Etudes gujarati : Société, langue et culture

11-02-09

‘THE GLOBAL SINDHI DIASPORA’

A CONTEMPORARY READING

Mark-Anthony Falzon

University of Malta

Sindhis must be one of the most widely dispersed and diasporically inclined, so to

speak, in/from South Asia. Sindhis draw upon a fascinating history of mobility and

transnational enterprise, and today are found in scores of places worldwide. As such,

one would expect the rich literature on ‘immigrant communities’, which flourished

from the 1970s especially in Britain, the USA and Canada, Australia, and the

Netherlands, to have picked up their scent. For some reason, however, it is only since

recently that the Sindhi diaspora is attracting scholarly attention. Sustained works

include monographs by Markovits (2000) and Falzon (2005), and paper-length

contributions by Haller (for example 2005).

In this paper I argue that the term ‘Sindhi diaspora’ subsumes three different

moments of mobility, each of which emerged out of particular historical

circumstances and had its own logic. Against this backdrop, I then discuss some of the

patterns of contemporary Sindhi economy and society, with an emphasis on their

transnational aspects. First, however, a caveat on my use of the term ‘Sindhi’, which

is based on two overlapping distinctions. Primarily a religious one: the paper is about

Hindu Sindhis as distinct from Muslims, Christians, and other groups living in Sind,

Pakistan, who are also ‘Sindhi’ on the basis of linguistic and regional criteria. The

bulk of the Hindu population of Sind left the newly-fledged Pakistan in 1947–8, and it
 2

is these people I worked with1 and talk about today. Even then, however, the term

‘Hindu Sindhi’ is not fully satisfactory since, as pointed out somewhat crossly by

Devji in his review of my book (2006: 4435, my parenthesis), ‘this new community

[the Hindu Sindhis of this talk] was only made possible because its menial castes had

been left behind in Sindh where they still languish’. That said, I believe my

terminology - and therefore (arbitrary) research boundaries - to be justified, in the

sense that, if we may speak of imagined communities (Anderson 1991), it is fair to

say that Hindu Sindhis living in worldwide diaspora outside of Pakistan today

constitute such a community.

Empirically, then, the paper traces some of the more salient recent historical

experiences of Sindhis. Theoretically, I seek to re-read my earlier work in light of

recent understandings of, first, cosmopolitanism, and, second, the space of global

objects such as diasporas. With respect to both cosmopolitanism and diaspora, my

argument is that Sindhis constitute a model of ‘a cosmopolitan diaspora’ and at the

same time feed into our developing understanding of the multiplicities and

discontinuities - of both cosmopolitanism and global space.

Pax Britannica and the development of the Sindwork trade diaspora

As Brown & Foot (1994) point out, many of the protagonists of today’s Indian

diaspora come from regions and groups with traditions of mobility that stretch back

many centuries. In our case, the contemporary Sindhi diaspora is to some extent a

1
I did anthropological fieldwork with Sindhis in Mumbai, London, and Malta intermittently between

1995 and 2001; this was supplemented with archival research in various collections, notably the Royal

Commonwealth Library in Cambridge, the British Library in London, and the National Archives in

Malta.
 3

creature of the nineteenth century. Sometime around 1850, indigenous traders -

mainly from the town of Hyderabad - discovered that there was a foreign market for

the native handicrafts of Sind (‘Sind works’), and the ‘Sindwork’ trading diaspora was

born.

Hardly out of nothing, however. Take the memoirs of one Seth Naomul

Hotchand of Karachi (1804–1878), written largely as a hagiography to extol the

greatness and success of his family. Even if Seth Naomul’s claims that by 1804 his

family owned agencies and firms ‘at about 500 places’ in northern India and around

the Arabian Sea, or that the members of his family and their gumashtas (agents)

‘acted in perfect concord’ (1915: 48), are not to be taken literally, the memoirs present

a model of trade and mobility that was to find much fortune with Sindhis. They

effectively show a firm, controlled by a paternalistic hub in urban Sind, expanding

geographically by locating agents and/or relatives at strategic trading points which, in

turn, often acted as depots fostering further expansion.

Prior to the conquest and ‘annexation’ of Sind by the British in 1843, the

country had been ruled by the Talpur Mirs and most of the fertile land owned by a

Muslim aristocracy of powerful waderos (landowners - see Cheesman 1981a,b).

Trade and commerce was, however, substantially in the hands of Hindus, who ranged

from large-scale urban merchants like Seth Naomul’s family to village banias (small

traders). A number of sources hold tantalising clues to the deep-rooted tradition of

mobile trade in and out of Sind. The geographical accounts of Al Idrisi (1100–1166),

for example, with their vivid descriptions of entrepôt towns and mobile merchants

(see Ahmad 1960); the poetry of Shah Abdul Latif, with its depictions of overseas

trading voyages in the eighteenth century; or the records of the Dutch East India

Company (VOC), detailing the shrewdness of Sindhi traders who bargained hard by
 4

playing the Dutch against their English competitors (Floor 1993–4). On much firmer

ground, thanks to Markovits’ (2000) peerless scholarship, is our knowledge of

Shikarpuri Hindu shroffs (bankers), who plied their hundi (promissory note) trade into

central Asia notably during the Durrani hegemony. The point is that Sindwork, albeit

an innovation, was one of many trajectories of mobile trade that the region nourished

in the long term.

The emergence of Sindwork is one of those instances of entrepreneurship and

inspiration which cannot be ‘explained’, but which make better sense when located

within a specific historical heuristic. One factor which made up this matrix was the

deposition of the princely court of the Mirs by the British at annexation, which

effectively destroyed a traditional market for luxury and artisanal goods, particularly

in Hyderabad - the capital and seat of the Mirs as also the point of origin of Sindwork.

This meant that the merchant-purveyors to the court had to look elsewhere to sell their

wares, and in the 1840s that elsewhere was the world of Pax Britannica, with its new

technologies of mobility and communication and its penchant for ‘curios’, soon to be

sharpened by the various world expositions and fairs. By the 1860s we find mention

of Sindwork merchants in various places worldwide, from Malta to Japan to Panama.

Sindwork traders (commonly called ‘Sindworkis’) plied mainly two types of

wares - textiles and artisanal works/souvenirs, the latter usually known generically as

‘curios’. Originally, these included goods of Sindhi artisanal manufacture such as

glazed pottery and lacquer work. This trade took them to British-controlled harbour

towns, which at the time were cauldrons of mobility and a certain cosmopolitanism.

The Sindworkis soon noticed that there was a liking for all sorts of oriental-looking

goods and especially the textiles, ceramics, and various other items of Japanese

manufacture (this was the heyday of Japonisme). The consequences were far-reaching
 5

in that Sindwork changed from being a case of traders selling the products of a

relatively tiny source back in Sind, to a network of merchants capable of tapping into

a much larger world market. Two examples will suffice to make my point. The first is

from Malta, where in 1887 the firm Pohoomull Bros. applied to the colonial

authorities for the release from customs of one case containing ‘Oriental goods and

some fancy weapons as knives, daggers, etc.’.2 By 1917, however, one of the many

lines Sindworkis plied in Malta was the export of Maltese lace to Batavia (Java),

Johannesburg, and a host of other places.3 The second example concerns Japan, where

by the end of the Meiji period (1868–1912), Sindworkis had breached local markets to

the extent that in Yokohama, for instance, they controlled a substantial chunk of the

Japanese silk trade (Chugani 1995). These examples are crucial in the sense that they

show how, in a couple of decades, Sindwork shifted from being essentially a peddling

venture reliant on the artisans of Sind, to a network of merchants and well-organised

firms that did not depend on any one source, but was explorative, innovative, and

expansionist.

Sindwork in pre-Partition Sind was exclusively a trading diaspora (I use the

term in Abner Cohen’s [1971] sense), with a social network based in Sind. Sindworkis

may have spent most of their lives travelling and doing business in the various

locations of their kothis (firms), but their operations were invariably rooted in

Hyderabad, in three ways. First, with respect to the firms themselves, the decades post

1860 saw the consolidation of well-established firms, notably the renowned ‘Three

Cs’ - Choitram, Chellaram, and Chanrai. These employed people on a contractual

2
Petition to the Chief Secretary to Government (CSG) 4949/1887, National Archives (Malta).
3
Petitions to the Chief Secretary to Government (CSG) 2947/1917 and 1886/1917, National Archives

(Malta).
 6

basis, the old gumashta system eventually giving way to salaried employment on

three-year written contracts. The substantial numbers of young bhaiband men

required for the Sindwork operations were recruited through social and kin networks

in Hyderabad. Second, the mobile aspect of Sindwork was a male prerogative, which

means that the extended patrilocal family life in which it was embedded, and which

made it possible, was located in Hyderabad. It was to these families that Sindworkis

homed in from all over the world when their contracts expired, and through them that

the social networks so essential to the growth of kothis were cultivated. Third, the old

capital of the Talpurs got a new lease of life as the profits of Sindwork were invested

in lavish havelis, and social venues, such as the Bhaiband and Rotary clubs,

burgeoned. The Sindwork diaspora may have been independent of Sind in strictly

commercial terms, but was very much rooted in it in social ones.

This, then, was the situation of Hindu Sindhis in the first half of the twentieth

century. In the north, the shroffs of Shikarpur went about their ancient moneylending

business, now increasingly looking south rather than towards central Asia. The south,

and especially Hyderabad, was where the specialized Sindwork diaspora had its heart,

pumping blood to scores of locations worldwide. In the villages and small towns,

Hindu banias ran their trade and moneylending businesses. The boundaries between

the three were to some extent porous (Sindworkis often recruited the sons of village

banias, for example, and their kothis made use of the Shikarpuri banking services),

but by and large the distinctions were fairly clear cut.

They also had to do with caste. Apart from occupation, Hindu Sindhis aligned

themselves along two major criteria: jati based on birth and kinship metaphor, and

regionality. The issue is complex, for these distinctions overlapped and are not always

easy analytically to disentangle (for a detailed discussion see Falzon 2005). To deal
 7

first with jatis, the two that one encounters most frequently are the bhaibands and the

amils, followed by bhatias, sahitis, Brahmins, chhaprus, and bhagnarees. Amils were

generally involved in clerical-administrative duties. The Talpurs had employed Hindu

amils as their munshis (scribes) and revenue collectors. With the arrival of the British,

the amils carved out a niche for themselves based on their past specialization. They

took to the professions and later the civil service and by the beginning of the twentieth

century had successfully cultivated the image of a Westernised, English-speaking,

literate elite.

Unlike amils, bhaibands were seldom employed in salaried labour. Instead,

they concentrated in the commercial sector. The word ‘bhaiband’ itself means

‘brotherhood’, and the usage was therefore something along the lines of ‘brotherhood

of (Hindu Sindhi) traders.’ The Sindworkis of Hyderabad were drawn from the

bhaiband jati, and they were certainly the most successful and mobile - this

characteristic mobility of Hyderabadi bhaibands is to be specifically noted, because it

has had a profound influence on the contemporary situation outside of Sind. The large

part of the Hindu population in Sind therefore belonged to the bhaiband jati, although

in the smaller towns and villages the local traders and moneylenders were known

simply as banias (traders/moneylenders) or even hatvanias (‘small banias’). Even

today among Sindhis, each surname is usually associated with a particular jati

although, clearly, knowledge of this sort is never foolproof. The general point is that

until 1947, the Hindus of Sind were very much differentiated into types based on the

criteria, of region, caste, and occupation. There was no sense of a single ‘Sindhi

diaspora’ as we know it today, the closest being the Sindworkis of Hyderabad which

made up a specialized trading diaspora.

 8

Partition and subsequent settlement and re-diasporization

Partition changed all of that, and as such represents the second key moment in the

making of the Sindhi diaspora (alternatively, ‘the second in a series of Sindhi

diasporas’). The complex dynamics of that political set of events cannot concern us

today; the upshot, however, was that by early 1948 the exodus of Hindus from Sind

was in full swing. There were basically two trajectories out of Sind. Substantial

numbers of Hyderabadi Sindworkis found it relatively straightforward to relocate to

the various countries where they already had business and other assets. This is not of

course to downplay the trauma of Partition for these people - one should keep in mind

that extended families and other social/kinship networks were firmly rooted in

Hyderabad until 1947. In any case, Sindworki families (rather than male merchants)

appeared on the scene in Malta, Japan, Gibraltar, and elsewhere, directly after

Partition.

For the less historically-mobile Hindus, the exodus brought about different

challenges. In the (rather special) case of the better-established amils, the natural

choice was Bombay, where they already had good connections (Sind had been part of

the Bombay Presidency); in the city, they soon established themselves as

professionals or civil servants. The rest - the bulk of the Hindu Sindhi refugees, that is

- found Partition harder to weather. All but a very few moved to India and

southwards, a good proportion eventually converging on Bombay. Refugee camps

were set up outside the city at Kalyan (today’s Ulhasnagar, a town still synonymous

with Sindhis) and there was some degree of mutual soccour, led mostly by well-

connected amils. The 1951 Census gives us an idea of the numbers involved - 408,882

‘displaced persons’ were enumerated in the three states of Bombay, Saurashtra, and

Kutch, 82.4 per cent of whom had come from Sind. Most of these eventually settled
 9

down in cities and towns around India, and took to small business - notably textiles

wholesale, manufacture (especially in Ulhasnagar), and retail. According to the 1991

Census, a total of 1,551,384 persons in India today describe Sindhi as their mother

tongue. Of these, 91 per cent live in urban areas, especially in Maharashtra, Rajasthan,

Madhya Pradesh, and Gujarat (in that order).4 By far the most significant,

demographically and culturally, area of settlement in South Asia today is Bombay

(Mumbai), where Sindhis enjoy a fairly high profile with respect to the commercial,

social, and cultural life of the city. Sindhi restaurants, ‘colonies’ (residential enclaves

- see Falzon 2004), film financiers, actors and directors, and well-known

businesspeople and property developers, are synonymous with Mumbai.

By the early 1950s, therefore, the Hindu Sindhis who had left their homeland

were settled in scores of countries around the world in the case of the Sindworkis, and

in urban India in the case of other groups. The stage was set for the third major

episode of Sindhi diasporization, which kicked off in the early 1950s and gained

momentum in the last decades of the twentieth century. This time the historical

context was not Sind, but rather South Asia and the patterns of mobility that

characterized the region (Sindhis had become ‘Indian’, as it were). The main

destinations were the countries of the ‘West’ which at the time were major receiving

locations - Britain, Canada, the USA, and Australia.5 Later, and especially after the oil

crisis of 1973, thousands of South Asians - including Sindhis - moved to the Gulf

4
Whereas in Maharashtra, Madhya Pradesh, and Gujarat, the overwhelming bulk of Sindhi speakers

live in urban areas, it appears that in Rajasthan 38 per cent of Sindhi speakers come from rural areas.

This anomaly derives from the fact that a good number of Sindhi speakers in Rajasthan are not in fact

Partition refugees, but natives of the region.

5
One should note that, up to the enactment of the Commonwealth Immigrants Act of 1962, citizens of

the successor states of British India had the right to enter and settled in Britain (Peach 1994).
 10

countries. In Dubai, for example, Sindhis at one point dominated the textiles trade,

and they were also involved in ‘re-exporting’ consumer items from South Asia and

East Africa via Dubai to India, Russia, and elsewhere (Weiner 1982). More recently,

there has been a marked flow of highly-qualified Indians to the West in pursuit of

opportunities in the information technology sector, and again this includes a fair share

of Sindhis.

There is one final large-scale population movement (‘movements’, actually) of

Sindhis which is worth recording. Partly this concerns that old venture, Sindwork,

which continued to siphon people away from the sub-continent, on employment

contracts with the firms, well into the late twentieth century. Many of these eventually

set up their own businesses and in turn recruited more personnel. Also linked to

Sindwork are the multiple ‘experiences of rediasporization’ (Boyarin, as cited in

Clifford 1994) that Sindhis in, say, East Africa and Fiji went through as a result of the

specific politics of those locations and their effect on immigrant populations; in this

sense East Africa is of course the textbook case (see Gregory 1993). As one of my

bhaiband informants, whom I met dispensing free homeopathic cures in Ulhasnagar,

said, ‘Our family has been through so many partitions. We lost property in Lagos, and

Cambodia and Saigon in Indo-China. Now we are operating mainly from Manila in

the Philippines, although I have cousins in business in many other places’.

It is in this sense that the term ‘Sindhi diaspora’ in fact subsumes so many

different experiences. Each of the key moments discussed above is, as Clifford (1994:

302) put it, ‘embedded in particular maps and histories’. Somehow, however, this

heterogenous composite is rendered vertebrate by Sindhis themselves, who speak of

‘the diaspora’ and ‘Sindhayat’ (‘Sindhiness’). So much so, that the different moments

of mobility described above are homogenised into a continuous and unifying history
 11

which threads together Sindwork, Partition, late twentieth century migratory projects,

and so on. It is to this cohesive element that I now turn my attention.

The production of translocality

The Partition exodus and later cycles of mobility went a long way in shaping who

Sindhis are today. The second half of the twentieth century was characterized by a

number of processes from which emerges the contemporary Sindhi ‘community’.6

Telegraphically:

• Partition fostered a sense of Sindhayat through the device of common

dispossession of the homeland - a sense of ‘we are in it together’

• although the older (not ‘timeless’) jati distinctions still matter, there is a

growing sense of Sindhayat itself as a jati

• with respect to religion, two key processes are, first, the tendency of Sindhis

(especially Indian-based ones) to move towards a universal Hinduism, and,

second, the active construction of Jhulelal as ‘the Sindhi god’

• whereas in pre-Partition Sind there was a clear distinction between the

globally-mobile Sindworkis (and to a lesser extent Shikarpuris) and the more

located Hindu populations, the broader-based migratory processes decribed

earlier meant that most Sindhi families can now think of themselves as in

some way or another ‘mobile’ and ‘in diaspora’

• this model brings with it an active production of translocality, with respect to

kinship, marriage, travel, etc. The centre of this production is the city of

Mumbai.

6
I use quotes for ‘community’ because for the best part of twenty years anthropologists have not been

able to imagine communities other than imagined ones. Hardly imaginative, but I am no exception.
 12

Let us start with Sindhayat and its contents. The defining feature of Sindhayat, as

identified by my informants wherever I went, is what I have called (Falzon 2005) its

‘cosmopolitanism’. By this I mean two things: first, actual geographical spread

throughout the world; second, a very particular relation to places and the social

diversity that comes with them. Sindhi cosmopolitanism, therefore, is both a

geographical reality and a mindset, a way of relating to the world (see Vertovec and

Cohen 2002, Falzon forthcoming). In tangible terms, it is expressed and represented in

a number of ways.

Take religiosity, which is very telling for at least two reasons. First, for purely

ethnographic interest; second, because the data feed into my general thesis that

following Partition, the various disparate elements have tended increasingly to think

of themselves as forming part of a single, coherent diaspora. Historically, the

strongest religious current among Sindhis is Nanakpanth, which follows the teachings

of the first guru of Sikhism, Nanak (1469–1539), but not later developments such as

Khalsa or the figure of Gobind Singh (1666–1708), the tenth guru. Unlike Sikhs

‘proper’, so to speak (and this, admittedly, is a gross simplification), Nanakpanthis

also follow various devotions usually thought typical of Hinduism, such as the

worship of Hindu deities. Many Sindhis practice some form of vegetarianism and

home puja; most homes have a small mandir (shrine, temple) containing images of

both Guru Nanak and various Hindu gods, notably Lakshmi and Ganesh. Communal

places of worship, in India and elsewhere, differ in the relative prominence they attach

to the Guru Granth Sahib (the Sikh holy book) and murtis (statues of Hindu deities).

Likewise, religious calendars vary according to particular devotional leanings, but

usually incorporate both Sikh as well as Hindu auspicious days - notably dates like the
 13

gurpurabs (anniversaries of Sikh gurus), Mahalaxmi Sagra, and Ganesh Chaturthi.

Rituals include both the ceremonial reading of the Granth Sahib, as well as for

instance arati. In short, Nanakpanth practice among Sindhis is a variable field - and

this will come as no surprise to scholars of South Asian religion.

Apart from the ‘standard’ Nanakpanth and Hindu beliefs and rituals, a number

of devotions, usually linked to particular saints, enjoy a healthy following. In this

respect Satya Sai Baba, the Chinmaya mission, the Holy Mission of Guru Nanak,

Radhasoami, and the Sadhu Vaswani Mission, are especially noticeable. Last not

least, a number of Sindhis cultivate relations with Sufism, usually through an

attachment to pirs (masters) that may bind families as murids (followers) and pirs for

several generations.

So far so eclectic. However, and this is of cardinal importance, there is a

wealth of evidence pointing towards a general shift among Sindhis worldwide towards

a notion of a unified, reformed, ‘mainstream’ Hinduism. For instance, a number of

Sindhi pundits (usually of Brahmin extraction) I spoke to in Mumbai were critical of

what they saw as religious eclecticism, and maintained that Sindhis would do well to

focus on ‘pure Hinduism’. This view is commonly appropriated. Once, for example, I

overheard two worshippers at a temple complain about a ‘right Sindhi hotchpotch’; it

transpired they were referring to the juxtaposition of Granth Sahib and murtis. There

is a long and complex history behind this shift, and I should point out that it is not

specific to Sindhis; when, for instance, Gujarati khojas began moving in large

numbers to Mumbai and beyond during the nineteenth century, they discarded local

Kutchi and indeed Sindhi elements of their religious texts for the universal narratives

of reformed Hinduism (Devji 2006). With respect to Sindhis, however, one of the

reasons is undoubtedly the influence of the ‘saffron wave’ of Hindutva (see for
 14

example Hansen 1999, Vertovec 2000, Deshpande 1998), which made its presence

felt both in India (significantly, particularly in Mumbai via the Shiv Sena) and in the

countries of the diaspora. There are also older legacies such as the re-interpretation in

the nineteenth century of Hinduism and the spread of the Arya Samaj among Indians

worldwide (Baumann 1995). The shift may also have to do with receiving contexts;

in, say, centralizing Catholic contexts like Malta, it is easier and safer to be

‘recognizably Hindu’ than to appear eclectic and fragmented. I am not of course

arguing for some straightfoward notion of syncretism, but rather for a shift that is

partly situational-contextual, partly the legacy of long-term shifts in South Asian

religious practice and politics.

One facet of this shift, and therefore of the crystallization of a unified

diasporic imaginary, is the post-Partition reinvention of the god Jhulelal. In Sind,

Jhulelal had long been revered by sections of both Hindus and Muslims, although in

different ways. (There are indications, for example, that the anti-Muslim currents that

accompanied the deposition of the Talpurs, Jhulelal came increasingly to be

represented by Hindus as a mythical hero who saved them from Muslim tyranny.) In

any case, in pre-Partition Sind the devotion to Jhulelal was limited and somewhat

localized in the north of the province. In the 1950s, however, a group of Sindhis in

Bombay led by Larkana singer and cultural entrepreneur Ram Panjwani decided to re-

invent Jhulelal as ‘the god of Sindhis’ - of all the Sindhis in diaspora anywhere in the

world, that is, irrespective of jati or region of origin back in Sind. Gobind Malhi, who

had been one of Panjwani’s closest associates, described the idea to me when I met

him in Mumbai in 2000, as an attempt to ‘provide a thread to the scattered beads and

make a necklace’.
 15

Several decades later, it seems that the project has succeeded. Sindhis mandirs

worldwide now venerate Jhulelal, hold special rituals like the bahrano sahib, and

celebrate Jhulelal Chand as particularly auspicious. My informants invariably

described Jhulelal as a ‘community god’. He appears on Sindhi shop signs, Internet

sites, business cards, and so forth - in short, he has become the symbol of a unified

diasporic Sindhayat. Since he is also thought to be an incarnation of Vishnu, this

brings Sindhayat neatly in line with a universalising Hinduism. It also effectively

grafts it onto the notion, itself a recent historical product, of Hinduism as a ‘world

religion’. I quote Masuzawa (2005, 200):

‘World religions’ as a category and as a conceptual framework initially developed in

the European academy, which quickly became an effective means of differentiating,

variegating, consolidating, and totalizing a large portion of the social, cultural, and

political practices observble among the inhabitants of regions elsewhere in the world.

Masuzawa is saying that the imaginary of a finite number of world religions, each a

formally equivalent cultural system, is a product of a specific history. In the centuries

of European and therefore Christian world hegemony, a particular model of ‘world

religions’ has spread, under whose influence adherents of previously loosely affiliated

local cults have sought or have been encouraged or compelled to conform to the

prescriptions of some ‘great traditon’ (in this case ‘Hinduism’). At the same time,

‘great traditions’ themselves have been modified to fit the requirements of the

template provided by the world religion model (Cook et al. 2009).

Religion among Sindhis today, then, thrives on the image of a bounded

transnational group united under the incorporative banner of ‘world Hinduism’ - a

‘global systems’ trope which happens to dovetial very beautifully with representations

of a worldwide diaspora. It is fair to say that the situation is one of ‘less Sindhi, more

Hindu/Indian’. For example, at a number of events held as part of the EU Year of

 16

Intercultural Dialogue 2008, Sindhis in Malta were invited to ‘represent’ (and

‘celebrate’) the Hindu Indian ‘community’ in Malta. In this sense they are imagined

as purveyors of the East and its alternative lifestyles such as yoga, vegetarianism,

‘Hindu texts’, etc.

There are other aspects that produce the contemporary ‘community’. Travel

and the geographical conquest of space are of prime importance. The vernacular

histories produced both at popular-oral level and in print by indigenous authors

invariably emphasize the dispersal of Sindhis, and their sense of adventure and mobile

enterprise. Malkani (1984: 169), to cite a typical example, writes that ‘others had

found a Sindhi enterprise even on Falkland Islands near the South Pole’; likewise,

Buxani (as cited in Panjwani 1987: 95, my parenthesis) jests that ‘A Sindhwarki

[Sindwork] post has not yet been set up in the new Antarctica settlements of scientists

but a Shikarpuri who operates gold business in Alaska is believed to be working on

it’.7 It is hardly surprising that this wanderlust is shared by Sindhi gurus. 90-year old

Dada J.P. Vaswani, for example, is well known as an ‘international saint’; the

initiatives of the Sadhu Vaswani Mission, such as the annual ‘Meatless Day’, are

equally cosmopolitan in scope.

Another self-attribution by Sindhis, which dovetails with my argument that

Sindhayat has become a cosmopolitan tag, is that of adaptability. In terms of religion,

it is commonly believed that Sindhi beliefs and practices are especially ‘open-minded’

7
Note that ‘Shikarpuri’ and ‘Sindworki’ are being used interchangeably, under the general category

‘Sindhi’. It is interesting that, whereas pre-Partition vernacular literature tended to focus on specific

jatis (see for instance Narsain 1932), after 1947 we increasingly come across generalising writings on

‘Sindhis’ (see for instance Panjwani 1987, Hiranandani 1980, Malkani 1984). This supports my

argument that one of the long term effects of Partition was to create a sense of unifying Sindhayat for

Hindu Sindhis in diaspora.

 17

and accommodative - at times to a fault, in the sense that a lack of identity and

bounded specificity is inevitably a corollary of openness. Linguistically, too, I was

often told by my informants that Sindhis are especially adept at learning new

languages, and that this has been pivotal both to their success as diasporic

entrepreneurs and to their integrative abilities wherever they are settled. What I find

especially interesting is that indigenous narratives explain this openness with

reference to the specifics of Sindhi history, as a group located at the border between

south and central Asia. Linguistically, it is thought that the fact that Sindhi may be

written in the Urdu as well as the Devanagari script, and that it enjoys the phonetic

legacy of both types, gives speakers an edge over historically less syncretistic

traditions. Rather like religion in fact, as Nanakpanth and Sufism are often invoked as

exemplars of a historical, and specifically Sindhi, eclecticism. (I should emphasize the

situationality of this notion, which also means that it can co-exist, peacably if not

always happily, with broader shifts towards a more rigidly-defined Hinduism.) The

point is not to establish whether or not Sind and Sindhis were/are particularly

syncretistic and accommodative, but rather to note how they knead together elements

of their history in order situationally to represent themselves as a cosmopolitan type

that has been long in the making.

Another central aspect of Sindhi cosmopolitanism is kinship, actively

manufactured through transnational marriage matching. The biographies and

genealogies of Sindhi families are characterised by a ‘circulation of women’, so to

speak, across space. Like many other north Indian groups, Sindhis are patrilocal.

Women are brought up with the idea that marriage involves becoming part of another

family and leaving the natal home; a daughter is parai jai (literally, ‘belongs to

someone else’), and the ideal wife is one who respects her husband’s agnates,
 18

particularly his parents. In the case of joint families - which are still important,

sometimes episodically so, even as the nuclear family model makes its presence felt -

she will also live with the husband’s family. Which does not, of course, mean that

married women are cut off from their natal families. On the contrary, and

notwithstanding the notion that daughters are ‘married off’, their long-term location

between two families brings these families together - to attend each other’s feasts and

important occasions, to exchange gifts, and, sometimes to do business with each other

(I came across several instances in the field of brothers-in-law business partnerships).

Further, what is especially significant is that, since marriage among Sindhis is often

arranged across long distances, this results in a long-term process of circulation of

women between the various countries and locations where Sindhis are settled. It is not

just a family and home which women leave at marriage, but quite often a country.

This is especially true in the case of bhaiband and/or Sindworki business families,

who still generally have the most transnational kinship diagrams of all jatis/groups.

I have so far borrowed the Levi-Straussian term ‘circulation of women’. It

would be a mistake, however, to think that women are in any way passive actors

waiting to be transnationally circulated by men. Rather, the broad and cosmopolitan

networks of affines and agnates - what Kelly (1990), referring to Gujaratis, calls

‘transcontinental families’ - that make up a typical Sindhi family are very much

forged by women. First, the fact that girls are brought up thinking that marriage

involves translocation, does not mean that they are ready to move anywhere. On the

contrary, variables like the size and type of Sindhi settlement, ‘open-mindedness’ with

respect to women’s rights and aspirations, affordability of domestic service, and such,

are all taken into account by women when sizing up a potential match, and therefore
 19

destination.8 Second, the type of transnational information that goes with such long-

distance matching is very much held and exchanged by women. The knowledge that

‘aunties’ - especially bhaibands and/or ones that are already embedded in

transnational families themselves - have of individuals and families, marriageability

and eligibility, and reputations, can be breathtaking in its detail and geographical

scope. The rounds of lunch parties, ‘kitty parties’, and satsangs that tend to occupy

better-established women, serve as venues for the active exchange of such

information. In sum, if women are ultimately circulated, they also do quite a bit of

circulating themselves. The Sindhi diaspora, although synonymous in the popular

imagination with Black Label-guzzling mobile businessmen, is in at least equal

measure the manufacture of enterprising women.9

There remains one facet of the Sindhi cosmopolitan diasporic imaginary,

without a discussion of which my argument would not be complete. It concerns the

notion of ‘home’ - which, in Safran’s (1991) landmark ideal-typic definition of

diaspora as well as subsequent formulations (see for instance R. Cohen 1997,

Vertovec 2000), is one of the key criteria of such a phenomenon. In the case of

Sindhis, the original homeland is of course now part of Muslim Pakistan. Save for an

increasingly slim generation of people who remember pre-Partition Sind, there is little

sense of an emotive relation to it (although, as discussed earlier, the memory of Sind

is important for Sindhis to define who they are, and in what ways), and in no case is

there a narrative of eventual ‘return’ or ‘restoration’. Does this mean, therefore, that

the Sindhi diaspora lacks a centre? Not necessarily. What has in fact happened is that

8
Although marriages are arranged, most parents have learned from experience that it is unwise to

disregard their sons’ and daughters’ views.

9
On gender and the Indian diaspora, see for instance Rayaprol (1997).
 20

India, and in particular Mumbai, has come to be seen as, if not the real thing, at least

the cultural heart of Sindhis worldwide.

Demographically, Mumbai represents the single most significant settlement of

Sindhis anywhere in the world. Apart from the hundreds of thousands of Sindhis who

actually live in the city and its suburbs, most families in diaspora have and cultivate

some connection to it. December in Mumbai has become something of an institution

among Sindhis, as thousands converge on the city to meet friends and/or relatives, to

enjoy themselves, to attend the numerous weddings and social occasions, to visit the

renowned Sindhi pundits based there, and possibly to advance marriage-matching

projects at the various ‘marriage bureaux’ run usually by well-connected ‘aunties’. As

one pundit put it to me, ‘Mumbai is like a huge sea into which numerous small rivers

flow’. He might have added ‘and from which’, for it is the case that, even as business

reputations, personal and family biographies, and indicators of wealth and success,

flow into the city from all over the world, they are eventually re-distributed - possibly

changed by virtue of interaction - to the pilgrims’ points of origin. In this sense, to use

a notion from Merleau-Ponty’s writings on the theatre (1964), Sindhis going to

Mumbai become voyantes-visible, viewers that are themselves visible. Which is why I

was so often reminded that a majestic appearance and the requisite expensive sari and

jewellery at a big Mumbai wedding, goes a long way into affirming the transnational

reputation of a family; it is also why so many jokes circulate in the city about the

‘showiness’ of Sindhis. Again, women have a crucial part to play in this game.

I am now in a position to discuss that aspect of Sindhi diaspora that so fires the

popular imagination, especially in India: the extent and resilience of global business

networks.
 21

Global business, mobility, and exploration

Is there such a thing as ‘Sindhi business’? In the sense of some single structured set of

operations with limited and clearly-bounded lines of trade, definitely not. Sindwork

may have come close to that, at least in its early stages, as may the Shikarpuri shroffs

with their banking practices. Contemporary Sindhi-owned businesses, however, are as

variable as the entrepreneurial drives behind them, their contingent biographies, and

the local contexts they operate in. In London, for example, Sindhis are active in the

financial agency, import and export, wholesale, retail, and property development

lines, to name but a few. In Malta, the descendants of the original Sindworkis have

diversified into textiles, furniture, industrial supplies, and retail. In Mumbai, textiles

wholesale, small-scale manufacture, import, retail, and property development are

among the hundreds of lines that Sindhis ply. The list is proverbially endless, the only

constant being that Sindhi business is variable, explorative, and enterprising in its

innovation.

At the same time, empirical research into the connective possibilities offered

by transnational kinship and ethnicity, and the ways in which these relate to economic

activity, suggests that, in business as in other things, Sindhayat matters. In this I am

fully in accordance with Markovits (2000: 284) when he states that ‘globally, Sindhi

businessmen have remained a community of traders ... international trading linkages

created by the dispersion of Sindhi families across the world are the key ... to the

success of Sindhi firms’. I shall now proceed briefly to outline how this is so.

First, there can be no question that the popular association between the Sindhi

diaspora and business is well founded. Like other Indian commercial groups (see for

instance Lachaier 1997 for Kutchi Lohanas), the main occupational distinction made

by Sindhis is that between (self-employed) ‘business’ and (salaried) ‘service’. This

 22

distinction has deep-rooted historical antecedents, fostered in part by a British

enumerative-orientalist modality (see Cohn 1996). It is also jati-related, as with for

instance bhaibands and amils. The point here is that this distinction is not neutral but

hierarchised, in the sense that, with the possible exception of well-established

professional amil families, Sindhis in general see business as a superior occupation to

service. Further, one notes that post Partition it is Sindhis rather than specific jatis that

have become associated with business enterprise.

In practice, the upshot is, first, that the numbers bear out the model, and

second, that there is a marked tendency for Sindhis ‘in service’ eventually to set up

their own businesses. A detailed statistical examination is beyond the scope of this

talk (see however Falzon 2005) but my approximations (I emphasize) from my sites

indicate that 95 per cent, 70 per cent, and 65 per cent of Sindhis are in self-employed

business in Malta, London, and Mumbai respectively. (The variance has more to do

with specific diasporic trajectories than with receiving contexts.) With respect to the

second point, Sindwork provides an excellent case study. I said earlier that Sindworki

firms continued to draw upon the salaried labour of Sindhi recruits well into the

twentieth century. However, these recruits were seldom ‘content’ with being

employees. The majority of them eventually sought to strike out on their own, and in

fact many Sindhi businesses operating worldwide today were originally set up by

erstwhile employees of Sindwork firms. So much so, that joining a Sindwork firm was

considered a potential stepping stone from ‘service’ to ‘business’.

Which brings us to corporacy, or to what makes ‘Sindhi business’ so particular

- in Markovits’ (op.cit., my emphasis) terms, to the ‘international trading linkages’ it

is synonymous with. The conventional wisdom posits a logical, and sometimes

empirical, connection between commerce and diaspora, in the sense that it is easy to
 23

see how transnational linkages of kinship and Sindhayat (which, at least in as much as

it also summons a metaphor of shared blood, can be see as an extension of kinship)

can double as relations of trade. Indeed, it is very common to find Sindhis (both

family and co-ethnics) doing business together - both in a corporate way, as partners,

and as temporary exchange relation or longer-term series of relations. In a typical

actual example, a retailer running an import/wholesale business in Liberia is sourced

by a London business operated by his son, who in turn is partly sourced by a maternal

cousin based in the US. These linkages are only very seldom incestuous, and rather

are usually combined with others involving non-family and/or non-Sindhis. In the

above example, the London business is also sourced by British manufacturers and a

Brazilian owned and based company. However, the measure of social control - and

therefore trust - afforded by kinship and Sindhayat certainly makes ‘inward-looking’

trading linkages desirable, especially when it comes to credit. We might keep in mind

the intersection between transnational marriage, Mumbai and marriage-matching, and

individual/family reputations - which means that word of dishonest business practice

spreads rapidly through the diasporic network, and will usually come to haunt the

guilty party in kind and much more.

The combination of transnational connections and readiness to apply them to

business makes Sindhis what they are. It is not that they are a unique case of

translocally-organised jati. The literature on Indian commerce is rich in examples of

commercial castes. Hazlehurst’s Punjabi Banias (1966), C.A. Bayly’s ‘geographically

extended kin groups’ (1978), Timberg’s Marwaris (1978), Mines’ Kaikkoolars

(1984), and Rudner’s Nattukottai Chettiars (1994) are among the more memorable

examples, and indeed I would argue that there is a mismatch to be addressed between

such empirical knowledge and the tendency of generalising theory on caste to over-
 24

emphasize embeddedness in locality (Falzon in prep.). Even so, Sindhis are justifiably

renowned for their culture of mobile commerce, exploration, and business resilience.

The typical Sindhi business is positioned - however loosely and flexibly -

within a transnational network. This means that knowledge of markets and

opportunities, sometimes from across the world, is available to a very significant

extent. I was constantly struck, for instance, by just how up-to-date informants were

with current business prospects and conditions in a range of localities. This is not a

matter of ‘as it happens’, but rather a characteristic that individuals cultivate and

produce in an ongoing process - being ‘in diaspora’ is a practice rather than a given

state of affairs. As a London businessman put it to me, ‘Making money has a lot to do

with spreading out members of the same family. The big Sindhi families, those

leading the field in business, are all dispersed and living in several countries.’ Given

this translocal knowledge and familiarity with markets, it should come as no surprise

that Sindhis in general are ready to move around, open new businesses and/or try new

lines; the general impression, quite empirically sustained, is that Sindhi businesses

‘come and go’ in the short term, but endure in the long term. If they do not actually

translocate in search of good fortune - and many are very stable and settled in

particular places - they may shift sources, enabling them to experiment with different

products and price differentials. Unstable political contexts, for example, which offer

a potential for good profits due to minimal competition and regulation, have

historically tended to attract Sindhi businessmen, who can afford to take risks not

least because assets and operations are seldom invested exclusively in such places. An

interesting case I came across was the French-Dutch island of Saint Martin’s in the

Lesser Anitilles. In the early 1970s, only a handful of Sindhis did business there; as

the Caribbean began to take off as a mass tourism destination, however, several
 25

hundred flocked there from all over the world and today they control a significant

slice of the tourist trade. Such examples show that the link between commerce and

diaspora is not just an old piece of rhetoric - it really matters to ‘be’ everywhere (in

person or by proxy) and be ready to sell anything.

Does this therefore mean that the Sindhi diaspora is one big happy family,

united in its love for Jhulelal, intermarrying, and doing business together? Not

necessarily. In fact, the link between commerce and diaspora tends to be rather

episodic in nature. During fieldwork I encountered a contradiction. On the one hand,

people told me that doing business together (as kin and/or co-ethnics) was a thing of

the past, a symptom of a pre-modern mentality, and that they would never dream of

being partners or extending credit to a Sindhi. However, when these same individuals

narrated their individual biographies, it usually emerged that they had, at some point

in their career, done the undreamable. There are points at which collaboration

becomes important - collectively, as in the cases of the immediate post-Partition years

and the expulsion of South Asians from East Africa in the 1970s, or individually, as

when one needs that little initial credit to strike out on one’s own, or to resuscitate an

ailing business. Then, transnational kinship- and ethnic-based solidarities may become

temporarily and situationally crucial - as others fail, one might say. This is why Sindhi

businesses are reputed to have the ability to ‘come back with a bounce’, although not

necessarily to the same place and in the same line. This is also what Susan Bayly

(1999: 320) means when she writes that, ‘in uncertain times, a wide range of

“modern” Indian businesses have continued to find that profit margins can be

protected or enhanced by pooling assets and sharing information with kin and caste

networks’. The quotes around ‘modern’ assume, as my informants did, that kin, caste,

and ethnic solidairities have little place in modern business, which presumably runs
 26

on faceless bureaucracy and bloodless exchanges of currency. If we stick to this

modern:pre-modern dichotomy, which is probably analytically useful in this case, my

argument in sum would be that the two very much co-exist, and each gains

ascendancy or loses out episodically.

For a Sindhi in business, it does not matter all the time to be part of a

transnational diaspora; it may well matter, however, when one needs it most. Without

sounding instrumentalist or indeed cynical, it makes sense for Sindhis to invest in

Sindhayat in the long term - through marriage, visits to Mumbai, extending credit to

family and other Sindhis, and so on. Whether one is dancing to Bollywood tunes and

having a good time at a wedding party, celebrating bahrano sahib on the Chennai

waterfront, or phoning a cousin to ask for a grace period on an unsold shipment, it all

ultimately falls into place. It may be worth referring to Diouf’s (2000) work on

Senegalese Murids. Murids constitute a successful trade diaspora and display many of

the characteristics – such as linguistic proficiency and commercial networks –

discussed in the present work. They are tightly organized primarily on the basis of

ritual and location, and their cosmopolitanism is very much what Cohen (2003) calls

an ‘institution of stability-in-mobility’.

Cosmopolitanism in practice

In conclusion, I wish to take you back to the notion of cosmopolitanism. It is clear

that we are dealing here with what Robbins (1998) calls an ‘actually-existing’

cosmopolitanism – an empirical/ethnographic instance rather than a theoretical

concept, that is. To my mind, one of the reasons why there is a real and urgent need to

look at these practical instances is that scholars have tended to privilege

cosmopolitanism in its philosophical sense at the expense of more vernacular types

 27

which seem to be open to the accusation of banality. However, quite apart from the

fact that this distinction is uncomfortably reminiscent of otherwise-obsolete high:low

culture dichotomies, there are important reasons why cosmopolitanism in practice

matters. First, all cosmopolitanisms are to some extent actually-existing in that they

are located within some historical and geographical framework; thus, as Stuart Hall

(2008) argues, even the towering universalistic political projects of Kant turn out to be

‘harnessed back’ to a very specific ideological and cultural movement, namely the

Western Enlightenment. Second, actually-existing cosmopolitanisms deserve our

primary scholarly attention, since their actual existence presumably means that they

are actually consequential (as opposed to Utopias, which are just that). Third, to look

at the maps and histories – and therefore the limits and limitations – of particular

cosmopolitanisms is partly to absolve oneself of the nagging feeling that

contemporary social science, in waxing lyrical about ‘flows’, ‘fractals’, and ‘fluidity’,

is (once again, some might say) being the handmaiden of contemporary regimes of

production, accumulation, and power – in our a case a neo-liberal ‘globalizing’

agenda in which the ‘emphasis falls more on individualist aspirations and

universalizing norms’ (Pollock et al. 2000, 581). Related to this is the argument that

looking at practical (mundane) cosmopolitanism will serve to redress the balance in

favour of various groups for whom belonging everywhere (and therefore nowhere) is

not a choice but a predicament. ‘Elite’ political models of world citizenship have their

rightful place, but they do not explain everything. Finally, understanding the social

organization of cosmopolitanism is very much about overcoming trite local-global

dichotomies. Smith’s work (2001) on ‘agency-oriented’ ‘transnationalism from

below’ in contemporary cities constitutes an important point of reference here. In a

 28

way, this talk seeks to do for people (ethnic groups) what Smith does for places

(cities).

Talking about actually-existing cosmopolitanisms, however, raises the

perennial spectre of definition. In this sense the difficulty that so preoccupied Weber

is still with us. In a nutshell, if all cosmopolitanisms exist within, and therefore take

on some of the characteristics of, specific histories and geographies, this very

specificity seems to preclude us from generalizing in any useful way. Scholars of

cosmopolitanism (notably Vertovec & Cohen 2002) have, quite successfully I think,

sought to circumvent this problem by creating inclusive typologies – rather than

exclusive definitions - of cosmopolitan attributes which the various empirical

instances we observe to some extent share. I have argued that ‘Sindhiness’ displays

the following attributes:

• A worldwide distribution.;

• a problematic relation to the nation-state;

• an attitude or disposition which means that Sindhis, to quote Vertovec &

Cohen (2002, 13), can ‘end up anywhere in the world and be in the same

relation of familiarity and strangeness to the local culture, and feel partially

adjusted everywhere’;

• a marked competence at operating within translocal business networks;

• and, finally, kinship practices based on marriage alliances beyond the person’s

immediate locality and which therefore result in cosmopolitan genealogies.

It is important to stress that, taken individually, none of these attributes leads to

cosmopolitanism. Given the ongoing inflation of the term (as has happened with

‘diaspora’ – see Brubaker 2005) for instance, it is clear that self-ascription alone
 29

cannot be indicative of a tangible cosmopolitanism. Equally, just as travel does not

necessarily broaden one’s horizons (as anyone who has enjoyed the dubious pleasure

of tourist enclaves knows), worldwide distribution per se does not make the

cosmopolitan grade. Taken in conjunction, however, these family resemblances mean

that belonging to the Sindhi ethnic group constitutes a powerful and well-trodden

trajectory into the (currently desirable) cosmopolitan way of engaging with the world

– even if mitigated by factors such as caste, resources, location, occupation, and level

of education.

This ongoing dynamic is partly the result of an issue which all actually-

existing cosmopolitanisms must face, namely that of a problematic relation with

locality. To my mind, one reason why cosmopolitanisms are burgeoning is that

contemporary processes are drawing more and more people into tricky legal, political,

economic, and cultural relations with established models of location. The fact

remains, however, that in spite of all the routes and rhizomes, a majority of groups

still define themselves in terms of localities in which they claim some sort of

historical continuity. In the contemporary world, states and the groups that ‘belong’ to

them, nations, are probably the most important manifestations of such localized self-

definition. Not surprisingly, cosmopolitans have always found themselves in awkward

positions vis-à-vis formations like nation states and city states, and their requisite

allegiances. The problematic leads to the internal dialectics I discussed earlier, the

protagonists of which can sometimes get quite carried away. In the case of Sindhis,

for example, a very few cultural entrepreneurs have gone so far as to suggest the

recreation of an exclusively-Sindhi homeland, two of the nominations being the desert

of Rajasthan and the Nicobar Islands (!)

Perhaps more interestingly, these colourful relations with locality can produce
 30

dividends. Historically, for instance, the strangeness/alterity factor may have helped

Sindworki businesses in the sense that their worldwide connections, and therefore

geographically-eclectic wares, ultimately made them stand out as purveyors of the

exotic. Sindworki shops may have seemed strange and ‘out of place’, but it was

precisely this characteristic which gave them the edge over less outward-looking local

establishments. This comes close to what Vertovec & Cohen (op. cit., 7) call an

‘aesthetic cosmopolitanism [based on] ... forms of consumption’, and can be seen in

the kimonos sold in Sindworki shops in Mediterranean harbour towns in the 1920s, or

the incense sticks and ‘ethnic’ carvings that go down so well with customers today.

The problematic relation to locality as a factor which fosters what I have been

calling the ‘internal dialectics’ of actually-existing cosmopolitanisms, cannot be seen

in isolation. In the case of Sindhis a number of variables such as caste and divergent

histories of mobility have left individuals and families within the ethnic group in very

different structural positions. There are big differences between the ‘old money’

Sindworki families, with their webs of kinship stretching across the globe and

reserves of mobile capital, small entrepreneurs who run textiles businesses from small

offices in Ahmedabad, and amil civil servants living in Mumbai. The structural

differences also mean the production of different spatializing discourses and practices.

On a global level, for instance, Sindhis have shown a marked reluctance to engage

with local politics (generally by being on good terms with whoever is in power,

irrespective of partisan alignments.) On the other hand, Sindhis settled in India have

readily involved themselves in Hindu nationalist politics, directly (as in the case of

high-profile politicians like L.K. Advani), or indirectly (a significant number of

Sindhis settled permanently in Mumbai cultivate close relations with the Shiv Sena).

This involvement has a history and we find that the RSS (Rashtriya Swayamsevak
 31

Sangh, a vaguely para-military Hindu nationalist organization) enjoyed considerable

support among amil civil servants in pre-Partition Sind. The ambiguity can partly be

understood as a factor of the significance, to the ethnic group, of particular localities:

For social, linguistic, historical, and political reasons, India and, say, Britain mean

very different things to Sindhis.

It is crucial, therefore, to de-essentialize and de-homogenize ethnicities, not

least cosmopolitan ones. Once cosmopolitanism is put into practice and embedded

within actual social forms, it ceases to be normative and straightforward and develops

a whole new dialectical dynamic which brings into play a host of variables. This

complexity is behind the insistence of scholars like Vertovec & Cohen (op. cit.) to

think of cosmopolitanism as a form of imagination rather than an essential quality.

For Clifford (1998), it is precisely this idea which gives cosmopolitanism (both as

analytical concept and empirical practice) the edge over competing notions like

multiculturalism. It is not as if Sindhis are condemned to be cosmopolitan; they may

choose to organize themselves so, and quite often they do.

It also sometimes happens that groups and/or individuals within groups

become simultaneously more and less cosmopolitan. With Murids, successful and

mobile traders with business stretching from Strasbourg to Chinatown in New York

seek to enhance their chances to be buried in the hallowed ground of Touba in

Senegal (Diouf op. cit.). With Sindhis, the traders that ‘make it big’ with respect to

wealth, business networks, and transnational kinship connections, tend to engage in

practices that seem antithetical to cosmopolitanism – ‘pure’ Brahminical lifestyles

(see Bayly 1999), stricter endogamous practices, ostentatious investment in specific

localities, etc. All this seems contradictory, but I have argued that it is not. On the

contrary, it is only by looking at its apparently non-cosmopolitan forms of

 32

organization that cosmopolitanism as a practical way of engaging with the world

begins to make sense.

References

Ahmad, S. M. 1960. India and the Neighbouring Territories in the Kitab Nuzhat Al-Mushtaq Fi’

Khtiraq Al-‘Afaq of Al-Sharid Al-Idrisi. Leiden: E.J. Brill.

Anderson, B. 1991. Imagined Communities: Reflections on the Origin and Spread of Nationalism

(revised ed.). London: Verso.

Baumann, M. 1995. Conceptualizing Diaspora: The preservation of religious identity in foreign parts,

exemplified by Hindu communities outside India. Temenos, 31, 19-35.

Bayly, C.A. 1978. Indian Merchants in a ‘Traditional’ Setting: Benares, 1780-1830. In The Imperial

Impact: Studies in the Economic History of Africa and India (eds) A. Hopkins & C. Dewey,

171-93. London: The Athlone Press for the Institute of Commonwealth Studies.

Bayly, S. 1999. Caste, Society and Politics in India from the Eighteenth Century to the Modern Age

(The New Cambridge History of India IV.3) Cambridge: Cambridge University Press.

Brown, J.M. & R. Foot. 1994. Introduction: Migration – The Asian Experience. In Migration: The

Asian Experience (eds) J.M. Brown & R. Foot, 1-11. St. Martin’s Press & St. Antony’s

College, Oxford.

Brubaker, R. (2005), The ‘diaspora’ diaspora, Ethnic & Racial Studies 28:1, 1-19.

Cheesman, D. 1981a. Local Power and Colonial Authority: The Pax Britannica in 19th Century Sind.

History Today, 31, 23-6.

Cheesman, D. 1981b. Power in Rural Sind in the late Nineteenth Century. Asian Affairs, XII (I), 57-67.

Chugani, A. G. 1995. Indians in Japan: A Case Study of the Sindhis in Kobe. Unpublished M.A. Thesis

submitted to the Graduate Division of the University of Hawaii, May 1995.

Clifford, J. 1994. Diasporas. Cultural Anthropology, 9 (3), 302-38.

Clifford, J. (1998), Mixed Feelings, in P. Cheah & B. Robbins (eds.), Cosmopolitics: Thinking and

feeling beyond the nation (Minneapolis and London: University of Minnesota Press).

Cohen, A. 1971. Cultural strategies in the organization of trading diasporas. In The Development of

Indigenous Trade and Markets in West Africa (ed.) C. Meillassoux, 266-81. Oxford: Oxford

University Press.
 33

Cohen, R. 1997. Global Diasporas: An Introduction. London: UCL Press.

Cohn, B.S. 1996. Colonialism and its forms of knowledge: The British in India. Princeton

University Press.

Cohen, A. (2003), Custom and Politics in Urban Africa: A study of hausa migrants in Yoruba towns

(London: Routledge).

Cook, J., Laidlaw, J. & Mair, J. 2009. What if there is no elephant? Towards a conception of an un

-sited field. In Multi-sited Ethnography: Theory, praxis, and locality in contemporary

research (ed.) Mark-Anthony Falzon. Aldershot: Ashgate.

Deshpande, S. 1998. Hegemonic Spatial Strategies: The Nation-Space and Hindu Communalism in

Twentieth-Century India. Public Culture, 10, 2, 249-83.

Devji, F. (2006), ‘Portrait of a Remarkable Community’, Economic and Political Weekly, 21 October

2006, 4435–7.

Diouf, M. (2000), The Senegalese Murid trade diaspora and the making of a vernacular

cosmopolitanism, Public Culture 12:3, 679-702.

Falzon, M.A. (2003), ‘Bombay, our cultural heart’: Rethinking the relation between homeland and

diaspora, Ethnic and Racial Studies 26:4, 662–83.

Falzon, M.A. (2004), ‘Paragons of Lifestyle: Gated communities and the politics of space in Bombay’,

City & Society 16:2, 145–67.

Falzon, M.A. (2005), Cosmopolitan Connections: The Hindu Sindhi Diaspora, 1860–2000. New Delhi:

Oxford University Press. (First published in Leiden & Boston by Brill Academic Publishers,

2004.)

Falzon, M.A. (forthcoming), ‘Ethnic Groups Unbound: A case study of the social organization of

cosmopolitanism’. In Cosmopolitanism in Practice (eds) M. Nowicka and M. Rovisco.

Arlington: Ashgate.

Falzon, M.A. (in prep.), ‘Caste and space: Towards a disembedding of classical South Asian

categories’. (Preliminary draft read at the Anthropology Department Senior Seminar Series,

Cambridge, March 2006).

Floor, W. 1993-4. The Dutch East India Company (VOC) and Diewel-Sind (Pakistan) in the 17th and

18th centuries. Karachi: University of Karachi.

Gregory, R. C. 1993. South Asians in East Africa: An Economic and Social History, 1890-1980.
 34

Boulder, CO: Westview.

Hall, S. (2008), Conversation with Pnina Werbner, in P. Werbner (ed.) Anthropology and the New

Cosmopolitanism: Rooted, feminist and vernacular perspectives. ASA Monograph No. 45

(Oxford: Berg).

Haller, D. (2005), ‘Let it flow: Economy, spirituality, and gender in the Sindhi network’,

Anthropological Theory 5:2, 154–75.

Hansen, T.B. 2001. Wages of Violence: Naming and identity in postcolonial Bombay. Princeton and

Oxford: Princeton University Press.

Hazlehurst, L. W. 1966. Entrepreneurship and the Merchant Castes in a Punjabi City. Monograph and

Occasional Papers Series No 1, Duke University.

Hiranandani, P. 1980. Sindhis: The scattered treasure. New Delhi: Manohar.

Hotchand, Seth Naomul 1915. A Forgotten Chapter of Indian History as described in the Memoirs of

Seth Naomal Hotchand, C.S.I., of Karachi 1804-1878 (Written in Sindhi in 1871, translated

by Rao Bahadur Alumal Trikamdas Bhojwani, edited and with an introduction by H. Evans

M. James). Exeter: William Pollard.

Kelly, E. 1990. Transcontinental families - Gujarat and Lancashire: a comparative study of social

policy. In South Asians Overseas: Migration and ethnicity (eds) C. Clarke, C. Peach & S.

Vertovec, 251-67. Cambridge: Cambridge University Press.

Lachaier, P. 1997. The Merchant Lineage Firm and the Non-Invisible Hand: Pune, Maharashtra. In

Webs of Trade: Dynamics of business communities in western India P. Cadène & D. Vidal

(eds), 23-52. New Delhi: Manohar.

Malkani, K.R. 1984. The Sindh Story. New Delhi: Allied Publishers Private.

Markovits, C. 2000. The Global World of Indian Merchants 1750-1947: Traders of Sind from Bukhara

to Panama. Cambridge: Cambridge University Press.

Masuzawa, T. 2005. The Invention of World Religions: Or, how European universalism was preserved

in the language of pluralism. Chicago: Chicago University Press.

Merleau-Ponty, M. 1964. The Primacy of Perception Northwestern University Press.

Mines, M. 1984. The Warrior Merchants: Textiles, trade, and territory in south India. Cambridge:

Cambridge University Press.

Narsain, S.J. 1932. The Amil Community of Hyderabad (Sind). Unpublished Thesis submitted to the
 35

School of Economics and Sociology, University of Bombay.

Panjwani, R. 1987. Sindhi ain Sindhyat. Bombay: The author.

Peach, C. 1994. Three Phases of South Asian Emigration. In Migration: The Asian Experience (eds)

J.M. Brown & R. Foot, 38-55. St. Martin’s Press & St. Antony’s College, Oxford.

Rayaprol, A. 1997. Negotiating Identities: Women in the Indian diaspora. New Delhi and New York:

Oxford University Press.

Robbins, B. (1998), Actually Existing Cosmopolitanism, in P. Cheah & B. Robbins (eds.)

Cosmopolitics: Thinking and feeling beyond the nation (Minneapolis and London:

University of Minnesota Press).

Rudner, D. W. 1994. Caste and Capitalism in Colonial India: The Nattukottai Chettiars. Berkeley:

University of California Press.

Safran, W. 1991. Diasporas in modern societies: myths of homeland and return. Diaspora, 1 (1), 83-

99.

Smith, M.P. (2001), Transnational Urbanism: Locating globalization (Oxford: Blackwell).

Timberg, T. A. 1978. The Marwaris: From Traders to Industrialists. New Delhi: Vikas.

Vertovec, S. 2000. The Hindu Diaspora: Comparative patterns. London: Routledge.

Vertovec, S. and Cohen, R. 2002. Introduction: Conceiving cosmopolitanism. In Conceiving

Cosmopolitanism: Theory, context, and practice (eds) S. Vertovec and R. Cohen, 1–22.

Oxford: University Press.

Weiner, M. 1982. International Migration and Development: Indians in the Persian Gulf. Population

and Development Review, 8, 1, 1-36.

-------------