack of Hindu peasant support had reduced it to competing, not always successfully, with the Hindu Mahasabha for the votes of Hindus in Punjab's towns. The analysis in ‘Deserted collaborators’ understood the Hindu Jat switch to Congress in 1946 as a temporary response occasioned, first, by the polarisation of politics at the All-India level, and, second, that it was linked to the factional weaknesses which followed the death of Chhotu Ram who had dominated the community's politics.⁴⁵

Subsequent studies

David Gilmartin's seminal work Empire and Islam: Punjab and the Making of Pakistan was published in 1988.⁴⁶ It too emphasised the role of the colonial state in bolstering the local Muslim landholding elites⁴⁷ who had coalesced in the Unionist Party before deserting it for the Muslim League.⁴⁸ Both parties ‘built their rural support primarily within the same local factional structures that had characterised rural politics since 1937’.⁴⁹

Empire and Islam also took our understanding in a different direction. The volume pointed out the contradictions between rural expressions of political order arising from the political hierarchies entrenched by the Punjab colonial state and a cultural definition of power based on Islamic ideals. The Pakistan campaign starkly contrasted these commitments.⁵⁰ The Muslim League, however, lacked the rural base to destroy the traditional method of conducting politics. The adaptation to the rural political milieu ‘compromised the moral foundation of Pakistan’.⁵¹ Nonetheless, the paradoxical mix of religious rhetoric and manipulation of kinship networks and landlord politics gave the Muslim League campaign its potency. ‘The triumph of the Muslim League,’ Gilmartin has argued, ‘thus simultaneously defined a new symbolic foundation for the state and affirmed the same structure of mediation on which the colonial state had rested.’⁵² The resulting underlying tensions have compounded the difficult processes of nation-building and state-construction in Pakistan.⁵³ The same elite families who had been active in the landlord politics of the Raj adapted themselves to the post-colonial political landscape.⁵⁴

David Gilmartin's arguments have influenced a later generation of South Asian scholars. The ways in which the Pakistan idea was debated and secured popular support have been taken up, for example, in Venkat Dhulipala's recent research.⁵⁵ Dhulipala, drawing on Gilmartin's attention to the significance of symbolic Islamic foundations for political order, points out that the impact of religious symbols and authority intensified when they worked alongside an appeal for a religious homeland, however vaguely defined. The correspondence below illustrates the Punjab Muslim League's realisation of the importance of this linkage. The Punjab Muslim League requested pirs and Sufis to:

Help the Muslim nation of India in its present life and death struggle, by their sincere prayers and by exhorting their followers to sacrifice their all in the cause of the attainment of a free and independent Muslim India (italics added).⁵⁶

Subsequent studies on the Hindu Jats have also enriched the understanding of the community's relationship with the Unionist Party. Nonica Datta's 1999 monograph not only revealed more about Chhotu Ram's personality and world view, but also explored the process of community identity formation and the important role played by qaumi (community) narratives.⁵⁷ It is particularly useful to read Datta's work alongside the article Prem Chowdhury published earlier in Modern Asian Studies which focuses on the 1946 Punjab elections. Chowdhury revealed that there were similarities between the Muslim League's call for Pakistan and the Congress's for azadi as solutions to wartime consumer shortages.⁵⁸ She also pointed out the impact among servicemen that was generated by the Congress support for the INA men.⁵⁹ The exploitation of grievances in a region that had been hit by famine as recently as 1938–1941 enabled Congress, despite its urban roots, to make a temporary breakthrough.

We also now know far more than in the 1980s about the Punjabi Sikh community and its ambiguous relationship with the Unionist Party and Congress. Chhanda Chatterji points to the complexities of Sikh responses to the constitutional and political problems thrown up by the Second World War.⁶⁰ This point is taken up with more analytical depth in Singh and Shani's Sikh Nationalism.⁶¹ Torn between leftist, communitarian, and nationalist politics, Sikh leaders pursued contradictory policies in the face of what community leaders saw as the existential Pakistan threat. Ultimately, when the Boundary Award left the community cut off from rich farmlands and historic shrines and gurdwaras, violence was the only means by which to assert Sikh interests.⁶²

What is important, however, is not the notion simplistically put forward by some Pakistani writers that the Akali Dal Sikh leaders conspired against their own interests by not reconciling with the Muslim League.⁶³ This approach in any case ignores the extent to which Jinnah's assurances may have lacked credibility in a mirror image of the Congress's trust deficit with the Indian Muslim community. Rather, it is the insight that the end of empire reduced the space for all parties that pursued interests that complicated the British drive to achieve an All-India settlement. The Sikhs had their British sympathisers within Punjab, including the final governor, Sir Evan Jenkins, but their demands were viewed as a ‘small nuisance’ in the official mindset.⁶⁴ The fates of the Unionists and the Akali Dal were sealed not in Punjab but in New Delhi. They were not alone in being bypassed. Ultimately, even Jinnah found that he had to accept a ‘moth-eaten’ Pakistan.

Studies of British Punjab gathered pace during the 1990s. These provide further and more detailed context for the Unionist political system. Colonial Punjab was understood as a garrison state that had profound legacies for post-colonial Pakistan's democratic consolidation.⁶⁵ Imran Ali earlier theorised Punjab's political economy in terms of ‘stunted nationalism’ arising from the stagnation of productive forces and under-development.⁶⁶ Other scholars took up this theme in a revisionist understanding of the notion of a contented peasantry resulting from a benign colonial administration in the Punjab region.⁶⁷ This points to the fact that we should not overlook that the Unionist Party faced challenges not only from religious nationalism, but also from agrarian radicalism. Raghuvendra Tanwar has pointedly declared that ‘The Unionist Party was born at the top and had little to do in its origin with the “pains of labour and the ground realities”.’⁶⁸ In colonial Punjab around 8 per cent of the population owned 45 per cent of the land.

The emergence of politicised religious Punjabi identities, alongside fluid syncretic social formations has been a major recent research focus of the flourishing fields of Sikh and Punjab studies.⁶⁹ Issues of identity, and political expression, have all been discussed in Malhotra's and Mir's insightful collection, Punjab Reconsidered.⁷⁰ Despite its multi-disciplinary approach to the concept of Punjabiyat (Punjabiness), many of the contributions raise issues present in colonial-era works by Temple and O'Brien that described instances of shared religious expression.⁷¹ Navtej Purewal has looked to the Mughal era for earlier insights into traditions of musical devotion which question notions of entrenched Muslim/Sikh conflict.⁷² Instead of monolithic Hindu and Muslim communities being pitted against each other, Neeti Nair, in her seminal study Changing Homelands, reveals that there was a struggle for power to alter or maintain the status quo involving the dominant Khatri and Arora Hindu castes, Muslim Sheikhs, and Arains.⁷³ Nair argues that it was not religious difference but ‘shockingly petty political differences’ that caused partition in the final instance.⁷⁴ This argument can be usefully read alongside Farina Mir's argument in The Social Space of Language that conventional histories have overemphasised the transforming impact of religious reform in Punjab.⁷⁵

This recent scholarship contrasts with the limited new insights deriving from research on the Unionist Party within Pakistan. Samina Yasmeen has produced a series of articles arising from her 2006 thesis, ‘Communal Politics in Punjab (1925–1947)’.⁷⁶ Her colleague Kishwar Sultana at the Allama Iqbal Open University produced a short piece in 2011 on the Jinnah-Sikander Pact.⁷⁷ Unsurprisingly, more convincing work has been produced on Mian Fazl-i-Husain, a figure who is less problematic for nationalist historiography than Khizr Tiwana.⁷⁸ Tahir Kamran drew on the Mountbatten papers held at the University of Southampton to produce the most insightful study.⁷⁹ His analysis of Punjab politics during March–August 1947 pointed out both the mistrust that prevented a Sikh-Muslim League agreement following Khizr's resignation and the role played in this by the Muslim League High Command. During negotiations with Jenkins, the Punjab governor, Swaran Singh, the former development minister, bluntly informed him that ‘The Sikhs had no plan of being treated as serfs under Muslim Masters and felt they were strong enough to defend themselves.’⁸⁰ The leadership of the Punjab Muslim League had its hands tied in any case as Jinnah was not prepared to risk negotiations that could jeopardise the Pakistan demand. Peace in Punjab had to wait until an agreement had been reached at the centre.⁸¹ Yaqoob Bangash in a forthcoming publication takes scholarship in a new direction with his study of the role of Anglo-Indians in the Unionist Party and their manoeuvrings in Punjab politics in 1946–1947.⁸²

Otherwise, there have been Pakistan-based studies that reproduce contemporary Muslim League criticism of the British and Khizr Tiwana's actions after the 1946 polls.⁸³ Maqbool Ahmad Awan drew extensively on my biography of Khizr Tiwana, published in 1996 and which was translated into Urdu.⁸⁴ His article did not however add anything new.⁸⁵ Similarly, general studies of the Unionist Party reproduced the empirical findings of the 1980s. All that was new was an accompanying nationalist slant.⁸⁶ Iftikhar Malik's mid-1990s examination of Muslim politics was more reflective in examining what he saw as the inherent contradictions within the Unionist Party's espousing of a linguistically and economically undifferentiated ‘ruralist’ Punjabi identity.⁸⁷

During the past three decades, there have been numerous works on late colonial India's provincial politics that offer comparative insights into Punjab. The Muslim League branches in Sindh and the North West Frontier Province clearly adopted similar methods of political mobilisation to those in Punjab.⁸⁸ The complex and confused non-Muslim responses to the Pakistan demand emerges clearly in studies of Bengal.⁸⁹ Partha Chatterjee's comment on the Bengali Hindu elites’ responses being ‘contingent and taken on the assumption that Pakistan was in the offing’ could as well have been made with respect to the Akalis’ shifting stance in Punjab.⁹⁰ Studies of political developments in Bengal have also reinforced the understanding that the demise of the All-India parties’ rivals was neither an inevitable, nor a linear, process. There were contingencies and twists and turns along the way that culminated in the ending of fluid patterns of politics. Ultimately, the Unionist Party faced a similarly unfavourable environment to Huq's Krishak Praja Party in Bengal.⁹¹ The Unionist Party clung onto office longer because its power was more entrenched in both government and society. It thus provides a potentially more viable model for ethno-religious conflict resolution in contemporary South Asia.

Punjab unionism as a model of consociational democracy

In ‘Back to the Future’, published in the mid-1990s, I returned to a consideration of the Unionist Party, but in terms of the model of consociational democracy.⁹² Consociationalism is especially associated with Arend Lijphart's scholarship.⁹³ The definition and central arguments regarding consociationalism have been vigorously debated.⁹⁴ According to its theory, the maintenance of stable democracy in states riven by linguistic, ethnic, and religious cleavages depends on elite accommodation. This is institutionalised in such devices as grand coalition, mutual veto, proportionality, and segmental autonomy. Lijphart sees consociationalism as a normative as well as an empirical model. In his work on power-sharing in 1980s Northern Ireland, Brendan O'Leary differentiated between externally politically engineered systems, which he termed ‘coercive consociationalism’, and the more sustainable voluntary type. Can the Unionist experience of power-sharing, if it was a form of consociationalism at all, be better understood as an example of ‘coercive’ rather than ‘voluntary consociationalism’?

The Unionist Party's two decades of dominance undoubtedly indicates a large degree of success for intercommunal political accommodation, despite the increasing role of religion as political identity. The Unionist Party captured over 68 per cent of the seats in the 1937 provincial elections. Even in 1946, when the Unionists are traditionally depicted as being ‘crushed’ by the Muslim League, they still received over 20 per cent of the total votes and were able to exclude the Muslim League from office. Political accommodation could thus pay electoral dividends even in a society like Punjab which, by the end of colonial rule, was witnessing a high degree of polarisation along religious lines.

Moreover, the Unionist Party successfully operated such consociational devices as proportionality in public service recruitment and decision-making. The June 1942 Pact between Sikander and Baldev Singh provides a classic example of consociationalism at work. It increased Sikh representation in those government departments in which it fell below the fixed communal proportion of 20 per cent and endorsed Sikh ‘segmental autonomy’ in giving jathka meat the same status as halal meat.

When Sikander formed his cabinet in 1937, it bore the hallmarks of a grand coalition in keeping with classical consociational theory. The Unionist Party itself contained, as we have seen, a 13-strong Hindu Jat wing led by Chhotu Ram. Its Hindu and Sikh coalition partners, the Hindu Election Board and the Khalsa National Party, together numbered 25 members. In all the government accounted for 120 out of the 175 Assembly members drawn from all the religious communities. During the Second World War, the Unionists and their traditional rivals the Akali Dal drew closer together. Baldev Singh, who negotiated on behalf of the Sikhs in 1942, owed his seat in the Cabinet to Akali support, although he was the leader of the United Punjab Party which contained some Muslim and Hindu members as well as Sikhs.

As in other consociational systems, the Unionist Party's institutional arrangements were built on elite-level cooperation. The formation of the Unionist Party in 1923 resulted from the calculations of the Muslim politician Mian Fazl-i-Husain and the Hindu Jat leader Chhotu Ram that cross-communal cooperation held out the best prospects for power and their communities’ advancement under the political conditions established by the British. The Unionist Party's successive Muslim leaders Sikander and Khizr continued to hold this view, despite Jinnah's increasing drive to consolidate the Muslim community behind support for the Muslim League.⁹⁵

Linked with the point above, the Unionist case appears to support consociational theory's argument that political leaders can act autonomously of social forces and communal cleavages. Political scientists use the term ‘structured elite predominance’. Khizr remained committed to the Unionist cause in the face of an increasingly personalised campaign that portrayed him as a ‘traitor’ to Islam.⁹⁶ Some Pakistani authors have claimed that he lacked the imagination to adjust to the new political circumstances arising from the Muslim League's growing popularity.⁹⁷ It could equally be argued that the powerful lure of office and the desire to economically benefit the agriculturalist Muslims’ interests, rather than stupidity, prompted his action. From the time of Khizr's public breach with Jinnah in 1944, a younger Muslim rural counter-elite emerged that was more in tune with shifting public sentiment in favour of the Muslim League's approach. Khizr's ability to act independently of a growing political culture of communal cleavages became increasingly compromised. This brings us back to one of the key discussions in Francis Robinson's debate with Paul Brass regarding Muslim separatism in the UP. As British rule drew to its close, Jenkins termed Khizr ‘a general without an army’.⁹⁸ Yet until his resignation on 2 March 1947, the Unionist Coalition Government maintained peace in the Punjab, while communal conflict racked such areas as Calcutta, Noakhali, and Bihar. ⁹⁹

Political developments in Punjab during the Unionist period have at least partial similarities to the consociational model. It could be argued that the Unionist Party's position in Punjab politics was more analogous to the Congress one-party dominant system of the 1950s than to a consociational grand coalition, although it operated along those lines internally. Moreover, questions can be raised about whether we should talk of the rural grouping as a ‘modern’ party with a coherent ideology and policy aims. When the Unionist Party was founded, all 32 of its Muslim members came from leading landlord or pir families. Was the Unionist ‘Party’ for much of its history more a shifting combination of local elites whose influence was independent of party structure?

Turning to Brendan O'Leary's conceptualisation, the political engineering of the colonial state is clearly present in establishing the framework for Unionist rule. The franchise remained very limited (only 13 per cent of the population could vote). Its weighting meant that only 25 per cent of the electorate were non-agriculturalists. The distribution of seats meant that no single religious community could dominate.¹⁰⁰ Punjab Unionist politics could thus be seen as an example of ‘coercive consociationalism’. Aside from the internal vulnerabilities attendant on this type of power-sharing,¹⁰¹ the British establishment of this political framework was primarily to serve imperial interests. Punjabi interests were secondary. The British willingness to override Punjabi interests in the quest for an All-India settlement in 1946–1947 was an important factor in the overwhelming of the Unionist influence, along with the rise of religious nationalism. Until the eve of independence, British officials in Punjab and a dwindling number of Unionist leaders were resolutely in favour of a united Punjab because they feared the chaos of partition. Here one must be careful not to see the Unionist Party consociationalism as solely a colonial construct, or a ‘collusion of class interests’.

Modern scholarship has revealed that the people of Punjab knew how to live with difference.¹⁰² They did not require colonial prompting to look beyond communitarian identities; the langar open to all on the Tiwana Kalra estate symbolised this as did the mandir which Umar Hayat Khan Tiwana had constructed for his Hindu workers. Most of Kalra's key administrators were Hindus. Each winter, in a hectic three of four days, all communities dammed the Jhelum with sandbags to ensure the water supply to the private canal system.¹⁰³ The Tiwana lancers, the family regiment, comprised Hindu, Muslim, and Sikh squadrons.

Private correspondence in the period 1925–1943 between Sardar Sadhu Singh Tiwana, a member of the Sikh Tiwana family in Patiala, reveals the persistence of family cross-community ties in late colonial Punjab.¹⁰⁴ These letters were given to me by Justice Kulwant Singh and had not been previously available to historians.

On 11 March 1937, Khizr wrote from his Beadon Road Lahore residence thanking Sardar Sadhu Singh Tiwana ‘for his kind letter of congratulations’ on his appointment as minister of public works and local self government.¹⁰⁵ Sardar Tiwana continued the correspondence on 21 July, accompanying his letter with a consignment of mangoes. During 1937 and 1938 he wrote several letters to Umar regarding his hopes for promotion in the service of Patiala state, terming the Tiwana chief, ‘the anchor-sheet on whom the safety of my voyage of service in the state depends’.¹⁰⁶ Umar contacted Sir Liaqat Hyat Khan, the Patiala prime minister, and in September 1938, his Sikh relative was able to achieve his cherished ambition as a District and Sessions judge.¹⁰⁷ The correspondence with Umar did not end then, as at the beginning of 1943 he wrote congratulating him on his son's appointment as prime minister. He concluded his letter with the words, ‘It is a matter of great pride for the Tiwana community as a whole to see the illustrious young man of this family to occupy this post of responsibility and prestige at so young an age. We the members of this family living in this part of the Punjab wish to pray to Almighty God to shower all his bounties upon him and grant him full success in his new career.’¹⁰⁸

Sardar Sadhu Singh Tiwana was not the only member of the Patiala branch of the family in contact with their Muslim kinfolk. In an interview in December 1993 Justice Kulwant Singh informed me that his father, who was a deputy superintendent of police at Bhatinda, met Umar on several occasions.¹⁰⁹ While the significance of such manifestations of mutual affection and regard should not be over-exaggerated because of their transactional undertones, they point to the persistence of a more complex Punjabi society than is often acknowledged when it is viewed through the prism of partition.

Conclusion

Seventy-five years after independence and partition, Punjab remains the cornerstone of Pakistan, but the nation still searches for the Islamic foundation for political order that will enable it to become a new Medina. Rural ‘electables’ operate in a world of duniya alternating between the biddings of rival parties and the ‘establishment’ in Pakistan's hybrid democracy.¹¹⁰ Nationalist historiography struggles to fit the Unionist Party experience within its narrative. In Haryana, the other main locus of Unionist power, ‘the Jat communities, continue to draw on their collective memories of being oppressed, exploited and subordinated by the so-called superior castes’.¹¹¹ The ‘Modi-wave’ which swept Manohar Lal Khattar to power in 2019 has receded in the wake of the farmers’ protests as Jat agrarian interests are privileged over the Hindu nationalist discourse in the state's politics.¹¹² Recent research on Sikh political identity has reiterated the significance of Punjabi Unionism as a ‘modern consociational arrangement’.¹¹³ One may conclude that while the Unionist Party may not form a normative model for consociationalism, it points to the existence of culturally and socially rooted plural political expressions in late colonial India that have been obscured by subsequent nation-building processes in the Punjab region.