224 68 202MB
English Pages 158 [85] Year 2011
Ethnography at the Frontier
Ugo E. M. Fabietti
Middle East, Social and Cultural Studies This series focuses on encounters, conflicts and transformations from the 15th century to the present and beyond. It invites works from various disciplines of the social sciences and humanities that consider the adoption, development, or reconceptualization of basic social and cultural phenomena in the Middle East, as broadly conceived, at a町time fror川叫fteenth century to the present, particularly in consideration of future implications. The series pnv1leges studies adopting the anthropological and historical approach in order to analyse the symbolic systems and social practices of the various societies of this area of the world.
Directors Wendy Shaw, University of Bern
Ethnography at the Frontier Space, Memory and Society in Southern Balochistan
Irene Ma筒,University of Lausanne
Editorial Committee Vincent Bisson, Researcher of political geography at the Observatoire des P叮S Arabes in Paris Mounia Bennani-Chraibi, Professor of political sciences, University of Lausanne Nancy Um,Associate professor in Art History, University of Binghamton Mauro Van Aken, Assistant professor. University of Milano-Bicocca
Etudes culturelles et sociales sur le Moyen Orient Cette collection vise a approfondir la reflexion sur les rencontres, les conflits et les transformat10ns qui ont eu lieu au M叮en-Orient a partir du XVeme siecle jusqu'au present et meme au-dela. Elle
souhaite recevoir des propositions venant des sciences sociales et humaines prenant en consideration le developpement ou la reelaboration de phenomenes sociaux et culturels fondamentaux dans l'aire du Moyen-Orient, con♀u dans son acception la plus large, a n’importe quel moment de !’histoire entre le XVeme siecle et l'age contemporain avec une attention particuliere a leurs futures !'adoption,
implications. Elle met l'accent sur l'approche socio-anthropologique et historique dans le but de produire des analyses des systemes symboliques et des pratiques sociales des differentes societes de cette region
du monde.
Directeurs Wendy Shaw, Universite de Bern Irene Maffi, Universite de Lausanne
Comite scientifique Vincent Bisson, Chercheur de geographie politique a l'Observatoire des Pays
Arabes a Paris
Mounia Bennani-Chrai:bi, Professeure de sciences politiques, Universite de Lausanne Nancy Um, Professeure associee d’histoire de l'art, Universite de Binghamton Mauro Van Aken, Professeur assistant d’anthropologie sociale, Universite de Milano- Bicocca
“@暴
二皂
PETER LANG Bern・ Berlin・ Bruxelles・ Frankfurt
PETERLAr叫G
am Main・ New York・
Oxford・ Wien
Bern・ Berlin・ Bruxelles・ Frankfurt
am Main・ New York・
Oxford・ Wien
Bibliographic information published by Die Die Deutsche Nationalbibliothek
lists this
Deutsche Nationalbibliothek
publication in the Deutsche National-
bibliografie; detailed bibliographic data is available
on the Internet
Table of Contents
at<h忧p://dnb.d-nb.de人
British Library Cataloguing-in-Publication Data:
A catalogue record for this book is available from The
British Library, Great Britain
Library of Congress Cataloging-in-Publication Data Fabietti,
Ugo E.M
Ethnography at the frontier space, memory and society in southern Balochistan I Ugo E.M.
Preface
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1.
Spaces and Places
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
11
2.
Time and Water
31
3.
Equality and Hierarchy
45
4.
History and Models
71
5.
Blood and Force
6.
Nation and Memory
.
.
VII
.
.
.
:
t?:d? ?::;.:?:?:i::?::?:i??,;?;:di叫t阳c阳elles
et sociales
Fabiettr.
su
Introduction s
汇:::叫::l;?;吕!工:z
: ;?;:?:;;?::,d:?:??:;;立i♂宫 :
: LE由nology--Pakistan--Balochistan.2. Balochistan (Pi汰istan)一E出nic relations. 3. Baloch tan (Pakis阳1)--Social lifi and ・
:???;?:?o;???叫 2011036663
.
.
93
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
107
.
Bibliography
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
131
Index.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
135
.
.
.
.
巳骂骂::?f?!;乙:过:胁。例adella如nt肌叫邮e storia in Bal.毗阳 Cover design:
Illustrations
Thomas Jaberg, Peter Lang AG
ISSN I 663-6384
ISBN 978-3-0343-0422-
I
。Peter Lang AG, International Academic Publishers, Bern 2011 Hochfeldstrasse 32, CH-30l2
Bern info@peterlang.com,www.peterlang.com,
www.peterlang.net
All rights reserved.
?↓?:;;?t1::i:ri?:??:;iγpr !!:.::!?::?::£?:??:;:;::汇::;::i:iE'.'!::Ii?::?Y:??::fr::?;?:?;,mg
Printed in Switzerland
in
Al
Preface
This book basically comes together as the result of an eight-year research, carried out in Balochistan from 1986 to 1994. More than fifteen years have passed since I last saw this region. Since then, the ups and downs affecting this Southern province of Pakistan by the shores of the Arabian Sea, have rarely reached Europe, the West and likely the outer world. The social and political background - an uneasy one, even at the
time毛t the research - got worse over the years. The strain from Paki- stani government on the one hand and the pressure from Islamic funda- mentalism on the other, concurred to exacerbate an already politically uncertain, socially conflicting and economically depressed situation. Balochi people, an Iranian-speaking group spread out over Pakistan, Iran and, to a lesser degree, Afghanistan, share some common traits with other so-called “nations without a state”,in the words of the Ital- ian sociologist Alberto Melucci. Claims for independence and unifica-
come up as an effect of the British colonial rule under whose ide- ology a nationalistic feeling had taken root among those who came in
tion,
contact with the colonial establishment, have always been frustrated, ever since the Great Partition in 1947. Sure enough, Balochi nationalis- tic movement did not originate as a form of imitation; it took a run-up thanks to ideas coming from the outside, which soon met a fertile pre- existing background within the independence claims of some poten二 tates and proto-states of the region, the most famous and important of
which being Kalat Khanate (1638-1955). While such independent political bodies were ruled by tr,iba_l elites or landlords, over the past forty years there has been a populariza- tion石f those same autonomist claims, which are now shared by a vast majority of the Baloch. Nowadays, any of the demanding for autonomy arising in the area, is vigorously fought back by the Pakistani government, while Islamic fundamentalism, particularly aggressive in this country, takes on the role of“moralizi吨”由people of Balochi
VIII
Ethnography
at the
Frontier
Sunni, yet little inclined to conceal political statements with the Ian- e ?f re?igion. Tal er to accoun powenng the region - Makran, the Southern end of Balochistan, bor- ders Iran to the East while Afghanistan lies a little further away to the North - and Ma训avellically speal斗?n refuse to make a parade of religious discourse for their political claims, o the very blind spot of their strat< is
Introduction
g1
・
m町ar
by n? means
book on geopoliti??\atl the呻ects of S↓ut rn Balocl i阳1 ultur圳in?f》on
Still, this is
anthropology
and口art
a
lar
nt reseat
put it:ζthe locus of s叫
W灿n山s?ace of several consecutive years I 1s?ent som
:;
:??\!l!???i?乙?i川江;=??F乙:;但也:??:?; :?1:?k…Europ叫to仪cl吨e tl叫hts and ir阳nation "on our res一,as
they would
say.
」出;;?ce?o!:??:?:::?P{:江:e:?t?:,!:??c?;“气川l':?ys
t;)?f.l:叩;:t?:t?fti?:ft'.??1;1;??£
?{?f
:
[???
:?::!
;;:〔iu工:.a::dt?:d,二::注二f;江士4飞: is for 出?::二乌:1:;e飞::工::飞??;k
?:?i:i??:供.
t;:?:口1! o:? I lS reg1?m straddling ?cf::r?I;?r .s a out ?;j!\?t{?t:}I'.:?f the Middle East and the Indian ?::口??:?:丁:
t
go deeper mto some叩cs clear up…e passages 1;.111 ifA???????:?;-?;:?叫出:;ti?;?i?;??f omeo山rs. Cl叩的山mselv…ave :!??;;“0阳s1on to
slightly different
way
f
h
I
I.
.
向of eth叫呐ic ?:?:i:l ??;二:??::::?? February 2011
be en arranged in a
so as to
make tl叫S
Ugo E.M. Fabietti
eastern part of southern Balochistan, Makran, which faces the is an arid region furrowed by semi-permanent rivers along which mankind has been settled for at least five thousand years. In oases supplied by water from rivers, and by subterranean chan- nels from the lower slopes of the nearby mountains, cereals and date palms have been cultivated from time immemorial. Seen from the air, these valleys, the desert and the cultivations, display a sharp contrast between the ashen-yellow of the arid lands and the bright greens of the oases that follow, often in wide bends, the courses of rivers. Gath- ered into one of these river valleys, the Kech, is the majority of the
The
Arabian Sea,
region’s
population.
Beyond the oases, desert. Much farther north, beyond the Central Makran Range and north-west towards Afghanistan, Balochistan is the land of large nomadic tribes. But in Makran there is no large-scale nomadism. Small groups of shepherds, families gathered for the period of a season, pasture their goats and their rare dromedaries in the moun』 tains, from which they get their name: Kuh-i-Baloch, the Baloch of the mountains. The local communities are in fact, in Makran, made up of settled farmers and have highly stratified characteristics in social, polit- ical and economic terms. This stratification has in recent years wit-
nessed signi岳cant changes, especially since Makran became a perma- nent part of the Pakistani state (the process, as I shall be recalling later, took place in two stages, in 1947 and in 1955). stratified society therefore, in which the relationships of depend- ency and servitude aiso extended to a substantial form of slavery. This institution, which included the use of agricultural and domestic labour, was fuelled bv trade between the east coasts of the African continent,
A
the Persian G?lf and the Arabian Sea ports, chiefly Gwadar1. Although officially abolished by the British with their entry into the region (at 1
Gwadar was until
1958 an
Omanite
enclave.
Ethnography at the Frontier
Introduction
the close of the 19th century), slavery nevertheless left its mark both in terms of social relationships (often transformed into domestic ser- vice) and in the actual appearance of some of its inhabitants, many of
headquarters of one of the most powerful “families" in the region. Its members were the descendants of a group which, having arrived three centuries earlier from the east (or so the stories said), had imposed its politico-military supremacy on the peasants of the valley. Having set- tied along the course of the Kech and in some of its adjacent valleys, the ruling families had divided control of the valley amongst them- selves by setting up a system of dominion founded, to a large extent, on the use of force. They kept their independence for a long time, before being absorbed by other political formations and, finally, by
2
whom have typically“African”somatic features.
After an initial rapid tour of this region (1986) I settled, the follow- ingyea巳in a village not more than thirty or so kilometres from Turbat, the administrative centre of the Kech District, on the route to the Ira? nian frontier less th乳n one hundred kilometres awav.
The practical and ethnographic reasons for this choice were justified. The village - with about two thousand inhabitants in all - c卢uld rea- sonably offer a context neither too broad nor too narrow in which to carry out“anthropological" research. Situated a few hundred metres from that route, the village formed an almost compulsory transit place ?or all those who, for a great ma町different reaso叭needed to travel trom east to west and vice versa, across that wide valley of Kech. Their purposes included work, to be sure, but also commuting with Iranian
Balochistan for not always confessable reasons, such as traffickin1! in P;t叫motorcycles, video-recorders and alcoholic drinks, and pe;haps also espionage (by Iran and by Pakistan itself). There were also pil- gri户s, heading for or on their way back from Turbat, the main oro;in- c叫town? near which (in a place called Tamba) was the zik ...
・
(an
y
I山mccong吨ation of rem。“hiite origins and targe时in r
;t
rea…by non-Balochi in吨ralists), and migrants to the已ulf countries, 阳mg or returning to their homes in viii咿S all over the surrounding area ..1?inall民阳e were t叫叭politicians e咿ged in electoral prop- ?gancta, g?vernment or pro-free Balochistan student orotesters_ wan- dering sufis,但d immigrants from other parts of Balo?hi;;;? ??' f;… .
much
?1ore distant regions. Despite all this bustle、“modernitv" did see血,m some ways, truly a long way off. Outside -the main adr tr巧?vetoγn, for a radius of hundreds of kilometres、there was no elec- tnc1t民0.nly precarious or hopeless communicatio?s, and a shorta1!e of drinkmg water. No hospital, very few schools (with woefull唁y总 ・
wded叫刊ry
the Anglo-Indian Empire. The lords of the place I had chosen had lived, up till the mid-2。th century二in a castle (kalat) whose remains still dominated the village. By that time irremediably in ruins, that castle nevertheless tended to fasci- nate anybody approaching the village from the east. Indeed the fascina- tion issuing from those ruins bathed the whole scene - the bright green of the palms, the contrast between that and the yellow-grey colour of the desert, the (not always) bright blue sky, the iron-coloured hills on the other side of the valley and, naturally, the crumbling tower of the castle itself- in a truly unique atmosphere. Constructed in ghel-o-sang, a mixture of clay, rubble and straw, and never restored during the pre- vious fifty years, that building seemed to be crumbling like a choco- late cake exposed to excessive heat. Every now and then I used to go up, incautiously, to its summit to observe the valley, the oasis and the
from above. To carry out research in the shade of those ruins would have given, I thought, a frame, and at the same time a particu- lar “tone”,to the whole ethnographic undertaking. In reality it was a chance encounter, and the interest shown by a number of locals in what I was planning to do, that prompted me to settle in that village. There I was able for many years to enjoy the cordial, generous, collaborative - and naturally “assiduous”一hospitality of a local family. About them, and about other figures whom I met over the years, I have spoken in
village
the various chanters of this book2. to
must conf已ss that, at least initially, the interests that urged me work in Makran were random, complex and not always the fruit
2
Their names have
l
ul斗:;垃::::飞i古::?t:?:??:
the吵th cen
3
I
,:;立!::;::二?;.c??了二uch of all
been changed.
Ethnography at the Frontier
Introduction
of any specific set research plan. The scientific mission which I was a part of, and from which in practical terms I detached myself, not with- out problems almost from the start3, had in 1984 launched a research survey close to the Iranian side of Balochistan, namely in the Strait of Hormuz area. Owing to the war with Iraq in which Iran was still engaged, it was not possible to follow up the project which consisted in the study of forms of settlement present in the region. The directors of the archaeological mission decided therefore to transfer the research to the other side of the border, into Pakistan.
the flow of Afghans (prevalently Pashtun) into Balochistan contrib- uted to create friction between the latter and the Baloch, who conse- quently started to become a minority in the population of Balochistan intervention in Afghanistan, from as a whole. Following the
4
了he
durmg
comp?ex and manifold
interests,
which gradually emerged
my subsequent stays, depended on the other hand in the first
instance upon the reality of the region:“a frontier" in the most clas- sic sense of the term, a frontier in the Lattimore style: 1) the absence of
power capable of keeping under control the centrifugal forces of particularism; 2) the discontinuous pressure of outside p;wers and their tendency to rule through intermediaries belonging to hegemonic local groups; 3) the existence of a close-knit web of relations and inter- ests between the local and the foreign powers. Makran has since antiquity been an area of transit between the Middle East and India, a kind of “corridor”in which highly dive si- fied forms of cultural and social life are encountered. as well as svstems of adaptation. 4 Durin the unabated Baloch nationalist autonomism had a local
redoubled while
3
4
This was the Italian Historical and Archaeological Mission in Makran, offi- cially supported by the Italian Ministry of Foreign Affairs. Of all the members of that mission I wish to recall here Ottavia Schmidt di Friedberg. A scholar of the Muslim world, Ottavia left us long before she could give the scientific ommunity what she would certainly have given had she lived long Agricultural communities were present here during the neolithic nd bronze ages. At the. time of the Achaemenids, Makran was part of the twentieth satrapy of the Persian empire, the one to which the Greeks gave the name Gedrosia. It was inhabited by the Maka, who appear in bas-reliefs on the royal pal e of Persepoli and whose likenesses can still be retraced on the faces of v Makranites. Alo叩}盯ecoasts of山Arabia山a live山d创叫ants of th:
K::{;e泣ic巳N叫川lex and的admira
5
OTAN
the 199。s, Balochistan, including its southern part, Makran, has today more than ever retained its “instability”and at the same time its posi- tion as a crucial piece on the intricate contemporary middle-Asian chessboard. It is a “frontier" region in the traditional sense of the term, a crossroads of conflicting strategic interests where rival powers vie for control, today as in the past, of the passage from central Asia to the Arabian Sea.
My interests were not however geo-political. For me Balochistan was not only a“frontier”area on account of the opposite interests concentrated on it. Rather, like most of the “world’s outskirts”,Balo- chistan, a land immensely rich in resources yet poor and thinly popu- la时,was part of出at va川niverse which, initi the logic and reproduction of capitalisn:i, was looking for a way of con- fronting as best it could the diktats of globaliza吵l.的la叫to use a slightly- overworked formula, might be described? as where tradition and modernity, the forces of the global market and those of tribal life came togethe?; where different types of logic, and a variety of “cul- tures" i且teracted to harbinger the social forms of a large part of the globe at the turn of the 21st century. Balochistan as a“frontier region”, astride two worlds in constant movement; a “frontier" of human sur- vival where every“local”carried with them splinters of a planetary
globality that i?posed, then as now, continual social, ethical, eco- nomic, identitv-related and other reformulations ... Indirectly, the book reflects this situation, not without repercus- sions even ?n the way of “conducting ethnography”. Living in the midst of the Baloch, right from the beginning I knew that mine had to be an“ethnography of the contemporary”. We are always, it is true, ethnographers of a heγe and a now. Recognising this does not however, in itself, entail falling short of that distancing, as Johannes Fabian (1983)
;
ves moreover from the name given by mahi khoran,“fish-eaters”,later translated by
the Persians to fish the Greeks with ichtyophages.
Makuran (assonant to mahi khoran) still
called,
by the
locals, at the
is
the
way in which the region was
beginning of the
XXth century.
often
Ethnography at the Frontier
6
called
it,
of anthropology
from the
“other”.
Carrying out ethnogra-
phy
in a given space and time (the heγe and the now) does not shelter us from the illusion of having to do with someone who is irremedia- bly distant and remote, in space as well as in time. To recognise that
we
are instead talking about realities caught in the whirl of globali- zation means to interrupt this illusory process. The societies that we once studied under山conviction that山y belonged to o山r flows of temporality appear by now definitively part and parcel of ouγhistory.
On this
“frontier”,and bearing in mind this idea of an ethnography of the contemporary, the distinction between “simple" societies and “complex" ones, which l叫lo吨dominated our stulies, was me_lt like
bound to ?now in the sun. The constr以ion of e山10g呻hic即resen-
tat10ns relating to supposed “simple”social and cultural realities, that had ultimately stiffened societies and cultures into typologies ;hile overshadowing their real social and cultural dynamics, no longer made sen?e. So let us by all means talk about an anthropology of '?omplex societies”,but ,.on -?・ondition that this term - complexity- is connoted /, }s a synonym t?r“mterconnection”,“globalization" and, I was saying,
contemporary”-
is午r邮reason_ that conduc叫an ethnography of contempo- tor me meant having to take into consideration the histori?al dimension. This is not a paradox. If we can say that there is no his- tory without a horizon of sense within which to retrace events. we can also say出矶山re can be no anthropology wi山ut a二 ;;te; ?;…id;;; 崎and即resent the tempor均discontinuous nature of斗tures we It
r.怠neity
study. Thus, in confronting the objects of med叫odically di_s ibu时in ti 丁o
my research (which I have
:epresent山ir temporally discontinuous character, and have tri。d to
brmg it out through
the representations山t I was与叫ying. For this …_I often talk in the book about historical and social “imaεina tion", the “power of models,,,‘‘memory”,representations of “ch口σe” and “social th.ought" These are in fact山expressions with whi?h I 』a飞:e ?ried to mc?rporate the events and temporal flow into the his- t?ncal s??se - and that of my interlocutors - whilst at the same time makmg sure my models would be continually compared with the
mme
Introduction
7
representations of history, of change, and with the forms of memory produced by the imagination of those whose lives I had set out to study.
At the time of my research (1986-1994), works on Balochistan in gen- eral, and on Makran in particular, were very few. Nor, it seems, have they multiplied in the past fifteen years. The majority of them dated from the colonial period and consisted for the most part in ethno- graphic accounts by the India Office and of travel reports by British political agents. Some North American anthropologists had however worked, in the years 196。一70, between Iranian and Pakistani Balo- chistan, dealing respectively with nomads and with settled peoples. Some time after returning from my first stay in Balochistan, I wrote to an anthropologist who had worked“teen years before in a place not very far from the village in which I had decided to do my research. I accompanied my letter with a charming photo of the place in ques- tion. My intention perhaps, with that letter and image, was to try to convey to an older and more expert colleague from across the Atlan-
my satisfaction with having picked a “terrain”which, since his stay
tic,
had seen no other researchers approach that region. A few weeks later I received an envelope from the United States, containing nothing but the photocopies of a short article (with dedication) by my colleague.1 The title of the article spoke, as they叫二 volurr Be Anthγopologist - evidently intended to echo the name of Kipling's卢 there,
celebrated short story, the unfortunate tale of two British adventur- ers. Spurred by a mixture of greed and a sense of cultural superior- ity (as well as by a desire for power vaguely heralding Conrad’s Kurz), they had set out to establish contact with a population in the north- west tip of the Anglo-Indian Empire, with intent to become their rul- ers. 5 Just as Kipling’s tale was that of a failed design on personal domin- ion, a metaphor in its turn for another failure, that of imperialism, my colleague’sarticle was likewise the story of a “failure": the one which, towards the end of 1968, the author and his wife, also an anthropologist, had encountered. In concise and lucid terms, my colleague explained the reasons that had prevented him and his wife from pursuing their 5
R. Kipling, The
Man Who Would Be King,
1888.
Ethnography
8
at
the Frontier
Introduction
them to leave that area in favour of another, less tur- bulent and “difficult”part of Balochistan. Published in 1979 in a jour- nal that today might be defined as “geopolitical”,the article stated its author’s disillusionment with the possibility of “doing research" in certain parts of the planet, and in particular in areas where the pres- ence of a“strange”figure like that of the anthropologist might 卢oke not so much the ghosts of a colonial past, but in which the anthropol- r ogist could become part of a complex local history, made up of tribal and factious strife, intrusions by regional potentates and groups of research, obliging
卫otables, myths and beliefs relating to the farangy (“westerners’''), as well as recurrent actions by major - near and dis阳1t - foreign po亏ers. In short, the Great Game. colleague, who shall remain anonymous, ? concluded the brief article with words to this effect:
My
As
anthropo_logists
we should not be too surprised if our well-i activities may be disliked by our subjects and by
and usually mnoc斗ous
political elites that issue visas and permits. ers may be simply depressing (as in the
and
a
I di斗
in
the
Even if our experience as re case of the field.;ork which mv wife
Balochis叫,we must moderate our perso叫relationιwith
recogmtio刀,仇the卢r叩lace, of the historical foγces that created ou; dilem-
mas [italics mme].6
When
arrived in Balochistan I too was gripped by the “historical had contributed, and still do, to shape that corner of south Asia. I mention those forces often in the book. Nevertheless I believ my :esearch in Balocl让i I
fo飞ces”that
albeit
tortuous and sometime?“depressing”way of
nitive relationship
with a social reality different to
entering into a cog
-
my own.
The subjects on which I concentrated my research are those dealt with more or less directly in the chapters of this book. Chapter I, Spaαsand P与叫examines山relation between local soci町叫spa叫the latter
bemg understood in a physical as well as a social
sense.
As 1 have tried
6
?of?::u:・::l・:::h:?r???:.;i:!:i:r:?ew?:?';! ?:::?e\?:?;l??::::v:!.:? ) nd anthropology years before, to devote himself to sculpture.
to show, for the
Baloch of Makran, space
9
is
a
complex network,
rang咀
ing from matters of social morphology (the distribution and nature of residential agglomerates) to forms of representation of history (local theories of the origins and position of communities in the area).
In Chapter II, Time and WateηI have expounded a complicated system of distribution of the water resources used in Makran, the func- tioning of which had not been understood by other authors. I have also however tried to demonstrate that this system, the cornerstone of agri- cultural activities, also operated for centuries as a system of reproduc- tion of economic and neighbourhood relations, a “web of water”which had for a long time kept the communities “on their feet". In Chapter III, Equality and Hierarchy, I discuss two opposite and complementary principles, on which the communities of southern Balochistan are founded. Equality and hierarchy are the two poles of a tension on the basis of which, in the course of time, and through the historical changes affecting the societies of that area, local communi- ties came into shape at the time of my researches. The chapter deals with the way in which the locals represented themselves, in the climate of accelerated change that characterised the life of those communi- ties, in relation to the socially shared values of equality and hierarchy. Chapter IV, Histoγy and Models, is the result of a twofold “con- cern”. On the one hand the desire to explain, in a more circumstanti- ated manner than some of my predecessors had done in the past, the appearance, from a certain point in time, of a form of political power that had been widespread until quite recently. On the other the entirely theoretical concern to discuss the applicability of models for the inter- pretation of the historico-social dynamic in contexts different to that of their original development. Chapter V, Blood and Force, is devoted to the analysis of a story about the passage from one form of political rule to another. It is a popular and fairly well-known story in Makran, in which I believe the elements of a local representation of power and society can be retraced, with reference to such local “cultural principles" as force,
blood, honour and the assimilation of foreigners.
Chapter VI, Nation and Memoη,focuses on the representation, gradually created in the course of time
by Balochi nationalism,
of the
10
Ethnography at the Frontier
Balochi people and nation. of historical
It is
also a discussion
on how the formation
memory can be constructed from local elements、but also
from hegemonic “discourses”that arose with globalisation’and, ? lier still, under colonial domination.
PA a
F3
ear-
《L
e ed
1n qAM
-G D-
--EBA
句AW
RL ρL Qd
L’espace saisi par l’imagination
ne peut rester l’espace
ind衍汪rent livre a la mesure et a la reflexion du geometre. Il est vecu, non pas dans sa positivite, mais avec toutes les partialit白de l’imagination. 1
Gaston Bachelard, La poetique de l'espace, 1957
The first thing that strikes one on observing the river Kech from hills that accompany it on its long journey to the Arabian Sea is
the the iron- its the banks, sharp contrast between the green vegetation along grey hillsides and the ashen whiteness of the valley that opens between these and the river. The villages dotted along this green snake that winds slowly south-east have the same colour as the valley, because their houses are built with the stones and clay of the desert soil. From the height of the hills this scene lends itself to questioning by a gaze that makes the contrast between inhabited and cultivated space on the one hand, and desert on the other, the main criterion of recognition. If we come down from the hills and approach the oasis and villages, the contrasts change. the sharpest is between the oasis and the vil- lage, between the tilled space and the inhabited land, whilst the valley and hills flanking it on either side disappear from the eye and the mind. The perspective onto space changes too: as we gradually move into the oasis new scenes present themselves, and the gestures of the people living inside that space grow visible. It also becomes possible to listen to the speech and stories of those who work, live and imagine there.
Now
1“Space
that has seized upon by the imagination cannot remain indifferent space subject to the measures and estimates of the surveyer. It has been lived in, not in its positivity, but with all the partiality of imagination”. Engl. Translation, The Poetics of Space, Beacon Press, Boston Ma优1994: xxxvi.
Ethnography at the Frontier
Spaces and Places
Human beings assign to space symbolic values and meanings that depend on the way they experi point of view and in terms of imagination. A space is as men and women 斗se it and think of it. To study this space the ethnographer must erase the isotropic space of the surveyor and replace it with the soace ren- dered significant by阳ial action and thought. 士hese brief c?n浏erations link up with山m町questions山t arose in the course of my research. They very soon thickened around the theme of the social perception of inhabited space and tilled space, and around the very particular“precedence”which the la阳・ seem; to me to have taken over the former as regards the development of the forms of settlement to be found in the region. This theme, and other “imoressions” 也就I began to develop during山research, to白ched on mat卢s such as the dispersion of patronymic groups across the territory and, naturally,
A consenting “prisoner" of “my family”,the recipient of their kindness
12
阜
the叫la阳
the _one hand and those wl山ich I formed for myself, topics meanwhile assumed a relevance that was for for example the impact
on ?he other. Other
me decisive. such as
of山system of ma叫ementof山till?i? a…
on山…od?ls of r创dence, or the role played by山juridical po叫卢nof
m
the constitution of those models. Finally,山political factor wo1;1e干 as decisive to the form which the occupation
of the space had assumed
in this region. These are themes without any immediately perceptible link oth红than that, of c户ur风of the fact that they are all ph n situated 1s?ace. In reality二as we shall see, they have many points of act and mterconnection. The important thing is not to forget that all the forms of叩・esentation of space
m
are,“lea二so山v se;? e, forms of “practical”life. By the term“P肌tic?l”I want to md1cate not only the “doing" that matches the material mana2"ement of resou附s, but also the continued use of symbols, represent-??i???-?nd 了o?:ed in
whereby human beings give a meaning to what ?hey are doing. Whatever ethnog叩?ers may claim to recount,由eir position does 甲‘free them from the mescapable reality of being astride two worlds. ιo 'recount" therefore how the Baloch of Makran - at least manv ideas
巧。二e
whom I have known person
means to recogmse that my own conception of spac晶 gain “reflexively involved" with that of my interlocutors.
u1eiγspace,
13
which never could have been repaid, I found it difficult to get them to realise that I did not need to be with them all the time. The obsessive care which they showed at the beginning towards their guest was cer- tainly mingled with a fear that something might have happened to me, as a result of my own imprudence or due to some inconsiderate person. But nothing ever happened that might have con岳rmed their fears. And so, as the weeks went by, their attention towards me slackened, and with their growing confidence and familiarity (aided by the decrease in their initial curiosity about me), they began to feel less obliged to be with me always. It was when I was able at last to enjoy this freedom that I began to go for long solitary walks in the village oasis, a place which until then I had always visited in the company of my hosts. A silence, broken only by the song of the occasional bird, enveloped those fields often flooded and surrounded by rows of date palms. Like bright green gigantic umbrellas, the palms cast their shadow across all the edible species of plants which, as archaeology shows, the time-hon- oured toil by the inhabitants of this region had succeeded in growing despite what is certainly not a clement climate and a by no means gen- erous soil. Every now and then, the lowing of a gok, the small humped ox characteristic of these parts, or the sharp thud of a branch cut off by a peasant’s axe, marked the presence of men at work. But encoun- ters were frequent. Small groups of ploughmen or hoers, isolated indi- viduals supervising the conditions and working of the canals, intent on opening or closing wooden sluices, on building or pulling down small dams of earth to allow or prevent, depending on the hour and on the owner’s rights, the flow of water into the fields. The oasis was an inhabited world. And time to time, in a tumult of visual contrasts in which the present underwent the irruptions of
memory and overlapped
that almost geometrically tidy landscape, the brightness of certain summer’s days it was possible to make out, from the top of a hill, against the background of a Tuscan landscape of abrupt colour con- trasts between the yellow of stubble fields and the green of cultiva- tions crops, the silvery grey of the olive trees and the dark stain of the wood, the men and animals yoked to carts, diminished by the
came images of
my adolescence when, in
?’41’
Ethnography at the Frontier
Spaces and Places
distance like the岳gurines of a model, or like characters in certain Sien- ese 14th-century paintings. In that play and contrast between the pre- sent and memory, it seemed to me that it was always the second t-hat
Having explored the oasis, I began after a while to venture, alone or in company, out into the streets of the village, arousing that“foreigner- effect" that consists in attracting the curiosity at times somewhat exhausting of one or two adults, in addition to that of swarms of chil-
14
gained something. But I had to change
and
in that contrast, the
my mind, because in that play
memory itself of things known
changes, is transformed and leads to fresh conclusions. Travellers are oft n sus-
pended between a present composed of things known and a past that anchors them to the known, and every contrast between past and pre sent leads them back to what is known, but onlv to know it a2:ain in a different light. I had a clear sensation of that o二e day. While马re were ?ittin乎 near a house in the village, a puff of wind brought me a famil- iar odour, an aroma that immediately grew clear and distinct in my mind: basil. Looking arour I noticed, unde时1e wall next to which had squatted, a small plant similar to that of Mediterranean basil. But it was darker, with thicker and more lanceolate leaves. I oulled one off and, snif fi吨it, recognised at once what was for me that unmis- takable aroma. None of my friends however knew what the plant was, and a町possibility of its being u时to flavour food was r由江out. Fo; me, bound by history and memory to the Mediterranean, the smell of basil could conjure l not only the familiarity of certain foods for山 p 子ooking of which this plant is abundantly used, more or less accord- mg to regions. It was also an odour that reminded me of stories and images hidden by time and which, reawakened by that potent aroma, w?re waft?d mo…ividly than ever back into my mind. This vividness or memories was also accompanied almost at o e y how that te毗r lit plan l刚at anchored me to a s??ce and to a of my own had sufficed to establish a gap between me and “them”, ・1
m了Balochi friends. An awareness also of how for them other aromas. other impressions and other memories could have performed the role. which now that solitary little shrub was performinσfor me: taking back,”s, t干山kn今wn, but also of re-presentin;邮kn wn i? light of relative experience, expected always to be compared with what ?s 口?t known, and to derive from it the lesson of an inevi协le "oartial-
旷on川art.Be巾乌is not邮山ultimate sense of a川吵ology and ?ught not
this
perhaps always to be the outcome of the choice of
movmg towards the elsewhere?
dren.
I
15
later to encounter other curiosities and other swarms that regret the attentions of those Baloch, which were on the
was
made me
whole always sobre and discreet. It was during these walks, not always solitary, along the canals of the oasis and in the dusty village lanes that I tried gradually to give shape to what had until then remained a vague impression: the extraor- dinary contrast that reigned between the order and cleanliness of the oasis and the disorder, neglect and abandon that characterised the inhabited space. Whilst in the latter you could have truly found eve- rything, in the oasis you could not have seen as much as a piece of paper larger than a postage stamp. In the oasis everything was“meas- ured”,neat and tidy, exploited down to the millimetre. The dykes that wound their way through the fields cultivated with cereals, pod and green vegetables, medicinal herbs and fruit trees, were kept clean and tidy with a painstaking and constant, almost obsessive care. The more delicate crops (some cereals, certain pod vegetables) were located in the shade of the big palms which were in their turn freed, with methodi- cal daily obstinacy, of dried-up branches and bits of bark. Here every- thing was artificial but nothing was ornamental. That the oasis was a garden, as well as a place in which date palms, rice, broad beans, toma- toes, medicinal herbs and onions (plus some thirty or so other edible plants) could be understood from the attitude displayed by the locals towards this place. That attitude could not be reduced simply to a“pro- ductive relation”,but also contemplated a “ludic”type. On Friday hol- iday afternoons it was not rare to come across families enjoying the shade and cool away from the dusty village streets; clusters of ado- lescents, all strictly male, or of young bachelors meeting in the shade of a mango or of a zeitun tree (which contrary to the Arabic root of the name does not here stand for the olive, but for the guyava tree); or adult men gathering to chat and drink tea before the sunset prayer, a rite, as I was very soon able to realise, many Baloch often and, more than often, readily skipped.
Ethnography at the Frontier
16
At the time of my researches the oasis was still, despite the rise of activities other than agriculture, the place in which were concen- trated the economic interests of the community and the place where social life
became most open and
intense. Despite the appearance of (transport, building, commerce, public despite the migrant labour (in Pakistan as
new activities and new trades
contraband etc.), Gulf region), the purchase of water rights often remained the favoured investment for those who had accumulated a bit of monev. No building stood within the oasis, with the exception of an old small mosque which seemed almost intended to bear witness, wedged by a few dozen metres into the thickness of the vegetation, to something offices,
in the
more than a human imprint in those places. The habitation was therefore distinctly separate from the oasis. At first, this struck me as odd, since I could find no reason for this read- iness to sacrifice living in a calm and shady place instead of under the hot sun and dust. This perplexity was very quickly dispelled however,
when I began to relate the fertility of its land with the water resources
available. I realised in fact that the people did everything in their power to reserve for their crops even the tiniest pocket of land.“Objective”
reasons (limited water resources and poor soil fertility) thus explained this sharp separation between an inhabited space - the village - and
one一the
oasis. But what were the con叫uences of this state on the way the locals conceived their relationship with the two space_s? How could a work place become - as a true co?tradiction in terms tor our mentality - a place in which to relax, meet, talk and exchange con岳dences far from the dust and garbage of the village, far, that is, from their abited space? Our “culture" (urban and othe;wise) has for at least centuries by now distinguished between worked space and inhabi时space, and whilst it does not always include in d
a tilled
of affairs
the space of ludus, of play, it certainly does not situate the latter in the space of labor, from which indeed it tends to distinguish it through the (fairly recent) representation“of free time" - free of course. from work. If“doi吨e山1ography”means, as someor叫as said,“山i卢g to put oneself in other people’s shoes" (without moreover ever cmite succeed- ing), to account for this apparent contradiction one need;d to ask o self what these two spaces represented for the Baloch. Having reached
Spaces and Places
this
for
point
them
17
we may in fact wonder: what for them is an oas比,and what
is
a。illage?
Until now I have called“oasis" what the Baloch call mu岭,a term taken from the Arabic legal vocabulary meaning “property’:“rule”. Amulk is in fact, juridically speaking, the complex of the cultivated area on which various individuals exercise their property rights. The limits of a mulk are not rigorously de岳ned in terms of space, and the name of a specific mulk is that of the “nearest" village. Maitag-e-mulk is thus the name of the “oasis" of the village in which I stayed longest. The notion of mulk does not however indicate a place consisting only of cultivated fields and palm trees. The mulk is always criss-crossed by a grid of canals that carry water to the crops according to a complicated mechanism of divided water resources founded on four different “systems”,adopted in as many periods of the year. 2 These canals irrigate the fields of the mulk with water from the kariz and from the kaur-jo. The former are subterranean channels that bring to the mulk water collected below the hills a few kilometres away; whereas the latter are open-sky channels that deviate river water to the crops. We therefore have two fundamen- tal elements of the worked space: the mulk and the canals. On the side of the inhabited space lies the village. Maitag is only one of the different ways in which the Baloch call a not too large settlement, in contrast to shahr (city), a term imported from the Persian. For vil- lage, the term bazaar is also used, this too being of Persian derivation, which however also maintains its meaning of “market气 After a more attentive consideration of the use of the term maitag, we discover that it is employed to indicate, as well as village, the nomad encampment. This may sound odd, because among the peoples with a strong tradition of mobility, the “village”and the “encampment”also remain sharply distinct. As among the Beduin of Arabia who contrast qaria, village, with manzil, encampment. is this not so among the Baloch of Makran? In order to answer this question it is perhaps neces- sary to step back in time, to the period prior to the arrival of the Brit- ish in the region, hence to the end of the nineteenth century.
Why
2
See the chapter “Time and Water".
Ethnography at the Frontier
Spaces and Places
The political restlessness of these places is certainly no recent matter. At the time of my research - and things have certainly not changed since then, but have if anything grown more complicated- this in;ta- bility was translated for the most part into manifestations of local
beginning of this book, had by then for many years been in ruins. But as recently as the end of the Second World War it had housed, in addi- tion to the family of the kakim, at least fifty persons including employ- ees, clients and “attendants”,a not too distant memory of the merce- naries with whom these lords surrounded themselves to make sure the laws of the Khan of Kalat and their own were obeyed. Historical accounts tell us that, resulting from the political instabil- ity and frequent clashes between factions associated with the diverse families of hakim, the villages were agglomerations of housing made of woody shrub stems and woven mat. To this day some homes, usu- ally the poorest, are made of this material. In the village only the kalat, the castle, represented the “landmark" of the settlement. Everything else was in fact more or less fluid, mobile and unstable, like the polit-
18
nationalism and opposition to the central Pakistani government3. Nev- ertheless up until the end of the nineteenth cent?ry, the turbulent political climate was principally due to rivalries among the chieftains (hakim) of the various villages. Even though formally subordinate to the Khan of Kalat, these local rulers were forever at loggerheads to γiden their spheres of influence and impose their rule over varyingly large or small portions of the Kech valley. They lived in their castles, as did the ex-lords of the village in which I was conducting my research. With _one of them, fairly advanced in years, I had a few brief occa- sional encounters. For his age and by local standards (he told me he was nearly eighty), he was a man of some“culture”. But he had some- thing of an absent-minded, rather dreamy air, though it was hard to
if.th与was due to his no lo耶r young ?ears or ra山r to his having remamed, as certam among my friends maintained,“out of time”. His fell ow villagers described him to me as a man who had stoooed to mar- ryi吨a woman of low ra此,so that no one had ever wanted to marry 中e two daughters he l叫ad had by h优Of hir hrst encounter when, after he had asked me where I came from. and when I had answered that my coun町was Italy, his fa臼lit -up: and with a laugh that may or may not have been one of self-satisfaction with his knowledge, or rather a mockery of myself, he exclaimed:“ah! Ah! ... Italy... massolini!…”Initially disconcerted by that reaction, I realis?d lat红白at the old hakim had said nothing ;pecial. He and I were living m a sort of “warped”historical perspective. For that old Baloch, the story of his relations with Italy had stopped at the time in which the British were at war with the Axis To tell the truth, the British dominion first. and the annexation of 山吨ion to the Pakis阳1 state la民r, l叫transformed the descendants of that class of form叫ominators叫diligent阳collecto二:in shadow of their former selves. Their kalat, the castle mentioned at the tell
.
ical
climate of the valley.
political instability seemed to me to be a long-lasting feature of the “production of locality" typical of this region. It is therefore not to be ruled out that the vernacular language even in the presence of agricultural populations, may have received an instability of the kind
Thus
form of a conservation of those terms that seem rather to belong nomadic tradition. This possibility seems to be corroborated by a number of elements. Firstly, by a sort of “nomadic memory" which characterises the peoples of this region. Aside from the way in which things really went, the Baloch are inclined to reconstruct their past in terms of migrations and nomadism. It is this idea of mobility that dom- inates their view of their own history, and hence of their actual origins. Furthermore, the use of the term bazaaγas an alternative to maitag seems somehow to mirror the same imprint. Although in the past it may
in the
to a
have been introduced into the region, we can be sure that bazaaγis not a Balochi term, but as I have said, Persian. As a place of trade, it must not be identified immediately with a permanent settlement. Just as there existed in pre-Islamic Arabia places occupied solely by markets (for example in the neighbourhood of Mecca) and not inhabited settlements, and just as in Algeria and Morocco there existed until very recent times uninhabited places in which periodical markets were installed, it is pos- ?ible出矶tl
3
See the final chapter of this book.
19
but temporary, just like their villages gathered around the kalat.
20
Ethnography
at the
Frontier
Spaces and Places
But the kalats were not the only“fixed”elements of a highly “mobile” sp?ce. There were in fact, as I have said, the mulk, workplaces in which subsistences were produced; and there were also the kaγiz and the kaur-jo, the canals that carried, then as today, water from the hills and river to the cultivated lands. It is around these elements that the lives of populations revolved. So it was the places of tilled space that cons titu以山real landmarks in d of this reg10n.
Every
now and then I tried to anchor my suppositions and the cer-
tainties of
my interlocutors
to something “practical”that might con- a…orco斗fute them. Thus I would ask anyone passing with :n air of gomg out mto or returning from work in the fields, towards which place they were heading or from where出ey had c?me. Unfailingly, men, women and chindren indicated their destination or orovenance bv the name kariz or kauγ-jo, whose wa阳irrigated the la?ds cultivated by them:“man Sadbad-a rahwin'・γ‘I am going to Sadbad”(the name of the kariz);“man Sadbad-a pidak户,“I have come from Sadbad”. In no case, not even in that where the fields were situated in the mulk of another village, did he or she use (as we probably would)出e name of hat village or of that particular mulk. The P?sition of such pre-eminence which the canals had in organ- ising the image which the locals had of their experienced space did not 叫lain the reason why it was山names of山kariz that p;evailed
?叫?f t巳e oasis or of山village. My answ盯出1er as山followin吁 g he irngat1on networ』c was an element of space onto which was focused what might be called山commu叫r's‘‘s?cial a阳tention".… :
Each of these canals carries a quantity of water that is redistributed
among an often very large number (even hundreds) of individuals, who ??e
h
owners of water shares that can also be v町U叫ual. Furthermore owners of?en live i? different villages Tl山is is not only because a
se
:;;乙:t?7;:;二: ???:1:1士;;;;;;:;立;?::!!?!?:
of in山heri阳叭buying and selling and
marria;;i;?;ti:;;口??-(;i;):i;
mely fragmented. Thekaγiz and the kaur-jo, with their sub-channels, ditches and fur- ther branches, are thus something more than just an irrigation network
21
enveloping the fields of oases. When plans to build a kariz a_re drawn up, the necessary funds are collected among all those prepared to invest in the undertaking. The ownership of the resulting water resources is then divided on the basis of the amount of money paid. The canal itself is maintained by dividing the expenses proportionally among those claiming rights to the water acquired in that way. If we were to say what a kariz or a kaur-jo resembles we might say, by resorting to an analogy, that they are vaguely reminiscent of a condominium. From this “condominium”,many individuals are excluded. The rea- sons for that exclusion may vary greatly: the lack of sufficient funds to be able to share in the construction and maintenance of the canal; the sale of water sharing rights; con岳scation for legal reasons; exclu- sion from an inheritance, etc. It is in the differential possession of water shares that the social strati岳cation was clearly manifested, at one time more than today. Shares owned bv an individual can be rented or sub-rented. Con- sequently, many owners are not directly concerned with their fields, whereas several tenants, day workers, servants, and generally persons without ownership rights, till the land and manage its irrigation. Open- ing and closing the sluices at a given time of day or night to let water out of a canal into the neighbour’s fields rather than one’s own; supervising the canal’s capacity level; registering possible dyke breakages; clearing the canal of detritus that may obstruct the regular flow of its water; and to check - above all - that nobody deviates water to his own岳elds before the established moment. All these operations involve the farm-
complex of services and counter-services. Management of the network is carried out day and night (by men only), and is such that those who oossess or cultivate on behalf of others allotments adjoining one anoth?r have to assume an attitude of mutual responsi- bility. This does not however rule out conflict within a macro-system regulated by expiries, rights, duties, services and counter-services. I have supplied an analytical picture of the operating logic of this system in the following chapter “Time and Water”,where I have also tried to illustrate the importance which this system had, and in part still had at the time of my research, on the process of social integration at a level reaching beyond the village. ers in a
irrigation
22
Ethnography at the Frontier
In the course of my inspections of the oases and channels that irri- gated them, I soon realised that in Makran the idea of landownershio was expressed through signs different to those familiar to me: land surfaces that can be calculated and often recognised by the presence of fences, walls, hedges, nets and the like. In Makran the “borders” between the areas of these owners could not by any means be defined on the basis of lines (albeit imaginary) of boundaries, but rather on the basis of the water shares possessed or managed. In Makran in fact - at least in the river valleys - there is no true form of agricultural prop- erty exercised on land as such; here the regime of ownership concerns water or, more precisely, the time taken for the water of a生iven canal to flow. The borders of a property con叫uently correspond?t。the land which the owner or the tenant can irrigate with the water at their dis- posal. I had a demonstration of this fact whenever I asked someone to show me the extent of their property within a mulk. This person did not show me fields or fences, nor did he trace lines on the 品2:round or on a piece of paper; still less did he mention surface measur?ments. If he felt like it and had the time, he would take me to see a canal or, in contrary cases, he would list for me the water runnin2: time fractions of which he was the owner in one or more canals. Ti。民able to see the canal was however decisive, because not all the canals have the same capacity, therefore possession of a certain water share can be more or !ess impor阳tdep叫ingonwhe山r the ca叫i川hich it is possessed has a greater or smaller capacity. The result of all this was a- fine prac- tical demonstration of the fact that if all human societies oossess an idea of limits and boundaries, it is not on the basis of山s乌me social experience that they develop an idea of that kind. And that the physi- cal borders, even where we would expect them to exist on the basis of our experience, may be missing without thereby undermining the idea of the limit and of the boundarv. The centrality of the wo巾d叭ce therefore appeared to be山prod- uct of several factors not only of a historical heritage, of a“nomadic” culture that continues, in its vernacular. to asc山e le;s?r imoortance to the permanent charac优r of a settlemen;; of the mobility of ;opulations as a consequence of the political instability typical of the period prior to the arrival of the British; and finally, the “precipitate" of a structure
Spaces and Places of social interactions
23
shaped by the system of production and by the around it, having
complex of services and counter-services that develop the mulk and its canals as its hub. This complex of relations continually constructed
around the mulk
and the irrigation network is however also, as I have said, the result of another factor: the type residence pattern adopted by the newly formed families.
The inhabited space, the
village,
does not“contain”a public space
as
we understand the term. This fact is to be borne in mind because in the West public space is that created in our cities in a continuous dialectic of thrusts and counter-thrusts (often conflictual) between public and private interests. In this part of Balochistan on the other hand the vil- lages consist of one-room houses built against a perimeter wall, at times with several dozen metres per side. What “remains outside" between one enclosure and another is “public" space. This model seems to be developing, but on a wider scale and through the use of diverse mate- rials, that of the period in which enclosures and homes built previously by means of a mud paste, straw and stones (ghel-o-sang), and later in cement, were originally made with matting and shrub stems. There is absolutely no“planning”in the arrangement of these houses, which often makes it necessary, to reach a place only a few dozen metres away as the crow flies, to make long diversions. The inhabited part of the village, therefore, contains nothing “public”,
mosques and, in the village that I am talking about, the bazaaηin a corner of which there was also a“cafe" - here called some- what fancifully utel. It was frequented mostly by a small number of cus- t户mers, some of them habitues, others passers-by, and the occasional bizarre or dubious character. Once I met a fellow there who, dressed to the nines with a woollen shalvar kamiz, astrakan hat and sheepskin waistcoat and gun slung over his shoulder, was sitting at 35 degrees in ?ther than the
Accompanied by a squad of armed men with unsavoury faces, was told he was from another village - went around m that manner because “he feared a vendetta" (bir) - they added - for something that had happened years earlier and elsewhere and about which I was never able to find out anything at all. the shade. 丁ha?
character一I
E no
24
’h
舍EL’
σhu
rA 而d
PA
’n
VJ
而d
’h 舍’ι,
舍’L’
ρ』
EA rA
on
・・i φ・・
户』
Spaces and Places
rA
The transvestites (bagha) on tour from Karachi occasionally also appear in these parts, whilst in the provincial capital Turbat,- some thirty kilometres awa民they can be encountered much more frequently. When I tried to find out more about local sexual tastes, my friends laughed loudly and appeared rather evasive. But on the whole I under- stood山t the presence of山se individuals (and consequently their activity) was not considered all that scandalous. When I speak of the absence of public space, I refer therefore to a physical space and not to a social one understood as a sohere or arena of comparison and discussion. The public, as a form of r;lation and dis- course, may certainly have a place as a support, but it may also have none.“Public" means here what“is shown" or, in other ;ords. what is not kept “concealed”,for example, by the secrecy of domestic- walls. The house (log) is the hub of family life. In it live for the most past T二tended" families (likewise very significantly called log). These fam- 1hes are composed in general of a married couple (polygamy, though admitted in theory by virtue of Islamic law, is somewhat rare among the Baloch), their sons/daughters, the spo山es of these and their off- spring, according to a fluid residential pattern whereby husbands can in may cases reach their wife’s familv. ・
The women spend almost
the whole of their lives at home. That mainly to women of “high-ranking”families, although the ?ense of honour (izzat), to which this practice is constantly linked, it 1s a sen!iment generally widespread at all social levels. The iarge enclo- sures that surround domestic space serve precisely to allow ; greater freed可m of movement to the women of the family (who however fre- quently go out to visit neighbours or“relations" in other villaεes). The neighbourhood here assumes a great importance and a syste? ?f pre- cise rules underlies the contacts which adults on the one hand. and 户e-adolescents on the other, can en阳山n with persons of the卢ppo- site sex. The neighbourhood does not imply close family ties, but only a syster口of services and counter-services鸣arding the卢ost ordinary aspects of daily life. Often, in the walls separating two logs, narrow passages are opened, the upper part of which is lowered just enough to prevent the eyes of a neighbour falling continuously on what is going on in the nextdoor enclosure. applies
25
The home is therefore a space protecting domestic intimacy, closed on the outside and accessible only to relations or to wati people, indi- viduals allowed - if the conditions exist (by the fact of being relatives or servants, or neighbours, but of a certain type - woman, adoles- cent, etc.) to enter. To guests and even to the male friends of the men in the house, this is not permitted. These persons are in fact received or lodged, as in
my own case,
in a special part of the building, the
(a Hindi word). This area, when it exists, consists of a room, a latrine, and a small courtyard built outside the main enclosure and whose doors and windows do not communicate with the log proper,
betek
but give onto the exterior. My friends insisted on the idea of the log being a protective space, not only of family intimacy but of the “honour" (izzat) of the women in the family. The “honour complex”,so highly developed in the dis- course of Mediterranean and Middle Eastern anthropology, makes the house the place in which the woman is at once free and a prisoner, a place ill-suited to men (who are looked upon as nothing short of a nui- sance when they meddle with daily domestic affairs), whose sphere of action should instead be that of a public confrontation. This separa- tion of female and male spaces (which specialised anthropology has transformed, by superimposing it on the question of honouηinto what Appadurai called a“gate-keeper concept”) is in reality a constant leit- motif of the rustic and peasant world in general, a characteristic of
which anthropologists, coming mostly from middle class and urban backgrounds, have never experienced or of which they have lost the memory. The taciturn and introvert Alam, who was not lacking how- ever in a considerable sense of humour, was often the target of his wife Yasmi的吗优She reproacl叫him for spending his days in the enclo- sure of the log and not going to the village centre to open his medici- nal shop, pretentiously called the parmasi (f is pronounced p). It was not Alam’s idleness that infuriated his wife, but the fact of his loaf-
ing in a space habitually subject to the control
and organisation of the
women of the log. On these occasions Alam would deal with Yasmin’s …nes by going off grumbli吨to his shop, a microscopic room facing the
main street of the village. He would open it reluctantly, as if aware day was not the right one for business. On suchlike days, in
that the
26
Spaces and Places
Ethnography at the Frontier
effect, customers were very rare, because their passage increased or dwindled according to market days, in Maitag or in the villages not too far off. This was the real reason why on certain days Alam did not w气nt to go to work. There were in fact days on which Alam, without bemg prompted by anybody, went off punctually to his shop. At nine 0’clock sharp. He knew, on those days, that he would receive cus- tomers. A coming and going of men, women, children and old people lasted for a good part of the morning. Pills against malaria, one or two analge气ics, 气it干min tablets with a “mysterious" air, and a few equally mysterious m1ections which Alam gave with nonchalance in a small and dusty interior whose dubious septic cleanliness aroused in me a vague feeling of horror. On the days of “forced opening" of the shop on the other hand, Alam would sit around for hours in a foregone and resigned idleness, rocking on the old metal chair one of the two ( or three that I saw in those parts in the space of eight years), leafing through a pape马humming something or happy, as when some friend of his would arnve and, seeing the shop open, would drop in for a chat.
Alam enjoyed the visits of a suf i (su卢) friend of l山is with γhom he would readily c阳・se-abou gious matters. In fact Alam cordially detested all the mullahs in the ??ighbourhood and further, made fun of their “religion”,which for In particular
him was nothing but politics
...
But the space, for my Balochi friends and “informants”was also a see- nario onto which to project a pas吨e of hi
Next
to the social strata into which until a few decades before arrival the Mal王ranite population was divided (hakim, baloch, hizmatka厅,there existed a sizeable number of more or less numer-
my
ous groups of descent, the zats. The name of each zat
・
阳兰the pla叫
eιO every zat belong individuals in relation
By and large, these are endogamic groups and
to patrilinear de
become increasingly so, However, except for山
le产st. ideal!y, 干s the social hierarchy rises. asc叩t1ve cntenon of patrilineal descent,山ir individuals behave
4
On the origin and meaning of this partition see chapter
3.
in
27
compliance with the principle of bilateral descent. Both their status and their goods are in fact inherited both paternally and maternally, although the latter is less important as regards the inheritance of goods二whilst it may instead be decisive in determining the status of an individual. The breadth of these groupings is very varied, ranging from a few thousand individuals in certain cases, to a few hundred in others. Not all the members of a given zat have an identical status. Some “families" may be rich, others poor. In fact the zats do not collectively manage a heritage of resources held to be the property of the whole group, as water might be for example. So their nature differs sharply from that
groups of traditional ethnographic literature. Many moreover may be said to belong to a zat nominally, espe- cially when a group is particularly numerous and thus very much dif- ferentiated internally in social, economic and other terms. of the corporate
individuals
Spatially speaking, the zats are heavily scattered. Situations of dis-
persion of groups similar to the
one we have here are not rare in other
contexts. Jacques Berque thus interpreted the local explanation given in the past of the dispersal of a number of eponymous North African
The dispersion of groups beari吨the name of their supposed ancestor was seen by the locals themselves as a process of continual fission involving a “projection”。f segments in space. Berque maintained that these were pure conjectures, in that“not only does that e_xplanation not conceal - as he wrote - anything historical at all, but does not even make any reference to history" (1953: 265). Berque pro- posed therefore that that dispersion be considered, aside from the rea- sons that might have caused it, as a phenomenon endowed with “a value of classification”,as “a play of verbal mutations" to which the locals would resort to claim ties with one group rather than another. The shift made by Berque from the “concrete" to the abstract or, as might be said, “fr?m the fact to the sign气could also apply in our own case. But it has to be_ recognised that here, unlike Berque's North Africa, the histori- cal dimension - albeit highly conjectural - is brought into play often and readily by the locals. The Makranites have in fact two theories about the fragmentation of zdts across the territory. The first, which was particularly fashionable groups.
common
Spaces and Places
Ethnography at the Frontier
28 at the
time of
my research (and I shall say why later on5), is
founded
on the migrations of the Baloch during the past centuries. According to that theory, the various zats would appear to have scattered while advancing from north-west eastwards, leaving minor groups behind, families and individuals who in some cases would have later consti- tuted the settled nuclei and growth of the zats in different points of the territory. It is a theory which to account for a morphological real- ity combines the image of Baloch society as one in movement (indeed a nomadic society) with the (agricultural) metaphor of dissemination. This interpretation combines the migratory factor with demographic g?owth and entrusts the criterion of patrilinear descent with the task of conserving the historical identity of these groups. The second local theory of the dispersal of groups across the terri- tory refers back instead to the lack of precise rules as regards the res- idential pattern of the new couples. This theory puts the historical
dimension definitely on a secondary level, and emphasizes the dimen- sion which we might define as、n川ural" ina…uch as it helps to
explain the f吨mented presence of zats on the territory. It in fact ;efers to the ambilocal nature of residence and considers it to be the main cause of dispersal of patronymic groups.
In this region, new couples have in fact a tendency to settle with the one or other of the two spouses, on the basis of considera-
f干mily_ of
may vary above all in relation to contin2:ent factors. even if for山most part conn ed with the ma吨emen? of山bride's pos- tlons that
sible property.
In this second type of explanation too、I was faced with a“partial 山th”,seeing that it neglected to take in;o consideration the卢sons that P:ompted individuals to repeat the
ambilocal pattern of residence. of groups across山阳itorv is not in fact由result ot unique and exclusive causes as might be ?igration or the model of
T?e d叩ersion
residence. The same local theory of dispersion that dates the later back to the model of residence is in its turn the “precipitate”of a series of combined factors such as the regime of landed rroperty, the exploita- ion of agricultural resources, inheritance, matrimonial strategies and 5
See the last chapter of the book,“Nation and Identity".
29
dynamics. It only remains for me, therefore, to consider in question of residence. I shall refer here only to the form which residence assumes in a context of adaptation made up of per- manent settlements and of irrigated agriculture. These are in fact the forms of occupation of space and of exploitation of resources respec- tively, that have in the long term shaped the material and mental land- political
this light the
scape of the region’s inhabitants. Where this form of adaptation dominates, the landed regime is based, as I have said, on the individual ownership of water resources. These resources, I have also pointed out, are the object of continual transac-
both in the form of sales and purchases, and in that of marriage compensations (jaiz) and also, in the form of legacies. Some research『 ers maintained, even several years ago, that the extreme“individuali- sation" of landed property among the Baloch of Makran may account for the small number of endogamic unions on the one hand and for the absence, on the other, of corporate groups proper (Pastner, P. and C. 1972). According to these ethnographers, the individualisation of rights to ownership of water, together with the limited availability of that ele- ment, would urge men to engage in working the land of other owners on the basis of varying agreements. This situation would be responsible account, by and large, according to this line of interpretation, for “dif- fused and fragmented patterns of both economic links and marriage alli- ances" (Pastner, P. and C. 1972: 133), and likewise for the inhibition of any tendency “towards a pattern of marriages based on close kin endog- amy geared to the maintenance of uni岳ed family estates in land" (idem). This perspective, in itself correct, shows however only one aspect of the problem, namely the reason why men tend not to marry women too ‘close”to them, but rather to join with women who, in many cases, do 吧。t even belong to their own group of descent (zat). But what then are the consequences of this fact in terms of the constitution of domestic units? What is of interest here is in fact to see the other side of the issue: the fact that these men tend to marry women who possess, or whose families posse风吨hts of access to re?ources situa时in irriga时areas different to those of their husband. This means that, in many cases, men marry women from other villages. Which brings back into play the dynamic of residence as a factor of dispersion of zats across the territory. tions,
Ethnography at the Frontier
30
The men tend
to move into the family of the wife resident in other with a frequency proportionate to the importance of the worn- an’s present, or future, ownership of water resources on the mulk of the place in which she lives. This pattern of residence is connected with
2
villages
Time and Water
the fact that in Makran the women see their rights to inheritance rec- ognised, both in the paternal and in the maternal line, and to a much greater extent than in other regions of Balochistan itself. Sometimes, due to the combination of the inheritance by both parents, certain. women find themselves owning even conspicuous properties, the man- age.ment (hence了public”) of which however is precluded from them and, consequently, entrusted to their husbands (or brothers prior t 0
The technology of crop production did not stand alone. By a process offus ion with the social structure and dominant forms ofpolitical control, it reacted, changed, and responded to pressures generated by the whole society. Kirti
N. Chaudhuri, Asia before Europe,
199。.
山time of削r marr
to the wife’s family’s home, and therefore of ten in a villa2"e diff eren to their own, can more easily control her interests and c nsequently
the neγly established family. The log in which I was a guest for many years 1s a typical example. Alam and Aqil were the husbands of two sisters whose father was a fairly important owner in the Maitag mulk. It was therefore to山wives’home出at they had moved from ;'m咿S different to that in which they now lived.
When I
first arrived in Makran, I was immediately struck by its curi- ously complicated irrigation system, by the canals and ditches that
transformed its oases into lush green places with bubbling streams running through them. Most of all, I was intrigued by the still more complicated way in which the water was shared among owners or ten- ants, by the manner in which it was inherited, sold and bought. This complexity became apparent to me very gradually and with effort. Exhausting interviews with owners, with my hosts and their friends, had brought me into a world made up of hours, half-hours and quarter hours (all things that referred to the running time of water possessed by individual owners) so complex that I felt ?lmost sure I would never have sorted the matter out. From time to time I used to go with Dosten, his brother Hamal (Alam’s sons) and their friends to bathe in one of these canals. The water was extraordinarily cool and clean, and it only took a few immer- sions to come away regenerated. None of those boys knew what swim- mmg meant. One cannot swim in the canals, as they are too shallow. One can dip into the water at some points where, next to a weir or ditch,
山w阳forrr
immersed, tickled by the microscopic岳shes that live in these Thus, among complicated岳gures relating to the capacity of 丁hese canals and my regenerative baths, the matter of understanding Just how that system worked remained a mystery. There was no doubt at all about its functioning, but if in a society there is something that pletely
canals.
E no
32
舍-tw
’h
σhu
rA 44
PA
tn
vd
丙d
+・・
φL
h
AL
EA rA
on
・・且
φ・・
ρ』
Time and Water
rA
functions - and that irrigation system did indeed undoubtedly “func- tion”- one had to try to explain how this something could function ...
Field 1959; Pastner
that
terized
As sometimes happens during岳eldwork, the image of something we are, or are not looking for, takes shape almost suddenly i口 our mind, as with a puzzle whose pieces, after lying scattered across a table for days, suddenly fit together in a flash. Thus it was for me in case of the logic of the functioning of that irrigation system, which other previous, partial, imprecise, or decidedly misguided descriptions th二
had made it even more dif :ficult to comprehend. I believe on some nights my mind must have exhausted just about all my resources to get to the bottom of the problem (and also those of my friends in trying to understand what it was I was trying to find out ... Until - and I remember still as if it were yesterday the faces of mγhosts a_nd surveys
that
).
illurr
betek of Alam and Aqil - I don’t know which of the usual partici- Rants ?n our讪nog呵hic sessions, pronounced a word: hep问. To that lmgenng memory that had never abandoned me、and which at times see?ed to have gained the upper hand over山present, did not that word pronounced - heptag - sound similar to the Greek epta learnt at high school? It was that“seven" - or rather、“week”一that had at last
阳时thi
Everything that followed is, I would dare to say, the effect pro duced by that “seven”on my capacity to reason. And as a result, I was able almost at one blow (though it took me a few days to get the maths right) to fit together the jigsaw pieces, which consisted of time a口dγater.
Af阳山t, I set out山results of my research on the water-
shanng system
my
in use at the time of stay in Makran. Before going into details, it is however necessary to supply, or to recall, some of the general characteristics of the agriculture in this region.
The study
of time representations among the Baloch of Makran has never been the scope of my research. Rather, my interest in local “uses” of time derived from the fact that I realized, from the very beginning of
my于l?work,
that as
I
told, the
avail灿descri伊ions of山local
;?t…a叫system was large!川阳仙ctory. The logic upon wh讪 s
system
is
founded - I realized - had been totally misunderstood,
or overlooked,
by
a
number
of previous ethnographers
and Pastner McClure
33
(BDG
1906;
1972).
As we have seen in chapter one, Central Makran is an area charac- by the presence of large oases, set along the rivers crossing the region along the east-west axis, and forming a virtually uninterrupted In these areas cultivation of date palms, cere- and vegetables is the main source of income for the settled popula- tion. Here, the agriculture is made possible by the presence of irriga- tion systems consisting of open channels (kaur-jo) bringing water from the rivers (when water is present), and of underground channels (kariz) that bring water from the distant foothills. I also insist in reminding the readers that in the Kech area, land is not an asset in itself. The value of land depends on how much water is available for its irrigation. In fact, agricultural property in Makran is 口ot calculated in terms of land surface, but in “quantity of water". It is the possession of water that quali岳es a person’s agricultural property. To the time of my stay, kauγ-jo and kariz were built by squads of skilled workers (kannat) who in many cases seemed to pass down their strip of cultivated areas. als
craft from father to son. These squads worked with “t「aditional”,one might say Neolithic, techniques if it were not for the use of iron-tipped
No
and shovels. energy other than human (digging) and animal (conveyance of materials and rubble) was used in this labour: donkeys and camels. Motor-driven transport was beginning to replace animals, hut the essence of the work was still done with very ancient techniques. The construction of these canals, for example a kariz one mile long, required the employment of labour equal to thirty persons for a whole ye可(to simplify, I shall from now on refer only to the kariz). A certain number of associated persons participated in :financing the building of a kariz, supplying the necessary money to cover the costs of materials, implements and labour. In proportion to his invest- ?1ent, each individual had the right to a fraction of the hydro capac- 1?Y of each kariz, after subtraction, when the work was complete, of tne part reserved by tradition to the person who supervised and guar- anteed the proper distribution of rights among the owners (essadar), whilst also serving as general overseer to maintenance. That of the sareshta, as the men who perform these tasks are called in this area, ?icks
34
Ethnography
at the
very often a hereditary appointment and a position of prestige and At the time of my research, such an appointment was usu- ally held by a high』ranking individual. It is not a duty like another、as can be imagined from the wealth that goes with the二ppointment ?f a sareshta, derived from the possession of a by no means negligible frac- tion of the water carried by a kariz. Add to this the fact that an indi- vidual may be sareshta for more than a kariz. which enables him to dispose of resources which in their turn bring -him money to buy, as a plain sharer, water quotas in other canals. is
privilege.
All these variables that I have pointed up here combine to make the system extremely difficult to grasp in its totality, logic and articulation,
both岛r an out.s?de observer as the ethnog即her may be, and for川1di- vidual involved that system. We can leave out the ethnographer who, by the nature of his or her role, is “external”to the system by definition. But in the case of the locals, it has to be said that an essadar usuallv has a somewhat“limited" vision of山system itself, since in the ma{o向 of cases he knoγs only those owners who possess water shares “tempo rally close" to h1?. As I have said, only the “supervisors" of canals, the sareshta, knows detail how the water-shares of a canal are divided. However the logic of how the subsystems function, as I have described
m
m
阳“叭does not
system. Although they agree on (after endless discussions) the general ic:?r? of the functional logic of the sys臼m outlined by me, the ?a;?shtas fsee the system as formed by a succession of quotas assiirned within a given time span: one or two weeks. They do not represent?the system i I1 the abstract sense - model 二:ally, this is not because the Baloch are without logico-formal capac’ 1t1es, and therefore in飞apable of supplying an abstract “explanation" of how the system functions. In fact, the modelling is not a problem that ncerns them, but one that concerns the ethnographer, who not only attempts to understand, but also tries to explain.
!he mone叫zation
of the local economy favoured transactions and mve_stments agriculture, but clearly only a small part of the pop- lation could be engaged successfully in this type of activity. That
m
Time and Water
Frontier
35
small part was composed of people from the well-to-do former hakim and baloch categories of people. Most of the population was excluded
from it, and I had the impression that a steadily increasing number of people were joining the already landless, that is to say without rights to water. And this was happening despite the fact that many ex-emi- grants had invested a part or all of their savings in the purchase of water shares. Although on the whole there seemed to have been a decrease in the number of big landowners, probably as the consequence of vari- ous factors, the principal among which was the low rentability of land and a corresponding tendency to invest in other sectors ( construc- tions, transport, commerce, etc.). Furthermore, money economy may have been an incentive to sell a small or a radically shrunken property, owing to the impossibility of getting su凶cient resources out of it in kind to maintain a family. Smaller was the property, more accentuated was its fragmentation, due to the effect of heredity. Heredity, in fact, caused a fragmentation of rights to water which, in Makran, seemed to be happening much more quickly than elsewhere in Balochistan due to the fact that in
Makran women inherit from their parents.
have already noticed that many of the inhabitants of the oases had an extremely detailed empirical knowledge of the ways in which the
I
water had to be divided into units and sub-units. They knew when it should be let out of a sub-canal so that it would change direction
moment
by the rights to time fractions. They were from the speed of the water, if somebody was takmg away water from his neighbour without the right to do so. T.hey al川new, with the aid of山sun and of water-operated vessel- chronometers (by then increasingly replaced by watches), how to cal- at
the
established
cap?ble of discovering,
culate exactly the division of
water into its sub-units. Despite this empirical ability, the oasis farmers were never capable 干f offering a complete and coherent picture of the systems of distribu- tion of time fractions among those entitled to them, that is, they did not k斗ow all the holders of rights to the water of a particular kariz or ?aur-;o. And indeed why should山y have? The sareshta alone knew every smallest detail how the rights to the water supplied by a given kariz were divided up. Sar吵ta :;ere usually able to reconstruct, by
m
36
Time and Water
Ethnography at the Frontier
purely mnemonic means (even if they kept unofficial registers in their houses), the whole picture of rights concerning a particular kariz. This is all中e 1:1-ore川 watermakaγiz were subject to progressive and frequent fractioning as a result of legacies and of sale and purchase contrac低It was com?on that three or four (and even more) dozens of individuals shared the water of a kariz and, what was more difficult to remember. their ri2:hts
can be borrowed and/or ren时,whilethewa阳咄are of a? individu could be exchanged by this one, with the water-share of another in
different
moment of the day.
a
The time has now come to observe the working of the irrigation system in detail and according to the logic on which it is founded. In the literature it is possible to find a certain number of descrip- tions of such a system. But, as I told, I believe that such descriptions re misleading as they are founded upon an incomplete knowledge of the whole process.
For出e sa牛e of clearness, we m1风first of all, keep in rr of general pomts The capacity of a canal is calculated in hangam. The hanrza ??sic standard measure which is divided into .
do
n
・
subm毗ples. Th?ha咿n correspond to a fixed amount of water. It is the volume of
carried by山chann叫elin
d
?ust consider t_wo hatψms: one from dawn rrom sunset to dawn.
to sunset, a二d- ar…her
A ch斗nnel, a kariz, is not described as“having" two ha俨gams. U
karzzes are desc削as having 14 hat r;? e amount of wa叫・ !hat makes two .
ha咿ms
a day, 14
ek (hφtag). T巨1s is the basic point that has always be derstood by the prev1ou b Th b
1
amount
is
always the
Four patterns, or sub-systems, are in use in Kech valley: bunjui, tallaη hapt, and bamprosh. Every owner has the right to dispose of his water according to the pattern in use at that moment. For the sake of clarity I shall assume that the water of a channel is owned by fourteen individuals, that is, one owneγfoγeach hangam. Of course owners could be less in number, even only one, but as a matter of fact, they are always much more than fourteen, as many of them have rights on submultiples of hangam as a consequence of water sales, herit- age and “marriage compensations" (jaiz), which are a constant element in the social and economic life of this region. As already mentioned, in a kariz of fourteen hangams there are usually more than thirty owners, and can easily be up to fif ty.1
As the aim of the different patterns is to enable owners to make the water according to the general productive conditions of the different seasons, people will take (but in the winter) their water at different times of the day (24 hours).
best use of their
Water-rights are calculated on the basis of the investment of each individual in the construction of the channel (in our example, there- fore, all the essadars paid the same quota - except for the sareshta - since they had the right to one hangam each), and the order of access
among owners is established by drawing lots at the beginning of every ye牛 So, let us con圳er山first and most simple of tl时our patterns or sub-systems in use in southern Balochistan, the bunjui. to water-shares
.
han?ams
be 叫咱ct,叫e basisso?♂;:::ionin;:二;二:;二dmust of seven days. .
of water available, the unit of measure hangam, a measure, I insist, of water-running time). effective
37
consid-
Water-shares are distributed among the essadars during the seven 出ys of山week. A巾V的week山distribution sta仲a?a;n ac,、
mg to山pa7ti 1叫a阳n in use at that season "t 」f h; ;!倪志tem ..
z工址:?;?:?h乙:i??飞:?!::??;飞?:!0!;:E?: ::t::
1
The smallest sub multiple of hangam (12 hours) an individual can, in theory, own, is tas (thirty minutes). 24 tas is r hangam. Tas (English cup, French tasse, Italian tazza) is also the name of the special copper vessel which, laid down in a basin full of water, sinks (in thirty minutes) due to a little hole at its bottom. As thetas sinks, one tas (yek tas) is lasted. One tas and half (45 minutes) is half abba (nim abba) and one abba (yek abba), i.e. rh”’is half tassu (nim tassu). Therefore one tassu, is hours. Two tassu (6 hours) correspond to half hangam 3 (nim hangam). Four tassu is one hangam (12 hours).
E tn no
38
+EL’
σhu
-n
rAmd
pi
VJ
a
舍L
舍L
’h
户lv
RArA
on
・・i
舍ELW
Time and Water
ρL 俨A
Bunjui means“fixed”. This pattern is adopted in winter, when crops to be irrigated very frequently. Supposing that the shareholders are fourteen, as much as the number of the hangams, each will have the right to water once a week. Giving a number to each shareholder, (r, 2, 3, 4, ... ,14) the pattern will be as
Saturday
do not need
NIGHT: Sunday
in the figure.
Sunday (yek shambe)
owner 1 owner 2
DAY NIGHT
owner 3 owner 4
DAY NIGHT
owner owner
Days: 7
owners: 14
etc.
Friday (jumah)
TOTAL
(1 (1
(1 (1
Friday
13
(1
14
(1
hangam) hangam)
hangams: 14
means “free”. As the beginning of the hot season approaches (usually in March)’. bunjui is replaced by the tallar pattern. Tallar is
Tallar
adopted when
rice 1s
sown and needs
to be irrigated as frequently as possible, especially in daylight. All the fourteen owners will be allowed t? take wa阳from the break of day till sun叫seven of them from dawn to noon, and the other seven from noon to sunset. This ?h干t th与 daylight hangam is divided among the fourteen essadars, that is mto tourteen parts (seven from dawn to noon: seven from noon to sunset). At night, owners will share two by tw二the har叼m of each night, as during the night hours rice叫uires less wate;6than in川th ?ot hou of tl叫ay. E
uninterr hangam only once in the week, always by night. Therefore, at the end of the week each owner will have taken his full share of water, that is 1 hangam (r/r4 x 7 = 7/r4 = rh, plus rh, total r
TOTAL
I
(tot.
I
(1
owns 3 and4
(1
hangam) hangam)
hangam) hangam)
DAY:
Morning: owns.
(1
and 14
(1
NIGHT:
owns
DAYS:
owners: 14
7
1-7;
owns 8-14 13
hangam) hangam)
hangams: 14
Haptmeans“seven". The hapt pattern is applied when rice crops need less water than in the sown period. The hangam of daylight is divided into seven parts, three and half from the daybreak to noon and three a.nd half from noon to sunset. In this way the first seven essadars take the water one day, the other seven the following one (that is on alternate
At山end of山week, every about half a ha咿m. say about because due to the logic of the pattern, those owners who take water the first day of the week, take water also on the last day of
own川巧ot
days).
I
week; while those who get their share the second day of the w'eek, seven days will receive water three daγs out of seven. In fact, in tl时apt system the turn will be completed ?ot in a week, but in a fort- night. This is due to the fact that the same seven owners who “open” the weekly turn, also “close" it. Each 户wner will get his other half hangam by night. Each night the owners have access to water two by two and once a week (as in the the
after
tallar system).
Saturday
DAY:
NIGHT:
ctays
hangam).
(tot.
Morning: owns. 1-7; Aftn.: owns: 8-14
Aftn.:
hangam) hangam) hangam) hangam)
Morning: owns. 1-7; Aftn.: owns: 8-14 owns I and 2
etc.
,
DAY NIGHT
DAY:
NIGHT:
Every day the water is accessible only to two owners: one during daylight that is, from daybreak till sunset, and one in the ni豆ht time, that is from sunset to dawn. Let us begin with Saturday, the first day of the week for the locals. So the bunjui works as follo;s: Saturday (shambe)
DAY:
39
Sunday
DAY:
NIGHT:
owners: 1-7 owners: 1 and
(1
2
owners: 8-14 owners: 3 and 4
(1
(1
(1
hangam) hangam) hangam) hangam)
40
Ethnography Monday
DAY:
NIGHT:
owners: owners:
at
r- 5
Time and Water
the Frontier
7
and 6
(r (r
hangam) hangam)
labour,
41
only one pattern was in use, mostly the bunjui, which
is
also
the easiest to deal with.
etc.
have outlined these patterns in an abstract and highly schematized way. In reality, as I pointed out, the essadars never own exactly the same share of water. Furthermore, in some areas, there is such an abun- dance of water that people divide the water of a kariz into two m句or sub-channels, obtaining 14 hangams each. In these cases, the kariz is said to have 28 hangams (it took me a lot of time to understand why some kaγiz "had" 14 hangams while others “had" 28). We must also consider that within each pattern in use at a given period of the year, essadars are free to make individual agreements about exchanges (or rent) of water shares according to their personal needs of the moment. The different patterns have been conceived with I
Friday
TOTAL
DAY:
NIGHT
owners: 1-7 owners: 13 and 14
Days: 7
owners: 14
(r (1
hangam) hangam)
hangam:
14
Barn prosh means“exchange”. This sub-system is introduced start- ing ?r?m September, when rice has grown tall and the hot season declmm?. According to this pattern day-water (1 hangan water (1 hangam) are divided among owners. Three essadars will take 7 their share during the day, while three will take it in the night. One o t of seven will get some water in the daylight and so night. The next 24 hours is the turn of the other seven essadars. This means that each owner gets 2/7 hangam at every turn (the 2 hangams are divided among owners), that is one day out of two, which corre- 7 sponds to about 1 hangam in a week. I say about because here also, as _the hapt pa阳n, two weeks are necessary to complete the u叫as ?? me same seven owners open and close each weekly turn. Saturday
DAY:
NIGHT: Sunday
DAY:
NIGHT:
owners 1-3 and 4 (partial) owners 4 (partial) and 卜7 owners 8-ro and 11 (partial) owners II (partial) and 12-14
(1 (1
(1 (1
hangam) hangam)
aim of enabling the owners to make the best use of their water according to the general productive conditions of each season (climate, kind of crops raised); but within such frameworks, people are abso- the
and tend to establish agreements on a personal basis. Finally, notwithstanding the geographical location of the region - that is, in a tropical arid area - the duration of days and nights varies according to the different seasons of the vear, so that the distribution of water- shares needs to be continuously adjusted. It will be noted, for example, that during the hottest periods, the pat- lutely free
terns of distribution adopted favour an increased frequency of access to water resources, whilst that frequency decreases during the cooler
hangam) hangam)
s??sons.
Now, if we consider that many“real" essadar (hence not “fic-
etc.
t1t1ous”like
Friday
than one hangam (the average ownership quota considered model), the system’s logic seems even more “functional".
TOTAL
DAY:
NIGHT:
owners 1-3 and 4 (partial) owners 4 (partial) and 5_7
(1
Days: 7
owners: 14
hangam:
(1
A sub-system
hangam) hangam)
bunjui, for example, would mean that a part of the essadars would have, throughout the period in which the subsystem is active, the_ disadvantage of irrigating their fields during the day (when evapo- like the
14
fou明terns on wl灿he whole water sharing system r飞:口:/;?1 some areas, where rice crops and other cultures which need
a lot of
water had been abandoned due to low rents or lack
those of our model) possess water quotas of often less by us as our
of
others would be forced to irrigate them by should always be remembered that the water is not calculated ?n volumetric units, but in units of time). Although the bunjui system lS材op时in wint叽when the temperature diffe:ence between night
ration
is
干ight. (it
rapider), whilst the
Time and Water
Ethnography at the Frontier
42
and day is not so high as in other seasons, disparities are in any case c_reated between those benefiting from water used during the d乳y and those using it at night. These “inequalities" therefore are compensated by reversing, during the subsequent week, the order of access to water shares or, as the Baloch say,“by repeating the last hangam”. Let us go back then for a moment to the bunjui system pattern. The essadar with an odd number ( 1小5, ... ) get water in daytime, whilst those with even num?ers between sunset and dawn of the next day. The turn e阶with essa1aγ?o. 14 which, according to the logic of the system, ought to receive the next water share after one week and again during the night. By “repeating the last hangam”on the other hand, he will have access to water the following morning, so that the turn between even and odd essadars is reversed: for one week the even numbers will take water in the daytime, and the odd numbers during the night. The “repetition of
the la川a1ψm”is a functional device serving mainly as a r:iore equal utilisation of resources by those possessing smaller shares than others. It must also be stressed that this flexibility of the system has other technical advantages. Essadars in possession of shares during what are
for
fo了
them咀convenie旷hours can exchange them
for o?h:r owners
whom that exchange may be advantageous. Or, thanks to the “rep-
hangam" of the week previous to the incoming one ?mal! owners have the possibility of not always working at nigh? (say et1tion of the last
m wmter) ?r al:'ays in daytime (in summer). If in fact access to water in the day飞1me m summer an advantage in productive terms (in that is
cro?s re户户e water during the hottest hours, hence when they most need it), It IS ne飞ertheless a disadvantage in terms of work nves ??ca?se t?e e?vironmen叫conditions are cer阳1ly not the b创(tem- peramres the sum?er here can easily reach 45-5。°??C). Furthermore, if the fact of possessmg only daytime water shares is an advantage in .h严??ops do not have to be left without water during山hottest hours, ・
m
Is
true, con丫ersely, that this fact
l
n
IS
may be a disadvantage
seeing
hour cycle in which
evap。”
the penod of the twenty-four quickest.
归t山s is
adopt the best solutions in terms of productive yield, but also curb precisely the disadvantages that may arise from those same
users to to
solutions.
However, from this modelling we can also understand that the logic underpinning these systems allows the various owners to be abso- lutely free and to make individual agreements in order to derive fur- ther advantages from the use of respective shares. If, for example, a certain essadar possesses a water share greater than the necessities of the crops he wishes to grow, he can lease a part of it to another owner who instead has a greater need, or who intends to augment his shares with other supplementary ones. As it is conceived, the system is highly synchγonous. Its functioning involves (and requires) a succession of turns which must be respected and which require, at some moments of the day and night, the owners (or some other trusted person) to be present. We could be authorised to suppose that the time-cycles upon which such an important activ- ity as agriculture is founded, may shape the social life of these popula- tions. We could be tempted to argue on the analogy of our own expe- rience and to conclude that a highly “synchronised" labour system has immediate and compulsive effects on the organization of individ- ual and community life.
We must indeed
stress the role of the system in shaping social rela- demands the respect of rules that must be obeyed in order to carry on activities vital to the community and to avoid conflict. But it is also true that people act in such a way that the strictness of the system ca? be adapted to the particular and practical needs of individuals. Let us omit for a while the abstract image of the system presented .
t1ons, as
it
and think of a number of essadars three or four times more than which I introduced to make thine:s clearer; that those individuals own, as a ma阳of fact. smaller wa;er-sha…and of dif- ferent durations in time:肌abba.…u and let us also think that very often ma町of them o?n wa?er-share二 in different channels and above,
the fourteenth
.
了v?r
乙:2己?江;t::t:二?r?e:J:,:;:工;::;i?:s???:::;工?;二击’
43
:
diffe陀n
川onsequence of such considerations, the svstem loses the cha ac of a叫stract and well-balanced model and finds itself at the ω阶 r叫of a complex series of in阳创s,ne斗,options and choices which
Ethnography
44
at the
Frontier
3
oblige the people involved to manipulate the fixity of the turns estab- lished by drawing lots at the beginning of every year. Every kind of transaction, every kind of arrangement is allowed. By exchanging their own turns, individuals reformulate the framework of the system as a
Equality and Hierarchy
whole. Turns and water-shares can be combined or divided according to the agreements enabling individuals to manage their time activities,
masse des individus q山la compose悦..... mais, avant par i’idee qu'elle se fait d’elle-m在me. Et sans doute, ii aγγive q饨’elle hesite sur la maniere dont elle doit se
economic,
social, political, domestic and religious spheres, avoid- ing the necessity to submit themselves to the “times”imposed by the logic of the system which is rigid and synchγonous in its very nature. In this way the system is transformed from a synchγonous, to an asyn- chronous system. Such a-synchrony of the individuals regarding a fixed time sequence, does not undermine the logic upon which the system is built, namely on the division of water flowing conceived of as a computable and frac- tionizable entity. From this point of view the a-synchronous charac- ter of the system (and of one individual with respect to another) does not question the synchronous basis of the system itself, which refers to a quantitative and measurable conception of time. It is important to stress that the “social”effects of the utilization of water are in no way limited to the activities that turn on its exploitation and distribu- tion. Essadars looking after their own quotas of water often go from one canal to another, even to ones located in distant villa2:e-s. Thus the irrigated fields become places of meeting, exchange, dialogue and negotiation between individuals who do not see each other regularly or who do not know each other at all and this allows them to consolidate their ties or establish new ones. While on holiday evenings the canals and their intersections turn into gathering places for groups of men or whole families, who are able to岳nd a moment of tranquility and respite from the noise and dust of the village by some rustling brook. Using a metaphor, we could say that here the community appears to b气enmeshed in a “web of water”sustained by a framework of tempo- ral sequences. In this case the liquid element is no longer the basis of a view of society as being made up of groups organized-on the principle of “genealogical distance" represented by the zats, but the mainstay of a form of social life that hinges on the desire to impart “自uidity”to a - system which is otherwise rigid in its conception.
in
Car Uγie societeγi'est pas simplement coγistituee par la
-
tout,
co nee。oi:γ. Elle se seηt
en des sens di・。erge时s. eclatent, or.以ieu none悦γe
tiγaillee
Mais ces coηflits, qua叫ils
i’ideal et laγea litιmais e时γe idea以diffeγents, eηtγe celui d汗1ieγet celui d,’aujourd’hui, entγe celui qui a pour lui i’autoγite de la tγadition et celui q以est seulement eη
voie de
de饥γii:γ.1
Emile Durkheim, Les formes elementaires de la vie religieuse, 1912
When I talked with my friends and hosts about Balochistan, about the way
of life of its people and their cherished hopes of a better life for themselves and for the younger generations, the clash between a her- itage of hierarchy, dominion, privilege and exclusion, and a present made up of novelties, the desire for emancipated participation - but also of resistances and new forms of discrimination - became partic-
The past seemed in fact to have dragged an obstacle into formed by inequalities that were hard to overcome but which nevertheless somehow had to be abolished sooner or later. It seemed to me then that two fundamental ideas always prevailed in 斗larly acute.
the present, a weight
1“For a society is not simply constituted by the mass of individuals who compose it .... but above all by the idea it fashions of itself. It is undoubtedly true that it hesitates over the manner in which it ought to conceive itself; it feels itself drawn in divergent directions. But these conflicts which break forth are not between the ideal and reality, but between two different ideals, that of yesterday and that of today, that which has the authority of tradition and that is still evolving”,Engl. Trans. Emile Durkheim, The Elementary Forms or Religious L二舟,Oxford University Press, 2008: 317-318.
46
F』 舍L ’且
no aPAtnvd a σhu
rA
φL
舍EL・
In
A--
EA rA
on
--A +・・
Equality and Hierarchy
AL rA
what my interlocutors were telling me. To these two ideas they referred constantly in the construction of their lines of argument: equality and hierarchy. One of the most significant aspects of Makran’s history, as I have remarked, is its highly stratified society. That, however, does not altogether preclude an ideal of equality among adult male individu- als. This ideal of equality bears little resemblance to the way we might 出ink of it in山W悦For amo吨山Baloch, it remains anchored to the conception of personal honour (izzat), which in its turn is corre- lated to a habitus constructed around the岳gure of the adult man as the defender and guardian of his family and their interests. Equality and hierarchy were not of course concepts which the Baloch employed as such. Rather, they are concepts through which an observer like me “summed up" and translated Baloch notions, moods and moti- vations expressed symbolically and acted upon pragmatically. Equality and hierarchy can be seen as two “ideals”on which Baloch society was founded. At the time of my stay among them these two ideals were“solicited”,so to speak, by factors creating new social, eco- nomic, and political dynamics, and they were the refe;en friends 阳d to make such change comprehensible. From山:ir contex… tual enhancement also stemmed, however, their ambiguity, opacity, and problematical nature. To analyze the contextual use of “equality”and “hierarchy”entails attem?ting to understand and describe Baloch society in a・ way that not only portrays concepts as people intentionally used them, but por- t:.ays them in such a way as to reveal whether or not they involved a d吨uising effect (Gellner 1973). This is not to attribute t卢these ideas the nature of “false consciousness”,but rather to suggest that notions, 6户。ods and .?1oti?ations which such ideas sum up, were not always transparent" to their users, and that this opacity was an effect of the way in which they were for the most part unconsciously used. _Fredrick Barth, in a classical study, showed how, in Nord阳1 Balo- ch1stan, Pu斗un ethnic identity was maintained a; individuals repro duced practices associated with a particular system of values (Barth 1969b). Si…ilarly, Baloch identity springs from respect for certain types of be巳av1our and from adhesion to values that correspond to that behaviour. Although Balochistan’s encapsulation by Pakistan fuelled
47
Balochi nationalist feelings, these continue to perceive their distinc- tion from other groups as founded on practices and values invariably
grouped under the expression razm-e-Baloch, or“Balochi custom". Central among the types of behaviour and values included in this cat- egory are those associated with the idea of “honour”,or the respect- ability of the individual, a sentiment expressed, as told, by the term izzat. Flowing into the concept of izzat are other notions such as those of individual autonomy, hospitality, and the protection of the women in O肘’s family. These notions correspond respectively to the ideals of the freedom to make decisions, generosity, and the conservation of male respectability, all of which, in a circular way, flow back into the concept of izzat. This complex of ideals and compliance with a particular code of behaviour constitute the foundation of the egalitarian ideal in Baloch society. Adult men recognized themselves as“equal" inasmuch as they
were assumed to share values that they perceived they can achieve through their individual behaviour. This representation of equality
on
a definition of the individual which leaves aside the social each single person. This allowed the idea of equality to per- vade all strata of society which, as described below was, and I’m sure still is, sharply hierarchical. At this level, then, the idea of equality “embodied" that of hierarchy. It also dominated it in that it was capa- ble of causing all members of society to recognize themselves through
rested
status of
common ideals.
The complex of ideals and related behaviour making up the con- formation of what might be called a “moral community”. Here “moral community" means a group of individuals who, aside from specific distinctions of wealth and status, and inde- pendent of ties of obligation and dependence, consciously and vol- U斗tarily abide by a standard that puts them all on the same plane. In this sense, the moral community is formed around a public model of values and behaviour that produces the mutual recognition of individ- cept of izzat underlies the
uals as“equals”.
This could be ascertained when sometimes, half seriously half jok- mgly, dire与tly or“obliquely”,I asked interlocutors to define them- selves (their“identity”,that is “who they were"). Despite the great .
my
48
Ethnography at the Frontier
Equality and Hierarchy
differences that distinguished individuals in terms of status and‘ they themselves often maintain, of origin, the Baloch alway v ally admitted that they felt equal on the basis of their co?mon Ian- 骂uage and their shared system of values. In other words, Baloch iden- t1ty sprung, as everywhere springs, from the recognition of c values and ideal behaviour, and this recognition was founded on the code (language) by means of which these“indicators" were reoroduced. ,
49
popu- term, and the hizmatkar, or“dependents." This division of the defined is only one latio口 of Balochistan into different strata of which as baloch does not stultify the appropriateness of the definition of local culture as being “Balochi"
on the whole. In the words of Stephen a于d
These valu三s are r eprodu阳cedp鸣matically in pract 10ur compllant to norms transmitted between members of the same ?eneration and between those of different generations. But with an important point to be made: the “reasons of the world" shaoe individ-
term‘Balo ch’is used to designate a dis- tinct cultural status vis-a-vis other cultural groups in Pakistan, such as the Pushtuns, the Brahui, etc., baloch .... refers to a specific social status within a stratified social system. That is, the cultural status of Baloch is shared by all Balochi speakers and by all social strata, but the social status of baloch refers to the broad middle layer of the soci- ety- pastoral nomads in the hinterland and independent agricultural-
飞al_att
ists
their turn contribute to the ideal (and practical) reshaping of the world.
While the moral community and Baloch identity were founded p an idea of叩ality, social, economic,叫political relations in Makran were founded on a hierarchic image of society. The old struc ur f relations based on the patron-client relationship was declining, but nonetheless, deep differences of wealth and status remainf>cl o?? as it might be be阳S剧,assumed new forms. Th??? cliff?;二二二儿’ inheri阳ce of a not too distar即ast and阳啦stirr;d-b;二?h?-??ds of change”,they continued to determi?e in lare:e
me川二the
svste
?f social relations. This hierarchic model con??:s时;i?h-th;-e. ?li;?r- 1an one upon whi中山moral community was based, and so ti二iden- tity and social existence of individuals was determined bv a sort of double regime. this is p户ssible can be叫lained by sta;ting from the realization that the consistency and clarity of social r俨prese ns d二pend in lar measure on山} is in this sense that hierarch可and equality can each produce a disgui s
How
・
泣:古ct.
Ins叫“仰anr
a society, they show themselves to be consistent, in that they e intend?d to regulate two different symbolic
and pragmatic spher Accordm_g to most of the travel and ethnograpl山icli阳
rani Baloch1stan, the local population the 又Xth century, into three social
was divided till the middle o f hakim or as o
strata: the
Arabic root of the word,“rulers”;山baloch in the stri…二;二ofth
Carroll Pastner,“While the
in the oases" (1977: 121). Until the beginning of the twentieth century, to these three “classes" one more, that of the ?hulam, or slaves, had to be added. The free-
dom of slaves was gra二ted after the region became a part of British India (1892), but fo; a long time slavery was in fact maintained. It is not known how many slaves existed at that time, though their use was
For example, the region of Kharan, bordering on Makran, seems to have attracted the designs of the Khan of Kalat in the eig?- teenth century“for its slaves and its camels" (BDG). Slaves were used in agricultural work and, very probably, in the hard and dangerous work of digging underground canals. This social asset, which Makranis themselves recognized, is usu- ally, and quite rightly, contrasted with that of the other areas of Balo- chistan, where a “tribal" structure with “chiefs" (sardars) and geneal- extensive.
formally, to the presence of social groups on the basis of descent, is still in place. Descent groups do exist in Makran, but, as I said previously, they by no means carry the same social weight and function that they carry in other areas. Although they have a patronymic character, i.e., groups founded on a rather vague idea of descent of individuals from a common male ancestor whose name thev bear, they do not constitute what in anthropological literature are known as ''corporate”groups. These descent groups, or zats, are not collective owners of common ogies
corresponding,
at least
recruited
nor do their members have, by the simple fact of belongi11g one of them, the right of access to any sort of common resources. In
properties, to
Ethnography at the Frontier
50
Equality and Hierarchy
Makran men and women personally acquire and privately own rights ?o wat?r a户d land. Beyond the nuclear or extended“family”。og) there 1s a cohesive group, referred to as the khandan, towards whose mem-
hers an individual should behave with solidari?y. This is an ego-cen- ter叫group, or kindred, formed by all the living kinsmen of ;n indi- vidual male or female acquired by birth on both the paternal and the
maternal side. The khandan is, however, a structure of relations which assumes strategic and ideological importance only at the upper levels of the hierarchy. Since the status of an individual is determined bilat- erally, alliances among high-ranking families must maintain the status
of the forebears of a couple. 卫ach ?f the出ree ca吨ories o川rata making up“trad让ion al”Mak- ram society co:11sisted of a number of zats, referred to very often as ‘tribes" in the literature (BDG 19。6; Field 1959) or even as“castes”by the Baloch themselves.
Right from the
start, these
lem of “translation”.
groups had for
me
presented a real prob-
Whenlbega斗my research I had a fairly misleading idea of this type of social grouJ?.mg. For as I have said, ethnographic litera u lated zat as “tribe”or “caste”,two terms which in themselves cla d.esignate things v町different to one ano出er. Whend…二?-?; inte卜 ・
views I tried to ?nderstand山real nature of the groupings which my friends called zat, they told me that it meant“descendants of the same ancestor" Translat:d in this way二the term zat could have very well become for m w1 nh my prev10us experiences of research among the 。r me, Beduin of A灿ia, a synonym for叮in吨e”. The term zat had been also tran?lated as“caste”by the surveys of British func th二 late nineteenth centur予As the Baloch themselves were anxious to f.omt out, some zats were effectively‘superior’,others ‘ir由rior'. Even ?mea子es can be ‘S叩erior' or ‘inferior’,but the lineal mean- mg the representation “from within" which in the lirideology, ea2"e soc・ we have of山se groupings, is explicitly咿litarian. Th;;?:s?;?s con ・
・
cerned‘l王now'
that山re are more powerful lineages a?d。由rs that and poorer ones (Peters 199。);“holier" and
re less so; richer lineages ‘less
holy" (Gellner 1969).
look deeper into the meaning of the term zat, my resorted (with exasperation) to the English word caste. The
When I friends
51
term zat
is
tried to
a “translation" of the
Hindi jat
(“descent”,“race”,“stock’:
and that, due to a phonetic mutation that seems recurrent in the languages of the Indian subcontinent of Iranian stock, it changed the j into z. In realty, as currently presented, zats have little in common with the “castes" proper (jat) of Hindu India, which are hierarchi- cally established according to criteria based on degrees of ritual purity (D?mont 1966). If however we leave aside the fact that the Baloch zats are not hierarchically arranged on a ritual basis, they are nonetheless arranged hierarchically within the society. Although to my knowledge,
“progeny”)
of that hierarchy exists, there are“superior” It was therefore this hierarchy that induced the British to assimilate them to Indian castes. The fact that the Baloch“translate" zat as “caste" meant and involved a number of things. Firstly, that not having what Nadel has called a “knowledge of the current usage of words”,I needed a“translation”- Furthermore, that the translation supplied to me by the Baloch referred to a term not from the Balochi language but from English (caste), which, besides being in its turn a derivation of the Portuguese casta, referred to a context largely extraneous to Balochi culture: that formed by ?he existence of a social hierarchy expressed and founded upon an“idea of ritual purity”. This translation was therefore not only “inexact”, but reflected the presence of a linguistic hegemony exercised over the Balochi by the English language (zat = caste). The processes of transla- 丁ion, in fact, always and in any case imply exchanges of force betwe宁n languages and between cultures. So, the very fact that I went so tar as to define a zat as a‘patronymic group’witnesses that not only did
no
explicit codification
zats
and “inferior" zats.
my
word" not exist in own language (lineage and caste very soon proved inadequate), but that I had been compelled to“widen" my tran.slation by resorting to a periphrasis. I had to find “new ways”into my language, to experiment with fresh means of expression.
the “right
As a matter of fact, the image of Makrani society as divided into thre与 strata, is inconsistent with the oatterns of interaction between social groups. In my opinion the use of this model by external observers as
52
Ethnography
at the
Equality and Hierarchy
Frontier
the “true representation”of local society is a typical case of the objecti- fication of native categories. The岳rst thing to point out is that the1divi-
Makrani population into the three strata of hakim. baloch. and hiz.matkar was not originally an indigenous conceptio卢. These categories were imposed for fiscal purposes by the bureaucracy of the sion of the
Kalat the
Khana风which domina时Makran between山吨htee?th and
half of the twentieth centuries. This fact has been clearlv rec- ognized by some authors (Pastner 1971b), who nevertheless t elaborate. In fact the division of the population into three strata is over- simplified, though not completely inconsistent with the way Makranis conceived of the social differences in their society. Hakim (ruler lor?) was a term used locally even before山period of K山’s domi- nat10n. It applied to every individual or group capable of establishing power over a certain area, and it did not refer to particular patronymic first
l
groups (zat). The Gichkis, who had for a long time dominated Maitag from their castle, for example, have long been considered hakim both by Makranis and by the Khanate of Kalat. Groups fighting against them for local supremacy sometimes succeeded in obtainin2" a share of power, and as a con叫uence山y were called hakim by由;eople and ognized as such by the Khan of Kalat. But if such groups lost their power they would no longer be considered rulers. and thev reverted to
balochi established control over Makran, along with the Gichkis, groups r ognized as hakim were the ausherwani, the Mirwari, and the Bizenjo. Groups that were strong enough to have influence with the Khan- ate or local hakim could, like the hakim themselves. escane taxat' Not all the groups that succe价d in avoiding fiscal…;;?l?-a????d attain山i ion from the Khanate. When the British arrived in Makran the hakim were the local representatives of Kalat, while other ore-eminent baloch groups, the Kah叫ai o飞 Dasht (a regi?n south of K;ch valle;;ere plu vial agriculture is dommant), for example, avoided taxation but never acquired the official status of hakim. !he third category in Baloch society, the hizma向γ,consisted of vanous groups of landless pe干ple. It included nomadic p;st alis叽dl e dor岳shermen of the Arabian Sea coast, artisans such blacksmiths as
N
53
and domestic servants (khanezat) or others obliged to work for landowners. In Makran, and especially in the mountain- ous area around the valley of Kech, there were few nomads at the time of my research. It seems, however, that they lived scattered in small
(lori),
stoq叫ellers,
grou area even in the past.
The minor importance of nomadism in this
area
their is in fa叫时ably due to the presence of potentates who founded
power on the control of an agricultural population that could be taxed ;nd mobilized for military purposes. Nomadism, in this area, could have been a “refugee”response to a power (that of the hakim) consid- ered irksome and oppressive (Fabietti 1996b). Hizmatkaγcategory also included groups like the darzada, who were reputed to be descendants of the original inhabitants of the region. According to local lore, the Baloch imposed themselves on this agri- cultural population but this interpretation seems, however, to be an a posteriori justification of the social hierarchy. We can therefore see that the hizmatkar category was composed of groups very different both
occupation and social status (BDG Makran: rn7-8). Membership in making up the hizmatkar was conceived of as ascribed. To be darzada, loγi, or khanezat meant essentially to be born of a dar- zada, loγi, or khanezat father. The Makrani population was thus partitioned, during the domina- tion of Kalat, into three distinct categories on the basis of both political position and wealth. In the Makranis’own conception of this hierar- chy a hakim was a man of power; a baloch was a member of a Balo- chi zat and a free man; and a hizmatkar was a member of a group car- rying out a menial occupation. Though this model is a product of :he notions Makranis do have regarding social hierarchy, it is not in tact consistent with the way Makrani society actually worked in the past or worked at the time of mv岳eldwork. It should be stressed fir;tly, that these three categories were not so ‘closed”as native (and anthropological) representations of the hierar- chy suggest. There is, for example, a fairly common form of hyperg?-
in
the groups
amy in Makrani society in which men of higher rank marry women oi The offspring of such unions hold an intermediate status between that of their father and mother. Such considerations of status
lower rank.
54
Ethnography
at
and rank are very important throughout Makrani society, even if at present they seem to be less so than in the past. It would be a mistake to think that this kind of consideration concerned only prominent political groups. Status ascription was particularly important in mar- riages between high-ranking groups, i.e., landlords or groups heavily involved in the struggle for power. 2
The second factor that must be taken into account is the real svstem of discriminating amor groups of differ ma?eon山basis of由“amount”of izzat (“honour" or~eputation”) 二ac? ?roup is perceived to hold. The “respectability”of an individ叫is m this case assessed differently from when the issue is community or
“ethnic" identity or shared social values. In the latter case all M二1kra nis are considered equal, and their izzat will decline onlv as a conse- quence of behaviour deemed inadequate in terms of Bal与ch cultural values. In the case of individual status, honour is ascrip-
determined by a pers?n’s membership in a smaller social group, e.g. zat or khandan. As it 1s commonly conceived in Makran, tl盯1ai川e阳n tively
nant ot a group's status is its location in the three-tiered social hierar- chy. All zats co:°sidered to be baloch, for example, would enjoy equal status under this conception. In reality, however, not all zats belonging to the larger baloch cat- 宁gory enjoy the same status. Inequalities exist between ;hem not or由 property owned by the members of the zat, but also in their colle?- tive reputatio且Such differences a阳10t freq阳1tly or explicit stated by people who are more concerned to emphasize the equality of me n who belong to the same moral community. These differences can be inferred from. behaviour in matters such as marriage, political and
m
veryday relat10ns. Regarding this last ?oint, I recall that my hosts, although they had a “modernist" (and some ways“progressive”) vision of society and politics, became rather guarded when they felt exposed to pos- sible involvements with persons of lower rank. I had oroof of this when, bei吨wi阳lt阳1sp叫I looked for someb?-d; who might
m
2
Gick?s, for example, were not taxat10n.
Equality and Hierarchy
the Frontier
1叫ownerψut山irw叫山川句came from
me
55
to the villages and oases of the valley to interview hosts, who farmers, builders, traders and local administrators. one summoned transport at possessed no means of all, who, after service as a soldier in Oman, had invested his savings in a pickup taxi with which he plied the road from Turbat to Mand, almost performed the task that had been on the Iran border.
accompany
My
Muhammad
Muhammad
assigned to him.
With an astrakan hat and dark
glasses
aged to see what colour his eyes were) he drove me far days through the Kech valley and back to Maitag.
(I
never man-
and wide for Muhammad was a taciturn man and told me very little indeed about his life. Upon our return, Muhammad, whom I had paid for his services, was dismissed fairly abruptly by my hosts. Some while later, when I again needed
Muhammad directly to enlist his again on a short trip similar to the first. When I got back, Aqil took me aside and, in a roundabout way told me it would be better “not to mix too much with certain persons" (Muhammad was never mentioned by name) who “did not have a good reputation”,because "it was not prudent”,“could be risky”,and so on. It was clear that Aqil feared nothing at all for me, otherwise he would not have bothered to contact Muhammad in the first place. Aqil feared only for himself and his family and his fears had to do solely with the possibility that, through me, an undesired relationship might have been established between Muhammad and them. What Aqil was afraid of, in short, :'as the possibility of being involved, despite himself, with persons of lower status, as Muhammad certainly was compared to them. transport, I
thought of contacting
help. I set off
Although most members of baloch zats assert their Baloch identity on the basis of their ancestry, some claim descent from non-Baloch. Mem- ?ers of so叫1igl『ranking zats claim d侃侃from Arabs, Persians, or trom groups from other parts of Balochistan. While they claim differ- ent origins to that of the Baloch, they declare themselves to be Baloch j了all effe侃On the one ha叫then, they assert山ir membe灿ip in ??e sai:ne moral community as 。山r Baloch, and thus that tl町吨ard er阳n of equ y叩ramo削On川the other ??e ut an accent on their difference, i.e., on their foreign origins, which
ha训
.
lil
Makran is synonymous with being a“conquering race”・
The attribution of common descent to all the Baloch zats satisfies the conditions for conservation of the moral community. In reality the status of a zat derives from the position it occupies in a social hierar- chy shaped by factors having little, even at a representational level, to do with genealogical purity. Thus we have a picture of a system of relations founded at once on ideas of equality and hierarchy. This equality, which as the mainstay of Baloch identity seems conceptually to subsume the idea of hierar- chy, gives way to the latter when switching from the representational level (corresponding to an image of society as founded ?n the sharing of given values) to the operative one. It is on this operative level, i.e., on the level of the practices shaping the system of economic, marriage and political relations, that the hierarchy subsumes equality. At this point we must ask ourselves in what way these two representations of society were affected at the time of my research, by the rapid changes that were taking place in Makran. Makranis were aware that
their society was caught between two con- trasting social forces. the one hand, thev sensed that social rela- tions and tradi?ional models of behaviour w?re facing pressure from
On
new events and circumstances; on the
other, they perceived as inade- quate the rhythm with which their society was mobilizin豆to meet the challer
Equality and Hierarchy
Ethnography at the Frontier
56
es cl阳angel邸as
crea时Baloch attri
backwardness of the people, which in their opinion was caused by the Pakistani government’s lack of commitment to any genuine develop- ment in the region. The uneasiness felt in the face of change was deep and immediately perceptible to outsiders. To understand the solutions the Baloch have developed to cope with the “discomfort of change”,we must first grasp the nature and extent of that change. What during my stay in Makran was perceived as the
Makrani society was the more or iess direct prod- of t予ree faαs: the s阳dy decline of the system of patron-di…rela- t10ns; the monetization of the economy; and the encapsulation of the region by Pakistan. brief illustration of山processe;山t have led to a virtual total monetization of the region’s economy will help explain the other two factors.
‘change”affecting 斗ct
A
Decisive steps
57
toward monetization took place in the beginning of
century, when the British reinforced the Kalat Khanate’s presence and a new system of tax collecting was installed. Those taxes were paid in cash, and although this did not monetize the economy, it was certainly a step towards a more radical transformation. More recently, the rise of emigration and market-oriented agriculture have caused more extensive monetization. The growth of market-oriented agriculture in Makran has resulted from the region’s encapsulation by Pakistan and the subsequent creation of limited road and communica- tion networks. The export of dates to the Gulf States is another con- tributing factor, and, though that trade is an old one, growing num- hers of Karachi-based companies had begun buying dates for packing and export in recent years. Another factor in the monetization of Makran's economy has been emigration, and the consequent entry of a part of the population into a wage-earning labour circuit. Emigration is not new to Makranis. For centuries there has been movement of peoples between the coast of Balochistan and the Gulf countries, southern Arabia, and east Africa, to say nothing of India e Central Asia. In the 196。s with the explosive growth of the oil economy in the Gulf countries emigration became a the twentieth
large-scale
phenomenon.
There are no fully reliable statistics available, but it can be estimated that in the 198。s there were not less than I抖,。。。Makranis working outside Balochistan, which was one tenth of the region’s population ?Bal叫and Khalid叨。:4日). The overseas workers were men aged tro? twenty to forty, and tl町made up a substantial part of Makran亏 productive population. An idea of the importance of山money earned by emigrants for the local economy can be had by comparing the annual wage of a Pakistani worker at that time, around 670 with that of an ¥ em?grant worker in the Gulf area, around¥ 3250. Many of Makran’s
my
?migrants were, until a few years before arrival, Baloch who had army of Oman. Since the end of the 1980s however, had stopped recruiting troops among the Makrani population and had gradually phased out its remaining Baloch regulars. This has been a
Joined the
s阳
by their mercenary
menfolk.
Oman
58
FU
舍’-
’h
no
σb
r
nr
尚品
tn vd
a
φL
←・・
zn ρL Fr 俨A
on
舍L
--I
Some migrant workers were recruited through specially-authorized government or private agencies, but a fairly large percentage of them found work through unofficial channels. Men who left the villages or nomad encampments for the Gulf countries usually staid abroad for a period of three to岳ve years. Nearly all left their families behind in order to increase their savings and because of the restrictive immigra- tion laws enforced by employer countries. Younger, unmarried men saw emigration as a means of making sufficient money to pay for their marriage compensation and, at least initiall民for the maintenance of a family. In the late 198。s emigrants had to pay Pakistani Rupees凡。。。 (about¥ 360) to obtain the necessary documents to work abroad, an amount that was equivalent to岳ve average monthly earnings of a Paki- stani worker (in Makran a worker on daily wages was paid Rs 4。to Rs 户per day). This forced many to give up the idea of emigrating. Others, however, resorted to taking loans from moneylender飞usually at roo per cent interest.
Because remittances constituted a major element of Makran's econ- omy, emigration had been a further cause of change. It had not only sparked a rapid increase in the size of the wage-earning labour force, but it had also contributed to the restructuring of the traditional pro- duction relations in the region. Emigrant labour had introduced a con-
siderable mass of money into the system of local production and trade. This in turn had boosted the demand for consumer goods imported from other regions of Pakistan, Iran, and Southeast Asia. As a conse- quence, the development of the local economy had augmented the cir- culation of money, which was by that time the predominant medium of exchange, especially in transactions connected with the supply of labour. Additio户ally, as the government had developed infrastructure and social services, so Makran’s building sector had grown, as had the 干umber of go?vernment employees. Very often the;e government employees came trom regions outside Makran, and this irritated the
community which suffered from chronic’unemployment. Thanks to these elements - remittances from emiεrants abroad, the growth of demand for i口W时goods, and governm?nt spending? the local economy had moved sharply towards monetization. local
Equality and Hierarchy
AL rA
59
While men from all strata of Makrani society have emigrated, most have been hizmatkar. This had enabled individuals without means of subsistence to find alternatives to wage labour in the agricultural or building sectors of the local economy. Emigrants who had managed to accumulate enough money tended to invest in the agricultural or retail sectors, sometimes with a fair degree of success. The purchase of water shares or the opening of small shops were the most frequent types of investment among ex-emigrants. In some cases men had saved sufι cient money to buy a vehicle (particularly sought-after were Japanese- made six-seater pick-ups) with which they made a business of trans- porting passengers.
Another factor that had led to the monetization of Makran's econ- omy is that for the past fifteen years before my research the region had become an important transit route for heroin smuggling. Local smug- glers, accustomed to profitable trade with colleagues in the Gulf coun- tries, exported heroin and imported motorcycles, whisky, cigarettes, gold, and VCRs which were resold at about half their market price. Although local drug consumption was also increasing rapidly, large profits were made from the transportation of heroin to the Arabian Sea coast. Heroin reached, jus・t as probably nowadays, the coast from the North-West Frontier Province after crossing Balochistan by camel or truck (Akbar 1989). A solid network of intermediaries had developed along the trails crossing the region. At the top of this network were individuals capable of controlling drug traffic on a gigantic scale, often with the interested connivance of local of岳cials. Thanks to the mass of money in their hands, these traffickers had become leading entre- preneurs in agriculture, the building trades、and, in the coastal areas, fish processing. They were import;nt dist;ibutors of resources and
had large numbers of individuals more or less directly in their employ. Due to出eir ability to mobilize men and岳阳ices;山influence of these traffickers
was considerable indeed, and politicians in search of sought alliances with them. One of the fac- 吵favoring the development of the drug trade was the high unem- pioyment rate among Makranis. This drug traffic had created consid- r中le _w叫h in a short time and contributed to the disruption of the supporters or consensus
region's
previous social order.
Ethnography at the Frontier
60
Equality and Hierarchy
Local consumption of heroin had not spared even my hosts' family. Shortly after my arrival, when in the evening by the faint light of an oil lamp, I would sit in a circle with my friendly hosts and their friends
who had come to see“the foreigner”,every now and again a young man would appear, having silently crept into the betek. He would sit
one
behind the circle of guests, his head hanging down. Some- times he would eat a few leftovers from our dinner. Then he would leave, in silence or whispering a timid shap washbat-e, good night. After a number of these rare apparitions, I plucked up courage to ask Dosten who was the young man with the jet-black hair, fine features, sad reddened eyes and a slightly darker skin than that of his family. He was Ahmed, Dosten’s youngest uncle and hence the brother-in-law of Alam and Aqil. Ahmed, Dosten explained to me, had succumbed to the use of heroin. He was married to Zubaida, a lovely young girl from another village by whom however he had not yet had any chil- 斗ren. And a good thing too, commented Dosten, for how could chil- dren grow up with a father on drugs? Whenever Ahmed appeared I to
side,
sensed a slight与hange of mood among those present, sense of embar- a rassment and pity, or perhaps vexation towards that young man who had let himself be dragged down and caused great concern among his family. But I never saw anyone address him brusquely or reproach- fully. Let us say he was for the most part ignored. His fleeting appear-
by吨ht often}叫to do with his “outings" in the company of ?ther young men who, like Ahmed, were on drue:s of various kinds,
61
from groups considered to be high status and accustomed to a high standard of living, had become poor. Despite shortages of agricultural labour resulting from emigration and the tendency for those who remain to shift to construction or commerce, unemployment is, as told, high in Makran. The unemployed are mainly young and educated to middle school levels. There are few professional opportunities for them, though in recent years some had found oppor- businessmen. Others,
facilities that supply telephones, and other services. Officially these con- tracts were granted on the basis of competitive bidding, but in reality they were granted on the basis of criteria ranging from personal con- nections to outright bribery. These connections were often based on traditional relationships of solidarity associated with consanguinity or other affinities. The pursuit of alliances among members of groups of equal status and wealth, or the rivalry that arose between them, did not from this point of view differ much from the social scene of previ-
tunities as
government contractors, constructing
water, electricity,
ous decades. But when patron-client relations were dominant the stakes were the control of water and land. Controlling these resources meant controlling men and thereby gaining access to their labour and their support in the event of conflict. The patron-client relationship allowed anyone having a substantial retinue to demand armed service from his
the event of conflict. This service was called sγen bandi, or up the loins”,with reference to the act of putting on weapons. Today the stakes are the control of the channels through which it is
clients in
a巴ces
“girding
b山princip均heroin. On one of my returns, havi且g not seen Ahmed I asl叫Aqil how he was. Aqil replied evasively, that he had oved to some relatives of his. The only certain thing I found out was that Ahmed’s wife, Zubaida, had died shortly before. I asked how could that have been possible. They told me she had been killed by the fever, a于d that her arms had been suddenly covered with red sp?ts. I believe 1t must have been meningitis.
possible to obtain contracts or other economic and political bene岳ts. At the time of my researches, a new social figure w今emerging from t?at rapidly evolving situation: the mas归or呗g”(man). In truth, the masteγwas a figure that had already been present in this region for some time. The term applied to anybody powerful and rich enough to command respect. Alam’s father-in-law, Dosten’s grandfather, had in
r叫p飞ar,
Thed与line oft时itional servile relations and the patron-client system ed a new type of social mobility and social deoendence. Some ind?vtduals belonging to groups not particularly wdl-off in the old ocial structure
had managed to become successful tradesmen and
山P叫been con耐e时a mas的precisely because of山influence and authority emanated by his social position and personality. But these new mast古γs, often looked upon as outright parvenus, were characters heavily involved in illicit dealings (smuggling, heroin traf岳cking). They were able to create their own private retinues which restated, on bases very different to those of the past, the logic of dependency founded
62
Ethnography at the Frontier
on the patron-client relationship. The power of these masters was not con岳ned to the control of resources and of a local clientele, but was grafted into the system of political and economic connivances between elements of central government and the emerging local powers. When an uncle of Dosten, who lived in Dasht, died, a few days later (con- dolences can be offered within forty days of the death of a member of the family) there appeared a heavily escorted personage, with an air of somebody who kept himself to himself. He stayed a few min- utes to pronour山d印1al formulas. As he was goi吨away I took a photo of him and his entourage, which seemed to upset him consid- erably because he lingered a while and asked Aqil after drawing him
to one side, who I was and what I was doing in those parts. Evidently Aqil must have reassured him that I was neither a Pakistani ooliceman nor some sort of western police agent busy hunting down traffickers
and the like ...
My friends subsequently dissuaded me from asking too
many questions about these ties, since they considered the busin卢s to be “very dangerous”,for me but also for my hosts. At this point we can briefly sum up what we have dealt with so far. An
exami.nation of the way people of Makran perceived the social differ- ences between groups has revealed two important facts. The first is that the traditional division of the local population into three strata only
approximated the actual position of each group in the social hierar- 如and the second is tha aescent groups, run counter to the egalitarian ideal upon which Baloch id二ntity and the moral community were based. The consequence of this was that individuals depended, as far as their social ??sition in the broad sense was concerned, upon two削ithetic a?l contradi伫 tory representations of society: one founded on equality and the other n ?ierarchy. The social and representational reality of Makran was much more complex than that received via Kalat. which. thouεh not completely inconsistent with山actual situation. had b卢卢iveιsort of objective reality both by Makranis and by臼;阳 ?his is not the biggest obstacle to be overcome in understanding the mcongruencies of this system, i.e., the ambieuous situation in讪ich 叩ality and hierarchy stood side ・by side. T? say th斗-;;?; one, now
Equality and Hierarchy
the other prevails,
63
depending on the context, does not reduce the issue gap always exists between models of behaviour
to a realization that a
and actual behaviour. In the following I will refer to a particular case how equality and hierarchy encompass each other accord- ing to the context of discourse. to discuss
I resided, had a population of about two thousand some dozens east of Turbat, the main centre of the Makran Division in the middle of the Kech Valley. Here, as I told, date palms, cereals, and a large variety of vegetables, fruits, and other products are grown, using underground (karez or qanat) and open
Maitag, the village
and was located
(kaur-jo) irrigation
channels. Until the first half of the nineteen cen-
was the residence of one of the most powerful families of Makran, which till the first part of the twentieth century held a promi- nent position in the political life of the region. This family belonged to the Gichkis zat and its members enjoyed the title of hakim. The hakim ?esided in a fort (kalat) while the inhabitants of the village lived in mud houses or semi-permanent huts and shelters made with wooden poles, tree branches and mats. In the nineteen centurv there were verv few mud houses in Maitag or other villages in the ;egion, because a・rmed conflict and political turbulence resulting from the rivalries between hakim forced people to remain mobile. It was only in the 192。s that the number of mud houses in Maitag increased. In time, because of its location near the main road connecting Turbat with Mand near the Iran border, the village grew. About twenty years tury Maitag
before shops,
my research the core of the settlement. includine its bazaar、its and some m叫hous风was swept aw今by a flo;d仕om a止mi-
permanent river (kauγ) running not far from the kalat. Since then all new houses and shops at Maitag have been built far from the kaur and
The latter, which formerly the J dominated the village and now lay outside of its core. 叫aitag was inhabited by people of different origin, status, and occu- pation. Most of them were Makrani Baloch. small landowners or ten- ants, a仙m由ers of what was once山hi;matkar cla低Th the kalat.
ha习ar,
als?
people from the northern, Brahui、areas、as well as from other a;ea and ・Dasht. Some were from
r吨ions of Makra叩。tably山巳oastal
64
Equality and Hierarchy
Ethnography at the Frontier
Penjab, others from Iranian Balochistan. Land was almost entirelv cul- tivated by people who did not own any rights to water and ther?fore worked for those who did. Darzadas, whose status was lower than that
oft:ie切l叫sometimes owned plots of I叫and rights to w就叫ut such cases were rare. Shops in the bazaar were owned by people who possessed a certain amount of capital, that is, by those who got sub- 斗antial rent from their agricultural lands, or who had savel money during years spent in the Gulf countries. Alam. for examole. had opened his medicinal shop with the proceeds of the agricultu;e ?n his and his wife Asia’s proper时s, although the last time? I called 。“im he told me he no longer ran the shop. I couldn't quite understand the reasons for this decision of his. Perhaps business was not so good any more, or maybe he had simply got fed up with the whole thing. I recall ?nly that when I asl叫him why he had not gone into the construction business, as his brother-ir卜law Aqil had done during my absence, he replied in a subdued tone:“man azand mardom”“I am free man a Artisans were rare, due to the introduction of i?dustrial oroducts in 山market, though traditio叫blacksmi山continued to卢nufact and repair agricultural tools and forge iron points and blades while a goldsmith repaired and produced necklaces, bracelets and rings using '’
...
very thin
foils
As we have
of gold.
group or zat in Makrani society had a repu- (izzat) which distinguishes it from others. Among zats whose reputations were equal a kind of “communication” ?ook pla.ce. By “comm1 ication I mean especially inter-marriages because It was mos?ly in this field that the ideology of honour maked itself felt. But marriage was also an occasion when families conserved or reinforced their economic and political position. Due?o山idea of honour and reputation出hat per ded the Makrams stated that they married members of their own zat or wome n ?ation
and
seen, each
level of
honour
?f zats holding equal status, so that no decrease in status could result ?rom marr吨e. Generally speaking, though, zat endogamy was noth’ ng more than an ideal put forward by the locals to assert their own status. In fact marriages between members of zats of different status di斗 occur. Such marriages were mostly between men of lower Balo- chi zats and women belonging to darzada or khanezat groups (both
65
which were included in hizmatkar category). At the bottom of the ladder this was not a problem, but further up the hierarchy, mar- riages with people belonging to a different zats were carefully consid- ered. Some former hakim like Sheh Murid who lived their love story to the end by marrying women of low status must now face the unwill- ingness of the young men of their own zat to marry their daughters, whose status was considered to be lower than theirs. These, however, are extreme cases which were mentioned by the locals as examples of how dangerous it could be to deviate from such a fundamental social of
social
principle as respect for
hierarchy.
In Maitag, at the top of the ladder were a few baloch families
which
seemed in the past to have surpassed in reputation and probably in wealth the families of the traditional rulers (hakim). These families, amongst which was that of the father of Alam’s and Aqil’s wives, enjoyed their economic and political position due to the fact that their members acted on behalf of the same past rulers (hakim) of Maitag. They acted as“counselors" (wakil) of the hakim. But thanks to shrewd marriage strategies, they did not let themselves be identi- fied with unpopular representatives of the Kalat. After the partition of the subcontinent in 1947 and the subsequent integration of the Kalat Khanate into the Pakistani state (1955), ancient orders were unhinged. Former hakim lost much of their local influence and “reputation" but the Balochi families that had acted as their wakil didn’t, as they man- aged to keep the hakim “at bay". Some of them had managed to get themselves elected motabeηsomething like “mayor”,a post formally recognised by the Pakistani state although subject to the authority of the
commissioneηappointed and sent by the central state. In the course of time these families established, by means of marriage alli- ances, relations with powerful Balochi families in other areas, some which were former enemies or rivals of the hakim of Maitag. At same time, as I told, they kept their distance from the local hakim, wit斗 whom they never intermarried, and so did not suffer a dam- a?ect即utation as did 山hakim, whose methods of r always appreciated by the people of Maita?. The ha走im non only 飞f
the
?:ac
ne peasants
and servants, impose punishments, and confiscate water,
Ethnography at the Frontier
66
Equality and Hierarchy
On
land, and cattle. the other hand, because of their social position these high-status Balochi families had no partners in Maitag with w且om to conduct marriage exchanges. Theirs阳 of course, make them desirable partners to members of other local
Baloch zats, but they avoided marrying their men and women with other people of Maitag. Instead, the men of these families married women of their own or of different zats of equal status. This means they tended to marry“outside" Maitag. 号iscussions with my fri叫s
and members of these }峨-status families
dawn of new social dynamics through which Makran would have been able to join the process of '‘modernization". Generally speak- ing, this attitude ended up, in fact, as a“modernist”discourse. More lucid than the other, this attitude exploited every element that could
the
the transformations affecting Balochi society as a marriages in this regard were cosidered those which created ties to families that had connections with government, with individuals who had educational qualifications that enabled them to assert themselves in the liberal professions, or with groups holding power and wealth locally or elsewhere. This attitude sought, within a new political and economic frame-
link local reality to
whole. Desirable
about marriage agreements revealed that two different ootions. which 。”rinciple do not contradict each other, corresoonded ?to tw。 differ-
work, to reformulate the egalitarian ideal that constitutes
ent perspectives
tity.
on the future of the
ern Balochistan in general.
local
com卢unitv and of
south-
The ideal that marriages sh。uld take place
?nside山zats seemed to express, even if confusedly, a general hostil- ity towards elements which might threaten the status quo. This attitude did not go beyond a vague proclamation of equality among all Baloch, and it can be thought of as the byproduct of “traditionalism" or“con- servatism" It coinci?ed with a generic refusal of any出ing ti阳night undermine Makram autonomy, Balochi culture, local customs, and a more general sense of identity, morality, and the certitude 山t山 world was not falling to pieces. It was perhaps a passive response to the trend of ch:nge. The other "ideal”,which saw intermarriage involving different zats as an appropriate strategy, was an expression of similar ??斗cerns about change and the possibility that Ma』;rani society could disintegrate. This attitude, however, did not correspond to a 'passive approach towards questions like identity, culture, ethics, or the certi- tu des on which the meaning of life rests, however. Rather、it was the
expression of an“ac的e" response towards山se charσes一i? the sense it reflected an attempt to neutralize their poten?ially disruptive effects on the local social assets through a reinterpretation of some of 山principles on which the community was fo??由d. This attitude was, in fact, part of a more general disposition which was not exclusive to those people whose aim was to improve their status and reputation that
through careful marriage politics. ideological strea
Such a disrosition was part of a new
67
Baloch iden- The encapsulation of Balochistan by Pakistan has, as we have seen, produced new dynamics in the distribution of economic and political resources. This situation and the response to it did not represent an absolute break with Makrani political history. Makran has often been ruled by external powers which local people treated with some aver- sion, though such feelings did not, of course, prevent local rulers from ?ea_ling with肌h powers in order to gain their favou肌As did hakim m th二 past, so too did those groups that have acquired enough strength to :ake an active part in the process of change. Here we are faced, however, with another and more general con- tradiction between behaviour and ideals. It consists of the adoption of
from the egalitarian model that kind of cohesive ideology upon which Balochi identity, and the related moral community, were based. This is because ideal norms, or .models, contain in themselves elements that make such a manipu- lation possible without questioning the main assumption upon which they are founded, in this case the notion that “all Baloch are equal”・ ?f w飞毗e into account the concepts that they employed explicit"1y or implicitly to manipulate the norm, we will realize that what appears 丁o be a contradiction between norm and behaviour、for the Baloch is not. Those concepts include the equality of all Baloch; the impor- tan?e of kh仰zdan; and the transformation of societv in which the old zdt en gamy and the marital灿tion of g叫S is marriage strategies that clearly deviated served as a
口乳::; ?:::?d
Equality and Hierarchy
Ethnography at the Frontier
68
This 血anipulative process started by assuming that in an egalitarian is, a society in which individuals could assert themselves independently of birth and descent, zats were a relic of the past. Here the egalitarian ideal subsumed the hierarchic one, and was used in the context of a“modernist”line of argument which said that zats were ‘relics of the past" because they were a hindrance to the intermingling of the people, and that a new, dynamic Balochi society could not be founded on such a、ste-like system”. Instead, peopl二should m盯r according to their “inclinations”. Such inclinations had little to do, of ?ourse, with “personal feelings" and the like; rather, they concerned
the decline of the patron-client relationship, and the power and resources, had compelled groups tradi- tionally occupying the upper levels of the hierarchy to search for “ade- quate”alliances in order to maintain their position and status. Once again the hierarchic ideal, this time “spoken" by the modernist dis- course, encompassed the egalitarian ideal. By its rejection of the zats, the modernist discourse seemed to r可ect the ascription of status and thus the hierarchical ideal. Dismissing the zats as “relics of the past" did not mean, however, that the modernist perspective tried to undermine the central pillar of the Makrani social social relations,
society, that
redistribution of
the stra吨ic choices that allowed people to
structure
improve their…tus, w创h, educational standards, and political influence ... Aqil was a real theo- ris\of thes二 opti<?ns. When I met him for the last time in 1994 he told me he was very busy getting his two sons (who had lived for years in Quetta) into administration and health". ..
While the search for greater political influence was of m幻or concern only
to high-status Baloch, these aspirat amo吨the population in general. They were both the pro卢ises a口d the premises of a future free from the fetters of underde;elooment which, in the eyes of Makranis, had long
shackled Balochistan. n the current context of rapid change, this positive and active approach to develop- men巧山t could change the currency of social assets and cultural values
stated new principles to which appropriate social behaviour should con- form. Those who - like Aqil - spoke from this “modernist”perspective ?nd who looked upon the zats as social fossils focused their concern on
俨cr叩io斗of平arriage叫political bonds more in keepi吨with伽 cnang1吨situation. Within this perspective the zats as groups tradi- io叫ly mves时with social status、should slide back1:?:round to
make way for the kha,均n as山focus
int二
the
for strateσic conside;ations. The“quali旷of or的kha,仇叽the ego-center向ro?p of variable span formed by all the living bilateral kinsmen of an individual. was a key con阳cern of l蚓1-r
be acquired through “right”or“wrong”marriage choices. T丰eunce…intie.s associa时with this stra鸣y were usually avoided
by
“m
69
its productive and political rela- namely a marked social strati岳cation conceived of in terms of status hierarchy and reputation. Marriage strategies, as well as the system of economic and political alliances, in actual fact always took
and that which supported
tionships,
its higher levels. devalued the zats, and therefore descent, as a factor that ascribes status, the modernist discourse tried to use new concepts to l叫plausibility, and hence moral acceptability, to the inequalities mherent in society. These concepts (the archaic nature of zats, mar- ria?e choices according to a person’s own“inclinations”,patterns of soc月1 interaction adjusted to change, etc.) contradicted the image of hierarchic society. Precisely because they resonated with a modern- 1st ideology that spoke to the aspirations of many Baloch, these con- cepts appeared to re-enhance the egalitarian ideal upon which the tra- ditional moral community rested. The status hierarchy, traditionally conceived and rendered explicit as a scale of honour among zats, was represented now in a different guise, i.e., as a hierarchy based on con- trol of resources and relations with the (despised) central power. At its higher levels at least, this hierarchy was a copy of the one founded upon izzat, so, if carefully analyzed, we find that this new pattern did not represent a complete break with the past, inasmuch as it reactivated the id叫so叫hie叫忖1st when it?urported to deny it.
that
hierarchy into account, especially at
When
it
飞
斗u?kheir的words cited at the beginning of邮chapter S盯ed pre l:Ise,1.Y to anticipate its conclusions. Makrani societ founded on dual ψhtarian and hierarchic ideals, clashed with the .rr?ality of “change” ,
Ethnography at the Frontier
70
and remained the prisoner of a dilemma which it seemed unable to escape. Just when it placed one ideal before the other, it was driven back towards that one which it professed to have put in the background. These two ideals remained in conflict but, due to the way白马were contextually used - equality serving as the foundation of the moral community and identity, and hierarchy as the pivot of the political and economic order - the contradiction between them was n卢rer revealed as su?h. Nevertheless,
because the politico-economic order and
H
QAM
+L
or VJ
4m tG
-G ρL
Qd
Coloro’e quali solamente per fortuna diventano di privati principi, con poca fatica diventano,
ma con assai si mantengano.
the
communi_ty had begun to change under the pressure of new cir?umstances, the concepts of equality and hierarchy had also had to undergo a process of reformulation. The discourse which I have called “modernist”had, more than any other, set itself this task. But its con・ tradictions show how difficult it is for a society to shake off the ambig” uous character of certain ideas that for centuries、due to this same ?mbigui吼have forrr does not imply that everything is the same and nothing has changed in Makran. Rather the situation is precisely the opposite; nothing is as was, nor _will it ever be so again. The point is that social change, when o_ccu叭does not only entail an adjustment of society’s conc今ts to new circumstances. As Durkheim seems to tell us. the oroblems of a soci- 町do not arise from a direct confrontation betw;en ideals and r叶 ity, b?t between 斗ifferent ideals by which humans envisage山nature
MO
Niccolo Machiavelli, IL Principe,
1
1513
…。ral
it it
of their
own soc:1Ptv
一…一---・‘-阳 ??
The lunar landscape that frames the lush green oases of Makran would not suggest that in this land an ancient civilisation once thrived. And yet the facts of archaeology here tell us exactly the opposite, dating right back to the 3rd millennium the presence of communities that appear to have served as a connection between the Indus Valley and
BC
the Iranian plateau.
The “dizzy”time of archaeology is not that, more circumstantiated and modest, of the anthropologist. Nevertheles, while I was trying to understand what forces might have modelled in time the society I had
on returning. And so it seemed to me, between and an account, a colonial and an ethnographic document, that these forces might, beyond the connotations of socio-political and economic dynamics, have assumed the features of armed violence. The stories told by the locals and the historical documentation seemed fairly explicit on this latter point, albeit soaked in the aura of legend. When I was trying to put together the pieces of these stories, from t?e叫ive of memory I would extract tra乌s, hints and reminiscences bef?re a
me, history kept
history
of my own personal“encyclopedia”. The ethnographer’s never neutral, but conditioned by forms of “pre-comp hension". Thus I began to gain the sensation that the formation of the that were
eye
i斗
p干rt
fact 1s
1“Pr・
;了e citi?:斗s 10n on their
who become princ叫urely by good扣rtune do so with littl own part; but subsequently they maintain their pos
by considerable exertion"
Harmondsworth, 1961:
53.’
i
The Prince, Transl. by G. Bull, Penguin Books,
Ethnography at the Frontier
72
History and Models
potentates which prior to the arrival of the British had arisen, fought oneano出er and declined one after the other, might be read in the light of what Machiavelli wrote on the subject of the rise and fall of pri斗ci-
paliti优Sheh Murid and Bahar Khan we问on this point, my princi pal sources. The first, an elderly man (he said he was eighty) reduced to semi-poverty; the second a half-crazy pleasure-loving岳fty-year-old with eyes wild with addiction to some local drug, a heavy drinker of whiskey and of “tourist expeditions" to Thailand (when reminding me of these he would wink meanir ly... ). Bahar Khan, unlike the o?h叫 seemed to have got along fairly well thanks to his properties, in the oasis and elsewhere. Listening to the stories of their“families’,heirs to those th干t had gained power by driving out others, only in their turn to be dnven out, seemed to me like re-living the pages-of The Prince, in which Machiavelli explains how potentates are“conquered, main- tained and
lost”.
So when Shah Murid and Bahar
Khan spoke of the glories of their families, or when fallen Gichkis, the time of their words ended up t?ansported into a ‘model" derived from a book written in sixteenth century Tuscany by a republican Florentine in exile, with a Romagnolo lord as his
Do
minor
reader. But,
I
wondered, can
a
Makran
of the 17th-19th centuries
be
disguised as Renaissance ltalv? This doubt tormented me for quite some time, until I realised that, to explain phenomena like these to my readers (the succession of polit- ical eli?es that had asserted themselves by force in southern Iranian Baloch1stan), other colleagues had relied on interpretative models des- urned from the works of authors previous to or .Lnot much later than Machiavelli and known in the West at least as much as him. This real- seemed to resolve the dilemma by shifting it onto the plane ot consensus by the a?-thropological community: it authorised me to e.mul_ate the undertakmg and to set out another“model”. That realisa- tion however still left a further question open: can history be recon-
叩m ciled
with models?
Some colleagues who in earlier years had conducted research on 出e Ir句:ia.n side of sou阳・n Balo仙阳had dealt w汕ethnographic tenals many ways similar to mine. more
m
m
However, in my opinion
73
by the models than by history, they had discarded the latter from their “ethnographic representations”,putting forward interpre- tations of the process of social stratification typical of southern Balo- chistan which not only failed to take into account - I repeat - the his- fascinated
dimension, but by that very fact could not even explain any ethnographic data at all. Those data were instead important in grasp- ing the nature of the potentates that had established themselves in the torical
Makran. History was therefore essential model or, as Max Black has said, t。 “help us to highlight what would otherwise be neglected, to place the relative emphasis on details - in short, to see new connections”(Black arc of a
couple of centuries in
to“verify" the
所2:
productivity of the
237).
Without more ado, let me say that I have attempted to interpret the ethnography of the Balochi potentates by getting the models used by my colleagues to dialogue, so to speak, with mine. If my interpreta- tive model was based on Machiavelli, those of my colleagues were built on other authors: Ibn Khaldun and Hobbes respectively. For them, the theories both of Ibn Khaldun and of Hobbes constituted - the former explicitly, the latter indirectly - the key to an interpretation of the appearance of potentates in southern Balochistan. The
intr口sion of foreign powers and their albeit discontinuous pres- ence in the region contributed to give Balochistan the character of
what Owen Lattimore called a卢ontier (Lattimore 1940). The“frontier nature" of Balochistan in general and southern Balochistan in partic- ul?r, stems from the following factors: 1) the absence of a local politi- cal power capable of mediating and keepin? under control more or less 飞飞rrr
Q阳ntin.ous pressure of outside powers and their tendenc through mtermediaries belonging to traditionally hegemonic local groups; 3) the existence of a close-knit web of relations and intere s
??t?ee叫e rulir盯roups at loc仙vel and山for咿powers. Inf叫 mtrus1ve role played by these powers has always been accompanied, thern Balochistan, by fierce competition for political supremacy monglo叫ruling groups. The latte; always f。但1t one the hope of establishing privileged relations with the outside powers.
丁ne
Ethnography at the Frontier
74
History and Models
Nevertheless, within this general picture, Balochistan offers some- γhat different scenes as regards local forms of political organization. !n some cases these forms seem to have been moulded bv nteraction
between nomads and settlers. Thus, describing the situ;tion in a dis- of sou山r川ranian Balochis阳,Brian Sp;one刊xplains山strat- i岳cation of the population into “rulerγ(hakim) and peasants, here trict
called “villagers”(shahri):
a tribal, pastoral group of people invades a settled在gricultural area ... the l叫ing family of山invaders becomes the focus of p'?w们tl it invades, stays there and becomes the dominant power in the area. The invad- ers become therefore the de facto rulers of both peoples: those that entered the a.rea with it - the pastoralists - and those who were in the area when it came- ・
・.
the peasants (Spooner 1969: 147).
commumties. This
fact generates social differentiation
family叫卢nomadic ?
between
such
tribesmen. These remain pastoralists and n no control 今ver agricultural resources as the leading family does. So Spooner contmues:“The‘group feeling' (Ibn Khaldun’s asabiyya, that is the solidarity of the tribe’s members) now comes to a crisis and the l叫ing fam?ly becomes dynas旷(ibid:m). Related to this proces产 of stratification is in fact the tendency by the leading family to marry mto other families of the same status that have established thee same type otf power in other places or, if such families are not available, to marry“inwards”. The title of “ruler”(hakim) is 牛ept withi? the dynastic fam句which is in therefore able t )r e its power vis-a-vis the rest of the tribesmen. The cl唁'nastic family, ‘whe?e?? be岛r ha斗岛ug?t岛r and defended movable二op叫,onl that is, livestock - it 1s now mterested in defending immovable plots of cultivable land and .crops and peasants" (ibidem). Such a“model (sic!) is capable - Spooner concludes - of bein'I, devel・ 咿dfurth叫吨to a cycli山t吨e of reinvasio
?
w
The local population referred to by Spooner was seen for a long time distin2:uished into three strata: the hakim, the baloch in the strict peas- sense (or landowners, or nomadic shepherds); and finally landless
as
according to the various places, hizmatkar, (dependents), or tiakib (servants). As we have seen, this division of the population is typical of a region in which the peasant communities occupy oases where cereals and, above all, date palms are cultivated. These communities have been contrasted, as far as political organisa- tion, social morphology and the pattern of land occupation and the ants called,
shahri (villagers),
communities in the
north of Balochistan are concerned. The
latter, as
I have had occasion to explain in other parts of this book (see especially Ch. 3) were historically characterised on the other hand by a type of
social
According to this perspective explicitly drawn on Ibn Khaldun's ( Al M1:"qaddima), as time goes on the leading family of山invaders cquires more power and wealth, for it can control pastoral as well as agricultu?al resources, by acting as an intermediary between the two
75
organization that conjugates the typical features of “segmentary with the principle of “centralisation”,namely with the pres-
societies"
ence of chieftains (sardars) exercising a substantially autocratic power. A distinguishing feature of the organisation of communities in south- em Balochistan (both Iranian and Pakistani) is the historical presence of a settled elite that rules over a stratified community composed, as I have said, of peasants and nomads, customers and servants and, up till
twe?tieth centur民slaves.2 The origin of this type of stratification outlined by Spooner is retraced to a “model”fairly well-known among“Middle Easternist”
the early
Ibn Khaldun's Al Muqaddima was intended to explain and decline、in “medieval" North Africa. of political bodies led by noma elite叫1ich with the passing of ti?e .;ere replaced by others according to the same process. Ibn Khaldun's model “works”however as lonε as we can reasona- bly establish the existence of “t山al" links betv乌; anthropologists. the
birth
创飞nd
丁he
nomadic groups gravitati吨on the agricultural areas. Its contexts where that situation is satisfied has given results
application to
2
I
Kech valley the political role of the nomads has always the past three centuries, a farily marginal one. This fact is xplamed, firstly by the existence of a prevalently agricultural economy, but recall that
been’. at
in the
least in
al阳,byt:』二pr阳 ofη10b il a .
g
strong military force.
76
RL
争EL-
’h
no
σhu
rA 44
(Musil 1928; Montagne 1932;
PA
’H
VJ
而d 舍L
’h ←・・
户IV
RAFA
Boucheman
On
舍L
ρ』
History and Models
rA
1939; Rosenfeld 1966;
Fabi-
Al Rasheed 1991). When however the links between the lead- ing-family and the nomads cannot be proved or do not exist at all, application becomes problematical. Furthermore, this schema, as it applied by Spooner, eludes the consideration of mechanisms可hichin concrete terms may eventually allow one leading-family to take over from another. All that the model does is to highlight the alternation of etti 1984;
its
is
dynastic families
the head of potentates. first to -have resumed the Khaldunian model in an“Asiatic”context. But he was perhaps the first to have applied it outside the Arab-Berber world with reference to which it had been previously applied. at
Spooner was not the
“If牛K_』aldunian model is desumed so frequently it is because someth1斗g”must exist which justifies its application to ethnographic data. This“something”cannot be only and exclusively a represen-
by the ethnographer, which derives solelv from their habit of constructing models. This something that justifies such a frequent
tation
app!ica
accent however is placed on the岳gure of "proto-state institution" born to medi- the hakim, the pastoral nomads and settled agricultur- ate the opposite interests of function needs to refer to the exist- in order to ists. A model which ence of social relations founded upon the solidarity of blood relations (even fictitious ones) (Ibn Khaldun’s 'asabiyya), is in fact substituted by another, based on the idea of a“social pact”. On the basis of this second perspective, the opposition between the two communities apparently springs not so much from two different ways of exploiting resources. Indeed, this difference in itself could set up a profitable state of exchange. Nevertheless, as Philip Salzman has settled agriculturists.
latter, namelv the existence - when demon- between settled and nomadic communities gravitati吨。n the agricultural zon优It is in situations in which thi condition is飞atis岳ed that the model may be considered valid. However, where山existence o? ties between the ruli吨family叫出nomads cannot be proved, or m cases where such ties are manifestly non-exist- e叽the m?de! loses much of itsαplanato仰ower.From山high level of generahsat10n in which the model is situated, its application to a con- ext like ours leaves us for example in the dark as regards the concrete dynamics出干t determine山usurping of one ruli吨family by another.
- of
“tribal" links
The lac丰 of historical data to support山model in its process therefore makes its application unsatisfactory and highly sp???lative. The
lack of reliable historical data
what has brouσth to refuse this hypothesis and 叫ut forward another,叫licitly fo;nded upon struc- ural and functional considerations. The explanation of the formation 川平se stratified p?litic们nits by settled elites has been叫ht, m this case too, in the interactionledbetween nomadic pastoralists and is
The
“ruler”,as a
maintained in introducing this second interpretative model, this sit- uation of mutual advantage is placed in doubt by two destabilizing facts: on the one hand, that the nomads are militarily stronger than the settlers; and on the other, that the settlers can cope with periods of drought more successfully than the nomadic pastoralists can. These two circumstances pose a constant threat to the stability of relations between the two co血munities, because
explanatory power of the
strable
77
there are differential
consequences of the frequently recurring droughts. The
baloch (in this case the nomads), directly dependent upon pasturage for the welfare of their animals, suffer the consequences of drought more heavily than the shahri (villagers) who This means that during a
depend upon the relatively reliable irrigation water. drought the baloch will be deprived relative to the
shahri
131).
...
(Salzman 1978:
I? consequence the nomads could, the resources necessary to them
by using
force, gain possession of
but, being in no way inclined to take the_ place of the shahγi as farmers, they desist from destroying the con- dit1ons that ensure them the supplies they need. The pastoralists could therefore benefit from peaceful exchange during good years, as would the shahri, but would tend to be predatory during bad years, a fact that would make the shahri quite wary if not completely uncooperative just about all of the time. ?tis difficult to see how the ambivalence of the relationship and the instabil- 1ty of the system could be avoided without some kind of mediating institution
(ibidem, itaics mine).
History and Models
Ethnography at the Frontier
78
This mediating institution is the hakim. This double unbalanced situation (both at an economic and political level) just described is in fact presented as corresponding to a“hypo-
impersonal “structural and functional conditions”,as Salzman says, that is, from the mutual advantage that pastoral nomads and agricul- tural settlers
can both derive from
its
presence.
thetical, decentralized, acephalous,‘anarchical' society" (ibidem).
What we can observe is
in actual fact, according to Salz?an, the pres- ence of “rulers”(hakim) who mediate between different groups. -It is this mediation, thanks to which they were able to procure benefits for both the communities, that gave the rulers thei; dominant posi- tion over both.
In
my opinion this
explanation of the hakim’s special position is too rooted in a Hobbesian theory of power. Due to the lack of his- torical information, it is right to focus on“structural and functional" aspect.s of山pastoralis卧川er-villagers relationship.τ’he whole盯gu ?ent, h?:'ever, sounds somewhat abstract. It presents the hakim’s posi- t10nas扩it were the result of a more or less explicit agreement between
nomads and
the peasants.
It is
as if nomads
and peasants,
having
valued d肌osts and benefits of their opposition, had d?cid讪o invest a third party with a special power. The “hypothetical, acephalous, anar- chical" situation in which d吵aloch and山仇加is阳1cl vi叫吁is on another, reminds us of the Hobbesian “state of nature". in which the
阳u
protection to a third party. Likewise, baloch and shahη, i豆 order to eliminate the .threat of disruption in their mut叫ly adva二时eous but unstable relations, renounce their political autonomy to the advantage
of山hakim who, by this means, acquires a special mediating power
between the two groups. However, what here becomes reality.
was
in
Hobbes pure fiction
What can be criticized in this explanation is not of course that it puts the accent on“structural and functional" relations as such. But rather, the fact that these relations take absolutelv no account of the social origins of山characters on stage, and of山:\akim in particular. Here the hakim not only does not have the same oriρins as the pasto-
nomads (it is ford ?ork in such a contex趴but does not even originate from the villagers 口e seems to turn up out of the blue or, to be more precise, from the ral
On
the basis of eth- Where, then, do these hakims actually come from? nographic and historical data referring to Iranian Makran, the “rulers"
N
in question appear to have belonged to the ausherwani group. This was a “family" of fighters and freebooters very famous all over south-
who, from the eighteenth century, had imposed their on a number of oases in the region. Originally they were not set- tlers, nor were they pastoral nomads. It is a fact that the Nausherwani formed a group which, like Gichkis and others, took advantage of the widespread unrest in Balochistan towards the middle of the 18th cen- tur予In particular the ausherwani exploited to their own advantage the interference by the Afghan kingdom of that time in local affairs (B.D.G. 1906b). That interference was not an isolated instance but a recurrent pattern that saw numerous powers looking across the “fron- t?ers”of Balochistan. Among these, to mention those active in the past three centuries, were the Persian Shah, the Mughals of Punjab, the Afghan Kingdom, the Talpur of Sindh, the Khanate of Kalat (not to mention the British Raj, Pakistan, Iran and the Soviet Union). The Nausherwani are not a unique case. Other groups equally exigu- ou.s in numbe鸟but aggressive and militarily prepared, established their rule i且 various areas by the use of force. Such groups relied now on ern Balochistan,
rule
much
the
79
N
now on settlers to retain their supremacy after having estab- them由es in pow优Beca肌of山ins协ility of the吨ion, and its “frontier”character, rule by such potentates tended to be encapsu- ?at小ofte川1ly in name - in wid叫olitical bodies th们的also Detter organized on a military and administrative level. I believe that as
nomads, ?ished
regards the
process of formation of those potentates the foreign origin of由?e local“rule旷has not been川ωωι1to proper aαount. T』1s consideration can p盯e the way towards the adoption of a per- to those inspired by models dating from Ibn Khaldun nd Hobbes. In fact, the chieftains who, until quite recent times, ruled ver the oases of southern Balochistan - and who today, although they
spective different
have lost their
status as
hakim often retain a still considerable influence
History and Models
Ethnography at the Frontier
80
some oases - were all descendants of groups of looters that preyed on the region, far from the radius of action of political formations of the Iranian area to the west and of the Indo-Ganges area to the east. in
This fact is to be related to the overall situation characteristic of the western part of the Indian subcontinent between the mid-巧th and mid- r 8th centuries, in other words the period in which the activity of these groups seems to have been most intense.
Next
not a question of establishing relations of ascendancy ausherwani and similar groups, but it may reasonably be supposed that these groups arose from a very similar dynamic. The ausherwani were, as far as one can gather, of uncertain origin (BDG 1906a: 64), but other similar groups - such as the much more powerful Gichkis of Makran - claimed rajput origins3. After having become untrustworthy in their masters’eyes, they gave rise (whenever they could) to self-sufficient potentates, who were in Certainly
it is
between these mercenaries, the
At this point the interpretative
γ句・as
besian
and even of landlords
who were
jealous of their independence,
(zarr
demands of states and principalities. At times these armed groups escaped the control of their masters and would then begin to act quite
岳seal
autonomously. In his study of the “military labour market" colonial India, D.
in
pre・
可ung ag?in _into quite
ongm, (Kolff
a “portfolio"
of pos巾le sources of income. This was soldiers of Raj put, Baloch or Pathan a powerful and warlike body ... flourished for several generations
m the towns
?ase
practice free
...
where Muslim
weakness of the Khaldunian and Hob-
more evident. These models do not satisfactorily clarify the origin of the Balochi potentates, for they do not take into models seems
consideration violence as a pivotal local factor in the constitution of power a时,above all, do not treat their subject as something closed to
any“external”influence. ?t is at
this
point that the theories of Machiavelli begin to acquire
importance. Machiavelli dedicates some chapters of The Prince to mer- cenary militias. Here he singles out a number of features belonging to
the? which could be referred to the mercenary groups stationed in GuJ.rat at the time of the Mughal empire. As with Machiavelli’s merce-
On the other side the same Persian 17th century text, the Mirat斗, Ahmadi, quoted by Kolff himself, describes thei/ sphere of action as They became a source of resistance ... in this very region, in thorny places and ra:ines. They attacked villages, drove away cattle, escorted Nazims (Muslim
pnnces), took responsibility for collecting peshka均(tributes) from zamind- ?rs (landlords) on a small salary, got enlisted as recruits in the army for a few days, served the faujdars (military governors) ... they thus maintained them・ selves They could afford to refuse service in the MuQ:hal armies and did not even accept service outside Gujrat. Only later on did some of them,卢nd- ing themsef.τ,es in reduced circumstances, try their luck in other subahs (proψ inces) and made brai ・
those of the subcontinent,
no loyalty or inducement to keep them on the field apart from the little and this is not enough to make them want to die for you. They are only too ready to serve in your army when you are not at war; but when war comes they either desert or disperse (Machiavelli 1961: 78-79). there is
they are paid,
follows:
・
any control.
nanes, so also for
199。:5).
・
of
H. Kolff writes:
At the Gujrati end of the great caravan routes worked by the Balochi, these pe?ple would undergo quite a transformation over time, their activities wid- th二
N
N
to the military organization of the centralized states, the Sub- continent witnessed, between the mid-巧th and mid- r 8th centuries, a thriving of mercenary corps in the service of various“princes”(Hindu
and Muslim nawabs)
81
bei?g able to oper剖e in tl comp可1阳s阳es or bands, some groups, such as precisely the Naush- and the Gi州s,p时必l向mdanou削巾lo仙tan. Since ;?::Ill uere were no centralized political bodies here. thev could more easily
时ot
raid山吃ri叫turalset山?ents a叫if possibl?, se??le山re加mselve?
,
3
Even though, as recently
I explain in chapter been treated as secondary.
5,
for political reasons this claim has
Ethnography
82
and establish
their
own power.
at the
Cases of the kind are recorded in sev- southern part, where the
eral places in Balochistan, especially in its
population tends to be concentrated in the oases and pastoralism is of lesser importance than in other regions.
was in fact on areas inhabited by an agricultural population that the designs of these groups were concentrated. The last few centuries of local history seem to be characterized by a continual overlapping of groups that had arrived from outside with the intention of dominat- ing the rural population. That these designs were concentrated on the agricultural population is an easily explicable fact: for unlike the pasta- ral nomads, a community with only a small number of farmers can be subdued and controlled more easily and, most of all, is susceptible to produce a regular surplus which can be appropriated by aspiring rulers. According to the “frontier character" of this part of Balochistan, these groups succeeded in imposing their rule in the shadow of the m勾or powers which, nominally, held their suzerainty over the region. They ruled over both nomads and settlers, were stratified, exacted taxes, and had the monopoly of force. This force however did not con- sist of the nomads' military support, but rather of their own arms and of the support of armed bands in their service. From this point of view }he Gichkis of Makran, in the power till the end of 19th c卢tury (albeit 'representing”of the Khanate of Kalat) are exemplary. It
The Gichkis
demonstrated that as they appeared on the scene, they ousted other groups which in their turn had seized power before them. It is not very clear what sort of authority was in for卢in sou伽rn Makran at the time of their arrival. There was va2:ue talk of a malik (lit- erally“ki吨”) who governed the region. In any c;se nothing authorises 4
historically
analyse in chapter 5 a locally well known legend which narrates how they s;ized power. Such a legend is significant in showing how the Gichkis brought the region under their control by the use of force (B.D.G. 1906a).
I
83
conclude that the groups in power in Makran before the Gichkis by the same means and likewise came from abroad. 5 Gichkis, just like the Nausherwanis, could not rely on a tribal ret- inue, nor were they as numerous as the nomads and the cultivators of the area. They arrived as a band of pillards. Diplomacy and marriage alliances were also among their weapons. But the Gichkis seized power by force and held it by force. They established alliances with powerful “families”of the region, played one group against the other, and elimi- us to
had settled there
dangerous rivals. According to local tradition, the Gichkis were groups. This fact was repeatedly con岳rmed to me by my hosts. The families which at the time ruled over the oases in the region cherished the illusion that, having expelled the rulers of the moment, they would assume power in their turn. This kind of relationship with other dominant groups reinforced the Gichkis’posi- tion, but as time passed, they had to defend what they had obtained from attack by their former allies. The local Balochi families in fact - or so I was told - rose up against them time and time again. This point reminds us of what Machiavelli has to say in The Prince. nated
helped by a number of local
..
a
・
is always compelled to injure those who have made him the new subjecting them to the troops and imposing the endless other hardships
prince
ruler,
which his
new conquest entails. As
a result you are opposed by all those you occupying the principality, and you cannot keep the friend- ship of those who have put you there; you cannot satisfy them in the way they had taken for granted, yet you cannot use strong medicine on them, as you are in their debt. For always, no matter how powerful o肘’s armies, in order to enter a country one needs the goodwill of its inhabitants (Machiavelli 1961: 35). have injured in
arrived in the region in the course of the 17th century after coming - so
fro? the north of Makran, where they had settled tradition says - from Rajputana (or Rajasta吼4 It is
History and Models
Frontier
Clearl川1able川eep their previous
allies at
b矶the Gi州s needed
They had
relations with some local influential fam- but no far-flung relationships with the settlers, they had
n飞w supporters. ?lies,
whom
with nomads. As their own mi ital 时sufficient f叫1e purp叫they began ↓c7uit a mi1??r;?:?: ded, nor
5 丁创r
s
a.siz
rrept1tously a“cyclical”(“Khaldunian”) close the objecvt within a repetitive
model
type of story.
that
would be
liable
t0
History and Models
Ethnography at the Frontier
84
from outside. They had to rely on mercenaries of different origins, that is settlers, nomads but mostly strangers, and slaves. The characteris- tic of these people was that they did not form a group of individuals having a common social origin. They were given food, shelter, clothes and arms by the hakim and lived in the mud forts. These “armies" formed a permanent armed group, deployed to make sure the hakim's rule was respected, to get taxes and fines paid, and opponents reduced to silence. They were an army in the strict sense of the term that, even under British rule, was still present in the village forts. Thus, we might say, the mercenaries of former times began in their turn to enroll mer- cenaries themselves.
The Gichkis’power, of course, did not rest only on force. As I said before, the Gichkis were able to make alliances with other hakim and to earn the consent of a small part of the settlers. These settlers were
who in their turn ruled over a servile stratum of landless peasants. By developing ties with some of these landowners, they created an intermediate social class between them- selves and the peasant mass, including both masters and the land- less. This intermediate class had functions of an“administrative" sort
Balochi landowners,
composed
(wakil).
In this respect, however, we must recall what Machiavelli said prince who conquered a domain with the help of local people: This
is
what happens:
as
the weaker powers give
soon
as a
powerful foreigner invades
of
a country,
a
all
him their support, moved by envy of the power which
?as时叫omina时山m. So as far as山se weaker powers are concerned, h?
has no trouble at all in winning them to his side, because of their own accord they straight away merge with the state he establishes (Machiavelli 1961: 38).
Notwithstanding their pre-eminent position, Gichkis were not spared uprisings stirred up by other “rulers" of the area. As Machiavelli wrote: You can
you win over one of the barons. There always exis! These ... can open up the state to you and facilitate your victory. But subsequently, when you want to maintain your rule, you run into countless difficulties, as regards both those who have helped you and those you have subjugated. Nor is it enough for you to destroy the ruler's easily invade if
ma!?ontents
who want
a change.
family,
because there
still
remain nobles to
85
raise insurrections (Machiavelli
1961: 46).
groups tried, in some cases with success, to some- how evade a tax system felt by them to be a burden. When my friends spoke to me about their Khaudaui“relations" (meaning the zat from which the wife of Sheh Omar, Alam’s and Aqil's father-in-law came) who lived in the arid - but rich - region of Dasht, they never failed to stress that these,“prior to the arrival of the British, did not pay taxes to the Gichkis勺This fiscal system was in effect tough since the tax revenue served to maintain the rulers and their retinue of armed men (servants and customers) to pay the persons in the intermediate class In Makran the rebel
and to boost the finances of those foreign powers which at more or less claimed sovereignty over the region. Under the threat of local competitors not always easily subdued, the Gichkis were driven, though against their will, to form steadily regular intervals
with those outside powers. The history of their relations Khanate of Kalat (whi?h survived until 1955, even after the partition of British India in 1947) is, from this point of view, exemplary. stronger ties
with the
During the 18th century the Gichkis formed ties with their power- f?l neighbour in such a way that, when they eventually felt sure of their supremacy at a local level, they attempted to free themselves from that power. The Khan of Kalat, however, sacked their territories. The Khanate of Kalat was interested一in these territories for the same reasons that, more than a century earlier, had induced the Gichkis to take po?- session of them: viz. the presence of a peasant population concentrated 6
The Dasht region lies south of the river Kech. There are no permanent rivers there, and no hills on whose slopes wells might have been dug from which to carry water by canals to the crops. The system of cultivation is consequently founded here on the construction of large earth dams which, as Arnold J. Toynbee noted in Bet如een Oxus and Jumna of 1961 (report of a journey between Afghanistan and Pakistan in the 195。s), were a “pale memory" of those built in stone by the local populations centuries earlier, and to this day visible not far from the Arabian Sea coast. Thanks to these dams water was, and still is collected and used in the rainy season to flood fields sown with corn.
History and Models
Ethnography at the Frontier
86
in agricultural settlements from which they could extract a surplus regularly. In the impossibility of getting free of this potentate of cen-
Balochistan dominated by an aristocracy of Brahui language, the Gichkis had to agree to hand over 50% of their revenue from taxes, tributes and fines. In exchange the Gichkis managed to get themselves recognized as representatives of the Khan of Kalat in the region. But they were also forced to accept its troops and “agents”. In this way they reinforced their position towards the other local “rulers”but, to be able to do so, they had to yield to a powerful foreigner. Yielding to a powerful foreigner like the Khan of Kalat meant, for the Gichkis, establishing favored relations with him and, in conse- quence, to satisfy his requests in fiscal terms. This was translated into a subdivision of the population into three “tax belts" (hakim, baloch and hizmatkaγ). The division of the population into three “states" was in effect a fiscal innovation introduced by the Kalat and became the repre- sent干tion which the locals had for a long time of their own society con- cealmg, as I say in chapter 3, the existence of subdivisions within each of the three categories. The indirect presence of Kalat, furthermore, was translated into a hardening of pressure on local groups (Baloch) which, in their turn, grew still more hostile towards the Gichkis as rep- tral
resentatives of the Khan. The situation was in this way determined by a series of elements set out in circular fashion, the effect of which古as to make the rulers'
position increasingly precarious. In fact the Gichkis failed to eliminate internal competitors backed by the Kalat, nor to render them innoc’ uous by special favors. They in fact had to subject them to tax pres- sures in compliance with the requests of the Kalat itself, so that they found themselves back in the situation in which、as Machiavelli says, being able neither to satisfy them nor to des.troy them [the local competitors] you lose the state as soon as the opp。rtunity presents
‘・
? ?
itself"
(Machiavelli 1961: 46). a matter of fact, the Gichkis never actuallv“lost the state”in favour of their competitors. But they did lose it i? favour of山Kalat, which made itself the guarantor of their pre-eminence at local level. In a passage of The Pγince Machiavelli talks of the shrewdness which a prince, who has conquered a new area by force of arms, must use in
As
dealing with those
who
87
have called him in with the hope of preserv- own power over a competitor or ruler dis-
ing or incrementing their liked
by them:
watch is that they do not build up too much strength and too and with his own strength and their support he can easily hold down those who are powerful and so make himself, in everything, the master of the country (Machiavelli 1961: 39).
All he has to
much
authority;
en passant, that these statements by Machia- can be referred both to the Gichkis and to the Khan of Kalat. The former in fact were transformed from “invaders" into “invaded" when they found themselves having to bargain with the Kalat rather than risk forfeiting their local supremacy. For its part, the Khan of Kalat adopted towards the Gichkis the same attitude as the latter had assumed at the beginning towards the local “rulers”. He appointed them to be his rep-
It is
interesting to note,
velli
but took care also to reinforce those groups which, tradi- opponents to the Gichkis, could counterbalance the latters’ leanings towards autonomy (as with the Khaudai of Dasht
resentatives tionally
possible
mentioned before). The case of the Gichkis therefore, reminds us not so much of the cyclical succession of the tribal elites as it appears in the work of Ibn Khaldun; nor of what hobbesians assumewith regard to the delega- tion of power to a third party. Gitchi’s destiny reminds us rather that they gained power by the use of arms, like the man who, according to Machiavelli, acquired a飞ew principality by his own arms and prow- powerful ess”(1961: 49), but lost it when they resorted to the aid of a stranger. Asking for help from a powerful stranger, maintains M号chia- 才elli, is always dangerous. Speaking in fact, of auxiliary troops he in ract wntes: to ... are involved when you call upon a powerful state to come your defence and assistance ... In themselves, auxiliary forces can prove useful and reliable, but for those who recruit their services they are almost always
Auxiliaries
disaster. You are left in the lurch if山y are defeated, and in their they are victorious (Machiavelli 1961: 83). a.
power
if
88
Ethnography
at the
History and Models
Frontier
89
mentioned the possible limits of this reading of the ethnographic and on the presence of potentates in southern Makran at the beginning of the chapter: decontextualisation of the model and mod-
terms of a Khaldunian or Hobbesian model, we would be looking at societies “blocked" by a logic refractory to any interaction with the
elling of real events.
fact that
I
historical data
Can we now see its
advantages?
It is, as we have seen, a reading different to that which might have been conducted through the application of models derived from Ibn Khaldun and Hobbes. This difference consists, essentially, in the pos-
by following Machiavelli’s indications, of taking into consideration factors such as: the use of militarv force and intrusion from abroad as elements essential to the constitution of social forma- sibility acquired,
tions of this kind. This latter fact has enabled us to account for the existence of a power that cannot be explained either through the Kha!-
dunian model or through that inspired by the Hobbesian pretence, since those models do not take into account certain imoortant ethno- graphic and historical data. The use of the Khaldunia口 model seems inadequate to the extent that, in our context, it fails to meet the condi- ?ions which might justify its application: the presence of “tribal刊e between nomads and settled communities. As regards the Hobbesian model on the other hand, it must be stressed that, although it points up the role played by the hakim in reducing the political and ecologi- cal imbalance influencing the relations between nomads and settlers, its application does not clarify the question of the social origins of those - holding and wielding power within the local potentates. A further advantage of the reference to Machiavelli’s model is that while the application of Khaldunian and Hobbesian models is trans- lated into the construction of contexts that are“closed”and, so to speak, folded onto themselves, Machiavelli’s model enables us to con- sider the “social formations" in question as“open”entities. Instead of being subjected solely to the cyclical dynamics of their internal repro-
duction, these societies were“immersed”in the flow of events which, as such, acquire form and meaning as a result of the wider system of social, political and economic relations to which these societies belong. That system transcends local reality and corresponds to a broader one of social formations, with which the communities of southern Balo-
chistan entered into an“interactive”type of relationship. If the char- acteristic dynamic of the communities in question were represented in
the Khaldunian model, the image obtained is in marked by a constant and cyclical renewal of settled elites by nomadic elites. Whereas in the case of the Hobbesian model, the structure of these societies would be the product of a kind of “eter- nal conflict" between two communities, the nomadic and the settled. outside. In the case of
of a world
In both cases these are
homeostatic models, indebted to a theory of bal-
by a structural-functionalisi perspective. adoption of a perspective inspired by the suggestions contained in Machiavelli’s The Prince allows us to grasp the histori- cal dimension that links these communities to other social formations, characterised by different forms of social, political and economic rela- tions. In the case in point, by bringing the communities of southern ance inspired
Instead, the
Makran into a dialectic relation with other
“societies”,namely
with
para-statutal formations, this perspective enables us to explain their instability, conflict and internal strati岳cations.
statutal or
the explanatory power of a model depends structure as on the context of its application. The Khaldunian and Hobbesian models are here applied to contexts which, if those models are to function, must necessarily be represented as simple and closed contexts. They are contexts devoid of contacts with the outside which, as we have said, find the explanation for their exist- ence in themselves. However, societies must not live isolated from one 飞nother, but immersed in the flow of history which, beyond the event,
In this case too, therefore,
not so
much on its
presented as an interconnection between systems. Which, in their turn, instead of being “simple”,are “complex”and “open”. It is there- fore only by bearing in mind complexity and openness that can 15
we
以plain phenomena which, otherwise intended (i.e. as elements belong- ?ng to simple and closed systems), turn out to be inexplicable when sub- Ject to
the "ethnographic
As far as !
test”・
am concen叫the adoption of a model inspired by the ideas
to prove and explain the seizing of power by by one or two other similar groups with reference to
of Machiavelli served the Gichkis and
90
Ethnography
at the
History and Models
Frontier
period and in relation to equally particular political circum- which saw the formation of mercenan; armies in India between the 15th and the 18th centuries. This hi…rical contextual- ization can, I believe, clear away the impression that the adoption of a perspective of this type may end up reinstating, albeit in a different form, the idea of a repetitive and cyclical historical process formed by a given
stances: those
牟
the alternation of conquering foreign elites. Generally speaking, the resumption of “classic”models in ethno- anthropological analysis, and likewise the application of these same 1:1odels to situations historically and geographically remote from the contexts of their original development (14th-century Maghreb, 17th『century England and Renaissance Italy) cannot be rejected only because these models would be “decontextualized”. Their possible,
and legitima风applicatio叫epends on the fact that those m;dels are, for better or worse, part of the conceptual equipment at our disposal when dealing with matters similar to those discussed in this chapter. These models are in fact part of our act叫faculties of comprehen;ion of realities that are for山most part extraneous to us, and 飞ven if出 comprehension
is, in large measure, a pre-comprehension (at least in ?he sense ?hatγe seewl四we can see through our own categories), the tact remams that we must nevertheless try to comprehend something. I could not, of course, inform friends of all these things. They
my
recounted. Whilst I “modelled”what they told me. So the whole of this Machiavelli business remained exclusively in my head, without my hosts ever having been informed as to who Machiavelli or. still less 平lentino, was. This is a case in which, I admit, ethnography叫1ed he greatest possible distance “from the ooint of view" of mv inter- locuto低But it was a risk I had to accept. Fo川hilst it is true th川1e application of models always entails the risk of distancing our objects from tho_se contributing to keep them alive, to give them up would mean los1吨sight of one of山essential dimensi。ns of anthropologi- cal knowledge, namely its fundamental comparative inspiration. The pow二r o? m?dels (especially if山se are compared with one another) consists their capacity to let us“see new connections”. in their met-
m
power, as Max Blac』c maintain To put it blun anthropology, they help us to understand realities of which we cannot apl阳orical
91
an experience which, even if we did have both in its complexity and in understood it, could in any case not be application of models corre- the of view its generality. From this point have a true direct experience;
sponds to the construction of representations and, in the case in point, of ethnographicγepγesentations. That these representations may vary according to the models that we apply to that co?text is an established fact. This does not mean that all representations are“equal”;on the contrary, some representations exist which are more convi口cing and others that are less so. The real issue, then, is knowing how to recognise the “impact”of models, their capac- ity to produce “likely”representations and interpretations of what we
Now
want to know. I believe that capacity consists more in managing to explain the variety of these elements than in ignoring them, as on the contrary other models seemed to me to have done.
5
Blood and Force serve as my text - my jumping-off point into a variety of assertionsτvhich, with it as base and background, I hope to have accepted as relating in some responsible way to a certain peculiar social reality I have had some access to but most of my readers will have not. It will
Clifford Geertz,
Found in
开・anslation:
On the Social History of the Moral Imagination,
1977
It goes without saying that, since Foucault, power has been everywhere. The fact remains, however, that power is continually subjected to com- ment and narrated within “emblematic" stories. These, and many other things, contribute to the formation and reproduction of a“commu-
nity of
memory”
Southern B二lochistan is a land rich in stories and legends passed down. During discussions with my hosts and friends, I asked them questions which they answered not only by giving their interpretation of the facts, but by’telling stories that for them exemplified the sub- of our conversations. Thus whenever I tried to look deeper into question of social strati岳cation and its transformations, I would be told some“story”or other which, in the eyes of my interlocutors, 牛plained the stat? of things past or present. From the mesh of ques- t1ons and answers there nearly always in fact emerged “a story”,which
jects
the
showed the good will, interest, and certainly also the desire to make things clearer to themselves, on the part of my friends in their desire to“clarify" their point of view. Whilst however for my friends and interlocutors the issue was reduced to narrating the origin of something (in this case of the groups 阳hadm地the吨io的history), my own point of view was that o?. someo?e wishing, in the light of these stories, to find out what “game" was hemg played by those who, to use a very generic expression, I would
94
Ethnography at the Frontier
Blood and Force
de岳ne as “the symbolic elements peculiar to a cultural context”. These might be seen as“apriorisms”,areas of meanin2: constructed (histor- icall抄around terms or concepts good to be th…ght of, beca…问 are good for thinking. In this case good for thinking about the dimen- ?ion of power. Thus, having diligently“prepared himself气as was often his custom before answering my questions, one day Aqil showed me a book in Urdu, whose author had in the early twentieth century col- lected and transcribed from the Persian a story precisely about the sub- ject we had been discussing.“Found in translation”、the story was that ?f the disappearance of a power that had for centu二ies re…ed a州n large measure shaped that society which, after the permanent incorpo-
ration of the region into the Pakistani state thi caught up in the whirl of social change, political dissent against the cen- 叫ized state, the worsening of religious issues and the ev?r more acute gap between the state of infrastructure and the imoatient and disorien- tated imaginary of the younger generations. J.
As
we are tal』cing about an area of stratified com ?斗nities that derive much of their present physiognomy from relation I
巳ave said hi阳・to,
95
conquerors originating from India; c) the Bu Said (or Buleidi), who came from Arabia in a period shortly prior to the facts narrated; d) a beggar; e) the lord of Kech, Malik Mirza; and f) two young Gichkis. I shall introduce, in that order, the story as such; then a socio- cultural comment on the story itself, and finally an interpretation of it in relation to the current political situation and to the question of the production of local identity. The story runs as follows: Gichkis, invaders
The Gichkis (warrior conquerors) came from India. They were Rajaput. There were forty of them. They were horsemen. They stopped at Panjgur. One day, while they were shooting at a target, one of their arrows struck a passing beggar in the foot. After his wound had healed, the beggar went to the lord of Kech, Malik Mirza, and asked to be compensated for the wicked- ness of the Gichkis who, he said, were not even Muslims. And so Malik Mirza slaughtered the Gichkis, with the exception of two little boys. The boys were hidden by the Baloch, who claimed they were their own children. Thus the two boys went to live with the (nomadic) Zahari tribe in Jahlawan (a desert region of eastern Balochistan).
When Malik Mirza came to hear of this, he took pity on the two boys and gave them an area of Panjgur, called the
‘valley
of the Gichkis’,as the price of blood
maintained over山lo吨period with“exterr " poli忱ical forma tions. Among the latter, in chronological order, was ;he Khanate of Kalat, the Anglo-Indian empire, and the Pakistani state.
(bah a,‘blood’and ‘price of blood’). here the two boys took Balochi wives. The son of Malik Mirza’s son [also called Malik Mirza], then became lord of the Kech. At that time the Bu Said arrived (from Arabia) and settled in Buleida.
in this soecific case, er巾ryev创V町well』known locally, but today
Gichkis, the Bu Said came to an agreement with the latter to eliminate Malik Mirza, and to divide the region into two parts: Panjgur for the Gichkis and
S干1ps
By “production of historical discourse" I mean,
comments on a semi- ?artly contested in its official (textualized) version. The event refers to 山seizing o? power, during the seventeenth century AD, by a wa俨 nor group onginating from eastern regions that are today part of India. In order better to describe the setting of the events narrated in this story, I should mention that Makran is an arid re2:ion in which how- ever two mportant and densely populated agricultural areas are situ- f ated: PanJgur in the north and Kech in the south. T与main characters i叫is semi均e灿・y tale, a bit like the actors n a play, are: a) the Baloch, in other words the “autochthons"1; b) the 1
Balochistan, and especially its southern part, Makran, is a region in which groups of山most diverse origins came together: Indian, centr?l Asi矶East
When they heard
And
of the struggle between the once lord Malik Mirza and the
?echfor山Buleidi. Accordi时y, they killed Malik Mirza and叩lit the region
mtwo.
The Gichkis however demanded an extra part. Kech was larger than Panjgur, whilst they had grown more powerful. This was what Mar Singh, leader of the Gichkis, told the
Bu
Said.品
The Bu Said did not accept, and so began the struggle between the Gichkis and the
Bu Said.
Arab and Iranian. so that the notion of“autochthony" refers exclusively di…nsion which the locals have of th;ir relationship with territory. Balochi is a language in the Iranic group, as are Farsi, Kurdish
African,
to the self-perceptive
the
晶
and Dari.
96
Ethnography at the Frontier
Blood and Force
The story concludes with victory by the Gichkis who, from the time to which the events narrated by the story (the seventeenth century) alleg- edly refer, and up till the middle of the twentieth、maintained. amidst al阳nating fortun叭a strong influence over a la咿part of M;k肌 Th气interest of the story
lies
in the fact that
it all
appears to
rotate
around two notions: baha (blood) and dawa (force): notions to which the Baloch assign (in a manner as explicit as it is implicit) a fundamen- shaping their society. Since they are two elements pivotal to their social and political life, the Baloch have made them a; many tal role in
points of reference in their political vision which is, for this very reason, en?nciated in the narration of the historical origins of local p卢wer. Or, to be more exact, of the origin of the power of those who wielded it up till the arrival of the British first and of the Pakistani state later. In the first part of what follows I shall comment on the story as told from the text
found in translation.
it may be said that combined in the story are a historical vision, a political conception, and a form of social ethics.
Generally speaking,
Th二 story stresses first of all the Rajaput origin of the conquerors. The R句句ut, as we have said in an earlier chapter, are a larg乌 ethnic-lin- 乎u1st1c gro_up origir叫ly from Rajastan (now eastern Indi动from which tor centuries mercenary troops had been recruited by kingdoms and potentates of the Indian subcontinent. The Gichkis asserted themselves in .?akran between山seventeenth and eighteenth centuries. So their ongms can be retraced to the same as those of the mercenary bands th凭ope旦时on their own and were engaged, during that p卢iod, in raids agamst settled farmers far from the influence of the centralised political formations controlled by Hindu rajas and by Muslim nawabs. The fact that the conquerors were“fortv" refers to the function
serve1 by邮num以出Persian, and mo卢generally Indo- Iranian raditwn, of conveymg an idea of multitude and force, as well as beauty and splendour.
The Gichkis“were riderγ,inasmuch as thev were mobile and pred- 斗。与The .fact that in the story出ey are pres?nted as horsemen-may ?ignny an mtention to emphasize their nomadic and bellicose spirit, contrast to the sedentary and soi-disant “peaceful" character of the
m
97
is perhaps also meant to allude were no women among them, and therefore that they were strangers having no fixed and lasting ties with the local society, hence in a rather ambiguous situation. They appear separated
farmers.
This characteristic of theirs
to the fact
that there
from the autochthons, militarily strong but socially barren, marginal and “incomplete”.
Panjgur is the area in which the conquerors ended up when and also the point of departure for their incursions first and for their actual expansion later. Finally,
they arrived in the region,
The Gichkis’pastimes were in keeping with their profession. Their entertainment was archery. But they wounded a poor man, spilling blood. Why precisely a poor man? It is known that predators did not spare the most destitute strata of local populations. Nor have the Baloch forgotten the ferocity with which the dominators of the past compelled their subjects to pay taxes and dues,岳rst to their advantage and later to the benefit of those to whom they had been forced to yield in order to become their “representatives" at a local level2. The appeal lodged
by the beggar to the lord of Kech, and the words used by him,
region was characterised by the presence of a more cen- power than that exerted in Panjgur. Also, it stresses the invad- ers’adherence to a religion (Hindu?) different to that practised locally.
signal that this tralised
The reaction from Malik Mirza, then lord of Kech, was quickly man- ifested: he exterminated all the Gichkis, but unknowingly spared two children. It was from these two that the future lords were to descend, 飞nd it is thanks to them that a new political order was to be established
m the region.
On the basis of the moral principle whereby the enemy’s orphans must be spared and, if possible protected, the Baloch therefore adopted the boys and, for a question of “honour”(izzat), made a false statemen t ?he1 erl创rd Ir o仰d山ho严in is wa) e Baloch inc 〔? →e1r community二and this act marked a decisive point in the relations 2
refer to the subjection of the Gichkis to the Khan of Kalat, who in the mid- eighteenth century imposed his rule by force over eastern Makran.
I
Blood and Force
Ethnography at the Frontier
98
between the conquerors and the autochthons. In fact the incorpora- tion of the two young Gichkis into the local community sanctio卢ed a relationship founded on the dominion of one side and the consensus
and support of the other; or, to put it more precisely, of one part of the Balo ch 3. It is known that the Gichkis established privileged links on the political and economic, as well as marriage level, with a number of high』ranking Balochi groups. During this spell spent by the two young Gichkis in the desert is possible to see something resembling a liminal period experienced with nomads before returning, as settlers and founders of a conquer- ing dynasty. In effect, the protection accorded to the young men by the Baloch may be interpreted as a political allegory, expressing the alliance between the conquerors and the locals. But it is quite possi- bly also a cultural allegory. Adoption is in fact the device whereby the future rulers were to be incorporated into local history. It was during this (liminal) period spent in the desert that the two young men (a kind of metonymy of the Gichkis) lost their traits as “foreigners”. In 山desert, and during山years of cohabitation with the Baloch, they learnt :heir language, customs and lifestyle (razm) and, although the story does not say so but gives it, as we shall see, to be understood, they were converted to the local version of Islam, the zikr. 4 it
When Malik Mirza, lord of Kech, came to hear of these facts, he could not vent his wrath on the two boys. He was prevented from doing so by moral reasons. The razm-e-baloch, or customary law, honour (izzat) prevented him from persecuting the two young orphans and sons of an obliterated enemy. Malik Mirza did not merely spare the boys' lives, he went further. In keeping with the principle whereby he who has killed shall
pay
Although today the Baloch insist on their common origins, some groups claim to have non-Baloch ancestors. This “contradiction”is explained by the fact that whilst to the outside world (Pakistani government, foreign guests, etc) they tend ?o present themselves as“cultu町rally and 1山istorically homogen u
within their own sphere they tend instead to maintain active criteria of politico・
which refer to the diversity of their respective origin;. Those to high-ranking groups usually claim Arab ?r Persia川escent, or to descend from Balochi groups originating from different regions of Makran. In the past, and up till the beginning of the z。th century, some of these groups social differences
belo叩ng
of吁0吨n" origin, though not particula句阳阳。叭;cquir讪high status
and a politically relevant po白山n by acting as representa?ives (wakil) of the Gichkis. Today, some of their most prominent members are among those most involve斗
4
in卢ambivalent relations
the Pak1stam state. Zikr is the name given to the shi'ia.
The
cult dates
that li此lo叫reality to the s?川阳时of
from Ismailism. hence of a “branch”of from the work of Muh…?ad Jaunpu凡a preacher
a derivation
a “blood
price" (baha), to the victim’s family (so as to avoid
makes them a gift, by way of compensation (baha) of a part of the Panjgur oasis, and to be exact, an area also known today as
vendetta, bir),
the Gichkis". of Kech legitimises the Gichkis’rule over Pan- act dictated by a sense jgur, hence their right to settle in Makran. of guilt, honour, or fear of a future vendetta? Aside from these moti- vations, a gesture of this kind has a precise social meaning: the baha (“blood price”) paid reflects a political “equivalence”between the very Gichkis, by now firmly installed in the region, and the locals. important fact is that the two persons that founded the future dynasty adopted by the nomadic Baloch were not two young people of differ- ent sexes, but two males. They did not therefore give rise to the Gichkis community by joining with.them, but by “mixing" their blood with the “valley of
In this
way the lord
An
A
that of the
Baloch.
Thev in fact took their wives from the latter, thereby
we might
say, ・the“rite of passage" that legitimized their per- manent incorporation into the local community and inaugurating a new political scene, founded, precisely, upon the accord between con- querors and (a part of the) locals. By means of this step in the story, historical imagination integrates },he Gi州s with local society, put叫them on a level of c1川d homogeneity" with the Baloch. This is not however simply assimila- closing,
3
99
In fact this episode, besides reflecting the emergence of a new polit- ico-social locality characterised by complex relationships of dominion ?ion.
from the Penjab but active in the Farghana valley, in Afghanistan 16th century. Today the zikr seems to have a following of about 20% of :he local population, whilst in the 18th century it accounted for the vast maiority of Muslims present in Makran (in which Hindu and other religious
?riginating
m the
groups also lived). Their
main sanctuary lies
in
Tamba, not far from Turbat.
Ethnography at the Frontier
100
Blood and Force
which see a part of the Baloch themselves participating in the manage- ment of power, also sanctions the hypergamy of the Gichkis themselves
The status of “takers of wives”was for a lon2: time characteristic of the Gichkis and to this day its social “echoes”巳an be
vis-a-vis the Baloch. felt,
albeit attenuated
by the distance
in time, in the communities
of
the region.
Another important point. The marriages contracted by young Gich- with Baloch women concord with their conversion to the local5
kis
Islam.
Whilst the Gichkis reinforced themselves in Panjgur and extended their
dominion from the valley that bears their name, the Kech came under the control of Malik Mirza’s grandson (who bears the same name as that of his grandfather). But in the meantime the Bu Said arrived from
Arabia and settled in Buleida, a secondary valley of northern Kech. Like the Gichkis, the Bu Said also came to be included in local soci- ety, seeing that they were to adopt their language and customs, whilst also taking the
name Buleidi, after that of the place in which they had
settled.6
Let us now consider the elimination of the lord of Kech, Malik Mirza’s grandson who had donated a part of his rule to the young Gichkis as the price of blood. Even nowadays, the Baloch maintain that payment of the haha does not constitute an act capable of absolutely guarantee・
?ng卢eventuality of a v叫etta (bir), although to take reve咿after havmg received the baha means failing to keep a point of honour (izzat). In any case the story tells us that Gichkis and Buleidi allied them・ selves to shake off the power of Malik Mirza’s grandson. Nevertheless 5
6
The Gichkis, before they were subdued
in the 18th century by the Khan of Kalat, were zikri. It was precisely Gichkis’adhesion to the zikr that was adopted by the Khan of Kalat (orthodox Sunnite) as a motivation for waging war (jihad) against them. The Buleidi, descendants of a certain Bu Said according to tradition, are in rea.lity of uncertain o吨in. The story that we are examining considers them to have come from southern Arabia (Oman but it is possible that they may in reality have come from the Afgl阳时ion of
Helm卢
101
was not only conquered, but also physically eliminated. His murder might also be read as a vendetta against him by the Gichkis. However, the story clearly says that it was the Buleidi, and not the Gichkis, who killed him. The fact of attributing to the Buleidi the elimination of Malik Mirza might constitute an expedient to take the blame off the Gichkis (and we shall be seeing why) precisely when, by allying themselves to the Buleidi, they seem to have failed to keep a point of honour. The struggle against the lord of Kech concluded with the partition- ing of Makran: the region of Panjgur going to the Gichkis, and that of Kech to the Buleidi. the latter
Having become stronger than their Buleidi allies, the Gichkis took possession not only of Panjgur, but also of Kech, hence of the whole d则ern Makran. In this passage of the story might be read a fur- ther step towards the de岳nitive exculpation of the Gichkis for having slain Malik Mirza, and therefore for having disobeyed a fundamental point of local custom. In fact the final struggle against the earlier allies, the Buleidi, seems to redeem them from having been in reality accom- plices to the latter in the death of Malik Mir?a, an evidently “politi- cally thinkable" but “morallv execrable" fact for the Baloch. Thus the nar:ation concludes with the struggle that permanently sanctioned the takmg of power by the Gichkis, who succeeded in their intent thanks to the
'W_e
use of force (dawa).7
come
now to
the interpretation
of山story in
the
cont创ofth
s1tuat1on t<、cl组v -一,___
.,.,
This story ;eems to reveal, I was saying, the symbolic centrality of two key elements in the process of formation of political power and
7
The struggle be臼nGichl山is and Buleidi is p川ally doct削ed ir
st
ica !
terms (cf B.D.G. r9。6, VII: 48-49). At one point, the Buleidi were supported by 闪ad?r, the then Shah of Persia. I川1at period (first half of the 18th century)
Nadir aimed in fact to extend his influence towards India、while counting on alliances with a number of local potenta邸,such as that ;f the Buleidi.
Ethnography at the Frontier
102
success: blood and force, baha and force are we actually talking about?
Blood and Force
dawa. But what blood and
what
If the force always appears in the form of the irruption of violence into political dealings, the blood is not only that spilt in times of con- flict.
Intl
103
on the existence of dominant groups politically. Nor does it explain the existence of institutions, mental attitudes and behaviours that are nev-
in a hidden form, in the narration itself. The power possible through the symbols of what founds narration makes Power is not only force and violence. It is also continuity, stability, ertheless present, albeit
it.
for a crime
committed, and as the seal of alliance between Gichkis and Baloch, and, finally, as a key to the “naturalisation" of the latter. Whilst force and violence always go with blood, the opposite is not 如case. In effec飞blood does not appear only as a nψtive element. In the story,
serves as a metaphor for the inclusion of foreigners in local was in fact thanks to a “mixture of blood”bet亏een Gichkis and _Baloch that the political unity of the region was achieved, inas- much as it was after the union of Gichkis m?n with Balochi women that the invaders were able legitimately to settle in the region as soon as :h?y had adopted the local language, u吨es and religi马1. 1t 1s moreover a fact that for the Baloch. blood is a ootent metaphor it
society. It
for desce叽乌r the genealogical continuity of a group, for i臼}…oge- neity“of sentiment”,and for its cohesion in the face of external threat. And it is always blood that has to be paid (or repaid) once it has been spilt. As for the dawa, force, this appears in the story whenever there is a change in 山political scene. Everything happ.,ens as if imagina- tion could not go so far as to think of political transformations as
anything but relationships “of force" that change violent! and with blood which, although spilt, is always expec时n one way 。r ano阳 to repair the damage caused by the use of force. If we try to isolate the 子rucial phases of this stor民we find that blood and for;e are perfectly integrated: where force prevails, blood always comes to re-establish the broken order. .
From this point of view the text of the story and delimits a context, a discourse in which it is possible to rec『 ognise a specific“imaginative universe" that rotates around the cen- tral symbols of the story - blood and force - which are terms good to be thought of and to think both of change and continuity, disruption and stability. The story therefore becomes a context in which the eth- nographer can try to read the interconnection between mental atti- tudes, behaviour and institutions, just as these are thought of by local historical imagination. For the ethnographer the story becomes a text in which someone has put into the form of a“historical" discourse certain processes of conceiving what I have called “areas of meaning”: consensus, discourse etc. defines
ways of alluding to power, political dynamics, descent, violence, passion, reciprocity,
A
a reflection of reality. narration of this kind mav on the other hand allow us to understand the way in which a commu卢it
you like,“in了ents': th? facts ?fits own l让istory and its与wn“culture means of an 1magmat1ve act1vitγthat starts from a dimension of expe・ T
1
An
rience. activity of this kind 'makes the narration itself an object of reflection for social tho吨ht. The narration does not dwell parti?ula向
com-
etc.
terms like baha and dawa, blood and force, what I have called meaning" have been historically built up, the way in which that happened is not neutral to be sure, but dependent upon a specific vision of power and of history itself. In fact, as I have said, this version If around
“areas
of the
of
story
today, at least in part, disputed. we have seen, highlights the foreign origins of the Gic.hkis, how p叫g盯is the place where they settled at the time of their arrival in Makran and from which they launched their expansion. According to the story, furthermore, it is precisely there that they had
The
is
story, as
an area in token of compensation (baha) which had taken - name from them. received
Here it is not an interpretation of the story that makes the latter simply
honour,
its
It is precisely these passages of the story that are today called into ques阳1 by those, in a mounting nationalistic climate, who ten斗 to convey a“unitary'’vision of the Baloch which ignores, or attaches ?nly secondary i卢por阳ce, to the differences of o吨in and status of the groups making up the local population. Talking about this story, while we were translating and commenting on its passages, not even
Blood and Force
Ethnography at the Frontier
104
Aqil and my other friends seemed too convinced that this was its whole truth. As in all our discussions, Aqil displayed careful consideration and a critical spirit. But others, politically more ardent than he, vehe- mently denied that the story had any grounds for truth. The fact is that the Rajaput origin of the Gichkis, openly recognised by all up till some twenty or so years ago, is today often denied to the a_dvantage of a presentation of them as an autochthonous group. Hence the current tendency which consists in considering the Gichkis to have originated from the Gichk valley二and not as foreie:ners after whom the - valley in question allegedly got its name. This “politics of memory”,which reverses the order of precedence between the name of the group and the name of the place, -can be ade- quately grasped only if considered in relation to the示rei丘ht whichsto li牛e ?his 1aent1ty川
i与s
car巧时ay in川the‘in吨ging ir
discussi旷of Balo C
ll i
.
牛ltl
the memory and reality of a very marked strati岳cation, and although this strati岳cation today also performs a major role in the reproduc- tion of social relations, the hierarchic factor, is minimized、when not actually denied whenever the discourse refers to an extra-local dim削en sion. The motivations for this attitude are naturallv diverse: the lower levels of so it
country an opportunity for upward social mobility; the upper levels think of a largely autonomous Balochistan in its relations to the cen- tral Pal飞ista?i government whilst cherishi吨the hope of occupying, or pre?ervmg, a coun町capable of entering into a dynamic of mo斗’ 可nisa?ion detached from the logic of the central Pakistani state, a pos1- t1on ot pre-eminence on the political and economic level, in many cases wi牛飞he prospect of being able to boost a status trad让io叫ly consid- erea supenor”.
m
..
Whatever the motivations of the diverse comoonents of local society, all these see in the compactness of Balochi“c?lture”and society the most unimpea干hable argument for legitimising their claims in relation to non-Baloch1 powers and figures. But social and cultural homogene・ ity can hardly ever be thought of outside an idea of common origin, as
105
day by the theses upheld by the nationalists. In a society that makes descent the fundamental principle of social cohe-
demonstrated to this
8
taken for granted. The Makran Baloch know well however that their society was formed after multiple superimpositions of groups with different ori- gins, and that the vicissitudes of the Gichkis are part of a long story made up of immi?rations and emigrations, superimpositions, coales- cences and inclusions as well as exclusions9. In short a case - a very important one - among others. Nevertheless today this awareness does not prevent them from thinking of themselves as a single homo- geneous group that has to reduce to the minimum every form of dif- ferentiation within it, so as to be able to claim its autonomy from the Pakistani state. A proof of this fact is the relatively recent tendency that consists of giving children Balochi and not Arab (“Muslim”) names. At the same time, it consists of reproducing on documents not the name of the patronymic group to which they belong (zat) but, very simply the wording “Baloch”. Naturally the Baloch are far from convincing. Their internal divisions, especially those among the m句or tribal sardars in the north of the country, in fact indicate the oppo- site. On a symbolic level, however, things point otherwise. Asserting a common identity is equivalent, amongst other things, to affirming a unitary“tradition”,and that in its turn entails the need to affirm a shared history. Thus, alongside the principle of common origins, e:pressed by the notion of“blood”(baha), there subsists the principle sion, this fact is
01 a
common history, studded with a series of violent political events by the notion of “force" (dau明). This alone however would
expressed not be
enough to fully describe the coalescence of different groups
with equally different origins. 8
Cf. the last
9
In this regard
chapter of this book. it
must be remembered that before the “partition" of 1947 numbers of Hindus had lived in Makran. (a group derived from the shi’ia who look upon the
between ??kistan and India, large The Ismaihtes themselves 牛gha
Khan as their spiritual guide), who in tl盯川}叫beenf幻向阳阳ous
?:!??li;a? le::?:??::!£:::・:?i::i?r?t:J on the Arabian Sea coast.
:;,?;;;:;:;E,h;f点:?;,
Ethnography at the Frontier
106
6 In a society like that of the Baloch, whose language first began to be written in the 1930s (using Urdu characters derived from Persian and still earlier from Arabic), the historical tales passed down orally havea political value that largely supersedes a reading inspired by a ,;folklor- ish”type of approach. Since these are concrete examples of the “bring- ing into discussion" of a historical vision, the storie; are subiected to periodical revisions in the light of the political context in which they
Nation and
Memory
L'oubli, et je dirai meme l'erreur histonque, sont un facteur essentiel de la creation d’une nation ..
Ernst Rena民Qu'est-ce qu'une nation? 18821
are used10.
Just as the text of the story that we have taken into consideration might be seen as the “normalised”narrative version of a oeriod in which the Gichkis were firmly in power over the region, s马the cur- ?ent :7ariants of that story
may be read as
m relation to present interests.
attempts to -readjust history
When I first arrived in
appeared immediately untidy, of its subsoil), overtly hostile to central power a叫despite the various su卢schools, quite sceptical about religious matters. Unlike what I had found towards the end of the 1970s in Arabia (a kind of “Middle Eastern Switzerland" in the ironic words of a colleague of mine then), southern Balochistan was (few know, per-
poor (despite the
haps,
gual
1987, Balochistan
enormous wealth
what it is today) a remote province - a “frontier”一of a multilin- and multiethnic state where the internal tensions of a local soci-
very busy transforming itself, were
e_ty
added to and intertwined with
}he contrasts between Baloch and Pashtu (in the north), Baloch and eastern”ethnic groups (Punjabi), and between Baloch and the central gover斗ment. All against the background of the Afghan war, which had witnessed the involvement in the first instance of northern tribal
groups and the Soviet
army, and then Iran and Pakistan itself, the and the emerging Islamic republics of central Asia; and lastlr, to the almost total ignorance of all (including myself), of the dis-
United States 丁urb
and their
allies was accompanied by a widespread liking for the who were seen in any case as enemies of Islamabad (allied t 0 the 巳SA)叫山refore as possible wedges in the edi岳ce of山Paki-
ica?s
Soviets,
10干ikewise,山“sto旷of the Indo-Eu叫ean o时ns of山B山h,dev仰d certain 1叫biased interpretat叫speaks for itself. This bias tak叫 :…干 ne migratory and expansionist theories of the 19th-20th century European orientalists for the purpose of establishing a wide rift between the Baloch and the other ethno-linguistic groups present in Pakistan today, notably the Panjabi
(cf.
the last chapter of this book).
stam state, 1
from which the Baloch seemed anxious to remain
:IF吨tful阳,叫I shall eve叩y historical erro巳fonr the
creation of a nation." Ernest R芒nan.“What !dent町Cla衍sic Readi1ψedite by Oxf,叫UK, Blackwell 2。斗
is
Vir川P
as aloof
Nations and Peco肌M二lden, MA, and
a Nation?”,’in
Nation and Memory
Ethnography at the Frontier
108
A
as possible. vague nostalgia for the British lingered among- the older generations (the British having left in 1947). When I arrived in Makran, I landed bang in the midst of a fear-
some anti-government demonstration by the BSO, Balochistan Student Organisation, which was secular,“Marxist" and openly anti-American.
These were not perhaps yet the times in which anti-Americanism was to coincide with “anti-Westernism". I recall however a kind of (highly informal) interrogation to which I was subjected by some twenty or so BSO activists regarding my“views on the state of the world”. They were trying, basically, to find out whether I was in the service of the Pakistani government. Alam and Aqil, who had welcomed me a few davs earlier into their home (or rather, into tl时10me of their wives) were' clearly on my帅- They already “believed”in my good faith and so in them, who enjoyed a fairly high reputation locally, I had useful allies. In the end I passed the test. Having thus passed my exams with the activists (ad10ngst whom there may have been a government informer), I was exempted from h:ving to deal with any“problem”that might have upset all my
research But, sho盯ly after I had settled in with Alam and Aqil, I unde附n an attempted“suspension”by the supervisor of the archaeological mis・
sion, whose headquarters was several kilometres away from “my vii- lage”The supervisor is an official from the Ministry of Culture, an.d keeps w_atch on the archaeological missions to make sure nothing removed or improperly used by foreign researchers. Being attached to that archaeological mission, I too therefore came under his scrutiny. With a series of “prudential;, lines of argument, the supervisor, a Pun・ jabi and shy little chap, a “family man" separated from his family by ?1°re than a thousand kilometres, and who looked upon his experience Makran as a punishment, told me it was“too dan1!erous to work in 山t village气that all things- consider叫“it was not ;,.o仙risking my life :o study the‘cultur 1s
m
斗巳?干c叭wi山10 fai向he was a Christian!), and host to progress and c1v1hsat1?n”(a chorus which many anthropologists must have heard repeated mcreasingly often in recent years by the authorities of the countries they work in). He told me, in short, that I was should pack
109
my bags, leave the savages and their bizarre habits to rejoin the archae- ological mission, true science (he had not understood what my profes- sion actually was), and real civilisation. Obviously the first thing I did was to inform my hosts of this embarrassing situation, omitting to mention the supervisor’s opinions of them. There was no need to do so. Indeed something unexpected happened. The Baloch of “my" village rebelled, through the local polit- ical authorities (and probably having arranged for the supervisor to one or two“warnings"). They invited him to a council and, told him in no uncertain terms that it was inadmissi- ble that a Punjabi should come and poke his nose into Makran. They said that to oblige the Italian to leave was a ploy to make sure nobody could report what the Balochi really were: proud, noble, honest, vir- tuous people, etc etc, and not that band of faithless robbers and louts which the government in Islamabad and its servile hangers-on (Pash- tun, Punjabi and Americans) would have the world believe. Eventu- ally the poor supervisor had to yield and to leave the village amidst the worst insults expressible in an Indo-European language of the Iranian sub-family. In actual fact, the supervisor, as he himself revealed to me some years afterwards when I ran across him in Islamabad, had been for叫to take that line of action after pressure received from two secret service agents. These had asked him to make sure I would not remain in contact with the Baloch. I have recounted this episode because it seems to me significant of the tension which, at the time of mv research, characterised the rela- receive in
my presence,
tion町ween山inhabitants of岛1akran and c?ntral governn all that ensued from those relations in terms of山vattit叫es a?d dis- courses adopted by them towards non-Balochi. In this chapter I would like to explain how the Baloch saw the ques- tlon of their“difference" from other Pakistani groups. Naturally this 月fferen町as叩e叫not in "differential"阳ms but in terms of ictentity, i.e. with reference to what were. for the Balo ch. the historico 了ultura aits that made them different卢the others. Their discourses, rer:ent and, as I have mentioned in the introduction to this book, full .
-
们
and “distances"
among Indo-European peoples, were part
of a
much
Ethnography
110
Nation and
at the Frontier
wider movement known as Balochi nationalism which, from 1960s had taken shape consistently in this frontier region.
the
late
Memory
The most convinced supporter of
forms.
this
111
second perspective
is
A nth “of retracing the of Nations (1986)’ he proposed, for the furpose exami- nations”,a systematic foundations and ethnic roots of modern to the perceptions and attitudes nation of “cultural forms, sentiments, memories, in myths, codified and degree in which they are expressed
gins
In the last quarter of the twentieth century the theme of identit)刊,stated chiefly in terms of ethnicity, received particular attention both from anthropologists and historians. In many cases that common attention was coupled with an interest in the reappearance of nationalism which, as a phenomenon by then considered obsolete in its “classic”form, had begun to assume planetwide dimensions (Gingrich and Banki z。。6). Since it had not spared even the “old Europe" (where it had been believed dead and buried), and due to its appearance in places where it was thought that it could not still, or no longer, manifest itself, this phenomenon led to a convergence of interests on the part of historians and anthropologists. History and anthropology thus found themselves examining the same phenomenon from very close points of view:“S・ tory being enriched by the influences of anthropology (the compara・ tive aspect) and vice versa (attention to the contemporary West). Henc? important works of synthesis produced in the last part of the twentl・ eth century and now regarded as classics in each of the two fields, bear evident signs of the other discipline (Armstrong 1982; Anderson 198J; Gellner 1983; Smith 1986 and others). Anthropologists and historians agree that nationalism, as an ological construct and political movement must be treated as a ide・
rela・
时ly肌ent phenomenon with s1.阳ti ally啃ω旷roots. Whils
however some, like Anderson and Gellner, regard as irrelevant the idea that
it
may be linked to a pre-existent identity sentiment, others,
Armstrong and Smith, claim instead that nationalism is
like
the product 0
{
a pr
would be
whether cultural elements exist which, in c tain circumstances, may engender collective movements and forms of to ascertain
f’
identity defined as“nationalisms”. Supporters of the first perspective maintain that only precise histor・ ical circumstances to be found in the modern West favoured the em?rg・ ing idea of a“national" beli
e
0川he other hand that recent and current natio叫isms are斗 dynamics and symbolic references, of previous idenuty
result, in their
and symbols" (Smith 1986: 53-4).
values
my part, I shall attempt to develop the theme of the connections between nationalism and ethnic identity, in relation to Balochistan. By the very fact of establishing that connection, I therefore accept Smith’s general indication. intent however is not exactly to develop
For
My
an“archaeological”perspective to
on Balochi nationalism in an 乳?t?m?t
common values and symbols
spot
rooted in a pre-existent ethnic ・・
Rather, r町purpos二is to highlight the logic of the national- developed in Balochistan in the second half of the twentieth century, and traces of which were much in evidence at the
ide?tity”. ist
"discourse”,as
my stay in “discourse"
time of
Makran.
with Foucault 1969) a number of enunciations, representations, images and symbols that tend to pro- duce coherencie; of meaning. And which thereby tend to mould, to By
I
mean
(in line
“discourse”which I pro- pose to consider is not therefore only the explicit discourse of nation- alists, consisting of programmatic statements, political appeals and verbal as well as written theorisations; it is also the discourse which consists of opinio叭d??l;?ations and f eeli耶,叫ressed by individu- als not directly engaged in politics but who, albeit only broadly, share the sentiments of the nationalists. The “discourse of Balochi national- give a
shape, the objects they refer to.
The
ism" …e了
it
With a
considers useful in creati吨与n image of the Baloch as a people
common origin, culture and history.
Balochi nationalism is a political, ideological and intellectual move- ment which arose within one of the largest “stateless nations" in South West Asia (estimates of the number of Balochi speakers oscillate between 18 worki?g in the Gulf countries and巧million,
including migrants
and Africa).
The
idea of Balochistan as a“nation”was probably
first
Nation and Memory
Ethnography at the Frontier
112
mooted in an article published in 1933 in the Karachi weekly A!Baloch, the official organ of the Anjuman-e Ittihad-e Balochistan (Organisa- tion for the Unity of Balochistan). According to that article,飞reater Balochistan”(as it is called even today by the nationalists) should have included (see the map below) in addition to Iranian Balochistan Indian (now Pakistani) Balochistan, the Sindh, the Bahawalpur region and Dera Ghaza Khan and Ismail Khan areas, considered the point 0 maximum expansion northwards by the Balochi during the sixteen h cen ry (Inayatullah吵87: 31-32). The idea of a“Grea阳r Balochi was one of the many projects formulated by those who, between 1920 and
{
[
and叨o, had been町i吨to picture山future of山I灿山?con- tinent after a hypothetical departure by the British. The 1930s proiects had no seauel. due to the intervention of the British intelligence who silenced the nationalists, and, after the partition of咧凡to tl川lfi ation (in two stages) of Balochistan by the new state of Pakistan. The 0 nationalistic spirit survived however in the decades that followed, t
noetus in the seventies. 山nalism raised its head、山refore, and developec and movement, in ;esponse to山presence intelle Ld later of the Pakistanis in the region. As an id1 local of Llmoveme瓜it resulted from the卢rging
more gener均,in山hi拟 of山Anglo-I?dian empire and its bo伽叩: mamta1m true、as Ernest Gellner way r other the n?tion and not the
m hat makes that山川io叫ist i削e always 阶川ov山5 stones, world叮ews and social practices existing_fγior to n.??1011 nsist of All these elementsαisting prior to nationa阳飞:sel叫 bols and behaviours which enable subjects to perceive,? (B川1969). ascr ption, as belonging to one and the same “gr up ultaneouslr
sym・
・
・
.
But these values, models,
stories, etc, are
not always
So we could say that the discourse of Balochi nationalism is founded memory-function, which takes elements of the memory- archive and organises them according to a purpose. This perspective on a form of
advantage, over Anthony Smith’s idea, of not having to co?- necessary and organic way the nationalistic sentiment with what he calls “cultural forms, sentiments, attitudes and perceptions expressed and coded in myths, memories, values and symbols”,that is to say, in already oγganised discourses. Within the memorv-function, these ideas, values and behaviours 、trieved”from the 血emory archive interact with other elements has the
nect in a
reasonin2"
make things less abstract, I shall try to show how the memory-archive element; come into the discourse of Balochi nation- of a alism. For its part, this discourse is organised in relation to ideas European origin and as a result of moulding-shaping, as Foucault sa户’ anew“object of reflection": the legitimacy of Balochistan as a possi’ In order to
it is
1
1999).
its
between local identity memories and impor?e-?
a use、whilst
changes of cultural memory, Aleida Assmann in fact called an“amor- phous mass”the accumulated unorganised memories which can be 、sed”only if taken into consideration by a“memory-function" that 、trieves”and bends them, so to speak, to its own purpose (Assmann
Baloch and of the peoples of the Anglo- Indian empire. Playing on the interconnection between these elements arising打om the memory-archive and ideas more or less recently imported, nation- alism organises its memory function in such a way as to bestow coher- 妇cy on its discourse, and fresh meanings which are the peculiarity of
ir二lo二hi丁1is-?orγand,
:tion
Indeed,
history of the
ideas山tl叫arrived with colonial rule. Also amor with colonial dominion、in addition to those conn m-state as developed in Europe during the叫e; d throu2"h the、rorks of western autho叫ma刽l ns
many of them
are, so to speak,“dormant”- or to use Assmann,“stored”in what she herself calls the an expression by Aleida “memory-archive". In her book Erinneγungsraume, on the forms and active.
which, as I was saying, Balochi nationalism had absorbed during and after the colonial era: the idea of a nation, as developed in Europe in the modern age, and the theories of western authors interested in the
fi
l
113
ble nation-state.
In a very general sense、it
can be said that the Balochi nationalists seek
to legitimate
by referring
their clai?s
chiefly to the following points,
Ethnography
114
Nation and Memory
at the Frontier
both the popular and the cultivated idea that the Baloch have of their own past history: a theme closely linked to the question of b) the ethnogenesis of the Baloch people cf the territorial or“fields of discourse”:a)
question and to a lesser extent, to the d) religious factor. Although the Baloch speak a language of “lndo- European"
family
related to Farsi (Persian), their roots are traced back, traditionally,
Mir Hamza, paternal uncle of the Prophet Muhammad. The
to
;efer-
ence to this mythic ancestor, chosen probably as such at the time of the Islamization of Balochistan by the Arabs (in the eleventh century of the Christian era), therefore sets the origins of the Baloch people at the dawn of Islam. The poems and legends passed down orally tell, in effect, a story of conquerors originating from the North West. These would appear in fact to have initiated the occupation of the area today called Balochistan i川he eleventh centurv. which was then com pie时during山岳ftee叫1 century under the l二二dership of the I句end ary“national”hero Mir Chakar Rind. In the course of the “teen th and sixteenth centuries, part of the Sindh and of the Punjab fell, in eff ?ct, ?nder the influence of Balochi chiefdoms, thereby b?osting the 斗at10nalist image of a“Greater Balochistan”. This is an example of how the memory-archive was exploited to foster, in the vision of the early 斗ationalists, the idea of a country whose borders would extend beyond those recognised today. Imagini?g Balochistan as a region inhabited by peoples whose ances・ tors
came trom an area situated between Mesopotamia and
the region
south of the Caspian Sea, theorists of the Balochi nation restated the idea of a conquest, or migration, which is not rare in social cosmologies seekin写to explain both the origin of a people and rule by one group over others. The idea of the ethnogenesis of the Baloch tends in fact to reassert both cultural unity and common geographic origin. This idea of population by groups originating from the west is con・ firmed, in the eyes of cultivated nationalists, by the works of western orientalists who, at the end of the nineteenth century tended to p时’ sent the origins of the Balochi people in a“migrato?y”perspective
(Janmahmad ever,
western
1988: 1凡25; Inayattullah 1987: 33). In their turn, how・ orientalists based their hypotheses on popular Balochi
115
and in particular on oral poetry, which began to be translated into western languages towards the middle of the nineteenth century (Dames 19。4, 19。7). The “migratory thesis" was a product of the historical perspective which in the meantime had developed in the West on the strength of comparative studies of Indo-European Ian- guages (Dames 1904: xiii). That perspective consisted essentially in an tradition itself,
how these languages spread
as a result of the “fragmentation" and language into derived culture Proto- Indo- European society, and interrelated societies, cultures and languages (Renfrew 1987). But it is not onlv from these cross-references between one tradition and another that 巳ultivated nationalism gets what it needs to“invent” a tradition, at least in the sense in which the term“invention" is under- sort stood by Hobsbawm and Ranger in their classic study of 1983. of local“sociology”(deposited in the “memory-archive”) also contrib- utes to its endorsement. As an example, I recall that during the period of my researches in the south of the country, the morphological char-
idea of of a
A
of Balochi society were seen as a proof of past migrations. As we have s州in Ch. 1: the dispersion across the territory of pat- ronymic groups (zat) was explained, at the time of my research, on the acteristics
basis of
themselves to sion
be in agreement when it was
my friends
often showed put to them that this disper-
a“theory of migration". Although
might have bee? chiefly an effect of the water ownership regim
阳1igra g.ms of si户n.
the Balochi in the region inhabited
by them was under discus-
Spontaneous sociology and more or less articulate nationalist the- merged, at least among my better educated friends,
ones were often into a
discourse that i二nded to
assume a form of self-evidence.
To what extent this interpretation could correspond, in political and ideological climate of that time. to the need to supply a “historically"
叫“so叫ogically" coherer叼icture of山situation of仙e same is hard to say. Certainly however this view of the past clung 阳山idea of a common oriσin whi?h,suffused the nationalist dis-
years
course even in i卢;-1巳;?- ;;di??i and among those
locals
?;r卢j
who fed on their books (Inayatullah i987: 35).
What, though, this discourse “forgets”is the diversity of origins, both
Nation and Memory
Ethnography at the Frontier
116
cultural and linguistic, claimed by the various groups of which the Baloch people are “composed”. That diversity was in fact referred to whenever it was no longer a matter of negotiating identity with exter- nal agents such as might be the Pakistani state or the international media, but within the context of local politics. In the latter case the superior status of some patronymic groups over others was still deci- sive (see Ch. 3 and Fabietti 1992). Especially in the south of the coun- try, which is more exposed to the influences of lands beyond the Ara- bian Sea, some high status groups claim, as we know, relatively recent non-Balochi origins: Arab, Persian, Indian. In the “independentist” climate of those days such claims would seem to sugge;t an expla- nation of social superiority enjoyed in the past. Nevertheless even in those, these distinctions often had an effective socio-political bearing on situations. The fact is that both the idea of a common oriεin and that of a different origin of the various patronymic groups, ii{ reality 写rojected two opposite and alternative images. And in an indirect but decisive way, they informed the political dialectic of Balochi society. These opposite ideas, whilst not excludin1? each other, were referred to, as I have pointed out in chapter three卢cording to whether it was intended to assert the principles of equality, or tho;e of hierarchy and
117
say the nationalists themselves, of unifying ex Anglo-Indian Although the question of the status of the Khanate relative to the British was mentioned by the locals in the course of our conversations with a detachment not to say indifference, it proved of great importance to the intellectual nationalists. The Khanate reached the acme of its expansion in the eighteenth century, but lost its autonomy under the British policy of indirect rule. However, its position vis-a-vis the British was different to that of many other kingdoms or chiefdoms encapsulated by the Anglo- Indian empire. Like all the other “principalities" under British dominion and ruled capable,
and now Pakistani Balochistan.
by Hindu (raja) or Muslim sovereigns (nawab), the state of Kalat, led by a khan with autocratic powers, had stipulated a treaty with the British
Crown. This (according to nationalists' opinion) allowed
it
to main-
own - albeit nominal -…condition as an“independent and sov- ereign state". The newborn Pakistan, however, seems to have ignored 怡detail. Deeming itself to be山legitimate l览ir to由at part of出 tain its
status, as the ideal criteria for regulating social relations.
which fell to its lot after the withdrawal of Great Brit- toαtend its sov盯e叫rover all tl the region, including Kalat. In 1955, under various types of pressure, both diplomatic and military, Kalat was finally absorbed into Pakistan
As
(Harrison 1981; Redaelli 1996). Kal川as a long way from being a
be noted, the nationalist issue makes reference to山past by 三ombining an epic passed down orally, with a number of historico’ 可ill
l?nguistic-cultural theories of western origins, and local stories about the origins of the groups which today make up the Balochi“people”・
But
the medieval epic, with its hero Mir Chakar Rind, the pro- tagonist of poems and ballads (Sardar Khan Baloch 1977) contains ‘mythic”elements, the reference to a recent oast assumes a more con- if
meaning that can arouse more immedi二te reactions. One such case is that of the vicissitudes of the Khanate of Kalat, which was finally abolished in 1955, when it ceased to exist officially as a semi-autonomous entity and became annexed to the state of Pakistan. The question of the Khanate is not only a thorn in the side of the crete
nationalists, but also
an essential point of reference when a凶rmingthe recent and historically well-documented existence of a political power
subcontinent
纠n,
it
saw
fit
cen叫ized state. The khan抖m-
was an autocrat within his tribe (Ahmadzai), but not in a pos1t1<?n to impose his will on the chieftains of other tribes that were part ot a self
kind of confederation (Swidler on the 1979). Nationalist intelligentsia, other hand, tend to exalt even to-day the khan’s function, as if this personage had been nothing less than a head of state. In its efforts to
assert the politically compact character of east Balochistan under the Khanate, the nationalist argument forgets that the loyalty of tribes to Kalat was sporadic and obtained by force. To affirm the cul-
sometimes homogeneous character of the region, ・the nationalist argument furthermore interprets in its own way the linguistic data, whereby the ?alat吨。山dlo吨been considered inh必ited by populations of ori- gins different to th丛e of the Baloch (Ahmad Yar Khan Baloch切5; turally
Janmahmad
1988: 162-165).
118
The view that the Khanate of Kalat offers a precedent of political autonomy achieved by the Baloch in fact comes up against a fairly big problem. It is well known that the Khanate developed in the centre of a Brahui linguistic area. The Brahui, according to linguists, is a lan- guage closely related to the Dravidian dialects spoken by groups which, according to the “migratory”theory of Indo- European peoples, were driven by the latter towards southern India. According to this theory Brahui language, spoken to this day in the Kalat region, would be the language of a Dravidian group which in the course of time mingled with neighbouring groups, assimilating their customs and institutions (Bray 1977). In Balochistan the Brahuiphones are concentrated exclusively in the region of Kalat, in what is for the nationalists the cradle of the first his- toric instance of political unity in Balochistan. Consideration of the Khanate of Kalat as a precedent of achieved political autonomy clashes, therefore, with a considerable problem, that of a“contradictory”his-
and linguistic non-homogeneity between Baloch and Brahui, between “conquerors" (the Baloch) and “dominators of the conquer-
torical
The questions
of the political unity of Balochistan and that of its cultural homogeneity are in fact, in the nationalist per- spective, closely connected. In effect, the “culture”of Brahuiphones is not distinguished from that of the Balochi groups any more than the regional differences exist’ ing between the latter. Brahui itself, for the matter, was influenced by Balochi, Pashtu and Urdu、in the same manner in which Balochi was influenced by these idioms. Brahuiphones are as a rule bilingual and there is nothing, aside from the language used in everyday comm?- nication, to distinguish them from other Baloch. Besides, they speak Balochi fluently and declare themselves to be Baloch. They do not even seem to have an ethnic “dual identity" which might be displayed according to context (Barth 1969). The Brahui are perceived and per- ceive themselves as Baloch. However, for the nationalists this factual identity of Brahui-spea?・’ ing groups is not sufficient. In their opinion the origins of those speaK’ ing the language of the region must comply with the historical role of the Kahnate of Kalat. In other words, it must be possible to attribute an ors”(the Brahui).
Nation and Memory
Ethnography at the Frontier
119
(and not so much a“culture") to the Brahuiphone valid enough them on the same plane as the Baloch. Not convinced by the theory of European linguists (who postu- lated the existence of a Dravidian residual linguistic island), nor even by those of the Brahui themselves (who consider themselves Baloch to all effects), the nationalists attribute to the groups of the Kalat region a different identity but a common origin. In a perspective of this sort, the Brahui would in fact be neither the descendants of a Dravidian popula- tion, nor of an autochthonous group, but of groups with origins simi- lar to those of the Baloch, and who arrived in the Kalat region during origin to
put
same period as that of the large-scale Balochi migrations. This is in those nationalist authors who strive to assert the contempo- rary nature of the Balochi and Brahui migrations, attributing the char- acter of a historical “proof" to the myth of the origin of both peoples: if the Baloch are de…ndants of Mir Hamza who originated from the Arab Peninsula、the Brahui are the descendants of a certain Braho, or the
explicit
from Aleppo and companions former (Janmahmad 1988).
Ibrahim, origiO:ating of the
in
arms and
faith
So for the nationalist argument, the elements legitimating the unify- mg role of the Khanate of Kalat lie in the common origins of the Bra- huiphone and of the Baloch, and thus in terms of a common histori- cal destiny. This began, once again, with the migration of conquering peoples from the west towards the east and was achieved in histori- cal times with the unification of a great part of Balochistan under the Khanate. Linguistic difference thus becomes an altogether secondary and almost insignificant element. For it is, so to speak, reabsorbed and .
erased in a
common original history.
The “cultured" discourse of nationalism (Janmahmad 1988: 259-69; ?nayatullah吵87: 35-36) taken into consideration so far reveals an 阶rtwining of ideas and representations belonging both to the ?干pe- nence of European nationafism and to the Balochi memory-archive; local both to山di牛ourse of European historical science and to the repr二sentation of recent political history.
sa民mto
This discourse floγs, we may
what UH Hannerz has called geocultural imagination.
In Hannerz’s words, geocultural imagination would be “a matter of fairly large-scale mapmaking”,a way of referring “to the distribu- tion of things cultural, somehow cultural, over territories and their
human populations" (Hannerz 2007). At the same time the expression is a way of “suggesting that [we are] focus- ing on the way we think geoculturally, about the world and its parts,
“geocultural imagination"
and the main features of those parts.”This is not an absolute novelty for anthropology, because the adherents to the German “diffusionist
school" and other scholars had already dealt, by now a century ago, with the distribution of cultural traits in the diverse cultura areas of the planet. Yet, continues Hannerz, in those days, a century or so ago, these were mostly activities of the ivory tower, where scholars would argue over matters of conceptualization and cat-
egorization mostly with their peers. In more recent times, it seems [... ] that the geocultural imagination has become more volatile, occurring in both aca- demic and public arenas and also crossing the boundaries betwe n them more readily,
In this
Nation and Memory
Ethnography at the Frontier
120
and more ambiguously.
-
way geo-cultural representations become
significant components in a transnational collective consciousness, a set of representations of the world which are circulated. received and debated in a world-wide web of social relationships, and which-again stimulate h her cul tural production.
They are representations affecting broad strata of world society, thanks
medium of television. prop- 199。s, had its maximum
to the spread of literacy and, especially to the
Geocultural imagination, during the
agators in a number of “organic intellectuals”:Huntington, Kagan, Fukuyama, to mention only the best-known. Some of them “served in
?ashingto旷in one administration or ano山r. Huntington has linked his name to the geocultural image of a“clash of civilizations”,where the reduction of culture to religion is a fundamental pillar of his argu- ment (Huntington 1996).
The image of the cultural dynamic offered by these theorists of geocul- ture is one of those which Clifford Geertz would readily have defined
121
as typical of “politologists in a balloon" (Geertz 1999). In fact, whilst we have to recognise the extraordinary power of persuasion exerted by it in terms of geocultural imagination, we must also underline its poor analytical capacity at an ethnographic level. There is no doubt that when the idea of a “Greater Balochistan" was enounced in 1933, geocultural imagination was already present among the Balochi nationalists themselves. We can, however, also note that
changed according to specific politi- and moments. It is in fact possible to note these changes comparing the various versions of this idea that have succeeded
the idea cal
by
of a Greater Balochistan
strategies
one another in time.
For the nationalists of
1933,“Greater Balochistan"
embraced, in addition to the Iranian and Anglo-Indian (now Pakistani) Balochistan, the Sindh, the region of Bahawalpur and should have
Dera Ghaza Khan and Ismail Khan areas. Strangely, Afghan Balo- not mentioned (and thus not even indicated on the annexed map). The reason for this omission is that at the time Afghanistan was already a sanctuary and a potential ally of the Balochi nation- alists, as it was also to have been in the years 197。-80, thanks to the then pro-Soviet and anti-Pakistan government in Kabul. In the mid- twentieth century on the other hand, the idea of a Greater Balochistan again included Afghan Balochistan, whilst the northern region of Dera Ghazi Khan had disappeared - but only to return in the r99os as a topic of debate. Set on the north east limits of what is considered the max- the
chistan was
expansion of the Baloch during the sixteenth centur民 became once again the focus of attention by the nationalists ?fter the possibility of important uranium deposits had been suggested
i牛um point of this
area
m the neighbourhood.
can thus be noted that geocultural imagination - the representa- dominant cultural forms - changed accord- mg to periods, but also above all to contingent political circumstances. It
.
丁ion
I? is
s1ve
of spaces“containing"
larg句a mental construction, whose ‘‘forrr the purpose”at differing times.
strategies “useful to
we were saying, these geocultural representations are ana }Yt叫Y ineffect sin创l叼ey叫1创he cultural compl叫y to a few ssential”traits. They are in fact promptly belied by practice. Ho.wever, as
-
,
122
Nation and Memory
Ethnography at the Frontier
I mentioned earlier the best-known case of geocultural imagination: that of the reduction of culture to religion, carried out by Hunting- ton in his celebrated book of 1996. It is interesting therefore to see the
place of religion in the Balochi nationalist discot1rse and in the effec- tive behaviour of those who share it. Balochi nationalism is in fact a secular movement, having no organic connections with Sunnite reli- gious circles.
The Baloch
are for the most part Sunnite Muslims, but a substantial minority are believers in Zikr, a form of worship dating to 16th cen- tury that has references to Shiite Islam and is particularly widespread in the south of the country as well as among Baloch migrants in Sindh and in the Gulf area (cf. Chapter 5, note 4).
Still in the eighteenth century二almost the whole of the population of southern Balochistan was of Zikri faith. Onlv then did this “con圄 k臼i?n”begin slowly to decline, so slowly出at b夕the叫of the nine teenth century the Anglo-British administrator-ethnographers’in recording their high number, were quick to consider them a veritable
“tribe”(B.D.G., 19。6).
-
The decline of Zikr towards the middle of the eighteenth century was a consequence of military intervention by the Kha口 of Kalat, which smashed the “separatism" of the lords (hakim) of Kech (who at the time were followers of Zikr) by conducting nothing short of a jihad against them. After the local lords, defeated and deor甘ed of much of their autonomy, had been converted to Sunnite orthodoxv. the Zikri双,ere left free to practise their worship. Although a part二f the population ?ith time did abandon Zikr, following the example of their loca巾 the presence of this cult nevertheless remained strong in the region. In the mid-199。s the principal Zikri sanctuary, not far from the city of Turbat, was still attracting yearly, at the end of the Muslim month of Ramadan, some thirty thousand pilgrims from all corners of Pakistan. Since the mid-eighteenth century there has been no further major contrast between Zikri and Sunnites. Zikri、as recentlv as in the 1990s, still had no difficulty at all in proclaiming山ir r二ion and prac山 ing it throughout Balochistan. However, from the mid-198os, Zikn have been the object of attacks, and not only verbal ones, by Sunnite
123
1987). Having come from regions outside Balochistan expeditions of non-Balochs into the south of the coun- and organized these connected with the Quranic schools situated in the mullahs, try, northern and eastern regions of Pakistan, attempted to raise the ques- tion of what they consider to be heresy. Their attempts however did not produce - at least at the time of my stay in Makran - the results expected by the promoters of these “punitive expeditions”. The Sun- nite Baloch in fact saw these raids led by “fanatical mullahs”(as they call them) as nothing more than one of the numerous plots woven by the government to foment disagreement among the Baloch themselves. I recall perfectly the attitude of many of my hosts towards these epi- sodes. Alam was among the angriest. He saw religion as a“political fact”,and regularly made fun of the various (including local) mullahs (whenever the bang (the call to prayer) resounded through the vii- lage air. Unfortunately, at the time of my stay, a number of dramatic events also occurred. Such as when Munir, one of the most conspicuous authorities in the Zikri community, suffered an attempt on his life, in which he and his wife were wounded and some of his guests were killed. During the years of my research (1987-1994), the friction between orthodox non-Balochs on the one hand, and Zikri and Sunnite Balochs on the other, had in fact several times risen to a considerable level of 中帆to a point where even山川1 power loc均r (and who, accord- mg to the nationalists, were deeply compromised with central govern- ment), had openly declared themselves in favour of religious freedom mullahs
(Ahmed
and of the Zikris’right to practise their beliefs. is the attitude of the nationalists towards the religious ques- The idea of the mullahs’raids as the product of destabilization plans by the Islamabad eovernment did not seem at that time to have serious foundations. B山if one looks at the reasons given by Paki- stani governments to justify the annexation of Balochistan, it can be 叭d阳山one adopted most f叫uently has been that of“religious unity”I.e.“purity" of the country. Pakistan means in fact the “Land
But what tion?
of the Pure”. .
One should not be surprised therefore if r川ionalism regards incurr- fanatical mullahs as manoeuvres aimed to create a state of
sions of
division and confusion among the Baloch. If religious fundamental- ism is recognized as an element of anti-Baloch policy, the nationalist argument shows a tolerance in matter of faith, and even a clear incli- nation towards secularism. And to the point where the jihad waged by
Khan of Kalat against the Zikri in the r 8th century, is considered a marginal episode, in keeping with the idea of the Khanate as a polit- ical precedent of historically largely achieved unity. the
This is just one example of how, on a practic?l level, representa- tions of a culturally homogeneous Balochistan with no internal dif- ferentiations disappear altogether in terms of geocultural imagination,
where again the propagandist element blots out all the others. We must therefore concentrate on the socio-political context within which the nationalist discourse operates.
In the light of the grand historico-cultural visions embraced by the Baloch nationalists, there exist in fact、within the latter. differe且t atti- tudes to the “Balochi question”. There are not only differing politi-
depending on one ideological choice rather than another example the contrast, in the叭。s, between Marxists and non-
cal options, (for
Nation and Memory
Ethnography at the Frontier
124
Marxists); there are also differences relating to the social history of the different areas of Balochistan itself. One difference which, from this angle, seems to me very important is that between the north and the rest of Balochistan. The North is by tradition the land of the great tribes and tribal leaders (sardar). The south is on the other hand char- acterized by the presence of a heavily stratified agricultural society 灿n_ded on the patron-client relationship. This difference has repe卜 cuss10ns on the type of relationship that exists between the Balochi community and the state of Pakistan. The tribes of the north interact with the external powers (now with the Pakistani state) through their traditional leader;. It is true山tin the south too, the key exponents of this policy belong to the upper echelons of the hierarchy, but the morphology of local society makes the following of a prominent leader less stable and reliable than in the north. In the north we岳nd a structure of relationships that falls roughly, as I have already said, into the category of “segmentary" societies. Every
large
"tribe”,qowm,
is
composed of
125
a series of lineages”,khel, each
own (Pehrson
1966: 17-18). From one of these khel comes the “tribal”saγdaηan office passed down by inheritance within a single lineage, and called, in fact, saγdar khel. The lineages making up
with a sardar of its
thought of as descending from a common ancestor, from all are“equidistant”. Indeed the sardaγkhel, that is to say the lineage to which the tribal chief belongs, is considered gene- alogically“closer" (i.e. descending in direct line) to the tribal ancestor. In southern Balochistan on the other hand, the khel system is absent, whilst the patron-client relationship prevails. This constitutes the foun- dation of a stratified society, in which individuals recognise their sub- a “tribe"
are
whom, however, not
other individuals. The relationship of dependency that many individuals without means to a mas位γ(literally “the big- gest”) is certainly binding but is in some ways less “constrictive" than tribal affiliation. The relationship between client and master is in fact configured as one of dependency which can be interrupted by means of ordination to binds
by a client, to a new master. The absence of “tribal”bonds makes the following of a mas位γpotentially more“unsta- ble”than that of a sardar and, above all, less “morally" pledged to his affiliation,
therefore
This could help to explain why the leading figures of Balochi have been - with rare exceptions - tribal sardars, and why the armed conflict waged in the seventies was mainly provoked by the large tribes of the north bound to their sardar (Harrison 1981; Redaelli service.
nationalism
1996).
The particular position occupied by the sardars, which in the past some of them to present themselves as champions of Balochi
allowed
ind二pendence, is nationalists
however not without contradictory aspects. There are who have been accused、both by the nationalists
sardars
丁?er
t?ve…attitude
to the Balochi question, allegedly determined by calcula- opportunity. Cases are cited of sardars who allegedly to灿1叫1e bu灿1g of roads and schools 。川1eir
tlons of P:rsonal
沪d问r毗s
with the excuse that too rapid modernization would have t阳1at叫fects o叫epo川tion. This kind of r伽al has been mterpreted by some nationalists as proof of bad faith on the part of the sardars, who, they allege, are afraid of loosing their privileges. These ?rntones,
丁ad
Nation and Memory
Ethnography at the Frontier
126
by upholding the idea of an autono- mous Balochistan whose population should be entitled to preserve
sardars reply to this accusation
its
customs and
institutions.
To consider
the sardaγ・s conservatives in bad faith is perhaps a bit too simple (some have paid for their anti-government attitude with prison, exile and even death). However, this situation does testify, on the whole, to the contradiction that exists between aspirations to unity and autonomy on the one hand, and to“tribal particularism" on the other. The ambiguity of the sardars’attitude seems to be not so much the product of personal calculations as of the position that they occupy in the context of relations between Baloch and Pakistani central gov- ernment. The sardars occupy a prominent position within a socio- politically“tribal”structure, seen by a part of nationalist thinking as an obstacle to the realization of its designs.
Balochi nationalism thus seems to be marked bv a twofold charac- ter, that of the tribes and that of Balochi“modernist”educated people who consider the so called “tribalism”to be a force contrary to the achievement of the region’s independence in a manner inspired by the idea of a Balochi culture and “national”society. However much one may strive to promote in various ways an image of the Baloch peopl?
from the tribal heritage, for example by af岳xing to persona documents not“tribal”names but very simply the wording “Baloch' released
1
’
as a“surname”一it is nevertheless true that anti-Pakistan mobiliza-
阳
not always clear ones,
made by the sardars (Harrison 1981; Titus 199o).
or violence on the part of an outside power can, as such, form of reaction self-proclaimed as nationalist. Balochistan had been subjected for centuries, if not for thousands of years, to the pres- sure of foreign powers, but of course this fact had never until recently colonization spark a
been the origin
know
I
in all phenomena definable as“nationalism”,the appearance of national unity and identity in Balochistan is not simply the effect of a demand from within local society. Rather, it is the result of an inner sentiment prompted by external forces. This forces coincide with a large number of factors. Firstly, the presence of the Pakistani st斗te
wh?ch tends, like any state, to impose its autho叫on山臼rritones under its jurisdiction. Then, there are the other states in the region, and the non-Asian powers that have at times exerted dramatic p?es- sure on Balochistan. It would however be reductive to maintain that
might be quali岳ed as“nationalism气 would not be shared by many nationalist intellectu-
of anything that
this idea
have mentioned, overvalue the function of the Kalat in constructing a Balochi national identity. This can be noted also from Internet sites2 in which the Baloch claim their auton- omy and the right to the existence of an internationally recognized Baloch community. With British and subsequently Pakistani rule, something however did change. Education reached Balochistan and schools, though over- crowded and with often ill-prepared teachers, were opened, with a University in Quetta, the capital of Balochistan. As a consequence, thousands of young people have had access to information and pro- als
the
who,
as I
process of
qualifications that do not fit a society of pastoral herders and palm growers. It is no coincidence that the two students' organi- zations, the Balochistan Student Organization and a dissident part of it, fessional
date
have several
thousand members and that,
still
more significantly, they
two most active political movements in Balochistan. It is true that some nationalists see the develooment of educational facilities 叫services by山Pakistan authorities二an attempt to favour certain 巾ses and thus to break up山solidarity of山B aloch to山ad van- tage of the government, besides being a quite insufficient remedy for the growth of the region. But this development of educational struc- repre?ent the
...
produced an entirely new situation. In it young Baloch aspire place for themselves in a world auite different to that of the generations. However they are frustr;ted in these aspirations, by
tures
As
127
has
to find a
past
a state
that
is
?l the仙 IIlOre
unable to e:uarantee access to
new conditions
of
life t 0
conditions which the presence of the state has The presence of a state as such today, tends in
true in the case of
contributed to create. 2
te;d???t:??mple
http://b山hhouse.blogspot.com/2叫叫er阶cove时
Nation and Memory
Ethnography at the Frontier
128
fact to set the condition for a greater mobility of individuals and for increased communication between the different sectors of society. In this sense those, like Gellner (1983), who consider nationalism to be a product of “modernity”,are right. So it seems that nationalism emerges where, with the widening of a situation of difference perceived at the ethnic and cultural level, a gap arises between the sy;tem of expecta- tions and the possibility of actually fulfilling those expectations. What
seems very important to me is that the higher cultural level reached among the younger population has created, above all, the conditions }or山memory-archive of the“amorphous mass”to be retri叫bya
memory tunction”at the service of a political project. Any reflection on a cultural otherness brings into play what Roy
Wagner
(1982) qualified as a reinvention of its
own
culture. Those
observing others and reflecting on others make use of analogies and metaphors inferred from their own cultural codes. Thus they end up, in a way, by “objectifying”their own culture, so as to make it appear in a different light to that in which it had been presented to them pre- viously: they “reinvent”it. But this does not concern only the ethnog- rapher. For his or her interlocutors, too, must understand the people they are dealing with, and to do that they too must resort to models and metaphors inferred from their own cultural repertoire. They too, !ike山讪nog即her, arrive at an objecti岳cation of their own c1川m nence its remvent10n. In thi? kind of“primary scene" presented to us by Wagner-the eth’ nograph1c encounter - the obiectification of one's own culture is trans- lated into the “incorporation”of concepts and meanings of other pe。” pl卢cultures into one's own. This is what l呻appens and on an inter-subjective level, in the ethnographic encounter. But it is also what happens in the “contact between cultures’,in a period in which history seems to“accelerate’,in which the transfer 。£ tech-
or叫
same cultural source, the discourse on Balochi nationalism man- by dipping into the local memory-archive, to reinvent tradition, and hence to invent a new one. But this process does not occur within an“equal”situation. The cir- culation of meanings, this kind of “traffic of cultures”,never happens ina“neutral space”characterised by equality of the interlocutors, but rather within a general picture characterised by specific relationships of force. As Talal Asad (1986) pointed out years ago, these relationships of force can also be translated, besides everything that we know very the
ages,
hegemony of“languages" by one culture over another. Serv- an identity being formed, and as such threat-
into
well,
ing the
“reinvention" of
Baloch, the western languages of political and historiography, like certain influences of Euro- pean nationalist tradition exercise, in the context examined, an action ened, as is science,
of
in fact that of the
linguistics
dominion (in the University of Balochistan, to
my great surprise, I
came across a historian interested in the Italian “Risorgimento"). This does not mean that the resumption and adoption, by Balochi
of European-sourced ideas, should be construed as an example of the direct dominion of western culture over the local one. Rather, it means that local culture is learnine: to think itself also through those ideas and those representations. Fr this point of view,
nationalists,
m
t?e山mes of Balochi nationalism, and the modes of削r articula- tlon into a discourse with claims to coherencv.、constitute an exam- 飞le ot what can be intended,时a〕r, by “speci岳;i让ist cal灿tions of aommance and dialogue"3 between different cultures; in a period in 哈哈,to borrow an ;xpression used by Christopher Hann,“‘hist y accelerates" 4
nologies, the adoption of a language ot' exchange (as for example Eng- lish), the circulation of information, the market and the media, enable
peoples to“communicate”with one another. There is nothing “opti- 户istic" in this way of considering things. Nor il there anything in the least pessimistic
about it. It is a fact. Like that whereby, in taking up themes of western scientific thinking, and ideologies originating from
129
‘‘JaaT
,d
KH
yaA
・1]
qm
a、句AU
门口。/
09
VAA毡’
ny nxu 『/
Z
龟,、J
Bibliography
Ahmad Yar Khan Baloch, 1975, Inside Baluchistan, Karachi, Royal Book Company. Ahmed, A. S., 1987:,“Islamic Fundamentalism, Sufism and Ethnicity in Pakistan: A Case-Study from Baluchistan", Journal of Central Asia,
Akbar,
X,
I.,
1:
37-61.
1989,“The
Heroin Coast", The Herald, Karachi, October.
M., 1991, Politics in an Arabian Oasis. IB Tauris, London. Anderson, B., 1983, Imagined Communities. Re卢ections on the Oγigin and Spγead of Nationalism, London, Verso.
Al Rasheed,
Armstrong,]., 1982, sity
of N.
Nations
befoγe
Nationalism, Chapel Hill, Univer-
Carolina Press.
Concept of Cultural Translation in British Social Anthropology", in Clifford, J., and Marcus, G. E. (eds), W:γiting Culture. The Poetics and Politics of Anthropology, Berkeley, Uni- versity of California Press 1986. Assmann,儿,1999, Eγinneγungsraume. Formen und Wandlungen des
Asad,工,1986,“’fhe
kulturellen
Gedachtnis郎、Miinchen. Beck.
Makran), Bombay, The
Balu?histan District Gazette卢,19。6, (v二l. VII, Times Press. Barker,.
M.A., and Mengal, A. K., 1969,
A
Course in Baluchi,
2 vols.,
MacGill University, Montreal, Quebec. Barth,丑,1969,“Pathan Identity and its Maintenance", in Barth, F. (ed.), Ethnic Groups and Boundaries, Boston. Little, Brown and Co. Institute of
B erque,J.,
Islamic Studies,
1953,“Qu’est-ce qu’une‘tribu’nord-africaine?”,inAA.VV.,
Eve户tail
de
l
'histoiγe
Cohn. Black,Max, 1962,
Hommage
a Lucien
F沙τ,re,
Paris,
Models and Metaphors・ Studies in Language and Phi- Cornell University P卢ss. 1977 [1907], The Brahui Language, 2 vols. Quetta, Brahui
losophy, Ithaca, Bray, D.,
vivante.
Academy.
Ethnography at the Frontier
132
The Predicament of Culture, Cambridge, Mass. Har- vard University Press. Clifford, J., and G. E Marcus, 1986 (eds), Wγiting Culture. The Poet- ics and Politics of Anthropology, Berkeley, University of Califor- Clifford, J., 1987,
ma Press.
Dames, L. M., 1904, An Ristoγical and Ethnological Sketch of Balochi Race, London, Royal Asiatic Society. Dames, L. M., 19。7, Popular Poetγy of the Baloches, London, Folk-
Bibliography
Gingrich, A.,
Dumont, L.,
1966,
Homo hierachicus, Gallimard, Paris.
Berghahn. Hann, C.
oco, Roma, La Nuova Italia Scientifica. Fabietti, U., 1996, Loγds of the Desert, Lords of the Frontier. Pastoral Nomads and Political Centralization in Arabia and Baluchistan, in Fabietti, U., and卫C. Salzman 1996, 426-436. Fabietti, U., 1996b,“Equality versus Hierarchy: Conceptualizing
Change in Southern Baluchistan”,in Titus, P. (ed.), Marginal- ity and Modeγnity. Ethnicity and Change in Post- Colonial Balu- chistan, Oxford University Press, Karachi: 3-27. Fabietti, U., and卫C. Salzman (eds) 1996, The Anthropology of开ibal and Peasant Pastoral Societies. The Dialectics of Social Cohesion and Fragmentation, Pavia,
?
Ibis.
Ferraro,丑,199。,“‘Household’and 'Family'. Some Ethno-Lexical Aspects in Baluchistan”,Newsletteγof Baluchistan Studies,只Fall.
H.吵59 An Anthropological Rec?nnaissance in West Pakistan, Cambridge, Mass, Peabody Museum. Foucault, M. 1969, L泣γcheologie du savoiηParis, Gallimard. Field,
1955,
Geertz, C.,吵73, Interpretation of Cultures, New York, Basic Books. Geertz, C., 1999, Mondo globale, mondi locali, Bologna, Il Mulino.
Gellner, E., 1973, Cause and Meaning in the Social Sciences, London, Routledge Kegan Paul. Gellner, E., 1983, Nations and Nationalism, Oxford, Blackwell.
&
M.
(ed.),
1994,
When Ristoγy Accelerates, London, The Ath-
lone Press.
Hannerz,丑,z。。7,“The rylines”,paper
Geocultural Imagination: Scenarios and Sto- Milano Bicocca University
read at a seminar held in
on March 20th.
In the Afghanistan's Sbado肌Baluch National- and Soviet Temptations, York, Carnegie Endowment for
Harrison, S.丘,1981,
ism
How
Fabian, J., 1983, Time and the Other. A例如opology Makes its Object, New York, Columbia University Press. Fabietti, U., 1992,“Power Relations in Southern Baluchistan. A Com- parison of Three Ethnographic Cases", Ethnology, XXXII: 89-102. Fabietti,以,1995, L'identita etnica, stoγia e critica di un concetto equiv-
and M. Banks (eds), 2006, Neo-Nationalism in Europe Perspectives卢om Social Anthγopology, New York,
and Beyond.
-
lore Society.
133
New
International Peace.
and T. Ranger (eds), 1983, The Invention of Tradition, Cambridge University Press. Huntington, S.卫,1996, The Clash of C£τ必zations and the Remaking o[World OrdeηNew York, Simon and Schuster. Inayatullah Baloch, 198凡The Pγoblem of "GγeateγBaluchistan". A Hobsbawm,
E.,
Cambridge,
St』,d
』mad
Ja
Pakistan, Kolff,
the
D.
日・
(J.
Baloch),吵风E叫5仰Bal吵ch Nation
S tru马le in
Quetta, Gosha-e-Adab. A.,
Military
199。,NaukaηRajput and Sepo)人The Eth,价istoγyof
Labour Maγket in Hindustan, 1450-1850, Cambridge,
巳amb吨e University Press. Lattimore,
O.,.i988.[r94。1],
The InneγAsian Frontiers of China, Oxford,
O?ford Umvers1ty Press. Machiavelli, N., :961 [1513], The pγince, transl. worth,
by G. Bull, Harmonds-
Penguin Books.
Montagne,孔,1932,“Notes
sur la vie sociale politique des duNejd”,Revue des etudes islamiques, 6.
Chammar
Pastner,
S,叫c,叨z,“Agriculture Kinship and Politics in South- Man,只1: 128-136. Pehrson, R. N., 1966, The Social Oγganization of the Marγi Baluch. ern Baluchistan",
ffd:::ed
and
Analyzed介om灿灿tes by主Bar吮Chicago,
P牛E.,帆The Bedoui叫Cyrenaica. Studi…Personal and J. Goody and E. Marx. Cambridge, 己:?i;:i吕立r?it;?e?[
Ethnography
134
at the
Frontier
Index
Redaelli,孔,1996, fl Khanato di Kalat. lnghilterra, Russia e Persia: questioni di frontiera (XIX-XX secolo), Milano, I.S.U. - Univer- sita Cattolica.
Cape.’
Renfrew, C., 1987, Archaeology and Language. The Puzzle of Jndo- Euγopean Oγigins, London, J. Ricks, T. M., 1989,“Slaves and Slave Traders in the Persian Gulf. r8t l and呗h Centur The Economics of the Indian Ocean Slave 1》-ade in the Nineteenth l
Centuγ;y,
London, Frank Cass.
Salzman,卫c.,r97s,“The Proto-State in Iranian Baluchistan”,in Cohen, R., and E. R. Service (eds), The Origins of the State. The Anthropology of Political Evolution, Philadelphia, ISHI. Sardar Khan Baluch, M.,切7-吵84, Literaγy Histoγy of the Baluchis, Quetta, Balochi Academy, 2 vols. Smith, A. D., 1986, The Ethnic Oγigins of Nations, Oxford, Blackwell. Spooner, B., 1964,“Kuch u Baluch and Ichthyophagi”,Iran, vol.2, 53-67.
Spooner, B.,
1969,“Politics,
Kinship and Ideology in Southeast Persia’;
Ethnology,界 Swidler, N., 1979,“Brahui Political Organization and the National State飞in Embree, A. T. (ed.), Pakistan's Western Borderlands, Kara-
Royal Book Company. Titus, P., 199。,Political Allignament of Baluchi Sardars in Relation to the Goverment of Pakistan,“Newsletter of Baluchistan Stud- chi,
A
measure) 3元43
abba (water
Agha
Khan
105
agriculture Al
32
Baloch (nationalist journal) 112
Al-Qaeda 107 Anderson,
Benedict
11。
Anjuman-e Ittihad-e Balochistan)
1
Arjun
Appadurai,
Balochistan (Organisation for the Unity of
12 25
Ar牛Peninsula 119 asatnyya
74,
77
uq
Asad, Talal
Assmann, Aleida 113 B
.
1es”,7,
Fall, 59-67.
and Modeγnity. Ethnicity and Change i牛 Post- Colonial Balochistan, Oxford and Karachi, Oxford Univer-
Titus,卫(ed.), 1996, Maγginality
s1ty Press, 1996.
bagha (male prostitute) baha (blood)
93
24
ff.吁
Bahawalpur 112. 121 切loch习,48, Baloch.
Baloch1
Wagner,孔,1982, The lnve
95,
5?’,53
bamprosh
w ater
(
11
(
C?rganisat叫盼)
IO
am\(Kuh-e-Baloch)
1
??s?叫认40 123
于/h'.c;re??fct;;> ?::aar 17,
erque, betek
86
nationalism 110
::::t}丁?!:!::: bang
54, 65, 75, 78,
?8,町,~,时,时
19,
23
Jacques 27
(guest’s
room)., VVlll) 25, 32
Ethnography at the Frontier
136
Index
bilateral descent 2 7
exchange
biγ(vendetta) 2 3
extended families
Black,
Max
137
40
24
73,” blacksmiths 5 2
blood (see baha) boundaries 22 Brahui n8, n9 Brahui (language) 86
Fabian, Johannes
5
(Afghanistan) 99
Farghana Valley Farsi
Buleidi 95, 1。。,IOI bunjui (water system) 3元38, 41, 42 Bu Said 95, 100
(language) 95 (military governorsin
faujdar
force see
Michel.”,III, IIJ
Foucault,
frontier 4, 5, 6,
73, 82, 1。7
C
Fukuyama, Francis 120
casta (Portuguese) 51 caste 〔English) 51
G
commissioner 65 Conrad, Joseph 7
Geertz,
Clifford VIII,
Gellner,
Ernest 11。,II2、128
ghel-o-sang
D
geocultural
Gichkis
daγzada (social category)门,64 85
93
ff.
-
n1-" vanc:s
'-Prv'.l ;:)CI
(a
53
Great
84,
106
85,阳川”b队97, 98,陈叭叭叫
humped ox) 13 Balochistan 112、IIA.. Ill 万寸,
small
Greater
dependency 1 Dera Ghaza Khan n2、Ill domestic. .u且\,,,Hi"'
(buildi鸣叫terials) 3, 23 imagination 1 IQ. .,, 120
52, 79, 83,
103, 104,
gok
da也,a (force) 9,
120
ghulam (slave) 49
Dari (language) 95
Dasht
XVII c. India) 80
dawa
Game
8
Gwadar 105
Durkheim, Emile 70
H
E
t::??〔叫州,26,札队5内,65,
emigration 57 equality 9, 46, 116 essadar (wa?er’s owner)
』:口二出:?;:ras;:;) ?1annerz,
33 34 36 37 40,制,42,43,44
thnog即h1c即resenta…1s' 73: 9/ ethnography of the contemporary 5, 6
Ulf
II
;6,认川I
9、120
:t(wat叫s叫3凡39丘,40 φta?
(week) 32, 36
hered1t icy
35
7川,川
Ethnography at the Frontier
140
marriages 67 marriage strategies 69 Mar Singh 95
p Pakistan 12 3 Panjgur 94, 95, 1。l, 103 Pashtu (language) 107
Marxism
124 maste叫“big man")白,62, 125 med (fishermen) 52 Melucci, Alberto VII
memory
9,”,I。4,
ro7
Index
patrilineal
descent 26
patron-client relations 56, 6。,69 patronymic groups 116
ff.
Shah 79
migrations 2 8
Persian
Mirat-i-Ahmadi
places 8, ff. practical life 12
8。
Mir Chakar Rind 114, 116 Mir Hamza n4, n9 model 6, 9,巧,63, 71 ff. 83, '
prayer 123 l
of behaviour 63 monetization of the economy
Mughals 79,也,81 mulk (litt. Property.“Oasis“) mullah (man of religion) 26
12, 128.
56,
57
17, 20,巧,3。
Prophet Muhammad 114 public space 2 3 Punjabi ro9
Q qanat (canal in Iran) 63
qowm
N
(hindu prince) n7 Rajaput 81, 95, 96, 1。4 Ranger, Terence n5 razm-e-baloch (balochi customs) 47, 98 residence patterns 2 3
raja
nationalism I吵,129 nationalist intelligentsia 117 79,缸,83
nawab (muslim prince)
(“tribe”,Northern Balochistan) 12 5
R
nakib (servants) 75 Nation 9, 1。7 ff.
Nausherwani
n
8。,96,117
nazim (muslim prince) 8o earc斗us (Alexand的admiral) 4 nomadic pastoralists 5 2,
N
s
・
77
Salzman, Philip 77, 78, 79 chief") 49, 75, 105, 124, 1巧,126 sareshta (a canal’s“supervisor”) 33, 34, 35 servile relations 60
saγdaγ(“tribal
。 oasis 13,
15, 16,
17
shahγi (villagers) 74, 75, 77, 78 shi 'ia (religious denomination)
Sindh 112
98
141
Ethnography at the Frontier
142
slavery 2
Smith,
Anthony
w
111, 113
smuggling 59, 61 dependence social mobility 60 social
Wagner, water 9,
sren bandi story-tellers 53 Sunnite 122 supi (sufi) 26
T tallar (water system) 37, 38 Talpur (rulers of Sindh) 79 Tamba 2, 99 tas (water measure) 3凡43 tassu (water measure) 3元43 time 9, 31 ff. Toynbee, Arnold 85
translation 50, 51
bagha
tribalism 126
99
u V vendetta 23 village 16
ff.
nomadic tribe in Balochistan) 95 zamindar (landlord) 80 zat 26 ff. 49, 5。ff. 64 ff. II5 - as a‘patronymic group’5l - theories of the fragmentation ot 27
of a leader) 61
-
synchronous (synchrony/asynchrony)
Urdu (language)
n ff.
Zahari (a
74, 75 (“girding up the loins”,the followers
2,
128
z
Spooner, Brian
Turbat
Roy
wakil (councillor) 65, 84, 98
6。
Soviet Army ro7 Soviet Union 79 space 8, r r ff.
transvestites see
143
Index
1。6
zikri (religious
43,
44
denomination)
2, 98,”’100,
122
ff.
Illustrations
的口五ω击飞鸟。“ω同叫“5。占bv巳kr〈-KS-dkrhωk巳48凶ω击旧。的ωωd口τ-ug的ω的付。川同口。叫“EZDE
A-1
A-2
Ethnography at the Frontier
Illustrations
UKS的民灿、NHAωfg。占比ztkrω4humc。5bSE
问333叫百二sz2…N口。552口问
A-3
A-4
Ethnography
at the Frontier
Illustrations
A-5
A-6
的叮g叫击队EH{tbpu也gud』“mD{{问
Ethnography
Illustrations
at the Frontier
的问口。A』ω叫ω口叫』υ。-duAK{》Auhc。ZHNLHuvmD--H
‘、- f『?"咱干
A-7
A-8
Ethnography at the Frontier Illustration
』、
(
飞主J? 、!工.J
'?
?
’:主,j
r7?\污、.?:/牛〈 ?[-.主’」,'r'丑,'lf..》
8:
Digging a kariz
Illustrations
A-9
A-10
Ethnography at the Frontier
Illustrations
A-11
A-12
Ethnography
at the Frontier
Illustrations
Illustration 13: Balochi
ω3525号叫自的叮{。ω〈UN【口。叫“EZDE
woman in full attire
A-13
A-14
Ethnography at the Frontier
Illustrations
ddNH{utNdu405ωQ
一司吕。。
gbpg晴的
24足以向
口243周口问告骂
・(hn∞¥42-ERSHES陆)52号。丐闰ω-345τtgEDT〈ω仨旧。kcsgυω的-Eω50ω仨
队叮旨以向N叫N〈古卫〈七EKE(mm小时)E5548飞向hgd203旧。告gω的问唱主之。口。53hZ口。υωMHUNn矿罢
〈凶的Z〈时间〈时〈
Z〈vH的HZ〈EOH阳〈、Z〈扫的问凶的
口时』PE币的
Z〈’HhHEυOJ〈闰
Z〈HZ〈M问问 〈凶的Z〈H闰〈MH〈
Z〈M问问
Z〈M嗣同
This book
is
the result of a field research carried out
community of agriculturists
in
what was
till
by the author among
a
recently the uttermost part of
Southern Pakistani Balochistan. It
deals with themes such as ways ofliving and representing spaces, constructing
memory, the heritage of a form of social stratification which shaped community relationships in the last three centuries, and, last but not least, the insurgence
of nationalism.
Furthermore, the book puts forward some theoretical proposals about the translation of cultural“models”,throughout a constant
the author’s
and
comparison between
his interlocutors’,alternating ethnographic气descriptions”
with reflexive arguments. Notwithstanding forces
which
visited
its
reflect
remoteness, Balochistan
is
today
at the
both local and “global”logics, pushing
by few adventurous
which could impinge on
confluence of
this land,
once only
travelers, in the focus of international interests
political evolution of this sensitive area straddling
South Asia and Middle East.
Ugo
E.
M.
Fabietti
is
Professor of Anthropology at the University of
Milan Bicocca and Director of the PhD. Programme in Anthropology. He carried out research in Arabia ( 1979-1980) and Pakistani Balo chis tan 1986-1994), (
and
is
the author,
among other articles and books, of Antropologia
L’esperienza e l'interpre阳zione
culturale.
(Rome, 1999) which focuses on the main epis-
temological topics of anthropology.
巴QU
nKMM川川m M别们们们川Mm
m配呗w
8UUUUOP
3川川川门门qu咒 斗川川川川川4白
生们们们尸3阳 ;|了
「吨闹闹闹闹闹4μ’
3川剧目0吨
叫配EEMW少』刀
GRAD