Вы находитесь на странице: 1из 260

Names

A Study of Personal Names,


Identity, and Power in Pakistan

s
es
Pr
ity
rs
ve
ni
U
d
or
xf
O
O
xf
or
d
U
ni
ve
rs
ity
Pr
es
s
Names
A Study of Personal Names,

s
Identity, and Power in Pakistan

es
Pr
ity
Tariq Rahman
rs
ve
ni
U
d
or
xf
O

1
1
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
Published in Pakistan by
Ameena Saiyid, Oxford University Press
No.38, Sector 15, Korangi Industrial Area,

s
PO Box 8214, Karachi-74900, Pakistan

es
© Oxford University Press 2015
The moral rights of the author have been asserted

Pr
First Edition published in 2015
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
ity
by law, by licence, or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
rs
address above
You must not circulate this work in any other form
ve

and you must impose this same condition on any acquirer


ISBN 978-0-19-940258-8
ni

Typeset in Adobe Caslon Pro


Printed on 80gsm Local Offset paper OR 52gsm Newsprint paper
U

Printed by _______________, Karachi


d
or
xf
O
Dedicated to my colleagues in the
SCHOOL OF EDUCATION,
Beaconhouse National University
and our time in Lahore

s
es
from September 2011 onwards

Pr
ity
rs
ve
ni
U
d
or
xf
O
O
xf
or
d
U
ni
ve
rs
ity
Pr
es
s
Contents

Preface and Acknowledgementsix

s
Translation and Transliterationxv

es
List of Abbreviationsxix

Pr
1. Introduction ity 1
2. Beliefs about Names 25
3. Names and the Islamic Identity 34
rs

4. Names, Power, and Class 54


ve

5. First Names, Nicknames, and No-Naming 70


ni

6. The Modernization of Names 89


U

7. Names as Traps: Name-Changing 108


d

8. Conclusion 126
or

Appendices135
xf

Glossary199
O

Notes214
References217
Index237
O
xf
or
d
U
ni
ve
rs
ity
Pr
es
s
Preface and Acknowledgements

The idea of writing this book dawned upon me in

s
March 2012 when I was reading some casual reference

es
to onomastics. I realized immediately that there was no
scholarly study of names. As my previous book From Hindi

Pr
to Urdu: A Social and Political History (2011) had been
published I was on the lookout for some new topic for
ity
research. And here was onomastics—an entirely new one.
At that time, since I had read nothing on names so far, I
rs
did not know what this kind of research entails. At that
ve

time I thought that, at the most, I would write only one


article on names as there would not be more material for
ni

anything else. I had a vague concept that names would be


indexed with identity: religious, rural and urban, class and
U

ethnic, etc. But beyond that there was nothing else I had
d

in mind. And, as usual, I had no funding and no research


or

assistants. I was, after all, in a faculty of education and


there was no way I could justify expenditure on this kind
xf

of research. But, of course, I was quite used to this situation


O

because my research had always been out of curiosity and


for personal gratification. So I started hunting for data and
for literature on the subject. In the latter quest the first
person who helped me was Dr Ijaz Gilani, Head of Gallup
Pakistan, who provided books which proved invaluable. As
for the first, the quest for data (names in this case), I have
many people to thank.
x Preface and Acknow ledgements

The first person who gave me data was Mr Sartaj Aziz, then
the vice chancellor of the Beaconhouse National University,
where I was employed. He gave me the names of students
of BNU, which was my sample of upper middle and upper
class students. Later, Mrs Nasreen Kasuri, Chairperson of
the Board of Governors of BNU, kindly agreed to give

s
me the names of some of the young children enrolled in

es
the Beaconhouse School System. Throughout the project
I was constantly helped by my colleague, Ms Amina

Pr
Gardezi, who got the names of the members of Lahore
Gymkhana, pointed out articles on naming, and also gave
ity
me much information about naming practices in Multan
and the Siraiki-speaking areas. I was also given names of
rs
elite children by Mrs Farzana Shahid, Registrar of BNU.
ve

Rural names were collected by Bilal Abdullah, a computer


expert at BNU who went to villages around Lahore to
ni

carry out field research for me. The same work was done
by Abid Hussain, my cook, for a village in Muzaffargarh
U

and Mohamamd Ilyas, who worked as part-time research


d

assistant for me assigned to collect names from the villages


or

of the Potohar area and some other names. As for lower


middle class names, I got them from my sister Tayyaba
xf

Azam, who is the principal of a charity school in Rawalpindi


O

called the Apna Model School. A breakthrough occurred


when Mr Tanwir Bukhari introduced me to, Gul Najam
Jamy and Ahsan Ali Mangi, both officials of the Benazir
Income Support Programme. They gave me names of about
800,000 people on this programme. This was a treasure
trove since these were poor people—so poor that they were
Preface and Acknow ledgements xi

in need of assistance by the state—from all the districts of


Pakistan. Names from Balochistan were given to me by Col.
(Retd.) Tahir Kardar, a personal friend and former course
mate. He is the only source of Baloch names since nobody
else provided me with any more names. Names from Sindh
were also sent to me by two personal friends, Lt  Col.

s
(Retd.) Khalid Khan and Captain (Retd.) Mohammad

es
Yusuf. The former owns an educational institution and the
latter was the principal of Cadet College, Larkana. The

Pr
names of Khyber Pakhtunkhwa were sent by Abid Zareef,
another friend and owner of a school. However, most of the
ity
names from KP were sent by the chairman of the Board
of Intermediate and Secondary Education, Peshawar. I am
rs
grateful to Dr Sarfraz Khan, Director of the Area Studies
ve

Centre, University of Peshawar who introduced me to him


and gave me the opportunity to learn about Pashtun names
ni

in the November of 2012 at Peshawar. Although I wrote to


such boards in other parts of the country nobody responded
U

to me. However, my secretary, Ms Shazia Irshad, was more


d

resourceful than myself because she got the names from the
or

BISE, Lahore, which I was not able to obtain despite my


efforts. Similarly, some of my other colleagues and students
xf

helped me meet people whom I interviewed and also got


O

me names from their schools. For instance, Mrs Surraya


Alam, Proprietor and Principal of the Toddler’s Academy
in Lahore, gave me the names of her young students;
Ms Humaira, a cousin of my wife and teacher in the Lahore
Convent, got me names of the students of her institution;
and Dr Waseem Anwar, Dean of Humanities at the FCC
xii Preface and Acknow ledgements

University, Lahore introduced me to people who could give


me information on Sikh names.

Yet another breakthrough occurred when I was given


the Humboldt Research Award in Germany for my past
research. I am much indebted to Professor Hans Harder,
Director of the Centre for South Asian Studies, University

s
of Heidelberg, who had recommended me for this coveted

es
award. I stayed from June till September 2012 and again

Pr
in July and August 2013 in Heidelberg and was helped
by Dr Christina Oesterheld in so many ways that I feel it
is impossible to thank her for her many kindnesses. The
ity
library of the Institute is quite rich and I was able to find
rs
much more material in it, and in a library in Mannheim,
than I thought I would.
ve

This book might seem to be a small monograph to the


ni

reader but it took a lot of very hard work which I had to


do alone. I also incurred many debts to so many people
U

that all of them cannot be mentioned individually. Let me,


d

therefore, thank all such people who helped me in arranging


or

interviews, providing names, filling in questionnaires


and answering questions relating to naming practices.
xf

Some parts of the manuscript were word-processed by


O

Shazia. Later, her successor, Ms Faiza Laeeq, checked the


manuscript to ensure that the calculations were correct. I
take this opportunity to thank both of them. Most of the
monograph was typed by Mr Shahid to whom I am much
indebted for working on weekends. All this effort would
have gone waste if the Oxford University Press (Pakistan)
Preface and Acknow ledgements xiii

had not accepted this manuscript for publication. I thank


Mrs Ameena Saiyid, the Managing Director of the Press,
for encouraging me and supporting my scholarly work. My
special thanks goes to Mr Hassan Rizvi who worked on
the manuscript of the book with me and gave me excellent
advice which improved the book. I also thank the other,

s
behind-the-scenes workers, in the press whose diligent

es
effort made it possible for you to hold it in your hands.

Pr
A word now about funding, without which research is not
possible. I generally do not write applications for funding
and when I did so once many years ago the funding was
ity
inadequate and took years to come through. Hence, this
rs
time I did not even bother to ask for it. Instead, I went
to my sole donor, my wife Rehana (Hana) and dipped my
ve

hands in the till drawing out about PKR 150,000 (USD


1=PKR 100). This money came entirely from our savings.
ni

For this I have to thank her sincerely. I am sure there are


U

ways of getting funding without loss of time, but I doubt


if I will learn them now. I will continue to do what I
d

have always done: think of something; start searching for


or

material; start interviewing people if the ethnographic


xf

method is to be used; read papers and books and start


writing.
O

I know this is not the way any really big project in the social
sciences or even small projects in the natural and applied
sciences can be carried on. I am not speaking of those kind
of projects. My work can only be limited since it is not
supported by funding and research assistants. However, my
xiv Preface and Acknow ledgements

reward is the pleasure I get when I write papers and books


and the tremendous feeling of gratification when they are
published. You see, I do not play chess and research is my
substitute for it.

Tariq Rahman
Lahore

s
January 2015

es
Pr
ity
rs
ve
ni
U
d
or
xf
O
Translation and Transliteration

The pronunciation of Persian and Arabic names in this

s
book is that of Urdu speakers and not that of Iranians

es
or Arabs. As the primary readership of this book will
be scholars of South Asia, familiar words and names of

Pr
persons and places have been transliterated using ordinary
Roman letters. Spellings have been standardized unless the
ity
names or phrases occur in a direct quotation or the name
is spelled in a certain way by its bearer.
rs

In the case of components of names from Arabic the


ve

following symbols have been used in the appendices where


such names occur, but not elsewhere in the book where the
ni

spelling conventions in Pakistan (such as the doubling of


U

the vowels to indicate length) have been used as they are


more familiar to the readers. The letters used below are
d

both capitals and lower case.


or

Ā as in ask (nasalized as ã).


xf

Ē as in Urdu/pet/=stomach (half high front vowel) (ẽ


O

nasalized).
Ī as in seat (ĩ nasalized).
Ō as in Urdu/log/= people (half high back vowel) (õ
nasalized).
Ū as in boot (ũ nasalized).
Kh as in Scottish loch/lox/(‫)خ‬
gh as in Afghanistan (velar fricative/Ɣ/or ‫)ﻍ‬
xvi Translation and Transliteration

Symbols in Persian,
Graphemes used Symbols used
Urdu, Punjabi, Pashto
in scripts in Arabic only
etc.

‫ز‬ z z

s
es
‫ﻅ‬ ẓ

Pr
‫ذ‬ dh

‫ض‬ ḍ
ity
rs
‫س‬ s z
ve

‫ص‬ ṣ
ni

‫ث‬ th
U

‫ﻕ‬ q q
d
or

‫ت‬ t t
xf

‫ط‬ ṭ
O

‫ڑ‬ ṛ

‫ح‬ ḥ

‫ہ‬ h h
Translation and Transliteration xvii

Other Symbols used for Arabic spellings

‫ و‬As a Persian/Urdu conjunction is transliterated as (-o)


whereas as an Arabic conjunction ‫ و‬is transliterated as
(wa).
‘ as in Arabic pronunciation of ‘Ali (glottal stop). This
is used to show orthography only in Arabic names in the

s
es
appendices.
’ hamza ‫( ء‬in Arabic it represents the glottal stop and

Pr
is used for the pause between two vowels. Its use varies
according to its position i.e. initial, medial, and final. For
South Asian speakers of Urdu and other languages it
ity
functions like a vowel i.e. schwa or/ǝ/). This is not used
rs
except in the appendices to indicate spellings.
ve
ni
U
d
or
xf
O
O
xf
or
d
U
ni
ve
rs
ity
Pr
es
s
List of Abbreviations

The following abbreviations have been used in the Glossary

s
and in the Appendices to show the origins of the names

es
and in a few places in the text. Please note that they do
not refer to short forms of the categories of samples in the

Pr
Annexures where these letters may coincidently recur.

A = Arabic
ity
B = Balochi
rs
BNU = Beaconhouse National University
ve

P = Persian
ni

Pa = Pashto
U

PBUH = Peace Be Upon Him. Used with the names of


the Prophet of Islam (e.g., Muhammad [pbuh],
d

the Prophet [pbuh]). Assumed to be used in this


or

book whenever he is referred to.


xf

Pu = Punjabi
O

S = Sindhi
Sr = Siraiki
Sans = Sanskrit
T = Turkish
H-U = Hindi-Urdu
O
xf
or
d
U
ni
ve
rs
ity
Pr
es
s
1
Introduction

s
es
Shakespeare’s lines ‘what’s in a name? That which we call
a rose/by any other name would smell as sweet’ (Romeo

Pr
and Juliet, Act 2: Scene II) is quoted very often by people
arguing that names are arbitrary labels which have no
ity
intrinsic, essential reality or meaning of their own. But
actually Shakespeare is arguing just the opposite as a careful
rs
study of the whole scene reveals. Let us revisit the scene.
ve

Romeo and Juliet, who have just fallen in love, are talking
to each other about their major problem which is the
ni

enemity of their families: the Capulets and the Montagues.


U

This is how the dialogue goes:

Juliet: O Romeo, Romeo! Wherefore art thou Romeo?


d

Deny thy father and refuse thy name!


or

Or, if thou wilt not, be but sworn my love,


xf

And I’ll no longer be a Capulet.


O

And a little later:


Juliet: ‘Tis but thy name that is my enemy.
Thou art thyself, though not a Montague.
What’s a Montague? It is not hand, nor foot,
Nor arm, nor face, nor any other part
Belonging to a man O, be some other name!
What’s in a name? That which we call a rose
2 Names: A Study

By any other name would smell as sweet.


A little later Juliet tells Romeo to ‘doff ’ his name. And Romeo
replies:
Romeo: Call me but love, and I’ll be now baptiz’d;
Henceforth I never will be Romeo.

The point is clear. Both want to escape a problematic

s
identity. Although she uses the name Romeo, what she

es
means is Montague. The Capulets and the Montagues are

Pr
two families feuding with each other hence the name of
anyone being born in those antagonistic groups represents
dire conflict. In short, the point here is that names are a
ity
matter of life and death and not meaningless labels. But
rs
Plato’s legacy, of which this is not an example, dies hard.
After all, had he not said in Cratylus that words have
ve

essential meanings of their own. Although Plato was talking


about language in general and not about proper names the
ni

idea is still alive in some form or the other in Pakistani


U

society. As we shall see later, there is a common belief in


Pakistan that names reflect the personality of a person.
d

Thus someone called Jamil (beautiful) would somehow


or

be good looking. Though all evidence is against such


xf

assumptions, people still believe what Plato’s mouthpiece


argued in Cratylus.
O

But whatever the significance of names, the fact is that


it is impossible to function without them in a modern,
bureaucratic society. The Pakistani newspaper The News of
25 November 2012 carried an article saying that orphans
cannot apply for an identity card in Pakistan because they
cannot fill in their father’s name. This means that, without
Introduction 3

certain names, one is a non-person—unable to drive a


vehicle, to travel abroad, to get good jobs and to buy
property (Shahbazi, 2012). But with the wrong name, one
might even be murdered as Romeo could because of falling
in love with Juliet.

Personal names are badges of group identity, indicators

s
of lineage, socio-economic class, level of modernization,

es
etc., in Pakistan and of course in other countries of the

Pr
world. Names construct identities based upon the belief
system of the community in which they occur. They are
indexed to our identities and roles in society and one way
ity
of looking at them is to trace out what Jan Blommaert,
rs
a sociolinguist, calls ‘orders of indexality’. Explaining the
concept Blommaert refers to Michel Foucault’s concept of
ve

‘order of discourse’ asserting that there is a ‘stratified and


ordered nature of such indexical processes’ (Blommaert,
ni

2010: 37). Just as the particular variety of the language one


U

uses, one’s accent or one’s script, indexes one’s identity, one’s


place in a certain social order and the identity one chooses
d

to display in a certain social setting, similarly our names


or

and alternative names do the same. Personal names also


xf

index one’s niche in the social order as they are indexed


to webs of power relations, both personal and group. They
O

are products of history and embody layers of existence of


a socio-cultural group for ages. In a sense, then, Pierre
Bourdieu’s theoretical construct ‘habitus’ describes them
well. As Bourdieu says:

The habitus—embodied history, internalized as a second


nature and so forgotten as history—is the active presence of
4 Names: A Study

the whole past of which it is the product. As such, it is what


gives [our] practices their relative autonomy with respect
to external determinations of the immediate present. This
autonomy is that of the past, enacted and enacting, which,
functioning as accumulated capital, produces history on the
basis of history and so ensures the permanence in change
that makes the individual agent a world within the world

s
(Bourdieu, 2009: 56).

es
Thus the name ‘Muhammad’ is as much embodied in

Pr
history of Islam as is ‘Ram’ or ‘Christina’ of Hinduism and
Christianity respectively. Both names index their bearer to a
ity
certain collective or group identity of a certain belief system.
Names can, therefore, be used to proclaim a desiderated
rs
ideological position, e.g., self conscious Islamization of
ve

names could indicate a stress on Muslim identity; or, as in


some cases, the hiding of an identity to counter perceived
ni

stigmatization, e.g., a Pakistani Christian’s use of Muslim


U

names; or, as in many cases, to construct a higher or a more


urban, modern social identity for the bearer of that name.
d

In short, personal names are important constructors and


or

signifiers of one’s identity and one’s place in the networks


of relationships in society. They are related to the belief
xf

system of a society and are, in the last analysis, indirectly


O

related to historical, cultural, and economic factors and how


members of a society construct their social reality through
them. They are very much part of the language—meaning
not a formal code such as Urdu, Punjabi, or Pashto—but a
way of ‘talking about’ human beings which actually helps
us construct our reality—the way we look at those human
Introduction 5

beings and, of course, how we feel for them and treat them
(Grace, 1987: 3).

In a modernizing society like Pakistan, naming is influenced


by the processes of modernity as manifested in this part of
the world. Here modernity brought up a new orientation of
the self: in the image of the seemingly educated, rational,

s
advanced colonial master (Anglicization); a reactive or

es
self-consciously modern, ‘pure’ Islamic self (Islamization)

Pr
and an assertive ‘authentic’ ethnic identity (ethnicization)
which were all versions of modernity (Eisenstadt, 1999;
Therborn, 2003). This point will be taken up briefly in
ity
the chapter on the modernization of names (chapter 6)
rs
so I will say no more about it except to point out that
there are at least three language systems with which the
ve

manifestations of onomastic modernity in Pakistan are


associated. First, English with which the construction of
ni

the Anglicized self is associated. Secondly Arabic, which


U

constructs the Islamic self. And thirdly, the indigenous


languages of Pakistan (Pashto, Sindhi, and Balochi/Brahvi)
d

which construct the ethnic identity. In the first case, as we


or

will see, it is not a case of borrowing English names, except


xf

nicknames and initial letters of proper names in some cases,


but more of assimilating the norms of naming; the order
O

of cataloguing peoples’ names according to Anglo-Saxon


rather than indigenous fashions. In the second case, that
of Arabic, there is direct borrowing. Indeed, the Arabic
language is considered the repository of Islam to the extent
that words of Arabic, not necessarily considered appropriate
for naming, are borrowed by Pakistanis to assert their
6 Names: A Study

Islamic identity. And in the final case of ethnic assertion,


the indigenous languages as well as heroic figures from
indigenous nationalist archives, are borrowed to supply
personal names.

In our weblike globalized world, names, like languages


(including accents, registers, varieties, scripts etc.), function

s
to index identities. As mentioned above, Jan Blommaerts’s

es
concept of ‘orders of indexality’, which outlines ‘the

Pr
dominant lines for senses of belonging, for identification
and roles in society’ is useful (Blommaert, 2010: 6). With
these concepts in mind let us see what work on names and
ity
naming practices are relevant for our study of personal
rs
names and identity in Pakistan.
ve

Review of Literature
ni

In a comprehensive review of scholarship on identity-


construction Cerulo argues that ‘Anti-essentialist inquiries
U

promote the social construction of identity as a more viable


d

basis of the collective self ’ (Cerulo, 1997: 387). Names,


or

though not mentioned in Cerulo’s article, are a crucial


variable for one’s own as well as other people’s perception
xf

of one as a member of a group, or sub-group. In most cases


O

individuals, cocooned in the sustaining culture of their


group, may never examine the relationship between their
personal names and identities. However, as research on the
psychological perceptions of identity show, an individual
may for some reason or the other be challenged to examine
these perceptions—especially if the identity labeled by the
name becomes problematic—and may resolve whatever
Introduction 7

problem he feels by negotiating their identity (Frable, 1997:


147–8). When this happens by the person changing the
name it is relevant for this study.

Much work on names has been done by scholars of


onomastics. They have analysed and explained the meanings
of names and studied the relationship of names with

s
society. Indeed there is a journal called Names: A Journal of

es
Onomastics published by the University of Louiseville since

Pr
1952 in the United States, which focuses on the scholarly
study of names. A synoptic view of its contribution to
onomastics is given by its editor Frank Nuessel (2013).
ity
Among other issues, there are studies of the names of
rs
characters in literature (Gasque, 1991; Palaeas, 2005; Abel,
2013; Robinson, 2013); nicknames (Moore, 1993; Holland,
ve

1990; Leslie and Skipper, 1990); forms of address (Wright,


2013; Murray, 2002); name-changing (Klerk, 2002; Jacob
ni

and Horn, 1998; Zhonti and Millward, 1989; Ragone,


U

2012; Blair, 1991) and the relationship of names with


other disciplines of the social sciences (Algeo and Algeo,
d

2000) and society in general (Moore, 1993). In addition to


or

all this, work that is highly relevant to our purpose comes


xf

from anthropologists and sociologists. Beginning from the


pioneering work on naming systems from that doyen of
O

anthropologists, Claude Levi-Strauss, in his book La Pensee


Sauvage (The Savage Mind, 1962: 172–216) one can name
many others with similar interests. Levi-Strauss examines
the basic issue of universalization and particularization in
human societies. He refutes the claims of philosophers
8 Names: A Study

and linguists, to be mentioned briefly later, that names are


meaningless labels as follows:

We need to establish that proper names are an integral


part of systems we have been treating as codes: as means of
fixing significations by transposing them into terms of other
significations. Would this be true if it were true, as logicians

s
and some linguists have maintained, that proper names are,

es
in Mell’s phrase, ‘meaningless’ in signification? (Levi-Strauss,
1962: 172).

Pr
An overview of sixty such societies compares how names
ity
are given; who gives them; whether a ceremony is held
when they are given; whether nicknames are used or
rs
not; how and why are names changed and so on (Alford
ve

1988). Another valuable study on onomastics is Bruck and


Bodenhorn’s edited volume entitled The Anthropology of
ni

Names and Naming (2006). Yet another book on names and


naming practices is Framing my Name (Kumar, et. al., 2010).
U

Like the other literature, it investigates these phenomena


d

in several developing societies (India, Cameroon, China,


or

etc.) but, in addition to that, it also problematizes the way


foreign students in Australia’s higher education perceive
xf

their names and the issues connected with them.


O

We now come to the position, generally held by philosphers,


that personal names are labels and do not indicate any
attribute of the person with the name (Ryan, 1981;
Gardiner, 1940; Mill, 1882) and that they are, therefore,
empty of all descriptive content. Mill says that the name
John ‘is not conferred upon them to indicate any qualities,
Introduction 9

or anything which belongs to them [all males called John]


in common’ (Mill, 1882: Chapter II, 34). But it does not
appear to Mill, or indeed to other philosophers, that the
name John does at least ‘connote’ (a term used by Mill)
that he is a male and belongs to the European (British)
civilization. John Searle, in his book on the philosophy of

s
language, while agreeing with the basic idea that proper

es
names refer to objects elaborates upon the imprecision or
‘looseness’ of their descriptive functions as follows:

Pr
But the uniqueness and immense pragmatic convenience of
proper names in our language lies precisely in the fact that
ity
they enable us to refer publicly to objects without being
rs
forced to raise issues and come to an agreement as to which
descriptive characteristics exactly constitute the identity of
ve

the object. They function not as descriptions, but as pegs on


which to hang descriptions. Thus the looseness of the criteria
ni

for proper names is a necessary condition for isolating the


referring function from the describing function of language
U

(Searle, 1969: 172).


d

This idea is also expressed by Mark D’Cruz, who also argues


or

that proper names are placeholders for sets of descriptions


xf

of an individual (D’Cruz, 2000: 746). Gardiner, after giving


a rather purist definition of proper names, qualifies it by
O

saying that there ‘are less pure proper names than the purest
because of the assistance that, on rare occasions, they might
give by their suggestion of sex, nationality, or country’
(Gardiner, 1940: 42). Philosphers, for the most part, never
get beyond the idea that a name refers to an object so that
it becomes recognizable by others. So when they mention
10 Names: A Study

identity they mean the distinctive features of single objects


(persons, mountains, rivers, etc.) which may exist in
space-time for finite periods, changing in characteristics
yet retaining a core which is considered stable enough to
constitute an identitymarker. This study, however, is not
concerned with these philosophical preoccupations. It refers

s
to group identities constituted by names and components

es
of names. A useful concept, borrowed from the work of
the French philosopher Michel Foucault, is that of the

Pr
‘technologies of the self ’, which he presented in a seminar
at the University of Vermont in 1982. Foucault mentioned
ity
how it was possible to transform oneself ‘in order to attain a
certain state of happiness’ by, among other things, the use of
rs
signs (Foucault, 1982/1988: 16). A name, obviously, is one
ve

such sign. It is normally chosen by the namegivers in order


to place a person in a certain social space with its attendant
ni

power relations, but sometimes people change their


names or adopt alternative names to construct a certain
U

desiderated image of the self. In this sense, then naming is a


d

powerful technology of the self. That is why some scholars


or

refer to the ideas of philosophers to argue that ‘naming


involves powerful speech acts, making history, constituting
xf

persons and the social relations and systems within which


O

they are embedded at least—families, communities, states,


and empires’ (Palsson, 2014: 620). In short, the ideas
of the philosophers of language—Wittgenstein (1953),
Austin (1962) and Searle (1969)—provide insights in
understanding how names work in society; what force they
possess and how they trigger our responses to their bearers
individually or as a class, i.e., a category such a group
Introduction 11

identity. This, however, will be taken up in the conclusion


of this book.

Like philosophers, some linguists too take a purist approach


by stressing how names are to be classified. Anderson, for
instance, classifies them as determinatives. This category
‘is characterized notionally as maximally referential and

s
thus non-predictable’ (Anderson, 2004: 470). But some

es
sociolinguists have moved to their relationship with culture.

Pr
Rymes, for instance, argues that they are ‘indexical of a
rich realm of cultural and personal associations’ (1996:
246). Taking the case study of a gang member of Los
ity
Angeles called ‘Little Creeper’, she argues that names may
rs
be criminalized and ‘hold both implicit meanings and
explicit referent’ (Ibid.: 258). This reminds us, as Maurice
ve

Bloch suggests in his study of tekonymy in a Madagascan


tribe that ‘the usage of names cannot be separated from
ni

pragmatics and that names are therefore used to “do” an


U

almost unlimited number of things’ (Bloch, 2006: 98). Such


insights are very relevant for Pakistani names as they have
d

implicit meanings and, of course, these meanings can be


or

used to do many things: to index a certain identity, to give


xf

respect, to insult and so on.


O

Although names are related in numerous and complex ways


with identity, a few examples may be helpful, for instance,
patrilinealism may be reflected in names. This was true for
the Ju Hoansi people of Botswana and Namibia during
the 1950s and 1960s (Draper and Haney, 2005) and it is
true for Pakistan where the children belong to the father’s
family and names are one way to indicate this aspect of
12 Names: A Study

their identity. Personal qualities which go into identity-


construction—bravery, generosity, intelligence—may also be
reflected by names. Thus, in Cheyenne society a man ‘might
take a name in addition to his present one’ that may reflect
such positive qualities (Moore, 1984: 302). Name-changing,
indeed, symbolizes ‘the emergence of new characteristics

s
or a new identity’ (Alford, 1988: 85) and ‘males take up

es
new names in societies while women adopt new names in
11 societies at marriage precisely because these are seen as

Pr
emergent identities’ (Alford 1988: 86).

In some cases collectivities use names in a bid to emphasize


ity
one or the other identity. For instance, the Afro-Panamanian
rs
residents of the island of Bastimentos have an official
Spanish-derived and an ethnic Creole-derived name. They
ve

use the latter for in-group solidarity (Aceto, 2002: 601).


The Meithei-speaking people of Manipur in India, who
ni

accepted Hinduism in the 18th century, use names as a site


U

for contestation between political ideologies. Those who


choose pre-Hindu indigenous names want to break with
d

India while those who use Hindu names want integration


or

(Chelliah, 2005: 169). In such cases one constructs a


xf

politically oriented group identity. In other cases, however,


as in the Gaelic communities of East Sutherland in the
O

Highlands of Scotland by-names rather than formal names


are used for in-group solidarity (Dorian, 1970).

In-group solidarity presupposes out-groups. These ‘others’


may see a collectivity as a monolith because of its use of
ethnic, religious, and racial labels as part of their personal
names. Thus, in present day Western countries Muslims,
Introduction 13

though differentiated by citizenship (Turkish, Iranian,


Saudi, Pakistani, Indonesian, etc.), sect1 (Sunni, Shia) or
subsect (Ismaili, Deobandi, Barelvi, Wahabi, etc.) 2 are
perceived as a group (Lauderdale and Kestenbaum, 2000)
and maybe discriminated against in employment as they
evoke, negative stereotypes (Bertrand and Mullainathan,

s
2004). Such situations have an effect upon the use of names.

es
In most cases immigrants (and not necessarily Muslims)
adopt names which enable them to pass for a local (Bursell,

Pr
2012). This, indeed, is the usual practice of workers in call
centres who have to interact with Western clients. These
ity
operators use short Western names (Bill, Jill, etc.) to present
an integrative Western identity (Mirchandani, 2004: 365–
rs
66; Rahman, 2009: 236). In some cases foreigners might
ve

want to retain the correct pronunciation of their names as


it is part of the identity they want to maintain and privilege
ni

(Lehiste, 1975). But such discrepant responses are related


to the personal or group-power of the persons in question.
U

The call centre worker cannot insist even on the name let
d

alone its correct pronunciation, while a foreign professor,


or

ambassador, or business tycoon can.


xf

In short, the study of names, including alternative names,


can provide insights into the construction and perception
O

of identity. These, in turn may be related to belief-systems,


power differentials and social stratification within societies
(Collier and Bricker, 1970; Treis, 2005; Barnes, 1980;
Beidelman, 1974). Changes in names—as among the Zulus
(Suzman, 1994)—are related to the identity of societies as
a whole.
14 Names: A Study

Despite the significance of personal names in understanding


the complex issues of identity, power and belief-systems,
there is no study of personal names in present day Pakistan.
Indeed, there are no major studies of Muslim names in
South Asia though there are several studies of the names
of other Indian communities (Emeneau, 1976), including

s
a book-length study of ancient names of Hindus which

es
does, however, contain much information on their naming
practices. The only reference to Muslims in this book

Pr
is in the form of a footnote about the importance of
certain letters as derived from sacred books, but for this
ity
the author refers to Richard Temple whose work will be
referred to below (Hilka, 1910: 43). References to Muslim
rs
names, however, are found in onomastic studies both by
ve

colonial scholars as well as later ones. The tone of the


British colonial officer-scholars is condescending and
ni

arrogant towards Indians, but they have provided valuable


information nevertheless. The pioneering study of Punjabi
U

names mentioned above is Richard Temple’s Dissertation


d

on the Proper Names of Punjabis (1883), which is based on


or

data gathered in the ‘Ambâlâ District and neighbourhood,


where the Hindu element largely predominates’ (Ibid.: 2).
xf

However, Temple devotes a whole chapter (IV) to Muslim


O

names (Ibid.: 40–51) which is partly relevant even today for


a scholar of Pakistani names. W. F. Sinclair’s essay, though
lacking detail, mentions that prestigious titles (or ‘caste’
names) such as Syed, Khan, and Sheikh—all proclaiming
foreign descent—have actually been appropriated by local
Muslim converts because they facilitate claim to a high
social status (Sinclair, 1889: 171–2). Unfortunately, the only
Introduction 15

study purporting to be on Muslim names—Colebrooke


1881—merely adds to the literature on Muslim (mostly
Arabic) names on which there is a lot of contemporary
material: Jordan (Abd-el-Jawad, 1986), Kuwait (Yassin,
1978), Turkey (Spencer, 1961), Central Asia (Abazov,
2007) and nicknames as family names among the Arabs

s
(Goitein, 1970). Colebrooke, therefore, does not add to our

es
knowledge of Indian Muslim names.

Pr
There is, indeed, more to be gained by reading scholars
working on Hindu names in India if one is trying to
understand Indian Muslim names. For instance, M. B.
ity
Emeneau in his article on Hindu names tells us that the
rs
Islamic heritage languages contribute many linguistic
components to Hindu names in north India such as Ram
ve

Baksh and Jawahar Lal ( Jawahar means pearls or jewels;


Lal means a red, precious stone) (Emeneau 1978: 117).
ni

Francis Britto in his study of personal names in Tamil


U

society tells us that Muslims do not use traditional Tamil


names ‘since most indigenous names have some connection
d

with Hinduism’ (Britto, 1986: 354).


or

The only major study of Muslim names which also looks


xf

at South Asian names, though in passing, is Anne Marie


O

Schimmel’s Islamic Names (1969). But Schimmel’s study


is actually on Arab (and some Turkish) names and most
of her observations are about the incorrectness of South
Asian Muslims’ names on account of their ignorance of
Arabic (Schimmel, 1969: 25, 28, 62 et passim). However,
there are other aspects of naming which Schimmel
does not touch upon. For instance, the concealment of
16 Names: A Study

identity is an important subject. However, there is only


one article, Theodore Wright’s, on name changing among
Indian Muslims in order to merge with the Hindu
majority (Wright, 2006). Two other studies, both of minor
significance, which are indirectly relevant to Pakistani
Muslim names are about Bengal (Dil, 1976: 51–71) and

s
Afghanistan (Waleh, 1964: 4–6). In short, there is a gap in

es
our knowledge of Pakistani names.

Pr
Objective
The aim of this study, then, is to describe Pakistani Muslim
ity
names, naming practices, and beliefs in order to understand
identity construction and the way power operates in
rs

Pakistani society. For the latter the relationship of names


ve

with identity—religious, ethnic, class, and regional—will be


traced out in some detail. Change of names as it relates to
ni

identity-construction—such as the aspiration for a stricter


U

Muslim or modern urban identity—among Muslims or the


avoidance of persecution or negative bias among Pakistani
d

Christians will also be given attention. This information,


or

it is hoped, will provide insights into the construction and


xf

perception of identity and their relationship with power,


class, and religion in Pakistan.
O

Methodology
First, a pilot survey was taken. For this, small samples of
less than 5000 names were taken from (a) the villages of
the Punjab and Khyber Pakhtunkhwa (KPK), (b) urban,
modern youth from Lahore and (c) lower middle and
Introduction 17

middle class people from the Punjab. The samples of village


names were taken from two villages near Lahore, one near
Muzaffargarh (South Punjab) and one near Rawalpindi
(Potohar region) (called Rural-F and M in Annexure B).
The sample of the names of upper middle class boys and
girls are students of the Beaconhouse National University,

s
Lahore. As they are fee paying students of an expensive

es
private university they belong to the upper middle classes in
income, lifestyles and hence, naming practices (BNU-B &

Pr
G). The last sample is from a charity school in Rawalpindi
called the Apna Model School (Apna-G.). As this school
ity
has very few boys, only a sample of female names has been
used for finding Islamic and caste components like in all
rs
the other samples. (For an explanation of these samples see
ve

Annexure-B).

The large samples of names for the detailed study were


ni

taken from all over the country. These names are from
U

the lists of students who appeared in the matriculation


(Class 10) examination of the Boards of Intermediate and
d

Secondary Examination Lahore (Punjab), Peshawar (KPK)


or

and Larkana (Sindh). For the names of the 1950s the


xf

matriculation gazettes of the middle 1950s and 1960s were


consulted and for those of the late 1990s, the gazettes of
O

2012 were consulted. Unfortunately, however, the gazettes


of Sindh for the 1950s and 1960s were not available. For
Balochistan no gazettes were available at all. Thus, for these
two provinces names were taken from voters lists of 2013
elections which contain the names of men and women and
their fathers. This leaves out the names of elderly women
18 Names: A Study

(named in the 1940s and ‘50s) from both these provinces


while samples of such names are available for the Punjab
and KPK. The names of the elite were taken from the list of
the members of an elite club in Lahore. For children from
this class the names of pupils of expensive private schools in
Lahore and Islamabad were taken. The names of very poor

s
and rural people come from the National Income Support

es
Programme, previously named after Benazir Bhutto. These
names are from all the provinces of the country but ages are

Pr
not mentioned (Poor-M & F). Details of these databases
are given in Annexure-B. ity
In addition to that a survey of a sample of the population
rs
was carried out through questionnaires in Urdu and English
which were distributed to 400 people. However, only 372
ve

filled them in mostly when they were in one location as


in a classroom. Out of these 57 had to be discarded as
ni

they were not legible. Thus only 315 questionnaires have


U

been used for analysis in this study. The respondents were


selected through non-random convenience sampling.
d

However, most of them (264) were students to whom


or

questionnaires were given in classrooms where all of them


xf

filled them in. The other 108 were people whom the
author met during field work. They filled them in if they
O

so desired. The respondents are from Lahore, Rawalpindi,


Peshawar and Hyderabad. It was not possible to go to
Quetta so there is no representation from Balochistan. The
sample has 82 males and 233 females. More females were
given the questionnaires because many of them change
their names on marriage so their input was deemed more
Introduction 19

significant than those of males for the purposes of this


study. Moreover, since the other data lacks female names
and the informants were mostly males, it seemed advisable
that more females should be asked for their opinions to
counteract the preponderance of male opinions which were
available. The sample has more educated than uneducated

s
people, whereas the reality of the country is otherwise.

es
This is a sampling bias because it is easier to get people
in educational institutions (clusters) than to find them in

Pr
other places. Moreover, it is difficult to get illiterate people
to respond to written questionnaires as they mistrust them.
ity
To overcome this problem such people were questioned
during casual conversation.
rs

Some information about names is found in Urdu


ve

literature which has been used here. Further information


on naming practices, beliefs, and the structure of
ni

names for both Muslims and Christians was obtained


U

through unstructured interviews and informants from


these communities. Only some of these interviewees
d

and informants have been identified in the section


or

on references. Most are not mentioned individually


xf

because the information was volunteered in the midst of


normal, spontaneous conversation in which many people
O

participated without being recorded, as that would have


made them less communicative. These interviews were held
all over Pakistan with the exception of Balochistan where
militancy precluded visits and interviews. However, Baloch
intellectuals and academics were interviewed in the other
cities of Pakistan. It is hoped that the use of these methods
will help identify naming patterns, practices, and beliefs
20 Names: A Study

which, in turn, provide insights into how Pakistani society


is organized and how it functions.

Names have components—Asif Ali Zardari, for instance


has three components—which have been used to calculate
the percentage of Islamic as well as caste or prestigious
components in names from different ethno-linguistic

s
groups and periods of Pakistan’s history. Religious compo­

es
nents are classical or indigenous. In the case of males, the

Pr
former are often the attributes of God used as personal
names; the attributes of Prophet Muhammad (pbuh) are
also used similarly; the names of the prophets and other
ity
people mentioned in a positive manner in the Quran; the
rs
names and alternative names (laqabs expressing qualities)
of the four orthodox caliphs Abubakr Siddiq (573–634),
ve

Omar ibn al-Khattab (579–644); Uthman ibn Affan (577–


656) and Ali ibn abu Talib (598–661); the Companions of
ni

the Prophet (sahaba); those who had seen the Companions


U

(Tabiun) or the names of those who had seen the Tabiun


(Taba Tabiun). In addition to this the names of Shia
d

imams (spiritual leaders), some of which are shared by


or

the Sunnis, have been counted among ‘Shia components’


xf

(See Appendix-1). The Shia imams and the components


associated with them—Jafari, Kazimi, Rizvi, Naqvi, Askari,
O

etc.—are also given in Appendix-1.

In this context it should be mentioned that these


components are actually shared by Sunnis ( Jafri. Int., 2013).
However, they probably occur more frequently among
the Shias and the public perception is that they are the
onomastic distinguishers of Shias from Sunnis. Other Shia
Introduction 21

names have components like—kalb (dog)—such as kalb-e-


Ali, kalb-e-Hussain, i.e, dog of Ali and Hussain—which
have been changed to qalb (heart) in the last few decades.
Tahir Abbas, a Shia Alim, said that this had happened but
that it did not make religious sense (Abbas. Int., 2013).

In female names, the Islamic components comprise the

s
names of Prophet Muhammad’s (pbuh) mother, wives, and

es
daughters; women mentioned in the Quran; women con­

Pr
sidered holy by Muslims; women companions (sahabiyat)
of the Prophet (pbuh) and the usual Shia and indigenous
components.
ity
All the components mentioned above, with the exception
rs
of the folk (indigenous) ones, are from what Gellner calls
ve

‘high Islam’ which, in the case of Pakistan, implies a strict


interpretation of the faith in which human beings are
ni

guided by the Quran and the traditions of the Prophet


but need no saintly intercessors. The ‘low Islam’ can be
U

considered folk Islam with belief in the intercessory power


d

of saints often called Barelvi in Pakistan (Gellner, 1995).


or

Components of this form of Islam refer to God as Khuda, a


Persian term for the deity, and often use Punjabi and Urdu
xf

words for gift or giving (Ditta, Baksh, Dad) and religion


O

(Deen). These are some of the folk (indigenous) religious


components mentioned above. Such names sometimes refer
to the belief, which is considered to be against the tenets
of fundamentalist and classical (high church) Islam, that
children are given by saints peer (plural peeran in Punjabi) or
a particular saint such as Sheikh Abdul Qadir Jilani called
Ghaus ul Azam (1077–1166) or the Prophet (pbuh) (Nabi,
22 Names: A Study

Rasool). The strict belief is that only God gives children and
intercession of anyone else is not possible.

In addition to that, certain names are categorized as


‘radical’. These are defined as those which are contingent
upon the view that the ‘West’ is hegemonic, exploitative and
anti-Muslim. Among these names are Saddam and Osama

s
and some typically Arabic components of names not used

es
in Pakistan in its early years. Saddam Hussain (1937–2006)

Pr
and Osama bin Laden (1957–2011) are seen in Pakistan to
have stood up to Western bullying. Hence they are iconic
figures among those who identify with radical, political (or
ity
Jihadi) Islam. Sometimes, people use them without having
rs
radical Islamic beliefs only because of a vague anti-Western
feeling or simply beause their neighbours value such names.
ve

Furthermore, the trend of adding Arabic components of


names, such as, –umme (mother of ), -ibn or –bin (son
ni

of ) and –abu (father of ) and other Arab names—which


U

custom was actually not in use in this country in its early


years except with the names of some revered figures
d

from Islamic history—are taken here as indicators of an


or

emergent, hardline Islamic identity. The names of Muawiya


xf

ibn Abi Sufiyan (560–652) and his son Yazid (647–683)


are included in this list as being indicative of a hardening
O

Sunni identity as these names are deliberately provocative


to Shias.

The calculation of caste and prestige components is based


upon counting the frequency of titles used as names (Khan,
Malik, Sardar, Syed, Nawab, etc.); caste (Bhatti, Raisani,
Afridi, etc.) and names referring to foreign prestigious
Introduction 23

places of origin (Bukhari, Hamadani) or great family names


(Bilor, Isphahani). In female names rural components (Bibi,
Bano, Khatoon, etc.) are used to index rural or old-fashioned
identities. The use of mens’ names, whether first names
or caste ones, in womens’ names is also an indicator of an
urban and modern identity but could not be computed. In

s
addition to this, frequently occurring first names have also

es
been counted with a view to understanding naming patterns
and the relationship of identity with names (Annexure M).

Pr
After this introductory chapter there will be a chapter
(No. 2) on the beliefs of most ordinary Pakistanis about
ity
names and naming practices and usages. This provides the
rs
background information which facilitates the understanding
of Pakistani attitudes towards names. Chapter 3 is about the
ve

influence of Islam on names and the ways in which names


index religious and sectarian identities. One major question
ni

which will be answered in this chapter is whether, and if


U

so, how the perceived Islamization of Pakistani society is


reflected in names? Chapter 4 studies the connection of
d

names with power and class, especially among the powerful


or

rural classes. Moreover, it also explores how caste and other


xf

group labels have started functioning like the Western


family (last) names. Chapter 5 is on first names and the
O

various changes some have undergone. It also examines


the process of Islamization of names by looking at the
popularity of Islam-inspired first names. In addition to that
the chapter also touches upon the phenomena of nicknames
and no-naming strategies. Chapter 6 examines the
modernization of names. Attention will also be given to the
24 Names: A Study

tension between seeking a modern identity through naming


and its opposite trend, i.e., seeking traditional names for
an authentic identity. The last chapter (No. 7) is about
‘name changing’ and some names are described as ‘traps’ in
it. The chapter looks at major reasons for changing names
among powerful as well as marginalized and powerless

s
groups. For the powerful, names are chosen to boost the

es
ego while for the weak and marginalized communities they
are ‘traps’ and one changes them to escape the negative

Pr
consequences for bearing certain kinds of names. Among
other things it looks at the adoption of Muslim names by
ity
the Christian community to escape being discriminated
against. In the end there is a conclusion followed by
rs
appendices and annexures containing necessary details
ve

and statistical figures about the frequency of occurrence of


certain types of components in names. These are followed
ni

by notes, a bibliography comprising a list of sources and a


list of interviewees. In the end there is an index to facilitate
U

the reader.
d
or
xf
O
2
Beliefs About Names

s
es
Muslims comprise 96.28 per cent of the population of
Pakistan (Census-R, 2004) and their names neither follow

Pr
the Arabic structural pattern described by Annemarie
Schimmel nor the Western one comprising a personal,
ity
middle and family name. Arab names have the terms,
kuniya, nasab, nisba, and laqab. Kuniya refers to being a
rs
parent, i.e., Abu (father of ) or umme (mother of ); nasab
ve

means being the son of (-ibn) or daughter of (-bint); nisba


is one’s affiliation to a place or sometimes a profession
ni

(al-Iraqi means a person is from Iraq); and laqab is a


U

positive alternative name or nickname, e.g., sadiq (truthful)


(Schimmel, 1989: 1–13). Such names are not used in
d

Pakistan except by those who want to assert a newborn


or

fundamentalist or radical Islamic identity as we shall see


later. Moreover, the Western type use of a fixed last or
xf

family name is also not firmly established in all parts of


O

the country. Indeed, like the Tamils (Britto, 1986: 349) and
the Hindus of North India (Vatuk, 1969: 256), Pakistanis
have varying surnames and the ‘caste name’—the use of
this concept for Muslims will be explained later—may be
optional. Thus, as in North India, ‘the notion of a particular
surname as a patronymic is not well established, and it
26 Names: A Study

is not unusual for the “surname” of father and son, or of


brother and brother, to differ’ (Vatuk, 1969: 256). Christians
too exhibit inconsistent usage of family names especially as
the name Masih (Messiah or Jesus) is falling into disuse
since it declares the religion of its bearer too openly in an
increasingly biased society.

s
Names are chosen from the internet among the urban

es
computer literate people; otherwise from books on

Pr
childrens’ names, which are widely available in Pakistan’s
cities, and by consulting elders; clergymen (maulvis, peers)
etc. A survey of the market shows that among such books,
ity
those inspired by Wahabism prevail, which is based more
rs
on Arab culture and an austere, rigidly monotheistic
philosophy. Thus these books disregard all local practices,
ve

local names, mystical beliefs about names influencing life


and personality and other abstract qualities. In some cases
ni

names are selected with the belief that they will counter
U

bad luck. Ghulam Jilani Barq, an intellectual and writer


of pre-partition Punjab, writes in his autobiography that
d

a pandit told him that, being born on a Saturday, he was


or

unfortunate. There was a Hindu antidote to this presumed


xf

bad luck but he preferred an Islamic cure which was to take


out the numerical value of his own name through Abjad 3,
O

find names of God with the same value, and then repeat
these names as frequently as possible (Barq, 1982: 178).

Yet another belief is that if a daughter is named ‘Bushra’


or ‘Naveed’, both meaning good tidings, the next child
will be a male. Another onomastic device to procure male
children, reported from South Punjab (Multan area), is to
Beliefs A bout Names 27

call a girl ‘bhiravan’ (brothers in the local Siraiki language).


In our sample we came across a girl with this name and
named for this purpose, who was employed as a maid in
the family of the Gardezis of that area (Gardezi. Int.,
2012). Indeed, there are many women carrying this name
in our sample from the BISP data (Poor-F). In some areas

s
names of ancestors are given preference. Pahore, a Sindhi

es
academic, told the author that his own name Ramzan, was
that of the eldest dead person in the family who happened

Pr
to be his grandfather. This is done to keep the memory of
the ancestors alive (Pahore. Int., 2013). Evidence of such
ity
practices are found in Urdu literature also. For instance in
a social novel about elite Muslim life in nineteenth century
rs
north India, Shamsur Rahman Faruqi tells us that Nawab
ve

Shamsuddin named his son (who later became the famous


poet Dagh) after his dead younger brother Mohammad
ni

Ibraheem Ali Khan. However, he also added that the child


would be called ‘Nawab Mirza’ and so the poet is known
U

as Nawab Mirza Khan Dagh Dehlavi (Faruqi, 2006: 206).


d

When a child is expected, lists of names are drawn up


or

by the elders of both sides of the family. The parents too


xf

have a list and, increasingly in urban families at least, their


choice is respected. Our survey shows that the father gives
O

the name in 33.02 and the mother in 19.37 per cent cases.
If one adds to this the relatives on the father’s side then
the percentage goes up to 66.03 per cent while the same
ratio on the mother’s side is 30.16 per cent (Survey 2012
at Annexure-N). This is quite expected in Pakistan’s male
dominated society.
28 Names: A Study

A given name could have several components, though only


a few are used. Ugly, disgusting, ridiculous and negative
names are of course, rejected, though in rural areas
sometimes such names are still given to save the child from
the evil eye. Such names are mentioned by Temple who
points out that the name khairati (of charity) meant that

s
a child was given to a mendicant (faqir) and then begged

es
back in charity (Temple, 1883: 27). Some such names like,
ghasita (dragged), kala (black), khadera (dragged), are still

Pr
in evidence in rural society. To indicate a lower onomastic
level these pejorative names are all in Hindi-Urdu, not the
ity
more prestigious Arabic or Persian languages. Interestingly
enough, Shamim Saifullah Khan, former principal of
rs
Aitchison College in Lahore and a scion of the famous
ve

Khan Abdul Ghaffar Khan’s extended family of Charsadda,


told the present author that he was given a female name
ni

(Shamim meaning breeze in Persian) as he was the eldest


boy and had to be saved from the evil eye (Khan, S PC.,
U

2012). The use of such inauspicious names has reduced


d

significantly and now they are hardly heard of among the


or

younger people in all areas of the country. Most people


give ego boosting names so that 206 respondents (65.40
xf

per cent) like their names very much and 73 (23.17 per
O

cent) simply like them. This makes a total of 88.57 per


cent respondents who like their names. Indeed, only a mere
9 (2.86 per cent) dislike them, while 17 (5.40 per cent)
neither like nor dislike them and 10 (3.17 per cent) ‘don’t
know’ (Annexure-N).
Beliefs A bout Names 29

However, Islamic injunctions, in common with most


other societies, prohibit names which would expose the
child to ridicule or discrimination (Abu Dawud Book
41: No. 4930) and also ‘offensive nicknames’ (Ibid., Book
41: No. 4944). It is well-known ‘that the child who
bears a generally unpopular or unattractive name may

s
be handicapped in his social interactions with peers’ (Mc

es
David and Harari, 1966: 458). Social psychologists who
have studied this phenomenon compared 88 white males

Pr
with unique names out of 10,000 psychiatric cases and
compared them to a matched group with popular names
ity
and found ‘a significantly higher frequency of psychoses in
the peculiar name group’ (Hartman et. al., 1968: 109). It is
rs
also suggested that ‘the social desirability of the first name
ve

correlates significantly with individual popularity status’


(McDavid and Harari, 1966: 457). For these and other
ni

reasons (religious, cultural, historical, etc.) most societies


have restrictions on naming. For instance, an Icelandic girl
U

instituted a court case against the authorities to enable her


d

to use her name Blaer Bjarkdottir (meaning light breeze


or

and daughter of Bjark). The Icelandic Naming Committee


has approved 1,853 names but it considered this name
xf

unfeminine, hence the court case (Icelandic Names).


O

Giving a name to the child also has religious significance.


Whether done by the family or someone with a reputation
for piety, the meaning of the name is considered. It is
supposed to be auspicious and preferably at least one
component of the name should be from Arabic, Persian or,
to a lesser extent, Turkish. These languages are considered
30 Names: A Study

Islamic by the common people. In our survey of 315 people


the respondents were asked to give the meaning of their
names. Out of them 116 (36.83 per cent) gave no answer,
wrote ‘Don’t know’ or wrote a name which was illegible
and had to be excluded from the survey. Out of the 199
(63.17 per cent) responses which could be read 140 (70.35

s
per cent) were from Arabic; 43 (21.61 per cent) from

es
Persian; 07 (3.52 per cent) from Urdu/Hindi and only 02
(1.00 per cent) were from Turkish while those from other

Pr
languages were 07 (3.52 per cent). In short, Arabic is the
major language of naming in Pakistan. But Arabic is not
ity
understood except by the clergy and a few other people
in the country. Out of those 199 who wrote the meanings
rs
of their names, 124 (62.31 per cent) knew the correct
ve

meanings while 75 (37.69 per cent) did not (Annexure-N).


In short, if we take responses to all the 315 questionnaires
ni

into account it appears that people generally do not know


the meanings of their names. But considering that Arabic
U

is usually taught without giving the meanings of the


d

words (the emphasis is just on reading the script of the


or

Holy Quran without any translation), it is significant that


most of those who wrote the meanings of their names did
xf

know them. This indicates that there is a desire to know


O

the meaning of one’s name even if one does not know the
language it comes from.

In some cases more elaborate procedures of choosing names


are followed. For instance, letters are taken out of the
Quran by opening the book and taking the first letter of the
seventh line. To this the numerological value of the date of
Beliefs A bout Names 31

birth is sometimes added and a name is chosen. However,


a recent book on Islamic names for the Muslim diaspora in
the West condemns these practices as un-Islamic (Qasmi,
2007: 12). Moreover it prohibits all nicknames and names
like Qamar, Badar, and Gulab—all natural objects—unless
they have an Islamic component (Ibid.: 13). It especially

s
goes out to outlaw Parvez, because the Iranian monarch

es
of that name tore up the Prophet of Islam’s (pbuh) letter
when it was presented to him (Ibid.: 14). This view was

Pr
endorsed by the Deobandi, Ahl-i-Hadith and the Barelvi
ulema interviewed by the author (Ashrafia. Int., 2013 and
ity
2013 b; Ahmad. Int., 2013; Qadri. Int., 2013). The ulema,
emphasize the use of the prefix abd saying that it must not
rs
be left out when using a name of an attribute of God, but it
ve

is not permitted for any other name. Thus, while raheem and
rahman both mean ‘compassionate’ in Arabic, the former is
ni

a name for both the Prophet (pbuh) and God while the
latter is only for God. Thus, in the strict interpretation of
U

Islam, while the prefix abd can be omitted before Raheem it


d

cannot be omitted before Rahman which should always be


or

Abd ur Rahman (Slave of the Compassionate One).


xf

According to Annemarie Schimmel, and confirmed by our


own data, Pakistanis have a tendency to leave out the abd
O

completely and to invent forms like Haqqi from Abdul


Haqq, Abdi from ‘Abdullah, or Hamid from Abdul Hamid,
(Schimmel 1989: 28). This is one of those several cultural
practices the ulema desire to change. Yet another belief is
that certain names are ‘heavy’ (bhari) or inauspicious. Out
of our survey only 24 people (7.62 per cent), changed their
32 Names: A Study

names because they were inauspicious (Annexure-N). Upon


further probing, this belief turns out to consist of a set of
allied beliefs with subtle differences. One form is that the
name is inauspicious and if one changes it the evil effects
it was causing would be avoided. For instance, if one was
intermittently ill, or prone to accidents or misfortune,

s
one could change these ill effects of the ‘heavy’ name by

es
changing it. The other is that names are not responsible for
misfortunes of this kind as they are predestined. However,

Pr
they do have an effect upon one’s personality. In other
words people with names like ‘Sher’ (lion) will be brave
ity
and so on. Among the ulema interviewed for this study,
the Barelvi, Deobandi, and Ahl-i-Hadith ulema all agreed
rs
that names do have an effect on the personality. However,
ve

the Barelvi ulema said they advised change of name if the


meaning was bad. The example given by Maulana Irfan
ni

Qadri was Javed which means ‘one who lives for ever’
which, he said, was presumptuous since no human being
U

was immortal (Qadri. Int., 2013). Qari Fayyaz Ahmad, the


d

Ahl-i-Hadith alim said that names did have an effect upon


or

the personality but they should be changed only if they


do not belong to a Companion of the Prophet (Sahabi).
xf

For instance the name Umar could not be inauspicious


O

and should not be changed but less sacred names could


be (Ahmad. Int., 2013). In Jamia Ashrafia, the biggest
Deobandi madrasah in Lahore, the senior alim, who refused
to reveal his name, said all things come from God so names
cannot be inauspicious. While talking to the author he
received a phone call from someone whose daughter was
continually ill and who wanted to know if it would help if
Beliefs A bout Names 33

her name was changed. The Maulana told him it would not
help and, turning to the present author, pointed out that
people were superstitious because their belief in God was
weak (Ashrafia. Int., 2013). However, a junior alim, called
Shahid Obaid, gave an impassioned rebuttal of this point
of view when the author was about to leave, claiming that

s
names did have an effect upon the personality. Moreover,

es
he added, he personally knew a man whose name meant
‘boiling’ in Arabic and, sure enough, he had boils on his

Pr
body (Ashrafia. Int., 2013 b). Allama Tahir Abbas, the
Shia alim in Jamia Imamia, a Shia madrasah in Lahore, said
ity
categorically that names are not inauspicious and there was
no need to change them to avoid misfortune. However, he
rs
did not reply to the question whether they had an effect
ve

on the personality or not (Abbas. Int., 2013). Other Shia


ulema, however, did not necessarily agree with him asserting
ni

that the consensus of opinion was that names did affect


personality.
U

These beliefs are found among the ordinary people as well


d

as the educated middle classes. Even among the highly


or

educated professionals some of these beliefs are found


xf

especially in case the family experiences misfortunes.


Pakistani society is still in transition from a medieval
O

worldview to a modern one. Indeed, even among the urban


elite, purely rational beliefs about life—names being only
a part of it—are rare. Thus, by and large, the above beliefs
describe Pakistani society for the most part.
3
Names and Islamic Identity

s
es
One of the anecdotes about the communal riots of 1947
when British India was partitioned into Muslim-majority

Pr
Pakistan and Hindu-majority India is as follows:
ity
A man asked another his name. Upon being told he killed him.

Though this tale is not mentioned in literature published


rs
on the tragic riots, it could well be true. In the Punjab alone
ve

between 500,000 to 800,000 people—Hindus, Muslims,


and Sikhs—were killed (Ishtiaq Ahmed, 2012: xxv) and
ni

Saadat Hasan Manto’s creative writings in Urdu tell us that


U

very often there was nothing to save or condemn a person


except the name—which in such cases served as the carrier
d

of communal identity. Indeed, as a refugee interviewee told


or

Dr Ishtiaq Ahmed, ‘East Punjab had become totally unsafe


for anyone with a Muslim name. It was similar for us with
xf

Hindu names to leave West Punjab in 1947’ (Ibid.: 42).


O

The boundaries of the Islamic identity are drawn up by a


number of outward manifestations, signals, and symbols
which are characteristic of Muslim communities. The
foremost of these are names, or components of names,
drawn from the names of God, the Prophet of Islam
(pbuh) and others associated with the birth and spread of
Names and Islamic Identity 35

Islam in the 7th century. Names in the Quran, the Hadith


(sayings of the Prophet) and early Islamic history are also
categorized as Islamic.

Pakistan has a religious society and this is evidenced, among


other things, by the way names are chosen. Annemarie
Schimmel mentions in her book (Schimmel, 1989) that

s
people do not have adequate knowledge of Arabic or

es
theological niceties except for whatever local customs and

Pr
lack of knowledge of Arabic might bring about, there is a
general desire to give Islamic names to children in Pakistan.
And, since Arabic names are considered synonymous with
ity
Islamic ones, books on children’s names overwhelmingly
rs
comprise such names even if they are not actually used in
society. The following data indicates this trend.
ve

Table 1
ni

Figures for the languages of Names are percentages


U

Urdu-
Total Arabic Persian Turkish Other Source
Hindi
d

Ahmad &
3870 75.89 16.43 0.47 4.78 2.43
or

Ahmad 2008
4592 68.00 20.00 3.00 7.00 2.00 Mohsin 2010
xf

3968 70.00 25.00 2.00 0.60 2.40 Sheikh n.d.


O

2848 71.63 23.28 1.33 1.83 1.93 Baloch n.d


6481 72.38 20.00 2.60 1.32 3.70 Raza 2003
Total of names
and mean of
21759 71.58 20.94 1.88 3.11 2.49
percentages of
occurrence.
36 Names: A Study

The books from which the above figures have been


compiled also have names which are not used in Pakistan
in the opinion of the author. These have been excluded
from the above table. The column under ‘other’ includes
Old Persian, Hebrew, Sanskrit, Pashto, Sindhi, Balochi, and
Brahvi as well as English and other European languages.

s
As we can see, the writers of these books of childrens’

es
names feel that names from Arabic are valorized and
preferred to those from other languages. Most of them

Pr
begin with Islamic injunctions on naming and some also
give the ninety-nine names of God and the Prophet of
ity
Islam (pbuh). The common identity marker of the Muslim
identity in Pakistan is the name of Muhammad (pbuh).
rs
This is probably because the Prophet Muhammad (pbuh)
ve

advised Muslims to keep his name (Bukhari c. 9th C


428–429) and the names of the prophets mentioned in
ni

the Quran. The Prophet (pbuh) also said that ‘the names
dearest to Allah are Abdullah and Abdurrahman’ (Abu
U

Dawud, Book 41, No. 4932). And, indeed, traditional


d

naming does follow these guidelines.


or

Conversion to a religion generally entails a change of


xf

names. Indeed, Richard W. Bulliet used the occurrence


of Islamic names in Iran, Iraq, Egypt, Tunisia, Syria, and
O

Spain to measure the process of conversion to Islam in


these areas (Bulliet, 1979). Conversion to Christianity in
early Byzantine Egypt too has been measured by looking
at names (Bagnall, 1982). In South Asia too such studies
have been attempted. For instance, Nur Mohammad, in his
Tarikh-i-Jhang Sial (The history of Jhang Sial) (1862) says:
Names and Islamic Identity 37

In the early fifteenth century 10 per cent of the recorded Sial


males had Muslim names; for the mid-seventeenth century,
56 per cent; for the mid-eighteenth century, 75 per cent; and
for the early nineteenth century, 100 per cent (quoted from
Eaton, 2004: 113).

So important is the giving of Islamic names as boundary

s
markers of the Islamic community that Maulana Ilyas

es
(1885–1944), the founder of the Tablighi Jamaat (a
movement for the preaching of Islam), considered the Meos

Pr
of Mewat (Alwar, Bharatpur, and Gurgaon in Haryana
in India) with their Hindu appearance and names in the
ity
1920s, an affront to his Muslim sensibilities, although many
of them had converted to Islam. About their names the
rs

gazette of 1878 declares:


ve

They call themselves by Hindu names, with the exception of


ni

‘Ram’; and ‘Singh’ is a frequent affix, though not so common


as ‘Khan’ (Captain Powlett, 1878: 38. Quoted from Ibbetson
U

1883: 135).
d

As the Islamization of the Meos progressed, the components


or

of names shared with the Hindus decreased and then


xf

disappeared (Aggarwal, 1976: 268). However, the difference


in the religious characterisrics of names remained one cause
O

of communal animosity in colonial India, as conversion


from Hinduism to Islam and vice versa illustrate. Examples
of the former are too numerous to be quoted. Among those
of the latter are the bestowal of the name Alakdhari on a
Muslim boy by Dayananda Saraswati of the Arya Samaj in
1877 (Sikand and Katju, 1994: 2215). After the Partition of
38 Names: A Study

1947 some Rajputs converted to Hinduism in Udaipur and


the village heads instructed teachers ‘to change the names
of their Muslim students to Hindu in the official school
records’ (Ibid.: 2217).

This anxiety about giving Muslim names in order to


maintain identity is illustrated in many situations. For

s
instance, when there was interaction between the Indian

es
Muslims and the British (Christians) during the beginning

Pr
of colonial rule, some Muslim women became either the
wives or concubines of Englishmen. But despite becoming
‘White Mughals’ in the sense of having adopted some
ity
cultural and sartorial values of the Muslim elite, these
rs
British men gave English names to their children from
these mixed marriages. Often there were two sets of names,
ve

Indian (Muslim or Hindu) and British (Christian) as in


Shamsur Rahman Faruqi’s Urdu novel on this period. In
ni

this the heroine, Wazeer Khanum, has two children with


U

Marston whose naming is described as follows:


d

The name Badshah Begum she had kept only after a quarrel
or

with Marston Blake otherwise he was prepared to hear no other


name than the Christian one: Sofia. After a lot of argument
xf

he suggested ‘Masih Jan’ as the Indian name but the baptism


O

was to be only with the name Sofia (Faruqi, 2006: 188).

Eventually the Muslim Maulvi officially gives the name


Badshah Begum to the child while the Christian priest
gives her the Christian name. This Muslim name was
important for Wazeer because, as she tells her dead
husband’s relatives, with these names the children would
Names and Islamic Identity 39

have the right to choose their religion; without the Muslim


names they would be only Christians (Faruqi, 2006:
215). Dalrymple also mentions that Colonel Dalrymple,
commander of the British troops in Hyderabad, married
Mooti Begum, a nawab’s daughter, and their children were
named according to gender. The boys were given Christian

s
names and brought up like the English while the only girl

es
from the marriage was called Noor Jahan Begum and she
was brought up as a Hyderabadi Muslim aristocratic young

Pr
lady (Dalrymple, 2002: 119–20). In short, Muslim identity
was inextricably linked up with Muslim names.
ity
Islamic Components in Names
rs

The first attempt at finding the frequency of Islamic names


ve

among South Asian Muslims is by R. C. Temple who writes:


ni

An examination of the Muhammadan names in the Census


tables shows that about half, or 48 per cent, of them are
U

religious, i.e., of Arabic or foreign origins, and that the


d

remainder, or 52 per cent, differ in no way from those of


Hindus (Temple 1883: 46).
or
xf

But Temple’s definition—‘Arabic or foreign origins’—is


problematic since, firstly, it is not the language (Arabic,
O

Persian, or Turkish) which can be called quintessentially


Islamic. And, secondly, it is not clear whether the percentage
Temple gives is of names or the Islamic components
themselves. Since most names have two components, the
two kinds of percentages would differ widely.
40 Names: A Study

For this study Islamic components are categorized as being


classical and indigenous as defined in the section on meth­
odology (Chapter 1). It is with reference to that section that
the frequency of occurrence of these components have been
calculated in different samples and presented in annexures
C & D and also in F & G.

s
es
The Islamization of Names in Pakistan
The resurgence of high Islam as well as other religions, such

Pr
as Christianity and Hinduism, suggests that the classical
claim of scholars that modernity entailed secularization
ity
as it did in Western societies, needs rethinking (Hefner,
1998). Ernest Gellner argues that in Islam ‘modernization
rs

on the one hand, and the reaffirmation of a putative old


ve

local identity on the other, can be done in one and the


same language and set of symbols’ (Gellner, 1981: 5). In
ni

a sense then, it is the totalizing project of modernity, and


U

the fact that it is armed with a claim to monopoly of


knowledge and rationality, that provokes Muslim theorists
d

to fall back upon a claim to divine wisdom. In short, the


or

‘difference’ evoked by modernity can be seen as a survival


xf

tool of Muslim societies against ideological conquest by


‘the West’. Modernity did bring in a certain uniformity
O

and standardization which, C. A. Bayly in his history of


modernity suggests, led to a ‘standardized form of Islamic
practice’ including the spread of certain names (Bayly, 2004:
18). These trends are observable among Pakistani naming
processes as we shall see below. This, it is suggested, is an
aspect of onomastic modernization in Pakistan.
Names and Islamic Identity 41

Pakistani names have become more Islamized now than


they were in the initial years of Pakistan. This is suggested
by the fact that more people have Islamic components in
their names taken in the 1990s than the ones in the data for
the 1950s. The increases are slight in terms of percentages
in all the provinces and in male names and they are more

s
pronounced in female names. However, in the case of male

es
names in Sindh the percentage of the Islamic components
in the 1950s is higher. This anomaly is explained by two

Pr
facts. First, the data for male names is much smaller than in
other provinces. Secondly, the components from folk Islam
ity
are more than any other province (Annexure F). This brings
us to the crucial concept of what is an Islamic name? This
rs
seems to have changed. It used to include names referring
ve

to the belief that sons are given by beings other than God
such as saints or Prophet Muhammad (pbuh). Such names,
ni

as well as names with non-Arabic words for God such as


the Persian word ‘Khuda’ and the Sindhi words ‘Waddo’ (big)
U

and ‘Bhallo’ (good) were common in the 1950s but have


d

decreased now as some of these are seen as being heretical.


or

This question of what changes have occurred in the


perception of an Islamic name will be visited in some more
xf

detail below. Let us now consider the processes at work in


O

bringing about the changes. There seem to be two processes


at work here. The first is increased emphasis on high Islam
as mentioned above. And the second is urbanization. Both
work in tandem, so it is not always possible to pinpoint
precisely why a name has been changed. The quintessentially
folk-Islamic names—Peeran Ditta (given by saints) or
Rasool Baksh (gift of the Prophet)—are also seen as rural
42 Names: A Study

and old fashioned, so it may be because of the new fashion


rather than high Islam that the name has been changed.
Other old fashioned or rural religious names for women
are: Muhammadi, Mahmoodan, Namazan (one who prays),
Jannat (paradise), etc. But these names are being phased
out as people get urbanized and literate. This is more of

s
a process of ashraf ization—the seeking of gentlemanly

es
status among Muslims, which is the equivalent of the status
enhancing Sanskritization (aping the behaviour of upper

Pr
castes) among Hindus which Srinivas mentions (Srinivas,
1952: 32)—than Islamization. Hence it is not clear that the
ity
higher percentage of Islamic components in names both
in the data for KPK and the Punjab from the early years
rs
(1950s) and the post Zia ul-Haq (1977–1988) era (1990s)
ve

necessarily represents only an increase in the Islamization


of Pakistani society (Annexures F & G). However, there
ni

are certain other indicators in naming practices which do.


For instance, the percentage of folk Islamic components in
U

names is an indicator of both possibilities: a shift towards


d

high Islam in naming or urbanization (even both). Let us


or

take the decrease of folk components in names first.


xf
O
Names and Islamic Identity 43

Table 2
Decrease in components of
folk Islam in Male Names in percentages
1950 1990
Punjab 1.99 0.29
KP 1.34 1.18

s
Sindh 2.23 0.49

es
Elite 1.34 0.35

Pr
Source: Annexure F. Calculated by taking the percentage of the Islamic folk
components out of the total number of components in names.
ity
As we can see, the percentage of folk components was
highest in Sindhi names. Hence, the decrease of Islamic
rs
components in names over time probably represents
ve

unrbanization and modernization in Sindh rather than lack


of Islamization.
ni

The most popular of the Islamic names in Pakistan as


U

all our samples suggest, are the names of the Prophet


Muhammad (pbuh) and his family (the Ahl-i-Bayt). This
d

family includes Ali, the cousin of the Prophet (pbuh) and


or

his son-in-law; his daughter Fatima and their sons Hassan


xf

and Hussain. Despite the fact that most Pakistanis are


O

Sunnis, the names of Ali, Hussain, and Hassan, though also


shared with the Shias, are among the most popular names
among Sunnis. They are also found on vehicles, roadside
cafes, religious buildings (madrasahs, shrines, etc.) as well as
on trucks as Jamal Elias, who has written a book on trucks
in Pakistan, has noted (Elias, 2011: 133). The exhortation
to Ali for help in Urdu, ‘ya Ali madad’ (Oh Ali Help!) is
44 Names: A Study

found on the lips of the people and on vehicles as well as


other places. In all except one of our samples the name
Muhammad (pbuh) occurs most frequently followed by
Ali. Details of the frequencies of the occurrences of these
names are given later and they will be discussed in chapter
5 which is on first names (Annexures-L and M).

s
Let us now revisit the concept of the Islamic name. Pakistani

es
society is traditionally religious but the interpretation of

Pr
Islam in practice in most of the country, especially in the
rural areas of the Punjab, is Barelvi, i.e., intercessory, folk or,
in Gellner’s phrase, ‘low church’. Hence the traditional use
ity
of Islamic components in names symbolizes a traditional
rs
rather than an aggressively fundamentalist or radical Islamic
identity. In addition to these names, older names, shared
ve

by Hindus and Sikhs, are still found in the rural areas.


Names referring to natural objects have been mentioned in
ni

Chapter 5. Here I will refer to religious names of the local


U

or indigenous kind. These refer to the child being a gift of


God, Prophet Muhammad (pbuh), and saints, especially
d

Sheikh Abdul Qadir, the Ghaus-ul-Azam:


or

Khuda Baksh (Given by God)


xf

Allah Baksh (Given by God)


O

Rasool Baksh (Given by the Prophet)


Nabi Baksh (Given by the Prophet)
Allah Ditta (Given by God)
Allah Dad (Given by God)
Allah Wasaya (Blessed by God)
Allah Rakha (Preserved by God)
Peeran Ditta (Given by the saints)
Names and Islamic Identity 45

Ghaus Baksh (Given by Sheikh Abdul Qadir)


Peer Baksh (Given by the saint)

In Sindh the names Nale Chango (the name of the Good


One), Nale Mittho (the name of the Sweet One), and Nale
Waddo (the name of the Great One) all refer to the Deity
and are used as first names of a religious kind (Somroo. Int.,

s
2013). A number of interviewees told the present author

es
that they had changed such names under the instructions

Pr
of their strictly religious madrasah teachers. For instance,
Abdul Akbar, an electrician working in Rawalpindi, had
the name Mohammad Akbar. He changed it when studying
ity
in a Deobandi madrasah because he became aware that
rs
it meant ‘Mohammad is Great’ whereas only God could
be great (Akbar. Int., 2012). This strict attitude towards
ve

names is part of the Saudi, Wahabi influence on Pakistani


culture. Saudi Arabia has recently banned 50 names, which
ni

according to the interior ministry of that country, ‘contradict


U

the culture or religion of the Kingdom’. Among these


names are: Abdul Nabi, Abdul Rasool (both meaning slave
d

of the Messenger). Western names or non-Muslim names


or

are obviously part of the ban. Such names include: Sandy,


xf

Rama, Maline, Elaine, Maya, Linda, Alice, and Lareen,


among others. What will come as a surprise for many
O

Pakistanis is that the names they had chosen for being new,
unique and yet Arabic (and hence Islamic) during the last
few years—Basmala (with the name of God), Iman (faith),
Nabiyya (female messenger or prophetess)—are also on the
banned list (Saudi Arabia, 2014).4 This ban is not known
in Pakistan but the theological reasons behind it—strict
46 Names: A Study

monotheism and the elimination of non-Muslim cultural


influences—have been affecting Pakistan for the last two
decades. It has affected the construction and perception of
the Islamic name. It has not only affected the rural folk
Islamic names which attribute agency to beings other than
God but also the urban perception of equating Arabic and

s
Islamic concepts and revered places as being Islamic. Thus,

es
both the village names Nabi Ditta (Prophet-given) and the
new city names Azaan (call to prayers), Iman (faith), Dua

Pr
(prayer), etc., are no longer seen as Islamic by the strict
interpreters of Islam. To conclude, the newly emergent
ity
purist attitude towards Islamic naming may be affecting
self-consciously strict Muslims in urban Pakistan though
rs
rural areas and the Westernized elite still has many such
ve

names.

Let us put this in the historical perspective. It is true that


ni

Pakistani society has been getting Islamized from even


U

earlier than Zia ul-Haq’s rule, but his nearly eleven years
of rule brought about the Islamization of the legal system
d

and increased the use of the Islamic narrative in the


or

country. International factors, such as the use of Islamic


xf

militants in America’s proxy war against the Soviet Union


(1979–1987) and later by Pakistan in the low-intensity
O

conflict about Kashmir with India (1990 onwards) also


increased the street power of Islamist pressure groups in
Pakistan (Mir, 2008). The shift of Pakistani society towards
the hardline and fundamentalist Islam is evidenced by
the personal appearance of young Pakistanis and their
increasing intolerance towards the minorities (Rahman,
Names and Islamic Identity 47

2012). In naming this is evidenced by the use of certain


new components of names such as Saddam and Osama as
mentioned before. This is an assertion of an anti-Western,
radical Islamic identity which is reflective of the narrative
of militant Islam. In 1999, an Urdu-language journalist Jalil
reported that the name Osama was being given to boys and

s
businesses in the KPK province ( Jalil, 1999). In our data

es
the names Saddam and Osama occur very infrequently in
the 1950s but are much more frequent in the names of the

Pr
1990s (Annexure-J). Morever, Saddam occupies the sixth
position in the most popular first names in Sindh among
ity
males named during the late 1990s (Annexure M-2). These
seem to be the result of a wave of sentiment in favour of
rs
Saddam Hussain and Osama bin Laden who were seen as
ve

heroes who had dared to stand up to the ‘West’. The other


aspect of Islamic radicalization in Pakistan is the rise of
ni

sectarian sentiment in Pakistan. This has an onomastic


aspect since Shia components of names are based on the
U

names of the imams of the majority sect of the Pakistani


d

Shias i.e., the twelvers (athana ashari). Ironically, however,


or

according to a scholar of Shia Islam, Dr Syed Hussain Jafri,


the components considered exclusively Shia are, in fact, also
xf

shared by the Sunnis ( Jafri. Int., 2013).


O

The battle between the two sects, in which the Shias


are generally the victims being in a minority, has been
going on since the eighties. The Shias initially wanted
to be exempted from the laws of alms collection (zakat)
implemented by Zia ul-Haq. However, eventually their
extremist organization, the Sipah-i-Mohammad clashed
48 Names: A Study

with the extremist Sunni organization, the Sipah-i-Sahaba,


that was formed in Jhang in 1985 specifically to combat
Shia assertiveness and create a Sunni state (Zaman, 2007:
119–139). The result is a sectarian war in which the Shias
have suffered the most (Ahmed, 2011). Names indicating an
anti-Shia identity are Muawiya and Yazid. In our samples

s
from the early years of Pakistan the names Muawiya and

es
Yazid do not occur at all, though in the names of the 1990s
there are sixteen (16) occurrences of Muawiya in the BISP

Pr
sample (Poor-M) and even in the middle classes there are
some people with this name (Annexure-J). The name Yazid,
ity
which is a metaphor for tyranny in Pakistan, seems to be an
instance of self-naming to proclaim an anti-Shia identity
rs
since this name was never used by South Asian Muslims.
ve

An incident, and reportedly not the only one of its kind,


about singling out Shias by their names and then killing
ni

them, was reported from Kohistan in the north of Pakistan


U

in July 2012. Reportedly a bus was stopped on its way from


Rawalpindi to Gilgit and the passengers were asked their
d

names. Those who had Shia components in their names—


or

even though some of these components such as Ali,


xf

Hassan, and Hussain are shared by Sunnis as mentioned


before—were off-loaded and killed in cold blood. Another
O

similar incident was reported by Dr Najma Najam, Vice


Chancellor of the Karakoram University in Gilgit, to the
present author when she reported that one of her own
faculty members who had a Shia name was stopped by
the security agencies, reportedly for his own safety after
this massacre. The danger such names pose is poignantly
Names and Islamic Identity 49

mentioned in the following passage by a Sunni Baloch


journalist part of whose name was Hussain. He says:

I should not be simply murdered for my parents’ short-


sightedness for not foreseeing that 30 years later this name
could get their child killed (Hussain, 2013: 6).

s
None of the Shias the author interviewed or talked to

es
said they had considered changing their names but Dr
Jafri did mention that the community was much disturbed

Pr
about names which, though shared with the Sunnis, were
widely perceived to be distinctively Shia and so potentially
ity
dangerous ( Jafri. Int., 2013). The Shias do have the doctrine
of taqiyya (concealment) which may be used when one is
rs
being persecuted but whether it is used or not is not clear.
ve

About the use of the names Muawiya and Yazid, which


ni

hurt Shia sensibilities so much, the ulema expressed the


following views. The Deobandi ulema said:
U

Hazrat Ameer Muawiya was a sahabi and it is blasphemous


d

to disrespect his name. And even Yazid defied the Christians.


or

There is nothing wrong with his name (Ashrafia. Int., 2013).


xf

The Barelvi and the Ahl-i-Hadith ulema conceded that


O

Yazid’s name was a metaphor for oppression, but added


that ‘the name of Muawiya is quite justified’ (Qadri. Int.,
2013; Ahmad. Int., 2013). The Shia Alim, Allama Abbbas,
said that such names were anathema in his subsect (maslak)
(Abbas. Int., 2013). As mentioned earlier, such names were
not used in South Asia in recent history and are now used
by anti-Shia Islamic militants. Asmatullah Muawiya, now
50 Names: A Study

a leader of the Taliban, was a militant from an extremist


anti-Shia organization called the Lashkar-e-Jhangvi.
Another person bearing this name, Shams ur Rahman
Muawiya, is a leader of the Ahl-e-Sunnat wal Jama’ (A
Sunni movement, which is also anti-Shia). Further, talking
about the abduction of his son in May 2013 Yusaf Raza

s
Gilani, former prime minister of Pakistan, said that ‘one

es
Abu Yazeed claimed his group had abducted Ali Haider
Gilani’ (Shah, 2013: 45). In all three cases these seem to

Pr
be instances of self-naming.

Another case of self-naming is the adoption of the name


ity
‘Khurasani’ by a number of Taliban militants including the
rs
Taliban commander, Mullah Fazlullah, recently. Khurasan
refers to that part of the early Arab empire which comprised
ve

Central Asia and parts of present-day Pakistan. Those


who take up the name are signaling a lost identity; an
ni

identity which is being reclaimed by the process of naming.


U

And one implication of this identity is the aspiration of


reestablishing the lost Muslim caliphate in this part of the
d

world. That is why, as Rahimullah Yusufzai points out, such


or

names have been adopted by a number of commanders such


xf

as Omar Khalid Khurasani of the Mohmand Agency and


Muhammad Khurasani whose real name was Khalid Balti,
O

etc. (Yusufzai, 2014). Earlier, being inspired by Abu Musab


al Zarqawi (1966–2006), an Al-Qaeda anti-Shia leader,
Pakistani militants added Zarqawi to their names. In some
cases they also added Ayubi being inspired by Salahuddin
Ayubi (1137–1193), a famous Muslim crusader (Yusufzai,
2014). These are classic examples of names being used to
Names and Islamic Identity 51

construct a certain self—an uncompromisingly Sunni, anti-


Shia, radical Islamic self—as part of a ‘technology of the
self ’ (a la Foucault).

The indexation of the Islamic identity with names has other


potentially dangerous consequences in Pakistan. While Shia
names may be identifiable, Ahmedi names are exactly the

s
same as the majority Sunni community. This is a constant

es
complaint of the anti-Ahmedi Sunni ulema who contend

Pr
that the Ahmedis pretend to be Muslims and deceive
people by their names. The pressure of the ulema resulted
in riots against the Ahmedis in 1954 and again in 1974.
ity
Zulfikar Ali Bhutto’s PPP government thereupon conceded
rs
the major demand of the ulema by declaring the Ahmedis
a non-Muslim minority. Later General Zia ul-Haq added
ve

Section 245-C to the Pakistan Penal Code whereby the


Ahmedis were forbidden to call their place of worship
ni

a mosque and to observe the outward symbols of Islam.


U

However, the names remain the same and this constitutes


a threat to the ulema who would like the boundarymarkers
d

of their community to be drawn even more strictly than


or

at present. Incidentally this ambiguity of the indexation


xf

of religious identity with names is presented in Hasan


Manzar’s novel in which Ahmad Baksh, the protagonist,
O

meets a girl called Zainab who tells him that they belong to
different religions. He asks his elder brother’s wife whether
the name Zainab occurs in any other religion except his
own (Sunni Islam) (Manzar, 2008: 87–89). She cannot
answer him but it turns out that Zainab’s family is Agha
Khani—a sect which is not accepted as being Muslim by
52 Names: A Study

Sunnis. In short, naming is a serious matter in Pakistan as


sect has become a potentially explosive subject claiming
human lives on an almost daily basis.

To sum up, this analysis of Pakistani personal names has


brought out that there has been an increase in Islamic
naming in the middle classes in the country since the

s
early years of Pakistan’s birth. This has involved not only

es
the use of traditional Islamic components from classical

Pr
or ‘high’ Islam, but also the revival of Arabic names from
the early period of Islam. Among such naming practices
is the use of quintessentially Arabic components of names
ity
such as –ibn (son of ) and –umme (mother of ). Such names
rs
have not been in use in Pakistan earlier, but the quest for
an authentic Islamic (Arabic) identity has brought them
ve

in circulation. Groups associated with radical or militant


Islam (jihadis) use Arabic norms of naming to assert their
ni

newfound Islamic identity. In the Lashkar-e-Tayyaba, a


U

militant group, the Arabic Kuniya (Abu, Umme), are used to


‘ensure secrecy’ (Mir, 2009: 58). Names such as Abu Hafsa
d

and Abu Nisar, not known in Pakistan earlier, are disclosed


or

to the press by activists of Jihadi groups (Rana, 2005: 67).


xf

Even the names of these groups and their publications


are in Arabic which advertises their relationship to Arab
O

culture, al-Qaeda, radical Islam, and Jihadi thinking. The


desire to express solidarity with fundamentalist and radical
Muslims, probably as a measure of resistance against the
‘West’, which is seen as hegemonic and unified, takes the
form of using these Arab-sounding names or the names of
two anti-Western Arab figures: Osama and Saddam. Yet
another measure of increasing religiosity is the cognizance
Names and Islamic Identity 53

of sectarian identities and names like Muawiya and even


Yazid, the latter a term of abuse among the Muslims of
South Asia, are now found in the names of the 1990s,
though they were almost missing in the names of the
1940s–50s. All these are naming strategies to assert and
privilege a radical or hardline Islamic identity and perhaps

s
an expression of the rising power of Islamist sentiments

es
in Pakistan. These are self-conscious and deliberate
attempts of a person to index his/her name; to exhibit the

Pr
most obvious badge that the person wants to the world,
compatable with what is perceived as the Islamic identity
ity
in Pakistan. More evidence of the Islamization of names is
in Chapter 5 which is about the most popular first names
rs
of Pakistani Muslims.
ve
ni
U
d
or
xf
O
4
Names, Power, and Class

s
es
As names have cultural capital (Lord, 2002), a name may
exalt or debase a person. Thus the giving of a name or a

Pr
nickname is an act of power. In Plato’s book Cratylus, as
Socrates points out, slaves cannot name themselves. And,
ity
for that matter, neither can children though for social and
emotional reasons they are generally given names that
rs
parents do not consider harmful or degrading. Barbara
ve

Bodenhorn and Gabriele vom Bruck, agreeing with this


assertion ‘conclude that ‘naming is about agency, about
ni

feelings of being the peoper person, and crucially about


U

mutuality’ (Bodenhorn and Bruck, 2006: 27). Alternative


names—by-names, nicknames, sobriquets, etc.—whether
d

chosen by the individuals or others are very common


or

(in 45 out of 60 societies) (Alford, 1988: 70). However,


power plays a very important role in this kind of naming.
xf

Successful leaders, writers, cardinals and poets might


O

be in a position to assume positive alternative names.


Ordinary people, however, are given alternative names (and
sometimes renamed) by others. That, after all, is the reason
why in 50 per cent societies, nicknames are depreciatory
(Morgan et.al., 1979: 5), In 45 per cent they are based on
physical abnormalities and in 48 per cent they are based
on behavioral abnormalities (Alford, 1988: 84). Women
NA MES, POW ER, and CL ASS 55

are given names, even positive ones, by their husbands—as


in the Tiv, Azando, and Masai tribes (Alford, 1988: 88)—,
also as in some families among the Muslim gentry (ashraf)
in north India where a new bride is given a ‘title’ (khitab)
by her mother-in-law. For Instance, in Shamsur Rahman
Faruqi’s novel mentioned earlier, Nawab Shamsuddin’s

s
sister Jahangira gets the title of ‘Jahangir Dulhan’ (the

es
world-subjugating bride). But this title is very similar to
the original name which usually was not actually the case.

Pr
In the same novel Wazeer herself gets the title of ‘Shaukat
Mahal’ (The Magnificence of the palace) by the Mughal
ity
King Bahadur Shah Zafar when she is married to his son
Mirza Fakhru (Faruqi, 2006: 767). Similarly, girls forced
rs
into prostitution were given new names by the powerful
ve

administrators (madams) of the house. Thus, in Mirza


Rusva’s Urdu novel Umrao Jan Ada the abducted girl-child
ni

Ameeran is given the more sophisticated form of the name


Umrao by Khanum Sahib (Ruswa, 1899: 38).
U

Powerless groups named by others have to settle for


d

labels—often short ones—for whatever names are conferred


or

upon them. Susan Benson, for instance, mentions the


xf

‘injurious’ naming practices of masters towards slaves


(Benson, 2006). In Jamaica indigenous people were given
O

‘day names’ (e.g., Quashie, meaning born on a Sunday)


which were considered pejorative (DeCamp, 1967). Even
when they were given European names ‘they were usually
in the diminutive form’ (Burnard, 2001: 334–5; also see
Puckett, 1938: 36 and 38). Burnard, quoting Croton, says
that ‘some names were obviously intended to demean.
56 Names: A Study

Croton discovered slaves on Worthy Park Estate called


Monkey, Villain, and Strumpet’ (Quoted from Burnard,
2001: 336). Indians settled in South Africa, responding
to their lack of power as individuals, adopted Anglicized
first names for themselves ‘in work places to pre-empt the
often humiliating nicknaming by white superiors of staff ’

s
(Hansen, 2006: 216). In South Asia too names like Ghasita,

es
Kala, Lallu, Buddhu, Chattan, Banaras, Ajab, etc. were
common in the rural poor (for more names of this kind

Pr
see Chapter 5). In some cases, despite the Islamic belief
in egalitarianism, working class people were forbidden to
ity
take the names of prominent local dignitaries (Nadvi, n.d:
45). Even the name Ali Mohammad, because of its Arabic
rs
meaning of ‘exalted and praised’, considered appropriate for
ve

the males of the gentry, was shortened to Mammad for a


servant. Such shortening also took place in Hindu names
ni

(for Marhatta names see Sinclair, 1889: 161; for Punjabis,


see Temple, 1883: 32). Indeed, simple names were the norm
U

in the Indo-European languages too although upperclass


d

names were compounds (Pulgram, 1947: 206).


or

In this context folklore and literature provide examples.


xf

Temple quotes the story of a poor boy called Parsu who


grew up to be called Parsa and finally, when he got some
O

money, Paras Ram in the following couplet:

Is daulat men tin nam


Parsu, Parsa, Paras Ram
(With respect to wealth you have three names
Parsu, Parsa, Paras Ram)
(Temple, 1883: 32)5
NA MES, POW ER, and CL ASS 57

So common was the use of abbreviated names that some


British officers did not believe that the real names of people
in inferior social positions could be longer. For instance
Richard Temple says:

Prisoner Ali Nawaz Khan of the police report is the ‘Alia


of the evidence, and that the witnesses Govardhan Das and

s
Durga Parkash are known as Gobra and Durga to their

es
friends, and I would remark that ‘Alia, Gobra and Durga are
the real names of these worthies, the grander ones being used

Pr
merely for the occasion. … (Temple, 1883: 15)
ity
In fact, the longer names were probably authentic but
were considered too prestigious to be used by the common
rs
people. They were, of course, used for respectable people.
ve

This phenomenon is also found elsewhere. For instance, the


Dou Mbojo of Indonesia also abbreviate their full Arabic
ni

names, which are used in official documents, to shorter


forms, e.g., Halimah becomes Lima and Abdur becomes
U

Dura ( Just, 1987: 314).


d

In Urdu fiction low-status people are never called by full


or

names. For instance in Ismat Chughtai’s story ‘Kaf ir’


xf

(infidel) when the Nawab calls her son by his full name
as the boy’s father did in formal writing, the mother is
O

touched:

‘Kaleem Uddin’—tears welled up in the eyes of Kullu’s mother


… only his father would have written Kaleem Uddin like this
(Author’s translation from Urdu) (Chughtai, 2008: 584).
58 Names: A Study

Sometimes poor people are addressed with reference to


other people so that their real names are virtually lost.
Thus in Ismat Chughtai’s story ‘Nanhi ki Nani’ the nani
(maternal grandmother) was never called by her own
name but was referred to as the daughter-in-law, mother,
or grandmother of the other members of the family

s
(Chughtai, 2008: 715). Names from the Punjab in 1947

es
follow the same pattern. Ordinary people go by one name
or nickname—Dullah (from Abdullah), Bhala (Mehraj

Pr
Din), Kaada (Mian Nuruddin), and so on (Ahmed, 2012:
241, 142, and 32) while respectable people have longer
ity
names like Mujahid al-Hussaini who said his surname was
‘based on my devotion to the ideas of Maulana Hussain
rs
Ahmed Madani’ (Ibid.: 512).
ve

Onomastic Politics
ni

As mentioned earlier, Pakistani names, like those of the


U

Tamils (Britto, 1986: 349) and the Hindus of north India


(Vatuk, 1969: 256), do not have a fixed family name and
d

even the ‘caste name’ may be optional. Thus fathers and


or

sons and brothers may not share the same last name, as
xf

will be illustrated in the present author’s case in Chapter 6.


Moreover, prestigious components in names are correlated
O

more with power than belonging to the same clan or family.


Thus, the dominant classes of Pakistan exercise power
through what may be called ‘onomastic politics’, i.e., the
use of prestigious components in names in order to assert
their high status and so their right to be natural rulers
(Bodenhorn and Bruck, 2006: 14). Higher status calls for
NA MES, POW ER, and CL ASS 59

more sonorous naming and high caste names are invoked.


In this context it is pertinent to mention that scholars
disagree on the use of the term ‘caste’ [used interchangeably
with ‘qoum’ (nation), ‘zat ’ (type), and ‘biradari ’ (fraternity)]
among Muslim communities in India. Sir Denzil Ibbetson,
who compiled lists of castes in the Punjab, remarked that

s
‘nothing can be more variable and more difficult to define

es
than caste’ (Ibbetson, 1883: 2; for a detailed glossary
see Ibbetson et. al. 1911). Hamza Alavi mentions that

Pr
‘all biraderis in the endogamous system have zat names.
Such names are used locally, as surnames, to identify the
ity
biraderis’. He goes on to argue that the existence of zat
names does not necessarily signify the existence of caste in
rs
contemporary society (Alavi, 1972: 26).
ve

Imtiaz Ahmad, while doing research in U. P., equates zats


with castes [Julaha (weaver), teli (provider of oil), etc.] saying
ni

‘caste and kinship co-exist and overlap’ (Ahmad, 1976: 342).


U

But caste is not to be understood as something primordially


designated, nor does it have the support of religion as it
d

has in the Hindu varana system. Chaudhary (1999: 10–14)


or

provides definitions of terms used in discussions of kinship


xf

units in Pakistan such as qoum, biradari, and sharika (used


in the Punjab) and opines that qoum and biradari are useful
O

concepts for understanding the kinship units in a village.


And the anthropologist Stephen M. Lyon, studying power
relations in such a village for his doctoral dissertation,
finds that ‘categories such as zat and qaum, both of which
may be used to indicate caste terms, are important’ (Lyon,
2002: 125). Perhaps it is accurate to say that South Asian
60 Names: A Study

Muslims ‘have ascribed status based on certain conceptions


of lineage that correspond to Hindu notions of caste’ (Ali,
2002: 602). But, while surnames can bind Syeds, Pathans,
Sheikhs, etc., they are often too loosely scattered to do
so except in an imagined community. However, following
Baarth (1969), ethnicity is best seen as ‘socially constructed’

s
so this act of imagination may be a crucial determiner of

es
self-concept and how others view one.

Pr
The boundaries of this ‘caste’ are porous. There is no certain
way of determining if someone is authentically a Syed or a
Pathan. And, of course, the trend of ashrafization increases
ity
the number of people who take up surnames proclaiming
rs
prestigious descent. Thus lower caste Muslims claim
they are Sheikhs (Goodfriend, 1983: 123) and the ‘tantis
ve

(weavers) of Bihar began to call themselves ansaris after


becoming Muslims’ (Sikand and Katju, 1994: 2215). Thus
ni

the 1901 census of the Bengal recorded many ‘Sheikhs’ who


U

claimed Arab descent. However, the Bengal was conquered


by the Pathans not the Arabs and, in any case, the claim to
d

this name was recent (Census-B 1901: Appendix 11; also


or

see Ibbetson, 1883: 153–4) which means that the name had
xf

been usurped to claim gentility.


O

In traditional South Asian Muslim society respectability was


conferred by components of names proclaiming a foreign
origin as well as high religious status. As the Prophet of
Islam’s family is the most prestigious, ‘ashraf ization’ is
achieved by claiming Syed status. Although ‘Syed’ and its
South Asian equivalent ‘Shah’ are inspired by Islam, they
NA MES, POW ER, and CL ASS 61

are seen as being ‘caste’, ‘qoum’ (nationality) and ‘zat’ (type)


in Pakistan. Thus, rise in social status is correlated with the
assumption of more and more sonorous titles to one’s name.
Gauri Viswanathan points this out as follows:

For instance, almost in a crude sort of parody, the gradual


upgrading of a low-caste convert like Meher Chand is seen

s
in the progressive combination of names and titles that he

es
acquires through conversion to Islam: first the name of Meher
Ullah, then Meheruddin, next Meheruddin Muhammad, then

Pr
Munshi Meheruddin, Munshi Muhammad Meheruddin
Ahmad, then finally Maulavi Munshi Muhammad Meheruddin
ity
Ahmad (Viswanathan, 1998: 162).
rs
Urdu literature provides us with other such examples
too. For instance in Bano Qudsia’s short story ‘Shanakht’
ve

(identity; recognition) a village boy’s name is degraded into


ni

a pejorative nickname, Bagi, while he is a worker (kammi)


in the village, but assumes the respectful form of Hussain
U

Ali Bagh when he becomes a famous poet in England. But


d

the treatment of his name in his childhood is described as


or

follows:
xf

It is not known whether his name was Hussain Ali or Bagh


Mohammad but the whole village called him Hussaina Bagi.
O

He was so inferior, so devoid of value or status that nobody


bothered about the correct form of his name. ‘Bat’, ‘pumpkin’,
‘vegetable’ were all titles of affection for him (Qudsia, 2007:
226).
62 Names: A Study

Other group names inspired by Islam connect with the


Sufic influences in India. The names evocative of venerated
saints in India—such as the pioneers of the Qadiri and
the Suhrawardy orders Sheikh Abdul Qadir Gilani (1077–
1166), Sheikh Shahabuddin Suhrawardy (1155–1191 or
1208?), Sheikh Moinuddin Chishti (1141–1230) and

s
Sheikh Ahmad Sirhindi (1564–1624) who was also called a

es
reformer (mujaddid)—are Qadiri, Suharwardy, Gilani/Jilani,
Chishti, Mujaddidi, and Sirhindi (Sirhind being a small

Pr
town two hundred miles from Delhi). All these derivatives
from prestigious figures or locations are seen as labels for
ity
groups rather than religious tokens. Certain other places
associated with pious or powerful people are also deemed
rs
as having social worth, e.g., Hamdan (Hamadani), Bukhara
ve

(Bukhari), Gardez (Gardezi), etc.

Lyon tells us that he encountered many people whose


ni

occupation was considered menial in Pakistan but they


U

pretended to belong to a more respectable social group.


The servant of his host, a Malik, called himself ‘Bhatti’,
d

though people outside the village of Bhalot did not accept


or

him as a Rajput (2002: 101). In another instance, ‘a man


xf

was a nai (barber) but his qaum was something different’


(Ibid.: 127). This desire to appropriate a higher caste status
O

may, however, meet with resistance. Lyon goes on to tell us


that ‘on one occasion a barber told me he was a Gujar. A
young landlord laughed at him and told him that he could
be Awan if he wanted, but not Gujar’ (2002: 134).
NA MES, POW ER, and CL ASS 63

It is perhaps this obsession with social status, as expressed


through titles and caste names, which makes the political
rural elite use several such components in their names.
Naming for this class is, in Bourdieu’s words, an act ‘of
institution’ (social class as institution in this case) (1991:
121) which constructs a person’s identity for others and

s
himself or herself to perceive. In this case the piling up of

es
prestigious components adds up to the social superiority and
the right to control others in society. Thus, the legislative

Pr
assemblies of Pakistan, both at the federal and the provincial
levels, have names which were titles and even now betoken
ity
leadership and authority in the rural areas. The rural gentry
and aristocracy comprise Khans, Sardars, Chaudharys,
rs
and Maliks. However, as mentioned by Stephen M. Lyon,
ve

‘the terms Malik, Khan, Choudry and Sardar all indicate


titles of respect, but in other contexts these may simply
ni

be names’ (Lyon, 2002: ii). Moreover, ‘it is not uncommon


to employ a Punjabi and a Pathan title (Sardar Ovais Ali
U

Khan)’ (ibid.: iii). An analysis of prestigious components in


d

the legislative assemblies up to March 2013 may be useful


or

for understanding how certain caste names indicate social


dominance in Pakistan—especially in rural Pakistan. These
xf

components can be seen in Table 3:


O
64 Names: A Study

Table 3
Balo­
National Sindh Punjab
chistan KP As-
TOTALS Assem- Assem- Assem-
Assem- sembly
bly bly bly
bly
Total Names (906) 256 105 359 63 123
Total components

s
1054 398 1162 219 328
(3, 161)

es
Density (3.03) 4.12 3.79 3.24 3.48 2.67
Prestige/caste

Pr
components given 329 130 409 82 94
below (1,044)
Khans 50 17 47 14 55
ity
Syeds 35 35 21 2 9
Sardars 18 3 24 4 04
rs
Chaudharys 21 Nil 42 Nil Nil
ve

Maliks 15 Nil 33 Nil 02


Eminent 29 12 14 19 2
ni

Caste/tribe/
141 63 221 42 22
biradari
U

Religious titles 20 Nil 7 1 Nil


Percentage 31.21 32.66 35.20 37.44 28.66
d
or

The percentage of occurrence of prestige and caste com­


xf

ponents in the names of these politicians is more than the


names of ordinary people having, on the average, three com­
O

ponents per name coming to an average of 33.03 per cent


prestigious/caste components in their names. The working-
class men of Pakistan—(Poor-M sample of 500,000 names
with 1,030,000 components, have only 12.68 per cent such
components, Annexure-I). Even among the lower middle
and middle classes they range from between 05 to slightly
NA MES, POW ER, and CL ASS 65

above 31 per cent. It is, however, more (22.20 per cent)


in the urban upper middle and upper classes. Even boys
ranging between the ages of four and 13 (i.e., those named
in the 2000s) have 11.60 per cent such components in their
names though, in general, people use such components
as grown-ups (Elite-B in Annexure-I). It should also

s
be pointed out that the Sindhis and Baloch use caste

es
components much more than Punjabis and even Pashtuns
(for Sindhi caste names see Suhrawardy, 2012). That is why

Pr
names from Sindh and parts of Baluchistan have more caste
components than those from the Punjab. Yet another point
ity
is that people tend to use such components when they
leave their immediate neighbourhood so that a Pashtun
rs
inevitably uses the prestige component ‘Khan’, which he
ve

does not necessarily use at home. It should, however, be


clarified that the title ‘Khan’ is not used as a title but loosely
ni

as a part of the name even among poor people. Also, that it


is not a Pashtun monopoly, contrary to popular perception,
U

but is quite commonly used by the Baloch and the Punjabis


d

and even in Sindh (Chaudhary. Int., 2012; Buzdar. Int.,


or

2012; Khalil. Int., 2012).


xf

Among the category of ‘well-known family’ the names of


landowning families are categorized under the head of
O

‘caste’. Families which have gained prominence through


business, such as the famous twenty-two industrialist
families of Pakistan, are included under this head. However,
some of their names, such as Dawood, Jalil, and Habib are
too common to be distinctive. Other names, such as Saigol,
Isphahani, and Adamjee may be used for this purpose. The
66 Names: A Study

change which came in this entrepreneurial class is that


‘the Karachi based Memon groups fell in ranking while
the Punjab based Chinioti groups rose’ (Rahman, Taimur,
2012: 220). This, however, has not given us any distinctive
family names as these new capitalist families use the caste
names common to other, less distinguished, members of

s
the same extended family. As mentioned earlier, titles once

es
exalted, are used by humble people as names. The nawabs,
for instance, were feudal6 noblemen and the name is still

Pr
used as a title denoting that status among the feudal and
political elite. However, the title becomes just a name like
ity
any other name among the poor. Some poor villagers called
Maliks or Chaudharies might well have been landowning
rs
families in the past and now they may be impecunious but
ve

the family name continues to be used for prestige. In short,


the prestigious components may be hollow in meaning,
ni

being mostly labels and pretences, although the Syed (Shah


Ji) is still given respect even if the family is impoverished.
U

The prestigious caste labels are what Taimur Rahman calls


d

‘zamindar quoms’ (Syed, Awan, Rajput, Gakhar etc., etc.)


or

(Rahman, Taimur, 2012: 116). The working class (kammi)


labels such as musalli (sweeper), nai (barber), kumhar
xf

(potter) etc., are so low on the social scale that these are
O

concealed rather than displayed. Thus they are absent


from names as a rule, though some trades, like ‘mallah’
(sailor) in Sindh are displayed. Thus the caste names which
are assumed in order to climb the social ladder are the
prestigious caste labels and not the working class ones.
NA MES, POW ER, and CL ASS 67

It must be reiterated that these are not last names in the


Western sense of the term since they are not used by all
irrespective of social status, but manifestations of one’s caste
status. However, the Pakistani upper middle class and the
middle class in the urban areas of the country is developing
a sense of the family name not necessarily based upon caste.

s
In short, in the rural areas prestigious caste names still

es
index a socially dominant identity, while in the urban ones
they are becoming family names on the Western model.

Pr
Rural identity is also indexed with certain other kinds of
first and second names. This point will be taken up in more
ity
detail in the next chapter on modernization of names but
rs
it may be mentioned briefly that some women use old
fashioned components which usually come from Muslim
ve

heritage languages but have now become indigenized.


Among these are: Mai, Bano, Begam, and Khatoon and
ni

morphemes like –nissa (women), –dil (heart), and –gul


U

(flower). The first five components are hardly ever found


in urban names. The others are still found in the middle
d

class and the name Gul is popular among the Pashtuns of


or

all classes.
xf

Besides having less or none of these components, urban


O

women generally have male names in the end so that in


the data of the 1990s hardly any woman has only feminine
names (Annexure-H). Even the prestigious components in
female names are Syed and Khan (earlier used for men)
nowadays though they used to have feminine endings
making them Syeda and Khanam in the early years of
Pakistan. While in India, ‘a strong tendency in favour
68 Names: A Study

of secular names’ has supplemented religious naming


(Mehrotra, 1979: 206). In Pakistan religious naming
has increased but, as Chapter 5 indicates, this naming
is different from the past. Population census reports of
Pakistan tell us that urbanization has increased in the
country from 17.8 in 1951 to 32.5 per cent in 1998 (in

s
Rahman, Taimur 2012: 169). This corresponds with the

es
decrease of rural components in names as immigrants to
the cities often change them because of culture shame.

Pr
Thus, while our pilot survey of 476 rural women named
in the 1940s and early 50s shows that 91.60 per cent had
ity
Bibi as part of their names, immigration to cities changed
this trend (Annexure E-1). Parents from rural background
rs
who sent their daughters to the Apna Model School had far
ve

less. Out of 282 female names from this sample only 9.57
per cent had Bibi in them (Apna G in E-1). On the other
ni

hand 389 girls from affluent families in Lahore studying in


BNU had only 2.57 per cent such names (BNU-G in E-1).
U

As for the 608 urban Punjabi girl children born in the late
d

2000s this percentage is only 0.99. And even this figure


or

is because of names with Gul, Nissa, and Dilcomponents


and not Bibi or Begum, etc. (Elite-G in Annexure-E-1). In
xf

short, rural, or old fashioned names are abandoned when


O

people migrate to the cities. Thus the reduction of rural


components in names is a certain indicator of urbanization.

Indeed, there is so much culture shame for these quintes­


sentially rural Punjabi names that a woman working as
a maid servant who had once told her employer that her
name in the village was Allah Ditti refused to acknowledge
NA MES, POW ER, and CL ASS 69

that as her name, insisting again and again that her name
was Saima. When confronted by her employer in private
she said that the village name was not to be told to an
outsider (the researcher) (Saima. Int., 2012). Such name-
changing out of embarrassment about old-fashioned or
rural names has also been reported for India (Mehrotra,

s
1979: 207).

es
Because of this piling up of prestigious caste labels the

Pr
Pakistani elite, especially the landowning political elite,
have longer names than the ordinary people (with an
average of 3.03 components versus only 2.06 in our
ity
sample in each name). Also, it is much more (33.03 versus
rs
12.68 per cent) prone to using prestigious components
functioning as family names than the masses. In this case
ve

names carve out a niche for one’s self in networks of power


being indexed to groups functioning as people who wield
ni

power in Pakistani society. In another development, in the


U

urban areas these badges of traditional power are being


transformed into something like the family name in the
d

Western, specifically Anglo-American, sense of the concept.


or

Here they are indexed to two identities: the historical one


xf

of an indication of belonging to a respectable family, i.e.,


not a working-class name; and the second identity is the
O

modern, Westernized one, which derives its power and


status from offering services which the modern world
needs. The two might appear antithetical but they are both
sources of power and prestige and, hence, complement each
other. In short, prestigious caste names still index a socially
dominant identity in Pakistan.
5
First Names, Nicknames,
and No-Naming

s
es
Pr
The choosing of names described earlier as well as beliefs
about names pertain to what are called personal or first
names in the modern Western systems of naming. At the
ity
risk of some repetition and overlap it should be clarified
rs
that the names Ali, Mohammad, Hasan, Hussain—the
most common Islamic components used in South Asian
ve

Muslim names—are used both as first and second names.


Even a name like Abdul Haq, literally meaning the ‘slave
ni

of God’, which should not be abbreviated or split into its


U

formative components according to strict interpreters of


Islam, is actually reduced to ‘Haq’ and starts functioning
d

as a family name if it is repeated among the sons’ names,


or

such as Zia ul-Haq, Fazl ul Haq and so on. In the present


xf

study, the criterion for calling a name a ‘first name’ for the
present study is actual usage. Ghulam (slave) is used with
O

many names but it is rare to call someone Ghulam so it has


not been counted as a name. However, Gul or Jan are used
as both first and family names and have been considered
so in this chapter.

First names are often based upon natural objects and


phenomena in several communities, such as the Palestinians
F IR ST NA MES, NIcknameS, A ND NO-NA M ING 71

(Atawneh, 2005: 155–165) and the Irish (M’Clure, 1879),


and this is the case in Pakistan’s rural areas and the working
classes. Flowers, trees, fruit, rivers, sea, plants and the days
of the week as well as months and important events of the
yearly cycle provide names for nearly all rural communities.
In Sindh the names which are based upon trees and fruit

s
are as follows: Kirar; Kandero; Soof; Daru; Laung; Rabail,

es
etc. (Somroo. Int., 2013).

Pr
Names based on flowers (gul, phool, chanbeli, kanwal, yasmin,
gulab etc), plants (boota), animals (asad, arsalan, sher), birds
(tota, shaheen, baaz), rivers (darya), sea (samundar), mountain
ity
(Pahar), cloud (badal), precious stones (lal, heera, zamurrad,
rs
neelam), precious metals (sona, chandi), days of the week,
religious events (Eid, Ramadhan, Muharram), etc. Names
ve

based upon weeks in Sindh are: Aacher (Sunday), Soomar


(Monday), Jummo (Friday). The equivalents in Urdu and
ni

Punjabi-speaking families are Itwari, Peeru, Jumma as well


U

as Jumrati (Thursday). However, not all names are shared


by all communities in Pakistan. For instance the name
d

Shubrati and Ramzani are mostly found among working-


or

class Urdu-speaking immigrants (Muhajirs) from India.


xf

The following table shows the use of names of animals,


O

natural objects, and days of the weeks and important


cultural events used by people:
Table 4

Pakistan

Elite-M

Elite-B
1950s 13,771 chistan

Punjab
Sindh
Balo­

KPK
1950s 2,950
B’Tan-M

Poor-M
Sind-M

500,000
(1950s)

(1990s)

(1950s)

(1990s)

(1950s)
22,464

65,331

63,013

42,692

(2000)
s
3,649

512
es
Sher (lion) 127 13 1724 115 101 1579 279 10 Nil
Boota
Nil Nil 145 31 36 01 01 Nil Nil

Pr
(plant)
Badal
02 Nil 96 Nil Nil Nil Nil Nil Nil
(cloud)
Darya
ity
13 06 296 Nil Nil 05 01 Nil Nil
(river)
rs
Samundar
37 Nil 44 Nil Nil 09 01 Nil Nil
(sea)
ve

Chattan
Nil Nil 19 Nil Nil Nil Nil Nil Nil
(rock)
Pathar
ni

Nil Nil 10 Nil Nil Nil Nil Nil Nil


(stone)
U

Baaz (eagle) 222 Nil 300 02 04 139 22 03 Nil


Juma
221 09 410 03 01 66 07 Nil Nil
(Friday)
d

Jumerat
or

02 Nil 14 Nil Nil Nil Nil Nil Nil


(Thursday)
Mangal
xf

08 01 24 Nil Nil 06 01 Nil Nil


(Tuesday)
Itwar
O

Nil Nil 22 Nil 01 Nil Nil Nil Nil


(Sunday)
Shuberat
02 Nil 03 Nil Nil Nil Nil Nil Nil
(a festival)
Eid
05 03 146 Nil Nil Nil 04 Nil Nil
(a festival)
Ramzan
(month of 49 81 429 139 245 26 26 04 01
fasting)
F IR ST NA MES, NIcknameS, A ND NO-NA M ING 73

These names are not, however, used by urban, Westernized


people and have become indicators of a rural background
and little or no formal education. Their decrease from the
1940s–50s indicates both urbanization and modernization
in Pakistani society. Also, the fact that they are found in
Balochistan, the least modernized of Pakistan’s provinces,

s
reinforces the association of these names with earlier

es
onomastic practices. In keeping with modern trends in the
samples of boys from the elite of Lahore none of the names,

Pr
except one case of Ramzan, exists (Elite-B in Table  4).
However, for men from the elite the names Sher (10 cases),
ity
Baaz (3) as well as Ramzan (4) are found (Elite-M in
Table 4). But this elite has people who have settled down
rs
in the cities as well as competent rural students who have
ve

entered the bureaucracy or the military by dint of their own


efforts. In any case, the name Sher as well as Baaz belong
ni

to predators and the macho culture of Pakistan values


these names even in the urban areas. One notable feature
U

is that, while urban names may be based upon the names


d

of animals and objects, they are in languages other than the


or

common peoples’ languages except among Sindhi, Baloch,


and Pashtun nationalists. For instance the names Shahbaz
xf

(eagle), Asad, and Arsalan (lion) are very much in use in


O

all classes. Similarly, while the name Chandni (moonlight)


is a rural name for women, the name Seema or Seemee
(silver in Persian) is used in the urban areas. Similarly, the
names Moti (jewel), Heera (diamond), Sona (gold) are old-
fashioned or rural but the Perso-Arabic equivalents Dur,
Almas, and Zareen are used by the elite too. Other rural
names for women not used in the urban areas are:
74 Names: A Study

Table 5
Elite-G Elite-F
Pakistan Poor-F
2000s 1950s
Chand 558 Nil Nil
Chandni 151 Nil Nil
Bakht 945 04 Nil

s
Chinar 147 Nil Nil

es
Chandi  81 Nil Nil

Pr
Although there are several guesses about the most popular
first names in Pakistan, this is the first time an attempt
is being made to calculate the frequency of occurrence of
ity
first names from fairly large data bases. Earlier lists, such
rs
as those available on the internet, do not indicate how they
arrived at the results. For instance, one such list gives a list
ve

of 100 most popular first names for boys and 84 for girls.
In this Ali has the first place among boys’ names while
ni

Muhammad is placed at the 23rd rank. This list does not


U

indicate the size of the sample nor the methodology of


the selection of the list. Another problem is that this list
d

counts Khan and Abdul as first names whereas the first


or

is a title used as part of a name and the second, meaning


xf

‘slave of ’, never stands alone but is always dependent on


another name which could stand alone and be considered a
O

first name (Names Pakistani, 2013). Our data, given below,


suggests that Muhammad is much more popular than this
list indicates.

An analysis of twenty-three samples suggests that there


are three trends in evidence from people’s names of the
1940s–50s till those of 2000. These are: Islamization,
F IR ST NA MES, NIcknameS, A ND NO-NA M ING 75

Arabization, and the quest for novelty. Arguably parents


and other name-givers try to balance the quest for novelty
with a number of other imperatives—a link with ancestors,
emphases on certain aspects of identity (Islamic, modern,
urban, etc.) and peer group pressure. Sometimes an
intervening variable, such as the sudden rise in popularity

s
of an iconic figure, produces a cluster of names not in use

es
earlier. The cumulative effects of these trends create fashions
and swings in naming from generation to generation.

Pr
There are, however, constants which operate more on male
names than female ones. For males the major constants are
ity
the names of the Ahl-i-Bayt which have been explained in
rs
Chapter 3. Although, out of the sum of all the samples (N=
795,283) the name of Ali occurs 135,240 times (17.01 per
ve

cent) while that of Muhammad 93,018 times (i.e., 11.67


per cent), this statistical fact does not indicate the real
ni

popularity of the name Muhammad. To gauge its popularity


U

it should be borne in mind that out of the thirteen samples


of male names, the name Muhammad occupies the first
d

of the top ten positions in eleven. In only two it occupies


or

the second and the fourth positions. However, the sample


xf

in which it occupies the fourth position happens to be the


sample of poor people (Poor-M), which is many times
O

larger than the others combined (N=500,000) and in


this Ali occupies the first position. Thus the number and
percentage of Ali appears to be the highest. However, the
name Muhammad is popular in most of the other samples
we have, so it may be said to be the most popular name in
Pakistan among males. But Ali definitely comes next to it.
76 Names: A Study

There is no sample in which it slips below the fifth position


and in two it occupies the top position. Similarly Hussain
and Hassan occur in most samples and Ahmad, a variant
of the name Muhammad, is a perennial favourite. The
positions of these names among the top ten most popular
names is given in the table below. Names which occur in

s
less than six samples are not considered here:

es
Table 6

Pr
Names in Top Positions in most samples
Out of Thirteen samples of Male Names
ity

which the name occurs


Number of samples in
Names
rs
out of
the ten POSITIONS
ve

Top
Names
ni

1st 2nd 3rd 4th 5th 6th 7th 8th 9th 10th
U

Muham­
mad 11 2 Nil Nil Nil Nil Nil Nil Nil Nil 13
d

Ali 2 6 2 1 1 1 Nil Nil Nil Nil 13


or

Ahmed Nil 4 3 4 1 1 Nil Nil Nil Nil 13


Hussain Nil Nil 2 6 1 Nil Nil Nil 1 Nil 10
xf

The name Hassan occupies the next postion but has not
O

been counted as it occurs in six samples (i.e. less then half )


but it is nevertheless very popular. The presence of these
names among the top ten positions leaves only about five
positions at the most to be filled in by other names which
one may, albeit inaccurately in some instances, call ‘secular
names’. In the case of female names, however, the iconic
F IR ST NA MES, NIcknameS, A ND NO-NA M ING 77

power of Islamic names (Fatima, Amina, Ayesha, Maryam,


etc.) is less compelling. They are popular but are not a
compulsory norm in naming.

Table 7
Names in Top Positions in most samples

s
of Female Names

es
which the name occurs
Number of samples in
Names

Pr
out of
the ten POSITIONS
Top ity
Names
rs
Total Total
1st 2nd 3rd 4th 5th 6th 7th 8th 9th 10th
= 10 = 10
ve

Fatima 3 2 Nil Nil 1 Nil 1 Nil Nil 1 7


Amina 1 Nil 2 2 1 1 Nil Nil Nil Nil 7
ni

Ayesha Nil Nil 2 Nil 1 Nil 1 1 1 Nil 6


U

As mentioned in other contexts earlier, rural names have


d

what may be called indigenous components perceived as


or

being religious. For instance, the word for ‘given’ in Sindhi


is ‘Dino’. Thus Allah Dino (God-given) occurs in the
xf

samples from Badeen (Annexure M-1). Among women


O

such components are Hawwa (Eve) which occurs in the


list of Badeen (M-1) and Jannat (Paradise) which occurs
among poor people from all over Pakistan (M-4) as well
as in Badeen (M-1). The name for the Muslim month of
fasting (Ramzan) is also a favourite among the rural poor
(M-1) and, though it occurs in other samples too, it does
78 Names: A Study

not occupy the list of the top ten most popular names as it
does in our Badeen sample.

Among Pashto-speaking people Gul, Jan, and Sher are


common male names. Gul, however, is also a favourite female
name. In the analysis of our samples names having Gul as
an inseparable component—Gulistan (garden), Gulfishan

s
(scatterer of flowers), Gulzar (decorated with flowers) etc—

es
have been ignored. Only the name Gul in isolation has been

Pr
counted. Even so it occupies the top position among female
names in our sample from Balochistan (M-3), the second
position among the poor people of the whole country (M-
ity
4) and the top position among lowest middle and middle
rs
class names of the 1990s in KPK (M-7). As for male names
the following chart illustrates the popularity of ‘Gul’ along
ve

with ‘Jan’ and ‘Sher’. The numbers given in the table below
indicate the position from the top among the top ten names
ni

in the following samples.


U

Table 8
d

Balochistan Balochistan
or

KPK-M KPK-F
-M -M Poor-M
(1990s) (1950s)
(1950s) (1960s–80s)
xf

Gul 2 3 8 7 2
O

Sher 4 7 Nil Nil Nil


Jan Nil 2 Nil 6 7
Source: Annexures M1, M3, M4, & M7.

However, all the above samples are either from rural areas
or belong to the lower middle and middle class families of
KPK.
F IR ST NA MES, NIcknameS, A ND NO-NA M ING 79

Both male and female names vary according to the fashion


of the period. Apart from the Islam inspired names
mentioned above, men from the elite named in the late
1940s and 1950s were called Javed, Iqbal, Tariq, and Saeed.
Their grandchildren, now aged six or seven, are called
Fahad, Ayan, Saad, Faisal, etc. There is a trend towards

s
Islamization—Hamza, Usman, Omar, Abdullah, etc.—but

es
more so towards Arabization. The quest for the unique
name which brought Tariq and Iqbal in fashion during the

Pr
1950s, has now brought in Abdullah, Hamza, and Zain
among young boys (M-5). ity
Among the lower middle and middle class Punjabis named
rs
in the late 1940s and 1950s, the fashionable male names
were Iqbal, Aslam, Anwar, Rashid, and Nazeer. Among
ve

those named in the 1990s, one finds only Iqbal in the list
but the other names which have replaced them are Islamic:
ni

Usman, Raza, Hassan, Bilal, and Abbas (M-6). Among


U

women of the same year the 1950s names were all ‘secular’
except Fatima. In the 1990s Ayesha, Amina, Maryam, and
d

Rabia are all counted among Islamic names. However,


or

female names have changed entirely. Instead of the 1950s


xf

Khalida, Parveen, Sultana, and Surayya, etc. we have in the


1990s Iqra, Sidra, Saba, Kiran, and Anam. These names are
O

unique and many of them are considered Islamic, (though


not necessarily by the ulema) because they are used in the
Quran—Iqra being the Quranic command to read—or
because they represent something which is considered
sacrosanct.
80 Names: A Study

In KPK, among the same classes of people, the most


popular 1950s names were often the same as those of the
Punjab: Parveen, Naseem, Shameem, etc. In the 1990s,
however, four religious names Ayesha, Amina, Fatima, and
Maryam made an appearance. Then there are new names:
Hira, Sana, Huma, and Saba. Out of these Hira, being the

s
cave in which the Quran was revealed, is considered Islamic

es
by people. In short, the same process of name givers seeking
more religious names as well as uniqueness as is found in

Pr
the Punjab, also occurs in KPK (M-7).

Female names among the elite bear strong evidence of


ity
the processes of Islamization, quest for uniqueness, and
rs
Arabization. In the names of the 1950s, for instance, there
is only one religious name: Amina. Among the girls named
ve

in the 2000s there are Fatima, Ayesha, Amina, Zaynab,


Maryam and Khadija. There are new names too such as
ni

Eiman, Rida, and Minahil which were unheard of in their


U

grandmothers’ time (Annexure M-5). But this Islamization


is not the invocation of traditional Islamic names. The
d

latter do exist but in far fewer samples than corresponding


or

male names in our samples.


xf

The frequency of occurrence of first names in our samples


O

suggests that personal names have become more Islamic


and Arabized. There is also a drift away from Persianization
to Arabization. Names of Arabs—Saud, Faisal, Fahad,
Saddam, etc.—are also in fashion as Arabization is
associated in the public mind with Islamization. As these
names are modern in the sense that they were not used in
the 1950s, the new trend of Arabization is equated with
F IR ST NA MES, NIcknameS, A ND NO-NA M ING 81

modernization too. Thus, a girl called Hira would probably


be in her twenties, whereas women called Rehana or Sabiha
would probably be in their fifties. Of course the constants—
Muhammad, Ali, Fatima, Amina, Ayesha, etc.—could be
of any age.

Other popular names have changed too since the late

s
forties and early fifties of the last century. Three very

es
popular Pakistani male names—Tariq, Khalid, and Javed—

Pr
went out of fashion. This is illustrated in the following data
from men named in the 1950s and in the 1980s from the
upper and upper middle classes of the Punjab.
ity
rs
Table 9
Elite-M (P) BNU-B
ve

Name
(1950s) 3,649 (1980s) 512
Javed 77 (2.11 %) 01 (0.20%)
ni

Tariq 78 (2.14%) 06 (0.17%)


U

Khalid 107 (2.93%) 05 (0.98%)


d

In short, apart from the Islamic constants, first names


or

change according to fashion but are more liable to change


among women than among men.
xf
O

Nicknames
Nicknames have been interpreted in terms of ‘social action’,
i.e., how we negotiate relationships, power differentials,
and social roles in a given context in society (Leslie and
Skipper, 1990; Moore, 1993). We have mentioned the
role of nicknames in the social structure of other societies.
82 Names: A Study

In Pakistan, as in other societies, they perform the same


functions: ego-boosting and ego-deflation. In most cases it
is the latter function which is in evidence. Sometimes the
criticism or insult can be so subtle or indirect that a dog
may be called by such names instead of the person, in order
that he won’t feel too bad. This is the verbal strategy used

s
by the Baatombu of northern Benin (Schottman, 1993). If

es
the nickname is invented by the person himself it is usually
meant to protect his self image or maximize his feelings

Pr
of gratification about himself. If invented by others they
may be meant to serve the same purposes if the people
ity
who coin and use it look up to the person. However, if the
coiners are hostile or contemptuous towards the person,
rs
the nicknames they use may be personally derogatory
ve

and reduce his gratification. Notwithstanding Islamic


injunctions against opprobrious nicknames mentioned
ni

earlier, they are the norm in Pakistani society. However,


unlike other societies mentioned below, nicknames never
U

become formal names in Pakistan. In Palestine names of


d

people may have started as nicknames, and sometimes


or

may have even been derisive ones based upon body


parts, personal qualitites, etc. (Atawneh, 2005: 151–5; for
xf

other Arab societies see Goitein, 1970). This seems to


O

be the case of early Chinese names also (Goldin, 2000).


In Pakistan, however, such names remain nicknames.
Nicknames, then, indicate the power of naming and even
the ‘tyranny’ of it (Morgan et. al.,1979). But this ‘power’ is
not formal authority conferred by any official institution
or society. Sometime these appellations are the weapons
of the weak; or those placed temporarily in subordinate
F IR ST NA MES, NIcknameS, A ND NO-NA M ING 83

roles. For instance school teachers, though formally placed


in authority over their students, might be the object of
ridicule as expressed through the nicknames given to them
by their pupils. In this case the pupils use nicknames as
the weapons of the weak against their teachers (Crozier,
2004). Nicknames are a very common phenomenon in

s
Pakistan. Our survey indicates that 167 (53.02 per cent)

es
respondents said that they were given nicknames. But these
were probably pet names which did not upset the ego of

Pr
the respondents. Names given as a joke, which can really
hurt the ego were not even reported by 123 (39.05 per
ity
cent) respondents while 18 (5.71 per cent) gave no answer
at all. Only 70 (22.22 per cent) confessed that they were
rs
given such nicknames and 25 (7.94 per cent) said they had
ve

been given such names for a short time. Indeed, it seems


that the 79 (25.08 per cent) who said they had never been
ni

given such names seem to be concealing these unpleasant


epithets that they had faced.
U

The names invented by a person, such as the pen name


d

(takhallus), will be mentioned later. Other fictitious names,


or

called aliases, are used by criminals such as ‘commando’,


xf

‘captain’ and ‘general’ in Pakistan. These are leaders of


criminal or militant members of political parties and
O

gangs who take up these aliases to proclaim their courage,


militant credentials and leadership qualities. In some cases
these aliases are used by the followers or those who look
up to such characters. Not all such aliases apply only to
desperadoes. Sometimes, a nickname like Pari, Guriya
may be used for a pretty little girl by her doting family or
admiring friends.
84 Names: A Study

In the Punjab some nicknames are reserved for certain


physical qualities. For instance, a boy with light eyes will
invariably be nicknamed Billa (tomcat). The corresponding
female nickname Billi (female cat) is usually changed
to Billo. However, it is a common nickname and is not
reserved only for girls with light eyes. That is why the

s
implications of the following popular song:

es
Asan te jana billo de ghar

Pr
Kinne kinne jana Billo de ghar
(I have to go to Billo’s house;
Who all have to go to Billo’s house)
ity
—offended the parents of so many girls and also doubtless
rs
sometimes resulted in their daughters becoming the
ve

recipients of unwanted attention from boys.

Nicknames always conform to the phonological rules of


ni

the local language. Thus the guttural sounds of Arabic


U

or Persian (the ovular stops and fricatives) are not used


in nicknames. Instead, corresponding sounds from the
d

indigenous languages of Pakistan such as Punjabi, Sindhi,


or

Pashto, and Urdu are used. Nicknames are also shorter


xf

than formal names from the Islamic heritage languages.


Thus Fazal becomes ‘Phajja’, Abdullah becomes ‘Dulla’,
O

Ghulam Rasul ‘Soola’, Rafeeq ‘Pheeka’ and so on. In KPK


the vocative forms of names as well as nicknames take the
vowel a, i.e. Sahib becomes Sahiba, Gul becomes Gulla,
etc. As in the Punjab, nicknames are common among
Pashto-speakers also as Malala Yousafzai, the girl who
F IR ST NA MES, NIcknameS, A ND NO-NA M ING 85

survived being shot in the head by the Taliban, tells us in


her biography:

In our culture we all have nicknames—aside from Pisho,


which my mother had called me since I was a baby, some of
my cousins called me Lachi, which is Pashto for ‘cardamom’.
Black-skinned people are often called white and short people

s
tall. We have a funny sense of humour (Yousafzai, 2013: 16).

es
Among the Westernized elite, as already mentioned,

Pr
nicknames follow Western (English) patterns of naming.
Here too, as in other groups of people, they express
ity
solidarity and affection if given by people who like and
care for the person and hurt and insult if given by rival or
rs
teasing peers. If meant to signal defects and shortcomings,
ve

especially physical ones, they are a cause of distress but


retain their group identity marking function nevertheless.
ni

In this sense nicknames are part of ‘onomastic politics’ at


U

the level of one’s social circle. The power of those who are
socially more powerful, even if this power is ephemeral,
d

can give names which can control, marginalize, insult, or


or

honour and praise an individual. Some of these nicknames


can be so insulting that one of our interviewees from South
xf

Punjab (Muzaffargarh) refused to divulge the nicknames


O

of the children of his village on the ground that these


were not fit for the author’s ears (Abid. Int., 2012). This
is probably why people tend to deny that they were given
nicknames as a joke (as opposed to being given nicknames
out of affection).
86 Names: A Study

No-Naming
In many pre-industrial societies the personal name is not
used by younger people for their seniors. Instead, kinship
and even fictive kinship terms are preferred. According to
Alford’s comparative survey of 60 societies the following
facts emerge:

s
es
Table 10
Taboo on the use of No. of Societies Percentage

Pr
Parents’ name 18 30
Spouse names 16 26
ity
Parents-in-law’s names 17 28
Source: Adapted from Alford, 1988: 106.
rs

Indeed, ‘addressing someone by name indicates equality


ve

with or superiority over that person’ (Alford, 1988: 102–3).


ni

In Pakistani society also the prerogative to address someone


by the first name while receiving ones own name with
U

honorifics (e.g,. Sardar Sahib, Nawab Sahib, Syed Salman


Shah Sahib, Shah Ji) is an indicator of power. The person
d
or

with greater power, whether because of age, wealth or


societal status, can use the first name of a subordinate
xf

while the subordinate will always reciprocate with name


O

plus honorifics or fictive kinship terms. As Lyon observes,


‘assymetrical relationships, and the culture of intervention,
have become the modus operandi of the society’ (Lyon,
2002:  3). Thus naming conventions are also indicators
of the hierarchy among donors (patrons) and recipients
(clients) in Pakistan’s patronage oriented network. Among
north Indian ashraf Muslims too, elders are never called by
F IR ST NA MES, NIcknameS, A ND NO-NA M ING 87

their names by younger people but kinship terms are used.


This pattern of behaviour is also shared by the Hindus
of North India among whom, besides the usual terms of
kinship, ‘the English terms “auntie” and “uncle” for related
friends or neighbours of the parental generation’ (Vatuk,
1969: 269) are also used. This is also true for urban

s
Pakistanis.

es
In Pakistan married couples, in common with other pre-

Pr
modern societies as indicated in the table above, e.g., the
Hidatsa (Barnes, 1980: 322), use either tekonyms (‘Munne
ki Amma/Abba’, meaning ‘the mother/father of the baby’),
ity
or simply a vocative, ‘sunte ho’, (meaning ‘do you hear?’) by
rs
a wife or bhalilok (good people) by a husband (Rahman,
1999: 172–82). Both spouses may also use the pronoun ‘un’
ve

(she or he) instead of a name, when referring to each other


while talking to someone else. This avoidance of names
ni

does not extend to in-laws nor is it as acute, as reported


U

for the Ethiopian Kambasta speaking people (Treis, 2005).


However, elder in-laws names are not uttered in Pakistan by
d

younger in-laws. Moreover, as in classical Greece (Schaps,


or

1977), women are not named in the rural areas or among


xf

conservative people. They may be referred to as androon-


e-khana, ghar wale (both Urdu terms for ‘the people of the
O

house’), kor walan (Pashto for the same) or bachhe (children)


by polite male friends. Among urban but traditional men
the formal term Begum Sahiba may be used, but the name
of the wife is neither used by the husband in front of his
male friends, nor indeed by the friends.
88 Names: A Study

As Pakistan is modernizing, names too are becoming


modernized in many ways. This, however, is the subject of
Chapter 6. Here it should be noted that no-naming is in
a state of transition or flux. The Westernized, urban elite
has adopted the Western fashion of using first names even
with colleagues and acquaintances, which is still not the

s
norm among more traditional people. This elite imposes

es
its preferences from a position of power and violates the
internal norms of Pakistanis about using honorifics and

Pr
not using the first name (Rahman, 1999: 183–223). While
many traditional norms and values of Pakistani society, such
ity
as honour killing, are completely indefensible, linguistic
politeness is certainly not. However, it is such innocuous
rs
norms which are challenged, rather than those which
ve

violate human dignity.


ni
U
d
or
xf
O
6
The Modernization of Names

s
es
Modernity has been associated with rapid social change
involving the use of Western categories of thinking,

Pr
categorization and behaviour in non-Western countries,
especially those which passed through the colonial
ity
experience. The construction of the Anglicized persona
in Western dress, speaking a Western language and
rs
using Western cultural markers is a familiar one. Names,
ve

alternative names, and patterns of naming under the direct


influence of this kind of modernity is called onomastic
ni

Anglicization here and is given attention below. But it


U

is not necessarily the similarity with the colonial master


which is an indicator of modernity. We can also understand
d

Partha Chatterjee’s idea of ‘our modernity’—the Indian


or

experience abstracted—as being different from the Western


one (Chatterjee, 1997). Our modernity ‘is the modernity
xf

of the once-colonized’ (Ibid.: 20). Hence we take many


O

features of modernity, especially those which empower


us like machines, computers, weapons and means of
communication and travel, but we also need to fall back
upon things to which we ascribe iconic value to mark our
‘differences’ from the homogenizing Western experience.
90 Names: A Study

Thus the notion which is most relevant for Pakistani


modernity, which is a subset of the South Asian one,
is that of multiple modernities (Eisenstadt, 1999). This
concept refers to ‘a multiplicity of cultural and social
formations going far beyond the very homogenizing
aspects of the original version’ which have manifested

s
themselves (Eisenstadt, 2000: 24). Thus we can talk of a

es
‘fractured modernity’ in India as Sanjay Joshi does in his
study of the making of the middle class in north India

Pr
( Joshi, 2001). Although about colonial Lucknow, the
argument that middle class modernity comprised pre-
ity
modern concepts of hierarchy masquerading as levels of
education and respectability, is relevant for Pakistan. The
rs
time dimension makes our modernity ‘entangled’ so that we
ve

can only grasp the Pakistani present in relation to the past


(Therborn, 2003). In the same way the assertion of ethnic
ni

identities, sometimes seen as a throwback to the past, is


one consequence of the consolidation of the modern state
U

with its monopoly over power and distribution of goods


d

and services. Communities have to organize themselves as


or

pressure groups around identity symbols such as language


in order to claim their share from the modern state. In
xf

that sense then, ethnicity is a product of modernity. This


O

imagined community—to use Anderson’s idiom (1983)—


is confronted by equally imagined groups based upon a
constructed identity. In short, Pakistan’s present day ethnic
assertion, expressed through a language based identity, is a
quintessentially modern phenomenon created by the same
forces of asserting differences in the face of powerful forces
of homogeneity, as is the assertion of religious identity.
THE MODER NIZAT ION OF NA MES 91

Since the construction of the modern Islamic identity has


already been studied in previous chapters (3 and 5) we will
focus on the effects of colonialism and the assertion of the
ethnic identity on Pakistani names in this chapter.

Onomastic Anglicization

s
Modernity was introduced to Asia and Africa through

es
the colonial experience. This experience was disruptive
of traditional cultures and identities as recorded in many

Pr
sources. One particularly eloquent expression of its effects
on names is given by Felly Nkweto Simmonds as follows:
ity
I belong to a time. The twentieth century. A time for
rs
fragmentation, a time of rebirth. I need to understand and
ve

know myself from that position. It is the only position I have


(Simmonds, 1998: 36).
ni

So her Zambian name, given by paternal grandparents,


U

Nkweto wa Chilinda, was changed to Nora as her parents


wanted to give her an English identity in addition to the
d

African one. Then, being educated in a Roman Catholic


or

school, she named herself Felicitas. The last name comes


xf

from her husband as his father carried Dutch genes. For


Indian Muslims, however, it was not so disruptive, at least as
O

far as names are concerned since the British neither forced


their own names upon them nor made any laws regarding
the use of family names, as were imposed by the Russians in
their empires and even the modernizing dictators of Turkey
and Iran. However, education and employment under the
British were the most powerful vehicles of modernization
92 Names: A Study

and both forced changes in fashions extending to language,


dress, food, entertainment, and a number of related issues.
As men went out in the colonial world to find jobs they
had to accept the terms of employment and this involved
changes in lifestyle, whether they liked them or not. But
to offset these apparent intrusions of modernity, Partha

s
Chatterjee argues, women were expected to preserve the

es
‘inner space of a community’ (Chatterjee, 1993: 147). They
were to preserve the Hindu, Parsi, or Muslim—as the case

Pr
may be—identity, while men could acquire the linguistic,
sartorial and culinary tastes and fashions necessary for their
ity
partial integration with the British rulers (Ibid.: 116–157).
Among South Asian Muslims, however, while women did
rs
not take to Western dress or other cultural patterns in large
ve

numbers even in the cities, they did take up what may be


called family names, which in their case meant male names.
ni

Let us put this change into context by remembering that


U

it was not carried out by decree in Pakistan, in contrast


to some other Muslim countries. In Turkey, Iran, and the
d

Central Asian states—the ruling elite carried out forced


or

Westernization of names. As mentioned above, Turkey


xf

the Kemalist Grand National Assembly of the Turkish


Republic passed a law on 21 June 1934 to the effect that
O

every citizen must have a family name (Spencer, 1961:


206). In Iran the surnames were introduced by Reza Shah
(r. 1925–1941) (UNICEF Iran, 2005). Now the state
supervises naming and does not allow names perceived as
belonging to non-Islamic cultures such as Jasmin, Sonia,
Sasha, Sanam, Manal, Neeli, Neena, Hana, etc., so as to
THE MODER NIZAT ION OF NA MES 93

counteract the Westernization of Iranian society which had


taken place in the days of the Shah (Persian Names, 2012).
Most of these names are used in Pakistan though only in
the upper, Westernized sections of society. In the Central
Asian Republics traditional Muslim names were made to
sound Russian—thus Rahman became Rakhmanov and

s
so on—probably ‘to destroy the tribal organization among

es
the Kyrgyz and by this strengthen the control over them’
(Hvoslef, 2010: 89). And now there is a movement for

Pr
the ‘Kyrgyzification’ of names to accentuate the national
indentity (Hvoslef, 2010: 91–92). Roughly similar trends
ity
can be found in the other former Soviet republics of
Central Asia (Abazov, 2007; Shahrani, 1993). In China
rs
names underwent a change for political reasons during
ve

the Cultural Revolution and when the revolution was


discredited, they changed again (Zhongti and Millward,
ni

1989).
U

In Pakistan the most obvious change among middle


and upper middle class women is that they adopt their
d

husband’s names after marriage, whereas in the precolonial


or

era, as well as among rural women even now, they suffixed


xf

components of femininity, such as Bibi, Bano, Khatoon,


Begum, or Mai. Nowadays in cases where the husband has
O

no fixed family or surname—unless it is a tribal or caste


label, the name taken by women is often his first name.7 In
addition, purely Western names like Sonia, Tania, Dushka,
Ramona are often given to girls while nicknames in the
Westernized elite are quintessentially Westernized both for
girls and boys. Even when names are chosen from Arabic
94 Names: A Study

or Persian they are untypical and unique as explained in


the previous chapter. In short, modernization has changed
the naming pattern for urban women in Pakistan and has
also affected the choice of first names and, even more
profoundly, the use of nicknames among both the males
and females of the same class.

s
Apart from taking up Western-sounding nicknames, other

es
stratagems to sound modern and Anglicized were adopted

Pr
by men. For instance, a local sounding pronunciation could
be changed so as to eliminate the double consonants. Thus
Jinnah, the founder of Pakistan, changed his name in order
ity
to discard a regional (Gujrati) identity and adopt a modern
rs
one. This happened sometime between 1893 and 1896 when
he was in London as the letters of that period indicate.
ve

The significance of this change does not escape Jinnah’s


biographers. Here is what Jaswant Singh says about it:
ni
U

There was also a transition in the way that Jinnah had not
just changed his name but also kept altering the spelling of
d

his name: It travelled from being Mohomedalli Jinnahbhai,


or

to a jettisoning of ‘Jennabhai’ and adopting to ‘Jinnah’; then


on to Mahomed Alli Jinnah, and yet again then, dropping the
xf

second ‘L’ from Alli, and still later adopting an additional ‘m’
O

to Mohamed, leading finally to the version of Mohammed Ali


Jinnah (Singh, 2009: 68).

Singh goes on to say that ‘this is not a small point’ as the


‘very young Jinnah was attempting to find exactly that name
which would fit the personality he thought that he would
be in the years ahead’ (Ibid.: 68). And this personality
would be that of a modern young English trained barrister.
THE MODER NIZAT ION OF NA MES 95

Other names changed by literary personalities during the


twentieth century are as follows:

Table 11
Original names Modernized version
Nazar Muhammad N. M. Rashid

s
Fazal Deen Anjum Roomani

es
Kazim Ali Waqar Anbalvi
Aurangzeb Qateel Shifai

Pr
Sanaullah Meera Ji 8
Dilawar Ali ityMirza Adeeb
Muhammad Omar Shaukat Thanvi
Mohammad Deen M. D Taseer
rs
Source: The above information is from Manto (1991: 413–415). More detailed
information on Taseer is in Mirza (1978: 9).
ve

Moreover, old patterns of naming consisting of rhyming


ni

middle names, common in ashraf families in the U. P. area


was dropped. These were naming patterns which repeated
U

components which distinguished one family from another.


d

Thus, in ‘Mai Dada’, a story by Asad Ali Khan, all male


or

names had the rhyme pattern—Uddin Khan. All young


men had to learn the shajrah (family tree) and woe betide
xf

the youth who got confused between the names of his


O

ancestors (one-Uddin Khan with another-Uddin Khan)


for this was tantamount to stigmatizing some of the
present senior males of the family as illegitimate. When
these patterns were discarded as being old-fashioned the
results were bizarre. For instance, the present author’s
family was a traditional landowning family from U. P. in
which the pattern was the repetition of –ullah: Ahmadullah
96 Names: A Study

(grandfather), Samiullah (father), Shafiullah (uncle) and


Rafiullah (another uncle). This was the pattern but not
the family name, which was identified by the caste name
of ‘Khan’—the family claiming origin in the present-day
KPK (Toru near Mardan to be exact). The pattern was
broken by the author’s mother who named her eldest son,

s
the author, on the pattern of her sister’s son. The new

es
pattern shared by the two cousins was Abdur Rahman.
Thus the author was Tariq Abdur Rahman and his cousin

Pr
was Khalid Abdur Rahman and, of course, Khan was always
added to emphasize the claimed Pashtun origin. When the
ity
family moved to Pakistan and was exposed to urbanized,
middle class, modern naming patterns the first part of the
rs
father’s name was added, so the author’s younger brother
ve

was called Ahmad Nurus Sami and sister Tayyaba Fatima


Sami. Further modernization abbreviated the names, even
ni

removing the caste names so that the siblings became


Tariq Rahman, Ahmed Sami and Tayyaba Sami. Upon her
U

marriage Tayyaba adopted her husband, Syed Azam Jaffar’s,


d

first name becoming Tayyaba Azam. Their children then


or

took the names of their fathers so that Tariq Rahman’s


son is Fahad Rahman and the daughter is Tania Rahman
xf

Kaudari (this last being her husband’s family name).


O

Ahmad Sami’s only child is named Maryam Ahmed and


Tayyaba’s children are Umair Jafar and Beenish Syed. In
short, neither the author’s siblings nor their cousins have
names which would place them in the same family.

Surprisingly, a similar story is shared by Janet Finch, a


British academic, whose father was Finch but who no
THE MODER NIZAT ION OF NA MES 97

longer shares this name with anyone ‘of the people whom I
regard as my close kin’ (Finch, 2008: 714). In her case this
has happened because people change partners and do not
necessarily change their previous name (if they are women)
while doing so. However, in the UK, which Finch studies,
naming is much more governed by rules, at least in theory,

s
than it is in Pakistan. Thus last names can provide a certain

es
network of relationships which, in Pakistan, is only possible
through using tribal labels, because these refer to such large

Pr
groups that they do not isolate individual family lines as the
name Finch does in Brtain. ity
Even in Pakistan’s villages a modern trend is evinced in
rs
another naming strategy, that is to discard first names
which are perceived to be old-fashioned. Among the
ve

female names these are: Reshman, Mukhtaran, Ghaseeti,


Basheeran as well as all names ending with Bibi, Bano,
ni

Khatoon, Mai and even Nissa. In regard to males, names


U

with the component Ditta, Deen, Ghaseeta and even


those prefixed with abd were categorized as being old-
d

fashioned. This modernizing trend is noted by humorists.


or

For instance, in the play ‘Dolly ki Aye gi Barat’ (Dolly’s


xf

Wedding Procession will Arrive) (Geo TV, 2011) even


the name Qasim is considered outdated, while names like
O

Razeen, Nail, Ayan, and Rameez for boys and Raheema and
Inaya for girls are considered fashionable. The name has to
sound modern and that means it should be unique, short
and from Islamic heritage languages if possible.

As mentioned earlier, rural identity is usually indexed


with certain kinds of first and second names. The most
98 Names: A Study

widely used components are given below. This data is from


the BISP sample (Poor-F) which number 300,000. The
following table illustrates this:

Table 12
Components of Names
Number

s
(N= 300,000)

es
Bibi 104, 634
Khatoon 24, 907

Pr
Begum 11, 100
Jan 8, 402
Mai 9, 270
ity
Bano 4, 708
rs
Total 163, 021
Percentage 54.34
ve

Source: Annexure-K
ni

In short, the main component which marks a woman as


U

belonging to a rural or old-fashioned background is Bibi.


Women from the large cities, such as Lahore, as expected,
d

have none of the old-fashioned components of names given


or

above nor do they eschew male names as surnames. Out of


a pilot survey from urban, upper middle class Westernized
xf

elite young women students of BNU only two girls have


O

both female names—Javaria Zahra and Zavaria Jannat—


but all the others have male names as surnames and none
of the old-fashioned or rural components are used. These
names have been noted in the sample called BNU-G by the
author but is not given in the annexures.
THE MODER NIZAT ION OF NA MES 99

In short, a name will indicate the socio-economic class and,


more accurately, the origins of a person. In urban areas, it is
this function of the name betraying the class of its bearer
that accounts for parents bestowing cosmopolitansounding,
modern names on their children. That is why there are no
Dins and Dittas in the Apna school data, while modern

s
names like Shagufta, Rabia, Anila, Sarah, Eiman and

es
Anam—all used for girls—are evident (noted by the
author in Apna-G, but not given in the annexures). Indeed,

Pr
much cultural shame is associated by the bearers of these
quintessentially rural Punjabi names, as has been noted
ity
before in the case of the maid servant Allah Ditti who
insisted upon concealing her name.
rs

There is an opposite trend also, that of seeking old-


ve

fashioned names to claim an authentic identity. This


manifests itself in three ways. First, the reduction of the
ni

authentic identity to an Islamic one and, hence show


U

regard for traditional Islamic names; secondly, to adopt


prestigious caste names or titles and thirdly, to use old
d

ethnic names to assert an ethnic identity. The desire for the


or

first is the reason why so many lists of names for children


xf

give names from the Arabic language, which were hardly


used in the early years of Pakistan. The second trend
O

results in the emphasis on mentioning one’s ‘qoum’: Syed,


Shah, Makhdoom, Rana, Chaudhary, Malik, Nawabzada,
Sahibzada and so on. Among the Sindhis and the Baloch,
tribal names are quite common such as Marri, Bugti
(Baloch); Bhutto, Rind, Paleejo (Sindhi). The Pashtuns
too use their tribal affiliations (Khattak, Afridi, Yousafzai,
100 Names: A Study

Aurakzai etc) as part of their names. It has been observed,


however, that among poor people from the same districts,
while some prestigious names such as Khan and Syed
might be used, other such names are in less frequent use as
compared to the practices in the upper and upper middle
classes. As mentioned in Chapter 4, prestige-naming, as it

s
might be called, is an upper class—especially rural upper

es
class—phenomenon.

Pr
Onomastic Ethnicization
The third corollary of ethnic naming is found among ethnic
ity
nationalists. Evidence exists that they choose morphemes
from their own languages to make modern terms (Rahman,
rs

1996: 153, 171, 177, and 207; Rahman, 1999: 263–90).


ve

The same desire impels them to go to their stock of old


names. In Khyber Pakhtunkhwa, not only the ubiquitous
ni

Gul, but also names with this component in it such as


U

Gulalai, Gulfam, Gulshan, Gulrukh, Gulmakai, etc. are also


common. The last name became famous because Malala
d

Yousafzai used it as a pen name as she describes below:


or

Hai Kakar told me it could be dangerous to use my real


xf

name and gave me the pseudonym Gul Makai, which means


O

‘cornflower’ and is the name of the heroine in a Pashtun


folk story. It’s a kind of Romeo and Juliet story in which
Gul Makai and Musa Khan meet at school and fall in love.
But they are from different tribes so their loves cause a war.
However, unlike Shakespeare’s play their story doesn’t end in
tragedy (Yousafzai, 2013: 130).
THE MODER NIZAT ION OF NA MES 101

Other specifically Pashtun names for girls are: Parkha,


Brekhna, Zarsanga, Wagma, Lema, Malghrala, Palwasha
etc. Pashto names for boys are Azmarai, Sangeen, Mirwais,
Zarghuna, Zalan, etc. Malala or Malalai for girls have
become distinctive for several historical reasons. Malala
of Maiwand, the iconic figure behind this name in the

s
Pashto speaking areas, was a Pashtun girl who is said to

es
have participated in the Second Afghan War (1878–1880).
She also wrote or recited inspiring Pashto poetry for the

Pr
fighters. There are even songs about this iconic Malalai
like the following lines attributed to Rahmat Shah Sayel
ity
of Peshawar:
rs
O Malalai of Maiwand,
Rise once more to make Pashtuns understand the song of
ve

honour,
Your poetic words turn worlds around,
ni

I beg you, rise again


U

(Quoted from Yousafzai, 2013: 11).


d

Another Malala is Malalai Joya (b. 1978), who secretly


or

educated girls during the Taliban rule in Afghanistan and


openly criticized the warlords in Afghanistan’s legislative
xf

assembly (for her autobiography see Joya, 2009). Yet


O

another young woman of that name was the Malala of


Bagram who died to avoid rape by American soldiers. And
when on 9th of October 2012, Malala Yusafzai of Swat
(b.  1997), who campaigned for girls’ right to study, was
shot in the head by the Taliban, she won the sympathy and
admiration of people all over the world. In her biography
102 Names: A Study

she tells us that ‘I was named after Malalai of Maiwand, the


greatest heroine of Afghanistan’. Later in the book she says:

In Malalai we Pashtuns have our very own Joan of Arc. Many


girl’s schools in Afghanistan are named after her. But my
grandfather, who was a religious scholar and village cleric,
didn’t like my father giving me that name. ‘It’s a sad name’,

s
he said ‘it means grief-stricken’ (Ibid.: 9–10).

es
But Malala’s father told the story of the Maiwand Joan of

Pr
Arc to all his visitors and his daughter loved hearing it and
loved ‘the way my name floated on the wind when people
ity
called it’ (Ibid.: 11).
rs
The author was told in a group discussion of names in
the University of Peshawar on 15 November 2012 that
ve

the Pashtun nationalists had started giving these ethnic


Pashtun names to their children since the last twenty
ni

five years or so. Moreover, Asfandyar Wali Khan, Head


U

of the Awami National Party, also told the author about


the presence of ‘Pashtun’ names in his party in order to
d

emphasize the Pashtun identity (Khan, PC., 2013). For


or

instance the ANP Provincial General Secretary, Tajuddin


xf

Khan’s daughter is called Gulalai and her cousins are


Gulmakai and Parkha. Yet another girl of the family is
O

called Zarghuna. Dr Sarfraz Khan, Director of the Centre


of Central Asian Studies in Peshawar University, said that
his two daughters were called Bela and Malala. But such
names were used even before Pashtun nationalism increased
their usage. For instance, Malala Yousafzai’s mother was
called Tor Pekai (black tresses) even though she probably
THE MODER NIZAT ION OF NA MES 103

did not belong to a politically conscious family (Yousafzai,


2013: 15). However, the increase in such names in identity
conscious circles is certainly part of the assertion of the
Pashtun ethnic identity which the Pashtun nationalists
supported (Khan. Int., 2012). Similarly, Jamal Shah, a
Pashtun nationalist who wrote under the pseudonym of

s
Baacha Khan (being a follower of Khan Ghaffar Khan

es
who was called ‘Baacha’=king) named his daughters Zuro
Sanga (meaning a golden branch), Meena Gabeena (which

Pr
could be translated as honey-sweet love), Tatara Goloona
(the flowers of Tatara, a mountain near the legendary
ity
Khyber Pass overlooking the city of Peshawar) and Sandra
Bakhtawara (a song and one with good luck)’. His son
rs
was called Batoor (translated as brave as well as eagle) and
ve

this tradition was continued with the grandchildren also


(Gulmakai and Breshna) (Yusufzai, 2013).
ni

It is reported that Baloch nationalists choose names like


U

Guhram, (in Annexure Poor-F & M this name occurs


17 times), Chakar (33 times in the above Annexure),
d

Balach (23 times) for boys and Zabud, Hani, and Rajma
or

for girls occur the most often. This is confirmed by the


xf

Baloch academics the author interviewed (Buzdar Int.,


2012; Baloch Int., 2012). In the Siraiki belt, where there
O

has been a demand for a separate province for about forty


years (Rahman, 1996: 173–190), this feeling of ethnic
nationalism has affected names also. According to Mushtaq
Gadi, a lecturer in Quaid-i-Azam University who is from
the Siraiki belt and is an ethnic activist, some activists went
back to pre-Islamic, Siraiki names because they rejected
104 Names: A Study

the Islamic identity with Arabic names that is imposed by


the state. The names they took up include: Rawal, Punnu,
Sajawal (24), Chandan, Sobha, Jeevan, Sachal, Rohal,
Momal, Seemal, Sancha, Soha, Sassi, Sawal (16), Sojha, etc.
(Poor- F & M). This too is confirmed by Mushtaq Gadi
mentioned above (Gadi, PC., 2012).

s
The most interesting development is in Sindh where the

es
name Dahir has been used as an assertion of Sindhi ethnic

Pr
identity, especially since the Sindhi nationalist leader G. M.
Syed declared that Dahir was a son of the soil and defender
of Sindh. In his deposition to the court he said in 1973:
ity
[The consciousness of the autonomy of Sindh] has emerged
rs
from time to time—sometimes in the shape of Raja Dahir,
ve

sometimes in the person of Dodo Soomro, sometimes in the


shape of Darya Khan and Makhdum Bilawal and Shah Hyder
ni

Sannai. It has expressed itself in the love and courage of Shah


Enayat (Syed, 2012: 8).
U

Dahir, it may be remembered, was the ruler of Sindh who


d

was defeated by Mohammad Bin Qasim (695–715) in 715.


or

Syed declared that Muhammad bin Qasim ‘was a barbaric


xf

and cruel ruler who came to Sindh as representative’ of the


notorious tyrant Hajjaj bin Yusuf (Ibid.: 205). In contrast,
O

he claims, ‘Raja Dahir is respected as a nationalist, as a


hero of Sindh, but the imperialists call Raja Dahir an
infidel. But Raja Dahir was a great man’ (Syed n.d.: 108).
He asserts that Dahir’s government was ‘free, peaceful,
well-organized, prosperous and civilized’ (Syed 2012: 205).
For Pakistani nationalists and Islamists since Dahir was a
THE MODER NIZAT ION OF NA MES 105

non-Muslim he does not deserve to be even remembered


let alone celebrated. It is precisely for this reason that
Sindhi nationalists take an opposing view and celebrate him
as a son-of-the-soil who resisted an outsider (23 names in
Poor-M). It should, however, be kept in mind that Dahiri as
well as Dahraj are castes or tribal labels in Sindh (Somroo.

s
Int., 2013). Other names which are used in Sindh are

es
Roopho, Hoshee, Dodo (16), Doda (36), Chanesar (16),
Bilawal (74), Sobo (07), Sojho, Marvi (53), Bakhtawar

Pr
(160), Leelan (27), Bagal (06), Meeran (18), Kilar, Kando
(17) and Bagh [Poor-M & F]. ity
The Soomras ruled Sindh between 1298 and 1300. One
rs
of this line of rulers, Dodo Bin Khafif Soomro III, is the
hero of an epic poem called Dodo-Chanesar by the poet
ve

Bhago Khan. The gist of it is that he died defending


his land against Sultan Alauddin Khilji’s (r. 1296–1316)
ni

army, which his brother, Kamaluddin Chanesar initially


U

supported but eventually also fought till death. This bit of


medieval history was considered highly inspiring by many
d

Sindhis. This is evident from the appreciative comments


or

and popularity of a quite succinct narration of the story


xf

that was recently published in a newspaper (Rashid, 2012).


Similarly, the story of Marvi the girl from Thar who died
O

resisting the advances of her abductor, the Soomra prince


Umar, elicits comments from Sindhi nationalists in blogs
and other internet sources suggesting her iconic value.

In Sindh caste names are used much more commonly


than in the Punjab, KPK or among the Urdu-speaking
community. Almost all middle class people, and certainly
106 Names: A Study

all of the higher classes, write their caste names. That


is why the percentage of such names is high in the data
from the Board of Intermediate and Secondary Education
from Larkana (2012) (Sindh-M 1990s, Annexure-I).
However, since there is a hierarchy of castes such names
are sometimes subject to change. According to Badruddin

s
Somroo, a professor of Mass Communication as well as a

es
Sindhi nationalist, when the Mohanas and Mirbahar (two
old tribes) became prosperous and socially prominent they

Pr
started calling themselves Mallah. The Maachis (originally
fishermen) became Solangis; Bhurgris become Talpurs
ity
and Kalhoros become Abbasis (Arabs). In short, ethnic
nationalist identity is also expressed through iconic names.
rs
He also said that Manganhars call themselves Pardesis
ve

(outsiders i.e those who have come from another land).


Those who claim foreign descent, of course, are proud of
ni

it as they are all over Muslim South Asia, so much so that


Syeds, even if they are leaders of Sindhi nationalists like
U

G. M. Syed, proclaim their Arab (Syed) origin and the


d

Pashtuns settled in Sindh call themselves Pathans, which


or

incidentally, is not a name used by the Pashto speakers who


call themselves Pashtun, Pukhtana, or Afghans. But the
xf

Sindhis I interviewed claimed that the name Pathan is also


O

used as a first name if a male child has a fair complexion


(Somroo. Int., 2013; Pahore. Int., 2013). Likewise a fair
girl can be called Rani, Soni, and Soomri in Sindh. These
names may be used as pet names as well as formal personal
names.
THE MODER NIZAT ION OF NA MES 107

Pakistani names, while appearing to be throwbacks to


the past in cases other than British patterns of naming,
are actually a product of the way modernity brought in
versions of the Islamic and ethnic identities into play. Thus
personal names are indexed to forces of multiple identities
and not necessarily to a Westernized model alone. In

s
short, the process of modernization is not influenced only

es
by Western norms and onomastic fashions. In addition to
these influences, it is also influenced by the processes of

Pr
urbanization, ethnicity, and contemporary interpretations
of Islam. In short, it is a product of the way modernity has
ity
been experienced in Pakistan.
rs
ve
ni
U
d
or
xf
O
7
Name-Changing: Names as Traps

s
es
Changing one’s name is a much commoner phenomenon
than most people imagine. Not only do many women

Pr
change their name on marriage, especially in Western
countries, as has already been mentioned, but people of all
ity
sexes change name either partially or wholly for many other
reasons. The most common one is the indexation of the
rs
name with a certain desiderated identity or the concealment
ve

of a problematic identity. The most extreme case of this


was the taking of male names by female members of the
ni

ruling family of Yemen before it became a republic. In


U

the perception of their society, the feminine name ‘is a


medium enabling contact with her through establishing
d

a bodily image’. So assuming male names ‘are defensive


or

devices’ used by these women to protect them from the


libidinous imagination of males (Bruck, 2006: 228). This
xf

does not happen in Pakistan but women are not called by


O

their names among the conservative parts of the country


as explained earlier.

People who have the power to name themselves construct


a desirable identity through this privilege. In Japan, for
instance, names were changed to maintain certain power
relations within the family (Nagata, 2012). In South Asia,
NA ME-CH A NGING: NA MES AS TR A PS 109

however, the most obvious cases have traditionally been


of powerful people such as kings, princes, great lords,
and poets who adopted new ego inflating names. Mughal
emperors, for instance, adopted a title when they ascended
the throne: Prince Salim became Jahangir (Subduer of
the World); his son Khurram became Shahjahan (King

s
of the World) and so on. And adopting poetic pen names

es
(takhallus) is an established tradition in Urdu poetry.
Sometimes the takhallus becomes so well known, as in the

Pr
case of the Urdu poet Ghalib (1797–1869), that the real
name Mirza Asadullah Khan is not used. The takhallus too
ity
can be changed. For instance, Ghalib’s first takhallus was
‘Asad’ but he changed it to Ghalib. Khwaja Meer Dard
rs
(1721–1785), another classical master of the ghazal, had
ve

the takhallus ‘Meer’ but changed it to Dard when another


famous master of the ghazal, Meer Taqi also adopted the
ni

former pen name. Sometimes the poets own name could


function as a takhallus, for instance in the cases of Iqbal
U

(1723–1810) and Faiz Ahmed Faiz (1911–1984) whose


d

names were Mohammad Iqbal and Faiz Ahmed Khan.


or

In some cases, especially in the film industry, names were


xf

changed in order to fit in better with a certain role. For


instance when a Muslim youth Ali Meer had to play a
O

Hindu mythical hero called Puran Bhagat, he was given the


name of Kumar (Manto, 1991: 415). In any case, during the
1930s–40s Muslim actors were given Hindu names because
Bollywood was predominantly Hindu. Thus the filmstar
Yousuf Khan became Dilip Kumar; Mumtaz Jahan Begum
changed her name to Madhubala; Mahjabeen Ara Kamal
110 Names: A Study

became Meena Kumari (Wright, 2006: 6); while the actress


Shahida became famous as Naina (Manto 1991: 415).
However, this trend has been reversed now with Shahrukh
Khan, Salman Khan, Amir Khan, and Saif Ali Khan, who
made the name Khan iconic in Bollywood. This trend also
existed in Pakistan. For instance, the filmstar Sudheer’s real

s
name was Zaman Khan. Hasan Askari, the Director of one

es
of the films in which he Sudheer acted, told the author that
the actor had taken up this name because his family did

Pr
not want him to join the film industry and he thought the
name would protect him from his family finding out about
ity
this, since they did not watch films (Askari. Int., 2013).
rs
While the takhallus is always in addition to the given
name in the case of poets, an ordinary person may change
ve

his name completely for other reasons. One reason, as


mentioned earlier, is that there is a belief that certain
ni

names are ‘heavy’ (bhari), inauspicious or ominous for the


U

bearer (for a similar belief in another pre-modern society


see Barnes, 1980: 314). In the case of Mohammad Waseem,
d

a well known Pakistani academic, his first given name was


or

Parvez. However, as he was a sickly child and one elder


xf

brother of his had died earlier, his mother changed it to


Waseem hoping that thereby the fate that gave him ill-
O

health would alter (Waseem. Int., 2012). Nawaz Attari, a


peon in an office in Lahore, said that names are changed
if the husband and wife do not live in peace. It is assumed
that the names are not compatible for marital harmony
and therefore they are changed (Attari. Int., 2012). He
also said that certain names, such as Bilal, are considered
NA ME-CH A NGING: NA MES AS TR A PS 111

‘heavy’ (Ibid.). Such superstitions are often criticized by


religious reformers. For instance, Syed Ahmed Barelvi when
preaching in Saharanpur in the early part of the nineteenth
century said sometimes if a child dies the parents do not
give its name to a second offspring they may have, fearing
that then, likewise this child may die too. He preached

s
that such superstitions should be given up (Nadvi n.d.:

es
45). In some parts of Pakistan unusual name changing
fashions persist. For instance, a case which is opposite to

Pr
the above superstition, in a prominent Gardezi Syed family
of Multan, a child’s name is changed to someone who has
ity
died in the family (Gardezi. Int., 2012). This, however, is
not the norm in other parts of Pakistan.
rs

Yet another name changing practice is that of adopting


ve

neutral sounding names among immigrants such as


Hispanics (Ragone, 2012), Gypsies (Silverman, 1991)
ni

and Iranians (Blair, 1991) in the United States. However,


U

sometimes the Anglicized ethnic surnames of immigrants


are the mistakes made by petty immigration officials
d

when these people entered the United States (Baird,


or

2006). This behaviour—that of taking Anglicized names


xf

and nicknames—is also exhibited by Pakistanis settled in


Western countries. For instance, a friend of the author,
O

Tariq Ahsan, settled in Canada has a daughter whom the


parents wanted to name Qurratul ‘Ain. However, when he
consulted English speaking Canadians he found that they
could hardly pronounce the name as it was too unwieldy for
them. It would also be difficult to write and would certainly
mark the child out as an outsider. So they chose the name
112 Names: A Study

Shehrezad—of the Arabian Night’s fame—and normally


used its abbreviation, Sherry (Ahsan. Int., 2012). Other
names often chosen by the Pakistani community are Maria,
Laila, Marina for girls and Haris (pronounced/Haeris/)
and Sameer for boys. However, since the community does
not want to give up its identity altogether, it does not give

s
typically Western names to children either. In a novel about

es
a Pakistani man who marries a Swedish-American wife
the son is called Homer. When this news is conveyed to

Pr
his elder brother’s wife she suggests a number of Muslim
Pakistani names questioning the Western name and
ity
equating it with ‘hammer’ (Manzar, 2008: 343). In real
life the present author has not heard of purely Western
rs
names being given to children though such nicknames are
ve

in regular use. The only exceptions are the names given


by some Pakistani liberals who sometimes choose Russian
ni

names (Tania, Dushka, Sonia) or Hindu ones (Usha, Amar)


which have either been indigenized as a modern fashion or
U

are taken as alternatives of such Muslim names as Ayesha


d

and Amr (bin al-Aas).


or

In addition to that, Westernized Pakistanis also choose


xf

Western sounding nicknames to facilitate merging in with


the mainstream community. These may be John, Nick,
O

Mary or Jill or Westernized abbreviations of their own


names: Kamal (Kaemel), Shahzad (Shez), Qurrat ul Ain
(Annie), etc. Indeed, most Annies were Arabic ‘ains (eyes);
Sherrys were Shahr Banos, Shehrezad, or Shahwar, etc. This
Westernized elite is described in the novels of the British
period and the early years of Pakistan (Farooqi, 1968: 9).
NA ME-CH A NGING: NA MES AS TR A PS 113

The same trend can be observed among the workers of


the call centres of Pakistan, as mentioned earlier, who take
up Westernized nicknames to sound American or British
(Rahman, 2009: 236). The only example of the opposite
trend—insisting upon a deliberately difficult indigenous
name as Lehiste describes (1975)—is mentioned in an

s
Urdu novel in which the names of the daughters are Miftah

es
and Zam Hareer and the son is Zabar Jad. The Pakistani
parents, who are living in Scotland, explain that they kept

Pr
these names because their Scottish acquaintances were
under the impression that Pakistanis had only a restricted
ity
set of names. The wife, however, has the same prejudice
about Scottish names, as the Scots have about the Pakistani
rs
ones and ridicules the repetition of John, Jim, Mike for
ve

boys and Anne and Mary for girls. In her own words her
children had been named deliberately to ‘twist the tongues
ni

of the British’ (Manzar, 2008: 402).


U

Onomastic destigmatization—a term used by Moa Bursell


(2011) precisely to denote name-changing among immi­
d

grants to Sweden—strategies such as the ones described


or

above for Pakistanis are, of course, common in similar


xf

situations all over the world. The most painful labels for
immigrants are ‘ethnophaulisms’ which are defined as
O

words ‘to deprecate a group of people’ (Nuessel, 2008: 29).


Pakistani children are sometimes simply called ‘Paki’ in
Britain thus robbing them of their names and personality.
Richard Johnson, an Anglo-Indian immigrant to Australia,
having seen children with different names being bullied in
schools, describes the naming of his son as follows:
114 Names: A Study

I voted for a name that was easy to pronounce and spell.


We also wanted our choice to be meaningful. My wife and I
shared excellent memories of a Catholic nun in our days at
school—Sister Kevin. So, our son was called Kevin ( Johnson,
2010: 123).

This practice is recorded among Muslims in India to mask

s
their identity in order to escape physical danger. During

es
the days of the partition of India—the late 1940s—the
well-known novelist of Pakistani origin, Zulfikar Ghose

Pr
(b. 1935), changed his family name of Ghaus which
immediately identified him as a Muslim. This is how
ity
Ghose explains his change of name.
rs
Ghose is a common Hindu name familiar to anyone who has
ve

been to Calcutta, while we are Muslims. But the India before


independence was a time of communal hatred; we found it
ni

convenient to be known as Ghose among Hindu communities


and Khawaja among Muslims. … And I prefer it. It is half
U

Muslim, half Hindu, half Pakistani, half Indian (Ghose, 1965:


139).
d
or

Nor did such situations stop arising after the partition.


During the riots in Mumbai in 1992 the victims’ (mostly
xf

Muslim) ‘names were picked out from apartments’ and


O

voters’ lists’ (Wright, 2006: 8). Thus some Muslims,


especially in the Gujrat area used Hindu sounding
surnames to avoid such situations. The Muslim youth from
North Africa involved in rioting in French cities in 2006
felt ‘penalized by their poverty, the colour of their skin and
their names’ according to a French security person ( Jessel,
NA ME-CH A NGING: NA MES AS TR A PS 115

2006). Nor were Muslims the only ones to feel that their
names were traps. After Indira Gandhi’s murder by her
Sikh guard ‘In Delhi, the names of Sikhs were taken out
from ration card lists by area ration shop owners’ (evidence
of a Sikh man quoted by Rumi, 2013: 67). And, of course,
the cases of minorities being killed on the basis of their

s
names has been recorded in detail elsewhere.

es
Sometimes pseudonyms have to be used because of the

Pr
social opprobrium attached to giving one’s real name in
public. Women often had to face this problem in South
Asia and had to use fictitious male names or hide behind
ity
their men folks’ names (Mrs Abdul Qadir for instance). A
rs
recent example is of Malala Yousafzai who had to use the
name Gul Makai when she started writing a diary because
ve

of the fear of the Taliban in Swat (Yousafzai, 2013: 130).


ni

In the coastal South and West of India, old Hindu sur­


names survived. According to Theodore Wright these—
U

Saradgi (Karnataka), Antulay (Maharashtra), Kutty (Kerala)


d

and Pookunhikoya (Lakshwadeep)—remain because ‘Islam


or

arrived peacefully through Arab merchants even before the


Turko Afghan invasions of the North’ (Wright, 2006: 6).
xf

This adoption of names for business purposes or to fit in


O

better with the majority community parallels the similar


trend of adopting Hindu names in Bollywood mentioned
earlier.

Such behaviour has been shown by other minority com­


munities also. For instance, during the 1940s and 50s, ‘some
50,000 Americans filed petitions with state courts each year
116 Names: A Study

seeking permission to change their family names; 80 per


cent of them were Jews’ (Silberman, 1985: 59). But other
communities, such as the Poles, also changed or simplified
their names when they immigrated to North America
(Klymasz, 1963). And, of course, immigrants of nearly all
communities use Western sounding nicknames to adjust

s
better in Western societies.

es
The Hijras of South Asia—including eunuchs, hermaphro­

Pr
dites, transvestites, and homosexuals—who wear women’s
clothing, also change their names ‘on being initiated into
the community’ (Lal, 1999: 129). Unfortunately, despite
ity
detailed descriptions of the Hijra way of life in Muslim
rs
societies such as Bangladesh (Hossain, 2012) and Pakistan
(Pamment, 2010), none of these authors has mentioned
ve

anything specific about the naming practices of Hijras.


However, field research by the present author suggests,
ni

that they do not reveal their earlier name, a male name,


U

to strangers. Three Hijras whom the author interviewed


in Rawalpindi gave their present names—Reema, Saima,
d

Meera—which were all modern, urban female names, but


or

not their original ones. They said that their new names
xf

were given to them by their spiritual leader, the Guru, and


that they did not remember their earlier names (Hijras.
O

Int., 2012). Claire Pamment names many prominent Hijras


of Pakistan and, apart from a few like Mohammad Aslam,
others are known as Bobby, Moti and Sanam, which are
feminine names. In the elections of 2013 there were hijra
candidates named Resham and Bindia, both female names.
NA ME-CH A NGING: NA MES AS TR A PS 117

However, surprisingly the latter’s Rajput caste, Rana, was


attached to the name.

‘Neutral’-naming among religious


minorities
Religious minorities are increasingly being subjected to

s
persecution or, at least, negative discrimination in Pakistan

es
(Salim, 2006; Rahman, 2012). Thus it is only to be expected
that they would adopt name-changing as a destigmatization

Pr
strategy. Rana Irfan, Manager Training in the City School
Head office, a Christian, told the author in an interview
ity
that their family always had Muslim, Christian, and
even Hindu names—the last because Sir Chohotu Ram
rs

(1881–1945), a pre-partition political figure, was their


ve

relative. Her own family has neutral names—Rana, Reena,


Naveed, Faheem, Amir—her own son is called Gibran
ni

and daughters Anam and Shayan. These names are shared


U

by Muslims and Christians. However, since middle class


families come from clans such as the Gills (Kashmiris) or
d

Bhattis (Rajputs), they retain these names from the past


or

and have not adopted them recently (Irfan. Int., 2012).


xf

John Alexander Malik, the Bishop emeritus of Lahore, told


the author that all his children had Muslim names and
O

even his wife had such a name. However, even in the odler
generation, he said, there were people with Muslim names
such as his own mother. He concluded that names shared
with Muslims have increased in the Christian community
(Malik. Int., 2014).
118 Names: A Study

Sarfraz Anwar, a Christian social worker, who was inter­


viewed on the telephone, said that there were two con­
tradictory trends in the Christian community. First, the
name Masih (Messiah) had decreased. However, this was
understandable since it was associated with the sweeper
community and was shunned by upwardly mobile Chris­

s
tians. He also said that Biblical naming had increased.

es
However, the examples he gave—Sara, Ilyas, and Joad—are
either in use among Muslims or sound like Muslim names

Pr
( Joad=Jawad). In his own case, his name Kenneth Naveed
Anwar, was changed to only Naveed Anwar which is shared
ity
with Muslims ‘to facilitate schooling’ (Anwar. Int., 2012).
rs
Confirmation of some of these trends comes from a sample
of 4,423 names of educated, middle class people from the
ve

Christian community in Lahore between the ages of 21


and 59 compared with a sample of 282 children from the
ni

same community between the ages of 4 to 19. The results


U

are as follows:
d

Table 13
or

Between 21–59 Between 4–19


Description of name
xf

N = 4,423 N = 282
All components of name shared
O

1,977 (44.70 %) 120 (42.55 %)


with Muslims
With at least one component
1,385 (31.31 %) 148 (52.48 %)
shared with Muslims
Names with Masih 244 (5. 52 %) 01 (0.35 %)
All components Christian 817 (18.47 %) 13 (4.61 %)
NA ME-CH A NGING: NA MES AS TR A PS 119

In short, the use of Masih has decreased among the


educated, middle class in the younger generation though,
of course, the working classes and the rural areas still retain
this Christian identity marker. Moreover, most Christians
now use nominal components shared with Muslims.
Indeed, those who use purely Christian names (i.e., no

s
component is shared with Muslims) have decreased from

es
18.47 to 4.61 per cent. Those using at least one Muslim
name, which can be used at certain places or with Muslim

Pr
companions, were 76. 01 per cent in the older generation
but have now increased to 95.03 per cent (figures obtained
ity
by adding the first and second rows). Interviews bear this
out in poignant detail.
rs

For instance, two janitors working in the Quaid-i-Azam


ve

University at Islamabad were more open about the


discrimination they faced when they revealed their real
ni

names, Tahir Masih and Shamoon (Simeon) Masih. For


U

instance, they were frequently stopped by the police and


interrogated. That is why, even without any formal change
d

of names in their identity cards, both use the surname


or

Nazeer (Tahir and Shamoon. Int., 2012).


xf

As for Pakistani Hindus, they are mostly residents of Sindh,


O

and are reconizable by their distinctively Hindu names


(Hari Ram, Ashok Kumar, Pooja Devi, etc.). According
to Harbaksh Makijani, an academic from this community,
Hindu children receive their names after a ceremony in
which the Brahmin priest calculates the horoscope (kundli)
of the child. This name is traditional, and the names which
they use and write down in documents are informal names.
120 Names: A Study

These are called ‘calling names’. When it was pointed


out to him that his own name had an Arabic component
(Baksh), he said that this was used by the Sikh community
in India and had been accepted as a traditional component
among Sindhi Hindus as well. He did mention, however,
that there was discrimination against Hindus so people

s
sometimes used names shared with Muslims.

es
However, in his opinion the process of globalization was

Pr
changing names so that Neha was used both by Muslims
and Hindus (Makijani. Int., 2013). Such kind of names,
which can be called neutral names, were observed in the
ity
data from the districts of Sindh where the Hindus form
rs
clusters. For instance, in the data from Sindh comprising
360 names of Hindu rural, working class women, there were
ve

at least ten names shared with their Muslim neighbours.


Among these Meena (10 names) was a favourite, followed
ni

by Raj Bai (6) and Meeran (5). The name Beena (3) is
U

also a shared one, as are Veena, Seemi, Anita, Soni, Heer


and the more modern sounding Neha (Badeen-F). In the
d

data for students of matriculation from Larkana (Sindh-M


or

and F 1990s) the religion is mostly given in the case of


xf

Hindus and sometimes also for Muslims. Thus a name


which could be shared by both Muslims and Hindus such
O

as Reeha would be marked as ‘Reeha Bhojwani, Hindu’


in the case of a Hindu girl. There are a few examples of
shared names besides the favourite Reeha. These are: Hina,
Kiran and Kanwal but, as mentioned before, in all cases the
word Hindu is attached in the matriculation gazettes. The
presence of ‘calling names’ in the Hindu community was
NA ME-CH A NGING: NA MES AS TR A PS 121

also testified by Haroon Khalid who has interviewed many


Hindus in the course of his research on the minorities of
Pakistan. In his book, A White Trail (2013), which is the
outcome of his research, he writes about a Hindu from
Lahore whose ‘family and friends have adopted dual names
to dilute their religious identities ‘ (Khalid, 2013: 11). Later,

s
in face-to-face interaction he confirmed to the author that

es
a number of Hindus did share names with Muslims and
even got official documents with such names. One such

Pr
person is Raza, a practicing Hindu, who traces back his
Hinduization to his friends’ Muslimization during Zia
ity
ul-Haq’s rule (Ibid.: 24). However, they also had parallel
purely traditional Hindu names as well (Khalid. PC., 2013).
rs
Another educated young man from the Hindu community
ve

of Sindh, called Pirbhu Satyani, told the author that he


hired a flat with the help of his Muslim friend since people
ni

asked awkward questions when they saw his Hindu name


on his identity card. He said that some people did take
U

up Muslim sounding names just to get along (Satyani.


d

Int., 2014). This was further confirmed by Dr Munawwar


or

Chan, Chairman of the Pakistan Hindu Rabita Council,


whose own first name ‘Munawwar’ (radiant) is shared
xf

by Pakistani Muslims (Chan. Int., 2014). Indeed, in his


O

case the name ‘Chan’ (moon) is also shared by Muslims


and used to be common in rural areas of the Punjab and
Sindh. In Bangladesh too Hindus had to take up Muslim
names in order to avoid discrimination. During the military
action in 1971 a Hindu physician concealed his real name,
Sudhamoy, and assumed the Muslim name Shirajuddin
122 Names: A Study

Hussain (Nasrin, 1993: 10). This, indeed, is the fate of all


bearers of names caught in situations of conflict everywhere.

Some people, generally those with a developed social


conscience and having the security of belonging to the
educated middle class, deliberately adopt composite Hindu-
Muslim names in order to advertise their support for a

s
composite non-communal identity. One such Indian man is

es
Inder Salim ‘who bears a composite, self-constructed name’

Pr
a musician from Kashmir whom Raza Rumi, the Pakistani
intellectual, met in Delhi (Rumi, 2013: 160). In Pakistan
too there is Preetam Giani, originally named Mahboob
ity
Ghani by his parents, in Abbottabad (Giani. PC., 2013).
rs
And, of course, the legendary poet and writer of songs for
Hindi movies, Gulzar, is Sampooran Singh but prefers this
ve

Persian-Urdu name to show his liking and support for Urdu


poetry and a composite culture. Another such example is
ni

Dr Shreekant Gupta, a supporter of Pakistan in his writings,


U

who, according to Raza Rumi ‘allows me to call him “Ali”


(a testament to his composite Delhi identity)’ (Rumi, 2013:
d

286). But such self-naming is iconic since these individuals


or

want to use their names as a statement against a narrow-


xf

minded, communally oriented society. Some such names


are also given by parents for similar reasons. Hence, writes
O

Supriya Pattanayak, ‘A Hindu friend named his son and


daughter Aftab “sun” and Mahtab “moon”. This kind of
cultural integration is not acceptable to religious fanatics’
(Pattanayak, 2010: 69). That it is also not acceptable in
Pakistan is brought out by a report by the International
Dalit Solidarity Network. The report is about the caste,
NA ME-CH A NGING: NA MES AS TR A PS 123

class and religious discrimination against the lower caste


(Dalit) Hindu community in Sindh and lower Punjab. A
woman of this community called Alya Oad got a Bachelor
of Business Administration (BBA) degree and went to
Karachi where she found a job. When her colleagues found
out that she was not a Muslim, they turned against her. The

s
treatment meted out to her is summed up as follows:

es
“You are cheating people with your name. Why have you got

Pr
a name as a Muslim girl? You are a low caste Hindu, and we
didn’t know that”, said a senior colleague. Then it got worse:
“You are so beautiful because your mother slept with a Muslim
ity
man”, read a note on her desk. Oad wrote her resignation on
the reverse and left the office’ (Dalit, 2000: 6).
rs
ve

As Harbaksh Makijani revealed, he and his cohort were


embarrassed by their Hindu names in their youth. Now,
ni

according to him, Pakistanis have become so exposed


to Hindu names in Bollywood films and the Indian TV
U

dramas that some names are no longer strange for them


d

(Makijani. Int., 2013). Yet, as the experience of Alya Oad


or

above testifies, the quest for not being onomastically


marked can lead a person from the Hindu community of
xf

Pakistan into serious trouble. Moreover, it is only in the


O

urban elite that a few names used in Indian Hindi films


are shared with Pakistani Muslims and this elite also uses
some Western names. This is the exception not the rule.

The Parsi community too had distinctive Persian names to


which they added their father’s name before the modern
era. Nowadays their names reflect the influence of their
124 Names: A Study

migration to Gujerat (also spelled Gujarat and Gujrat)


as well as their participation in business and trade during
British rule. According to Eckehard Kulke

The new surnames are partially of Gujerati and partially


of English origin: Unwala (wool dealer), Kapadia (cloth
merchant), Merchant, Jhaveri (jeweler), Motivala (pearl

s
dealer), Biscuitwala, Readymoney, Paymaster, Shroff (cashier),

es
Batliwala (sale of bottles), Ginwala, Sodawaterwala. Mondawala
(wine dealer), Dubash (shipping agent), Davar (judge), Master,

Pr
Mehta (teacher), Dastur (priest), Commissariat, Doctor,
Engineer, Contractor etc. (Kulke, 1974: 51, Note 11).
ity
Some names, being of Persian origin, such as Rustam,
rs
Parvez, Shahryar, etc., are shared with Muslims. Even some
occupational surnames such as Mulla (Muslim clergyman)
ve

are shared though here it means a person with skills in


ni

writing and reading Persian. Moreover, when the Parsis


served Muslim rulers they were given the same titles as
U

Muslim noblemen such as Sir Fareedun ul-Mulk (Prime


d

Minister of Hyderabad state under the Nizam) and so on.


or

Even now in present-day Pakistan, though the surnames


are distinctive, many Parsis share first names with Muslims.
xf

Parsi women share names like Sadaf and Farah with their
O

Muslim sisters (Khalid, 2013: 183). The Sikhs attach


the suffix ‘Singh’ with their names but some first names
such as Iqbal are shared with Muslim neighbours and the
author was told of a Sikh lady called Amina Kaur, though
Amina is a typically traditional Muslim female name.
Unfortunately, this information could not be confirmed.
However, there are Sikhs named Iqbal, Boota, Ghaseeta,
NA ME-CH A NGING: NA MES AS TR A PS 125

Kala, Jarnail, Patthar, Chattan, Sher and so on, all which


names are shared with Muslims. According to Inderjit
Singh, an academic in Canada, ‘traditionally Sikh names
were drawn from both Persian and Sanskrit sources. I grew
up with names like Iqbal Kaur, Resham Singh, Attar Singh,
etc. My father was Kesar Singh and one of his sisters was

s
Kesar Kaur’. However, he added that the names of the

es
present generation are from Sanskritic roots. But this is the
case in India (Singh, PC., 2013). In Pakistan some such

Pr
names are still used, though of course, Singh and Kaur are
always attached to signal the Sikh identity. Whether the
ity
shared names are actually used in times of danger, and even
to determine their percentage in regard to overall usage,
rs
needs further research.
ve

To sum up, the point is that Muslim (or ‘neutral’) names


are used by religious minorities as a destigmatizing strategy
ni

in order to conceal their religious identity because revealing


U

it could result in discrimination which they tend to avoid.


It should be added that this kind of naming strategy is
d

practiced by all communities who lack power in a certain


or

situation. The intention of the weaker community is not


xf

to deceive the stronger one, but to be as safe as possible


under the circumstances. Indeed, in a world without
O

discrimination, without danger, without malice there would


be no need for such strategies except for fitting in better
or expressing solidarity with one’s companions and friends.
But Pakistan is surely not that kind of world yet.
8
Conclusion

s
es
We have come a long way from the rhetorical question:
‘What’s in a name?’. We now know that it may be life or

Pr
death; safety or peril; the battering or the boosting of ego
and one’s self-respect as well as the image one wants to
ity
present to the world—that it is an important technique
for the ‘technology of the self ’. One way of looking at
rs
names is through the ideas of John Austin in How to
ve

do Things with Words (1962), Ludwig Wittgenstein’s


Philosophical Investigations (1953) and John Searle’s Speech
ni

Acts (1969). Austin says statements are not only true or


U

false they also make changes in the world. These he calls


‘performatives’ (ibid.: 6). He then distinguishes between
d

locutionary, illocutionary and perlocutionary functions.


or

The first refers to the ‘act of saying something’ in certain


situations like phrases ‘I Christian, thee XYZ’ or ‘The
xf

court is adjourned’. However, these function if there are no


O

‘infelicities’ attendant upon the utterance of these phrases


(Ibid.: 14). If, for instance, the person uttering them does
not have the authority to do so, i.e., bestow a name upon
a Christian baby—suppose he is only pretending to be a
priest—the condition of felicity does not obtain and the
act does not take place. Once everything is right these
CONCLUSION 127

phrases have illocutionary force (Ibid.: 98). But if there are


unintended consequences—say the child is from a Muslim
family and his family picks up a fight with the person who
pretended to be a priest—that is the perlocutionary force
of the utterance. It is in the last in which we are interested.
Austin’s definition is, therefore, higly relevant for us:

s
es
Saying something will often, or even normally, produce
certain consequential effects upon the feelings, thoughts, or

Pr
actions of the audience, or of the speaker, or of other persons
(Ibid.: 101).
ity
Searle elaborates upon ‘illocutionary acts’ (Searle, 1969: 54–
71) pointing out that the illocutionary cannot be reduced
rs
to the perlocutionary (Ibid.: 71). The relevant argument
ve

for our purposes here is that both the set classified as


performatives and other kinds of statements, with truth
ni

value or not, can have consequential effects and generally


U

do. For instance a message on the phone that one’s son has
got admission in Oxford University has a visible effect on
d

the parents, but it does not actually do anything; it merely


or

reports an event which has taken place. So, the entry point
to understanding the perlocutionary force of names is not
xf

to attempt to classify them in any of the categories created


O

by Austin.

Names have an ‘ontological weight’ as Pina-Cabral


postulates (2010), but this is given in ways which are
distinct from other kind of statements. It appears to me
that names are different from all existing statements—
128 Names: A Study

performatives and declaratives, etc.—in that they state


nothing, perform nothing, have no truth value in the
normal sense of the term, but yet they do things; they
have consequential effects. The point is that as soon as any
name is uttered it has effects on the audience because it
exercises its power upon listeners. While other utterances,

s
composed of sequences of sounds denoting things being

es
done (performatives) or told (declaratives), function as a
class. Personal names are units of power (or lack of it) on

Pr
their own. Thus the same name can have different effects.
The name Ali can be of one’s boss, one’s son, one’s servant,
ity
a despised enemy, one’s childhood friend and, of course, the
fourth Caliph in Islamic history. In each case the response
rs
to the presence of the owner of the name or reference to
ve

him will have an effect upon the hearer which differs from
case to case. Each of the possessors of the name, and of all
ni

names, are indexed to an identity, a history of usage and has


a habitus and symbolic capital (which may be negative, i.e.,
U

baggage) which makes the hearer respond differently. Only


d

in some cases and situations, however, personal names may


or

function as a class. Take the case of the Hindu-Muslim riots


of 1947. At that time just having the name Ram Chand
xf

(Hindu name) was enough to get a person killed in Lahore,


O

and having the name Ahmad Khan (Muslim name), killed


in Amritsar. In such circumstances names index an identity
and their very utterance has the unintended force of getting
their bearers in trouble. Similarly, having the names Bibi,
Mai, Allah Ditti, Reshman, or Bismillah Jan signal a rural
or at least an old-fashioned identity. Also components like
CONCLUSION 129

Syed, Sardar, Wadera, Jam etc., act upon hearers as if they


denoted a class identity for the bearer. But in such cases
too, it is only if the bearer is in a group or one does not
know him that such names index a group identity and have
perlocutionary force of some kind. If the person is known,
then individual characteristics may become stronger than

s
class ones or, even if they do not, they do at least modify

es
one’s reaction to the name.

Pr
One may hypothesize that the listeners have a map of the
world which carries a power grid and that is where the
name fits in. If it is of a less powerful person it causes no
ity
sense of awe, reverence, admiration, or fear. If it is of an
rs
equal it may cause happiness or dismay depending upon
one’s relations with that person. If of a beloved it may
ve

cause feelings of romance or lust; and in the case of an old


companion or spouse, a sense of ease, familiarity, affection
ni

and so on. In short a name will produce some effect in the


U

world; it has both illocutionary and perlocutionary force.


This is the capability that is at the centre of sacred names,
d

which are used with special reverential phrases and which


or

are widely believed to have power of their own. The idea


xf

of using a name to conjure with—such as that of some


powerful magician or occult figure—probably comes from
O

such associations. Thus the names of people index the


hearer to the power they command: their identity, their
social role, their capacity to gratify or withhold gratification
and so on. In short naming is a language game as the
concept is used by Wittgenstein:
130 Names: A Study

And there is a language game of inventing a name for


something, and hence of saying, “This is …”and then using
the new name. Thus, for example, children give names to
their dolls and then talk about them and to them (1953:
Section 27).

But the very notion of a ‘game’ is, as Wittgenstein said:

s
‘a concept with blurred edges’ (Ibid.: Section 71). This

es
means that the number of games you include in a set may
vary so much that there are only similarities, no exact

Pr
resemblances. Therefore, such sets are not firm or finite.
Such sets are called leaky sets. And our classification of
ity
personal names as being some kind of sequences of sounds
that have perlocutionary force too is based on some likeness
rs

with similar sequences—even if such other sequences occur


ve

in finite classes and are much more elaborate—but not


exact resemblances. This particular language game, the
ni

articulation of phonetic sequences, which indexes a person


U

as a being in the world, is sensitive to power differentials


and other societal restrictions. Thus in some societies the
d

less powerful (or younger person) cannot use the name


or

of the more powerful one without attaching honorifics


xf

or kinship terms. And sometimes the power differential


is so great that the mere name of a person inspires awe
O

in the hearers. This treatment of names is very close to


Jan Blommaert’s concept of languages carrying ‘orders of
indexality’, which ‘define the dominant lines for senses of
belonging, for identities and roles in society’ (Blommaert,
2010: 6). Basically, like language—which includes varieties,
accents, scripts, registers, etc.—names are indexed in our
CONCLUSION 131

minds with identities as well as power differentials. These


are of a general, group category type such as Mohamamad
(Muslim), Ram (Hindu) and Kaur (Sikh female) or a
particular, individualizing type such as Mohammad Nawaz
Sharif (the Prime Minister of Pakistan). To function
meaningfully in any society, we carry both orders of

s
indexality and attach it to the bearer of the name calling

es
forth a response conditioned by differentials of power, the
compulsions of ideology, worldview, and personal history.

Pr
To conclude this monograph, it should be noted that,
besides inspiring such inchoate theoretical ideas as given
ity
above, this analysis of Pakistani personal names has
rs
achieved some pioneering results. For instance, it has
brought out that the fact that naming practices in Pakistan’s
ve

modernizing society are related to perceptions of identity


of various kinds. Certain names, for instance, are indexed
ni

to class identities, i.e., rural and urban; and within these


U

two broad categories to other forms of social stratification.


Others are correlated with ethnic identity and with the
d

tension between modernization and the desire to seek a


or

traditional identity. Urban naming is governed by the desire


xf

to seek novelty without, however, losing consciousness of


the meaning altogether. Rural naming is traditional with
O

the imperative that a name must have a meaning. However,


there are areas in which the actual knowledge of the
meaning is so defective or non-existent as to create absurd
and meaningless names.

As naming practices are related to power differentials one


naming strategy—an upper-class one—is to assert one’s
132 Names: A Study

high social position by giving big names with prestigious


honorifics, titles and clan affiliations. But names may also
carry the burden of belonging to a lower, discriminated
against or stigmatized group and then they may function as
traps which must be escaped from. Hence name-changing
serves to create a shield to save the ego from the battering it

s
receives through insensitive banter or more serious negative

es
discrimination. Another strategy to conceal one’s lack of
power is to adopt the names of the powerful. This explains

Pr
the Christian, and to a much lesser extent the Hindu,
community’s use of neutral Muslim sounding names in
ity
Pakistan to avoid being discriminated against. And this
is exactly what Pakistani Muslims do when they find
rs
themselves in weaker positions while dealing with Western
ve

clients, such as Pakistanis working in call centres or trying


to find employment in the Western job markets.
ni

In short, this brief study of names delineates the manner


U

in which power and identity are experienced in Pakistan.


Names are constructors of identities and are indexed with
d

religious, sectarian, class, rural and urban, gender, modern,


or

conservative, and ethnic identities. And if one delves deeper


xf

into names they will reveal the social realities of Pakistan


more and more to us. That is why serious research in
O

onomastics would be welcome. This study merely scratches


the surface as pioneering works generally do. For instance,
there could be further exploration of the philosophical
aspects of naming in relation not only to the ideas of Austin
and Wittgenstein mentioned very briefly above, but also to
the precepts of other great philosophers of the world. Even
CONCLUSION 133

more to the point, this monograph leaves many interesting


areas in onomastics itself unexplored. For instance, a
satisfying account of nicknames and their functioning is
missing. Moreover, the differences between the nicknames
of different ethnic communities and classes need to be
investigated. Traditional genealogical tables (shajras) have

s
not been touched upon though they can tell us a lot about

es
name-changing, naming patterns, and the modernization
of names. Indeed, if we go far back, analyses of names

Pr
can tell us about the rate of conversion to Islam in what
is now Pakistan. In short, there is much that a scholar of
ity
onomastics can do in serious studies of personal names in
Pakistan. Then, of course, there are place names, names of
rs
pets, names in literature and so on. Indeed, the scope of
ve

onomastics is so vast that I am surprised that I thought of


this kind of research so late in life.
ni
U
d
or
xf
O
O
xf
or
d
U
ni
ve
rs
ity
Pr
es
s
O
xf
or
d
U
ni
ve
rs

Appendices
ity
Pr
es
s
O
xf
or
d
U
ni
ve
rs
ity
Pr
es
s
Appendix-I

Islamic Components

s
1. Names of God are given in the order established by tradition

es
by the Ulema and are not in alphabetical order (see Mufti
2012: 89–100). The Arabic determiner –al is added in all

Pr
names. Arabic orthography is indicated by symbols. NC
means ‘not counted here’ as the same name has been counted
elsewhere. The language is Arabic for all these names.
ity
No. Name Meaning
rs
1. Allāh/Ullāh Proper name of God.
2. Raḥmān Good, giver, beneficent.
ve

Owner, possessor, sovereign, supreme ruler,


3. Mālik
supreme spirit.
ni

4. Quddūs Righteous, purified, perfect, holy.


U

5. Salām Peace, protector, preserver.


6. Mō’min To render secure and safe, trustworthy, faithful.
d

7. Muhīmun Protector.
or

8. ‘Azīz Mighty.
Force, compeller, to employ power, to coerce,
xf

9. Jabbār
omnipotent.
O

Proud, majestic, noble, eminent, illustrious,


10. Mutakabbir
magnificent.
11. Khāliq Originator, creator.
12. Bārī Evolver, deity, highest God.
Designer, sculptor, fashioner, to give form, to
13. Musawwir
fashion, to model, to sculpture
Forgiver, condoner, absolver, indulgence, remission
14. Ghaffār
of sin.
138 A ppendices

No. Name Meaning


Destroyer, vanquisher, one who forces and
15. Qahhār
compels, subduer, oppressor.
Bestower, giver, donor, generous, gracious, one
16. Wahhāb
who gifts, concedes.
Sustainer, nourisher, maintainer, supporter, pro­
17. Razzāq
vider, one who provides resourses for subsistence.

s
es
Opener, initiator, one who unlocks, reveals, who
18. Fattāḥ explains, expounds, the key, the beginning, victo­
rious, conqueror.

Pr
All knowing, omniscient, intelligent, wise,
19. ‘Alīm
learned, sagacious.
ity
Constrictor, contractor, restrictor, curtailer, astrin­
20. Qābiz
gent, one who refrains, who grasps, takes away.
rs
Expander, dilating, amplifying, stretching out,
21. Bāsit
making capacious, joyful, merry.
ve

Abaser, one who humiliates, humbles makes


22. Khāfiḍ
lower.
ni

23. Rāfē’ The exalter, one who elevates, makes eminent.


U

24. Mu‘izz One who honours.


25. Mudhill One who dishonours.
d

26. Sami’ All hearing


or

27. Baṣīr All seeing


28. Ḥakam The judge
xf

29. ‘Adl The just.


O

Subtle, fine, delicate, refined, gentle, elegant,


30. Laṭīf
graceful, moderate.
31. Khabīr Aware, knowing, well informed, experienced.
32. Ḥalīm Forbearing, mild, gentle, patient, long suffering.
33. ‘Azīm High, lofty, enormous, very big, great.
34. Ghafūr All forgiving.
35. Shakūr Grateful.
A ppendices 139

No. Name Meaning


Exalted, noble, sublime, high elevated, top most
36. ‘Alī (NC)
summit, upper part.
Great, large, immense, proud, haughty, arrogant,
37. Kabīr
grievous, high majesty.
38. Ḥafīz Protector, defender, guardian, caretaker.

s
39. Muqīt Nourisher, sustainer, preserver, maintainer.

es
Reckoner, measure, value, one who calls to
40. Ḥasīb account, one who evaluates, recompenses or
rewards.

Pr
Dignified, illustrious, highly esteemed in rank or
41. Jalīl
dignity, glorious.
ity
42. Karīm Noble, generous, liberal, benign.
43. Raqīb Watchful, expector, opponent, competitor, enemy.
rs
44. Mujīb Responsive, expecting, granting.
ve

Ample, capacious, all embracing, vast, broad,


45. Wāsē’
comprehensive, all inclusive.
ni

All wise judge, physician, learned authority,


46. Ḥakīm
knowledgeable.
U

47. Wadūd Loving, affectionate, beloved, friendly.


Excelling, Surpassing in glory, lauded, praised,
d

48. Majīd
honourable.
or

Resurrector, awakener, cause, motive incentive,


49. Bā‘is
impulse.
xf

50. Shahīd Witness, evidence, affirmer, testimony.


O

51. Ḥaqq Truth, reality, certainty, proven, confirmed, just.


Trustee, pleader, attorney to whom one entrusts
52. Wakīl
one’s affairs. Counsel who represents.
53. Qawī Strong, powerful, mighty.
54. Matīn Firm, solid, forcible.
55. Walī Master, patron, guardian, friend.
56. Ḥamīd Praised, praiseworthy, commendable.
140 A ppendices

No. Name Meaning


57. Muḥ’sī Reckoner, counter, calculator.
58. Mubdī Originator, one who initiates.
59. Mu‘īd Restorer, renewer, renovater, one who brings back.
60. Muḥī Giver of life, raiser from death.
61. Mumīt Deadly, killer, creator of death.

s
62. Ḥayī Alive, living, animated.

es
63. Qayyūm Stable, fixed, lasting, permanent.
Finder, discoverer, perceived, existing, inventor,

Pr
64. Wājid invention, discovery, existence, personality,
perception.
Sublime, exalted, high, praised, noble, glorified,
65. Mājid
ity
honourable.
66. Wāḥid Unique, singular, original, incomparable, isolated.
rs
67. Aḥad One, oneness, unity, individual.
ve

Eternal, perpetual, to repair, to have recourse,


68. Ṣamad
succor and salvation, high, sublime.
ni

69. Qādir Able, capable, potent, almighty, prevailing over.


Able, overpowering, strong, authoritative, pos­
U

70. Muqtadir
sessing power.
Preceding, prior, to advance, first part, antecedent,
d

71. Muqaddam
anterior, having priority.
or

Procrastinated, to delay, last part, latter part,


72. Muakhkhar
posterior, subsequent, following.
xf

73. Awwal First, foremost


O

74. Ākhir Last, final


Manifest, visible, evident, obvious, external,
75. Ẓāhir
outward.
Hidden, covert, invisible, internal, intrinsic,
76. Bāṭin
recesses of mind and heart.
77. Wālī Governor, ruler, patron.
78. Muta‘āl Most exalted, highest.
A ppendices 141

No. Name Meaning


Good, truthful, pious, justified, innocent, having
79. Barr
good faith.
80. Tawwāb Forgiving, acceptor of penitence.
81. Muntaqim Avenger, punisher
82. ‘Afū Forgiver, pardoner, mild, clement, indulgent.

s
83. Ra’ ūf Forgiving, benign, pitiful, compassionate.

es
Mālik ul
84. Sovereign, administrator, superpower.
Mulk (NC)

Pr
Dhuljalāl
85. Lord of overwhelming might and bounty.
wal ikrām*
Divider, disperser, apportioner, analyzer, one
ity
86. Muqsit who breaks something into equitableparts, just
measurer.
rs
Collector, compiler, combining, completing,
87. Jāmē’
synthesis, make whole, collective, all.
ve

88. Ghanī Independent, self-sufficient, rich, wealthy.


ni

89. Mughnī To make independent, to enrich, make wealthy.


Forbidding, inhibiting, refusing, preventing, dis­
U

90. Mānē’
suading, hindering.
91. Ḍār Injurious, hurtful, distresses.
d

Profitable, beneficial, advantageous, wholesome,


or

92. Nafē’
good.
93. Nūr The light.
xf

94. Hādī To direct, to guide, leader, spiritual guide.


O

95. Badī’ Novel, rare, incomparable, wonderful.


96. Bāqī Everlasting, permanent, eternal.
97. Wārith Owner, master, inheritor.
Guiding in the right way, right path, holding a
98. Rashīd
right belief, one guided or directed aright.
Having great patience, enduring, suffering, mild,
99. Ṣabūr
self restraining, to wait, to content oneself.
* counted as two names Jalāl and Ikrām.
142 A ppendices

Names of the Prophet (pbuh)


(PBUH = peace be upon Him is used by Muslims after these
names if used for the Prophet of Islam). The language is Arabic
for all these names. NC= Not counted because of overlap.

No. Name Meaning


1. Muḥammad Highly praised.

s
2. Aḥmad Most commendable.

es
3. Ḥāmid Praising

Pr
4. Maḥmūd Praised
5. Qāsim One who distributes.
6. ‘Āqib Following, the last
ity
7. Fātiḥ (NC) Conqueror opener
8. Shāhid Witness
rs
9. Ḥāshir Who gathers people
ve

10. Rashīd (NC) Well guided


11. Mashhūd Witnessed
ni

12. Bashīr Bringer of good tidings


U

13. Nadhīr Warner


14. Dā‘ī Caller
d

15. Shāfi Healer


or

16. Hādī (NC) He who guides right


17. Mahdī (NC) He who is well-guided
xf

18. Māḥī He who wipes out (infidelity)


O

19. Munji He who saves or delivers (from adversity).


20. Nāhi Safe
21. Rasūl Messenger
22. Nabī Prophet
23. Ummī Unlettered
24. Tīhāmī From the Tihama
A ppendices 143

No. Name Meaning


Hāshimī
25. Family of Hashim
(NC)
26. Abtaḥī Belonging to al-Batha
27. ‘Azīz Noble, dear
Ḥarīs
28. Full of concern for you

s
‘alaikum

es
29. Ra’ūf (NC) Mild
30. Raḥīm (NC) Merciful

Pr
31. Ṭāhā Meaning not given but it occurs in Sura 20:1.
32. Mujtabā Elect
33. Ṭāsīn Sura 27:1. Meaning not given.
ity
34. Murtaẓā Content
35. Ḥā’mīm Beginning of suras 40–46. Meaning not given.
rs
36. Musṭafā Chosen
ve

37. Yā’sīn Sura 36:1. Meaning not given.


38. Aulā Worthier, most worthy
ni

39. Muzammil Wrapped


U

40. Walī (NC) Friend


41. Mudaththir Covered
d

42. Matīn (NC) Firm


or

43. Muṣaddiq Who declares for true


44. Ṭayyib Good
xf

45. Nāṣir Helper


O

46. Manṣūr Helped (by God), victorious


47. Miṣbāh Lamp
48. Āmir Prince, commander
49. Ḥijāzī (NC) From the Hijaz
50. Nazārī Meaning not given.
Quraishī
51. From the clan of the Quraish
(NC)
144 A ppendices

No. Name Meaning


From the tribe of that name called Muzar in
52. Muḍarī
Urdu.
Nabī-at-
53. The Prophet of repentance
taubā (NC)
54. Ḥāfiz (NC) Preserver

s
55. Kāmil Perfect

es
56. Ṣādiq Sincere
57. Amīn Trustworthy

Pr
‘Abdullāh
58. God’s servant
(NC)
59. Kalīm Allāh He to whom God has talked
ity
60. Habīb Allāh God’s beloved friend
61. Najī Allāh God’s intimate friend
rs
62. Ṣafī Allāh God’s sincere friend
ve

Khatim al-
63. Seal of the Prophets. The last prophet.
anbiyā’ (NC)
ni

64. Ḥasīb Respected


65. Mujīb (NC) Complying, replying
U

Shakūr
66. Most grateful
(NC)
d

67. Muqtaṣid Adopting a middle course


or

Rasūl ar-
68. The messenger of mercy
xf

rahmat
69. Qawī (NC) Strong
O

70. Ḥafī Well-informed


71. Mā’mūn Trusted
72. Ma’lūm Well-known
73. Ḥaqq (NC) Truth
74. Mubīn Clear, evident
75. Muṭī’ Obedient
A ppendices 145

No. Name Meaning


76. Awwal (NC) First
77. Ākhir (NC) Last
78. Zāhir (NC) Outward, external.
79. Bāṭin (NC) Internal, inner
80. Yatīm Orphan

s
81. Karīm Generous

es
82. Ḥakīm (NC) Wise, judicious
83. Sayyid (NC) Lord

Pr
84. Sirāj Lamp
85. Munīr Radiant ity
86. Muḥarram Forbidden, sacred, taboo.
87. Mukarram Honoured, venerated, respected.
rs
88. Mubashshir Bringer of good news
ve

89. Mudhakkir Who makes remember, preacher


90. Muṭahhar Purified
ni

91. Qarīb Near


92. Khalīl Good friend
U

93. Mad‘ū Who is called


d

94. Jawwād Generous, magnanimous


or

95. Khātim Seal. Last. One who ends something.


96. ‘Ādil Just
xf

97. Shahīr Well-known


O

Shahīd
98. Witnessing, martyr
(NC)
Rasūl al-
99. malāḥim The messenger of the battles of the last day
(NC)
146 A ppendices

Names of Prophets in the Quran


No. Name Meaning
1. Ādam Dust
To learn—as he used to learn and teach the
2. Idrīs
heavenly scrolls.
3. Nūh One who cries a lot, one who gives solace.

s
4. Hūd Guidance, pious deeds.

es
5. Ṣālih Pious, good, right.
6. Lūt Attach, adhere, pleasing, likable.

Pr
Literally means ‘father of the people’ or ‘most
7. Ibrāhīm
affectionate father’.
It is said that this name is composed of two
ity
words, -isma’ meaning listen! and -il meaning
8. Ismā‘Īl
Allah in the Syriac or Hebrew language. In other
rs
words “Hear and grant my prayer”.
ve

9. Ishāq One who laughs. To be far away.


Derived from ‘Aqb’ which means heel, behind,
10. Y‘āqūb
ni

end. Supplanter—holds the place of another.


11. Yūsuf The beautiful. Extremely sad.
U

It is from the word ‘sh`ab’ which means big


12. Shu‘aib nation or from the word ‘Sh`ib’ which means
d

valley.
or

Unable to locate meaning in Arabic but the


13. Ḥārun
Hebrew meaning is high, superior, exalted.
xf

14. Mūsā Water and tree. One who is taken out of water.
O

Derived from ‘Wud’ which means beloved, friend,


15. Dāwūd
companion in Arabic.
16. Sulaimān Derived from ‘Salam’ which means Peaceful.
Derived from ‘Awb or Aaba’ which means return,
17. Ayyūb penitent. Also given as persecuted in some
sources.
A ppendices 147

No. Name Meaning


Literally means “possessor of a double requital
18. Dhul Kifl
or portion”.
Derived from ‘Uns or Ans’ which means
19. Yūnus
affectionate.
Unable to locate meaning in Arabic but the

s
20. Ilyās Hebrew meaning is, ‘My God is the lord’.

es
Beloved of God.
Al-Yāsā Unable to locate meaning in Arabic but the
21.

Pr
(Elisha) Hebrew meaning is, ‘My God is salvation’.
Unable to locate meaning in Arabic but the
22. Zakariyāh
Hebrew meaning is, ‘The Lord remembers’.
ity
23. Yāhyā The living.
Leader, blessed. Mixture of colour red and yellow
rs
24. `Isā
with white.
ve
ni
U

People mentioned but not as Prophets


(Language is Arabic)
d

No. Name Meaning


or

1. ‘Imrān Fully inhabited, full.


xf

Khiḍr
2. (Khizar in Guide, green, subtle.
O

Urdu)
3. Luqmān Perfect wisdom.
4. Zulqarnain With two horns.
148 A ppendices

Orthodox Caliphs (Language is Arabic)


No. Name Meaning
Abū Bakr Father of youthfulness.
1.
Siddiq Truthful.
‘Alī Exalted.
2.
Hyder Lion, brave.

s
‘Umar Long-lived, elevated.

es
3. Fārūq One who discriminates, distinguisher.
Khattāb One who preaches.

Pr
‘Uthmān Servant of God.
4.
Ghanī Generous.
ity
Sahaba
rs
Companions of Prophet Muḥammad (pbuh)
(Language is Arabic)
ve

No. Name Meaning


1. ‘Abbās
ni

One who frowns, stern-faced, morose.


2. Abū Huraira Father of cats.
U

3. Abū dhar Father of fragrance.


4. ‘Ammār Cultivator, mild, agreeable, benevolent.
d

Cultivate and inhabit a land, worship, visit


or

5. ‘Amr
frequently.
6. Bilāl Satisfies thirst.
xf

7. Ḥamzā Lion.
O

Ploughman, cultivator. (The name Haris with


the (s) ending means guardian and protector.
8. Ḥarith (NC)
However, there is no distinction in the pronun­
ciation of the two words in Urdu).
9. Ḥassan Handsome, good.
10. Ḥussain Handsome, good.
11. Khālid Eternal, glorious.
A ppendices 149

No. Name Meaning


12. Mu‘āz Protected.
One who is hasty about everything-not given in
13. Mughīrā
mat sources.
14. Muthāb Who is difficult to be defeated, difficulties.
15. Nu‘mān Blood.

s
Gifted, bestowed. Most sources do not give any
16. Owais

es
meaning.
17. Sā‘d Felicity, good luck.

Pr
18. Sa‘īd Happy, fortunate.
19. Thābit One who exists.
20. Salmān Healthy, safe, wholesome.
ity
21. Sa‘ud Fortunate.
22. Ṭalhā Kind of tree.
rs
23. ‘Ubaidā Servant of God.
ve

24. ‘Ubayd Faithful, small slave.


25. Urvāh Support, handle.
ni

26. Wāqid One who burns.


U

27. Waqāṣ Warrior.


28. Yasir Wealthy.
d

29. Zaid Increase, growth, abundance.


or

30. Zubair Strong, powerful, smart.


xf

Tabiun (Language is Arabic)


O

No. Name Meaning


1. Alqāmā Height.
Command, order. The meaning with an ‘ain is
2. Amr (NC)
given elsewhere.
3. ‘Attā Gift.
4. Ḥamām Pigeon, dove.
150 A ppendices

No. Name Meaning


5. Jurayj Probably from Gregory.
6. Masrūq That which is stolen.
7. Mujāhid Warrior, one who strives.
8. Muṣ‘āb Restive, made difficult.
9. Muslim Submitting oneself to God.

s
10. Qais To scale, to measure, to estimate.

es
11. Shihāb Meteor, shooting star.
12. ‘Ubaid (NC) Little slave.

Pr
Quranic name for boys and girls meaning lion,
13. Urwā (NC) ever-green tree and handhold. It is mentioned in
the Quran twice.
ity
14. Zaid (NC) Increase, growth, abundance.
15. Zubair (NC) Strong, powerful, smart.
rs
ve

Taba-Tabiun
ni

No. Name Meaning


1. Anas Very sociable; friend.
U

2. Awza’i Sources do not give a clear meaning.


Bukharī
d

3. From the Central Asian city of Bukhara.


(NC)
or

4. Hambal Purity.
xf

5. Sufiyān Owner of ships, rich.


6. Shafī’ (NC) Mediator.
O

Shia Components (Language is Arabic)


No. Name Meaning
1. Ḥasnain Hassan and Hussain.
Kalb/Qalb
2. Kalb=dog; Qalb=heart.
names
A ppendices 151

Names of Imams and components derived


from them used in Shia Names (Language is Arabic)
Compo-
No. Name Meaning Period nent of
Name
Exalted One. 600–661 ‘Alawi/
1. ‘Ali ibn-e-Abi Talib

s
Ibn-e=son of CE. ‘Alwi

es
2. Ḥassan ibn-e-‘Ali Handsome 626–670 Hassani
3. Ḥussain ibn-e-‘Ali Handsome 626–680 Hussaini

Pr
‘Ali ibn-e-Ḥussain
Ornament of
4. (Imam Zainul 658/9–712 Zaidi
Worshippers
Abideen)
ity
The
Muḥammad ibn-e-‘Ali
5. Revealer (of 677–732 Baqri
rs
(Imam al-Baqir)
knowledge)
Ja‘far ibn-e-
ve

6. Muḥammad (Imam Honest 702–765 Jafari


Jafar as–Sadiq)
ni

Musa ibn-e-Ja‘far
7. Calm 744–799 Kazimi
U

(Imam Musa al-Kazim)


‘Ali ibn-e-Musa (Imam The Pleasing
8. 765–817 Rizvi
d

al-Raza) One
or

Muḥammad ibn-e-‘Ali The God-


9. 810–835 Taqvi
(Imam al-Taqi) fearing
xf

‘Ali ibn-e-Muḥammad
10. The Pure 827–868 Naqvi
(Imam al-Naqi)
O

Ḥassan ibn-e-‘Ali Citizen of a


11. 846–874 Askari
(Imam al-Askari) Garrison
Muḥammad ibn-e-
The Guide; 869-
12. Ḥassan (Imam al- Mehdavi
The Proof unknown
Mehdi)
NB: These components are regarded as markers of the Shia sect but are shared
by the Sunnis also.
152 A ppendices

Indigenous Components of names


considered Islamic by Pakistanis
No. Name Meaning
1. Bakhsh (P) Gift.
2. Dād (P) Given
3. Dīn (A) Religion. Written as Deen in Urdu.

s
4. Dittā (Pu) Given

es
A high rank of mystics. The name refers to
the belief that a child is born by the blessing

Pr
5. Ghauth (A) of Sheikh Abdul Qadir Jilani whose name is
pronounced with an/s/and written as Ghaus in
transliteration from Urdu.
ity
The Presence. Refers to the Prophet of Islam.
6. Ḥuzūr (A)
Written as Huzoor in Urdu.
rs
Ilāhī (Urdu
7. God
ve

Ilahee) (A)
8. Khudā (P) God.
ni

9. Maulā (A) God


Old. Here it refers to a sufi saint. Written as Peer
U

10. Pīr (P)


in Urdu.
11. Rab (A) God
d

Saĩ Title of respect in Sindhi and Siraiki cultures.


or

12.
(Siraiki/S) Written as Sain in Urdu and Sindhi.
xf
O
A ppendices 153

Names from the Jihadi/Resistance


Muslim Sub-culture (Language is Arabic)
No. Name Meaning
1. Abū names Father of
2. Bin names Son of
3. Muāwiyāh Young fox

s
es
4. Osāmā Lion. Also spelled Usama.
5. Ṣaddām Brave. One who resists; contends.

Pr
Specific Names not used among Pakistani Muslims
6.
Arab names earlier. ity
7. Yazīd Becoming greater.
rs
ve
ni
U
d
or
xf
O
Appendix-II

Islamic Components in Female Names

s
Wives of the Prophet (pbuh) (Language is Arabic)

es
No. Name Meaning
1. ‘Ay’ēshā Life, vivaciousness, living, prosperous.

Pr
2. Ḥafṣā Young lioness, gathering.
3. Hind From India, female deer in Hebrew.
ity
4. Jawēriā One who spreads happiness, mysterious.
5. Khadījā Early baby.
rs
6. Mai’mūnā Safe, trustworthy, auspicious.
7. Māriā Bitter or sea of bitterness.
ve

8. Ramlā Prophetess. One who is embellished.


9. Reḥānā Fragrance
ni

10. Ṣafiā Lion’s share, pure, best friend, untroubled.


U

11. Ṣaudā Decent, mild tempered.


12. Zaynab (NC) Generous.
d
or

Daughters of the Prophet (pbuh)


No. Name Meaning
xf

Fātimā One who saves, motherly, one who abstains.


O

  • Zehrā Flower, beauty; also bright, shining, brilliant


1. (slightly different spellings in Arabic but identical
in Urdu).
  • Batūl Virgin, pure, chaste.
Kulthūm
2. With fresh and pink cheeks, beautiful.
(Ummē)
3. Ruqayyā Spell, superior.
4. Zaynab Name of a beautiful tree exuding fragrance.
A ppendices 155

Sahābiyāt
No. Name Meaning
1. ‘Afrāh Celebrations, joys.
2. Asmā Loftier, more eminent, supreme.
3. Barakāh Blessing.
4. Ḥamrā Red.

s
5. Hanī Delighted, content.

es
6. Khānsā Pug-nosed. (famous Arabic poetess)
7. Salmā Peaceful.

Pr
“High above” in Arabic. This was the name of the
8. Summayyāh
first martyr for Islam.
ity
It is an ancient Arabic word and the meaning
is not certain when used as a name, it is derived
rs
from the verb Amma, which means “he/she
intended to do something”, thus it may have
ve

meanings of “intention”. It is derived from the


9. Umāmā
A-M-M root from which many Quranic words
ni

are also derived such as  Ummah (nation, creed)


U

and Umm (mother).
Some sources say it is ‘the diminutive of mother’.
In Xosa it means ‘mother’.
d

It is an ancient Arabic name and its meaning is


or

not perfectly known. However, its root is ‘Ain-


xf

10. Umārā M-R, which has meanings of “to build”, “to


maintain”, “to tend to”, “to enliven”, “to inhabit”,
O

“to settle in”.


156 A ppendices

Other Women Associated with Piety


No. Name Meaning
1. Āminā Secured.
2. Ḥalīmā Gentle, patient, mild-tempered.
3. Rabiyāh Fourth.

s
Women in the Quran

es
No. Name Meaning

Pr
1. Āsiā Alive, hope. In Hebrew ‘work’.
2. Bilqīs Queen of Sheeba.
ity
3. Ḥājrā One who migrates.
4. Mariam Chaste. Star of the sea.
rs
5. Sārā Pure, happy.
ve

6. Zulaikhā Brilliant one.


ni

Arabicized/Radical/Islamic Names
U

No. Name Meaning


1. Bint names Names of daughters.
d

Ummē
or

2. Names of mothers.
names
xf

Shia Components of female names


O

(Same as in mens’ names)


A ppendices 157

Indigenous Components (Language is indicated)


No. Name Meaning
One who has been sustained or made to live in
1. Basāyī (U)
peace.
2. Dittī (Pu) Given i.e God-given.
Maḥmūdan Female form of Mahmud. This is a variant used
3.

s
(A) among the Muslim speakers of Urdu.

es
Muḥam­ Female form of Muḥammad. Variant used by
4.
madī (A) the above.

Pr
Rakhkhī One who has been sustained or made to live in
5.
(U-Pa) peace. ity
One who has been sustained or made to live in
6. Wasāyī (Pa)
peace.
rs

Names of Religious Concepts


ve

No. Name Meaning


ni

1. Ādhān (A) Call to prayers. Uḍun also means ‘ear’.


U

2. Du‘ā (A) Prayer, wish, invocation.


3. Firdaōs (P) Paradise. Many variant spellings.
d

4. Iram (P) Paradise.


or

5. Jannat (A) Paradise.


6. Marwā (A) A place in Mecca.
xf

7. Ṣafā (A) A place in Mecca.


O

8. Sidrā (A) A tree, a certain mystical location.


9. Ṭūbā (A) A tree in paradise, venerated.
Appendix-III

Identity Components in Women’s Names

s
es
Rural/old fashioned components

Pr
No. Name Meaning
1. Bānō (P) Princess. ity
2. Bēgum (T) Lady.
Lady. Said to be from Swahili meaning ‘lady of
rs
3. Bībī (A)
the house’.
ve

Dil names
4. Names having the word heart in them.
(H-U)
ni

Jān names
5. Life or dear.
(P-U)
U

6. Khānum (T) Female of Khan (chief ).


d

Queen, empress. Female form of Khan. Gerard


7. Khātūn (T) Clauson, an Orientalist, considers it borrowed
or

from Sogdian where it means a ‘landlord’s wife’.


xf

Mother. But also used as part of names in Siraiki


8. Mai (U-Pu)
areas and parts of the Punjab.
O

Nissā names
9. Names having women as a component.
(A)
A ppendices 159

Prestigious, Ethnic and Caste


Components of Names (both men and
women)
Prestigious titles used as Names
No. Name Meaning
Āghā
1. Title meaning respectable.

s
(T & P)

es
2. Bēg (T) Chief. Also spelled Baig.
Chaudhary

Pr
3. Chief.
(H)
4. Jām (S) Chief. ity
5. Khān (T) Chief.
Title meaning honourable. In Balochi it becomes
6. Khawājā (P)
rs
Waja.
Arabic meaning owner but used for landowners
ve

7. Malik (A)
in the Punjab.
8. Mīr (U) Used for Syeds. Also spelled Meer.
ni

Persian for Ameer Zada i.e son of leaders but


9. Mirzā (P-U)
used for Mughals during the Mughal empire.
U

10. Nawāb (U) Lord.


d

Nawābzādā
11. Son of a lord
(U)
or

12. Pāshā (T) Nobleman.


xf

13. Qāzī (A) Judge.


14. Raīs (U) Urdu for a rich man, notable. Also spelled Raees.
O

Sāhibzādā
15. Son of a notable person.
(U)
16. Sardār (P) Leader.
Shāhzādā
17. Son of a king.
(P)
18. Wadērō (S) Sindhi for a landlord
19. Wadērā (S) Same as above.
160 A ppendices

Prestigious components of names with


reference to religion (roots are from Arabic
unless otherwise indicated)
No. Name Meaning
1. ‘Abbāsī From ‘Abbas.
2. ‘Alvī From ‘Ali.

s
Those who helped the immigrant Muslims

es
3. Ansārī escaping persecution in Mecca to settle down in
Madina.

Pr
4. Chishtī Belonging to the Chisti order of the sufis.
Associated with the Caliph Umar Faruq. Also
5. Fārūqī
spelled Farooqi.
ity
6. Gilānī/Jilānī Associated with Sheikh Abdul Qadir Gilani.
rs
7. Hāshmī From the tribe of Banu Hashim.
8. Makhdūm Syed in Sindh.
ve

9. Nizāmī Associated with Sheikh Nizamuddin Auliya.


Associated with the mystic normally spelled
ni

10. Owaēsī
Owais al-Qirni.
U

Old in Persian. It refers to one’s mystic preceptor.


11. Pīr
Also spelled Peer.
d

12. Pīrzādā Son of a mystic.


or

13. Qādrī Associated with Sheikh Abdul Qadir Gilani.


Belonging to the Quresh tribe. Also spelled
14. Quraēshī
xf

Quraishi
15. Shāh Persian for king.
O

Associated with Abu Bakr Siddiq. Normally


16. Siddīqī
spelled Siddiqui.
17. Suhrawardī Associated with Sheikh Shahabuddin Suhrawardy.
18. Syed Leader, of gentlemanly status, decent family.
Associated with the Caliph ‘Uthman Ghani.
19. ‘Uthmānī
Normally spelled Usmani.
A ppendices 161

Prestigious Place inspired names


No. Name Association
1. Bukhārī From Bukhara in Central Asia
2. Dehlavī From Delhi in India
3. Gardēzī From Gardez in Iran
4. Hamadānī From Hamadan in Iran

s
es
5. Hijāzī From Hijaz in Saudi Arabia
6. Kasūrī From Kasur near Lahore

Pr
7. Madnī From Medina
8. Makkī From Makka (Mecca).
ity
9. Shīrāzī From Shiraz in Iran
10. Thānvī From Thana Bhawan in U.P.
rs
ve
ni
U
d
or
xf
O
162 A ppendices

Caste/Qoum Names
Balochistan
No. Name No. Name No. Name
1. Ahmedani 21. Jamali 41. Mastoi
2. Baloch 22. Jangi 42. Mazari

s
3. Bhurgri 23. Janwari 43. Mengal

es
4. Bijarani 24. Jarwar 44. Mirwani
5. Bugti 25. Jatoi 45. Muggo

Pr
6. Buledi 26. Jiskani 46. Nizamani
7. Buzdar 27. Khorkhani
ity 47. Pandrani
8. Chang 28. Khosa 48. Pitafi
9. Dasti 29. Khushk 49. Qaisrani
rs
10. Domki 30. Korai 50. Qalat
ve

11. Gabol 31. Kulachi 51. Qambrani


12. Gadai 32. Laghari 52. Raisani
ni

13. Gadhi 33. Lango 53. Rind


U

14. Gopang 34. Lanjwani 54. Sakhani


15. Gurmani 35. Lari 55. Sanjrani
d
or

16. Hasni 36. Lashari 56. Shahani


17. Hesbani 37. Lehri 57. Tauqi
xf

18. Isani 38. Lund 58. Umrani


O

19. Jagirani 39. Magsi 59. Zardari


20. Jalbani 40. Marri 60. Zehri
A ppendices 163

Sindh
No. Name No. Name No. Name
1. Abro 26. Jamot 51. Memon
2. Bhanbhro 27. Joyo 52. Miraei
3. Bhil 28. Junejo 53. Mirani
4. Bhurt 29. Kachelo 54. Mirbahar

s
5. Bhutto 30. Kakepoto 55. Mithani

es
6. Bughio 31. Kalhoro 56. Muggo
7. Buriro 32. Keerio 57. Palh

Pr
8. Chachar 33. Khaskheli 58. Palijo
9. Chandio 34. Kohli ity 59. Panhwar
10. Channa 35. Kumbhar 60. Pathan
11. Channar 36. Lail 61. Peechoho
rs
12. Chansar 37. Lakhani 62. Phulpoto
13. Chutta 38. Lakhmir 63. Rajpar
ve

14. Dahiri 39. Lakho 64. Sadhayo


15. Dano 40. Lanjar 65. Sahito
ni

16. Daudpoto 41. Lari 66. Samejo


U

17. Dayo 42. Larik 67. Sarki


18. Depar 43. Lohana 68. Solangi
d

19. Dero 44. Mahesar 69. Soomro


or

20. Hakro 45. Mallah 70. Tagar


21. Halepoto 46. Mangi 71. Talpur
xf

22. Hingoro 47. Mangrio 72. Unar


O

23. Hisbani 48. Manjhi 73. Wassan


24. Hur 49. Meghwar
25. Jakhio 50. Mehrani
164 A ppendices

Punjab
No. Name No. Name No. Name
1. Arain 25. Kardar 49. Ramay
2. Awan 26. Kassar 50. Rana
3. Bajar 27. Khakwani 51. Ranjha
4. Bajwa 28. Khatri 52. Rao

s
5. Bhalli 29. Khattana 53. Rathore

es
6. Bhatti 30. Khattar 54. Sahi
7. Bhullar 31. Kiyani 55. Sahni

Pr
8. Bhutta 32. Langah 56. Sethi
9. Butt 33. Lone ity 57. Shaikh
10. Chattha 34. Mahoon 58. Sipra
11. Chauhan 35. Mahtam 59. Siyal
rs
12. Cheema 36. Meghwar 60. Sulehria
13. Daha 37. Mela 61. Sumbal
ve

14. Dar 38. Meo 62. Talwar


15. Dogar 39. Mian 63. Tarar
ni

16. Ghumman 40. Minhas 64. Toor


U

17. Gill 41. Mughal 65. Toori


18. Gondal 42. Munda 66. Vandal
d

19. Gujjar 43. Nagi 67. Virk


or

20. Jagal 44. Nanda 68. Waryah


21. Janjua 45. Passi 69. Waseer
xf

22. Jat 46. Qaim Khani 70. Wattu


O

23. Kahloon 47. Raja


24. Kalas 48. Rajput

NB: Sumbal is a sub-tribe but also a first name. It has not been counted.
A ppendices 165

Khyber Pakhtunkhwa
No. Name No. Name No. Name
1. Afridi 13. Lodhi 25. Safi
2. Bangash 14. Mahsud 26. Shilmani
3. Barki 15. Marwat 27. Shiwari
4. Daur 16. Mashwani 28. Suri

s
5. Durrani 17. Mohmand 29. Swati

es
6. Jadun 18. Mullagori 30. Tanoli
7. Kakakhel 19. Muqbil 31. Tarin

Pr
8. Kakar 20. Nasiri 32. Tokhi
9. Kharoti 21. Nurzai 33. Yusufzai
ity
10. Khattak 22. Orakzai 34. Zaimukht
rs
11. Kuchi 23. Para Chamkani 35. Zardan
12. Kundi 24. Piracha
ve
ni
U

Famous Families
d

No. Name No. Name No. Name


or

1. Arbab 5. Minto 9. Oberoi


2. Bilor 6. Monnoo 10. Saigol
xf

3. Isphahani 7. Nangiana 11. Tiwana


O

4. Kheiri 8. Noon 12. Wala families


Annexure-A

Special Terms Used For The

s
Analysis Of Names

es
For counting the number of religious and caste components in
names the following abbreviations have been used.

Pr
Includes names used to indicate group identities also
Caste
called qoum, zat and biradari.
ity
The first four orthodox caliphs: Abu Bakr Siddiq,
rs
Umar Farooq, Usman Ghani and Ali ibn-e-Abi Talib.
Caliphs
Muslims prefix Hazrat (presence) as a conventional
ve

mark of veneration before their names.


Components in names e.g. Yusaf Khan has two
ni

Comp
components.
U

Daughters Daughters of the Prophet of Islam (pbuh).


Density The number of components per name.
d

Families Well-known extended family names.


or

Indigenous components found in folk Islam e.g. Ditta


xf

Folk
(given).
O

The ninety-nine names of God as given in Muslim


God
literature.
Women associated with piety such as Haleema, the
Pious
nurse of the Prophet.
Names inspired by prestigious places such as Bukhari
Place
from Bukhara.
A ppendices 167

Names of the Prophet of Islam (pbuh) except


those which overlap the names of God. PBUH is
Prophet understood to be used after these names if used for
him but not if used as part of other peoples’ names (see
ABBREVIATIONS).
Names of the prophets and other personages mentioned
Prophets

s
in the Quran.

es
Quran (W) Names of women mentioned in the Quran.
Components which may be indicators of radical

Pr
Radical
interpretation of Islam or Arabization.
Names inspired by religious associations but now used
Rel
ity
as group identity labels.
Sahaba Names of the male companions of the Prophet.
rs

Names of women associated with the above or


Sahabiyat
ve

contemporaneous with them.


Components predominantly used by Shias in North
ni

Shia
India and Pakistan (Naqvi, Rizvi etc).
U

Those who had seen the Sahaba and those who had
Taba-tabiun
seen the Tabiun (Taba-tabiun).
d

Prestigious titles now used as names e.g. Khan, Pasha,


or

Titles
Mirza etc.
xf

Wives Names of the wives of the Prophet of Islam.


O
Annexure-B

BRIEF DESCRIPTIONS OF PILOT and

s
MAIN SAMPLES

es
Pilot sample from the Apna Model School which
has lower middle-class children with very little

Pr
1. Apna tuition fees in Rawalpindi. Only the sample of
girls has been used since the number of boys was
too small. They were named in the late 1990s.
ity
Pilot sample from the Beaconhouse National
rs
BNU (B & University, Lahore which has upper middle class
2.
G) students. Both boys and girls were named in the
ve

early 1990s.
Names of voters in one polling station of the
3. BADEEN
ni

constituency of Badeen, Sindh. Age unknown.


U

Names of voters in one polling station of Loralai


Balochistan
constituency of Balochistan. Ages approximately
(M-1940);
4. known. The elderly males are from the 1940s
d

(M-1960)
and the other males and females from the 1960s
or

& (F-1960).
to the 1990s.
xf

Working-class, partly-employed and unemployed


men who receive financial aid from the
O

5. Poor-M government through the Benazir Income Support


Programme from all districts of Pakistan. Their
ages range between 20 to 60 years.
6. Poor-F Women of the same sample.
A ppendices 169

Middle and lower middle class males from


the Punjab whose names are in the 2012
Punjab-M matriculation gazette of the Lahore Board of
7.
(1990s) Intermediate and secondary Education. Their
ages being between 15 and 16, they were named
in the late 1990s.
Punjab-F

s
8. Girls of the above sample.
(1990s)

es
Pilot samples of elderly, middle-aged and young
Punjab-R girls from the villages of the Punjab near Lahore,

Pr
9.
(M, F & G) Bahawalpur (South Punjab) and near Rawalpindi
(Potohar) from the 1950s and the 1990s.
ity
KPK-M Boys of the same ages and classes in the gazette
10.
(1990s) of the Peshawar BISE, 2012.
rs
KPK-F
11. Girls of the above sample.
(1990s)
ve

Sindh-M Men of the middle-class from Sindh named in


12.
ni

(1950s) the early 1950s.


Sindh-M Boys of the middle-class from Sindh named in
U

13.
(1990s) the 1990s and early 2000s.
Sindh-F
d

14. Girls of the same sample.


(1990s)
or

Upper middle and upper class men from an elite


Elite-M
xf

15. club and an English-medium school in Lahore


(1990s)
named in the 1950s.
O

Elite-F
16. Women of the same sample.
(1950s)
Middle and lower middle class males from the
Punjab whose names were in the matriculation
Punjab-M
17. gazettes of the Lahore BISE of the 1950s and
(1950s)
1960s. They were named in the late 1940s and
1950s.
170 A ppendices

Punjab-F
18. Women of the above sample.
(1950s)
Names of males of the above classes in the
KPK-M
19. matriculation gazette of the Peshawar board of
(1950s)
1961. They were named in the 1950s.
KPK-F
20. Women of the above sample

s
(1950s)

es
Elite-B Boys from the upper middle and upper classes of
21.
(2000s) urban Punjab named in 2000s.

Pr
Elite-G
22. Girls of the same sample.
(2000s) ity
rs
ve
ni
U
d
or
xf
O
Annexure-C

CONSOLIDATED SUMMARY OF ISLAMIC

s
COMPONENTS IN MALE NAMES

es
PILOT SAMPLES

Pr
Rural-M Rural-M BNU- B
No. Names
(1950s) (1990s) (1990s)
No. of Names 477 204 462
ity
Components 983 373 1,189
rs
Density 2.06 1.83 2.57
ve

1. God 98 19 76
2. Prophet 239 34 205
ni

3. Prophets 20 8 25
U

4. Caliphs 43 7 89
d

5. Sahaba 82 98 110
or

6. Taba-tabiun Nil 2 02
7. Shia 01 1 18
xf

8. Folk 31 3 01
O

9. Radical Nil Nil 12


Total 514 172 538
Percentage 52 46.11 45.25
Annexure-D

CONSOLIDATED SUMMARY OF ISLAMIC

s
COMPONENTS IN FEMALE NAMES

es
PILOT SAMPLES

Pr
Rural-F Rural-F Apna-G BNU-G
No. Names
(1950s) (1990s) (1990s) (1980s)
No. of Names 476 131 279 389
ity
Components 955 261 565 860
rs
Density 2.01 1.99 2.03 2.21
ve

1. Wives 5 5 23 20
2. Daughters 22 6 15 33
ni

3. Sahabiyat 2 2 03 4
U

4. Pious 8 6 08 14
d

5. Quran (W) 1 2 09 18
or

6. Radical Nil Nil Nil 2


7. Shia Nil Nil Nil 12
xf

8. Folk 21 Nil Nil 01


O

Total 59 21 58 104
Percentage 6.18 8.05 10.27 12.09
Annexure-E-1

CONSOLIDATED SUMMARY OF RURAL/OLD-

s
FASHIONED COMPONENTS IN

es
FEMALE NAMES
PILOT SAMPLES

Pr
Rural-F Apna-G BNU-G
No. Names Elite-G
(1950s) (P) (1990s) (1980s)
ity
No. of Names 476 282 389 608
rs
Density 2.01 2.03 2.21 2.08
ve

1. Bibi 398 23 Nil 01


2. Bano 10 Nil 04 01
ni

3. Khatoon 2 01 Nil Nil


U

4. Begum 20 01 Nil Nil


5. Mai 04 Nil Nil Nil
d
or

6. Jan Nil Nil Nil Nil


7. Nissa names 02 Nil 03 04
xf

8. Dil names Nil Nil Nil Nil


O

9. Gul Nil 02 03 Nil


Total 436 27 10 06
Percentage
91.60 9.57 2.57 0.99
(out of names)
O Annexure-E-2
CONSOLIDATED SUMMARY OF ISLAMIC and CASTE COMPONENTS IN NAMES (BADEEN)
xf
Badeen-M Badeen-M Badeen-M
Badeen-M Badeen-M Badeen-F Badeen-F Badeen-F Badeen-F
Men Heads 1950–80s 1940s 1940s
heads caste heads Islamic Heads caste
Islamic Islamic caste
or
No. of Names 5,208 5,208
d 5,208 Names 4,584
Components 10,630 10,591 U 10,630 Comp 5,972
Density 2.04 2.03 2.03 Density 1.30
1. God 1,233 622 Titles 242 453 Wives 120 Titles 63
ni
2. Prophet 2,378 2,556 Religious 246 276 Daughters 242 Religious 22
ve
3. Prophets 260 370 Place 03 04 Quran (W ) 175 Place 02
4. Caliphs 968 577 Caste 840 1,192 Pious (W) 115 Caste 98
rs
5. Sahaba 431 253 Families Nil 02 Sahabiyat 41 Families 08
6. Taba-tabiun 7 3
ity
7. Shia 34 12 Shia Nil
8. Folk 126 471 Folk 59
9. Radical 12 02 Radical 12
Pr
Total 5,449 4,866 1,331 1,927 764 193
es
Percentage 51.30 45.94 12.52 18.12 12.79
s 03.23
O Annexure-F
CONSOLIDATED SUMMARY OF ISLAMIC COMPONENTS IN MALE NAMES
xf
MAIN SAMPLES
Balo- Balo-
KPK-M Punjab- Punjab- KPK-M Sindh-M Sindh-M Elite-M Elite-B
or
No. Names Poor-M chistan- chistan-
1990s M 1990s M 1950s 1950s
d 1950s 1990s 1950s 2000
M 1950s M 1940s
No. of Names 500,000 42,692 65,331
U 22,464 63,013 2,950 42,459 3,649 512 13,771 22,799
Components 1,030,000 89,270 148,057 53,717 135,621 6,632 119,178 10,198 1,138 27,925 27,701
Density 2.06 2.09 2.27 2.39 2.15 2.25 2.81 2.79 2.22 2.03 1.22
ni
1. God 61,989 12,610 9,209 4,183 18,919 508 9,718 640 116 5,634 7,745
2. Prophet 83,056 14,560 35,123 16,361 18,513 1,411 14,992 1,537 202 5,586 9,128
ve
3. Prophets 8,205 2,497 3,480 858 1,842 105 1,411 143 29 542 837
4. Caliphs 110,900 4,488 13,660 2,225 4,287 468 13,745 383 76 350 618
rs
5. Sahaba 59,424 5,415 12,201 2,905 4,929 274 5,803 545 77 322 568
6. Taba-tabiun 1,022 462 355 96 330 7 274 8 01 05 58
ity
7. Shia 6,508 175 2,330 245 221 65 600 126 13 35 47
8. Folk 28,888 1,053 443 1,068 1,814 148 595 137 4 700 1,654
Pr
9. Radical 3,346 350 726 15 4 Nil 546 5 18 11 02
Total 363,338 41,610 77,527 27,956 50,859 2,986 47,684 3,524 536 13,185 20,657
es
Percentage 35.28 46.61 52.36 52.04 37.50 45.03 40.01 34.55 s 47.10 47.22 74.67
O Annexure-G
CONSOLIDATED SUMMARY OF ISLAMIC COMPONENTS IN FEMALE NAMES
xf
MAIN SAMPLES
Punjab-F KPK-F Sindh-F Elite-F Punjab-F KPK-F Elite-G Balo-
No. Names Poor-F
or
(1990s)
d (1990s) (1990s) (1950s) (1950s) 1950s 2000 chistan-F
No. of Names 300,000 65,532 17,333 21,765 1,900 2,828 413 608 9,028
Components 609,000 134,011
U
29,994 45,868 4,313 5,938 838 1,265 16,601
Density 2.03 2.05 1.73 2.11 2.27 2.1 2.03 2.08 1.84
ni
1. Wives 2,859 4,731 962 697 65 76 9 34 224
2. Daughters 12,125 3,367 769 882 24 130 21 78 226
ve
3. Sahabiyat 1,045 1,089 447 419 rs 24 22 5 4 28
4. Pious 1,004 3,596 559 190 17 25 1 16 218
5. Quran (W) 1,625 1,599 546 483 30 61 4 10 97
ity
6. Radical 309 105 51 08 Nil Nil 1 2 Nil
7. Shia 162 193 9 16 27 16 Nil 5 Nil
8. Folk 310 42 6 86 01 02 4 Nil 12
Pr
Total 19,439 14,722 3,349 2,781 188 332 45 149 905
es
Percentage 3.19 10.98 11.17 6.06 4.36 5.59 5.37
s 11.78 4.84
O Annexure-H
MEN’S NAMES/GROUP LABELS IN FEMALE NAMES
xf
MAIN SAMPLES
Punjab-F KPK-F Sindh-F Elite-F Punjab-F KPK-F Elite-G Balo-
or
No. Names Poor-F
(1990s) (1990s) (1990s) (1950s) (1950s) 1950s 2000 chistan-F
d
No. of Names 300,000 65,532 17,333
U 21,765 1,900 2,828 413 608 9,028
Components 609,000 134,011 29,994
ni 45,868 4,313 5,938 838 1,265 16,601
Density 2.03 2.05 1.73 ve 2.11 2.27 2.1 2.03 2.08 1.84
1. Titles 4,744 1,517 640 491 194 48 7 61 113
2. Religious 1,615 1,048 418 958 98 95 4 25 159
rs
3. Place 89 136 13 42 18 5 1 3 05
ity
4. Caste 2,058 989 279 13,928 135 61 9 31 137
5. Families 04 19 21 17 34 03 Nil 3 Nil
Pr
Total 8,510 3,709 1,371 15,436 479 212 21
es 123 414
Percentage 1.40 2.87 4.57 33.65 11.11 3.57 2.5 9.72 2.49
s
O Annexure-I
CASTE/QOUM/ZAT/PRESTIGIOUS Components IN MALE NAMES
xf
MAIN SAMPLES
Balo-
or
Poor-M Punjab- KPK-M Punjab- KPK-M Sindh-M Elite-M Elite-B Balo-
No. Names chisan-
(Pak) M 1990s
d 1990s M 1950s 1950s 1990s
M 1940s
1950s 2000 chisan-M

No. of Names 500,000 65,331


U
42,692 22,464 63,013 42,459 22,799 3,649 512 13,771
Components 1,030,000 148,057 89,270 53,717 135,621 119,178 27,701 10,198 1,138 27,925
ni
Density 2.06 2.27 2.09 2.39
ve 2.15 2.81 1.28 2.79 2.22 2.03
1. Titles 120,780 3,012 8,971 3,008 19,768 2,621 6,904 1,014 58 3,740
2. Religious 318 2,330 3,077 1,523 6,141 2,679 1,574 359 35 1,011
rs
3. Place 105 87 21 43 37
ity 56 Nil 55 01 03
4. Caste 9,401 2,094 629 1,173 682 28,556 285 774 34 120
5. Families Nil 21 48 19 102 87 03 62 4 01
Total
Pr
130,604 7,544 12,746 5,766 26,730 33,999 8,766 2,264 132 4,875
Components
es
Percentage 12.68 5.10 14.28 10.73 19.71 28.53 31.65 22.20 11.60 17.45
s
O Annexure-J
Frequency of Occurrence of the Radical Islamic Components
xf
The frequency of occurrence of the names given below out of the total number of names is expressed in
numbers. The numbers in the parentheses are percentages calculated to two decimal places.
or
Elite-M
d KPK-M KPK-M Punjab-M Punjab-M Balochistan-
Poor-M
1950s U1950s 1990s 1950s 1990s M 1990s
N=500,000
3,649 63,013 42,692 22,464 65,331 13,771

195 1 282 245 08


ni
Saddam Nil Nil
(0.04) (0.001) (0.66) (0.38) (0.06)
ve
529 03 1 46 233 01
Osama Nil
(0.11) (0.08) (0.00)
rs
(0.11) (0.36) (0.007)

16 7
ity
Muawiya Nil Nil Nil Nil Nil
(0.00) (0.01)

13
Yazid Nil Nil Nil Nil Nil Nil
Pr
(0.00) es
s
Poor-M
O Elite-M KPK-M KPK-M Punjab-M Punjab-M Balochistan-
1950s 1950s 1990s 1950s 1990s M 1990s
N=500,000
3,649 63,013 42,692 22,464 65,331 13,771
xf
1,793 01 05 6 07 45
Abu Nil
or
(0.36) (0.03) d (0.01) (0.01) (0.03) (0.07)

800 01 U 02 23 07 42
Bin Nil
(0.16) (0.03) (0.00) (0.05) (0.03) (0.06)

Elite-F (P) KPK-F KPK-F Punjab-F Punjab-F


ni
Poor-F Balochistan
1950s 1950s 1990s 1950s 1990s
300,000 (F) 9,028
ve
1,900 413 17,333 2,828 65,532

234 13 77
rs
Umme Nil Nil Nil Nil
(0.08) (0.08)
ity (0.12)

75 01 7 28
Bint Nil Nil Nil
(0.03) (0.24) (0.04) Pr (0.04)

Sources: The above data is from both male and female samples of the years given above. Umme and bint names (e.g Umme Kalsoom)
and corresponding male names (e.g Abubakr), used traditionally in South Asia, are not an expression of a self-consciously radical
es
Islamic identity. As such, they have not been counted.
s
O Annexure-K
Number and Percentage of Rural/Old-Fashioned Female Names
xf
Punjab-F Punjab-F KPK-F KPK-F Sindh-F Elite-F Elite-G Balo-
Poor-F Badeen-F
1950s 1990s 1950s 1990s 1990s 1950s 2000s chistan-F
or
Total (300,000) (2,828)
d
(65,532) (413) (17,333) (614) (1,900) (608) (9,028) (4,584)

104,634 112 418 22 2,509 4 1 5,984 114


Bibi
(34.88) (3.96) (0.64)
U
(5.33) (14.480) (0.65)
NIL
0.16 (66.28) (2.49)

4,708 71 317 16 400 1 5 1 127 111


ni
Bano
(1.57) (2.51) (0.48) (3.87) (2.31) (0.16) (0.26) (0.16) (1.41) (2.42)
ve
24,907 55 20 13 11 3 3 54 124
Khatoon NIL
(8.30) (1.94) (0.03) (3.14) (0.06) (0.49) (0.16) (0.60) (2.71)
rs
11,100 377 13 86 399 1 9 78 49
Begum NIL
ity
(3.70) (13.33) (0.02) (20.82) (2.30) (0.16) (0.47) (0.86) (1.07)

9,270 02 11 08
Mai NIL NIL NIL NIL NIL NIL
(3.09) (0.07) (0.12) (0.17)
Pr
8,402 05 25 8 68 1 10 65 26
Jan NIL
es
(2.80) (0.18) (0.04) (1.94) (0.39) (0.16) (0.53) s (0.72) (0.57)
Punjab-F Punjab-F KPK-F KPK-F Sindh-F Elite-F Elite-G Balo-
Poor-F
O1950s 1990s 1950s 1990s 1990s 1950s 2000s chistan-F
Badeen-F

Total (300,000) (2,828) (65,532) (413) (17,333) (614) (1,900) (608) (9,028) (4,584)
xf
4,077 56 227 9 86 5 04 18 173
-Nissa NIL
or
(1.36) (1.98) (0.35) d (2.18) (0.49) (0.26) (0.66) (0.20) (3.77)

1,819 09 48 U 1 12 1 2 02
-Dil NIL Nil
(0.61) (0.32) (0.07) (0.24) (0.07) (0.05) (0.33) (0.04)

7,214 19 624 9 1,051 04 22 5 610 70


-Gul
ni
(2.40) (0.67) (0.95) (2.18) ve(6.06) (0.65) (1.16) (0.82) (6.75) (1.53)

448 117 54 7 15 10 04
Khanam Nil NIL NIL
(0.15) (4.14) (0.08) (1.69) (0.09) (0.11) (0.09)
rs
Names 176,579 823 1,746 171 4,551 14 55 13 6,957 681
ity
Percent-
58.86 29.10 2.66 41.39 26.25 2.28 2.89 2.14 77.06 14.86
age
NB: The percentages refer to the proportion of people having rural/old-fashioned components in their names.
Pr
es
s
O Annexure-L
Frequency of Occurrence of the Names of the Ahl-e-Bait
xf
Frequency of occurrence of the names given below out of the total number of names is expressed in numbers.
The numbers in the parentheses are percentages calculated to two decimal places.
or
Poor-M
d
Elite-M KPK-M KPK-M Punjab-M Punjab-M
1950s 1950s 1990s 1950s 1990s
N=500,000
N=3,649
U N=63,013 N=42,692 N=22,464 N=65,331
18,333 484 11,068 7,926 9,768 21,232
Muḥammad
(3.67) (13.26) (17.56) (18.57) (43.48) (32.50)
ni
102,640 210 3,421 3,053 1,774 8,733
Ali
(20.53) (5.76) (5.43) (7.25) (7.90) (13.37)
ve
28,013 126 1,522 1,209 1,833 2,121
Hussain
(5.60) (3.45) (2.42) (2.83) (8.16) (3.25)
rs
8,876 105 362 359 234 1,593
Hassan
(1.78) (2.88) (0.57) (0.84) (1.04) (2.44)
ity
UC-F(P) MC-F(KPK) MC-F(KPK) MC-F(P) MC-F(P)
WC-F
1950s 1950s 1990s 1950s 1990s
300,000
1,900 413 17,333 2,828 65,532
Pr
7,127 7 06 292 66 1,938
Fatima
(2.38) (0.37) (1.45) (1.68) (2.33) (2.96)
es
Sources: The above data is from both male and female samples of the years given above. s
O Annexure-M
BADEEN (SIND)—First Names   M 1
Male names 1940s–50s Male names 1960s–70s Female names 1960s–70s
xf
Total 5,208 Badeen-M 1940s Total 5,208 Badeen-M 1960s–70s Total 4,584 Badeen-F 1960s–70s
or
No. Names Frequency Percentage
d Names Frequency Percentage Names Frequency Percentage
1. Muḥammad 1,997 38.34% Muḥammad 1,453 27.90% Fatima 152 3.32%
2. Ali 374 7.18% Ali U 983 18.87% Noor 126 2.74%
3. Ahmad 255 4.90% Ahmad 470 9.02% Sakeena 96 2.09%
4. Hussain 142 2.73% Hussain 191 3.67% Amina 91 1.99%
ni
5. Ramzan 134 2.57% Hassan 104 2.00% Parveen 89 1.94%
6. Ibraheem 104 2.00% Noor 101 1.94% Mariam 80 1.75%
ve
7. Allah Dino 101 1.94% Raheem 84 rs 1.61% Hawwa 58 1.27%
8. Usman 98 1.88% Ramzan 81 1.56% Naseem 57 1.24%
Zaib-un-
ity
9. Ismail 93 1.79% Allah Dino 68 1.31% Nisa & 56 1.22%
Haleema
Jannat &
10. Hassan 90 1.73% Nawaz 63 1.21% 54 1.18%
Pr
Zarina
Names in the top ten
  65.06%   69.09%   18.74%
es
places  s
O SIND—First Names   M 2
Male names 1940s–50s
xf Male names Late 1990s Female names Late 1990s
Total 8,178 Sindh-M 1950s Total 42,459 Sindh-M 1990s Total 21,765 Sindh-F 1990s
No. Names Frequency Percentage Names Frequency Percentage Names Frequency Percentage
or
1. Muḥammad 2,776 33.94
d Ali 12,977 30.56% Parveen 624 2.87%
2. Ali 742 9.07% Muḥammad 5,743 13.53% Fatima 410 1.89%
U
3. Ahmad 615 7.52% Ahmad ni 5,340 12.58% Amina 375 1.72%
4. Hussain 304 3.72% Hussain 3,618 8.52% Anam 321 1.47%
5. Raheem 191 2.34% Raza 548 1.30% Saima 310 1.42%
ve
6. Umar 137 1.68% Saddam 515 1.21% Sana 306 1.41%
rs
7. Hassan 128 1.57% Amir 504 1.19% Shazia 254 1.17%
8. Ibraheem 123 1.50% Rahman 473 1.11% Asma 243 1.12%
ity
9. Usman 116 1.42% Hasan 456 1.07% Kainat 239 1.10%
10. Ismail 105 1.28% Imran 438 1.03% Hira 236 1.08%
Pr
Names in the top ten
  64.04%   72.10%   15.25%
es
places 
s
O BALOCHISTAN—First Names   M 3
Male names 1940s–50s Male names 1960s–80s Female names 1960s–80s
Total 22,799 Balochistan-M Total 13,771 Balochistan-M Total 9,028 Balochistan-F
xf
1940s 1960s–80s 1960s–80s
or
No. Names Frequency Percentage
d Names Frequency Percentage Names Frequency Percentage
1. Muḥammad 7,552 33.12% Muḥammad 4,546 33.01% Gul 540 5.98%
2. Gul 777 3.41%
U
Jaan 498 3.62% Fatima 171 1.89%
3. Noor 550 2.41% Gul 398 2.89% Noor 163 1.81%
ni
4. Sher 459 2.01% Noor 300 2.18% Akhtar 126 1.40%
5. Ali 407 1.79% Ahmad 265 1.92% Amina 94 1.04%
ve
Zahida,
6. Ahmad 354 1.55% Ali 227 1.65% 90 1.00%
rs Hameeda
7. Abdullah 340 1.49% Sher 218 1.58% Sabra 84 0.93%
ity
8. Raheem 326 1.43% Abdullah 193 1.40% Ayesha 78 0.86%
9. Ameer 324 1.42% Raheem 170 1.23% Sakeena 73 0.81%
10. Akhtar 286 1.25%  Ameer 149 1.08% Feroza 70 0.78%
Pr
Names in the top ten
  49.89%   50.56%   16.50%
es
places  s
O
ALL PAKISTAN POOR AND RURAL PEOPLE—First Names   M 4
Male names 1950s–1990s Female names 1950s–1990s
xf
Total 500,000 Poor-M Total 300,000 Poor-F
No. Names Frequency Percentage Names Frequency Percentage
or
1. Ali 102,640
d 20.53% Fatima 6,908 2.30%
2. Ahmad 41,260
U 8.25% Gul 4,185 1.40%
3. Hussain 28,013 5.60% Batool 2,291 0.76%
ni
4. Muḥammad 18,333 3.67% Zehra 1,896 0.63%
ve
5. Abbas 13,495 2.70% Parveen
rs 1,710 0.57%
6. Rahman 11,832 2.37% Kulsoom 1,053 0.35%
7. Hassan 8,876 1.78% Zainab 798 0.27%
ity
8. Gul 7,731 1.55% Aasiya 777 0.26%
9. Haider 5,089 1.02% Ayesha 682 0.23%
Pr
10. Iqbal 5,070 1.01% Jannat es 653 0.22%
  Names in the top ten places  48.48%   6.99%
s
O ELITE PUNJAB—First Names   M 5
Boys names 2000s Girls names 2000s
Total 512 Elite-B 2000s Total 608 Elite-G 2000s
xf
No. Names Frequency Percentage Names Frequency Percentage
or
1. Muḥammad 93
d 18.16% Fatima 62 10.20%
2. Ali 69 13.48% Noor 22 3.62%
3. Ahmad 49
U 9.57% Minahil 18 2.96%
4. Omar 19 3.71% Mahnoor 17 2.80%
ni
5. Abdullah 15 2.93% Ayesha 16 2.63%
6. Hassan 13 2.53% Amina 14 2.30%
ve
7. Ahsan 10 1.95% Zainab
rs 12 1.97%
Zain, Salman, Usma, Hamza
8. 8 1.56% Eiman & Mariam 10 1.64%
& Hussain
ity
9. Fahad, Ayan, Amir 7 1.37% Rida 9 1.48%
Tariq, Faisal, Faheem,
10. Rizwan, Saad, Saleem, 6 1.17% Khadija & Ramnha 8 1.32%
Pr
Haider, Jameel & Abid
  Names in the top ten places  56.45%   30.92%
es
s
O
Male names 1940s–50s Female names 1940s–50s
Total 3,648 Elite-M 1950s Total 1,900 Elite-F 1950s
No. Names Frequency Percentage Names Frequency Percentage
xf
1. Muḥammad 560 15.35% Amina 57 3.00%
or
2. Ahmad d478 13.10% Rubina, Sameena 38 2.00%
3. Ali 211 5.78% Yasmeen 37 1.95%
4. Hussain 126
U 3.45% Shahida 32 1.68%
5. Javed 116 3.18% Ghazala, Nighat 31 1.63%
ni
6. Khalid 109 2.99% Fauzia 30 1.58%
7. Mahmood 105 2.88% Ayesha, Shahnaz 29 1.53%
ve
8. Saeed, Hasan 85 2.33% Fareeda
rs 27 1.42%
9. Iqbal 84 2.30% Rukhsana 26 1.37%
10. Tariq 78 2.14% Shahla 25 1.32%
ity
  Names in the top ten places  53.50%   17.48%
Pr
es
s
O
PUNJAB LOWER MIDDLE & MIDDLE CLASS—First Names   M 6
Male names Late 1990s Female names Late 1990s
xf
Total 65,331 Punjab-M 1990s Total 65,532 Punjab-F 1990s
No. Names Frequency Percentage Names Frequency Percentage
or
1. Muḥammad
d
21,209 32.46% Iqra 2,465 3.76%
2. Ali 8,713
U 13.34% Ayesha 2,162 3.30%
3. Ahmad 5,361 8.21% Amina 2,014 3.07%
ni
4. Hussain 2,119 3.24% Saba 1,437 2.19%
ve
5. Usman 1,727 2.64% Anam
rs 1,432 2.19%
6. Raza 1,693 2.59% Sidra 1,208 1.84%
7. Hassan 1,545 2.36% Rabia 1,194 1.82%
ity
8. Bilal 1,267 1.94% Maryam 1,147 1.75%
9. Iqbal 1,065 1.63% Kiran 939 1.43%
Pr
10. Abbas 1,038 1.59% Sadia 932 1.42%
es
  Names in the top ten places  70.01%   1,038 22.77%
s
O
Male names 1940s–50s Female names 1940s–50s
Total 22,464 Punjab-M 1950s Total 2,828 Punjab-F 1950s
xf
No. Names or Frequency Percentage Names Frequency Percentage

1. Muḥammad 9,762
d 43.46% Akhtar 367 12.98%
2. Ahmad 3,134 13.95% Sultana 163 5.76%
3. Hussain 1,833
U 8.16% Parveen 145 5.13%
4. Ali 1,472 6.55% Naseem 127 4.49%
ni
5. Bashir 692 3.08% Shameem 120 4.24%
ve
6. Iqbal 469 2.09% Surayya
rs 89 3.15%
7. Aslam 434 1.93% Fatima 66 2.33%
8. Anwar 416 1.85% Khalida 58 2.05%
ity
9. Rashid 406 1.81% Jabeen 56 1.98%
10. Nazir 385 1.71% Razia 52 1.84%
Pr
  Names in the top ten places  84.59%   43.95%
es
s
O
KPK LOWER MIDDLE & MIDDLE CLASS—First Names   M 7
Male names Late 1990s Female names Late 1990s
xf
Total 42,692 KPK-M 1990s Total 17,333 KPK-F 1990s
No. Names Frequency Percentage Names Frequency Percentage
or
1. Muḥammad
d 7,926 18.57% Gul 962 5.55%
2. Ali 3,066
U 7.18% Ayesha 397 2.29%
3. Ahmad 2,723 6.38% Sana 319 1.84%
ni
4. Rahman 1,589 3.72% Amina 313 1.81%
ve
5. Hussain 1,209 2.83% Fatima
rs 289 1.67%
6. Jan 727 1.70% Mariam 284 1.64%
7. Gul 666 1.56% Hira 226 1.30%
ity
8. Bilal 558 1.31% Huma 218 1.26%
9. Saeed 557 1.30% Saba 210 1.21%
Pr
10. Amir 500 1.17% Saima 204 1.18%
es
  19,521 45.73% 3,422 19.74%
s
O
Male names 1940s–1950s Female names 1940s–1950s
Total 63,013 KPK-M 1950s Total 413 KPK-F 1950s
xf
No. Names or Frequency Percentage Names Frequency Percentage

1. Muḥammad 11,068
d 17.56% Akhtar 55 13.32%
2. Gul 4,429 7.03% Parveen 33 7.10%
3. Ali 3,359
U 5.33% Naseem 29 7.02%
4. Rahman 2,534 4.02% Sultan 12 2.91%
ni
5. Ahmad 2,235 3.55% Shameem 11 2.66%
ve
6. Saeed 1,251 1.99% Nasreen
rs 10 2.42%
7. Jan 1,884 2.99% Mumtaz 9 2.18%
8. Sher 1,564 2.48% Razia, Surriya 8 1.94%
ity
9. Hussain 1,522 2.42% Iffat, Khursheed, Rashida 7 1.69%
10. Wali 958 1.52% Fatima, Shahnaz, Fareeda 6 1.45%
Pr
  30,804 48.89%   180 43.58%
es
s
Annexure-N

SURVEY OF ATTITUDES TOWARDS

s
NAMES (2012)

es
The method used for taking this non-random sample and the

Pr
profile of the respondents has been described in the section
on methodology in the ‘Introduction’. However, it should be
added that the results for language need an explanation. Upon
ity
probing it was discovered that Punjabi speaking respondents give
Urdu as their mother tongue on the assumption that Urdu is a
rs
sophisticated form of Punjabi. Moreover, the English medium
ve

schools students of Lahore often report not knowing Punjabi


for which there is much shame. Thus the figures for Urdu in
ni

the Punjab are much higher than one would expect given that
its mother tongue speakers, concentrated in Karachi, are only 7.6
U

per cent (Census-1998: 107).


d

After this necessary explanation the answers to the questionnaire


or

with percentage figures is reproduced below (N= 315)


xf

1. Socio-economic Class
O

Lower middle 56 (17.78 %)


Middle 92 (29.21 %)
Upper middle 36 (11.43 %)
Upper 31 (9.84 %)
No reply 100 (31.75 %)
A ppendices 195

2. Age
11–20 132 (41.90 %)
21–30 132 (41.90 %)
31–40 24 (7.62 %)
41–50 07 (2.22 %)
51–Over 14 (4.44 %)

s
No reply 06 (1.90 %)

es
3. Education Level

Pr
a. Under Matric 117 (37.14 %)
b. Matric 18 (5.71 %)
c. Inter 14 (4.44 %)
ity
d. BA 85 (26.98 %)
e. MA 64 (20.32 %)
rs
f. Other 06 (1.90 %)
ve

g. No reply 11 (3.49 %)

4. Gender
ni

a. Male 82 (26.03 %)
U

b. Female 233 (73.97 %)


d

5. Income
or

a. Up to 5,000 39 (12.38 %)
xf

b. 5,001–15,000 55 (17.46 %)
c. 15,001–30,000 56 (17.78 %)
O

d. 30,001–70,000 69 (21.90 %)
e. 70,001–100,000 23 (7.30 %)
f. 100,001 to 200,000 36 (11.43 %)
g. Above 200,000 31 (9.84 %)
h. No reply 06 (1.90 %)
196 A ppendices

6. Mother Tongue
Urdu 109 (34.60 %)
Punjabi 76 (24.13 %)
Pashto 57 (18.10 %)
Sindhi 57 (18.10 %)
Siraiki 0

s
Balochi 01 (0.32 %)

es
Brahvi 0
Others 15 (4.76 %)

Pr
7. What is the meaning of your name?
Correct answer 124 (39.37 %)
ity
Incorrect answer 75 (23.81 %)
No reply 116 (36.83 %)
rs

8. Were you given a pet/nickname?


ve

Yes 167 (53.02 %)


ni

Some time 25 (7.94 %)


Never 89 (28.25 %)
U

Don’t want to write 29 (9.21 %)


No reply 05 (1.59 %)
d
or

9. Were you given a nickname as a joke?


xf

Yes 70 (22.22 %)
Some time 25 (7.94 %)
O

Never 79 (25.08 %)
Don’t want to write 123 (39.05 %)
No reply 18 (5.71 %)

10. How do you feel feel about your name?


a. I like it very much 206 (65.40 %)
b. I like it 73 (23.17 %)
A ppendices 197

c. Neither like/not dislike 17 (5.40 %)


d. Dislike very much 09 (2.86 %)
e. Don’t Know 10 (3.17 %)
f. No reply 0

11. Who gave you your name?


a. Grand Father (Paternal) 35 (11.11 %)

s
b. Grand Mother (Paternal) 26 (8.25 %)

es
c. Father 104 (33.02 %)
d. Mother 61 (19.37 %)

Pr
e. Family Friends 10 (3.17 %)
f. Relatives (Father’s side)
ity 43 (13.65 %)
g. Relatives (Mother’s side 34 (10.79 %)
h. Don’t Know 01 (0.32 %)
rs
i. No reply 01 (0.32 %)
ve

12. If you have changed your name what was your
previous name?
ni

Yes 151 (47.94 %)


U

No reply 164 (52.06 %)


d

13. Why did you change your name?


or

a. Previous name unlucky/ 24 (7.62 %)


Heavy/inauspicious
xf

b. Name too long 01 (0.32 %)


O

c. Fit in better with majority community 104 (33.02 %)


d. Avoid negative discrimination 03 (0.95 %)
e. I like my new name better 16 (5.08 %)
f. Fit in better with friends 04 (1.27 %)
g. It was fashionable 03 (0.95 %)
h. It was smart 07 (2.22 %)
i. Sounded beautiful 04 (1.27 %)
198 A ppendices

h. Sounded modern 04 (1.27 %)


k. Any other reason 14 (4.44 %)
l. No reply 131 (41.59 %)

14. Did you change your surname after marriage (women


only) If yes, did you adopt? FIRST or Surname.
First 45 (14.29 %)

s
Surname 27 (8.57 %)

es
No reply 243 (77.14 %)

Pr
15. Have you ever adopted a nickname yourself? If
yes, tick reasons. ity
a. To fit in better with friends 13 (4.13 %)
b. Fashionable 02 (0.63 %)
rs
c. Smart 10 (3.17 %)
d. Beautiful 14 (4.44 %)
ve

e. Modern 03 (0.95 %)
f. Margority (Fit in) 07 (2.22 %)
ni

g. Avoid discrimination 02 (0.63 %)


U

h. Like new name better 24 (7.62 %)


i. Any other reasons 26 (8.25 %)
d

j. No reply 214 (67.94 %)


or
xf
O
Glossary

Diacritical marks are given here to indicate orthography as

s
well as pronunciation. Please note that in the text of this book,

es
according to the orthographic conventions of Pakistan, long
vowels are depicted by doubling the vowel in question and not

Pr
the symbols used here. To avoid confusion some frequently used
names are spelled according to both conventions here.

Abbā (U) Father.


ity
Adīb (A) Writer.
rs
‘Affān (A) One who pardons.
Ahl-i-Bae’t (A) People of the house. This expression
ve

refers to Prophet Muhammad’s daughter


Fatima, her husband Ali, and their sons
ni

Hasan and Hussain.


U

Ahmadī (A) Follower of Mirza Ghulam Ahmed (d.


1908) who are also called Mirzais and
d

Qadianis. They are considered heretical


or

by the majority of the Muslim ulema and


were declared a non-Muslim minority
xf

by the Government of Pakistan in 1974.


O

Also spelled Ahmedi.


Aḥsan (A) Very beautiful, excellent, very subtle.
Akbar (A) Great; majestic.
‘Ain (A) Eyes.
‘Ajab (A) Strange; unique.
‘Allāmā (A) Savant; great scholar.
‘Amir (A) One who populates, to fill up. Summit
of a tree in Hebrew, ruler.
200 Glossary

‘Amīran (H-U) Rich (female form).


Anam (A) God’s grace.
Anbahi (Sans) Measure of any kind; metrical unit.
Meaning in Arabic not known.
Andrūn (P) Inside.
Anīlā (U) Inexperienced.
Anītā (Hebrew) Grace. In Spanish, graceful.

s
Anjum (A) Stars.

es
Ansārī (A) Helper; In Islamic history the people
of Madina who helped the Prophet

Pr
Muhammad and his companions when
they immigrated to the city of Yasrab,
ity
which later came to be called Madinat
un Nabi (City of the Prophet).
rs
Anwar (A) Luminous.
Ārā (P) Ornament, one who enhances beauty.
ve

Arbāb (A) One who preserves and takes care.


Arsalān (T) Lion.
ni

Asad (A) Lion.


U

Ashōk (Sans) Without sorrow, not causing sorrow.


Ashraf (A) High, upper caste, gentry.
d

Aslam (A) Safer; freer.


or

‘Asmat (A) Chastity.


Aurangzēb (P) Adorning the throne.
xf

Ayān (A) God’s gift, fortunate.


O

Azam (A) Great.


Azmarai (Pa) Lion.
Bāchā (Pa) King.
Bachchē (H-U) Children.
Bādal (H-U) Cloud.
Badar (P) Moon.
Bādshāh (A) King, sultan.
Glossary 201

Bāgal (S) Naïve woman.


Bāgh (P) Garden.
Bahādur (H-U) Brave.
Bahen (H-U) Sister.
Bakhtāwar (P) Lucky.
Balōch (B) From Balochistan. Origin disputed: ‘loud
cry’. In Sanskrit ‘power, high’.

s
Banāras A city in India.

es
Barq (A) Lighting, electricity.
Bashīr (A) Deliverer of good news.

Pr
Batūr (Pa) Brave eagle.
Bāz (H-U) Eagle. ity
Bēlā (H-U) A fragrant flower. In Hungarian ‘inner
part, within, white, distinguished’. In
rs
Sanskrit ‘time, limit, wave, violin’. It also
means portion of land between rivers in
ve

Pashto.
Bhābī (H-U) Brother’s wife.
ni

Bhagat (Sans) Devoted one.


U

Bhārī (H-U) Heavy.


Bhōgō (S) Luck.
d

Bilāl (A) Moistening.


or

Bilāwal (S) A Sindhi saint.


Billā/Billī (H-U) Cat (male and female).
xf

Bīnā (A) Clearsighted.


O

Bindiā (H) Mark on the forehead.


Bīnish (P) Wisdom, sight, intelligent. Also spelled
Beenish.
Birādarī (H-U) Fraternity.
Brekhnā (Pa) Lighting, thunder.
Buddhū (H-U) Fool.
Bushrā (A) Happy news.
202 Glossary

Būtā (Pu-U) Plant.


Chākar (B; Pu) Servant.
Chanbēlī (H-U) A sweetsmelling flower.
Chānd (H-U) Moon. Called Chan in Punjabi.
Chandan (H-U) Of the moon.
Chāndī (H-U) Silver.
Chāndnī (H-U) Moonlight.

s
Chanesar (S) Jam Chanesar, ruler 14th Century, in

es
the legend of Dodo Chanesar in Shāh
jō Risālō.

Pr
Chattān (H-U) Rock.
Chaudhary (H) Holder of fort. The origins are not clear.
ity
Chughtāi (T) From the race of Changez Khan.
Dādā (H-U) Grandfather.
rs
Dāgh (P) Mark of burning.
Dāhir (S) Proper name. Origin not known.
ve

Dard (P) Pain.


Dārū (S) Pomegranate. In Hindi it means
ni

medicine as well as alcohol.


U

Daryā (H-U) River.


Dēvī (Sans) Goddess.
d

Dhāt (H-U) Caste. Pronounced in Urdu as Zāt.


or

Dilāwar (P) Brave. One with a heart.


Dildār (P) One who pleases the heart, beloved.
xf

Diljān (P) Dear to the heart, beloved.


O

Dilshād (P) Happy of the heart; one who makes one’s


heart happy.
Dōdā (S) Fourhorned antelope.
Dōdō (S) A kind of bread. As a name some people
trace it to Swahili meaning plump and
lovable. Found in the tale of Dodo-
Chanesar.
Glossary 203

Dulhan (H-U) Bride.


Dushkā (Russian) Sweet, dear.
‘Enāyat (A) Favour. Also spelled Inayat.
Fahad (A) Panther, leopard.
Fahīm (A) Intelligent; wise.
Faiḍ (P) Grace, favour.
Faiṣal (A) Decisive, arbitrator.

s
Fakhrū (A) Pride of religion.

es
Faqīr (A) Beggar, mendicant.
Farah (P) Happy, happiness.

Pr
Fātēḥ (A) Triumph; conquest; opening.
Fayyāẓ (A) Very generous, liberal.
ity
Faẓal (A) Excellence, superiority, reward, grace.
Ghālib (A) Dominant.
rs
Ghar (H-U) House.
Ghasītā (H-U) Dragged.
ve

Ghulām (A) Boy, slave.


Guddī/ō (Pu) Doll.
ni

Guhram (B) Baloch proper name. Origin unknown.


U

Gul (P) Flower.


Gulāb (P) Rose.
d

Gulālāi (Pa) Like flowers, lovely.


or

Gulfām (P) Of the form of a flower.


Gulfishã (P) Scatterer of flowers. Normally spelled
xf

Gulfishan.
O

Gulistān (P) Garden (place of flowers)


Gulmakaī (Pa) Flower of the maize plant.
Gulrukh (P) Face as beautiful as a flower.
Gulshan (P) Garden.
Gulzār (P) Garden.
Guṛiyā (H-U) Doll.
Ḥajjāj (A) Argumentative.
204 Glossary

Ḥānā (Hebrew) Variant form of Hannah. In Arabic it


means ‘unique young woman’. Happiness.
Ḥarb (A) War.
Harī (Sans) One who steals or takes away [sufferings];
colour (yellow, green etc)
Ḥazrat (A) Presence, dignity, power.
Ḥinā (A) Henna.

s
Hīr (S) Cool breeze. Generally spelled as Heer.

es
Hīra (H-U) Diamond.
Hōshī (S) Origin unknown. In Japanese it means

Pr
‘star’.
Humā (P) A mythical bird which makes a person a
ity
king if it sits on his head.
‘Id (A) Muslim festival. Normally spelled as Eid.
rs
‘Ināyā (A) Help; care.
Iqbāl (P) Great.
ve

‘Irfān (A) Knowing, aware.


Ishtiāq (A) Longing, craving.
ni

Itwār (U) Sunday.


U

Jabīn (A) Forehead, face.


Jāfar (A) Rivulet. Also spelled as Jaffar.
d

Jahãgīr (P) One who subdues the earth. Normally


or

spelled as Jahangir.
Jahān (P) World.
xf

Jamīl (A) Beautiful.


O

Jān (P) Life; dear.


Jarnaēl (Pu) Variant pronunciation of the rank of
general in the army.
Jasmine (English) Name of a flower. Yasmin in Urdu.
Jāvēd (P) Eternal.
Jawāhar (P) Pearls.
Jī (H-U) Used for respect.
Glossary 205

Jīvan (Sans) Life. Also spelled Jeevan.


Julāhā (H-U) Weaver.
Jum‘erāt (U) Thursday.
Jum‘ā (A) Friday.
Kālā (H-U) Black, dark.
Kamāl (A) Unique; accomplished; skilful.
Kandērō (S) Thorny bushes.

s
Kandō (S) Thorny bushes.

es
Kanwal (H-U) Lotus.
Kāshif (A) Revealer, discoverer, pioneer.

Pr
Kēsar (P) Indigenized form of Qaisar which means
Ceasar. Ruler.
ity
Khadēṛā (H-U) Dragged.
Khairāt (A) Alms.
rs
Khālidā (A) Immortal, deathless.
Khānā (P) House.
ve

Kilar (S) Thorny tree with medicinal value.


Kiran (U) Ray.
ni

Kirar (S) Thorny plant.


U

Kōr (Pa) House.


Kullū (H-U) Dark complexioned.
d

Kumār (Sans) Young man, son.


or

Kumārī (Sans) Princess, used as a name for girls.


Lailā (A) Of the night, flowers which give their
xf

fragrance at night.
O

Lāl (H-U) Red.


Lallu (H-U) Foolish, incompetent.
Laung (S-U) Clove.
Lesna (Pa) Eye lashes.
Lian (S) Also spelled Leelan. From the tale of
Dodo-Chanesar. She was the wife of
Chanesar.
206 Glossary

Madhūbālā (Sans) Sweet maiden.


Madrassā (A) Islamic seminary.
Māh (P) Moon.
Mahal (P) Palace.
Mahjabīn (P) Having a brow like the moon; beautiful
girl.
Malālā (Pa) From ‘malal’ which is Persian for sorrow.

s
Malghrala (Pa) Diamond.

es
Manāl (A) A place where one finds something;
achievement

Pr
Mangal (H-U) Tuesday.
Manzar (P) Scene. ity
Māriā (A) Faircomplexioned women. Also spelled
Māriyāh.
rs
Marīnā (Latin) From the sea.
Mārvī (S) Like a moon. From the heroine of Shah
ve

Abdul Latif who was abducted by the


Somra prince Umar.
ni

Masīh (P) Jesus; Messiah.


U

Maulānā (A) My Lord; used for Muslim clergymen.


Mē’rāj (A) Staircase; ascendance.
d

Meher (P) Love, friendship, intimacy.


or

Miftāh (A) Key.


Mīnā Jabīnā (Pa) Honeylike love. Also spelled Meena
xf

Jabeena.
O

Minā (Sans) Fish.


Mināhil (A) Door to heaven.
Mīr (A) Incharge of water; a title used for syeds.
Also spelled Meer.
Mīrā (A) Aristocratic Lady, High born girl,
princess. Also spelled Meera.
Mīran (H-U) Pet name from Meer and Ameer=rich.
Also spelled Meeran.
Glossary 207

Mīrwais (Pa) Hero in a Pashto legend.


Mirzā (P-U) Abbreviated form of Ameer Zada (son
of a rich man)
Mo‘inuddīn (A) Assister of the faith.
Mōmal (Siraiki) Internal and external beauty. In the tale
of Momal and Rano in Shāh jō Risālō.
Mōnā (A) Wish, variant of ‘Hana’. In Sanskrit it

s
means ‘alone’

es
Mōtī (H-U) Gem.
Mumtāz (A) Distinguished.

Pr
Munnā (U) Small boy. Sonny.
Munshī (P) Clerk. ity
Murrāh (A) Bitter.
Mushtāq (A) Longing, desirous.
rs
Nā’il (A) Winner, one who finds what he/she
seeks.
ve

Nadvī (A) From the Islamic seminary Nadwat ul


‘Ulamā in Lucknow.
ni

Nainā (H-U) Eyes.


U

Najam (A) Star.


Najmā (A) Star (feminine).
d

Namāzan (P) One who says her prayers.


or

Nanhī (H-U) Little one (used for girls).


Nānī (H-U) Maternal grandmother.
xf

Nasīm (P) Breeze. Also spelled Naseem.


O

Navīd (P) Good news. Also spelled Naveed.


Nawāz (A) One who bestows favours.
Nazar (A) Eyesight. Vision.
Nazīr (A) Luminously beautiful; example.
Neha (Sans) From Sanskrit Sneha meaning ‘cherished,
beauty, adorable’. Others trace it to nehal
which means ‘rain’. In Marathi it means
‘good eyes’.
208 Glossary

Nīlī (H-U) Blue. Neel is indigo.


Nīnā (Hebrew) Little girl. Also spelled Neena. In Hindi
it means beautiful eyes.
Nūr Jahã (P) Light of the world. Also spelled Noor
Jahan.
Osāmā (A) Feline predator. Also spelled Usama.
Pahāṛ (H-U) Mountain.

s
Pajmā (B) Not known.

es
Palwashā (Pa) Ray.
Pandit (Sans) Hindu priest.

Pr
Pari (P) Fairy.
Parkhā (Pa) Dew. ity
Parvēz (P) Conqueror, lucky, eminent.
Parvīn (A) Variant Of Parveen: Pleiades, Cluster of
rs
stars.
Pathān (H-U) Pashto-speakers. An outsiders’ word for
ve

the races of Afghanistan and KPK which


call themselves Pukhtanā, Pashtūn or
ni

Afghān.
U

Phūl (H-U) Flower. Also spelled Phool.


Pūjā (Sans) Worship. Also spelled Pooja.
d

Punnū (Sans) One who does good.


or

Qādir (A) Capable, competent.


Qādrī (A) From Sheikh Abdul Qadir Jilani.
xf

Qamar (A) Moon.


O

Qaōm (A) Nation, group.


Qārī (A) Reader, reciter.
Qudrat (A) Power, control, domination, right.
Qurrat (A) Light of (Qurrat ul ‘Ain means light of
the eyes).
Rabail (S) A flower.
Rafī’ (A) Distinguished, high.
Glossary 209

Rafīq (A) Kind, ally, partner, friend.


Rahīmā (A) Female form of the Arabic Raḥīm used
in Urdu. Compassionate, kind.
Rām (Sans) Pleasing, supreme.
Ramaḍān (A) Muslim month of fasting. Also spelled
Ramzan.
Ramīz (A) Dignified. Another meaning in some

s
sources is spear. Also spelled Rameez.

es
Ramōnā (Spanish) Protector. In old German ‘protecting
hands’. It was popularized by Helen

Pr
Hunt Jackson’s novel Ramona (1884).
Variants in spelling are Romona,
ity
Romana, etc. In Pakistan the etymology
of the name given in popular books
rs
traces it back to romance, romantic, etc.
Rāshid (A) Wise counselor, righteous, mature.
ve

Rāwal (Siraiki) Title used by Rajputs and Brahmins but


now used both as caste and first name in
ni

Pakistan’s Siraiki-speaking area.


U

Razīn (P) Polite. Also spelled Razeen.


Rēsham (P) Silk.
d

Ridā (A) One who is helped by God.


or

Rīhā (P) Released, air. Also spelled Reeha.


Rīmā (A) White antelope. Also spelled Reema.
xf

Rōhal (Siraiki) Not known.


O

Ruphō (S) Not known.


Rustam (P) Strong, tall
Ruswā (P) Notorious.
Sa‘ādat (A) Honour.
Sabā (A) Queen of Sheba; breeze.
Sabīhā (P) Fair complenioned, beautiful.
Sachal (Siraiki) Truthful
210 Glossary

Saddām (A) One who confronts.


Safā (A) A place near Mecca.
Sāhib (A) Respectable. Used as a title in Urdu/
Hindi to show respect.
Ṣāi’mā (A) One who fasts.
Saif (A) Sword.
Sajāwal (Siraiki) Decorated.

s
Sājhō (S) Not known.

es
Samīr (A) Story teller; one who talks at night. Also
spelled Sameer.

Pr
Samundar (H-U) Sea.
Ṣanam (A) Beloved. ity
Sanchā (Siraiki) Not known.
Sandrā
rs
(Pa) Lucky song.
Bakhtāwarā
Sangīn (Pa) Hard.
ve

Santōsh (H) Tranquillity, peace, pleasure.


Sārā (Hebrew) Princess; In Arabic pure, happy.
ni

Sardār (P) Chief.


U

Sāshā (Russian) Man’s defender.


Sassī (S) The heroine in the story of Sassi-
d

Punnun in Shāh jō Risālō.


or

Sāwal (S) Summer.


Sehrah (?) Not known in other languages. In H-U
xf

the headgear of flowers and other deco­


O

rative items for a bridegroom


Shab-ē-Bārāt (P) The night of happiness. A special night
of religious significance before the
fasting month of Ramaḍān.
Shaēhzād (P) Son of a king. Also spelled Shahzad.
Shaēhzādā (P) King’s son. Also spelled Shahzada.
Shafī’ (A) One who offers salvation, redeemer.
Glossary 211

Shaguftā (P) Fresh.


Shāh (P) King.
Shahāb (A) Comet.
Shāhidā (A) Female witness; witness of God’s beauty.
Shahīn (P) Eagle. Also spelled Shaheen.
Shahr (P) City.
Shāhrukh (P) Beautiful countenance; royal face.

s
Shāhryār (P) The king.

es
Shāhwar (P) Worthy of kings.
Shamīm (P) Breeze.

Pr
Shams (A) Sun.
Shamūn (A) Name of Joseph’s step-brother (Simon
ity
in English).
Shānī (Hebrew) Crimson, red.
rs
Shaukat (A) Greatness, majesty, awe-inspiring.
Shehrēzād (P) Daughter of the city dweller.
ve

Shēkh (A) Variant of Shaikh: Head, chieftain,


teacher. Also written as Sheikh.
ni

Shēr (U) Lion.


U

Shubrāt (P) Indigenous variant of Shab-ē-Bārāt.


Sīmal (A) One who requests (?) In Hebrew it
d

means respectable, acceptable to God.


or

Also said to be of Indian origin. Mean­


ing is not clear.
xf

Sīmĩ (P) Silver, beautiful like silver. Also spelled


O

Seemi.
Singh (Sans) In Sanskrit Simha means lion. It is used
as a title by Rajputs, Sikhs. etc.
Sipāh-i-
(A) The soldiers of Muhammad.
Muhammad
Sipāh-i-Sahābā  (A) The soldiers of the Companions of the
Prophet.
212 Glossary

Sōbhā (Siraiki) Fitting, appropricate.


Sōbō (S) Conqueror, winner.
Sōfia (Greek) Wisdom.
Sōhā (S) Light.
Sōjhā (Siraiki) Not known.
Sōnā (H-U) Gold.
Sōnī (Pu) Beautiful.

s
Sōniā (Greek) Wise.

es
Sūf (S) Apple. Also written as soof.
Sultānā (P) Queen.

Pr
Suntī hō/hae (H-U) Do you hear. Also written hain.
Surayyā (Hebrew) Princess. In Arabic it means “bright
ity
starlight—the Pleiades, a group of seven
stars in the constellation Taurus”.
rs
Tāhir (A) Pure, clean.
Tālib (A) Seeker (of truth), Student.
ve

Tāniā (Russian) Fairy princess.


Tāriq (A) Morning star.
ni

Tāsīr (A) Result, effect, quintessential quality.


U

Tatārā Gulōnā  (Pa) Flower of Tatara.


Tayyab (A) Pleasant, well.
d

Tayyabā (A) Pure, clean.


or

Tēlī (H-U) Provider of oil.


Thanāullāh (A) One who praises God.
xf

Tōtā (H-U) Parrot.


O

Uddīn (A) Of the religion.


‘Umair (A) Living a long life.
‘Umrāō (U) Derived from the word ‘rich’.
‘Uthmān (A) The young one of a bird, one who sings,
beautiful. Spelled as Usman and Osman
in transliteration from Urdu.
Varanā (Sans) Layers, caste system.
Glossary 213

Vīna (Sans) Lute. Also written as Veena.


Wāgmer (Pa) Fragrance.
Wālā (U) Of; associated with; provider of; owner
of.
Waqār (A) Dignity.
Wasīm (A) Handsome; excellent; beautiful. Also
written as Waseem.

s
Wazīr (P) Minister. Also spelled Wazeer.

es
Yā ‘Alī madad  (P) Oh Ali Help Us!
Zabar Jad (A) Over, above, superior; gift of God.

Pr
Zabud (B) Not known.
Zain (A) Decoration. ity
Zālān (Pa) Shining.
Zam Harīr (A) Cold hell. Hareer also means silk.
rs
Zamān (A) Time, age, destiny.
Zamīndār (P) Also official responsible for collecting
ve

rent from land; landowner.


Zamīr (A) Heart, mind, conscience. Also written as
ni

Zameer.
U

Zamurrad (A) Green precious stone, emerald. The word


is said to have come from some African
d

language but this is not authenticated.


or

Zarghunā (Pa) sprouting out


Zarsangā (Pa) Golden branch.
xf

Zāt (A) See dhāt.


O

Ziā (A) Light; splendour


Zoyā (Russian) Living, vivacious. It is a common name in
Russian being a variant of Zoe (life). In
Arabic it means ‘God’s gift’; in Persian
‘twilight’ and in Sanskrit ‘shining’.
Notes

1. The major sects of Muslims in Pakistan are the Sunnis

s
and Shias. The census does not give figures for sects but,

es
according to Qasim Zaman, a scholar of Islam in South
Asia, ‘estimates about the size of the Shi‘i population range

Pr
from as much as a quarter to less than 2 per cent of the
total population, though the more accurate figure seems
to be closer to about 15 per cent (Zaman, 2002: 113).
ity
The term Shia is from Shiān-i-Alī or partisans of Hazrat
Ali (ra), those who believe that Hazrat Ali ibn Abi Talib,
rs
cousin of Prophet Muhammad (pbuh) and also his son-in-
ve

law since he was the husband of his daughter Fatima, was


deprived of his rightful position as the first caliph, by the
ni

first three caliphs whom the Sunnis regard as legitimate


and rightfully guided. Shias especially oppose the Omayyad
U

ruler Muawiyah ibn-e-Abu Sufiyan who appointed his son


Yazid as the caliph thus depriving Ali’s sons Hassan and
d

Hussain their right to be caliphs. Later Hussain was killed


or

in the Battle of Karbala (10 Oct 680) by Yazid’s forces [For


xf

the original sources see Tabari (fl. 839–923); for a highly


readable account of the battle see Hazleton, 2009: 157–211
O

Part 3 entitled ‘Hussein’; also see Jafri, 1979].


2. The Ismailis are a branch of the Shias. They believe that Ali
was succeeded by nine spiritual leaders (imams) while the
majority, called athana ashari (twelvers), believe in twelve
imams (Daftary, 1998). Deoband is a town in U. P. India,
with a famous Islamic seminary madrasah. The graduates of
this school interpret the Islamic law (Sharia) in an austere,
Notes 215

doctrinaire manner and they strictly deny the concept


of intercession by saints, except in the case of their own
pioneers, (Metcalf, 1982). Bareilly too is a city in U. P. India,
where Maulana Ahmed Raza Khan (1870–1920) justified
intercessory, mystical interpretations of Islam (Sanyal, 1996).
The Wahabis, followers of Abdul Wahab of Saudi Arabia
(1703–1792), is the name given to strict interpreters of Islam

s
in South Asia who call themselves Ahl-i-Hadith (Ahmad,

es
1966).
3. Abjad is a mnemonic device to commit the numerical values

Pr
of Arabic letters of the alphabet to memory.
4. The list of banned names issued by Saudi Arabia include:
ity
Malaak (angel), Abdul Aati, Abdul Naser, Abdul Musleh,
Binyamin (Arabic for Benjamin), Naris, Yara, Sitav, Loland,
rs
Tilaj, Barrah (To select, innocent), Abdul Nabi, Abdul
Rasool, Sumuw (highness), Al Mamlaka (the kingdom),
ve

Mallika (queen), Mamlaka (kingdom), Tabarak (blessed),


Nardeen, Sandy, Rama (Hindu god), Maline, Elaine, Inar,
ni

Maliktina, Maya, Linda, Randa, Basmala (utterance of the


U

name of God), Jibreel (angel Gabriel), Abdul Mu’een, Abrar,


Iman, Bayan, Baseel, Wireelam, Nabi (prophet), Nabiyya
d

(female prophet), Amir (prince), Taline, Aram, Nareej, Rital,


or

Alice, Lareen, Kibrial. <http://nameberry.com/nametalk/


threads/150695-Saudi-Arabia-Bans-50-Names>
xf

5. Also found in the following variant:


O

Daulat tere teen nam,


Chiddva, ched, soorakh khan,
[Wealth, you have three names,
Hole (pejorative form of the word), hole (normal form),
and Hole Khan (giving a respectable title to the word)].
Chidva is a pejorative word for hole. In ordinary Hindi-Urdu
the word for hole (ched) becomes the Persian soorakh when
216 Notes

a person gains social prominence and khan is added to the


name.
6. The term feudal is controversial as used for the landowning
elite of Pakistan as, among others, Taimur Rahman contends
(2012: 1–46).
7. Thus the present author’s wife, Rehana, used her father’s
name Tajuddin but took up the name Tariq after marriage.

s
This was later changed to Rahman, the author’s surname.

es
8. In Meera Ji’s case, the story is that he fell in love with a girl
called Meera. Because of her he also started appreciating the

Pr
poetry of Meerabai (1498–1547) (Manto, 1991: 60).
ity
rs
ve
ni
U
d
or
xf
O
References

Books and Articles:

s
Abazov, Rafis. 2007. Culture and Customs of the Central Asian

es
Republics. Westport, CT: Greenwood Press.
Abd-el-Jawad, Hassan. 1986. ‘A Linguistic and Sociocultural

Pr
Study of Personal Names in Jordan’, Anthorpological Linguistics.
28: 1 (Spring), 80–94.
Abel, Ernest L. 2013. ‘Dickensian Eponyms’, Names: A Journal
ity
of Onomastics. 61:2 ( June), 75–91.
Abu Dawud, Sulayman. (circa. 9th C) Sunan Abū Dāwūd Trans.
rs
From Arabic to English. Ahmad Hasan. 1997. Compendium
ve

of Islamic Texts. Quoted from <http://www.biharanjuman.


org/hadith/sunan-abu-dawud-english.pdf>.
ni

Aceto, Michael. 2002. ‘Ethnic Personal Names and Multiple


Identities in Anglophone Caribbean Speech Communities’,
U

Language in Society. 31:4 (Sept), 577–608.


Aggarwal, Pratap C. 1976. ‘Kinship and Marriage among the
d

Meos of Rajasthan’, Family, Kinship and Marriage among


or

Muslims in India. New Delhi: Manohar. 265–96.


xf

Ahmad, Afzaal & Ahmad, Ijaz. 2008. Nām: Bachchõ kē nām.


[Urdu: Names: Childrens’ Names] Lahore: Sang-e-Meel.
O

Ahmad, Imtiaz (ed). 1976. Family, Kinship and Marriage among


Muslims in India. New Delhi: Manohar.
Ahmad, Imtiaz. 1976. ‘Caste and kinship in a Muslim Village of
Eastern Uttar Pradesh’. Family, Kinship and Marriage among
Muslims in India. New Delhi: Manohar. 319–45.
Ahmad, Qeyamuddin. 1966. The Wahhabi Movement in India.
Revised edition. New Delhi: Manohar.
218 R eferences

Ahmed, Ishtiaq. 2012. The Punjab: Bloodied, Partitioned and


Cleansed. Karachi: Oxford University Press.
Ahmed, Khaled. 2011. Sectarian War: Pakistan’s Sunni-Shia
Violence and its Links to the Middle East. Karachi: Oxford
University Press.
Alavi, Hamza. 1972. ‘Kinship in West Punjab Villages’,
Contributions to Indian Sociology. 6:1: 1–27.

s
Alford, Richard D. 1988. Naming and Identity: A Cross Cultural

es
Study of Personal Naming Practices. New Haven, Connecticut:
HRAF Press.

Pr
Algeo, John & Katie. 2000 ‘Onomastics as an Interdisciplinary
Study’, Names. 48: 3–4, (Sep–Dec), 265–74.
ity
Ali, Syed. 2002. ‘Collective and Elective Ethnicity: Caste Among
Urban Muslims in India’, Sociological Forum. 17: 4 (Dec),
rs
593–620.
Anderson, John M. 2004. ‘On the Grammatical Status of
ve

Names’, Language. 80: 3 (Sept), 435–74.


Anderson, Benedict. 1983. Imagined Communities: Reflections on
ni

the Origin and Spread of Nationalism. Revised edition. London:


U

Verso.
Atawneh, Ahmad. 2005. ‘Family Names in Palestine: A
d

Reflection of Culture and Life’, Names. 53: 3, (September),


or

147–67.
Austin, John. L. 1962. How to do Things with Words: the William
xf

James Lectures Delivered at Harvard University in 1955.


O

Oxford: Clarendon Press.


Baarth, Frederik. 1969. ‘Introduction’, Ethnic Groups and
Boundaries: The Social Organization of Cultural Difference.
London: Allen and Unwin.
Bagnall, Roger S. 1982. ‘Religious Conversion and Onomastic
Change in Early Byzantine Egypt’, Bulletin of the American
Society of Papyrologists. 19: 105–24.
R eferences 219

Baird, Scott (2006), ‘Anglicizing Ethnic Surnames’, Names. 54:


2, ( June), 173–92.
Baloch, Imam Baksh n.d. Bachchõ kē nām. [Urdu: Childrens’
Names] Karachi: Shama Book Agency.
Barnes, R. H. 1980. ‘Hidatsa Personal Names: an Interpretation’,
The Plains Anthropologist. 25: 90 (Nov), 311–31.
Barq, Ghulam Jilani. 1982. Mērī Dāstān-ē-Hayāt. [Urdu: Story

s
of my Life] Lahore: Sheikh Ghulam Ali and Sons.

es
Bayly, Christopher. 2004. The Birth of the Modern World 1780–
1914: Global Connections and Comparisons. Oxford: Blackwell.

Pr
Beidelman, T. O. 1974. ‘Kaguru Names and Naming’, Journal of
Anthropological Research. 30: 4 (Winter), 281–93.
ity
Benson, Susan. 2006. ‘Injurious Naming: Naming, Disavowal, and
Recuperation in Contexts of Slavery and Emancipation’. The
rs
Anthropology of Names and Naming. Cambridge: Cambridge
University Press. 178–99.
ve

Bertrand, Marianne & Mullainathan, Sendhil. 2004. ‘Are Emily


and Greg more Employable than Lakisha and Jamal? A Field
ni

Experiment on Labor Market Discrimination’, The American


U

Economic Review. 94: 4 (Sep), 991–1013.


Blair, Betty A. 1991. ‘Iranian Immigrants’ Name Changes in
d

Los Angeles’, Creative Ethnicity, Symbols and Strategies of


or

Contemporary Ethnic Life. Cicala, Norman, UT: Utah State


University Press. 112–36.
xf

Bloch, Marice. 2006. ‘Tekonymy and the Evocation of the


O

“Social” Among the Zafimaniry of Madagascar’, The


Anthropology of Names and Naming. Cambridge: Cambridge
University Press. 98–114.
Blommaert, Jan. 2010. The Sociolinguistics of Globalization.
Cambridge: Cambridge University Press.
Bodenhorn, Barbara & Bruck, G. V. 2006. ‘“Entangled Histories”:
An Introduction to the Anthropology of names and Naming’,
220 R eferences

The Anthropology of Names and Naming Cambridge: Cambridge


University Press. 1–30.
Bourdieu, Pierre. 2009. The Logic of Practice. Trans. Richard Nice.
Stanford: Stanford University Press.
Britto, Francis. 1986. ‘Personal Names in Tamil Society’,
Anthropological Linguistics. 28: 3 (Fall), 349–65.
Bruck, G. V. and Bodenhorn, Barbara. 2006. The Anthropology of

s
Names and Naming. Cambridge: Cambridge University Press.

es
Bruck, G. V. 2006. ‘Names as Bodily Signs’, The Anthropology of
Names and Naming. Cambridge: Cambridge University Press.

Pr
226–50.
Bukhari, Abu Abdullah bin Ismail. (c. 9th C). Saḥīḥ al-Bukhārī
ity
[Arabic/Urdu]. Trans. Amjad ul Ali, Subhan Mahmood, Qari
Akhtar & Abul Fath. 2003. Lahore: Idara-e-Islamiyat.
rs
Bulliet, Richard W. 1979. Conversion to Islam in the Medieval
Period: An Essay in Quantitative History. Cambridge, Mass &
ve

London: Harvard University Press.


Burnard, Trevor. 2001. ‘Slave Naming Patterns: Onomastics
ni

and the Taxonomy of Race in Eighteenth-Century Jamaica’,


U

Journal of Interdisciplinary History. xxxi: 3 (Winter), 325–46.


Bursell, Moa. 2012. ‘Name Change and Destigmatization among
d

Middle Eastern Immigrants in Sweden’, Ethnic and Racial


or

Studies. 35: 3: 471–87.


Census-B. 1901. Report of the Census of India, 1901: Bengal-
xf

Extracts from District Reports Regarding Causes of Conversion to


O

Muhammedanism. Appendix 11, Vol. 6 Calcutta: Government


Press.
Census-R. 2004. Population by Religion: Pakistan. Islamabad:
Population Census Organization, Statistics Division.
Cerulo, Karen A. 1997. ‘Identity Construction: New Issues, New
Directions’, Annual Review of Sociology. 23: 385–409.
Chatterjee, Partha. 1993. The Nation and its Fragments: Colonial
and Post-colonial Histories. Princeton: Princeton University
Press.
R eferences 221

Chatterjee, Partha. 1997. Our Modernity. Lecture published by


Kuala Lumpur: Rotterdam/Dakar South-South Exchange
Programme for Research on the History of Development &
the Council for the Development of Social Science research
in Africa.
Chaudhary, M. Azam. 1999. Justice in Practice: The Legal
Ethnography of a Pakistani Punjabi Village. Karachi: Oxford

s
University Press.

es
Chelliah, Shobhana L. 2005. ‘Asserting Nationhood Through
Personal Name Choice: The Case of the Meithei of Northeast

Pr
India’, Anthropological Linguistics. 47: 2 (Summer), 169–216.
Chughtai, Asmat. 2008. Kulliyāt-e-Asmat Chughtāi. (ed. & comp.
ity
Tariq Mahmood) Lahore: Book Talk.
Colebrooke, J. 1881. ‘On the Proper Names of Mohammedans’,
rs
Journal of the Royal Asiatic Society of Great Britain and Ireland.
13: 2 (April), 237–80.
ve

Collier, George A. & Bricker, Victoria R. 1970. ‘Nicknames and


Social Structure in Zinacantan’, American Anthropologist. 72:
ni

2 (Apr), 289–302.
U

Crozier, W. Ray. 2004. ‘Recollections of Schoolteachers’


Nicknames’, Names. 52: 2 ( June), 3–19.
d

D’ Cruz, Mark. 2000. ‘A Theory of Ordinary Proper Names’,


or

Mind. 109: 436 (Oct) 721–56.


Daftary, Farhad. 1998. A Short History of the Ismailis: Traditions of
xf

a Muslim Community. Edinburgh: Edinburgh University Press.


O

Dalit. 2000. Report by International Dalit Solidarity Organization.


Retrieved on 24 March 2013. <http://www.dalits.nl.pdf/
Factsheet.Pakistan.pdf>
Dalrymple, William. 2002. White Mughals: Love and Betrayal in
Eighteenth-Century India. London: Harper-Collins.
DeCamp, David. 1967. ‘African Day-Names in Jamaica’,
Language. 43: 1 (Mar), 139–49.
222 R eferences

Dil, Afia. 1976. ‘A Comparative Study of the personal Names and


Nicknames of the Bengali-speaking Hindus and Muslims’,
Studies in Bengal. East Lansing.
Dorian, Nancy C. 1970. ‘A Substitute Name System in the
Scottish Highlands’, American Anthropologist. 72: 2 (Apr),
303–19.
Draper, Patricia & Haney, Christine. 2005. ‘Patrilateral Bias

s
among a Traditionally Egalitarian People: Ju/’hoansi Naming

es
Practice’, Ethnology. 44: 3 (Summer), 243–59.
Eaton, Richard M. 2004. ‘Approaches to the Study of Conversion

Pr
to Islam in India’, Religious Movements in South Asia 600–
1800. New Delhi: Oxford University Press, 105–27.
ity
Eisenstadt, S. N. 1999. ‘Multiple Modernities in an age of
Globalization’, The Canadian Journal of Sociology. 24: 2
rs
(Spring), 283–95.
Eisenstadt, S. N. 2000. ‘Multiple Modernities’, Daedalus 129: 1
ve

(Winter), 1–29.
Elias, Jamal J. 2011. On Wings of Diesel: Trucks, Identity and
ni

Culture in Pakistan. Oxford: Oneworld Publications.


U

Emeneau, M. B. 1976. ‘Personal Names of the Coorgs’, Journal of


the American Oriental Society. 96: 1 ( Jan–Mar), 7–14.
d

Emeneau, M. B. 1978. ‘Towards an Onomastics of South Asia’,


or

Journal of the American Oriental Society. 98: 2 (Apr–Jun),


113–30.
xf

Farooqi, Nasir Ahmad. 1968. Snakes and Ladders. Lahore: Watan


O

Publications.
Faruqi, Shamsur Rahman. 2006. Kayī Chānd thē Sar-ē-Āsmā.
[Urdu: Many Moons Were in the Heavens].
Finch, Janet. 2008. ‘Naming Names: Kinship, Individuality and
Personal Names’, Sociology. 42: 4, 709–725.
Foucault, Michel. 1982/1988. Technologies of the Self. (eds.)
Martin, Luther H, Huck Gutman and Patrick H. Hutton.
University of Massachusetts Press, 16–49.
R eferences 223

Frable, Deborrah E. S. 1997. ‘Gender, Racial, Ethnic, Sexual,


and Class Identities’, Annual Review of Psychology. 48, 139–62.
Gardiner, Alan. 1940. The Theory of Proper Names: A Controversial
Essay. London: Oxford University Press.
Gasque, Thomas J. 1991. ‘The Study of Names as Humor in
the Thirty-Nine Years of the Journal Names’, Names. 39: 3
(Sept), 217–24.

s
Gellner, Ernest. 1981. ‘Flux and Reflux in the Faith of Men’,

es
Muslim Society. Cambridge: Cambridge University Press. 1–85.
Gellner, Ernest. 1995. ‘Religion and the Profane’, Lecture in

Pr
Heidelberg in October. Transcribed by Caroline Schmidt.
Published Internationale Zeitschrift fur Philosophie. Gunter
ity
Figal and Enno Rudolf (eds.) 1/1996. Retrieved 30 Dec 2012.
<http://www.eurozine.com/articles/2000-08-28-gellner-en.
rs
html>
Ghose, Zulfikar. 1965. Confessions of a Native-Alien. London:
ve

Routledge.
Goitein, S. D. 1970. ‘Nicknames as Family Names’, Journal of the
ni

American Oriental Society. 90: 4 (Oct–Dec), 517–24.


U

Goldin, Paul Rakita. 2000. ‘Personal Names in Early China: A


Research Note’, Journal of the American Oriental Society. 120:
d

1 ( Jan–Mar), 77–81.
or

Goodfriend, Douglas. 1983. ‘Changing Concepts of caste and


status among old Delhi Muslims’, Modernization and Social
xf

Change Among Muslims in India. Delhi, Manohar, 119–52.


O

Grace, George W. 1987. The Linguistic Construction of Reality.


London: Routledge.
Hansen, Thomas Blom. 2006. ‘Where Names Fall Short: Names
as Performances in Contemporary Urban South Africa’. The
Anthropology of Names and Naming. Cambridge: Cambridge
University Press. 201–24.
224 R eferences

Hazleton, Lesley. 2009. After the Prophet: The Epic Story of the
Shia-Sunni Split. New York: Doubleday.
Hartman, Arthur A., Nicolay, Robert C. & Hurley, Jesse. 1968.
‘Unique Personal Names as a Social Adjustment Factor’, The
Journal of Social Psychology. 75: 1 ( June), 107–10.
Hefner, Robert W. 1998. ‘Multiple Modernities: Christianity,
Islam, and Hinduism in a Globalizing Age’, Annual Review

s
of Anthropology. 27 (1998), 83–104.

es
Hilka, Alfons. 1910. Die Altindischen Personennamen. [German:
Old Indian Personal Names] Breslau: Verlag.

Pr
Holland, Theodore J. 1990. ‘The Many Faces of Nicknames’,
Names. 38: 4 (December), 255–72.
ity
Hossain, Adnan. 2012. ‘Beyond Emasculation: Being Muslim
and Becoming Hijra in South Asia’, Asian Studies Review.
rs
36: 4: 495–513.
Husain, Mohammad. 2010. Piyārē Bachchõ kē Khūbsūrat Nām.
ve

Lahore: Idara Paigham ul Qur’an.


Hussain, Sajid. 14 January 2013. ‘From Quetta to Turbat’, The
ni

News.
U

Hvoslef, Erlend H. 2010. ‘The Social use of Personal Names


Among the Kyrgyz’, Central Asian Survey. 20: 1. 85–95.
d

Ibbetson, D., Maclagan, E. D. & Rose, H. A. 1911. Glossary of


or

the Tribes and Castes of the Punjab and N.W.F. Province. Vols.
2 & 3. Lahore: Aziz Publishers.
xf

Ibbetson, Denzil. 1883. Punjab Castes. Lahore: Sheikh Mubarak


O

Ali.
Icelandic Names. Retrieved 25 October 2013. <http://www.bbc.
co.uk/news/world-europe-21280101>
Jacob, James E. & Horn, Pierre L. 1998. ‘Comment vous
appelez-vous?: Why the French Change Their Names’, Names.
46: 1 (March), 3–28.
Jafri, Syed Hussain. 1979. The Origins and Early Development of
Shi‘a Islam. Karachi: Oxford University Press.
R eferences 225

Jalil, Muhammad. 1999. ‘World: South Asia’. Retrieved on 14


May 2012. <http://news.bbc.co.uk/l/hi/Southasia/390248.
stm>
Jessel, S. 2006. ‘France’s Morning After’, Catalyst. Accessed 28
September 2014. <http//:www.crl.gov.uk/default.aspx.LOC
ID>
Joshi, Sanjay. 2001. Fractured Modernity: Making of a Middle Class

s
in Colonial India. New Delhi: Oxford University Press.

es
Joya, Malalai. 2009. A Woman Among Warlords: the Extraordinary
Story of an Afghan Who Dared to Raise her Voice. London:

Pr
Simon and Schuster.
Johnson, Richard. 2010. ‘The Colour of Names, Others’
ity
Perceptions, my Reflections’, Framing My Name: Extending
Educational Boundaries. Altona, Victoria: Common Ground
rs
Publishing. 121–8.
Just, Peter. 1987. ‘Bimanese Personal Names: The view from Bina
ve

Town and Donggo’, Ethnology. 26: 4 (Oct), 313–28.


Khalid, Haroon. 2013. A White Trail: A Journey into the Heart of
ni

Pakistan’s Religious Minorities. New Delhi: Westland Ltd.


U

Klerk, Vivian de. 2002. ‘Changing Names in the “New” South


Africa: A Diachronic Survey’, Names. 50: 3 (September),
d

201–21.
or

Klymasz, Robert. 1963. ‘The Canadianization of Slavic


Surnames’, Names. Vol I-II ( June–December), 81–253.
xf

Kulke, Eckehard. 1974. The Parsees in India: the Minority as Agent


O

of Social Change. Munich: Weltforum Verlag.


Kumar, Margaret, Pattanayak, Supriya & Johnson, Richard (eds.).
2010. Framing My Name: Extending Educational Boundaries.
Altona, Victoria: Common Ground Publishing.
Lal, Vinay. 1999. ‘Not this, Not that: the Hijras of India and
the Cultural Politics of Sexuality’, Social Text. 61 (Winter),
119–40.
226 R eferences

Lauderdale, Diane S & Kestenbaum, Bert. 2000. ‘Asian American


Ethnic Identification by Surname’, Population Research and
Policy Review. 19: 3 ( June), 283–300.
Lehiste, Ilse. 1975. ‘The Attitudes of Bilinguals toward their
Personal Names’, American Speech. 50: 1/2 (Spring-Summer),
30–5.
Leslie, Paul. L & Skipper, James K. 1990. ‘Toward a Theory

s
of Nicknames: A Case for Socio-Onomastics’, Names. 38: 4

es
(December), 273–82.
Levi-Strauss, Claude. 1966. The Savage Mind. Letchworth,

Pr
Herfordshire: The Gordon City Press.
Lord, E. 2002. ‘Given Names and Inheritance’, Naming, Society
ity
and Regional Identity. Oxford: Leopard’s Head Press, 169–92.
Lyon, Stephen M. 2002. ‘Power and Patronage in Pakistan’,
rs
Unpublished PhD thesis, University of Kent, Canterbury,
England.
ve

Manto, Sa’adat Ali. 1991. Mantō Nāmā. [Urdu: The Writings of


Manto] Lahore: Sang-e-Meel Publishers.
ni

Manzar, Hasan. 2008. Dhanī Bakhsh Kē Bētē. [Urdu: The sons of


U

Dhani Bakhsh] Karachi: Scheherzade.


McDavid, John W. & Harari, Herbert. 1966. ‘Stereotyping of
d

Names and Popularity in Grade-School Children’, Child


or

Development. 37: 2 ( Jun), 453–9.


M’Clure, Edmund. 1879. ‘On Irish Personal Name’, Proceedings
xf

of the Royal Irish Academy: Polite Literature and Antiquities.


O

307–314.
Mehrotra, R. R. 1979. ‘Name Change in Hindi: Some
Sociocultural Dimensions’, Anthropological Linguistics. 21: 4
(Apr), 205–10.
Metcalf, Barbara D. 1982. Islamic Revival in British India:
Deoband, 1860–1900. New Delhi: Oxford University Press.
Mill, John Stuart. 1882. A System of Logic, Ratiocinative and
Inductive. 8th edition. New York: Harper and Brothers.
R eferences 227

Mir, Amir. 2008. The Fluttering Flag of Jihad. Lahore: Mashal


Books.
Mir, Amir. 2009. Talibanization of Pakistan: From 9/11 to 26/11.
New Delhi: Pentagon Security International.
Mirchandani, Kiran. 2004. ‘Practices of Global Capital: Gaps,
Cracks and Ironies in Transnational Call Centers in India’,
Global Networks. 4: 355–73.

s
Mirza, Mumtaz Akhtar. (comp. & ed.). 1978. Muqālāt-ē-Tāsīr.

es
[Urdu: Dissertations of Taseer] Lahore: Majlis-e-Taraqqi-e-
Adab.

Pr
Mohsin, Mohammad. 2010. Piyārē Bachchõ kē Khūbsūrat Nām.
[Urdu: Beautiful names of lovely children] Lahore: Idara
ity
Paigham-ul-Quran.
Moore, John H. 1984. ‘Cheyenne Names and Cosmology’,
rs
American Ethnologist. 11: 2 (May), 291–312.
Moore, Robert L. 1993. ‘Nicknames In Urban China: A Two-
ve

Tiered Model’, Names. 41: 2, ( June), 67–86.


Morgan, Jane, O’Neill, C. & Harre, R. 1979. Nicknames: Their
ni

Origins and Social Consequences. London & Henley: Routledge


U

and Kegan Paul.


Mufti, Aksi. 2012. Talāsh: Allāh Māwrā Kā ta ‘ayyun. Lahore:
d

Al-Faisal Nashran.
or

Murray, Thomas E. 2002. ‘A Further Note on the “Title+First


Name” Form of Address’, Names. 50: 4 (December), 263–73.
xf

Nadvi, Abul Hasan Ali. n.d. Jab Īmān kī Bahār Āyī. [When the
O

Spring of Belief Arrived] Karachi: Majlis-e-Nashariat-e-


Islam.
Nagata, Mary Louise. 2012. ‘Why Did You Change Your Name?
Name Changing Patterns and the Life Course in Early
Modern Japan’, The History of the Family. 4:3: 315–38.
Names, Pakistani. 2013. ‘Top 100 Pakistani Names-Pakistan’.
Retrieved 25 Oct 2014. <http//ww.studentsoftheworldinfo/
penpals.php3?Pays=Pakistan>
228 R eferences

Nuessel. Frank. 2008. ‘A Note on Ethnopaulism and Hate


Speech’, Names. 56: 1 (March), 29–31.
Nuessel. Frank. 2013. ‘ Sixty Years of Names’, Names. 61: 1, 3–7.
Palaeas, Maria. 2005. ‘The Importance of Being “Ernest” in Mary
Shelley’s Frankenstein: A Study in Literary Onomastics’,
Names. 53: 3 (Sept), 203–34.
Palsson, Gisli. 2014. ‘Personal Names: Embodiment,

s
Differentiation, Exclusion, and Belonging’, Science, Technology

es
and Human Values. 39: 4 ( January), 618–30.
Pamment, Claire. 2010. ‘Hijraism: Jostling for a Third Space in

Pr
Pakistani Politics’, The Drama Review. 54: 2 (Summer), 29–50.
Pattanayak, Debi Prasanna. 2010. ‘Establishing the Space
ity
of Naming’, Framing My Name: Extending Educational
Boundaries. Altona, Victoria: Common Ground Publishing, 69.
rs
Persian Names. 2012. Wikipedia. Retrieved on 26 Apr 2012.
Pina-Cabral, Joao de. 2010. ‘The Truth about Personal Names’,
ve

Journal of the Royal Anthropological Institute. 16: 2, 297–312.


Powlett, P. W. 1878. Gazetter of Ulwar. London: Trubner.
ni

Puckett, Newbell N. 1938. ‘American Negro Names’, The Journal


U

of Negro History. 23: 1 ( Jan), 35–48.


Pulgram, Ernst. 1947. ‘Indo-European Personal Names’,
d

Language. 23: 3 ( Jul–Sep), 189–206.


or

Qasmi, Abdul Shukoor. 2007. Rāhnumā-ē-Islāmī Nām. [Urdu:


Guide to Islamic Names] Karachi: Nadwa tul Qalam.
xf

Qudsia, Bano. 2007. Dūsrā Darwāzā. [Urdu: The Other Door]


O

Lahore: Sang-e-Meel.
Rahman, Taimur. 2012. The Class Structure of Pakistan. Karachi:
Oxford University Press.
Rahman, Tariq. 1996. Language and Politics in Pakistan. Karachi:
Oxford University Press.
Rahman, Tariq. 1999. Language, Education and Culture. Karachi:
Oxford University Press.
R eferences 229

Rahman, Tariq. 2009. ‘Language Ideology, Identity and the


Commodification of Language in the Call Centers of
Pakistan’, Language in Society. 38: 2: 233–58.
Rahman, Tariq. 2012. ‘Pakistan’s Policies and Practices towards
the Religious Minorities’, South Asian History and Culture. 3:
2 (Apr), 302–15.
Ragone, Agnes C. M. 2012. ‘Onomastics Among Hispanic

s
Migrants in South Central Pennysylvania’, Names. 60: 1

es
(Mar), 36–45.
Rana, Amir. 2005. The Seeds of Terrorism. London: New

Pr
Millennium.
Rashid, Salman. 12 May 2012. ‘The Legend of Dodo Chanesar’,
ity
The Express Tribune.
Raza, Asif (Comp). 2003. Nām Savārē Kām. [Urdu: The Name
rs
Facilitates Endeavour] Lahore: Khazeena Ilm-o-Adab.
Robinson, Christopher L. 2013. ‘What Makes the Names of
ve

Middle-earth So Fitting’, Names. 61:2 ( Jun), 65–74.


Rumi, Raza. 2013. Delhi by Heart: Impressions of a Pakistani
ni

Traveller. New Delhi: HarperCollins Publishers.


U

Rusva, Mirza Hadi. 2005. Umrāō Jān Adā. [Urdu] Lahore: Sang-
e-Meel.
d

Ryan, Eugene, E. 1981. ‘Aristotle on Proper Names’, Aperion: A


or

Journal for Ancient Philosophy and Science. 15: 1 ( Jun), 38–47.


Rymes, Betsy. 1996. ‘Naming as Social Practice: the Case of
xf

Little Creeper from Diamond Street’, Language in Society. 25:


O

2 ( June), 237–60.
Salim, Ahmad. 2006. Equal Citizens? Islamabad: Friedrich
Naumann Foundation.
Sanyal, Usha. 1996. Devotional Islam and Politics in British India:
Ahmed Riza Khan Barelwi and His Movement, 1870–1920.
Delhi: Oxford University Press.
230 R eferences

Saudi Arabia. 2014. ‘Is Your Name Banned in Saudi Arabia’.


Belfast. Retrieved on 01 May 2014. Telegraph.<http://www.
belfasttelegraph.co.uk/news/world-news>
Schaps, David. 1977. ‘The Woman Least Mentioned: Etiquette
and Women’s Names’, The Classical Quarterly. 27: 2: 323–30.
Schimmel, Anne Marie. 2005. Islamic Names. Lahore: Sang-e-
Meel.

s
Searle, John R. 1969. Speech Acts: An Essay in the Philosophy of

es
Language. Cambridge & New York: Cambridge University
Press, 2011 edition.

Pr
Shah, Benazir. 27 Sept 2013. ‘Stiff Upper Lip: Interview of Yusaf
Raza Gilani’, Newsweek. ity
Shahbazi, Ammar. 2012. ‘What’s in a Father’s Name?’, The News
on Sunday: Encore, 30.
rs
Shahrani, Nafiz. 1993. ‘Central Asian and the Challenge of the
Soviet Legacy’, Central Asian Survey. 12: 2: 123–35.
ve

Sheikh, Farhan A. n.d. ‘Ilm ul Ā‘dād kī Raōshnī Mē Islamī


Nām Encyclopedia [Urdu: Islamic Names in the Light of
ni

Numerology] Lahore: Rubi Publications.


U

Schottman, Wendy. 1993. ‘Proverbial Dog Names of the


Baatombu: a Strategic Alternative to Silence’, Language in
d

Society. 22: 4 (Dec), 539–54.


or

Sikand, Yoginder and Katju, Manjari. 1994. ‘Mass Conversions


to Hinduism among Indian Muslims’, Economic and Political
xf

Weekly. 29: 34 (August 20), 2214–19.


O

Silberman, Charles. 1985. A Certain People: American Jews and


their Lives Today. New York: Summit Books.
Silva, E. B. 2005. ‘Gender, Home and Family in Cultural Capital
Theory’, British Journal of Sociology. 56: 1: 83–103.
Silverman, Carol. 1991. ‘Strategies of Ethnic Adaptation: the
Cae of Gypsies in the United States’. Creative Ethnicity,
Symbols and Strategies of Contemporary Ethnic Life. Cicala,
Norman, UT: Utah State University Press, 107–21.
R eferences 231

Simmonds, Felly Nkweto. 1998. ‘Naming and Identity’, The


Feminist Critique of Language: A Reader. London & New York:
Routledge, 33–7.
Sinclair, W. F. 1889. ‘Indian Names for English Tongues’, The
Journal of the Royal Asiatic Society of Great Britain and Ireland.
21: 1 ( Jan), 159–78.
Singh, Jaswant. 2009. Jinnah: India-Partition Independence. New

s
Delhi: Rupa & Co.

es
Spencer, Robert F. 1961. ‘The Social Context of Modern Turkish
Names’, Southwestern Journal of Anthropology. 17: 3 (Autumn),

Pr
205–18.
Srinivas, M. N. 1952. Religion and Society among the Coorgs of
ity
South India. Oxford: The Clarendon Press.
Stern, Stephen & Allan, John (eds.). 1991. Creative Ethnicity,
rs
Symbols and Strategies of Contemporary Ethnic Life. Cicala,
Norman, UT: Utah State University Press.
ve

Suhrawardy, Kamran Azam. 2012. Aqwām-ē-Sindh. [Urdu: The


Nationalities of Sindh] Lahore: Takhleeqat.
ni

Suzman, Susan M. 1994. ‘Names as Pointers: Zulu Personal


U

Naming Practices’, Language in Society. 23: 2 ( Jun), 253–72.


Syed, G. M. 2012. The Case of Sindh: G. M. Sayed’s Deposition for
d

the Court. Lahore: Fiction House.


or

Syed, G. M. n.d. Sindhu Desh: A Study its Separate Identity through


the Ages. Karachi: G. M. Syed Academy.
xf

Tabari, Abu Jafar Muhammad Ibn-e-Jareer (al.fl. 839–923).


O

Tarikh al Rusul wal Muluk [Arabic: The history of prophets


and kings]. Vol. XIX. The Caliphate of Yazid bin Muawiyah
trans. and annotated. Howard, I. K. A. Albany: State
University of New York Press.
Temple, Richard. 1883. Dissertation on the Proper Names of
Panjabis: With special Reference to the Proper Names of Villagers
in the Eastern Panjab. London: Trubner.
232 R eferences

Therborn, Goran. 2003. ‘Entangled Modernities’, European


Journal of Social Theory. 6: 3: 293–305.
Treis, Yvonne. 2005. ‘Avoiding Their Names, Avoiding their
Eyes: How Kambaata Women Respect their In-laws’,
Anthropological Linguistics. 47: 3 (Fall), 292–320.
UNICEF. 2005. UNICEF Iran Information Series.
Vatuk, Sylvia. 1969. ‘Reference, Address, and Fictive Kinship in

s
Urban North India’, Ethnology. 8: 3 ( Jul), 255–72.

es
Viswanathan, Gauri. 1998. Outside the Fold: Conversion,
Modernity, and Belief. Princeton, N. S.: Princeton University

Pr
Press.
Waleh, A. H. 1964. ‘Names and Titles in Afghan Families’,
ity
Afghanistan News. No. 80: 7, 4–6.
Wright, Theodore P. 2006. ‘Southasian Muslim Naming for
rs
Identity vs. Name-Changing for Concealing Identity’, Indian
Journal of Secularism. 10: 1 (Apr–Jun), 5–10.
ve

Wittgenstein, Ludwig. 1953. Philosophical Investigations. 1986


edition. Trans. From the German by G. E. M. Anscombe.
ni

Oxford: Basil Blackwell.


U

Yassin, M. Aziz. 1976. ‘Personal Names of Address in Kuwaiti


Arabic’, Anthropological Linguistics. 20: 2 (Feb), 53–63.
d

Yousafzai, Malala. 2013. I am Malala. with Christina Lamb.


or

London: Wiedenfeld & Nicolson.


Yusufzai, Rahimullah. 17 Jun 2013. ‘Barrister Baachaa’, The News.
xf

Yusufzai, Rahimullah. 16 Nov 2014. ‘All in a Name’, The News


O

on Sunday: Encore.
Zaman, Muhammad Qasim. 2002. The Ulama in Contemporary
Islam: Custodians of Change. New Jersey: Princeton University
Press.
Zhongti, Lu & Millward, Celia. 1989. ‘Chinese Given Names
since the Cultural Revolution’, Names. 37: 3 (Sep), 265–80.
R eferences 233

Interviews and Personal Communications:


Abbas, Tahir. 16 Jan 2013. Interview with a Shia Alim of Jamia
Imamia, Lahore.
Abid, Husain. 09 Mar 2012. Cook in Lahore. From village Lal
Wala near Muzafargarh.
Ahmad, Qari Fayyaz. 14 Mar 2013. Telephone interview with an
Ahl-i-Hadith Alim, Lahore.

s
Ahsan, Tariq. 10 Sept 2012. Telephonic interview of a personal

es
friend settled in Ottawa, Canada.
Akbar, Abdul. 26 Mar 2012. Interview with an electrician from

Pr
Hazara now working in Rawalpindi, Rawalpindi.
Anwar. Int. 6 Mar 2012. Telephonic interview with Kenneth
ity
Naveed Anwar, Rawalpindi.
Ashrafia. Int. 08 Jan. 2013. Interview with the senior Alim at
rs
Jamia Ashrafia, Deobandi madrasah, Lahore.
Ashrafia. Int. 08 Jan. 2013b. Interview with Shahid Obaid, Junior
ve

Alim at Jamia Ashrafia, Deobandi madrasah, Lahore.


Askari, Hasan. Int. 04 Apr 2013. Open ended interview with a
ni

well-known director of Urdu and Punjabi films, Lahore.


U

Attari, Nawaz. 08 Mar 2012. Office attendant in Lahore, from


village Sirari near Muzaffarabad.
d

Baloch, Inayatullah. 28 Aug 2012. Senior official in the Centre


or

for South Asian Studies, University of Heidelberg, Germany.


Buzdar, Wahid Baksh. 08 May 2012. Assistant Professor of
xf

Balochi and Brahvi, NIPS, Quaid-i-Azam University (QAU),


O

Islamabad.
Chan, Munawwar. Int. 08 Nov 2014. Interview with the
chairman of the Pakistan Hindu Rabita Council, Lahore.
Chaudhary, Azam. Int. 13 Nov 2012. Interview with an Associate
Professor of Anthropology, NIPS, Quaid-i-Azam University
(QAU), Islamabad.
234 R eferences

Gadi, Mushtaq. 19 Dec 2012. Lecturer at NIPS, Quaid-i-Azam


University. E-mail message in response to questions by the
author.
Gardezi, Amina. 30 April 2012. Assistant Professor at BNU,
Lahore. Key source of the naming practices of eminent
landowning and Syed families of South Punjab.
Giani, Preetam. 20 October 2013. E-mail message in response

s
to questions about his name to the author.

es
Hijras. 2012. Author’s open ended interviews of Hijras
(Transgender/Transvestite) on the Streets of Rawalpindi and

Pr
Lahore.
Itfan, Rana. 20 March 2012. Interview with the manager
ity
training, Head Office, the City School, Lahore.
Jafri. Int. 16 Sept 2013. Telephonic interview with Professor
rs
Syed Hussain Jafri, former director of Pakistan Studies Centre,
University of Karachi.
ve

Khalid, Haroon PC. 16 Nov 2013. Conversation with the author


of a book on the minorities of Pakistan, Lahore.
ni

Khalil, Hanif. 13 November 2012. Interview with an Assistant


U

Professor of Pashto at NIPS, Quaid-i-Azam University


(QAU), Islamabad.
d

Khan, Asfandyar Wali. PC. 02 October 2013. Various conversa­


or

tions with the author with the head of the Awami National
Party, Dhaka, Bangladesh.
xf

Khan, Sarfraz. PC. 15 Nov 2012. Conversation with the author,


O

University of Peshawar.
Khan, Shamim Saifullah. PC. Dec 2012. Conversation with the
author, Lahore.
Makijani, Harbaksh. Int. 20 June 2013. Interview of a Sindhi
Hindu academic of the University of Sindh, Jamshoro, Bara
Gali Campus, University of Peshawar..
Malik, John Alexander. PC. 17 Dec 2014. Conversation with the
Protestant bishop emeritus of Lahore, Lahore.
R eferences 235

Pahore, Mohammad Ramzan. Int. 20 June 2013. Interview of


Sindhi academic from the University of Sindh, Jamshoro, Bara
Gali Campus, University of Peshawar.
Qadri, Irfan. 05 Jan 2013. Written answers to questions from an
Alim of a Barelvi madrasah, Ghausia Rizvia, Lahore.
Saima. Int. 20 Oct 2012. Interview of a maid servant employed
by Adiah Afraz, Lahore.

s
Satyani, Pirbhu. Int. 07 Dec 2014. Interview of an educated

es
person from the Sindhi Hindu community who has worked
in Lahore, Lahore.

Pr
Somroo, Badruddin. 20 June 2013. Interview of a Sindhi
academic of the University of Sindh, Jamshoro, Bara Gali
ity
Campus, University of Peshawar.
Singh, Inderjit. PC. 03 Nov 2013. E-mail message on Sikh
rs
names from Canada.
Tahir & Shamoon. Int. 08 May 2012. Interview with Tahir
ve

and Shamoon Masih, janitors at Quaid-i-Azam University,


Islamabad.
ni

Waseem, Mohammad. Int. 23 April 2012. Interview with a


U

professor of political science at the Lahore University of


Management Sciences (LUMS), Lahore.
d
or
xf
O
O
xf
or
d
U
ni
ve
rs
ity
Pr
es
s
Index

A Colonialism 91

s
Components 10, 15, 17, 20–4,

es
Abubakr 20 28, 34, 37, 39–44, 47–8, 52,
Africa 56, 91, 114 58, 60, 63–70, 77, 93, 95, 98,
Ahl-i-Bayt 43, 75 118–19, 128

Pr
Ali 20–1, 27, 43–4, 48, 50–1, Caste 17, 64–5
56–7, 60–1, 63, 70, 74–6, 81, ity Islamic 21, 39–44, 52, 70
94–5, 109–10, 122, 128 Old-fashioned 98
America 46, 116 Prestigious 20, 58, 63, 66–7, 69
rs
Amina 77, 79–81, 124 Rural 23, 68, 98
Anthropology 7–8 Shia 20, 47–8
ve

Apna School 99 Conversion 36–7, 61, 133


Arabic language 5, 99
ni

Austin 10, 126–7, 132 E


Awami National Party (ANP) 102
U

Ayesha 77, 79–81, 112 Elite 18, 27, 33, 38, 43, 46, 63,
65–6, 68–9, 72–4, 79–81, 85,
d

B 88, 92–3, 98, 112, 123


or

Balochistan 17–19, 73, 78 F


xf

Bangladesh 116, 121


Bengal 16, 60 Fatima 43, 77, 79–81, 96
O

British 9, 14, 34, 38–9, 57, 91–2,


96, 107, 112–13, 124 H
Officers 57
Rule 124 Hadith 31–2, 35, 49
Hebrew 36
C Hijra 116
Hindi 28, 30, 35, 122–3
Call Centres 13, 113, 132
238 Index

I L

Identity 2–7, 9–14, 16, 22–5, Linguists 8, 11


34–41, 43–5, 47–53, 61, 63,
67, 69, 75, 85, 90–2, 94, 97, 99, M
102–4, 106, 108, 112, 114, 119,
121–2, 125, 128–9, 131–2 Madrasahs 43

s
Christian 119 Malala 84, 100–2, 115

es
Ethnic 5, 91, 99, 103–4, 131 Masih 26, 38, 118–19
Islamic 6, 22, 25, 34–5, 37, 39, Meo 37
41, 43–5, 47, 49, 51–3, 91, 104 Modern 2, 4–5, 16, 23–4, 33,

Pr
Modern 23–4 69–70, 73, 75, 80, 90–1, 94,
Pashtun 102 ity 96–7, 99–100, 110, 112, 116,
Rural 67, 97 120, 123, 132
Sikh 125 Modernity 5, 40, 89–92, 107
rs
Urban 16 Muawiya 22, 48–50, 53
Western 13
ve

Illocutionary force 127 N


Iran 36, 91–2
ni

Islam 4–5, 21–3, 31, 34–7, 40–4, Names 1–133


46–7, 51–2, 60–2, 70, 79, 107, Arab 22, 25
U

115, 133 Chinese 82


Folk 21, 41–6 Christian 39, 119
d

High 21, 40–2 Family 15, 23, 26, 66–7, 69–70,


or

Islamization 4–5, 23, 37, 40, 42–3, 91–2, 116


46, 53, 74, 79–80 First 23, 44–5, 47, 53, 56, 70–1,
xf

73–5, 77, 79–81, 83, 85, 87–8,


J 94, 97, 124
O

Hindu 12, 15, 34, 37, 56, 109,


Jesus 26 115, 117, 119, 121, 123
Jinnah, M. A. 94 Islamic 15, 31, 35–7, 39, 41, 43,
46, 77, 79–80, 99
K Muslim 4, 14–16, 24, 37–9, 45,
70, 93, 112, 117–18, 121
Khurasan 50 Neutral 117, 120
Killing(s) 48, 88 Pashtun 101–2
Index 239

Pen names 109 Q


Popular 29, 43, 76, 78, 81
Rural 69, 73, 77 Quran 20–1, 30, 35–6, 79–80
Sikh 125
Sindhi 43 S
Siraiki 103
Urban 67, 73 Saddam 22, 47, 52, 80

s
Western 13, 45, 93, 112, 123 Sanskrit 36, 125

es
Nickname(s) 5–8, 15, 23, 25, 29, Scotland 12, 113
31, 54, 58, 61, 70–1, 73–9, 81– Sects 47
5, 87, 93–4, 111–13, 116, 133 Sindh 17, 41, 43, 45, 47, 64–6,

Pr
No-naming 23, 70–71, 73, 75, 77, 71–2, 104–6, 119–21, 123
79, 81, 83, 85–88 ity South Africa 56
Soviet Union 46
O Sweden 113
rs
Syed, G. M. 104, 106
Onomastics 7–8, 132–133
ve

T
P
ni

Tekonymy 11
Pakistan 2–3, 5–6, 11, 14, 16, Titles 14, 22, 61, 63–4, 66, 99,
U

19–22, 25–7, 30, 34–6, 40–1, 124, 132


43, 45–8, 50–3, 58–9, 61–5, Turkey 15, 91–2
d

67–9, 71–5, 77, 82–4, 86–8, 90,


or

92–4, 96–7, 99, 107–8, 110–13, U


116–17, 121–5, 131–3
xf

Partition 26, 37, 114, 117 Ulema 31–3, 49, 51, 79


Pashtun 65, 67, 73, 96, 99–103, Urdu 4, 18–19, 21, 27–8, 30, 34–5,
O

106 38, 43, 47, 55, 57, 61, 71, 84,


Perlocutionary force 127, 129–30 87, 105, 109, 113, 122
Philosophers 7, 9–11, 132 Usman 79
Prophet Muhammad (pbuh) 20–
1, 36, 41, 43–4 W
Pseudonym 100, 103, 115
White Mughals 38
Wittgenstein, Ludwig 126
240 Index

Y Z

Yazid 22, 48–9, 53 Zia ul-Haq, General 42, 46–7, 51,


70, 121
Zulu 13

s
es
Pr
ity
rs
ve
ni
U
d
or
xf
O