Stay informed: Sign up for eNews Subscribe

The Social Space of Language Vernacular Culture in British Colonial Punjab

Read Chapter 1


Forging a Language Policy

The East India Company's half-century of vigorous territorial expansion in India began with the marquess of Wellesley's governor-generalship in 1798 and culminated in the annexation of the Punjab in 1849. The Punjab's Sikh kingdom of Lahore (map 3), established by Ranjit Singh in 1799, had proved a particularly capable adversary, and the Company conquered the region only through a combination of political intrigue and military might. The new colonial administrative unit that resulted from this conquest-Punjab province-encompassed all or parts of what had been the Mughal subas (provinces) of Lahore, Multan, and Kabul before they had been wrested by the Sikhs. Colonial Punjab as established in 1849 was thus composed of the area's five doabs, or inter-riverine tracts, the territory immediately to their east and west (the trans-Indus and the cis-Sutlej territories), and the frontier areas of Peshawar, Leia, and Hazara. In 1858 Delhi and its environs were added to the province, and in 1901, the frontier areas were separated to form the Northwest Frontier Province (map 4). As these transformations suggest, the Punjab has been variously constituted throughout the modern period. The waxing and waning of its administrative borders notwithstanding, the Punjab has a geographic-cultural core, as this book demonstrates, whether conceived as an axis connecting the region's major cities, Amritsar, Lahore, and Multan, or more broadly as the five doabs and the cis-Sutlej territory. [Place maps 3 and 4 near here.]

The annexation of the Punjab in 1849 has for historians of modern South Asia always marked an important moment in India's colonization, for with Punjab's inclusion Company territories spanned the length and breadth of the subcontinent, albeit with Indian-administered states interspersed. With the transition to Crown Rule in 1858, the map of India (or the political realities it represented) did not change dramatically, though the subcontinent was now refigured as British India and Native States. Whether focusing on the Company period or that of Crown Rule, generations of scholars have documented the impact of British colonialism on Indian political, economic, legal, social, and cultural life. Indeed, the Punjab provides especially fertile ground for such study. One can plausibly argue, for instance, that the effects of colonialism were more pronounced in the Punjab than elsewhere, since by the time of its annexation the Company was already well practiced in administering Indian territories to its own advantage. Additionally, the Company instituted a more authoritarian administrative structure in the Punjab, one that was largely retained after 1858. Historians have dubbed this the "Punjab school" of administration and have singled it out for its marked authoritarianism, paternalism, and Christian evangelism. The administrators of this Punjab school were primarily responsible for the province's vigorous incorporation into the greater British enterprise during the late nineteenth-century "high noon" of colonialism, when Indian state and society were unabashedly manipulated to Britain's economic and political ends. Together, these factors contributed to colonialism's tangible, indelible impact on Punjabi society.

Language is undoubtedly a critical arena for the operation of colonial power, and this was as true in colonial India as in other colonial contexts. C. A. Bayly has argued further that language is central to establishing colonial power. In his Empire and Information, a study of late eighteenth- and nineteenth-century India, Bayly eloquently argues that the colonial state's ability to access indigenous networks of information and adapt them to its own ends was critical to its success. This access rested on colonial officials' linguistic abilities in Persian, Sanskrit, and local vernacular languages. These were linguistic skills they did not by and large possess, however, forcing them to rely on native intermediaries and interpreters, a situation most administrators found disconcerting. Given the obvious advantages of their own linguistic competence over their reliance on intermediaries, Company officials took pains to learn, codify, and ultimately teach Indian classical and vernacular languages in colonial institutions, in both England and India. Bernard Cohn has persuasively argued that such institutions helped officials gain the "command of language" that was crucial to the consolidation of colonial power in India. If language (competency) was critical to the colonial enterprise, as Bayly and Cohn contend, then so was language policy-that is, which language(s) to adopt for administration. The history of language policy in colonial Punjab, where the language of administration was carefully considered for its ability to integrate the province into broader structures of colonial authority, bears out this important relationship between language and empire.

The policy established by the colonial state in Punjab designated Urdu as the language of provincial administration. This had a number of consequences, the most evident of which is that this designation-which initially applied only to the administration of revenue collection and justice-led to the adoption of Urdu in other branches of government. Less evident is the impact of colonial language policy beyond state arenas to influence Indian society in new and important ways. Most significantly, language policy had a decisive influence on literary production in the Punjab, on its print public sphere, and its print culture more broadly. That the state's language policy impacted print culture is perhaps not surprising given the role of colonial actors in promoting print in nineteenth-century India. As in many other parts of India, it was Christian missionaries who first introduced print in the Punjab (in the early nineteenth century). At first, Indians had exhibited little enthusiasm for the technology, and Indian publishing ventures were few and far between. However, once the region was annexed, its colonial administrators saw potential benefit in an active "native" press and helped to spur Indian publishing enterprises by promoting the establishment of presses and purchasing much of what they produced. This publishing industry would in twenty years become the foundation of an incipient indigenous public sphere, though this was surely an unintended effect of colonial policy and patronage. By the 1870s, newspapers were increasing in number and circulation, and book publishing was thriving, with thousands of volumes produced each year, representing together an array of Indian opinion.

Almost all of these newspapers, and certainly all the commercially viable ones, were published in the Urdu language, as were a majority of the books published in Punjab. Given India's robust indigenous literary traditions and colonial efforts to cultivate and modernize vernacular languages across India, it is no surprise that the principal language of publishing in Punjab was an Indian vernacular. But the dominance of Urdu requires explanation. Elsewhere in India, vernacular publishing reflected the principal vernacular language(s) of each province, but Urdu was not prominent in the Punjab. Most of the region's inhabitants spoke Punjabi, the main colloquial language of people from different class, caste, and religious backgrounds, and a language with a rich literary tradition. Urdu, by contrast, had no significant spoken or literary history in the Punjab prior to the establishment of the colonial state. In colonial Punjab, then, the almost complete dominance of Urdu in certain genres of literary production and in the print public sphere, and its preponderance in print culture more generally was an outcome of the state's language policy. Marking this considerable impact of colonialism on indigenous practice is not my chief aim, however.

The more significant aspect of the history of language, literary production, and print culture in late nineteenth- and early twentieth-century Punjab is not that of colonialism's impact, but rather of its limits. The "success" of the colonial state's language policy, documented in this chapter through Punjabis' adoption of Urdu as the language of certain literary genres and the print public sphere, was only partial. Thus, for example, at no point during its rule in Punjab could the colonial state govern without recourse to Punjabi, colloquially at least. This resilience of the Punjabi language, despite a colonial policy clearly aimed at replacing it with Urdu, marks an important limit to colonialism, as does Punjabi literary production, and Punjabi print and performance cultures. Indeed, this resilience-recovered in this chapter from the margins of the colonial archive-presents a first indication of the vitality of the Punjabi literary formation during the late nineteenth and early twentieth centuries.

Early Printing in Punjab

Before turning to language policy, it is helpful to examine the state of publishing in early and mid-nineteenth-century Punjab, since colonial policy had a decided impact in this arena. Print technology had been available in India as early as the sixteenth century, but it was not until the nineteenth century that printed texts were widely disseminated. Across India, Christian missionaries were among the first groups to use printing presses, since they saw publishing as an effective evangelizing tool. This was true in the Punjab as well, where the American Presbyterian Mission established the first printing press, in 1836 in Ludhiana. The press initially had only two type fonts: roman font, used primarily for Romance languages, and Indo-Persian, with which texts could be produced in Arabic and Persian, as well as in a number of Indian vernacular languages, including Urdu, Kashmiri, and Punjabi. With these two fonts, the mission published Christian scripture in English, Urdu, Persian, and "Indo-Roman" (Urdu in roman characters). Subsequently the mission designed two more fonts: Gurmukhi, used principally for Punjabi, and Devnagari, used for Sanskrit and Hindi, which it used beginning in 1838 to produce publications in Punjabi and Hindi. Choices about what languages the press should publish in, and what scripts to use, appear to have been grounded in missionary (and colonial) conceptions about the links between languages, scripts, and religious communities, discussed below. Be that as it may, the Ludhiana Mission Press is significant to the history of publishing in the Punjab because it was the region's first printing press and, perhaps more importantly, because of the sheer volume of materials it produced and distributed. Despite numerous setbacks, in its third year (1838) the Mission Press published seventy thousand volumes of twenty-four titles, comprising well over a million pages. In 1840 the press printed fewer volumes (just over thirty thousand), but these accounted for a total output of some two million pages. These were in English, Urdu (in Indo-Persian script), Hindi (in Devnagari script), Punjabi (in Gurmukhi script), and also Kashmiri (in Indo-Persian; due to a sizable migrant Kashmiri community in the Punjab).

The American Presbyterian Mission may have brought a new technology to the Punjab, but this did not spark a revolution in indigenous publishing there. It took almost fifteen years before an Indian-owned press was established, and then only in response to the provincial administration's invitation to do so. Hursookh Rai, an experienced printer from the Northwest Provinces (NWP), accepted the government's invitation and in early 1850 established the Kohinoor Press in Lahore. Rai then launched Kohinoor, an Urdu-language newspaper whose editorial slant was, not surprisingly, decidedly sympathetic to government concerns. The Kohinoor Press was not the first press the Company government had helped establish in the Punjab, however. As early as the 1840s the British Resident at Lahore facilitated and financed the city's first English-language press, the Chronicle Press. For the Chronicle Press, too, the Company had turned to an experienced printer, in this case Muhammad Azim, who had been associated with the Delhi Gazette. Once established, the press published The Lahore Chronicle, a journal meant explicitly to further British policy. Although its exact date of inception is unknown, the reminiscences of Lahore resident H. R. Goulding reveal that The Lahore Chronicle was available in the late 1840s.

Through such Company initiatives, an Indian-owned-and-operated press was slowly established in the Punjab, and after 1850 presses were launched with increasing frequency. Through the latter half of the nineteenth century, presses were founded in Lahore and Amritsar, and also in smaller cities such as Multan, Sialkot, Jhelum, and Rawalpindi. These engaged in two kinds of publishing: newspapers and periodicals, and books, both in a variety of languages.

Urdu publications dominated, however. According to a survey of late nineteenth-century Punjab newspapers, while periodicals were published in Urdu, English, Punjabi, Hindi, Persian, Arabic, and Sindhi, and some newspapers were composed in more than one language, Urdu was the predominant language of the press. In 1876, for example, of the thirteen most important newspapers and periodicals, seven were in Urdu, four in English, and two in Arabic. In 1883 Urdu was the language of eleven of the thirteen vernacular newspapers published in Lahore. By 1901, 186 vernacular newspapers and periodicals were being published in the province, of which 137 were in Urdu. Taking an aggregate view, of the 413 periodicals published in the Punjab between 1880 and 1905, 343, or about 82 percent, were in Urdu. While some of these Urdu newspapers were short-lived or had limited circulations (sometimes only a few hundred), others were among the Punjab's most important commercial newspapers. The most significant were the Akhbar-i Am, started in 1870, and the Paisa Akhbar, started in 1887. Both were published in the capital city of Lahore and both had healthy circulations; in 1903 the Paisa Akhbar's weekly edition reached 13,500, an astonishing number for the time, given that the 1901 census documented the literacy rate for Lahore district (with a population of approximately 1.162 million) as a mere 4.4 percent.

Alongside newspaper publishing in the Punjab, book publishing, which remained overwhelmingly lithographic rather than typographic until well into the twentieth century, also commenced in the late nineteenth century. Government records indicate that books published there were even more linguistically diverse than the newspapers. Between 1867 and 1896, publishers produced books in Arabic, Hindi, Kashmiri, Marwari, Pahari, Pashto, Persian, Punjabi, Sanskrit, Sindhi, and Urdu, with Urdu publications again outnumbering the others. Numbers give some indication of Urdu's dominance vis-à-vis other publication languages (table 1). [Place table 1 here.]

Although precise information about titles and content is unavailable, anecdotal evidence based on surviving texts suggests that Urdu publishing encompassed a range of genres and subjects, from poetry and fiction, to history, religion, and science.

The preponderance of the Urdu language in late nineteenth-century Punjab's newspaper and book trades is an anomaly given that Urdu was not widely spoken in the province. Few indigenous sources detail the Punjab's spoken languages, whether in precolonial or colonial times, but those few that do show that people there did not by and large speak Urdu. Colonial records for the late nineteenth century concur. Similarly, the Punjab does not appear to have been a prominent site of Urdu literary production. This begs two questions: what were the norms of spoken and literary language in precolonial Punjab? And why did Urdu come to dominate the latter arena, at least, in the late nineteenth century?

Precolonial Linguistic and Literary Practices

A sketch of precolonial language use and literary production in the Punjab provides an important context for understanding the changes wrought by colonialism in these arenas. While charting the latter is possible through the survival of manuscripts, charting the former is challenging since few sources of the kind later available for this purpose-ethnographic surveys and censuses, for example-exist for the Sikh period, or the Mughal and Sultanate periods before it. These difficulties notwithstanding, a composite picture suggests that there were a "diverse collection of languages, different languages for different people on different occasions," in David Lelyveld's words. Lelyveld was writing of the Mughal court, but his words resonate here as in precolonial Punjab, like in much of India historically, there were colloquial, liturgical, sacred, court, and literary languages, some of which overlapped and some of which did not.

The earliest indication of the language(s) spoken in the Punjab is derived by inference from the poetry of the Sufi saint Shaikh Farid (1173-1265). Farid's poetry is among the earliest in the Punjabi literary canon, and we can infer that it represented the contemporary vernacular language of Pakpattan in central Punjab, where he established a Sufi khanqah, or hospice. This inference is based on broader patterns of Sufi practice in South Asia, where Sufis are known to have composed poetry in local spoken languages as a means of disseminating Sufi ideas to a largely uneducated, lay, local populations. The next available indication of spoken language in Punjab-or its central districts-comes some centuries later from janam-sakhi literature, traditional biographies of Guru Nanak (1469-1539), Sikhism's founder and first guru. Composed from the sixteenth century on, in an anecdotal style, many janam-sakhis contain dialogue between individuals who speak in Punjabi (albeit what we would today consider medieval Punjabi), though Persian dialogue is occasionally recorded as well. The Dabistan-i Mazahib (School of Religions), a mid-seventeenth-century Persian treatise on India's religions whose author spent time in Punjab, makes reference to the language of the Sikhs, which he terms "Jataki." Jataki is identified in nineteenth-century sources as both the language of the Jats, an important caste group in the Punjab (see chapter 4) and as a "corruption" of Multani, spoken in southern Punjab and closely related to Punjabi, if not a dialect of the language. Sujan Rai Bhandari's Khulasat ut-Tawarikh, a late seventeenth-century Persian-language history of India, also describes Punjabi as the spoken language of the area.

Punjabi was only one language in use in the region, however, as the reference to Persian in the janam-sakhis suggests. Persian had been the official court language-and thus the language of administration-for precolonial regimes from the turn of the second millennium. Introduced as a court language by the Ghaznavids when they established control over the Punjab during Mahmud's reign (997-1030), Persian remained the court language through the reigns of the Delhi sultans and the Mughals, and even during the era of the Sikh kingdom of Lahore. The status of Persian as a court language helped ensure that with the expansion of Muslim rule in the subcontinent, Persian was widely adopted as a language of letters. Indian elites were thus educated in Persian, both as a means to employment in state apparatuses (as was the case with Guru Nanak) and to partake in a high culture that emanated from courts and was emulated in society more generally.

This process unfolded in Punjab from the Ghaznavid period onward. From approximately 1000 C.E. through the era of Sikh rule, Persian remained an important language of literary production, used for poetry and prose, including nonfiction works such as histories, memoirs, and court chronicles. Beginning with the poetry of Mas'ud Sa'd Salman of Lahore (d. 1121), countless Persian literary works were produced in Punjab, many of them significant to India's Persian canon. Focusing on the immediate precolonial period makes the point adequately, however. Almost all major extant or known historical works produced in early nineteenth-century Punjab were composed in Persian, including Khush-Waqt Rai's Ahwal-i Firqah-i Sikhan (1811, a history of the Sikhs from their origins); Diwan Amar Nath's Zafarnama-i Ranjit Singh (c. 1830s, a court chronicle of Ranjit Singh's reign); Sohan Lal Suri's Umdat ut-Tawarikh (c. 1830s, a history of the Sikhs) and Ibratnama (c. 1840s, a poem on political events at the Sikh court from 1840-43); Ghulam Muhiyuddin's [Bute Shah] Tarikh-i Punjab (c. 1842, a general history of the Punjab); and Ganesh Das's Char Bagh-i Punjab (1849, a political history and geography of the Punjab). What little we know of the newspapers or protonewspapers produced in the years just before annexation suggests that these, too, were written in Persian.

Persian may have dominated certain genres of precolonial literary production in the Punjab, including history, but other languages also had a presence. Certainly, Sanskrit and Arabic were used in religious rituals and for religious scholarship. Their significance is suggested by a tradition of learning in these languages that extended into the late nineteenth century through indigenous schools. Precolonial Punjab also had a vibrant vernacular literary culture, though not in Urdu. While some see the Punjab, and Lahore in particular, as the cradle of Urdu literature because of the reputed (but now lost) vernacular divan (collection) of Mas'ud Sa'd Salman of Lahore, most scholars agree that Urdu's roots lie in the Dakani compositions of Sufi and court poets from the fifteenth- and sixteenth-century Deccan. In north India, classical Urdu literature (principally poetry) came into its own only in the mid-eighteenth century. From then until the disruptions caused by the rebellion of 1857-58, Urdu's major literary centers were Delhi and Lucknow. While Urdu literati also settled in other north Indian cities, including Murshidabad, Patna, Banaras, and Calcutta, neither Lahore nor any other city in the Punjab's central districts figures prominently in Urdu literary history until the colonial period.

A vibrant Punjabi literary culture flourished in precolonial Punjab, however, one that can be traced back to the eleventh century. Most scholars agree that Punjabi's literary foundations were laid in the eleventh and twelfth centuries, with the compositions of religious figures such as the Nath saint Gorakhnath (c. eleventh century) and the Sufi Shaikh Farid. Mohan Singh Uberoi, an influential literary historian writing in the 1930s, sees Gorakhnath and Farid as the first participants in a Punjabi literary era that spanned the eleventh to the fifteenth centuries, which he calls "Old Punjabi." He includes the compositions of the bhakti saint Kabir (1440-1518), and argues that the period is marked by heavy use of a dialect known as Western Punjabi or Lahnda. Punjabi compositions during this era used a range of poetical forms and genres, some indigenous and some adopted from Persian and Sanskrit. Uberoi identifies thirty-eight such genres, among the most significant being dohra (rhyming couplet), kafi (couplet), var (lay or war ballad), shloka (stanza of four parts of equal length), chaupai (four-line stanza), and kabit (quatrain comprised of thirty-one or thirty-two syllable lines).

Not all historians of Punjabi literature would anchor the tradition as definitively in the eleventh century or include bhakti poets such as Kabir. Christopher Shackle, for example, acknowledges the importance of Shaikh Farid, but suggests that the discernible literary roots of what he calls modern standard Punjabi are much more recent, dating from the sixteenth century. In contrast to Uberoi, who wants to see Guru Nanak's compositions, for example, as squarely in the Punjabi literary canon, Shackle suggests that the Adi Granth (which includes the writings of Nanak and other sants) and other Sikh sacred texts are better described as being in "the sacred language of the Sikhs," which he argues draws many of its elements from a local source (vernacular language-Punjabi, presumably), but is "certainly not 'Old Punjabi.'{hr}" These distinctions may seem overly academic, but they point to the contested history of Punjabi literature, including contests over the place of the Punjabi language in twentieth-century politics.

While scholars may thus disagree about whether Sikh sacred scriptures such as the Adi Granth and the Dasam Granth are central to a Punjabi literary canon, as Uberoi would have it, there is more scholarly consensus on the inclusion of a number of sixteenth-, seventeenth-, and eighteenth-century poets who wrote in a range of verse forms. Among these are Bhai Gurdas Bhalla (c. 1550s-1635), Shah Husain (c. 1530s-1600), Sultan Bahu (1629-91), Damodar (c. early seventeenth century), Muqbal (c. mid-eighteenth century), and Waris Shah (c. mid-eighteenth century). While Bhai Gurdas and the Sufis Shah Husain and Sultan Bahu all composed in traditional genres, vars and kafis, respectively, the three latter poets composed qisse, at the time a new verse genre of Punjabi composition. All these poets used existing folklore or popular tales in their poetry, whether composing Sikh, Sufi, or "secular" poetry. The earlier part of this early modern period also saw the first Punjabi prose compositions, though these were limited to the single genre of janam-sakhis. The latter part, the eighteenth century in particular, is marked by a proliferation of Punjabi literary production (though we can also surmise that more manuscripts from then have survived). Uberoi argues that during these centuries a stylistic change is also discernible, as we see more compositions in the language of the Punjab's central districts as opposed to Western Punjabi. One text that employed this new register was Waris Shah's qissa Hir-Ranjha (1766). Universally acclaimed as the greatest achievement of Punjabi letters, this text alone explains why the eighteenth century is recognized as an unequivocal high point in the history of Punjabi literature.

Most scholars date Punjabi literature's modern period to the turn of the nineteenth century. A focus on Punjabi literary production in the early nineteenth century helps provide a richer picture of literary conventions on the eve of colonial rule. Punjabi compositions from this era use a number of verse genres, the most significant being the qissa and the var. The most celebrated poet of the period, Hashim Shah (1735-1843), like many of his contemporaries composed in a variety of genres. He is best remembered, however, for his Punjabi qisse, among them the romances Sassi-Punnun, Shirin-Farhad, Sohni-Mahival, and Hir-Ranjha. Ahmad Yar (1768-1842) also enjoys renown for his Punjabi qisse. Indeed, he may be this period's most prolific poet, having produced some forty different Punjabi qisse during his lifetime. A teacher of Arabic and Persian and a member of the Sikh court, Ahmad Yar provides particular insight into the polyglot and varied literary production of the day. He composed in Arabic, Persian, and Punjabi, and his subject matter ranged from medicine (Tibb-i Ahmad Yari) to history (Shahnama, a Persian chronicle of Ranjit Singh's court), Islam (Jang Ahmad), and romance (Hir-Ranjha, Sohni-Mahival, Laila-Majnun, among others). A number of Ahmad Yar's contemporaries shared his linguistic breadth, underscoring the fact that many Punjabi poets could work in more than one language.

Hamid Shah Abbasi (b. 1748) was, like Ahmad Yar, both a polyglot and a recipient of court patronage. Abbasi composed Punjabi qisse and enjoyed the support of the Raja of Nurpur (a Punjab hill state). Abbasi's most significant literary contribution is the qissa Hir-Ranjha, a text in which Abbasi self-consciously places himself within a particularly Punjabi literary tradition. His contemporary the poet Vir Singh also composed a qissa Hir-Ranjha (1812), and enjoyed the patronage of Maharaja Karan Singh of Patiala, the most important independent state in nineteenth-century Punjab. Other qissa poets of the era were from more humble origins. Jog Singh ([Qissa] Hir Jog Singh, 1825), for example, was a sadhu (mendicant). Although he appears to have enjoyed little literary fame, fortune, or patronage during his lifetime, his text inspired a number of future Punjabi poets who placed him prominently in their own literary genealogies.

Vars were also an important genre in nineteenth-century Punjabi literary production. As a genre, they date to at least the era of the Adi Granth (sixteenth century), when they were first recorded. The most historically significant var of the early nineteenth century is likely Jang Hind Punjab (The War of Hind and Punjab), by Shah Muhammad (1780-1862). Muhammad served in the army of the Sikh kingdom of Lahore, and his text describes the Anglo-Sikh war of 1845-46 in which he fought. Many other significant vars were written in the immediate precolonial period, among them Qadir Yar's Hari Singh Nalwa, Puran Bhagat, and Rani Kolkila, and Sahai Singh's Hari Singh Nalwa.

In precolonial Punjab, then, as in much of precolonial north India, Persian served as a literary language and the only language for certain genres (in the Punjab, all prose other than religious exegesis). The Punjabi literary tradition was coeval with this Persian literary tradition, but had its own vitality and lineage. In the second half of the nineteenth century, in other parts of north India, writers increasingly used vernacular languages for genres where Persian had once dominated such as history, chronicle, or memoir. This was also true in the Punjab, but there litterateurs adopted Urdu, not Punjabi, as the language of choice for these genres. The genre of history exemplifies this shift. Whereas in the Punjab history had long been written in Persian, in the late nineteenth century a spate of Urdu histories appeared (to my knowledge, no Punjabi language histories were produced). Some of these histories were produced at the behest of the state, examples being Noor Ahmad Chishti's Yadgar-i Chishti (Chishti's Memories; 1854) and Tahqiqaat-i Chishti: Tarikh-i Lahore Ka Encyclopedia (Chishti's Inquiries: An Encyclopedia of Lahore's History; 1867). Other examples are the local histories produced in the 1860s at the behest of British district commissioners as they were carrying out their settlements of various Punjab districts. The results, such the Tarikh-i Jhelum (A History of Jhelum) and Tarikh-i Zillah Montgomery (A History of District Montgomery) were composed in Urdu. A number of the histories of the region from the late nineteenth century were not commissioned by the colonial state, yet they, too, were in Urdu. Kanhayalal's Tarikh-i Lahore (History of Lahore; c. 1871), and Mufti Ghulam Sarwar's Tarikh-i Makhzan-i Punjab (History and Record of Punjab; 1884) are paradigmatic of this literary shift.

What historical processes account for Urdu's dominant place in Punjab's literary and print cultures? Why did it dominate newspaper publishing to the almost complete exclusion of newspapers in other languages? One important element to consider is the disruption in the literary culture of Delhi (and Lucknow) caused by the colonial response to the rebellion of 1857-58. Frances Pritchett, in her fine work on Urdu literary culture during this period, argues that these events witnessed the destruction of the patronage system-grounded in the ustad-shahgird (teacher-disciple) relationship-that was so important to Urdu's literary vitality in Delhi. She goes so far as to suggest that the aftermath of 1857 destroyed not only the patronage system but "in fact the whole [Urdu literary] culture." With the demise of Delhi as a site of Urdu patronage, Lahore became an important center of Urdu literary culture as Urdu poets and men of letters migrated there in search of patrons. The historian of Urdu literature Ram Babu Saxena refers to Lahore during this era as a "resort of the exiled men of letters" from Delhi, largely because the Punjab provincial government was based there, and it was often with the colonial state that men like Altaf Husain Hali (1837-1914), one of Urdu's greatest modern poets and critics, found employment. Notwithstanding the important cultural implications of 1857 and the British response in its wake, it is not the dislocation of Urdu literati from other parts of north India that accounts for the dominance of Urdu in colonial Punjab's world of literary production and print culture. More important were the policies of the colonial state, and the most significant of these was its language policy.

Colonial Interventions

When the East India Company annexed the Punjab in 1849, it instituted a new and unique administrative structure, perhaps owing to both the difficulty with which the area was conquered and its military-strategic significance as a buffer zone between the Gangetic plain and Central Asia. Rather than incorporating the new territory into one of India's three existing presidencies (Bengal, Madras, and Bombay), the Company created a three-member Board of Administration, whose officers were responsible directly to the governor-general in Calcutta. The Company gave the board extensive and centralized power and worked assiduously to establish its rule and assimilate the province into the broader structures of colonial governance in India. Setting a language policy was crucial to both of these priorities.

By 1849, the Company had both precedent and a policy to follow regarding what language(s) to use in administering Indian territories. Through its initial years of rule in India, the Company had used Persian as the language of local administration, in keeping with precolonial practices. Act 29 of 1837 replaced Persian with Indian vernacular languages. While administration at the highest levels would continue to be conducted in English, the act called for administration at provincial and local levels to be in a local Indian language. The proponents of Act 29 had persuasively argued that Indians should be administered and adjudicated in a language they understood, and therefore Persian, which was only accessible to an educated elite, would no longer suffice. While the act did not specify which languages should be adopted in any given province, it was clearly grounded in the idea that the spoken language of an area should be its administrative language. Given the linguistic complexity of the subcontinent-where, as one Indian proverb has it, spoken languages shift dialects as frequently as every four kos (four miles)-this was hardly feasible. Therefore, in instituting the act, Company officials canvassed administrators about the language practices in their jurisdictions and then adopted the vernacular language they deemed most widely spoken in a province or in a number of its districts. On this basis, it was decided that most of the NWP's inhabitants spoke some variation of Urdu. In contrast, in Bengal three languages were chosen (Bengali, Oriya, and Hindi), since clusters of districts reported different languages as their predominant vernaculars.

Despite the seeming clarity of this policy, a discrepancy emerged between colonial intent and practice when it came to instituting vernacular languages for administration. The intent of Act 29 was to make the language of government accessible to the governed, but in practice this effort was vitiated in at least two ways. First, when the state chose a language for local administration, it necessarily privileged one dialect over others. People's ability to understand and use the administrative language therefore depended on its relationship with their own dialect. Second, the register of vernacular language employed by the government in its courts and administration often remained inaccessible to the majority of people because it was heavily laden with terms and vocabulary far removed from everyday usage. Most relevant to the present discussion, though, is that with Act 29 the Company endeavored to make the vernacular language of a province, or particular districts within it, the official language of these places.

When the Board of Administration considered a language policy for the Punjab, it did as Company administrations had done elsewhere: it canvassed local officers, in this case in each of the province's six divisions, and asked which language was "best suited for the courts and Public Business." Officials in the province's western half, which included the western portions of the Punjab and frontier areas (today the Northwest Frontier Province of Pakistan), suggested Persian. Those in the eastern half suggested a shift to Urdu. In both cases the board concurred and thus instituted a two-language policy for the province in September 1849.

In 1854 the board abruptly changed the two-language policy, motivated by new civil service rules that put certain officials in jeopardy of losing their jobs. In November of that year, a mandatory civil service exam was announced that for the first time required all candidates to pass a test in the official language of the courts in which they were employed. Many court officials in western Punjab were not fluent in Persian and so relied on interpreters; under the new rules they faced dismissal. Officers in districts where Persian was the language of the courts petitioned the board to institute Urdu in its place, in the belief that Urdu could more easily be mastered. The board agreed, designating Urdu the official language of colonial government across the Punjab. This designation survived the transition from Company Rule to Crown Rule in 1858, and remained in place until the end of colonial rule in 1947.

The colonial language policy in the Punjab contrasts with that adopted in other Indian provinces in that there is little evidence that Urdu was used in the province in any meaningful way. Colonial officials were, by their own admission, aware of this. John Lawrence, the board's president, clearly indicated as much in a note written in 1849, just as the Persian and Urdu policy was being implemented. "It should be considered," he wrote, "that the Urdu is not the language of these Districts [and] neither is Persian." That Lawrence knew this is no surprise-colonial knowledge about language and linguistic practices had repeatedly affirmed that Punjabi was the colloquial language of the majority.

Colonial Knowledge Production about Indian Languages

The production of colonial knowledge about Indian languages had begun in earnest with the Company's transformation in the eighteenth century from a trading concern to a political power. Colonial officials' pursuit of such knowledge and their desire to master Indian languages was in part prompted by the exigencies of rule. The study of Sanskrit and Persian in the late eighteenth century, for instance, was crucial to the colonial project of establishing a legal system in India with distinct laws for Hindus (the Gentoo code) and Muslims (Anglo-Muhammad law). Thomas Trautmann's work has been crucial to recognizing another important context for the colonial production of knowledge about Indian languages: European intellectual preoccupations with the origin and significance of languages, nations, and races. This broader intellectual context is critical for understanding the work of scholars such as William Jones and F. W. Ellis, the "founders" of the Indo-European and Dravidian language families, respectively. The contributions of such early Orientalist scholars notwithstanding, it was missionaries-and those of the Serampore Mission in Bengal, in particular-who produced what is perhaps the most significant information about the languages spoken in India in the early nineteenth century.

Missionaries devoted themselves to the study of Indian languages in order to communicate directly with Indians, to translate the Bible into what they deemed to be the appropriate languages for targeting specific communities, and to produce philological materials that would help other missionaries-and often Company employees-to learn Indian languages. Indeed, missionaries played a significant role in producing knowledge that linked together languages, scripts, and religious communities in India in definitive ways, though not always explicitly. Reconsidering the early publishing efforts of the Ludhiana Mission Press makes these associations evident. Recall that the press initially had only English and Indo-Persian type fonts. Finding themselves in a predominantly Muslim locale-somewhat to their surprise since they thought they were going to the "land of the Sikhs" only to find that Sikhs were a mere 10 percent of Ludhiana's local population-the missionaries' first publication, A Sermon for the Whole World, was published in Persian, based on the assumption that this was a Muslim language and that it was the best language by which to proselytize, in print, to local Muslims. They would subsequently publish tracts in Hindi, aimed at the local Hindu population, and in Punjabi, aimed at the local Sikh population, but only with the arrival of Devnagari and Gurmukhi fonts, respectively.

Missionary associations of language with religious community are most clear in their conception of Punjabi, which for them was so closely associated with Sikhism that the two terms-"Shikh" and Punjabi-had by the 1830s long been interchangeable in missionary discourse. The Seventh Memoir Respecting the Translation of the Sacred Scriptures into the Languages of India, Conducted by the Brethren at Serampore, published in 1820, provides an example. In a section titled "Languages in Which the New Testament is Published," one finds alongside references to "Pushtoo" (Pashto), "Telinga" (Telegu), and a number of other languages, "the Shikh language," clearly a reference to Punjabi. Any doubts about the correlation between the two are dispelled by the writings of William Carey (1761-1834), a Serampore missionary and significant figure in the history of Punjabi, who often referred to the language as "Shikh or Punjabi." That the language was in turn associated with a specific script is clear from the fact that the Serampore missionaries, like the Ludhiana missionaries after them, only published Punjabi in the Gurmukhi script even though it was possible to record the language in Indo-Persian script, a practice that was clearly quite common based on existing Punjabi manuscripts (figure 1, for example). But the missionaries singularly associated Punjabi with the Gurmukhi script and Sikhs. The Gurmukhi script had been created by the Sikh gurus in the sixteenth century to record their Punjabi compositions. Because the gurus created the script expressly for Punjabi it is most closely associated with that language, though it can be used to record any South Asian language.

For missionaries, targeting any religious community necessitated using their language and their script. Based on their publishing, missionaries correlated the Indo-Persian script and initially Persian and then increasingly Urdu with Muslims, the Devnagari script and Hindi with Hindus, and the Gurmukhi script and Punjabi with Sikhs. Later in the century, language activists in north India would use precisely such (singular) conceptions of language, script, and religious community-put in place by the missionaries and influential among colonial officials-to their own advantage. They, too, would argue that Urdu in the Indo-Persian script was the language of Muslims and that Hindi in the Devnagari script was the language of Hindus. Sikhs would ultimately follow suit with the Gurmukhi-Sikh-Punjabi triad, the groundwork having largely been laid for them by colonial knowledge of Punjabi, knowledge first produced by the Serampore Mission.

The Serampore Mission in Bengal plays a particularly important role in the history of Punjabi because it was the first European institution to produce and disseminate knowledge about the language, a process in which William Carey played a significant part. In pursuit of his missionary work, Carey produced the first modern grammar of Punjabi in 1812, which laid the foundation for all future philological studies of the language. Carey's interest in Punjabi stemmed from his knowledge that it was spoken in large tracts of India's northwest. Indeed, the Serampore missionaries were keen to know which languages were spoken where in India so as to achieve the linguistic competence necessary for their work. Their efforts in this regard were nothing short of remarkable. In 1822, the Baptist Missionary Society published a language map of India based on information compiled by the Serampore missionaries (map 5). The map, the earliest of its kind to my knowledge, contains relatively detailed information about languages then spoken in the Punjab. It divides the Punjab into linguistic zones, and shows "Punjabee" as the language of central Punjab, "Mooltanee or Wuch" as spoken in southern Punjab, and "Hurriana" in eastern Punjab (map 6). As a representation of the earliest existing European data on languages spoken in the Punjab, the map documented two facts immediately relevant to the colonial language policy implemented some twenty-five years later: first, neither Persian nor Urdu was spoken there; and second, the languages identified as spoken in the Punjab were closely related. They can be understood as being on a continuum comprised of a single linguistic field: Punjabi. [Place maps 5 and 6 near here.]

The Serampore missionaries were not the only ones to create and circulate knowledge about Punjabi in the early nineteenth century. Missionaries of the American Presbyterian Church, who arrived in India in 1833 as the British were expanding into the Punjab, also took a keen interest. Although the East India Company would not formally annex the region for some years, the cis-Sutlej states (between the Sutlej River and Delhi) had already come under British "protection." William Reed and John Cameron Lowrie, the first American Presbyterian missionaries to arrive in India, decided to establish a mission in British-controlled Punjab because there was as yet no missionary presence there and because they thought it to be the land of the Sikhs, whom they considered particularly susceptible to conversion. The Punjab also appealed because they thought its location would afford an opportunity to spread the gospel into the heart of Central Asia through Kashmir and Afghanistan. By 1834 Lowrie had established a mission in Ludhiana, which, as already noted, established the region's first printing press. In addition to its publishing activities, the mission produced grammars, dictionaries, and guides to the language that would be the standard works for their generation, including Rev. L. Janvier's Idiomatic Sentences in English and Punjabi (1846), Rev. J. Newton's A Grammar of the Panjabi Language, with Appendices (1851), and Revs. J. Newton and L. Janvier's A Dictionary of the Panjabi Language, Prepared by a Committee of the Lodiana Mission (1854). Through these works, in part, the mission played an equally important role in sustaining a colonial conception of Punjabi as the Sikh language.

The colonial state's idea that Punjabi was the Sikh language had been established early in the nineteenth century as the Company, spurred by both security and pecuniary interests, turned its attention to India's northwest. This was an area that the British identified with the "Sikh nation" due to Sikh political control there under the leadership of Ranjit Singh (r. 1799-1839). As Lt. Col. John Malcolm stated explicitly in his 1812 Sketch of the Sikhs, information about this community was now vital to Company interests. "Although the information I may convey in such a sketch may be very defective," he wrote in what can only be regarded as false modesty given the tenor of the work more generally, "it will be useful at a moment when every information regarding the Sikhs is of importance." Malcolm's was not the first colonial study of the Sikhs, but he claimed special authority for his work: access to the Adi Granth. As with many of his Orientalist contemporaries who deemed texts critical to understanding India's religious traditions, Malcolm thought the Adi Granth was a key to understanding the Sikhs, that it would provide insights about their "history, manners, and religion." Through this emphasis on Sikh textual traditions, an approach embraced by subsequent state actors as well, Punjabi became identified as the Sikh liturgical language, though this was something of a misidentification. Sikh sacred scripture (e.g., the Adi Granth, the Dasam Granth, the janam-sakhis) comprises writings in a number of different Indian languages but is always recorded in the Gurmukhi script. As noted above, although Gurmukhi was created specifically to record Punjabi, it can be used to record any language, and was adopted for Sikh sacred literature during its period of canonization in the sixteenth, seventeenth, and eighteenth centuries. Thus, while not all Sikh sacred literature is in the Punjabi language, it is (almost without exception) recorded in the Gurmukhi script. This distinction was lost on most colonial observers in the early half of the nineteenth century, however, and in both Company and missionary understandings of the Sikh textual, linguistic, and script terrain established at that time, a binding association of the Gurmukhi script with the Punjabi language emerged. This, in turn, contributed to the identification of Punjabi in colonial discourse as the sacred language of Sikhs.

In addition to being understood as the Sikh sacred language, Punjabi was also identified as their colloquial language. While the former was only partially correct, the latter was more firmly grounded in fact since the overwhelming majority of Sikhs were native Punjabi speakers. For the colonial state, as well as the Christian missionaries who tried to convert Sikhs, this confluence of sacred and spoken language produced an abiding belief that Punjabi was specifically a Sikh language. The tenacity of this belief in colonial discourse is striking given the contemporary production of knowledge that made it plainly evident that Punjabi was spoken and written by a much broader spectrum of the population; it was hardly the language of Sikhs alone.

While in the first half of the nineteenth century missionaries had principally gathered and produced knowledge about Punjabi, in the latter half of the century colonial officials and ultimately the colonial state itself produced philological studies on languages used in the province. One important study was Outlines of Indian Philology (1867), based on the firsthand experience of colonial officer and amateur philologist John Beames. Beames served in India for thirty-five years (1858-93) and was sent to the Punjab for his first tour, arriving in the town of Gujrat in 1859. His language map confirmed the information published in the Serampore Mission's map earlier in the century: Punjabi was the language of much of the colonial province (map 7). Although Beames had pursued his philological interests as an amateur-his texts were published not by the colonial government but by private publishers in Britain-his work was well respected by contemporary scholars. No less a figure than George Grierson, the leading philologist of his age, wrote at Beames's death (in 1902): "Oriental scholarship has lost one of its most eminent interpreters." [Place map 7 near here.]

In time, the colonial state also developed an active interest in mapping Indian languages. Although the first censuses of the Punjab (1855 and 1868) contained no information on language(s), statistical information on indigenous language practices began to be collected by various colonial departments in the 1870s. For example, the administration report of the Punjab for 1873-74 recorded spoken vernaculars. A district-by-district breakdown from this report identified Punjabi as the spoken vernacular in every district of central and southern Punjab, and in a number of other districts as well.

As the colonial state's reporting on linguistic practices became more detailed over time, Punjabi was time and again confirmed as the region's primary spoken language. In the first comprehensive census of the Punjab, conducted in 1881, an overwhelming majority in central and southern Punjab returned Punjabi as their vernacular. Subsequent district gazetteers, citing figures on language drawn from this census, show that the Punjabi-speaking population of districts in central and southern Punjab ranged from 85 to 98 percent.

The most comprehensive linguistic study undertaken during the colonial period, George Grierson's Linguistic Survey of India, conducted and published between 1903 and 1928, corroborated the information collected by the Serampore missionaries, by John Beames, and by various administrative reports and gazetteers. Grierson reorganized the Punjab into two linguistic zones: east and west. In the east he identified Punjabi and a series of its dialects (one of which was the "Mooltanee" cited in the Serampore Missionaries' language map as a separate language) as the spoken language(s). In the west, Grierson referred to the spoken language as Lahnda. While he argued that Lahnda was distinct from Punjabi, most linguists today agree that Lahnda is best described as a Punjabi dialect. Dialect was something of a pejorative term compared to language in the philological discourse of the late nineteenth century (and perhaps to this day for language activists). Irrespective of the shifting terrain of language and dialect, the significant point is that in the early nineteenth century the Serampore missionaries identified Punjabi or closely related idioms, and not Urdu or Persian, as the spoken language of the Punjab, and throughout the late nineteenth and early twentieth centuries colonial surveys repeatedly concurred.

Deconstructing Colonial Language Policy

Knowledge about the Punjabi language produced by both missionaries and colonial employees, then, had two somewhat contradictory aspects, both of which help us to contextualize and explain colonial language policy in the region. The first is that from the early decades of the nineteenth century, missionaries knew that the region's inhabitants spoke Punjabi. This missionary knowledge was accessible to the colonial authorities. William Carey, author of the first Punjabi grammar, taught at the Company's college at Fort William from 1801. The Serampore language map had been published in Britain and would have been readily available to Company authorities. Provincial officials were also in touch with the missionaries of the American Presbyterian Mission at Ludhiana. In 1851 the Punjab government sought out Revs. Forman and Newton of the mission precisely for their expertise in Punjabi. Second, although the Punjab's population was religiously diverse-made up of Muslims, Hindus, Sikhs, and, increasingly from midcentury on, Christians-in colonial perception the Punjabi language was most closely associated with the Sikh community. The language policy adopted by the colonial state in the Punjab suggests that this latter association between language and religious community was more influential than the fact that most of the region's inhabitants spoke Punjabi irrespective of their religious identities.

Given colonial understandings of Punjabi as the language of Sikhs, officials feared that adopting it as the official provincial language might promote Sikh political claims. The nascent colonial state in the Punjab certainly did not want to promote the claims of its defeated rivals, particularly since after annexing the region it found its situation far from ideal. Among the Company's most serious political concerns in the mid-nineteenth century were the military threats it perceived from both within and beyond its borders. From within, the Company feared an uprising by the military of the Sikh kingdom, which had yet to be completely disarmed. Despite its victories in the Anglo-Sikh wars, the Company lacked confidence in its control over the Punjab. That insecurity is evident in a November 1849 report by Charles Napier, commander-in-chief of Company forces: "The Punjaub has been occupied by our troops but it is not conquered," he wrote. "We now occupy it with 54,000 fighting men and it is at present very dangerous ground." Napier's fear was based on his belief that the Punjab still had 100,000 Sikh soldiers, whose "courage has been no way abated by the last struggle," and who he thought "may some day unexpectedly use it" against the Company. These fears bore directly on the language policy implemented by the Company state. To institute Punjabi as the state language could have encouraged Sikh political aspirations, which the state wanted to avoid at all costs. If political concerns about Sikh resurgence suggested that Punjabi should be suppressed, then colonial attitudes about its lack of merit as a language helped justify the implementation of such a strategy.

Colonial officials raised doubts that Punjabi was capable of serving as an administrative language. Such attitudes are most explicit in a series of correspondence from the early 1860s, after language policy had been instituted in the Punjab, but they are hinted at in early archival materials and we can surmise that they formed a continuum with ideas established during the colonial state's earliest experience of rule there. Foremost among officials' opinions about Punjabi was that it was not a language at all, but "merely a patois of the Urdu." Unlike classical languages such as Sanskrit or Persian, which were held in high esteem, or vernacular languages such as Urdu, Tamil, or Gujarati, which were used for official purposes, Punjabi was represented as a derivative dialect. Another objection was raised on the ground that there was no established "standard" Punjabi. "There is no one standard Punjabee to fix as the language of the courts," one official argued, because "there are wide points of divergence between the patois of the tribes of the Ravee and that of those of the Sutlej."

Criticism of Punjabi included a perception that it was unsuitable for official uses because it "would be inflexible and barren, and incapable of expressing nice shades of meaning and exact logical ideas with the precision so essential in local proceedings." Some officials based their arguments against Punjabi on its lack of uniformity in its written form. Another erroneously argued, "Punjabee could not be written in the Persian [Indo-Persian] character." Some officials even asserted there was no tradition of writing in Punjabi whatsoever: Colonel Hamilton, commissioner of Multan, wrote, "The Punjabee has never been a written language in the Mooltan division. It is doubtful whether a man could be procured in the division who could write Punjabee correctly in any character."

Such sentiments about Indian vernacular languages had been voiced in other contexts. In the Bengal Presidency, colonial officials had earlier made remarkably similar comments about Bengali, Oriya, and Hindustani (spoken in the presidency's central-eastern, southern, and western portions, respectively). In considering which of these languages should replace Persian as the official language, Company officials had charged that vernacular languages were not standardized; they were "uncouth", "barren," and "unadapted to the conduct of judicial proceedings." They also deemed vernacular languages less efficient than Persian. Officers had also expressed fears that a change from Persian would be extremely detrimental to colonial administration because, in the words of one, finding "competent officers to carry on the business of the courts" would be impossible. Despite these misgivings, all three languages-Bengali, Oriya, and Hindustani-were adopted for use in the Bengal Presidency. The outcome of this earlier debate is helpful in assessing the degree to which the biases and misconceptions of colonial officials in the Punjab figured in colonial language policy there. While such attitudes are sure to have had a role, we can presume they were not decisive. The colonial state would surely have instituted Punjabi as the official language if it had served provincial policy, its "shortcomings" notwithstanding. But it appears that Punjabi did not further colonial officials' political aims. Not only could using Punjabi have helped prompt a Sikh resurgence, but there were decided political and administrative advantages to instituting Urdu instead.

One such advantage was that using Urdu (and initially Persian) allowed the Company to employ a cadre of experienced administrative personnel immediately upon annexation. In the initial years of colonial rule, the Company filled the ranks of Punjab's administration with British and Indian personnel who had served in other Company provinces, principally Bengal and the NWP. The Indians among them were primarily from the Gangetic plains and Bengal, and had no knowledge of Punjabi; their vernacular languages were Hindustani and Bengali. They did have a working knowledge of Urdu and Persian, however, from their service in other parts of north India, and this made their transition to service in Punjab relatively seamless. Similarly, British officers who took up positions in the new colonial province knew no Punjabi, but had a working knowledge of Urdu and, in some cases, a rudimentary knowledge of Persian.

An additional reason for promoting Urdu was that it aided the integration of the new province into the Company's Indian territories. Act 29 had declared Hindustani the administrative language in Bihar, the NWP, and parts of the Central Provinces. The script officially sanctioned in those provinces was Indo-Persian, and this made the "Hindustani" of much of north India coterminous with the language referred to as "Urdu" in the Punjab. In light of the widespread use of Hindustani, John Lawrence went so far as to refer to it as the lingua franca of India, an idea that had circulated in colonial circles for some time. That the Punjab was included in the area using this lingua franca facilitated the region's integration into the Company's Raj.

Using Urdu in the Punjab provided the colonial state a further advantage, though it is one that does not emerge from the colonial record until some years after the policy was instituted. Correspondence from the early 1860s, in which district officials were asked to judge the merits of replacing Urdu with Punjabi, reveals that some viewed Urdu as the language of Punjab's native elites. The colonial state's policy across India had been to foster native elites as intermediaries with its subjects at large. In the Punjab, the state identified the region's indigenous rural elite-the "Punjab chiefs"-as bulwarks of its power. In return for their support, the state buttressed the power of these elites through honorary titles and, more importantly, land grants. The state's language policy was part of this broader political equation. "If Punjabee is declared the Court language," wrote one official, "what is to become of the Chiefs who almost universally speak very fair Oordoo and the more educated classes who really cannot speak the veritable Punjabee?" The interests of this native elite-the "chiefs" and "educated classes" in colonial parlance-surface repeatedly in considerations of language policy. For instance, in 1863 the commissioner of Rawalpindi Arthur Brandreth argued, "Nothing would be gained by substituting Punjabee for the Urdu. If this were done it would be a retrograde proceeding," particularly because "Urdu is the language of the educated classes." Whether Urdu really was the language of Punjab's indigenous elite is an open question. Colonial linguistic data cited above suggest that is was not, as does an 1855 dispatch from the Court of Directors that stated, "Urdu [should] be made familiar, in the first instance, to the educated classes, and through them, as would certainly follow, to the entire body of the people, to the eventual supersession of inferior dialects." Nonetheless, it appears that officials in the Punjab insisted that Urdu was the language of the elites, and that they saw promotion of the language as a way to protect elite interests, and thereby their own.

Thus a combination of factors explain the colonial state's decision to institute Urdu as the Punjab's official vernacular language. The most important are those related to consolidating colonial rule: using experienced administrative personnel, facilitating Punjab's integration into Company territories, and supporting native intermediaries. British fears of a Sikh resurgence and the conception of Punjabi as a Sikh language surely played their part as well. While this specific set of issues may have influenced the language policy established in the Punjab, the implications of the policy reached well beyond these original considerations.

Colonial Education

The vernacular language policy instituted in 1854 targeted two specific areas of government: judicial proceedings and revenue collection. Once the language policy had been established for these arenas, however, its impact was felt well beyond them. As the colonial government expanded, so did the impact of its language policy. Education was one critical site where this influence was felt. Scholarship on education in colonial India has invariably focused on the Anglicist-Orientalist controversy of the 1830s, which pitted the merits of European subjects of study against the subjects traditionally studied in Indian schools. The Anglicists won that debate. Their victory was reflected in the nature of Indian higher education, which emphasized European subjects and was conducted in English. In the early 1850s, however, the Company decided to extend education to the masses. Colonial officials agreed that for primary education, English was best bypassed for vernacular languages. Which vernacular languages were adopted, and where, appears to have followed established policies about the vernacular language of provincial government.

The East India Company introduced education on a broad scale, on a trial basis, in the NWP in 1850. This scheme of vernacular education was considered successful, and in 1853 the government in Calcutta suggested that the scheme be extended to the Punjab. In response to Calcutta's overtures, Punjab's judicial commissioner Robert Montgomery proposed an education scheme for the province the following year in which he carefully assessed what the language of instruction should be. His minute bears quoting at length because it shows the impact of official language policy on the language chosen for education:

The language used might be either Punjabee with Goormookhee character, or with the Persian character, or mixed Hindustanee with Punjabee phrases and idioms and written in the Persian or Hindee character; or pure Oordoo or Hindi. The Judicial Commissioner recommends the disuse both of the Punjabee language and of the Goormookhee character. Goormookhee though of sacred origin is rapidly falling into disnature.... The currency of Punjabee as a spoken language is also diminishing. It is degenerating into a mere provincial dialect. Hindustanee [Urdu when written in the Indo-Persian script] is the prescribed language of the courts and of the Public Department [and] it is becoming familiar to the Upper and Middle classes and the rural population understand it nearly as well as their brethren of Hindustan though neither perhaps will ever be proficients in this or any other polished language. There is no reason to perpetuate the Punjabee or even check its decondence [sic] at the expense of the superior Hindustanee.... On the whole then the Judicial Commissioner would prefer to adopt the same language as in the N.W.P.'s [sic] namely Oordoo and Hindi and there by to secure uniformity in addition to the advantage which those languages passes [sic] over the less cultivated patois of the Punjab.

On one hand, Montgomery's remarks point to a logic similar to that employed in earlier considerations about the Punjab's administrative language: continuities with the NWP and the failings of Punjabi as a language. On the other hand, his comments reveal something new: Montgomery pointed to the adoption of Urdu as the "prescribed language of courts and of the Public Department" as part of his reasoning. With Urdu as the vernacular language of administration, it seemed prudent to use Urdu in colonial schools.

Chief Commissioner John Lawrence seconded Montgomery's opinion. His sentiments reflected a desire to change indigenous practices in education, in which children were instructed in "those languages which have been found most popular." Rather, he advocated that "in Government Schools.... The Persian Character, and both the Persian and Oordoo Languages should be adopted ... they are the languages of our Courts [the uniform Urdu policy had not yet been instituted] and of correspondence with the Native gentry.... The Punjabi is a barbarous dialect which if let alone, will gradually disappear." As it turned out, neither Montgomery's education proposal with its language recommendations nor Lawrence's opinion on the matter were implemented due to a parliamentary review of Company policies. Although never adopted, the Punjab government's education proposals of 1854 seem to have established that the language of administration-which by the end of 1854 was Urdu alone-would be the language of government-sponsored education in the province. When a system of mass education subsequently unfolded under the auspices of "Wood's Despatch," the language of education seemed a foregone conclusion, not even meriting discussion.

In response to parliament's call for the Company to implement a more comprehensive education policy for India's masses, the Court of Directors sent a despatch to India outlining just such a policy. Commonly known as Wood's Despatch, it called for a primary education system accessible to Indians of all classes, conducted in vernacular languages. In drawing the education system into broader networks of colonial rule, the despatch stated: "We have always been most sensitive to the importance of the use of the languages which alone are understood by the great mass of the population. These languages, and not English, have been put by us in the place of Persian in the administration of justice, and in the intercourse between the officers of Government and the people. It is indispensable, therefore, that in any general system of education the study of them should be assiduously attended to." The Court of Directors also called for the creation of a separate government department for education. Under its guidelines, the Punjab Department of Public Instruction (DPI) was formally created on 1 January 1856 and an education system set in place.

Under the charge of William Arnold, Punjab's first deputy of public instruction (and brother of famous poet, critic, and educationist Matthew Arnold), a new program of native education was established in the Punjab, with Urdu as its principal medium. From humble beginnings, the education system slowly expanded. In 1860, for example, 33,368 pupils received daily instruction in vernacular schools. As the lieutenant-governor pointed out to the Department of Public Instruction, this was a decidedly small number considering the province's population and the number of potential pupils, estimated at the time to be some 1.5 million. But these numbers steadily increased. According to DPI reports, 84,160 students were enrolled in schools in 1874-75; 105,549 in 1878-79; and 132,993 in 1884-85. The education system the DPI established was hierarchical, with village schools at its base. Then came tahsili schools (tahsil was a subdistrict, or a collection of villages), then zillah schools (which were to be situated at the headquarters of each district), and at the top of the scheme was a college in Lahore, Government College, established in 1864. While increasing enrollments suggest the success of this system, extremely low literacy rates in the Punjab throughout the late nineteenth century-by 1901 it was only 3.6 percent-indicate that the program fell far short of its aims.

While the goal of this scheme was education for the "masses," the system after the 1860s focused on the upper levels of education, not the village schools. Not only did the latter receive little government attention, but the curriculum adopted in them did little to further their popularity. As Tim Allender suggests in his study of colonial education in the Punjab, by the late 1860s, "even the most optimistic British administrator ... came to doubt the possibility of 'the masses' ever being drawn into government schools and their Western-based curriculums." This was particularly ironic since Punjab's rural masses footed much of the bill for the education system through a one-percent agricultural cess. The system's main beneficiaries in the end appear to have been Punjab's burgeoning middle classes.

The Court of Directors had opted to limit English education in 1854 for fear of creating a class of individuals whose best hope of employment was to work for the state. Instituting Urdu as the language of education and administration had much the same effect, however. As the system of Urdu education became firmly rooted in the Punjab, the colonial state recruited employees from among the graduates. The limited numbers and comparatively well-to-do backgrounds of such individuals provides yet more evidence that education was not serving the "masses" in whose interest the system had been established. Indeed, an 1868 report on education highlights this fact: "[Village schools] teach the Urdu and the Hindi, and many cases the Persian. It is thus seen that the Government schools cannot as yet work among the real Mass of the people, but address themselves more generally to the wants of those who seek for service in the courts of law as mukhtyars [attorneys], mahurirs [clerks], and amlas [the head native officer of a judicial or revenue court]."

That the benefits of the colonial state's education system accrued to a limited and privileged strata of society does not dilute its other significant effects. These were felt not only by those few who enjoyed access to education as students, but also by broader segments of Punjabi society through the manifold activities of the Department of Public Instruction. One of this department's ancillary activities was active patronage of Urdu literary production. Needing Urdu language books suitable for use in the classroom, the DPI set aside resources to produce this literature itself. It took its cue from the Education Committee, an all-India body that had charged provincial education departments with the task of creating a "vernacular literature" for use in schools.

Colonial Literary Patronage

The Punjab DPI promoted the production of vernacular materials that could be used as texts largely through its subsidiary department, the Government Book Depot. Scholars have documented the impact of government sponsorship on literary production in NWP and the Madras Presidency, but little attention thus far has been paid to the role of the colonial state as a sponsor and patron of Urdu literature in the Punjab. Promotional devices for Urdu literature included such things as rewards for commendable works, the creation of textbook committees to oversee the production of Urdu school books, hiring translators, publishing and translating works suited for use in schools, and publishing and subscribing to Urdu newspapers of various descriptions.

The DPI's activities in 1870 provide a case in point. During this one year, the department undertook the following projects: it hired three translators and compilers (at Rs. 200, 75, and 50 per month) to produce works suited for educational purposes; it spent Rs. 20,820 on rewards and salaries to authors of Urdu books, Rs. 3,264 on subscriptions to vernacular papers and periodicals, and Rs. 5,237 on approved works in English and Urdu. The DPI was not alone in these endeavors. The Punjab government directly assisted in the publication of works of general or special interest, and in some cases bore the entire cost of publication.

The Government Book Depot was particularly important to the DPI's efforts to procure schoolbooks, and the two departments worked in tandem to produce necessary texts. In careful consultation with the DPI, the Book Depot engaged in the production of Urdu texts, the translation of books into Urdu, the purchase of books, and the supply of books to government institutions. Statistics from the report on public instruction for 1860-61 give some indication of the impact the Book Depot's activities had on Punjab's book market. That report states the Book Depot bought 56,288 vernacular language books that year, of which 11,071 were printed at the government press. The remaining 45,217 were procured from private presses. At a time when the book trade was still in its relative infancy, the purchase of over 45,000 books likely made the Book Depot the biggest purchaser of books in the Punjab.

Precisely because the Punjab government became a patron of Urdu literary production, leading Urdu poets and writers of the day sought their fortunes in Lahore. The government's patronage drew men like Muhammad Husain Azad (b. 1830) and Altaf Husain Hali (b. 1837) in the late nineteenth century, two men whose power "over Urdu literature and criticism has been unequaled ever since." Both men had been based in Delhi in 1857, and found life there untenable after the British crushed the rebellion and with it Delhi's Urdu literary culture. Azad's family had supported the rebels, and when the British retook the city his father was executed and he was exiled. For two years he wandered across India-to Lucknow, Madras, Bombay, and other places-and in 1861 ended up in Lahore, where he was able to obtain a low-level job in the postmaster general's office. For the next three years, Azad actively sought a job in the DPI. He had composed his first book, a textbook for girls, expressly for this purpose. Azad finally secured his desired job and in 1864 was appointed a clerk in the DPI. This became a stepping-stone for him, and he later took employment at Lahore's Oriental College. Hali, too, came to Lahore in 1870 looking for work and secured a job with the Punjab Government Book Depot.

As significant as those government jobs, however, was the attraction of an environment fostered by Lahore's colonial officials that nurtured Urdu literature. The Anjuman-i-Punjab, or Punjab Association, was significant in this regard. Dr. Leitner, a prominent scholar of Arabic, Persian, and Urdu who headed Lahore's Oriental College, started the organization in 1865 to promote Oriental learning and vernaculars. The Anjuman was something of a literary society, and among its many activities it sponsored mushairas, or poetry gatherings, for Urdu poets. One series of such mushairas in the 1870s provided the occasion for Azad to put forth a stunning critique of Urdu poetry, one that marked a radical break with a literary past and the embrace of a new, modern literary sensibility. Azad's performance at this mushaira marks an important moment in the history of Urdu poetry. More significantly, such occurrences underscore the colonial state's efforts to foster Urdu literary production. Alongside providing jobs to Urdu litterateurs, sponsoring the production of a literature for use in schools, and awarding prizes for commendable works, the state also patronized Urdu through civil society institutions such as the Anjuman-i-Punjab. According to Pritchett, the Anjuman-i-Punjab "was actively supported by leading British officials, including the commissioner, the deputy commissioner, officers of the Department of Public Instruction, and even the lieutenant governor himself.... Most of its thirty-five original members were directly employed by the government." It was the Anjuman-i-Punjab, and Dr. Leitner's patronage in particular, that encouraged Azad's career.

The promotion of Urdu by the Department of Public Instruction, the Government Book Depot, and other government departments and organizations such as the Anjuman-i-Punjab underscores the impact of official language policy in various arenas of Punjabi society. Once the colonial state established the vernacular language it would use in the Punjab, state institutions used their resources to promote it. The impact of this should not be underestimated, particularly regarding the colonial state's role as a purchaser of Urdu texts and newspapers. Others have noted that the state's purchasing power in the late nineteenth century was a key component in the success or failure of indigenous publishing ventures in other provinces. Ulrike Stark, for example, has examined the role of the NWP government in the fortunes of the Naval Kishore Press in Lucknow, by far the biggest publisher in north India in the late nineteenth century. In the late 1850s, the colonial state commissioned Naval Kishore to produce schoolbooks in Hindi, Urdu, and the classical languages, and, subsequently, became the primary purchaser of those works. Although similarly detailed information is not available for late nineteenth-century Punjab, it is clear from records relating to the Government Book Depot that the Punjab government purchased thousands of Urdu texts per year throughout the late nineteenth century. It is reasonable to assume that the Punjab provincial government played the same critical role in the cultivation of Urdu literature and publishing as provincial governments in other parts of India played in fostering and influencing their designated vernacular literatures. Certainly, given the literary activity in Punjab's incipient public sphere, it seems the colonial state's influence loomed much larger in this regard than previously thought.

Punjabi on the Margins

So in fact the colonial state did not "overlook" Punjabi when it established its vernacular language policy for the Punjab, nor was it ignorant of the linguistic practices of the region's inhabitants. Colonial knowledge produced both before and after the language policy was set clearly documented Punjabi as the primary spoken language in much of the province. By instituting Urdu as the official language of provincial administration, the policy's architects were clearly committed to changing indigenous practice. That the state was trying to effect such change in Indians is not surprising; it engaged in similar projects, on a greater and lesser scale, throughout the nineteenth century and into the twentieth century with varying degrees of success. What is compelling about colonial language policy in the Punjab is how it was successful in some arenas and left others largely untouched.

That success can be gauged by Urdu's dominance in the region's print culture. As noted earlier, Urdu dominated commercial newspaper publishing. This is particularly significant because of the key role of the press in constituting the public sphere, even in colonial contexts. Once Urdu was established as the language of government, it became logical to use it as the language of an incipient public sphere. Urdu also accounted for the largest share of book publishing in the region due to the state's patronage of Urdu texts and its purchase of thousands of books a year for use in schools. Yet, the colonial state's purchasing power alone does not explain Urdu's dominance; colonial language policy had a tangible impact on literary production as well. Genres previously composed in Persian were now composed in Urdu. If colonial language policy was meant to influence indigenous practices, then the adoption of Urdu for certain genres of literary production-notably history-is another signal of the policy's success. It even succeeded in marginalizing Punjabi, though perhaps more in colonial discourse than in the everyday experience of Punjab's inhabitants. But if one of the policy's intended effects was to see Punjabi (that "barbarous dialect," in John Lawrence's words) "gradually disappear," then it was largely unsuccessful.

The language policy put in place by Punjab's colonial administrators accorded the Punjabi language no status. In an interesting twist, however, early colonial correspondence suggests that officials in Lahore and Calcutta did not share the same views on this matter. The Governor-General's Office in Calcutta clearly considered Punjabi requisite to effective administration. In the early 1850s, just as the first language policy was being implemented by the Board of Control in Lahore, the Governor-General's Office asked the board to create a committee to test colonial administrators in Punjabi. The board's response illuminates not only the difficulty it faced in fulfilling the governor-general's request, but reaffirms the personnel problems it would have faced (initially, at least) had it tried to implement Punjabi as the official vernacular of government. On behalf of the board, its secretary P. Melville responded to the Governor-General's Office: "I am directed to say that up to the present time there is no officer stationed at Lahore who has passed in Punjabee, but that there are two missionaries well acquainted with the language who are willing in the absence of qualified officers ... to take part in examinations.... A committee consisting of one of the Missionaries referred to, two officers who have passed Interpreter's Examinations and are known to have a good colloquial knowledge of Punjabee ... may be assembled to examine in Punjabee." The two missionaries were the Reverends Newton and Forman of the American Presbyterian Mission. It was subsequently found that Forman was, in fact, "wholly unacquainted with the language [Punjabi]," and thus Newton was the only European in the province deemed capable of administering a written examination. In response to this news, the Government of India urged the board to constitute whatever committee it could, since it was eager for examinations to be held on a regular basis. That way, it suggested, a cadre of government officers capable of participating in a Punjabi examination committee would be created. By commencing examinations as soon as possible, the governor-general felt confident that "there will soon be no difficulty in finding competent examiners."

Given colonial conceptions of the Punjabi language, it is perhaps no surprise that the format the Governor-General's Office outlined for the exam privileged the Gurmukhi script and Sikh sacred literature. The competency exam it outlined had four parts: translation from Punjabi to English, translation from English to Punjabi, grammatical questions, and conversation. The textual translations prescribed for the exam included selections from the janam-sakhis, and the exam was to be administered only in the Gurmukhi script. Whether the board ever constituted the committee or administered exams is unclear, but what we know from the subsequent archival record-that officials constantly wrote of the imminent demise of Punjabi and how Urdu would supersede it-suggests the board did not implement the governor-general's recommendations.

Even if the board was not inclined to accord the Punjabi language official status or to train and test colonial officials in the language, administrators on the ground soon realized that they could not successfully administer justice, collect revenue, or carry on other government functions without recourse to Punjabi. If colonial officials were to communicate with the local population, then Punjabi was critical. Punjab's Police Department, for example, considered Punjabi the vernacular necessary for service, and Punjabi proficiency was required of all police officers irrespective of the official language policy. The same was true of Punjab's Department of Public Instruction. Although the DPI had designated Urdu the language of the government's mass education scheme in the province, it required its officers to be proficient in Punjabi so they could communicate and implement DPI goals at the local level.

Colonial language policy in the Punjab thus had two tiers, one official and the other unofficial: Urdu was the official provincial language, serving as the language of record in courts, local administration, and education; Punjabi was necessary to effectively communicate and implement government aims. It is clear that colonial personnel had to learn Punjabi if their jobs involved interacting with the populace. By the century's end, the need for Punjabi proficiency was institutionalized, and members of the Punjab civil service were tested for fluency in spoken Punjabi. But official language policy was never changed to reflect this. Thus, although Punjabi was crucial to the practice of colonial administration, Urdu remained the province's only official vernacular language for the rest of the colonial period, and Punjabi would remain marginal to official discourse.

Colonial language policy that promoted Urdu thus had a decisive impact in the Punjab, on both its print public sphere and its patterns of literary production. But the one impact many colonial officials anticipated as a natural concomitant of their policy-the atrophying or death of Punjabi-did not occur. Punjabi remained the Punjab's principal colloquial language and the language of its literary culture, which thrived in the late nineteenth and early twentieth centuries. Given the nature and intent of colonial language policy in the Punjab, it is tempting to read Punjabi's survival as a form of resistance to colonial rule. This is not borne out by the evidence, however. Rather, Punjabi traditions are better understood through the rubric of resilience, an understanding that foregrounds how Punjabi helped constitute a broader vernacular culture that was relatively independent from, rather than overtly resistant to, colonial power.