December 28, 2007

Language in Pakistan

In the recent coverage of Pakistani political strife, I haven't seen much about the troubled linguistic and ethnic situation there. This is probably because the official ideology is to deny its existence. Thus according to Tariq Rahman ("Language and Politics in a Pakistan Province: The Sindhi Language Movement", Asian Survey, 35(11) 1105-1016, 1995):

Pakistan is a multilingual country with a population in 1994 of about 128 million. While multilingualism is not denied -- though the 1981 census contained no question on language -- the state denies the multinationality thesis endorsed by ethnonationalist leaders. The classical form of this thesis, argued by Gankovsky, is that there are four major nationalities in Pakistan: the Punjabi, Sindhi, Pakhtun, and Baluchi (Bangladesh, the former East Pakistan, has been left out here). To this list, the Siraiki was added in the 1960s and an effort to make Muhajir a nationality began in the 1980s. The official point of view is that there is one Pakistani nation united by the bonds of Islam and the national language, Urdu.

But Urdu is the native language of only about 7% of the country's population. Its imposition has caused "language riots" since the country's foundation, and language-focused unrest was at the center of the struggle that resulted in the former East Pakistan breaking away as Bangladesh in 1971. Other struggles about language policy have continued since that time, especially with respect to the role of Sindhi, the language of the region where Benazir Bhutto's Pakistan Peoples Party has its central power base.

Tariq Rahman ("Language policy, multilingualism and language vitality in Pakistan", Trends in Linguistics, 175:7 3-106, 2006) gives this language distribution in Pakistan as of the 2001 census:

Language Percentage of speakers
Punjabi
44.15
Pashto
15.42
Sindhi
14.10
Siraiki
10.53
Urdu
7.57
Balochi
3.57
Other
4.66

As Alyssa Ayres points out ("The Politics of Language Policy in Pakistan", in Brown and Ganguly, eds., Fighting Words: Language Policy and Ethnic Relations in Asia, MIT Press):

Conflicts over language identity are not merely about language: They are intertwined with struggles over power and access to it. The vast majority of Pakistan's rulers and policymakers have been Punjabi and mohajirs (settlers), while the military has been ruled by a Punjabi-Mohajir-Pathan nexus.

This leaves out about a third of the population in the remaining western half of the country, with the inhabitants of Sind at the head of the list. Ayres again:

The story of Pakistan's Sindhi language movement (and language riots) parallels that of the Bengali language movement from partition in 1947 through Benazir Bhutto's first regime (1988-90). During Bhutto's first term in office, tensions between Karachi's numerous ethnic groups exploded. ...

Sindhi, like Bengali, enjoyed regional hegemony throughout the time of the British Raj. It has long had a literature and a widespread presence both colloquially and administratively. ... Sind had been a separate province during the Raj. This was due in part to the Sindhi language movement of the 1930s, which had resulted in Sind separating from the Bombay presidency in 1936. This institutionalization of a Sindhi ethnic identity linked directly to language was therefore in place even before partition. Partition would trigger Sindhi ethnic mobilization for two reasons: cultural insensitivity and economic subjugation.

Partition brought massive demographic changes to the subcontinent. Karachi in particular saw an enormous influx of migrants from Uttar Pradesh -- the Urdu-speaking Mohajirs -- as well as from Punjab, Baluchistan, and the NWFP. At the same time, Hindus, who had comprised 64 percent of the population of Sind prior to partition, fled to India. Homes and possessions left behind in Karachi and other urban centers (as well as agrarian lands in the Sindhi interior) were claimed by Mohajirs. The results were striking: In Karachi, Mohajirs comprised 57.55 percent of the population in 1951; in Hyderabad, 66.08 percent. These cities were effectively cleaved in half and then populated by strangers.

When Pakistan came into being, Sindhis, like Bengalis, were surprised to find their language would be subservient to Urdu in the national order. This lower status offended Sindhi cultural pride. The problem was exacerbated by the inherent advantage afforded the newcomer Mohajirs, whose mother tongue was the national language; this gave Mohajirs a considerable advantage in seeking government employment.

A crucial event early in Pakistan's history helped to precipitate Sindhi-Urdu tensions: On July 23, 1948, the provincial government of Sind offered the city of Karachi to the federal government for use as the new capital of Pakistan. The federal government, headed by Jinnah, accepted the offer and then decided to reconstitute the city as a federal territory. When M.A. Khuhro, then chief minister of Sind, objected, he was dismissed by Jinnah on grounds of being both a poor administrator and a corrupt government official. Karachi thus became a federal territory with a heavy Urdu presence. Most important, however, the economic and cultural capital of Sind was perceived as having been hijacked by the Pakistani state. From the Sindhi point of view, these developments created a painful inequality: To obtain government jobs, Sindhis would have to learn a "foreign" language. At the same time, the newly arrived "foreigners" (i.e., Mohajirs) did not have to learn Sindhi to go about their daily lives in urban Sind, where most of them lived. There was no compelling reason for Mohajirs to integrate with Sindhis -- a situation that struck the latter as highly discriminatory.

There's a lot more to the story, but this should be enough to give you the general idea. And it's not only in Sind that language policy is a social and political issue, according to Rahman (see the original text for references):

[T]he privileging of Urdu by the state has created ethnic opposition to it. However, as people learn languages for pragmatic reasons, they are giving less importance to their heritage languages and are learning Urdu. This phenomenon, sometimes called ‘voluntary shift’, is not really ‘voluntary’ as the case of the native Hawaiians, narrated by Daniel Nettle and Suzanne Romaine, illustrates. What happens is that market conditions are such that one’s language becomes a deficit in relation to what Pierre Bourdieu, the French sociologist, would call ‘cultural capital’. Instead of being an asset it becomes a liability. It prevents one from rising in society. In short, it is ghettoizing. Then, people become ashamed of their language as the Punjabis, otherwise a powerful majority in Pakistan, are observed to be by the present author and others ... Or, even if language movements and ethnic pride does not make them ashamed of their languages, they do not want to teach the language to their children because they think that would be overburdening the children with far too many languages. For instance, Sahibzada Abdul Qayyum Khan reported in 1932 that the Pashtuns wanted their children to be instructed in Urdu rather than Pashto. And even this year (2003), the MMA government has chosen Urdu, not Pashto, as the language of the domains of power, including education, in the N.W.F.P. The same phenomenon was noticed in Baluchistan. Balochi, Brahvi and Pashto were introduced as the compulsory medium of instruction in government schools in 1990. Language activists enthusiastically prepared instructional material but on 8 November 1992, these languages were made optional and parents switched back to Urdu. Such decisions amount to endangering the survival of minor languages and they devalue even major ones but they are precisely the kind of policies that have created what is often called ‘Urdu imperialism’ in Pakistan.

In short, the state’s use of Urdu as a symbol of national integration has had two consequences. First, it has made Urdu the obvious force to be resisted by ethnic groups. This resistance makes them strengthen their languages by corpus planning (writing books, dictionaries, grammars, orthographies etc) and acquisition planning (teaching the languages, using them in the media, pressurizing the state to use them ...). Secondly, it has jeopardized additive multilingualism as recommended by UNESCO ... As Urdu spreads through schooling, media and urbanization, pragmatic pressures make the other Pakistani languages retreat. In short, the consequence of privileging Urdu strengthens ethnicity while, at the same time and paradoxically, threatens linguistic and cultural diversity in the country.

Perhaps the most important result of resistance to Urdu is increase the importance of English, as discussed by Ehsan Masood ("Urdu's last stand", 9/1/2007). Benazir Bhutto's Urdu was notoriously poor ("Benazir's poor Urdu inspires many a joke", ExpressIndia, 12/4/2007), and Pervez Musharraf's Urdu is apparently also often criticized, though Urdu is his native tongue.

[After writing this post, I saw that Tristan Mabry has an excellent Op-Ed in this morning's Philadelphia Inquirer, "In divided Pakistan, not all are mourning Bhutto". Tristan (a friend from his time as a grad student at Penn) knows these issues well. The quality of the information in his article is a striking contrast to the (complete?) omission of these matters from the rest of the media coverage of the situation in Pakistan, both before and after Bhutto's assassination. It would be nice to see his expertise published in more prominent newspapers, and mixed in with the usual talking heads on the national broadcast media.]

Posted by Mark Liberman at December 28, 2007 08:12 AM