Pakistan is a multilingual state the national language of which, Urdu, is the mother tongue of only a minority (7.6 per cent). Moreover, this minority (the Mohajirs) emigrated from India after the country was carved out of British India in 1947.

The official language of the state is English, the language of South Asia's erstwhile rulers - the British. Pakistan has seen a number of language-based ethnic movements in its short history (Rahman 1996). One of them, the Bengali Language Movement, of 1948-52, eventually led to the emergence of Bengali ethno-nationalism which led to the breakup of Pakistan into Bangladesh and Pakistan in 1971. After that, in January 1971 and July 1972, there were riots between the Urdu-speaking Mohajirs and the Sindhi-speakers after which antagonism between the two communities increased and the Mohajirs emerged as yet another nationality in Pakistan at the behest of the Mohajir Qaumi Movement (MQM). The province of Sindh 1 , home to both Mohajirs and Sindhis, has seen much ethnic tension since the middle 1980s when the MQM became a militant force to be reckoned with. Other ethnic groups in Pakistan also see language as a marker of identity. This constitutes the ethnic level of language as - a subject to which we will turn again.


Apart from the ethnic aspect of language, there is also the problem of administration, education, higher commerce, media - in short, the domains of power in the state and the civil society. Here too language is s sensitive issue in Pakistan. The use of English favours the Westernized elite while the use of other languages would bring in other candidates for power. This may be called the class question i.e., the way in which language relates to socio-economic class in Pakistan. Connection with this constitutes the individual level of empowerment through language i.e., how individuals seek to empower themselves by learning the languages of the domains of power - the civil and military bureaucracy, judiciary, education, commerce, media and so on. Connected with all these questions is the issue of ideology : How languages are used to further the ideological aims of elites and counter-elites? How they are associated with certain ideologies.

Survey of Literature

These questions are raised and answered in heated and highly polemical newspaper debates in Pakistan. The approach to the role of language in ethnicity has been through conspiracy theories.

Since the 1950s, when the Bengali Language Movement challenged the West Pakistani domination of the former East Pakistan, the people and the press in West Pakistan agreed that this was the work of Hindus, communists and anti-state elements who wanted to destabilize the state. Another theory, coached in equally unsophisticated and polemical terms which has been used to explain ethnic nationalism, is a version of the primordialist theory. It is assumed that people are born with a fixed identity - Punjabis, Pathans, Sindhis, Balochis etc - and, instead of becoming modernized and identifying with Pakistan as a whole, they remain 'backward' and insular. They never rise above their provincial identities and practice forms of nepotism and 'tribalism'. This theory reduced ethnicity to 'provincialism' and was used as the standard explanation for the Sindhi, Pashtun, Bengali and Baloch ethno-nationalism during the Ayub Khan era (1958-1969).

So strong was the hold of such theories on the minds of the intelligentsia of Pakistan that serious studies explaining ethnicity emerged only recently. One of the first such attempts was Tahir Amin's study of the ethno-national movements of Pakistan. Tahir Amin used modern theories of ethnicity, especially the instrumentalist theory, to explain that ethno-national movements are the products of the demand for a just share in goods and services in a modern state (Amin 1988). However, Tahir Amin's reference to language is inadequate and incomplete. Later, Feroz Ahmed, a Sindhi left-wing intellectual, wrote several articles on Mohajir, Pashtun and Sindhi nationalism which were later published as a book (Ahmed 1998). He wrote on the role of Sindhi and the language riots in Sindh but, by refusing to accept Mohajir ethnicity, he could not give an objective account of ethnic identity-construction in the light of the latest theories on the subject (see Hutchinan & Smith, 1996). That, indeed, is the problem of M.S. Korejo whose recent study on G.M. Syed, the leading Sindhi nationalist leader, fails to rise above the polemical level where the Mohajir identity is concerned and brings no fresh evidence on the role of Sindhi in the Sindhi ethnic identity formation or its assertion (Korejo, 2000).

Indeed, so little has the role of language been studied in the context of ethnic movements that Anwar and Afia Dil, a husband and wife team, published their history of the Bengali Language Movement only in 2000 (Dil & Dil, 2000). While this book provides historical details and draws on Bengali literature which no other writer in Pakistan does, it is short on theoretical insights into identity-formation and ethnicity.

The present author's book, Language and Politics in Pakistan (1996), presents an analysis of the role of language in the ethnic movements of Pakistan coming to the conclusion that language becomes an identity symbol under modern conditions when different collectivities compete for power and resources. The possibility of increased communication facilitate the manipulation of larger labels for group identities such as religion or language. These labels supercede, or push into the background, such pre-modern and smaller labels as kinship, tribal, class and occupational labels or markers of identity. Thus, while pre-modern Siraikis of South Punjab saw themselves as Multanis, Riasatis (inhabitants of the state of Bahawalpur) and so on, the term Siraiki is used for the whole collectivity now. In short, ethnic identity is constructed just as nationalist identities were constructed in Europe because of the presence of collective symbols, especially uniform and standardized print languages, as Benedict Anderson (1983) has argued.

Apart from the role of language in identity construction, there is the issue of its use in education and attitudes towards it. In this context, Shemeem Abbas has written on the strong presence of English in education and other domains in Pakistan (Abbas, 1993). Sabiha Masroor has carried out a survey of Punjabi students' attitude towards languages and comes to the conclusion that they rank English highest ; Urdu comes second and at the bottom is their mother tongue, Punjabi (Mansoor 1993). Apart from that there are some studies of the teaching of English (Malik, 1996) and the use of Urdu, both formally and informally, outside Pakistan (Javed, 1996).

But, there being so few linguists in Pakistan, there is not much scholarly research in the highly interesting field of the relationship of language with power, ideology and other interesting issues in Pakistan. This article is an attempt to fill that gap.

Background Information

The last census of Pakistan was held in March 1998 but its results have yet to be published. The census figures which are available are those of 1981. In that census the question asked was about the language 'commonly spoken in the household'. The results, expressed in percentages, are as under :

Punjabi 48.17 per cent
Pashto 13.14 per cent
Sindhi 11.77 per cent
Siraiki 9.83 per cent
Urdu 7.60 per cent
Balochi 3.02 per cent
Hindko 2.43 per cent
Brahvi 1.21 per cent
Others 2.81 per cent

Source : Census 1981.

Among the 'others' are more that 50 languages or dialects, most of them unwritten, which are given in Annexe 1.

The census does not mention English, Arabic nor Persian. But English is the key to power as far as the modern, employment-based domains of power are concerned. Without knowing English one cannot enter the most lucrative and powerful jobs, both in the state apparatus and the private sector in Pakistan. There are no reliable figures of the number of Pakistani who can use English. The Census of 1961, however, gave the figure of 2.7 per cent of the population (Census 1961 : IV, 30-32). This percentage should have increased because the middle class- or, rather the 'salariat' as defined by Hamza Alavi (1987) - has increased and the higher jobs require some competence in English. If those who have passed their matriculation examination, in which English is a compulsory subject, are considered to have literacy in English then the figure comes to 19.56 per cent in 1981 (Census 1981 : Table 4.6, p. 31). However, most matriculates from vernacular-medium schools cannot speak English and can barely read their textbooks which they tend to memorize. As such, those with fluency in English could hardly be more than 3 to 4 per cent of the population.

Urdu 2 , however, is much more widespread and not only the 20 per cent matriculates are quite proficient in it, but also the students of religious seminaries, madrassas, soldiers, as well as otherwise illiterate working class people living in cities pick it up and use it quite well. Urdu has spread so widely because it is used for inter-provincial communication, entertainment, media (T.V, radio, newspapers) and, above all, lower middle-class jobs all over Pakistan except in rural Sindh.

Arabic is understood only by a handful of religious people in the madrassas and a few academics and scholars connected with Islam or Arabic language and literature. Although Muslims learn to read the Quran - the 1981 census reported that 18.37 persons could read it (Census 1981 : Table 4.7, p. 33) - this reading is no more than recognition of the Arabic letters. They are not taught the meanings of words nor can they read Arabic words written without the diacritical marks used in the Pakistani versions of the Quran.

Persian too is only understood by a few experts. It is taken as an easy option by students in certain examinations leading to state employment but in general the students never get beyond memorization of a few passages.

The indigenous mother tongues of the people are either not taught at all (Punjabi etc) ; taught inadequately (Pashto, which is the medium of instruction up to class 5 in some schools and an optional subject in higher levels) or taught only in a certain area (Sindhi, which is taught in Sindh). However, some people learn them out of their own interest because books written in them, called chapbooks, are available in all the major cities of Pakistan. An American scholar, William Hanaway, and Mumtaz Nasir listed 940 chapbooks in Punjabi, Siraiki, Hindko, Khowar, Pashto, Sindhi, Persian and Urdu (Hanaway & Nasir, 1996 : Appendix A 441-615). Films and songs in these languages, especially in Punjabi and Pashto, are quite popular too.

English, Urdu and Sindhi are media of instruction in schools corresponding to a class-based division of Pakistan society. The elitist English-medium schools, where the teachers really teach in English and the students come from elitist backgrounds with exposure to English, are so expensive as to exclude lower-middle and working-class pupils. The Urdu and Sindhi-medium schools, as well as the few schools where Pashto is the medium of instruction at the lower levels, are run by the state and are quite affordable for most Pakistanis. Even more affordable, because they provide not only free education but even free board and lodging, are the madrassas which have central bodies which examine students in Urdu and Arabic. However, the madrassas of the Pashto-speaking areas use Pashto as the medium of instruction while those of the Sindhi-speaking parts of Sindh use Sindhi. In the Punjab and Balochistan, although Urdu is the formal medium of instruction, the explanation is often in the local language.

Data about the number of schools according to their medium of instruction is not available. The following chart is based on partial information about some provinces and the assumption that all ordinary state schools in Punjab, Azad Kashmir, Balochistan and the N.W.F.P use Urdu as the medium of instruction. The chart is as under :


Table 1
Provinces and Language of Instruction
Islamabad NWFP Sindh Balochi-stan Azad
FANA Federal Government Total
67,490 368 13,556 3,657 9,939 6,009 1,370 167 102,556
Nil Nil 36,750 Nil Nil Nil Nil Nil 36,750
Pashto Medium
(up to 5h only)
Nil Nil 10.731 Nil Nil Nil Nil Nil 10,731
No Information (NI) NI NI 1,394 NI NI NI NI 1,394
NI 48 NI 95 NI NI NI 82 225
Total 67,490 416 24,287 41,896 9.939 6,009 1,370 249 151,656

Source :

Sindh : SEMIS, 1998 ; Punjab : EMIS, 1998 : Tables 122 ; All other figures are from the Academy of Educational Planning and Management, Islamabad, 1999. Figures for Federal Government includes schools in military (or cantonment) areas for which the source is the Directorate of Federal Government Institutions, Rawalpindi.

There is an indeterminable number of madrassas as report by the press to be 6,761 (Nawa-i-Waqt October, 1999) while the government sources still quote the figure of 3,906 from 1995 (Directory, 1995 : 282). As mentioned earlier, the madrassas preserve Arabic more as a symbol of continuity with the past and of Islamic identity than a living language, though most of their graduates cannot function in Arabic (Rahman, 1999b). They can, however, function in Urdu which has spread through the madrassa network ever since the nineteenth century (Metcalf 1982) and is now associated with Islam and the Muslim identity in both Pakistan and India (Rahman, 1999a).

With these facts about languages and their use in Pakistan in mind, let us study how language is linked with issues of power and ideology in the country.

Language and Power : The Ethnic Dimension

Although Urdu was first taught formally by the British, it became an identity symbol of elitist Muslims (the ashraf) in India in the nineteenth century. This Urdu was highly Persianized in a process described by Amrit Rai (1984). As the British patronised Urdu, it was the language of the lower domains of power - administration, judiciary, education and commerce - in the former centres of Muslim power (especially in U.P.). In time, partly because of the competition for jobs and power between the Muslim and Hindu 'salariats' Urdu came to be identified with the Muslim identity while Hindi was identified with the Hindu identity (King 1994). The Urdu-Hindi controversy of the nineteenth and twentieth centuries was really part of the power struggle between the Hindus and the Muslims in undivided India (Rahman, 1996 : Chapter 4). As the controversy itself has been described in great detail (Brass, 1974 : 119-181 ; Gupta, 1970 ; Kerrin, 1972), we need not go into it here. It has been mentioned in order to make the point that Urdu was part of the ideology of Muslim separatism in India. The Muslim League government of Pakistan inherited this ideological view of Urdu and felt convinced that it could be a force, a major symbol of national integration, in the new country. As the federal government was located in West Pakistan and the bureaucracy as well as the military were dominated by Punjabis and Mohajirs, Urdu came to be associated with the hegemony of the centre. This meant that Punjabis were perceived by the people of East Bengal, Sindh, Balochistan and the North West Frontier Province to be dominating them.

The ethnic nationalists of these provinces reacted by consolidating their ethnic identities of which language was an important part. They also resented the domination of Urdu and launched movements of ethnic assertion which had strong components of language activism. The strongest movement was the Bhasha Ondolan, the Bengali language movement of 1948 and 1951. The movement itself has been described elsewhere (Alam, 1991 ; Umar, 1970) and need not detain us. The important point is that the state's ideology was challenged and language was the symbol used for the mobilization of a counter ideology - that of ethnic assertion. Although the language movement itself was diffused by 1955, it resonated throughout the whole period of united Pakistan. Indeed, it provided the symbolic force and cumulative antagonism against West Pakistan which finally culminated in the creation of Bangladesh.

The state's ideology of using Islam and Urdu as symbols of integration did not change even after the loss of East Pakistan though Pakistani historians now admit that policies were wrong adapted in the Eastern Wing (Zaheer, 1996). Thus language movements in reaction to perceived domination by the Urdu-supporting ruling elite caused the Urdu-Sindhi riots in Sindh in January 1971 and July 1972 (Rahman, 1996 : Chapter 7). The only language movement which decreased in intensity was the Pashto language movement. It decreased because the Pashtuns extended their trade and transportation network all over Pakistan and entered the salariat, especially the army, in fairly large numbers. Even so the ethnic political parties, such as the Awami National Party, show their resentment of the domination of the Punjab by trying to maintain their separate Pashtun identity of which Pashto is a part. Hence they keep up the demand for teaching Pashto and giving it more importance (Rahman, 1996 : Chapter 8).

The Siraiki language movement is probably a response to under development in southern Punjab. It is not a powerful movement being confined to the intelligentsia of this region. However, since the label of Siraiki has been used by the state for the language of this area, the people of southern Punjab use this term for their language rather than the terms 'Multani', 'Derewali' and 'Riasati' etc., which they used earlier. In short, for the theoretician, this movement provides insights into the phenomenon of identity-formation. The whole movement from local identity - such as Multani or Riasati - to a larger ethnic identity (Siraiki) is clearly exemplified by this movement (see Shackle, 1977 ; Rahman, 1996 : Chapter 10).

The other language movements, although small, are also responses to the state ideology of using Urdu as a symbol of Pakistani identity. For instance, in Balochistan the Balochi and Brahvi languages and the resistance literature written in them are seen as a response to the dominance of Urdu, the Punjabi ruling elite, and the Urdu-based Mughlai culture of north India (Rahman, 1996 : Chapter 9). Even in the Punjab, much maligned though it is by activists of all ethnic movements, there is a small movement to give importance to Punjabi. The movement is confined to a small intellectual circle of Lahore and a few other cities of the Punjab. Some of these intellectuals argue that the state, though dominated by Punjabis, uses Urdu and English because these two languages are the languages of elitist discourse and facilitate the elite to acquire power through employment. Moreover, Urdu is patronised in order to consolidate the elite's rule over the federating units even if this domination is at the cost of the mother tongue, Punjabi. This, they feel, is too heavy a price to be paid for maintaining Punjabi hegemony (Mirza, 1994).

Language and Individual Empowerment

The demand for learning a language is linked to empowerment. People demand a language if, after learning it, they can enter the domains of power through employment. This excludes that large number of Pakistanis whose power depends on owning land, real estate or by their position as heads of tribes or other groups. It also excludes the leaders of political or religious organizations who can often manipulate the written word but whose source of power is the faith of their followers or undefinable charisma. Even after these exclusions there are large sections of the population, mostly of the urban population, who derive their power, at least partly, from their ability to manipulate the written word in English and Urdu - the languages of the domains of power in Pakistan. This power is not directly proportional to one's competence in the languages but without the ability to read, write and speak these languages one cannot enter the elite cadres of the Pakistani salariat.

Language, then, is a coin and what it buys in the market is power. If one cannot write Urdu and English, one cannot get even clerical jobs in Pakistan except in Sindh. If one can write Urdu but not English one can get lower jobs in all the provinces of Pakistan. Higher jobs, however, are reserved for those who can read and write English. This state of affairs is related to the pattern of the distribution of power. When the Mughals ruled India, they used Persian in the domains of power forcing Hindus to learn this language and become Muslimized in culture (Faruqi 1998). Later, in 1837 when the British did away with the ascendancy of Persian by substituting English in its place in the higher domains of power, both the Hindu and Muslim elites switched to English and Persian declined (Faruqi 1998 ; Rahman 1999c). Still later, in Pakistan the westernized ruling sections of the salariat, which dominated the modernized sections of the armed forces and the civil bureaucracy as well as the media and the commercial institutions, did not allow either Urdu or any other Pakistani language to take the place of English. The ascendancy of English, therefore, reflects the ascendancy of those who happen to be powerful at the moment - the Western-trained cadres of the Pakistani salariat.

Contrary to popular perception, the languages of power - whether Persian or English - have never been imposed by the rulers of South Asia. Instead, they have been denied to the masses and rationed out to the middle classes. It is because they are seen as being empowering that individuals and groups demand them in the first place. In the Medieval Age, this demand was partly met by philanthropist individuals and institutions. For instance, the founder of the Sikh religion, Guru Nanak, was taught Persian by a Muslim philanthropist neighbour (Khan, Vol. 1, 1789 : 83).

In modern times, the state provides institutions for teaching the languages of power but in an obviously class-based and highly discriminatory manner. For instance, in Pakistan the mainstream public education is mainly in Urdu though in the Sindhi-speaking areas of Sindh it is in Sindhi. There are also some Pashto-medium primary schools in the Pashto-speaking part of the N.W.F.P. This vernacular mainstream, however, is for poor people. For the elite of power - the armed forces and state functionaries - the state has created a parallel system of education in which the medium of instruction is English for all subjects or, in some cases, all science subjects. At the top of these institutions are the cadet colleges and public schools. Most of these institutions are either directly or indirectly controlled or influenced by the armed forces. In addition to this, the armed forces run schools through their welfare organizations such as the Fauji Foundation (Army), the Bahria Foundation (Navy) and the Shaheen Foundation (Air Force). Some other state institutions, such as the railways, the customs department, the telephone and telegraph department, the police etc also run their own schools. The federal government also runs some 'model' and other schools in English. The buildings of these schools, their teachers and the facilities provided in them are much better than those provided by the government in its mainstream system. These schools charge higher fees from outsiders while students whose parents serve in the departments which run or influence the school pay less. This is how the Pakistani ruling elite itself infringes its own principle of providing education at public expense through the medium of the vernacular.

Apart from these institutions there are chains of highly expensive English-medium schools like the Froebels, the Beaconhouse and the City School System. Here the tuition fees ranges between Rs 1,500 to 3,500 per month. These schools cater for the elite of wealth. Even more wealthy people get their children educated outside the country or in the International American school which charges over US $ 10,000 per academic year.

In short, the Pakistani ruling elite itself creates and maintains a class-based, discriminatory system of schooling. In this system the majority of the population is either left illiterate or given vernacular-medium schooling which puts them at a disadvantage in the quest for empowerment vis-à-vis the elites of power and wealth. Moreover, the ruling elite does not only lack faith in its own education policy but also subverts it by investing in a parallel model of education from which it stands to benefit.

Sensing the usefulness of English as a language of power, the people go to great hardships to provide English medium education for their children. Thus there are schools claiming to be English-medium institutions all over Pakistan. They charge between Rs. 50 to Rs. 1000 per month and they provide education of so variable a quality that it defies classification. A number of religious organizations too now run such schools. They claim to combine Islamic socialization with skills in modern subjects and English.

Language Textbooks and Ideology

That language contains an ideological load bias is well known (Whorf, 1956). This, however, is not being explored here. What I would like to point out is that language and literature textbooks are used in Pakistan to disseminate ideological messages to students. Such messages are mostly given through the textbooks of history, social studies and Pakistan studies. However, language textbooks also reinforce the ideological messages in the other books.

These ideological messages are of three kinds. First, messages on Islam ; second, those on nationalism ; and third, those on militarism. All the messages consist of poems, stories, essays, exercises and so on in the textbooks of Urdu, English, Arabic, Persian, Punjabi, Pashto and Sindhi. The Islamic lessons are about the fundamentals of Islam ; Islamic personalities or events which glorify Muslim history. The nationalist lessons are about the Muslim leaders of the Pakistan Movement, the movement itself, and about Pakistan. They are written so as to create Pakistani nationalism and a sense of Pakistani identity. The third component, that of militarism, comprises lessons glorifying war in general and especially glorifying the wars between Pakistan and India in 1948, 1965 and 1971. The heroes of these wars are celebrated in many poems and lessons.

One special feature of these three ideological themes is that Islam is made to support nationalism of which militarism is taken as the chief expression and most desired value. In short, Islam is co-opted in the service of the state in a process described by Jamal Malik, though in other contexts, as the 'colonialization of Islam' (Malik, 1996). According to some analysts the purpose of this indoctrination is to create support for the state's militaristic, anti-India policies (Saigol, 1995). It has also been pointed out that Pakistani society is male-dominated, hierarchical and power-oriented and that these textual messages tend to maintain the prevalent pattern of the distribution of power (Ibid).

While a number of people, notably K.K. Aziz (1993), have looked at history and social studies textbooks for their ideological messages, the language-teaching textbooks have been explained from this point of view only by the present author. The percentages given below refer to the number of ideological items - all lessons and exercises - in the textbooks from class 1 to 10 in the public schools of Pakistan.

Table 2
Language of textbooks, in percentages
Language Content (in percentages)
Arabic 71
Urdu 40
Pashto 43
Persian 31
Sindhi 29
English 8
Source : Field Research in 1998 described above.

It should be remembered that only English and Urdu are compulsory languages in most schools. Thus, it is Urdu which is the primary ideology-loaded language in the Pakistani educational system.

Urdu is also an 'Islamic' language in the sense that under British rule it was adopted by the Islamic religious scholars (ulema) and clergymen (maulvis) to disseminate Islam (Metcalf, 1982). In Pakistan it is also the language of examination of the madrassas affiliated to Central Board such as the Deobandis, the Barelvis, the Ahl-i-Hadith and the Shias (see Nayyar, 1998 ; Rahman, 1999 : Chapter 5). It is also the language in which religious tracts are written and sermons are given. Moreover, it is the language of some of the most reactionary Urdu newspapers so that most of the material inciting people to fight in Kashmir, struggle against perceived Western domination or support Islamic fundamentalism is available in Urdu. This means that the ordinary literate person in Pakistan is much more exposed to right wing views which are meant to, and may, make him support anti-democratic and pro-war views.

Linguistic Responses to Modernity

As we have mentioned earlier, modernity, brought in by the British, increased the number of people who could be empowered through the manipulation of the written word. Earlier, the bureaucracy was smaller and it did not operate on the rational, Weberian model. Land was, and remains in the Pakistani countryside, the basis of power. Agrarian societies, with surplus produce to guard, are war-like and authoritarian (Gellner, 1988). Pakistan too is such a society. Hence the extended bureaucracy in Pakistan works more on the feudal and colonial model, which are both authoritarian and not fully rule-based, than on the European one. However, entry to the large bureaucratic network created by modernity is through literacy in the standard form of the language recognized as 'official' by the state. Since the British state recognized only English as the language of the elitist domains of power, it took the place of Persian. Thus one foreign language took the place of another such language. However, the British also recognized certain vernacular languages for official work in the lower domains of power. Among these were Urdu and Sindhi which are now used in Pakistan. The teaching of these languages in state institutions, the development of their standard, written variety and printing in them are all the consequences of modernity.

Another impact of modernity is the need to express newer forms of reality - such as new facts about the world investigated by the scientific methodology. New terms to express these new concepts had to be created. This is part of corpus planning and it has been carried on in Urdu, Sindhi, Pashto, Punjabi, Siraiki, Balochi and Brahvi in the last fifty years. The creation of these new terms are governed by certain imperatives out of which the ideological ones are important. In Urdu, for instance, the state uses Arabic and Persian roots to create new terms. This is in conformity with the state's ideological imperative of creation or emphasizing the Pakistani (rather than the ethnic) and the Islamic (rather than the secular) identity of its citizens. The ethnic nationalists, on the other hand, use the older roots of their indigenous languages to create new terms. This is because they want to emphasize their indigenous, ethnic identity rather than the uniform Pakistani one which the state would have them adopt (for details and examples of terms see Rahman, 1999 : Chapter 12).

Yet another impact of modernity is the change in world view which has come about in Pakistan. The reflection of this change as far as language in concerned takes many forms. First, the older Persian textbooks which reflected a magical, non-rational, pre-modern world view have been replaced by textbooks which reflect the modern point of view. This 'modern' view, however, is that of the ruling elites of the state. The older texts had tales in which cause and effect were not necessarily connected because the magical and the miraculous penetrated so deeply in one's life. The powers that he could curse or bless ; punish or reward ; kill or allow one to live for reasons which always remained inscrutable. This created or reinforced the idea that the world, like one's despotic rulers, was not amenable to rational or comprehensible codes of conduct. Moreover, the Persian texts were erotic and presented women as lustful, deceitful and cunning. All these ideas helped to sustain political despotism and male dominance.

The British intervention created a new kind of literature in Urdu and Sindhi. In the name of morality the medieval texts with their erotic context were condemned by the Victorian British administrators. In this they were outdone by Indian reformists, both the secular modernisers of the likes of Altaf Hasan Hali (1837-1914), and the ulema such as Ashraf Ali Thanvi. The new language and literary textbooks were more puritanical as ulema and the reformers desired. They were also more engaged with the mundane (society, morals, education, the ideal citizen, the good life etc). In short, the language textbooks of the state schools came to be controlled largely by the state and the middle class salariat which ran the state.

In Pakistan, this meant a devaluation of the indigenous languages of the people as well as their pre-modern, agrarian world view. The people were not taught their indigenous languages except in Sindh and their beliefs and practices were dismissed as superstitions. The people, however, did not leave their languages or views. They spoke them, of course, but this they could hardly help doing. What is surprising is that they learned to read their languages.

The indigenous languages of Pakistan were not officially used as media of education but books in them were written and there is evidence that they were used in educational institutions (Sheerani, 1934).

The present author has seen manuscripts or printed versions of such books in Punjabi, Sindhi, Pashto, Balochi and Brahvi in libraries and private collections in England and Pakistan. These books are about the rudimentary teaching of Islam, in praise of the Prophet of Islam and saints and about rituals of cleanliness and so on. Some of them are also about tales of legendary love affairs such as Heer and Ranjha. These books have been read by people in their houses since at least the 18th century onwards. Thus, there is an existing tradition of literacy in the indigenous languages despite official non-recognition.

This continues to this day as evidenced by the chapbooks in the indigenous languages on sale in the cities of Pakistan which have been mentioned earlier. The present author has read a number of these chapbooks. Like their predecessors, they too are about religion and legendary stories or love affairs. Some of them are also about astrology, magic, casting spells and writing letters. They cater for those pre-modern people who are still comfortable in the world of magic and pre-modern non-rational modes of thinking. However, the modern world is at hand and very powerful ; hence the need to learn to write letters in order to empower one's self.


Language is intimately related both to ideology and power in Pakistan. The state has seen Urdu as a marker of Pakistani identity and an integrative device in a country which has at least five major ethnic groups with their own indigenous languages. The ethnic groups have countered the centre which, in their perception, practices versions of internal colonialism. In this perceived counter-hegemonic struggle, the ethnic elites have used their indigenous languages as markers of identity and mobilizing devices. This struggle is portrayed as being purely ideological by the nationalist ruling elites which are Punjabi-dominated. The nationalist view is that Pakistan was carved out of British India as a homeland for Muslims and all attempts at weakening it or arguing that it is multi-national are against the ideology of Pakistan. These arguments negate the power struggle which takes place in multi-national states while emphasizing the ideological aspect of the struggle and portraying it in terms of pro and anti-Pakistan ideological commitments.

The ethnic ideologues emphasize the power struggle arguing that the fruits of power (water, electrical energy, development funds, jobs, shares in state organizations, etc.) is the crux of the issue. To this they add the ideological argument that ethnic groups or nationalities have distinct identities which enrich the federation and should not be dominated, marginalized, destroyed or assimilated to the point of disappearance.

Besides the level of the national or ethnic groups, there is the level of the social class. Language is very much a marker of the socio-economic class in Pakistan. English is associated with the upper-middle and upper class ; Urdu with the lower middle and middle class while the local, indigenous languages are the preserve of the peasantry, unskilled labourers and the working class in general. This refers more to official, formal interaction typically in urban contexts rather than to informal interaction. However, in Sindh there are areas where Sindhi is used even formally in domains in which Urdu is used in other parts of Pakistan. Being connected with class, English has the highest snob value followed by Urdu and other local languages. This equation does not remain constant all over the country. In Sindh and parts of the Pashto-speaking belt, pride in indigenous ethnic identity is strong enough to counteract the prestige of Urdu however much it may be in demand for purely pragmatic reasons.

At the individual level, the necessity of learning a standard, printed language is dictated by pragmatic considerations. People want to learn English and Urdu because, under the present circumstances, they need these languages to obtain employment i.e., to empower themselves. The knowledge of English opens the doors to the most lucrative and powerful jobs in the state, the private sector and the international bureaucracy. It also gives one prestige, influence through informal social contacts and a certain snob value. Urdu too gives access to jobs though only at the national level and not very high ones even then. It too is a marker of middle class status and a badge of education and urban socialization. In short, it too is empowering as far as middle class, salariat-based urban Pakistani society is concerned. The indigenous languages alone, other than Sindhi, do not give access to jobs in the salariat and the power which comes from them. Those who know only these languages are either restricted to working class jobs or, if they possess power, to quasi-feudal land ownership, tribal leadership, political manipulation or trade.

Besides being part of ideological claims and counter-claims and power struggles between ethnic groups and classes, languages are also associated with certain ideological biases in Pakistan. Thus, English is seen as the carrier of Western, liberal values ; Urdu is seen as an Islamic and Pakistani-nationalist language while the indigenous languages of the country are associated with ethnic nationalism and identity. However, English is being appropriated by the Islamic revivalists and under-privileged Pakistani groups who are impressed by the fact that it is a language of empowerment in the modern world.

To sum up, language is a very important factor in modern Pakistan. It is the key to the complex issues of the distribution of power between the ethnic groups, the socio-economic classes and individuals inhabiting the country. To understand how language provides the ideological and pragmatic basis for the different dimensions of these struggles at various levels is to understand at least some part of why such struggles take place.


ABBAS, Shemeem. 1993. The Power of English in Pakistan, World Englishes 12 ; 2 : 147-156.

AHMED, Feroz. 1998. Ethnicity and Politics in Pakistan. Karachi : Oxford University Press.

ALAM, Shamsul .1991. Language as Political Articulation : East Bengal in 1952, Journal of Contemporary Asia, 21 ; 4 : 469-487.

ALAVI, Hamza .1987. Politics of Ethnicity in India and Pakistan. Originally published in Pakistan Progressive, 9 : 1 (Summer). Quoted from ALAVI, H. and HARISS, John (Eds.) .1989. Sociology of Developing Countries. London : Macmillan Education Ltd. pp. 222-246.

AMIN, Tahir .1988. Ethno-National Movements of Pakistan : Domestic and International Factors. Islamabad : Institute of Policy Studies.

ANDERSON, Benedict .1983. Imagined Communities : Reflections on the Origin and Spread of Nationalism. London : Verso (reprinted edition), 1991.

AZIZ, K.K .1993. The Murder of History in Pakistan. Lahore : Vanguard Press.

BRASS, Paul R. 1974. Language, Religion and Politics in North India. Cambridge : Cambridge University Press.

CENSUS .1981. Handbook of Population Census Data. Islamabad : Population Census Organization, Statistics Division, Govt. of Pakistan, 1984.

_____ .1961. Census of Pakistan 1961 ; Pakistan : Populations, Tables and Reports. Karachi : Ministry of Home and Kashmir Affairs.

DIL, Anwar & DIL, Afia .2000. Bengali Language Movement to Bangladesh. Lahore : Ferozsons.

DIRECTORY.1995. Pakistan Deeni Madaris Ki Directory. [Urdu : The Directory of Religious Seminaries in Pakistan]. Islamabad : Islamic Education Research Cell, Ministry of Education.

EMIS .1998. Education Management Information System : Punjab. Lahore : Govt. of the Punjab.

FARUQI, Rahman .1998. Unprivileged Power : The Strange Case of Persian (and Urdu) in Nineteenth-Century India, The Annual of Urdu Studies, No. 13 : 3-30.

GELLNER, Ernest .1988. Plough, Sword and Book : The Structure of Human History. London : Collins Harvill.

GUPTA, Jyotirindra Das. 1970. Language Conflict and National Development : Group Politics and National Language. Berkeley & London : University of California Press.

HANAWAY, William & NASIR, Mumtaz .1996. 'Chapbook Publishing in Pakistan'. In HANAWAY, William & HESTON, Wilma (Eds.) .1996. Studies in Pakistani Popular Culture. Lahore : Sang-e-Meel. pp. 343-437.

HUTCHINSON, John & SMITH, Anthony D (Eds.) .1996. Ethnicity. Oxford & New York : Oxford University Press.

JAVED, Inamul Haq .1996. Baerooni Mamalik Mein Urdu [Urdu : Urdu in Foreign Countries]. Islamabad : Muqtadra Qaumi Zaban.

KERRIN, Dittmer. 1972. Die Indischen Muslims und die Hindi-Urdu Kontroverse in den United Provinces. Wiesbaden : Otto Harrassowitz.

KHAN, Seid Ghulam Hossein .1789. The Seir Mutagherin : or Review of Modern Times : Being an History of India as far Down As the Year 1783 the whole written in Persian. Trans. from Persian by Tota Manus. 4 volumes. Reprinted. Lahore : Sheikh Mubarak Ali, Oriental Publishers and Booksellers, 1975.

KING, Christopher R. 1994. One Language, Two Scripts : The Hindi Movement in Nineteenth Century North India. Bombay : Oxford University Press.

KOREJO, M.S .2000. G. M. Syed : An Analysis of his Political Perspectives. Karachi : Oxford University Press.

MALIK, Farida .1996. The Teaching of English in Pakistan. Lahore : Vanguard Press.

MALIK, Jamal .1996. The Colonialization of Islam : Dissolution of Traditional Institutions in Pakistan. Lahore : Vanguard.

MANSOOR, Sabiha .1993. Punjabi, Urdu, English in Pakistan : A Sociolinguistic Study. Lahore : Vanguard.

METACALF, Barbara .1982. Islamic Revival in British India : Deoband 1860-1900. Repr. Karachi : Royal Book Company, 1989.

MIRZA, Shafqat .1994. Pakistan Wich Man Boli Laher [Punjabi : The Mother Tongue Movement in Pakistan]. Interview by Maqsood Saqib in Maan Boli (January), 83-96.

NAYYAR, A.H .1998. Madrasah Education - Frozen in Time. In HOODBHOY, Pervez (Ed.) .1998. Education and the State : Fifty Years of Pakistan. Karachi : Oxford University Press pp. 215-250.

RAHMAN, Tariq .1996. Language and Politics in Pakistan. Karachi : Oxford University Press.

_____ .1996b. Language-Teaching in Pakistani Madrassas. In RAHMAN, 1999 : Chapter 5.

_____ 1999. Language, Education and Culture. Karachi : Oxford University Press.

_____ .1999a. The Politics of Urdu in India. Journal of South Asian and Middle Eastern Studies, Vol. XXII ; No 2 (Winter), 38-60.

_____ .1999b. Decline of Persian in British India, South Asia, Vol. XXII, No. 2 (June), 47-62.

RAI, Amrit .1984. A House Divided : The Origin and Development of Hindi-Urdu. New Delhi : Oxford University Press, 1991.

SAIGOL, Rubina .1995. Knowledge and Identity : Articulation of Gender in Educational Discourse in Pakistan. Lahore : ASR Publications.

SEMIS .1998. Sindh Education Management Information System. Karachi : Govt. of Sindh.

SHACKLE, Christopher .1977. Siraiki : A Language Movement in Pakistan, Modern Asian Studies. 11 ; 3 : 379-403.

SHEERANI, Hafiz Mahmood .1934. Punjab Mein Urdu Ki Sarguzasht Ka Aik Faramosh Shuda Varq, [Urdu : A forgotten chapter of the story of Urdu in the Punjab]. In MAZHAR MAHMOOD, Sheerani (Ed.), Muqalat-e-Hafiz Sheerani. Lahore : Majlis Taraqqi-e-Adab, 1966. Pp. 117-145.

UMAR, Badruddin .1970. Purba Banglar Bhasha Ondolan-o-Tatkalin Rajniti [Bengali : Language Movement of East Bengal and contemporary politics of the time]. Dhaka : Mowla Brothers (rev. ed), 1989. Vol 1.

WHORF, Benjamin Lee .1956. Language, Thought and Reality. CORROLL, John B (Ed.). Cambridge ; Massachusetts : The MIT Press.

ZAHEER, Hasan .1996. The Separation of East Pakistan. Karachi : Oxford University Press.

Annexe 1

Language Other Names Where Spoken Speakers Source
Balti Baltistani, Sbalti,
Purki (a variety of Balti)
Baltistan 27,000-300,000 (in 1992) SSNP-2 :8 & Grimes 1992 :739.
Badeshi Bishigram (Swat Kohistan) Not known Grimes, 737.
Bagri Bagria, Bagris, Baorias, Bahgri Sind (nomadic between India and Pakistan 200,000 (in 1993) Grimes, 737.
Bashgali Eastern Kativiri Gobar, Rumbur Valley (Chitral) 3,700 - 5,100 SSNP-5 : 134
Bateri Indus Kohistan 30,000 Breton 1997 : 200
Bhatneri North Eastern Traces Unconfirmed Grimes, 739.
Burushaski Billum, Kunjut, Khajuna Hunza, Nagar, Yasin Valleys (Northern Areas) 55,000 - 60,000 (in 1981) SSNP-2 : 37
Chilisso Chiliss, Galos Koli, Palas, Jalkot Indus Kohistan 1600-3000 Breton 1997 : 200 & Grimes, 739.
Dameli Gudoji, Damia, Dameoli Damel Damel Valley (Southern Chitral) 2000 - 5000 (in 1992) SSNP-5 : 11
Domaaki Domaski, Doma Mominabad (Hunza & Nagar) 500 plus (in 1989) SSNP-2 : 79
Dogri Punjabi, Pahari Azad Kashmir 1 million? Breton 1997 : 200
Dehwari Deghwari Kalat, Mastung (Central Baluchistan) 10,000 (in 1987) C-L. Breton 1997 :200 & Grimes, 739.
Dhatki Dhati Tharparkar, Sanghar (Sind) 100,000 plus (in 1987) Grimes, 740.
Gujari Gujari, Gojri, Gogri, Kashmir Gujuri, Gujuri Rajasthani Swat, Dir, Northern Areas, Azad Kashmir 200,000 - 300,000 (in 1992) SSNP-3 : 96 & Grimes, 740.
Gujrati Gujarati Karachi, other parts Sind Not known (44,000,000 In the world) SSNP-3 : 96 & Grimes, 740.
Gawar Bati Narsati, Narisati, Gowari, Arandui, Satre, Gowar-Bati Southern Chitral, Arandu Kunar river along Pakistan-Afghanistan Border 1300 - 2000 (in 1992) SSNP-5 : 156
Breton 1997 : 200 & Grimes, 740.
Gowro Gabaro, Gabar Khel (different from Gawri) Indus Kohistan (on the eastern bank, Mahrin village) 1,000 2,000 (in 1990) Breton 1997 : 200
& Grimes, 740.
Hazargi Hazara, Hezareh, Hezare'i Quetta 70,000 (in 1993) Grimes, 741.
(Breton 1997 : 200 calls it Marga)
Kalkoti None reported Dir Kohistan 4,000 (in 1990) Breton 1997 : 200
& Ethnologue
Kashmiri Keshur Kashmir & diaspora 105,000 (in 1993) Breton 1997 : 200 Grimes, 742.
Kati Bashgali, Kativiri, Nuristani (Chitral) Gobar Linkah Valleys 3,700-5,100 (in 1992) Grimes, 742.
Kamviri Shekhani, Kamoleshi.Lamertiviri, Kamik Chitral (southern end of Bashgal Valley) 1,500 - 2,000 (in 1992) SSNP-5 : 143 ;
Grimes, 742.
Khetrani None reported Northeast Balochistan Few Thousand (in 1987) Grimes, 742.
Kalasha Bashgali, Kalashwar, Urtsuniwar, Kalashamon, Kalash Kalash Valleys (Chitral) southern 2,900 - 5,700 (in 1992) SSNP-5 : 11
Kohistani Indus Kohistani ; Kalami ; Dir Kohistani, Kohiste, Khili, Maiyon ; Maiya, Shuthun, Mair. Indus Kohistan West bank of river, 22,000 (in 1993) Grimes, 742.
Koli Kachi, Kori, Kuchikoci (Lower Sind) around Towns of Tando Allahyar & Tando Jam 80,000 - 100,000 (in 1995) Grimes, 742.
Koli Tharadari Tharadari Lower Thar Desert 30,000 (in 1980) Grimes, 742.
Kalami Bashgharik, Dir Kohistani, Bashkarik, Diri, Kohistana, Dirwali, Kalami Kohistani] Gouri, Kohistani, Bashkari, Gawri, Garwi Kalam (Swat) Dir Kohistan 60,000 - 70,000 (in 1995) Baart 1997 : 4
Khowar Chitrali, qashqari, Arniya, Patu, Kohwar, Kashkara Chitral, Northern Areas, Ushu in Northern Swar 222,800 (in 1993) SSNP-5 : 11
Breton 1997 : 200
Grimes, 742.
Majhi Punjabi Lahore District Unknown. Refugees from India Punjab Grimes, 743.
Marwari (northern)

Marwari (southern)
Rajasthani, Meghwar, Jaiselmer, Marawar Rajasthani, Meghwar, Jaiselmer, Marawar South Punjab north of Dadu and Nawabshah Sindh and Southern Punjab 50,000 (in 1992)

50,000 (in 1992)
Grimes, 743.

Grimes, 743.
Lasi Lassi Las Bela District (South East Balochistan) few thousand Grimes, 743.
Ormuri Buraki, Bargista Kaniguram (South Waziristan) some in Afghanistan 10,000 - 40,000 SSNP-4 : 54
Od Odki Scattered in Sind & South Punjab 30,000 - 50,000 (in 1986) Grimes, 743.
Parkari Koli Negar Parkar town Tharparkar (southeastern Sind) 150,000 - 250,000 (in 1995) Grimes, 743.
Persian Farsi ; Madaglashti Persian in Chitral Dari, Tajik, Badakhshi Balochistan, Shishikoh Valley in Chitral Quetta, Peshawar etc 1,001,400 (in 1992) SSNP-5 : 11
Grimes, 740.
Pashai 5,000 Breton 1997 : 200
Phalura Dangarik, Ashreti, Tangiri, Palula, Biyori, Phalulo 7 villages near Drosh, Chitral possibly 1 village in Dir, Kohistan 8,600 (in 1990) SSNP-5 : 11
Sansi None reported North-western Sind Unknown Grimes, 744.
Shina Sina, Shinaki Gilgit, Kohistan, Baltistan 500,000 SSNP-2 : 93
Torwali Kohistani, Bahrain Kohistani Bahran (Swat) 60,000 Breton 1997 : 200
Turkmen Turkic Refugees in Pakistan scattered Grimes, 745.
Ushojo (Ushuji) Upper part of Bishigram
Valley in Swat
Kohistan (12 villages) 2000 (in 1992) Grimes, 746.
Uyghur Ugur, Ughir Near the Chinese Border, Scattered few hundred Grimes, 746.
Uzbek Turkic Refugees in Pakistan 50,000 (in 1993) Grimes, 746.
Vaghri Vaghri Koli Sind 2,000 plus Grimes, 746.
Wadiyara Koli, Wadaria North of Matli & Jamesabad 75,000 (in 1980) Grimes, 743.
Wanetsi Tarino, Chalgari Harnai (East of Quetta) 90,000 (in 1992) SSNP-4 : 51
Breton 1997 : 200
Grimes, 746.
Wakhi Kheek ; Kheekwar Wakhani, Wakhigi, Wakhan Northern ends of Hunza & Chitral 7,500 - 10,000 (in 1981) SSNP-2 : 61
Yidgha Yidghah, Lutkuhwar Upper Lutkoh Valley (Western Chitral) 5,000 - 6,000 (in 1991) SSNP-5 : 11



Urdu can be understood by those who know Hindi because both are really two variants of the same language. The script of Urdu is Perso-Arabic (nastaliq) while that of Hindi is devanagari. The Roman script has the advantage of being understood both by Urdu and Hindi users.