The Fall of Urdu and the Triumph of English in Pakistan: A Political Economic Analysis

John Willoughby^
Zehra Aftab*

November, 2020

Abstract

In this paper, we investigate both how the use of language in higher education in Pakistan has
evolved and why the medium of instruction remains a contested terrain. We focus on the struggle
between advocates for the use of Urdu and the use of English. By examining the repeated failed
attempts by high political authorities to replace English with Urdu, we demonstrate the
usefulness of Avner Greif’s evolutionary, path-dependent theory of institutional change. We also
argue, however, that Jack Knight’s focus on the struggle over resources is necessary if we are to
understand why the futile attempts to make Urdu the dominant language of education persist.

Keywords: Higher education, Education and Inequality, Language, Pakistan.
JEL codes: I23, I24, P16, Z13

^ Professor, American University; jwillou@american.edu
* Faculty Fellow, American University; corresponding author: aftab.zehra@gmail.com

word count: 7,400




The Fall of Urdu and the Triumph of English in Pakistan: A Political Economic Analysis

“Whoever controls the language, controls the culture.”

Dennis Peacocke

There are sharp differences within Pakistan's educational system: It is divided by medium of
instruction and social status. At the top end exists the traditional elite English-medium schools.
More recent entries include non-elite private English-medium schools aimed at the middle class
as well as a large number of Urdu-medium madrassas which largely cater to low-income students.
In between are public sector Urdu-medium primary and secondary institutions as well as public
sector schools which offer instruction in other vernacular languages such as Sindhi and Balochi.

This paper begins with setting the context in terms of providing an overview of the state of
language inequality in today’s Pakistan. We then provide an overview of Pakistan’s colonial past
and a discussion of how Urdu became associated with the Muslim struggle for independence.
This is followed by a brief outline of language policy shifts in Pakistan since the founding of the
nation in 1947. We review the historical record in order to explain both why the Pakistani
authorities have largely failed in their quest to establish Urdu as the universal language of
education and why this failure has led to several attempts to reassert Urdu’s predominance. In the
third section, we utilize the institutional analytical frameworks of Avner Greif and Jack Knight to
better understand the historical narrative of section II. We consider why clear policy
pronouncements in support of Urdu were never implemented, despite the ostensible support for
Urdu from the Pakistani religious establishment. Our final section closes by presenting a research
program which our analysis suggests, as well as considering some interesting theoretical issues
raised by this analysis.

I. Contemporary Language Use in Pakistan and its Historical Roots

Pakistan is a multilingual, multicultural society with more than 66 spoken languages
(Lewis et al, 2016). At the time of independence 56 percent of the population spoke Bengali, i.e.
the population of then East Pakistan; while the majority language of West Pakistan was Punjabi
(67 percent of West Pakistan) followed by Sindhi and Pashto; only a minority spoke Urdu.
However, Urdu, despite being a minority language, has emerged as the proclaimed national
language and English the language of official business.1
With reference to schooling, Pakistan is characterized by five broad (pre-university)
schooling streams using three different language categories: elite English-medium schools
(including military cadet schools), non-elite private English-medium schools catering to the
lower-middle and middle-income classes, government-run Urdu-medium schools, public
vernacular (mostly Sindhi or Pashto) medium schools, and madrassas (Islamic seminaries) which
mainly use Urdu. However, university education is mostly imparted in English. For many
university students who have received earlier education in another language and are not very
proficient in English, this is a particularly challenging situation. Moreover, English opens doors
to much coveted jobs in the military and civil service and gives not only a social, but also a
psychological advantage to those fluent in the language (Phillipson, 1992). In contrast, Urdu
comes with its own cultural history and ideology as it was Urdu that became associated with
Muslim identity during the struggle for independence. Hence, as these groups struggle for their
respective language, they are also struggling for their cultural heritage and access to economic
power. The adoption of language as the medium of communication for state business or private
business is both a product of educational policy and also a process that affects educational policy.

1. Pre-Partition Urdu becomes Associated with Muslim Identity

Before partition, the main political actors in the area that became West Pakistan
constituted the British colonizers, both Muslim and Hindu leaders (the former represented
Muslim League and the latter the Congress party), the Muslim Urdu-educated intelligentsia of
Delhi and Lucknow, the Muslim poets and ulema, and the local vernacular-speaking feudal
landholders and peasants (Rahman, 2011).
The British colonizers came to the Indian Subcontinent with the objective of resource
extraction, which required the imposition of Anglo colonial rule. When British imperial control
was established, British rulers replaced Persian with English as the official language in British
India during the 17th century (Powell, 2002). They also introduced English as a medium of
instruction as they wanted to create a local gentry that would help them administer colonial India.
(Rassool, 2007). However, English language education was only made available in major urban

1 Present-day linguistic make-up of Pakistan: Punjabi, 44.15%; Pashto, 15.42%; Sindhi, 14.10%; Siraiki, 10.53%;

Urdu, 7.57%; Balochi, 3.57%; Others, 4.66%. Census Report of Pakistan. Population Census Organization, Statistics
Division, Government of Pakistan. 2001. Table 2-7, p.107.

schools and higher education institutions while the education of the rest of the locals was
imparted either in Urdu or vernacular. An incentive to join the English-medium schools was the
opening of civil service positions for the local population in 1832, 41 years after the 1791 Act of
Native Exclusion (Spear 1958, in Rahman, 2006, p.30), for which the main selection criteria was
competence in English. The Congress and Muslim league leadership emerged from these English-
medium schools. Jinnah, in his freedom movement used the English-educated bureaucracy,
military, and judiciary, which had originally been in the service of the British Raj. This led
Hamza Alavi to dub the newly formed state of Pakistan “a vice-regal” state – a state that
continued to be ruled by the “salariat” in power: the military, bureaucratic and also landed elite
that continued its pre-colonial administrative practices (Alavi, 1972). Jinnah in using these very
intermediaries in his struggle for the Muslim national movement had made these social structures
even more strongly embedded in what emerged as the state of Pakistan (Nasr, 2001). It is not
surprising that these English-educated agents/actors opted for English as the language of official
state business, as this had been the language of business in colonial India.
Language has often been associated with national and regional/ethnic identity formation,
but in the case of India, Urdu became associated with religious identity -- the Indian-Muslim
identity -- between the late 18th and the early 20th centuries (see Rahman, 2011, for a detailed
discussion). This occurred despite the fact that Indian Muslims spoke a variety of languages
including Bengali, Punjabi, Pashto, Sindhi and Gujrati. Yet, it is modern Urdu that is associated
with Islam in the Indian subcontinent.
Linguistic historians trace Urdu’s ancestor as an Indic language which incorporated
words from local languages (bhasha) and Sanskrit, and its allusions were to India and the local
culture, though the script was Perso-Arabic (as opposed to Devanagari). The Islamisation of the
language began in mid eighteenth century as Muslim poets purged the language of its Indic
element: local (bhasha) and Sanskritic words were removed on the excuse that these terms were
“obsolete, unfashionable and rough”; moreover, Indian and Hindu cultural allusions and
metaphors were substituted with Iranian and Islamic references. As more than 4,000 local words
were banished, what emerged finally was highly Persianized Urdu full of Islamic cultural
references _ an identity marker for the educated Muslim elite of Delhi and Lucknow (Rahman
2011). This, in turn, alienated the Hindus, and led to the Sankritisation of Hindi, creating the
Urdu-Hindi divide (Brass 1974).
The association between Urdu and Islam, further developed during the British era: As
Muslim political power shrank, Muslim Ulema, along with poets and political activists, started
writing and publishing pamphlets in Urdu: it became the favored language for religious debate
between Muslim scholars, and, in time, emerged as a repository for Islamic literature. Thus while
the Urdu poets and Ulema together nurtured Urdu, and while it is true that Urdu has also been
associated with modernity and enlightenment (the Delhi Renaissance), it is the association with
Islam that predominates (Rahman, 2011).


2. Post-Independence: Colonial Legacy and the nation-building project

At the time of independence, Pakistan, like most other ex-colonial countries, was faced
with the problem of developing a language policy in a multilingual society. As in many new
countries, formulating an appropriate policy was complicated by different language groups
competing for recognition and status. The nation’s founders, as mentioned above, were
themselves trained in English. The military, judiciary, and the civil service were Anglicized
institutions who wanted to continue state business in English, but they also wanted to unite an
ethnically diverse population under the umbrella of a national language. With reference to the
national language, the struggle was initially between Urdu and Bengali: While the former had
become a symbol of Muslim unity, the latter was the majority language of East Pakistan (present-
day Bangladesh) which was the largest and most populated province of Pakistan. It would appear
incongruous that Punjabi, despite being a widely spoken regional language was not even in the
running. Punjabi was never used in the domains of power, not even at the time of Ranjit Singh,2
and Punjabis also supported the choice of Urdu as a unifying symbol of the federation. The Urdu
lobby thus comprised both the Urdu-speaking mohajirs, the middle-class Punjabi speaking
intelligentsia of Punjab, which claimed to give up their local vernacular in the interest of the
federation, and to a lesser extent, the similarly Urdu-educated intelligentsia of urban Khyber
Pakhtunkawa (KP) and Baluchistan. The Bangla leaders and Bengali majority of East Pakistan in
contrast advocated for Bengali as the national language.
Finally, the following language hierarchy emerged: English was the language of choice
for state business, Urdu was deemed the national language, and Bangla (with Pashto and Sindhi)
was relegated to provincial status. However, despite the assumed integrative appeal associated
with Urdu, the decision was opposed by the Bengalis of Eastern Pakistan who, being the majority,
were unhappy that their language, Bangla, had been relegated to the status of provincial language.
Murshid (1985) provides a detailed account of the Bengali movement in the early 1950s which
finally led to Bangla also being given the status of national language. However, the seeds of

2 Ranjit Singh was the most powerful indigenous Punjabi-speaking ruler of pre-colonial Punjab; like most other rulers of India, he
used Persian for official documentation; although, the informal conversation in the court was in Punjabi (Rahman, 2019).

resentment had already been sown, and in 1971 culminated in the majority of Pakistan (East
Pakistan) seceding from their once cherished homeland.

II. Medium Of Instruction _ Stated Policy Vs Practice: A Historical Overview

With respect to the medium of instruction, the Advisory Board of Education held its first
meeting in 1948 and resolved that the mother tongue should be the medium of instruction at the
primary stage. Moreover, it was also recommended that English be replaced by Urdu in the
universities (ABE, 1949), while secondary education should be in Urdu (ABE, 1955). As a
result, a number of institutions were established to do basic work in Urdu, from coining new
terms, to translations, to developing new tools and techniques to expedite its adoption as an
official language (Rahman 1996, pg. 233).
Since this first conference on education in 1948, the basic aspects of language policy
have remained constant. English, especially for higher education, is justified by the state, in the
interest of modernization because it is the language of science and technology. On the other
hand, Urdu is justified in the interest of Pakistani national unification. The documents kept
insisting that mother tongue is the best medium of instruction for a child but, except in East
Pakistan (until 1971, when it became Bangladesh), Sindh and some parts of Khyber Paktunkhwa
(KP), no mother tongues were used as medium of instruction.

1. Urduization vs English Bias: Ayub Khan’s 1960s

By the end of the 1950s, despite the efforts of the Urdu lobby to promote the use of Urdu
and the ruling elites apparent support of these efforts, it was English which emerged as the
dominant language in government and higher education. In fact, by this time the expanding
middle class (both Urdu-speaking mohajirs and Punjabis) -- initial Urdu advocates -- had also
realized that the best chance of acquiring power, social prestige and affluence was by joining the
superior civil services, the officer cadre of the armed forces or to get a professional degree: The
Central Superior Services (CSS) exams were held in English, and higher education was also in
English. Hence, the urban Urdu middle-class also had a strong incentive to be educated in
English. Not only the urban upper middle class but even feudal/tribal elites, though not literate
themselves, sought to educate their children in elite English-medium schools (Rahman 1997).3
Moreover, under the patronage of General Ayub Khan, who himself was an Anglicized military

3 Examples include Aitchison in Lahore and Burn Hall in Abbottabad.
dictator, the armed forces started developing their own schools—cadet colleges and PAF Model
Schools—to provide subsidized, English-medium schooling, to prepare students for careers in the
defense forces.4
In 1959 the Sharif Commission on Education defended the above-mentioned
government-subsidized English-language educational institutions in the name of efficiency and
modernization (CNE, 1959), However, the commission also recommended that both Urdu and
Bangla be used as mediums of instruction from Class VI onward and in this way, in about fifteen
years, Urdu would reach a point of development where it would become the medium of
instruction at the university level. The Commission had also stated that until Urdu was ready to
replace English, English should continue to be used for advanced study and research. This
statement allowed confusion to take root in terms of how and when and by whom it would be
determined that Urdu was ready to replace English. This was a convenient method of maintaining
the status quo, and English was given a fifteen-year extension (Khalique, 2006).
In 1966 students from less privileged Urdu-medium institutions protested against
government-subsidized cadet schools, and a new commission under Justice Hamoodur Rahman
was set up to examine student unrest and students’ welfare problems. The commission agreed that
cadet colleges and PAF schools violated the constitutional assurance that ‘all citizens are equal
before law (Paragraph 15 under Right No. VI)’ (GOP 1966, pg. 18) because teaching in English
excluded some students. Nevertheless, Judge Rahman’s commission defended these schools by
concluding that such schools “are intended to produce some better type of students who would be
more suitably disciplined and equipped for eventually entering the defense service of the country
or filling higher administrative posts” (GOP 1966, pg. 18). As a result these cadet colleges
actually multiplied from 1970 onwards. The Hamoodur Rahman Commission also criticized those
universities which had adopted Urdu as a medium of examination in BA for being over-zealous:
Karachi, Punjab and Sindh universities were criticized for allowing Urdu and Sindhi as languages
for instruction and sitting exams.
Despite these setbacks, the Urdu political advocates kept up their pro-Urdu movement,
demanding that signboards should be in Urdu (Pakistan Times, 21 February 1961), that
proceedings of meetings be in Urdu (Abdullah, 1976). However, despite all the efforts of the
Urdu lobby, the elitist officer corps of the higher administration, judiciary and the military kept
using English. Higher education, especially in scientific and technological subjects, also

4 In the words of Rahman, “the elite of wealth (feudal and tribal lords; business magnates, etc.) and the elite of power

(the military and bureaucratic elites) made arrangements to facilitate the entry of their children into the elite, thus
narrowing its base of selection, through promoting elitist schooling while professing to create equal opportunities for
all through vernacularization “ (Rahman 1997).
continued to be given in English. Urdu was allowed eventually for Arts (i.e. Social Sciences and
Humanities) and although the policy was couched in the language of popular demand and
facilitated access to higher education, it actually ghettoized the non-Science students and
disciplines since they bore the stigma of being culturally and intellectually inferior (Rahman
2019).
In 1969, there was a new government and a new committee (headed by Air Marshal Nur
Khan) was constituted to overhaul the educational system. This committee recommended that
Urdu and Bengali should be used as the medium of instruction by1975 (PNEP, 1969). This was
the first time that an official document acknowledged the fact that there is 'almost a caste-like
distinction between those who feel at ease in expressing themselves in English and those who do
not' (PNEP, 1969, pg. 14). It was observed that: ‘Not only does the use of English as the medium
of instruction at higher levels perpetuate the gulf between the rulers and the ruled, it also
perpetuates the advantages of those children who come from well-to-do families, . . . ‘(PNEP,
1969, pg. 3). However, the elite English medium schools (including the cadet colleges) remained,
and the New Education Policy (1970) left the task of examining 'the question of the change over
from English to the national languages' to a commission which would be established in 1972
(NEP, 1970, pg. 19). Thus, the incipient radicalism of Nur Khan was reversed as the status quo
asserted itself.

2. 1970s: The Democratic Bhutto Years

1971 marked the partition of Pakistan. In West Pakistan the democratically elected
Peoples’ Party formed the government with ZA Bhutto as Prime Minister. The 1973 Constitution
of the Republic was promulgated under Bhutto with Article 251 pertaining to language in
education. The article declared Urdu as the national language and pledged to further its
development; moreover, a time frame of 15 years was set for the replacement of English with
Urdu. Interestingly, the timing of the Constitution coincided with the lapse of the fifteen year
extension given to English by the Sharif Commission and hence refreshed that extension for
another fifteen years.5
Given Bhutto’s left-leaning socialist agenda and secular views, and his Sindhi roots, he
was looked upon with suspicion by the Urdu lobby which by now comprised a large religious
element in the Jamiat-e- Ulema-i-Islam (a religiously motivated party). Afraid of being

5 The constitution also recognized the linguistic rights of speakers of regional and minority languages by allowing the provincial
governments freedom to develop their languages.
categorized as the “Other”, and in an attempt to placate his opponents Bhutto succumbed to the
integrative appeal of both Islam and Urdu, while Sindhis demanded to promote and encourage
Sindhi as an official language in the province, in congruence with Article 251(3) of the
Constitution (see above), This culminated in the Urdu-Sindhi language riots of January 1970 and
July 1972. These riots were the response of the supporters of Urdu to what they thought was an
effort to dislodge them from their position and make Sindhi the dominant language for education
and administration (Amin, 1988). Finally, Sindhi was adopted as the official language of Sindh,
but little effort was made in real terms to give the language its due official status. In KPK and
Balochistan, similar efforts were made by the provincial governments. But, at the federal level,
and in elite schools, English reigned supreme (Rahman, 1997).
Bhutto chose to placate the Urdu lobby (who by now was mostly aligned with religious
parties) by announcing cosmetic Islamic measures rather than the less emotive, and more
controversial, strategy of giving Urdu the place of English in educational institutions (Rahman
1997, pg 18).6Thus, it was in Bhutto’s regime that religion came to occupy the public sphere and
different language lobbies were solidified into different groups: the Urdu lobby was with the
religious parties while the ethno-nationalists and the Anglicized elite were left of center, being
either inclined to socialism or liberalism, respectively (Amin, 1988: pg 244). The fortunes of
Urdu would now be connected more closely than ever before with the struggle between the
religious and the secular in Pakistani politics (Rahman, 1997, pg 18).

3 1977- 1988: General Zia ul Haq’s Pro-Urdu Stand

It was during General Zia ul Haq’s martial-law that both Urdu and Islam came into their own.
Zia himself hailed from middle-class Urdu-speaking background and therefore had the support of
Urdu mohajirs and other Urdu advocates who appreciated his policies of Islamisation/Urduisation
as part of his ‘centralizing ideology’. Now, Urdu was not only associated with Islam, but also
with authoritarianism. In 1979 Zia ordered that all speeches should be in Urdu and also set up the
Muqtadra Qaumi Zaban (National Language Authority) to consider ways and means for the
promotion of Urdu as the national language of Pakistan and to make all necessary arrangements
in this regard. By the end of 1979 many offices in Punjab began to use Urdu rather than English.
He also ordered that Urdu be the medium of instruction in all schools from grade I, such that by

6 By now, it was clearer than ever before: the Urdu proto-elite was with the religious right wing while the ethno-nationalistic proto-
elites and the anglicized elite were left of center, being either inclined to socialism (Amin, 1988: 244) or liberalism, respectively. The
fortunes of Urdu would now be connected more closely than ever before with the struggle between the religious and the secular in
Pakistani politics (Rahman, 1997, pg 18).
1989 the matriculation (10th grade) examination could be conducted in Urdu. Moreover, the
Ministry of Education instructed schools not to use the English-medium nomenclature, and
Islamic education was decided to be a compulsory subject until graduation. In light of the above
initiatives Zia ul Haq was declared the 'Patron of Urdu’, and such was the confidence of the Urdu
lobby in him that in 1981 at the Annual Urdu Conference at Lahore (27-28 Nov. 1981) the Urdu
lobby demanded that Urdu should be imposed through a presidential ordinance (Pakistan Times,
28 Nov. 1981). But, in the end, despite all the fervor and enthusiasm even the ‘Patron of Urdu’
and martial-law administrator, Zia ul Haq, could not purge Pakistan of the English language
either in the official domain or as a medium of instruction.7
Although many of the government and federal model schools did adopt Urdu from grade
1, the cadet schools and elite private schools remained English medium. The major argument of
the English lobby was that Pakistan would fall behind other countries if English was abandoned,
while the Urdu lobby insisted that sufficient books did exist in Urdu and more could be
translated. However, no practical steps materialized, and in 1983 the elite schools were given
legal protection to prepare their students for senior and higher senior Cambridge examinations
thus making the two parallel streams of education even more distinct. Instead, Urdu became a
compulsory subject in these schools until class 12. Moreover, on 11 October 1987 General Zia ul
Haq himself allayed any residual fears of the English lobby by declaring that English could not be
abandoned altogether (Rahman, 1997). According to Rahman, apart from a few editorials against
the continuation of English language schooling, the reversal of the 1979 education policy, the
biggest concrete step taken in favor of Urdu, was allowed to take place almost silently (Rahman,
1997, pg 198). Moreover, in 1987, despite the initial fervor, ministries were also instructed to
continue their proceedings in English.
Zia knew that the Urdu lobby would keep favoring him despite their disappointment in
his pro-Urdu stance; and Zia in the end realized that he could not alienate the Anglicized elite in
the long run. Hence, English reigned supreme and the net result were two parallel streams of
education: Urdu-medium and English-medium.8

4. Failed Attempts to Democratize English (Post 1989)

7
For newspaper articles related to the Urdu-English debate refer to Akhtar, B. M., Rahman, K. and Syed, M. 1986.
Qaomi Zaban Akhbarat Ke Aine Men [Urdu: The National Language as Reflected in the Press]. Islamabad: Muqtadra
Qaomi Zab.
8 Under different circumstances, perhaps the demand for Urdu emerging as the national language might have been

viewed as compatible with social justice and may have appealed to Pakistani leftists and liberals, but Urdu had now not
only become associated with rightist Islamic forces, but also with authoritarian rule which had even labelled ethno-
nationalists (vernacular supporters) as anti-state actors, hence, forever alienating the liberal, socialist forces.
 As mentioned above, at the time of Zia ul Haq's assassination the position of Urdu was
not much better than it was when he first took power reflecting the dominant classes monopoly
over English. By now the religious Urdu-lobby had completely alienated the leftist secular forces
and in 1989 Benazir Bhutto attempted to introduce English in all schools from class I as an
attempt at modernization, despite the fact that it did conflict with her party’s socialist agenda.9
This policy was hurriedly launched through a government notification, with no well-defined
implementation strategy. Not surprisingly, little effort was made by the educational planners and
school leaders in public sector schools to go beyond introducing English as a formality, mainly to
show compliance to orders.
General Musharraf assumed power in October 1999 through a military coup. Musharraf’s
modernization and “enlightened moderation” in religion replaced the more fundamentalist
policies of Zia ul Haq. His government reiterated Benazir’s pro-English stance supporting
English as the language of and for development (Shamim, 2007). However, again no proper
implementation strategies were adopted to translate these policy statements into practice in
schools in Pakistan.
In 2010, under the 18th amendment to the Constitution, education became a provincial
issue. This made the provinces more autonomous than before with the result that the Punjab
government under chief minister Shahbaz Sharif decided to support the local demand for English
in the public. The Punjab government passed an executive order converting a number of
government Urdu-medium schools to English-medium. The schools did not have sufficient
numbers of teachers to implement this policy in any meaningful way nor were the students
exposed to English outside school so the policy failed. In 2011 the British Council concluded that
teachers still taught in Urdu and Punjabi just as they did before this policy was declared (PEELI
2013, pg 22–23). In 2014 the decision was reversed, and today the confusion regarding medium
of instruction continues, and in practice, public schools can be Urdu, English, Sindhi, or Pashto
medium, depending on their location, with English being introduced in some public schools from
grade III and in some from grade V. Matriculation (i.e. 10th grade) examination may be taken in
Urdu or English, but Intermediate (i.e. 12th grade) examinations take place in English (Abbas
1993). Moreover, some English medium schools are now offering British O-level and A-levels to
their students, thus increasing the gulf between the two groups. Higher education continues to be
in English, and examinations for access to key government positions remain in English as well.

9 Benazir Bhutto’s civilian government announced that English would be taught in all schools from Grade 1 (it was

previously taught from Grade 4 in public sector schools).

 The above narrative illustrates that while the state/ruling elite apparently supported Urdu
because of its supposed integrative value, in the formal official domains it continued to support
English. This attachment to English maintained the ruling elites’ social distinction from the non-
elite,. In time, there seems to be a gradual evolution of the middle-class towards English as well
and the strength of Urdu advocates has weakened because of growing demands for English
instruction.. In fact, with time, most of the Urdu-speaking middle-class has in fact been coopted:
this is reflected in the growing demand and, in turn, mushrooming growth of low-income private
English medium schools of varying quality (ASER, 2006). But, this emphasis on English-medium
instruction has not weakened the distinction between the elite and non-elite as the quality of
teaching in these schools is limited. Therefore, the vicious cycle continues, and this unequal
education system continues to reproduce inequality.

III: Institutional Analyses of Historical Change: Using the Frameworks of Greif and
Knight to Understand Pakistan’s Language Conflict

Greif defines an institution as a system of norms, beliefs, organizations, and rules

“exogenous to each individual” but which “generalizes behavioral regularities”.10 In one sense,
language does seem to fit into Greif’s definition of an institution. Language can certainly be
viewed as a set of distinct rules of communication which is human-made but not chosen by the
individual. Moreover, to anticipate Greif’s theory of institutional change, the continual use of
language depends on the strength of payoffs in terms of facilitated communication with others.
On the other hand, there is an aspect of language which escapes the definition of an institution.
Language is not generated through a belief system or a set of norms in the same way that a pre-
pandemic handshake or procedures governing marriage might be. While the particular language
we use is learned, our ability to communicate in the enhanced way which humans can is a
genetically programmed capacity more similar to the capability of walking on two legs. The
issue of language as an institution becomes, relevant, however, when studying language choice or
the way in which different forms of linguistic expression are used in different social contexts.
Language becomes more like an institution when its use changes. In this sense, norms might
govern the words and grammar appropriate for communication within a social group, beliefs
might make one more likely to elevate one language over another and rules might regulate which
language is permitted to be used in particular government sanctioned contexts. Greif, by
distinguishing between rules and beliefs, places motivation at the center of the analysis and

10 The term system here is used to highlight that an institution is not a monolithic entity, but comprises of inter-related

elements that conjointly guide behavior.

argues that if prescriptive rules are to have an impact, individuals must be motivated to follow
them.
Our analysis begins with a consideration of how aspects of any institution – norms,
beliefs, rules and the distribution of advantages – are relevant to the Urdu-English choice in
Pakistan. Social norms are defined as informal rules of behavior which are not codified but are
reflected in the spontaneous behavior in the population. Beliefs attach judgments (either positive
or negative) to a particular form of behavior, and rules create government-sanctioned hierarchies
of behavior.
Modern institutionalist theories of change deploy a metaphorical game theoretic
framework which focuses on the payoffs from a particular transaction or series of actions. Greif
develops a theory of change by labeling structural features of a society as ‘quasi-parameters’ if
these features evolve over time and either reinforce or undermine the payoffs from a particular
transaction or activity. With respect to language choice in higher education, the activity or
transaction is the act of being educated in a particular language and the payoffs are the differential
benefits one gets from being educated in that language. The quasi-parameters that might disrupt
the structure of payoffs could be changing shares of population from different linguistic groups,
the rising educational attainment of the general population, and changes in the domestic polity
and economy with the global political economic environment. If, for example, demographic shifts
raise the percentage of the population which is Urdu-speaking or attend Urdu-medium schools
then one could argue that payoffs from using Urdu in higher education would increase. On the
other hand, the rising educational attainment of a population combined with the predominant use
of English in higher education could signal to non-elite families of those children attending
secondary school that these children too should have access to English instruction. Such changes
in the shares of population with a specific characteristic (Urdu competence or general educational
attainment) could signal a change in beliefs about the appropriate use of English, which in turn
might culminate in shifts in behavior with respect to language use in the context of higher
education. We propose the following four hypotheses that can both account for the continued
persistence of English as the language of choice in higher education as well as the continued
persistence of Urdu as a possible alternative. These conflicting long-term tendencies allow us to
deploy an additional institutional change argument by Jack Knight who, in contrast to Greif,
focuses on distributional struggles over rules, rather than the evolution of norms.
The first hypothesis addresses the quasi-parameter of rising educational attainment. We
posit that increased schooling strengthens the parents’ demand for English instruction for their
children, since this provides a potential pathway to eliter occupations.
 The second two hypotheses focus on the quasi-parameter differential birth rates for
distinct sectors of the population. In the first place, we argue that a higher birth rate in the
population more closely attached to political Islam will demand increased education in Urdu-
medium madrassas. In this case, the priority of parents lies less in the social advancement of their
children, but in ensuring that their children are embedded in a moral order which offers salvation
to all within it. History and contemporary events tell us that this motivation can be as powerful as
any direct material interest in social advancement.
Our third hypothesis is that an increase or decrease in the share of the Urdu-speaking
middle class will either increase or decrease the demand for Urdu as the medium of instruction in
higher education. Note that if this share of the population is growing, the second and third
hypothesis do not necessarily reinforce each other if the less elite Islamist population sees their
children as becoming educated in madrassas rather than morally threatening secular universities.
Our fourth hypothesis examines the quasi-parameter of global political economic
evolution. We hypothesize that Pakistan’s increasing openness and dependence on the world
economy and political economic culture increases the perceived payoffs from the use of English,
which in turn strengthens the demand for higher education English.
These propositions, if correct, allow us to construct an evolutionary dynamic which
anticipates the rise of two segregated educational systems – one culminating in the strengthening
of English-medium university institutions and supporting elite and non-elite secondary schools
and the other culminating in a network of madrassas taught in Urdu as well as intensive
instruction in Arabic.
This Greifian framework allows us to better deploy Jack Knight’s analysis which stresses
conflict over distributional shares as the major reason for institutional change. In his book,
Institutions and Social Conflict, Knight presents a study of the “ … rationality of social
institutions that place greater emphasis on the role of strategic actions (pg 19). Thus, rather than
conceiving social institutions as the product of efforts to constrain social actors within a
collectivity, Knight argues that “social institutions are conceived as a product of some to
constrain the actions of others with whom they interact.” This analysis puts much more emphasis
on power relations than Greif’s approach. As Knight states, “Those “actors with a relative
bargaining advantage can force others to comply with the equilibrium strategy of the strong actors
whether or not they want to do it and from this it follows that social actors follow the institutional
rule not because they are pareto improving but simply because they cannot do better than to
follow them.” (pg 127)
 The distributional issues associated with the struggle between Urdu and English are clear.
If a change in rules privileges Urdu over English as the language of government and business,
then we would expect to see a flow of competent Urdu speakers into important professional
positions at all levels of society. This in turn could push English speakers to more specialized,
niche jobs. Our hypotheses, suggest, that those advocating English would include not only the
elite Anglicized officials already ensconced in privileged positions, but also those newly educated
families of the middle class and lower middle class eager to have their children obtain English
competence. This might, (as we noted in the historical narrative) even include Urdu speaking
families. Faced with this opposition, advocates for Urdu could attempt to mobilize support from
those families and children attending Islamic madrassas. This linkage between Islamism and
demands for Urdu-medium higher education instruction reached its apogee during the rule of Zia
ul Haq, but since then, the link between these sectors of Urdu-based education has weakened.
The results of any distributional struggle depend on the ability to mobilize populations
which believe strongly that their future would be improved or threatened by any change in
government policy. One major source of power for advocates for the maintenance of English is
the simple fact that the medium of instruction in higher education is already English. This reality
confirms Knight’s argument that already existing institutional relations might force subordinates
to adapt to, rather than resist institutions. In addition, the existence of English instruction requires
parents to invest in English-language training for their children. The fact that commission’s
studying the question of language in universities always picked fifteen years as the point when
Urdu utopia would be reached suggests a great reluctance of policy makers to disrupt these plans.
While the term ‘crypto-morality’ i.e. “the secret adherence to one morality while practicing
another in public” (Greif, 2010, pg 229) was developed by Greif to describe how norms and
beliefs prescribed by the state can persist within an oppressed community, we argue that this
concept can be modified to describe the contradictory behavior of political elite who, on the one
hand, advocate Urduization but, on the other hand, take measures which permit their cohort’s
children to attain high quality education in English.11

11 Greif (2010) uses the concept of crypto-morality to demonstrate how moral beliefs persist in societies despite

dictatorial regimes and pressure for change: In hostile institutional environments minorities “pretend to hold the
institutionally sanctioned moral belief, while secretly holding and transmitting another morality to their children”
(Greif, 2010, pg 229). Hence, crypto-morality (“Crypto” from “kryptein”, Greek for “to hide”) is practiced by the
common citizens in the face of a hostile state. Greif (2010) refers to various historical episodes including the former
communist states to provide support for his narrative.

CONCLUSION

The institutional interpretive frameworks we have deployed have the advantage of

forcing the analyst to consider changes in long-term background factors which can restructure the
payoffs from a particular action and unleash distributional struggles for changes in rules which
can further cause net benefits to evolve. While Greif’s approach focuses on shifts in norms and
Knight considers struggles over rules, these approaches can be used together when certain
movements in quasi-parameters create contradictory tendencies. This analysis provides a way to
synthesize this modern, more general approach to social change to a more ‘old fashioned’,
Marxist emphasis on how the reproduction and intensification of social contradictions can lead to
distributional struggles. The advantage of this approach is that the analyst need not limit him or
herself to the issue of class division. Moreover, the lack of a teleological conviction on how
history should move allows the analyst to develop a more open-ended set of hypotheses about
institutional change.
In the case of Pakistan, we have argued that increased educational attainment in the
general population in combination with the continued ability of the Anglicized elite to maintain
English as the primary medium of instruction in higher education has reinforced the use of
English. Potential countervailing movements in quasi-parameters, such as the rise of Urdu
medium madrassas, have not undermined the elite use of English because Islamic education has
given rise to self-segregated educational systems in which the two student populations do not
interact. Moreover, the relatively small number of native Urdu speakers within the population of
Pakistan has also limited the ability of Urdu advocates to push forcefully for dramatic linguistic
changes within universities and colleges. Finally, we also suspect that political economic
globalization has also reinforced the perceived payoffs from the use of English.
We have also noted the disconnection between the struggle of shifts in the medium of
instruction at all levels of Pakistan’s educational system and the meager budgetary support for
such changes. In this sense, Hamza Alavi’s original argument that the state of Pakistan remains
an instrument of an elite remains salient, even if his identification of English with neo-colonial
rule is undercut by the growing non-elite popular support for the use of English.
These conclusions are tentative, but a strength of our use of these institutional theories of
change is that our claims can be empirically investigated both by surveying different sectors of
the population about their attitudes towards English education and by more closely exploring the
groups behind contemporary disputes on language policy. Understanding the linguistic fractures
bedeviling Pakistan is essential if Pakistan is to construct a more inclusive educational system
that can overcome rather than reproduce its deep social divisions.




 REFERENCES

Abbas, S. 1993. 'The Power of English in Pakistan'. World Englishes 12 (2): 147-156.

Abdullah, S. 1976. Pakistan mein urdu ka masla [The problem of Urdu in Pakistan]. Quetta:
Kuzak Publishers.

ABE. 1948. The Advisory Board of Education: First Meeting, June. Ministry of Education.
Karachi: Manager of Publications.

ABE. 1949. The Advisory Board of Education: Second Meeting, February. Ministry of Education.
Karachi: Manager of Publications.

ABE. 1958. Proceedings of the Seventh Meeting of the Advisory Board of Education, January.
Ministry of Education. Karachi: Manager of Publications.

Akhtar, B. M., Rahman, K. and Syed, M. 1986. Qaomi Zaban Akhbarat Ke Aine Men [Urdu: The
National Language as Reflected in the Press]. Islamabad: Muqtadra Qaomi Zab

Alavi, H. 1972. The state in post-colonial societies: Pakistan and Bangladesh. New Left Review,
I(74), 59–81.

Alavi, H. 2004. “The Territorial Dimension of Pakistani Nationhood”, pp. 85-102 in Pakistan:
Perspectives on State and Society, (Ed.) Abbas Rashid, SAHE: Lahore.

ASER. 2015. Annual status of education report: Pakistan 2015. Lahore: South Asian forum for
Education Development.

Baron & Byrne. 1987. Social Psychology: Understanding Human Interaction. Allyn and Bacon,
Boston.

Bourdieu, P. and Passeron, J. C. 1977. Reproduction in Education, Society and Culture. Beverly
Hills: Sage.

Brass, P. R. 1974. Language, Religion and Politics in North India. Cambridge University Press.

Census. 2001. 1998 Census report of Pakistan. Islamabad: Population Census Organization,
Statistics Division, Government of Pakistan.
CNE. 1959. Report of the Commission on National Education. Ministry of Education.

Constitution. 1973. Constitution of the Islamic Republic of Pakistan. Government of Pakistan.

GOP. 1966. Report of the Commission on Student’s Problems and Welfare and Problems.
Government of Pakistan, Ministry of Education.

Greif. A. 2006. Institutions and the Path to the Modern Economy: Lessons from Medieval Trade.
Cambridge University Press.
Khalique, H. 2006. The Urdu-English Relationship and Its Impact on Pakistan’s Social
Development. The Annual of Urdu Studies.

Knight, J. 1992. Institutions and Social Conflict .Cambridge University Press.

Lewis, M. P., Simons, G. F. and Fenning, C. F. 2016. Ethnologue: Languages of the world.

Nasr, V. 2001. The Islamic Leviathan: Islam and the Making of State Power. Oxford University
Press.

NEP. 1970. The New Education Policy. Ministry of Education & Scientific Research,
Government of Pakistan.

PNEP. 1969. Proposals for a New Educational Policy. Ministry of Education & Scientific
Research.

Rahman, T. 2019. Mother Tongue Education Policy in Pakistan. Book chapter: International
Handbook Policy in Asia. London Routledge.

Rahman, Tariq. 2011. Passports to privilege: The English-medium schools in Pakistan. Peace and
Democracy in South Asia, Volume 1, Number 1, January 2005

Rahman, T. 2006. Language and Politics in Pakistan. Karachi: Oxford University Press.

Rahman, T. 2005. Denizens of Alien Worlds: A Study of Education, Inequality and Polarization
in Pakistan. Oxford University Press.

Rahman, Tariq. 1997. The Urdu-English Controversy in Pakistan, Modem Asian Studies 31,1. pp.
177-2

Rahman, T. 2008. Language, Ideology and Power: Language-learning among the Muslims of
Pakistan and North India. Karachi: Oxford University Press

Rassool, N. 2007. Global issues in language, education and development: Perspectives from
postcolonial countries. Clevedon: Multilingual Matters Ltd.

Pakistan. Economic Survey. 2014-15. Economic Adviser's Wing, Finance Division , Government
of Pakistan, Islamabad

Powell, R. 2002. Language planning and the British Empire: Comparing Pakistan, Malaysia and
Kenya. Current issues in language planning, 3, (3), 205-279.

PEELI. 2013. Can English medium education work in Pakistan? Lessons from Punjab.
Islamabad: The British Council.