Full text of "Hindus of the Himalayas"

See other formats

•

Digitized by the Internet Archive

in 2011 with funding from

LYRASIS Members and Sloan Foundation

http://www.archive.org/details/hindusofhimalayaOOinberr

Hindus

the

Himalayas

Berkeley and Los Angeles

Hindus

the

Himalayas

BY GERALD D. BERREMAN

UNI VERSITY OF CALIFORNIA PRESS 1963

UNIVERSITY OF CALIFORNIA PRESS

BERKELEY AND LOS ANGELES, CALIFORNIA

CAMBRIDGE UNIVERSITY PRESS

LONDON, ENGLAND

LIBRARY OF CONGRESS CATALOG CARD NUMBER: 63-IO463

PRINTED IN THE UNITED STATES OF AMERICA

To My Mother and Father

Acknowledgments

This book is based upon research first reported in my dissertation,
presented to the faculty of the Graduate School of Cornell University,
in September, 1959. I am grateful to Professor Morris E. Opler for his
interest and suggestions in supervising the preparation of the disserta-
tion and for his encouragement thereafter to revise and publish it.
Others who provided helpful advice include Professors Allan R. Holm-
berg, Lauriston Sharp, Robert J. Smith, and Robin M. Williams. Mr.
J. Michael Mahar and Professor S. C. Dube gave liberally of helpful
counsel before the research was undertaken. Mr. A. C. Chandola re-
viewed linguistic and other materials in the course of the final revision.
Mr. James M. Sebring checked census figures and numerous biblio-
graphic references and helped in proofreading. To all these people I
am grateful.

The Ford Foundation, through its Foreign Area Training Fellow-
ship program, financed nearly three years of work which led to the
dissertation, including 15 months in India during 1957-1958, and six
months of analysis and writing following the research. Library re-
search and substantial rewriting in the preparation of this book were
supported by a University of California Summer Faculty Research
Fellowship and by part-time research appointments with the Himalayan
Border Countries Research Project and the South Asia Village Studies
Project in the Center for South Asia Studies of the Institute of Inter-
national Studies, University of California, Berkeley. Dr. Joan V. Bondu-
rant, then chairman of the Center for South Asia Studies, Dr. Leo E.

Vlll

ACKNOWLEDGMENTS

Rose, head of the Himalayan Border Countries Research Project, and
Dr. William L. Rowe, my colleague in the South Asia Village Studies
Project, have been encouraging and helpful in this research. The De-
partment of Anthropology at the University of California, Berkeley,
has been cooperative in accommodating to my part-time research ap-
pointments. I wish to thank all of these individuals and institutions for
their material and moral support.

My family and I are indebted to many people who were kind and
helpful to us in India, especially to Mr. and Mrs. James Alters, Mr.
and Mrs. Leon Elliott, Mr. and Mrs. T. D. Fordham, and Mr. and Mrs.
J. Suvanto and their families, all of whom were friends in need. In my
research I benefited greatly from the assistance of Mr. A. P. Sharma and
Mr. M. Basir. Mr. Basir's unfailing interest and untiring efficiency were
important factors in the success of the research.

I am immeasurably indebted to my wife, Evelyn, who bore with good
humor the difficulties of life under trying circumstances; who managed
our household, cared for our small daughter in health and illness, and
typed untold hundreds of pages of reference material and notes — all
while I was preoccupied with research and most of the time while I
was physically absent. Our daughter, Janet, shared without complaint
the unstable life of the field anthropologist's family and the crises,
maladies, and discomforts inherent therein (and several not inherent
therein). Few families are called upon to endure what they endured,
and they did so with a grace deserving of respect and my deepest
gratitude.

Finally, I am grateful to the people of Sirkanda, without whose
trust, friendship, and forbearance the research would have been im-
possible. In particular I appreciate the kindness and understanding of
my friends, Alam Singh and Safri.

Contents

Introduction 1

The Setting 9

The Economic Context 38

The Religious Context: The Supernatural 80

The Religious Context: Calendrical and Life-Cycle

Ceremonies 121

Kin Groups and Kinship 143

Caste 197

7
Intercaste Relations 229

CONTENTS

The Village Community

259

The Outside World: Urban Contacts and Government

Programs 294

Conclusion

Bibliography

Appendixes

I. Gods Worshiped in Sirkanda
II. Calendrical Rites in Sirkanda

III. Life-Cycle Rites in Sirkanda

IV. Sirkanda Kin Terms

Notes
Index

339
359

369

388

395
410

415

423

MAPS

1. Linguistic and administrative environs of Sirkanda 13

2. Sirkanda marriage network and chans 24

3. Sirkanda village settlement area 25

He who thinks on Himachal, though he should not be-
hold him, is greater than he who performs all worship in Kashi.
In a hundred ages of the gods I could not tell thee of the glories
of Himachal. As the dew is dried up by the morning sun, so
are the sins of mankind by the sight of Himachal.

SKANDA PURANA

NotC ON HINDI TERMS

English equivalents have in most instances been substituted for Hindi
and Pahari terms. Where pertinent, the Hindi or Pahari word has
been included in parentheses. Where a Hindi or Pahari term has
seemed more appropriate than an English one, it has been italicized
and denned (if it is not denned by context) in its first appearance.
Such words have been written as nearly phonetically as possible, accord-
ing to Sirkanda pronunciation, in Roman script, with long vowels (— )
and nasalization (~) indicated in their first appearance. Retroflex con-
sonants have been indicated by capitalization in their first appear-
ance. Pluralization has usually been indicated as in English with a
final "s" rather than as in Hindi. Some words which appear frequently
in English writings are spelled in the conventional manner rather than
phonetically, especially if they are proper nouns.

Place names of villages within the immediate vicinity of Sirkanda
and all personal names of individuals (except those in the Acknowl-
edgments) are pseudonyms. "Sirkanda" is not the name of a real
village, though it is the name of a well-known Pahari temple dedi-
cated to Devi, the most important goddess in the village described
here. The employment of pseudonyms is done for protection of privacy
rather than for concealment. Anyone who set about it could easily
locate "Sirkanda."

INTRODUCTION

In India, as in peasant societies elsewhere, the village community has
been found by many anthropologists to be the most manageable unit
for ethnological research. As a result of their experience no one would
now be likely to assert that the village is isolated, independent, or un-
changing, or that it can be studied meaningfully without reference to
its past and its extensions into other communities, towns, and urban
centers. In fact, dependence upon market towns or urban areas is
inherent in most definitions of peasant communities. Redfield, for
example, describes the peasant as a "rural native whose long estab-
lished order of life takes important account of the city. The account
that the peasant takes of the city or town is economic, political and
moral" (Redfield, 1957, p. 31).

Although no peasant village can be understood in isolation, it can
serve as a useful focus for research. A village, with its extensions into
the surrounding region, constitutes a functioning segment of rural
Indian society of a size amenable to anthropological research tech-
niques. Although no village is representative of rural India as a whole,
any village properly understood represents in a general way villages
of a particular type and area. Dube has commented that "what we
need today is a series of studies of village communities from different
parts of the country covering the many divergent patterns of organiza-
tion and ethos. Until this is done our picture of social systems in rural
India will remain vague and inadequate" (Dube, 1955, pp. 6 f.). Only
then will we be able to assess the validity of generalizations already

2 INTRODUCTION

set forth on the subject. The present study is, among other things, a
contribution in this direction.

The research reported here was in the nature of a community study
carried out from September, 1957, through August, 1958, in and around
Sirkanda,1 a village of the lower Himalayas of western Uttar Pradesh,
India.

The aims of the study were three: (1) to provide an ethnographic
community study in an important and previously unreported culture
area of India; (2) to analyze the functioning and interrelationship of
kin, caste, and community ties in a Hindu society known to be dif-
ferently organized in some significant respects than those of the ad-
jacent and well-known plains; and (3) to study the effects of recent
governmental programs and other outside contacts on a relatively
isolated and conservative Indian community.

The area selected for the research was the hill region near the town
of Dehra Dun. It was chosen over other, equally promising hill regions
because of ease of access, availability of competent medical care, and
adequate housing in nearby Dehra Dun — features lacking in many
hill areas but important in view of my plan to take my wife and in-
fant daughter. Other factors were that I had an advance contact with
a potential interpreter in the town of Dehra Dun and that town, as
district headquarters and an educational center, had useful libraries
and other documentary resources. The particular locality and com-
munity were chosen as suitable in terms of the combined requirements
of the type of community to be investigated, the plan of the research,
and the necessity of remaining within a reasonable distance of the
facilities of Dehra Dun.

In selecting a site for the research, I visited some 25 villages over a
period of two months, both in the hills and in the Dehra Dun valley
bordering the hills. Sirkanda, a village ten air miles from Dehra Dun,
was attractive in that it was large for a hill village, was fairly isolated
but accessible, had a caste distribution typical of the area, and had no
apparent important atypical features. During the period of research
I lived most of the time in Sirkanda and simultaneously maintained
a house in Dehra Dun or, for the last four months, in the nearby hill
town of Mussoorie. Except on visits to the village, my family occupied
the house in town. My characteristic routine during the research
period was to spend four or five days in the village followed by two
days in town. This provided a weekly opportunity to type notes, con-
sult library resources and official records, secure supplies, and get
needed rest.

In addition to my residence in Sirkanda, I had fairly extensive con-

INTRODUCTION 3

tacts with other hill villages in the immediate area. I passed through
other villages and intermediary markets en route to and from Sir-
kanda, and I paid occasional visits to surrounding villages, often in
the company of Sirkanda villagers. Contrasting hill regions were ob-
served in two treks, one from Simla to Chakrata, a no-mile trip in
Himachal Pradesh and Jaunsar-Bawar, to the west. The other was
from Mussoorie east to Tehri, a 40-mile trip through the heart of the
culture area represented by the people of Sirkanda. From these con-
tacts and similar ones in the Dehra Dun valley, I gained a first-hand
perspective on the place of Sirkanda in its own culture area and its
relation to neighboring ones.

Living conditions in the village were primitive even by local stand-
ards— a fact which led me to reject the idea of moving my family there
on a full-time basis. Except for a period of about four months when I
owned an unreliable jeep and could drive by a devious route to within
five miles of the village, access from Dehra Dun was by a foot trail nine
miles beyond the end of a bus line. The last five-mile distance to the
village was along a mountain trail climbing 2,700 feet. Access from
Mussoorie, location of my "town house" for the last four months, was
by a relatively level, but in places extremely rough, trail of 16 miles.
My village house consisted of three small connecting rooms, one of
which was occupied continuously by two to four water buffalo, and all
of which were inferior to those inhabited by most villagers. The only
food consistently available in the village was grain, milk, sugar, and
tea. The source of water was one-fourth of a mile distant over a rocky
trail. In short, Sirkanda was not suitable for protracted family living
by outsiders.

Throughout the research the services of an interpreter were utilized.
The language of the people is the Central dialect of the language group
known as Pahari. This is related to Hindi, and most villagers knew
Hindi well. I knew elementary Hindi and was able to carry on or-
dinary conversation and to comprehend much of what was being said
by informants. However, for research purposes an assistant conversant
with the language was needed. Two interpreters worked with me con-
secutively.

Problems of the research, especially as it was affected by the con-
trasting personal and social characteristics of these two men, have been
the subject of a short monograph (Berreman, 1962c). Suffice it to say
here that villagers were at first suspicious of us and our motives. The
nature of the rapport we established and consequently of the informa-
tion we obtained was heavily influenced by the villagers' perceptions
of us. The first interpreter, a young Brahmin, was helpful in establish-

A INTRODUCTION

ing friendly relations but, partly because of his status as a Brahmin of
plains origin, he inhibited people in their presentation of themselves
to us. Villagers are self-conscious about many of their religious and
social practices, which are unorthodox and even defiling by plains
standards. They were anxious to conceal these from the young Brahmin
in order to secure his respect. When it unexpectedly became necessary
to find a replacement as interpreter, a retired Muslim schoolteacher
was employed. Although his religion prevented him from achieving
close friendship with most high-caste villagers, he won remarkably
good rapport with villagers of low caste who, in any case, were the
best informants. Moreover, neither he nor the villagers felt compelled
to impress one another with their status or ritual purity. This led to
more candid information and observation among all groups than had
been possible when the Brahmin was present.

The primary sources of data were Sirkanda villagers themselves.
Initially they were, as a result of their suspicions, an unreliable source,
and many of the data obtained were in the nature of half-truths. In
the beginning there was consistent suppression of many kinds of in-
formation, especially that which the villagers felt reflected adversely
on their status as Hindus, and that which they believed might result
in additional taxation, legal proceedings, or other governmental in-
terference. Thus their subjects of conversation with me were limited
to the weather, agricultural techniques, and similarly innocuous topics.
Gradually we won their confidence and they gave information more
freely. A few individuals remained suspicious and uncommunicative
throughout the period of research; others became friendly but not
informative; some became both friendly and informative. In virtually
all instances, however, it was possible to check information with two
or more informants. Most villagers were at least occasional informants.

Low-caste people were freer with most kinds of information than
were high-caste people, apparently because they had a smaller stake in
the impressions others received of them and because they did not feel
threatened by the anthropologist and his interpreter. Low-caste vil-
lagers contributed a relatively large amount of information per capita,
but this was compensated for by the greater numbers of high-caste
informants. Information was always obtained from or checked by those
in a position to know whereof they spoke on a particular topic. In the
year of research there was an opportunity to observe many events at
first hand. However, some of those to be described, and especially some
religious and ceremonial activities, did not occur in my presence. De-
scriptions of such events were obtained from people who had par-
ticipated in them.

INTRODUCTION 5

Information was obtained primarily from observation and through
informal interviews, both directed and undirected. Interviews were
conducted wherever and whenever the occasion arose — in my house
or yard, at the houses of villagers, in the shops, at places of work of
artisans, in the fields, at the water source, on the trails, and so on.
Generally several people were present at once, contributing informa-
tion and attitudes and inhibiting one another's contributions. Such
interviews were not recorded on the spot. At the first opportunity they
were recorded in private. The approach could not be called participant
observation because real participation was a virtual impossibility in
most contexts. My interpreter and I were outsiders. Our participation
was limited largely to informal social situations, and even there we
were usually in the role of guests, invited or uninvited.

As time passed villagers became interested in our work and ac-
customed to our queries, and some were willing to have interviews re-
corded in their presence. This was done in the unreliable privacy of
my village house. Some objective and largely quantitative types of
data were gathered, notebook in hand, from all the villagers or from
a cross-section. These included genealogical and census materials,
economic data, and the like. Scheduled interviews had been drawn up
to get at attitudes toward the outside world and related matters, but
they were discarded. I felt, on the basis of experience in informal in-
terviewing and in collecting genealogical and other information, that
these would do more harm in terms of rapport than good in terms of
reliable data. More was to be gained, I believed, by getting this in-
formation informally than by drawing attention (and inevitably,
suspicion) to my interests by attempting a survey approach. This is an
unusually closed society, whose members are accustomed to concealing
and protecting themselves from the outside world. Data obtained by
survey techniques on matters about which villagers are sensitive
would, in all probability, have been of little value except as a kind of
projective test, and any attempt to secure such data would have jeo-
pardized further research efforts.

The difficulty of obtaining reliable data is reflected in the quality
of the official statistical sources of information presumably available
on the village. There are such records as voting lists, birth and death
records, census lists, school enrollment lists, livestock census, land
records, and tax records, which provide neat statistical summaries.
However, much of the information is inaccurate. On the current voting
list, for example, 142 names appear accompanied by the person's age
and name of father or husband. Among these are three people who
appear twice, at least eight who are incorrectly identified, seven who

6 INTRODUCTION

are unidentified, and at least two who appear also on the voting list
of a neighboring village. The list includes a mixture of full-time and
part-time residents and people who no longer reside in the village,
while some individuals in each of these categories are missing from the
list. Ages given are at best rough approximations. The village live-
stock census was conducted by a forestry officer, who was reported to
have looked at the cattle in the yards of one or two houses and mul-
tiplied by his guess as to the number of houses in the village. Land
records are more accurate, since they are based on measurements, but
they are inadequate to show current land distribution because informal
division of land has long gone unrecorded. The man responsible for
recording births and deaths in Sirkanda lives in another village and
misses many births and cases of infant and child mortality. Even the
school enrollment record which is kept in the village includes four
students whom the teacher has never seen (these are excluded from all
discussions and calculations with regard to the school in later chapters).
The record fails to list six students who attend regularly.

In view of the difficulties faced by those responsible for keeping
records both in terms of the reliability of their informants and the
pressure from their superiors, it is remarkable that they do as well as
they do. But documentary records are less useful sources of data for
research than might be expected. Each type of record has its own
specific types of limitations. Reliance upon such records is misleading
unless their limitations are recognized.

Literature dealing with Central Pahari-speaking people is extremely
scarce. The earliest account, and an indispensable one, is G. W. Traill's
"Statistical Sketch of Kamaon" (1828). E. T. Atkinson's three-volume
work, The Himalayan Districts of the North-Western Provinces of
India (1882-1886), is a comprehensive and accurate report on the area
and the people. It is the only such source, and most subsequent ac-
counts rely heavily upon it. To these two insightful civil servants we
owe most of what we know of these people. G. A. Grierson in the
Linguistic Survey of India (Vol. IX, 1916) provides invaluable lin-
guistic and historical data. E. S. Oakley's Holy Himalaya (1905) also
contains interesting material. Other useful works are the District
Gazetteers of Dehra Dun, British Garhwal, and Almora (the latter two
largely revisions of pertinent sections of Atkinson's volumes) by H. G.
Walton (1911a; 1910; 1911b), the Historical and Statistical Memoir
of Dehra Dun by G. R. C. Williams (1874), and to a lesser extent the
District Gazetteer of Naini Tal by H. R. Nevill (1904).

An informative informal description by one who knew the people
well is that of "Mr. Wilson of Mussoorie," contained in A Summer

INTRODUCTION 7

Ramble in the Himalayas, edited by Wilson under the nom de plume
of "the Mountaineer" (i860, pp. 121-232). Useful works by local his-
torians are: Garhwal, Ancient and Modern, by P. R. Bahadur (1916),
History of Garhwal (in Hindi) by H. K. Raturi (1928), and Kumaon
(in Hindi) by R. Sankrityayana (1958). L. D. Joshi (1929) compiled a
valuable account of customary law — primarily family law — in the
Central Pahari area, and V. A. Stowell (1907) contributed a source
book on land tenures.

With the exception of a brief appendix to the Gazetteer of British
Garhwal, none of the above deals specifically with Tehri Garhwal, the
administrative unit immediately adjacent, and for many purposes
most relevant, to Sirkanda. The books cited do, however, deal with
closely related peoples in neighboring Garhwal and the Kumaon dis-
tricts to the east.

The Pahari area is populated by five million Indo-Aryan speaking
Hindus. It is famous as the location of some of the most important
Hindu shrines, as the setting for widely known religious epics, as a
traditional region of retreat for religious figures, and as the site of
many points of historic interest. It is the home of the men who made
up the Garhwal Rifles, who won fame in two World Wars. It is the
location of "hill stations," where thousands of people from all over
North India go annually to escape the summer heat, and where state
and national governments and international political conferences have
met. It is the source of an increasing variety of natural resources. Yet
it has remained a virtual blank on the ethnographic map and is ig-
nored by most of those who write about North Indian culture and
society.

The situation has changed little since 1935, when S. D. Pant noted
that, "The Himalayas have for many years constituted a subject of
such profound and unique interest, and have attracted so distinguished
a company of explorers and investigators, that it seems all but incredible
that no adequate scientific account of the human geography and social
economy of the Himalayans has yet been written" (Pant, 1935, p. 9).
Pant himself has provided the first exception to this dearth of scien-
tific work in the area with his book, The Social Economy of the Hi-
malayans, a valuable general survey of Almora and Naini Tal districts
focusing upon the economy of the Paharis and Bhotiyas. This is the
only contemporary study conducted even partially among Central
Pahari-speaking peoples.

The only Himalayan hill people to have attracted the attention of
Indian anthropologists or sociologists are the polyandrous residents
of Jaunsar-Bawar, a small area in Dehra Dun District inhabited by

g INTRODUCTION

Western Pahari-speaking people. Evidently attracted by unusual mari-
tal customs and related features which have led to characterization of
the society as "a fossil of the age of the Mahabharata" (Munshi, 1955,
p. i), D. N. Majumdar and his students, notably R. N. Saksena, have
done some work in that region (Majumdar, 1944; Saksena, 1955).

Recently short reports of village studies carried out in hill tracts
farther to the west, in Chamba and Kulu, have appeared, by W. H.
Newell (1955) and C. Rosser (1955), respectively, and a book by Rosser,
The Valley of the Gods, is in press. The area is certain to receive fur-
ther attention.

The present study is, therefore, the first community study or ethnog-
raphy attempted in the Central Pahari area, and one of very few
studies of any kind that have dealt with Himalayan hill people. Be-
cause of this it has seemed worthwhile to include in this account a good
deal of ethnographic material which could have been omitted if it
were available elsewhere. There is no independent documentation for
much that is presented here. Without documentation and cultural
context, the reader would be faced with the problem which faced the
researcher in the initial phases of his work — he might fail to see ex-
isting relationships and he might assume nonexistent ones. The reader
familiar with India would be especially likely to jump to erroneous
conclusions about the degree of similarity or difference between the
community described here and other Indian communities with which
he might be acquainted. This would be likely to prejudice his estimate
of the credibility and applicability of the generalizations, analyses, and
conclusions presented. Intelligent appraisal can follow only from data
presented in cultural context. For this reason I have tried to present
some of the empirical data upon which these generalizations, analyses,
and conclusions are based.

The aims of this book, then, are both ethnographic and analytic: the
description of an example of a heretofore undescribed culture; the
analysis of social organization in a uniquely organized caste society;
and the analysis of reactions to planned and unplanned change in a
remote and unsophisticated village. These aims have been important
in determining the content and manner of presentation of the ma-
terials to follow.

THE SETTING

The lower Himalayas from western Kashmir to eastern Nepal are
populated by peoples sharing common and distinct cultural, linguistic,
and historical traditions. This study deals primarily with the inhabit-
ants of Sirkanda, one village in this broad culture area, and with their
neighbors and relatives in nearby villages. Sirkanda lies in the Hi-
malayan foothills bordering Dehra Dun and Tehri Garhwal districts in
Uttar Pradesh, India, 125 air miles north and slightly east of Delhi,
and 85 miles southwest of the Tibetan border. It is about ten air miles
northeast of Dehra Dun, a large town situated in a valley at the foot
of the Himalayan range, and seven miles east of Mussoorie, a British-
built hill station on the crest of the first ridge of the Himalayas north
of Dehra Dun. Distances in this region are deceptive because the ter-
rain is rugged and mountainous. The only means of travel within the
area is by mountain trails which average at least double the air-mile
distances. There are adults in Sirkanda who have never visited Dehra
Dun.

Natural Features

The Kumaon Himalaya or Himalaya West comprises the central
portion of the mountainous expanse from Kashmir to Darjeeling. It
constitutes the northernmost part of the Indian state of Uttar Pradesh
and is bordered on the east by Nepal, on the north by Tibet and on
the west by the state of Himachal Pradesh.1 It rises abruptly from the

THE SETTING

Gangetic plain at scarcely 1,000 feet above sea level to the perpetual
snows of the passes into Tibet at 16,000 to 18,000 feet and the peaks of
Bandarpunch, Trisul, Nanda Devi, and others reaching 20,000 to
25,000 feet. The main axis of the high or snowy Himalayan range is
an arc from Kashmir in the northwest, through and beyond Nepal in
the east. The region of interest here comprises the lower and outer
ranges known as the Sub-Himalayas, and specifically those portions of
the Kumaon Himalaya lying predominately below 10,000 feet and often
called the Kumaon hills.

Topographically the Kumaon hill area is one of mountains and
rivers. Since the mountains are geologically young they are precipitous
and rocky. The valleys are steep and narrow, the streams and rivers are
swift, and in the rainy season they become rushing torrents. The
mountainous area is separated from the plains to the south by barren,
rocky talus slopes below which is a strip of forest and marshland.

Along part of their length the Himalayas are paralleled a few miles
to the south by remnants of the ancient Siwalik range. Between these
ranges lie narrow valleys such as that of Dehra Dun, bordering the
Himalayan foothills for 45 miles and separated from the Gangetic
plain by the low but rugged Siwalik hills. Here, as in popular usage in
the area, "the valley" will be used to refer to the Dehra Dun valley,
some of the eastern portion of which is visible from Sirkanda. The
valley is about 2,000 feet above sea level, while the Gangetic plain
beyond the Siwaliks is not over 1,000 feet in elevation. It is the latter
plain that is here termed "the plains," in accordance with local usage.

Throughout the Himalayas, population density is restricted by the
terrain. Prime requisites for occupation by the agriculturists who
inhabit these mountains are topsoil that can be terraced and a steady
water supply. Alluvial fans, gradual slopes, or broad valleys are ex-
ceptionally favorable spots for habitation. Most of the population is
found in valleys and on the lower slopes of the mountains below an
elevation of 6,000 feet.

The Kumaon Himalaya lies across 300 north latitude and is there-
fore within the Temperate Zone not far from the tropics. Climate is
influenced by the high Himalayas to the north and the Gangetic plain
to the south. Three seasons are recognized and terminologically dis-
tinguished by the inhabitants: winter, hot season or summer, and rainy
season. Winter lasts from mid-October to mid-February and is char-
acterized by clear, cool weather, followed by colder weather and some
precipitation in December, January, and February. Temperature varies
with altitude, so that even during the coldest period little snow falls

THE SETTING 11

below the 6,000-foot level while considerable falls above that level. In
Sirkanda for about a month in winter, temperatures often fall slightly
below freezing at night, but they usually rise considerably during the
day. Summer begins in mid-February and extends through the middle
of June. In March and April localized storms occur. By mid-April the
hot, dry season is at hand and lasts until the monsoon breaks in the
latter half of June. Rains continue to come frequently and heavily and
temperatures remain fairly high until mid-September, when tempera-
tures begin to drop and rains cease. Annual rainfall varies greatly,
with the outer range of mountains catching the heaviest precipitation
as the clouds move in from the plains. Annual rainfall in Mussoorie
averages around 80 inches, while other hill locations report rainfall of
40 to 100 inches per year. Vegetation is profoundly influenced by al-
titude and rainfall.

Generally the slopes which face northward are more thickly wooded than
the southward slopes, as in the former the sun's rays only slant across the
surface, and the moisture is retained in the soil for a longer period. With the
decrease in elevation a gradual change in the composition of the forest
is observed. . . . The slopes above the Kali . . . present an excellent epitome
of this plant variation. One passes from the shisham (Dalbergia sissoo) and
sal (Shorea robusta) in the river beds, through oak and rhododendron on the
high hills, to firs, birch, and box on the still higher central ridges. (Pant,
1935. P- 37)

The fauna is varied and includes deer, goats, pigs, monkeys, foxes,
jackals, porcupines, bears, leopards, and an occasional tiger. A wide
variety of birds is to be found, including native wild chickens. The
streams abound in Indian trout. The wild life and terrain of the Ku-
maon hills have been vividly depicted in the well-known works of Jim
Corbett (1946), while the fauna has been described in detail by Atkinson
(1884a, pp. 1-266).

Administrative History

The western half of the Kumaon Himalaya is known as Garhwal
— the country of fortresses — after the large stone structures found
scattered through the region. It comprises the present districts of
Garhwal, Tehri Garhwal, and the Himalayan hill area of Dehra Dun
District exclusive of Jaunsar-Bawar (see map 1). Almora and Naini
Tal districts to the east are often called the "Kumaon districts" when
contrasted to Garhwal. In the Hindu scriptures Garhwal is referred to
as Kedarkhand. Kedarkhand is important in Hindu religion and my-

12 THE SETTING

thology as the place of origin of the two sacred rivers Ganges and Jumna
and as the site of famous places of pilgrimage including Hardwar,
Rishikesh, Kedarnath, Badrinath, Jamnotri, and Gangotri.

Rishis [sages] and ascetics in large numbers resorted to its silent valleys
for purposes of meditation or the instruction of their disciples. The final
scene of the life of the five heroes of the Mahabharata was enacted amid its
mountains and many place-names in the Alaknanda valley still recall the
memory of BhimSen and his brothers. Garhwal may still claim to be the
holy land of India: its valleys are full of ancient temples and there is scarcely
a ridge from which the wonderful spectacle of the snowy range is visible
without its humble shrine. Every year thousands of pilgrims from all parts
of India make their laborious way on foot along the via sacra of Badrinath.
(Bahadur, 1916, p. ii)

Little is known of the early history of Garhwal and the surrounding
areas either through written sources or local tradition.

Up to the time of Ajaiya Pala [about a.d. 1358] Garhwal was divided
amongst a number of petty Rajas. Every glen or hill, as formerly was the
case in the highlands of Scotland, was subject to its own chiefs who have
left no record behind except the moss-covered walls of their strongholds.
And although Ajaiya Pala is credited with having reduced fifty-two of these
petty chiefs under his own rule, we may well suppose that he was only the
first of his line to aim at more than a local supremacy, and that to his suc-
cessors is due the extension of the Garhwal power over the Dun, Bisahir,
and the tract now known as Tihri or foreign Garhwal. (Atkinson, 1884a, pp.
526-527)

For the end of the sixteenth century and later, more precise records
and dates are available. Then Garhwal appeared as an independent
kingdom at its greatest recorded extent. It remained thus until 1803-
1805, when the Garhwal Raja, Pradhuman Shah, was defeated and
killed in battle by invading Nepalese armies which pushed westward,
extending Nepalese suzerainty to the Sutlej River. Although the Nepa-
lese acquired a reputation for harsh, exploitative rule which reportedly
led to mass emigration out of the hills, their eleven-year rule left few
imprints in relatively inaccessible localities such as Sirkanda.

The British defeated the Nepalese in 1815 and, in taking control
from them, divided Garhwal into its present parts. The eastern half,
comprising 5,629 square miles, was placed under direct British rule
and was thenceforth known as British Garhwal or simply Garhwal.
For administrative purposes it was included with the districts of Naini
Tal and Almora, to its east, to form the Kumaon Division. The Dehra
Dun valley and adjacent hills in the southwestern corner of Garhwal
and the hill tract called Jaunsar-Bawar west of Garhwal were annexed

CENTRAL PAHARI DISTRICTS
(HIMALAYA WEST)

1. Dehro Dun District

2 Tehri Garhwal District

3 Garhwal District

4 Almora District

5 Naini Tal District

LOCAL AREAS

o io 20 miles

Cultural-linguistic

boundary
Administrative boundary

Map 1. Linguistic and administrative environs of Sirkanda. Partly adapted
from Grierson (1916), pp. 101, 373.

14 THE SETTING

to Saharanpur District, which was already under direct British rule.
Later the annexed sections became Dehra Dun District. The remaining
large portion of Garhwal, comprising 4,200 square miles between
British Garhwal and Dehra Dun District, was ceded to Sudershan Shah,
the son of the last of the Garhwal Rajas (who had died in the Nepalese
war). This section, called Tehri Garhwal, remained an independent
princely state under succeeding rajas until Indian Independence (1947),
when, after some delay, it was incorporated into Uttar Pradesh. Since
Independence these areas have remained administratively separated
as districts in Uttar Pradesh but are, of course, under a common law.
Throughout the period of recorded history, the inhabitants of Garhwal
have retained their social and cultural ties despite administrative
division of the area.

The People and Their History

The people of the Sub-Himalayan hills from western Kashmir to
eastern Nepal are referred to by the generic term Pahari (of the moun-
tains). While not a particularly precise term, it is a useful one and is
recognized throughout North India. Two major ancestral stocks are
generally believed to have contributed to the present Pahari popula-
tion. One, often assumed to have been an early, indigenous group,
now appears as the Dom or low castes. The other, described as an
Indo-Aryan speaking group, is presumably more recent and of Central
Asian origin. Its descendants, called Khasa or Khasiya, comprise the
present high castes of the hills. The term "Khasiya" is used by the
people in the region of this study, while "Khasa" is more frequent in
the literature about Paharis. Throughout this account, the Bhotlya,
a Mongoloid, Tibeto-Burmese speaking, and culturally distinct people
of the higher Himalayas are excluded from the discussion except where
specific reference is made to them.

Both Khasas and Doms are often described as internally relatively
undifferentiated. Khasas are divided into Brahmin and Rajput groups,
but interaction is more intimate between them than is usual on the
plains. Even intermarriage is tolerated. Doms are divided into several
endogamous groups ranked relative to one another and associated
with occupational specialties. However, occupational specialization is
remarkably variable, and many accounts describe them as formerly
less differentiated. One might speculate that at one time there were
two relatively homogeneous groups, the dominant, agricultural Khasas
and the dependent, depressed artisan or service group known as Doms.
These were probably groups of different ethnic affinities, but they

THE SETTING 15

could have been status groups originating from a common source.
Internal differentiation within each may have resulted from subse-
quent contacts with plains peoples — perhaps immigrants to the hills.
As a result of such contacts Khasas took the names and other status
characteristics of Brahmins and Rajputs, or in some areas they may
all have become "Rajputs" while the immigrants from the plains were
Brahmins. Meanwhile Doms may have subdivided according to oc-
cupational specialty as a result of high-caste expectations or their own
adoption of plains attitudes, or as a result of an influx of artisans from
the plains. Around Sirkanda it seems not improbable that the caste
of drummer-tailor-basketmakers (Bajgi) may be descended from the
archetypical "Doms" while the other specialties (such as blacksmiths)
were derived from immigrant groups or possibly from specialization
within the old Dom group (such as weavers). This, like the widespread
assertion that Doms preceded Khasas, is speculation and should be
interpreted as nothing more. With present evidence no better can be
done. We can safely say only that the origins and affinities of con-
temporary Pahari castes and occupational groups are largely unknown,
and that this fact has stimulated conjecture. Such conjecture has
centered most heavily on the Khasas, who are dominant in numbers,
wealth, and status and about whom some historical information is at
hand.

Khasas

"Sanskrit literature contains frequent references to a tribe whose
name is usually spelt Khasa, with variants such as Khasa, Khasha, and
Khasira. The earlier we trace notices regarding them, the further
northwest we find them" (Grierson, 1916, p. 2). They appear frequently
in the Puranas (ancient Hindu literature), and they figure prominently
in the Mahabharata.

We gather that according to the most ancient Indian authorities in the
extreme north-west of India, on the Hindu Kush and the mountainous
tracts to the south, and in the Western Panjab there was a group of tribes
one of which was called Khasa, which were looked upon as Kshatriyas of
Aryan origin. These spoke a language closely allied to Sanskrit. . . . They
were considered to have lost their claim to consideration as Aryans, and to
have become Mlechchhas, or barbarians, owing to their non-observance of
the rules for eating and drinking observed by the Sanskritic peoples of India.
These Khasas were a warlike tribe, and were well known to classical writers,
who noted, as their special home, the Indian Caucasus of Pliny. . . .

It is probable that they once occupied an important position in Central
Asia, and the Kashgar of Chitral were named after them. They were closely

l6 THE SETTING

connected with the group of tribes nicknamed "PiSachas" or "cannibals" by
Indian writers, and before the sixth century they were stated to speak the
same language as the people of Balkh. At the same period they had ap-
parently penetrated along the southern slope of the Himalaya as far east
as Nepal, and in the twelfth century they certainly occupied in considerable
force the hills to the south, south-west and south-east of Kashmir.

At the present day their descendants, and the tribes who claim descent
from them, occupy a much wider area. (Grierson, 1916, pp. 7-8)

The languages of the hill regions, like the peoples, are termed Pahari.

The word "Pahari" means "of or belonging to the mountains," and is
specially applied to the groups of languages spoken in the sub-Himalayan
hills extending from Bhadrawah, north of the Pan jab, to the eastern parts
of Nepal. To its North and East various Himalayan Tibeto-Burman languages
are spoken. To its west there are Aryan languages connected with Kashmiri
and Western Panjabi, and to its south it has the Aryan languages of the
Panjab and the Gangetic plain, viz.: — in order from West to East, Panjabi,
Western Hindi, Eastern Hindi and Bihari.

The Pahari languages fall into three main groups. In the extreme East
there is Khas-Kura or Eastern Pahari, commonly called Naipali, the Aryan
language spoken in Nepal. Next, in Kumaon and Garhwal, we have the
Central Pahari language, Kumauni and Garhwali. Finally in the West we
have the Western Pahari languages spoken in Jaunsar-Bawar, the Simla
Hill States, Kulu, Mandi and Suket, Chamba, and Western Kashmir. (Grier-
son, 1916, p. 1)

"It is a remarkable fact that, although Pahari has little connexion
with the Panjabi, Western and Eastern Hindi, and Bihari spoken im-
mediately to its south, it shows manifold traces of intimate relation-
ship with the languages of Rajputana" (Grierson, 1916, p. 2). There
is general agreement that the relationship between the Pahari and
Rajasthani languages is attributable to the movement of peoples,
notably the Gurjaras, between these two areas.

The earliest immigrants [into the Pahari tract] of whom we have any
historical information were the Khasas, . . . They were followed by the
Gurjaras, a tribe who invaded India about the sixth century a.d. and oc-
cupied the same tract, then known as Sapadalaksha. At that time, they
[like the Khasas] spoke an Aryan, but not necessarily Indo-Aryan language.
Of these Gurjaras the bulk followed pastoral pursuits and became merged in
and identified with the preceding Khasa population. Others were fighting
men, and were identified by the Brahmans [priestly class] with Kshatriyas
[warrior class]. In this guise they invaded Eastern Rajputana from Sapadalak-
sha and, possibly, Western Rajputana from Sindh, and founded, as Rajputs,
the great Rajput states of Rajputana. (Grierson, 1916, p. 14)

THE SETTING 17

Vincent Smith, quoted by Grierson, differs in his explanation of the
sequence in this relationship, but not in the fact of the relationship:

The Gujars, etc., of the lower Himalayas who now speak forms of Rajasthani
are in large measure of the same stock as many Rajput clans in Rajputana,
the Panjab and the United Provinces; . . . their ancestors emigrated from
Rajputana after they had acquired the Rajasthani speech; and . . . the most
likely time for such emigration is the ninth century, when the Gurjara-
Rajput power dominated all northern and northwestern India . . . (Grier-
son, 1916, p. 12; emphasis added)

In either case ". . . it is plain that down even to the days of late
Musalman dominion the tie between Sapadalaksha [the Western
Pahari-speaking areas] and Rajputana was never broken. And this, in
my opinion, satisfactorily explains the fact of the close connexion be-
tween the Pahari languages and Rajasthani" (Grierson, 1916, p. 13).

The Khasas were apparently among those participating in the large
movements of Aryan-speaking peoples into India. Therefore it is
probable that they entered between 1500 and 1000 b.c. from the north-
west. There is not universal agreement among scholars that this was
the route followed or even that there was large-scale immigration then.
F. E. Pargiter concludes from his research that "tradition or myth in-
dicates that the Ailas (or Aryans) entered India from the mid-
Himalayan region, and its attitude towards the N. W. frontier lends
no support to any invasion from that quarter" (Pargiter, 1922, p. 299).
Thus he believes that entry was from Tibet through Garhwal into
India. Probably archaeology can best resolve this question.

It is evident that the Khasas have been in the Sub-Himalayan hills
for a very long time. It is also true that there has been a continuing
immigration of people from the plains who have become absorbed
into the hill population. Grierson notes that the Gujars not only
emigrated from the hills to the plains, but

. . . there was a constant reflux of emigration on the part of the Gujar-
Rajputs from Rajputana and the neighbouring parts of India. These re-
immigrants became, as befitted their Kshatriya station, the rulers of the
country and today most of the chiefs and princes of die old Sapadalaksha
trace their descent from Rajputs of the plains. The re-immigration was
increased by the oppression of the Mughul rule in India proper, and there
are historical notices of tribe after tribe, and leader after leader, abandoning
their established seats in Rajputana and seeking refuge from Musalman
oppression in the hills from which they had originally issued to conquer the
Gangetic valley. (Grierson, 1916, pp. 14 f.)

l8 THE SETTING

Other non-Pahari people from the plains adjacent to the Himalayas
also frequently sought refuge in the hills and became merged with the
hill population.

Contact has long existed between Paharis and Tibeto-Burmese
speaking peoples, often referred to as Bhotiyas or Tibetans, who oc-
cupy the higher Himalayas bordering the Pahari region along its
northern periphery. In border areas physical and cultural intermin-
gling occurs. In the Western and Central Pahari regions such admix-
ture has been minimal, while to the east, especially in Nepal, it has
occurred to a much greater extent.

The physical appearance of the high-caste Paharis of the Central
and Western hills is often described in idealized fashion, as in these
words of a contemporary anthropologist:

The Khasas are usually tall, handsome, fair-complexioned (rosy or sallow),
possess a long head, vertical forehead, fine or leptorhine nose, hazel eyes
with a sprinkling of blue, curly hair and other features well-cut and pro-
portioned. The women are also comparatively tall, slender, graceful, of a
very attractive appearance and of extremely gay disposition. (Majumdar,
1944, p. 110)

The number of exceptions to such a description probably exceeds the
examples, and the range in physical types overlaps greatly with that
found in the Gangetic plain. However, there is perhaps a statistically
significant tendency in the direction suggested by Majumdar. Some
observers claim to find physical differences among contiguous popula-
tions of high-caste Paharis. Thus R. N. Saksena's description of the
Khasas of Jaunsar-Bawar: "Their physical features, fair complexion,
tall stature, aquiline nose and well-defined features of the face easily
distinguish them from their neighbours, the Garhwalis" (Saksena,
1955, p. 9). Such statements are based on stereotypes rather than ob-
servation.

The Khasas have often been referred to as a "tribal" people. This
term has not been defined in such a way as to include satisfactorily the
diverse groups it is often used to designate nor to exclude many of the
diverse groups it excludes. However, in India it usually refers to
peoples who are not Hindus or Muslims or followers of other major
religions, who do not have caste organization, or who practice a more
primitive economy than that of most Indian communities.

Whatever their origin, the Khasas of today cannot be considered a
tribal people by any of these criteria. They certainly are not aborigines
as that term is used in India to denote non-Hindu "original sons of
the soil," who live in a "primitive state of existence."

THE SETTING 19

The great mass of the population in Kumaon and Garhwal profess a
belief little differing from the orthodox Hinduism of the plains. . . . All
their feelings and prejudices are so strongly imbued with the peculiar spirit
of Hinduism that although their social habits and religious belief are often
repugnant to those who strictly observe the orthodox ceremonial usages of
Hinduism, it is impossible for any one that knows them to consider the
Khasas to be other than Hindus. There are several facts connected with
their history that show, whatever their origin may have been, the Khasas
have for centuries been under the influence of the Brahmanical priesthood.
(Atkinson, 1884a, p. 269)

Their unorthodox practice of Hinduism is well-known and long-
recognized. Grierson refers to the Laws of Manu: "Looking at the
Khasas from the Brahmanical point of view, he says (X, 22) that Khasas
are the offspring of outcaste Kshatriyas, and again (X, 44) ... he says
that [among other tribes, the] Khasas are those who became outcaste
through having neglected their religious duties . . ." (Grierson, 1916,
p. 5). "Even in the most orthodox writings the Khasas are looked on
more as heretical members of the great Aryan family than as outcaste
aborigines, and . . . from a very early period they have been recog-
nized as an important tribe in Upper India" (Atkinson, 1884a, p. 283).

Today the Khasas universally define themselves as Hindus who are
in some respects culturally distinct from others of their religion and
caste. This they attribute primarily to the rigors of life in the hills.
They display no pan-Khasa unity beyond recognition of the term and
certain cultural similarities. Recently there has been some pressure in
the Western Pahari area to form a separate hill state, but that is a
post-Independence phenomenon of educated hill men.

It is worth noting that in many areas, including that of this research
and Jaunsar-Bawar to the northwest, only Rajputs (Kshatriyas) refer to
themselves as Khasiya or Khasa. Brahmins, though apparently of the
same stock, do not there admit to the designation. In fact the term
Khasa refers historically to a Kshatriya group. However, Majumdar
(1944, p. 110) notes that "the Khasas or the Khasiyas who constitute
the high caste people of the cis-Himalayan region are either Rajput
or Brahmin. . . ." "Khasiya Brahmans" are referred to by Atkinson
(1886, p. 430), who states that, "Nearly ninety per cent of the Brahmans
in Kumaon belong to the Khasiya race and are so classed by the people
themselves. . . ." "Khas-Brahmins" are found throughout the Pahari
regions both as landowning and tilling agriculturists, like the Rajputs,
and as priests. Atkinson (1884a, p. 734) remarks that these Brahmins
really have no title to the name Brahmin, evidently reserving legitimate
application of the term to Brahmins of plains origin. In any event,

20 THE SETTING

Pahari Brahmins are indistinguishable from Rajputs in most respects.
Even caste distinctions between the two are not as rigid as in most of
India. It is likely that the division into Rajput and Brahmin castes
occurred after the Khasas moved into India and not before the earliest
references to them. Brahmin disavowal of the Khasa appellation may
well relate to its association with defiled, or at least very unorthodox,
Hindus.

Some Rajput Paharis deny that they are Khasas on grounds that
they are of plains origin. They say the term refers to degraded people
who are not really Rajputs, and they claim higher status for them-
selves. No such people reside in or around Sirkanda, and it is impos-
sible to assess the accuracy of their claims to plains origin. Probably
some are of more recent plains ancestry than avowed Khasas, but
others may well have adopted this claim as a means to status enhance-
ment.

Doms

The second large population group of the Sub-Himalayan region
is that of the alleged predecessors of the Khasa, commonly called Dom.
Little or nothing is known of the history of these people. They are
often described as having occupied the region at the time the Aryan
invasions occurred, and having then been pushed back, subjugated,
and assigned to their rigidly inferior social status by the conquering
Aryans. For this reason they are sometimes included in accounts of
aborigines. Majumdar (1944, p. 110) has suggested that the Aryans may
have brought the Doms with them when they entered India. Whether
they are related in any way other than by name to Doms of the plains
is unknown.

Doms are the low-caste groups of the Himalayan area.

In the hills and in the Dun they comprise all classes who do menial and
more or less degrading duties such as are performed by separate occupational
castes in the plains. They are a depressed race, seldom cultivate and practi-
cally never own land. (Walton, 1911a, p. 97)

"They have for ages been the slaves of the Khasiyas and been thought
less of than the cattle . . ." (Atkinson, 1884a, p. 370). They are the
"serfs of the Khasiya race in Kumaon, Garhwal and along the hills to
the westward as far as the Indus valley" (Atkinson, 1886, p. 443). They
constitute the artisan class and in fact are also collectively referred to
as Shilpkdr (artisan). They are untouchable (achut) whether they be
blacksmith, carpenter, musician, shoemaker, weaver, tailor, basket-
maker, or whatever. Fifteen to thirty castes of Doms are found in

THE SETTING 21

various parts of the Kumaon hills (Atkinson, 1886, pp. 444 f.; Bahadur,
1916, pp. 101 f.; Raturi, 1928, pp. 196 ff.). Some of these may be wholly
or partly derived from equivalent castes of the plains who have emi-
grated to the hills, while others are old indigenous groups or groups
derived from internal differentiation of the Doms.

Doms have had no distinct language in history or tradition, and their
religious and social beliefs and practices appear to be continuous with
those of the higher castes. Whatever their cultural heritage may have
been, it is now merged with that of the Khasiyas so that traces of their
separate origin, if any, can no longer be identified. Many writers at-
tribute the unorthodox religious and social practices of high-caste
Paharis to their contact with Doms, but there is no evidence to support
this explanation. At least one writer attributes the origin of the wor-
ship of Shiva, one of the great gods of modern Hinduism, to the
aboriginal low-caste Paharis (Bahadur, 1916, pp. 73, 122).

Aside from occupational and status differences, the most widely re-
marked distinctive feature of Doms is their physical appearance. The
stereotypes regarding Doms are essentially those regarding low castes
and "aborigines" or "Dravidians" elsewhere in India. Majumdar (1944,
p. xvi) refers to them as "... a dark-skinned, short-statured, flat-nosed
people who 'scourge the eastern districts of the province.' " Atkinson
(1884a, p. 370) notes their "exceedingly dark complexion." In the
Central and Western Pahari regions with which the writer is familiar,
these are physical stereotypes based on a grain of truth, but little more.
It is impossible for Paharis or others to distinguish Doms from high-
caste people accurately and consistently on the basis of physiognomy
alone. Majumdar is the only physical anthropologist to have worked
among Paharis. The results of his measurements show minor physical
differences between Khasas and Doms (Majumdar, 1944, pp. 181 ff.).
Genetic admixture between the groups has largely eliminated the
physical differences that may once have existed.

Census materials indicate that high-caste people outnumber Doms
ten to one in Tehri Garhwal District. This is almost exactly the ratio
in Jaunsar-Bawar to the west and in thirty villages related to Sirkanda
by marriage ties, although particular villages sometimes show wide
deviation from this average.

Language

The people of Sirkanda speak the language termed "Central
Pahari" by Grierson, a language which includes the local dialects of
the lower Himalayas between the Nepalese border and the Punjab,

22 THE SETTING

except for Jaunsar-Bawar, where Western Pahari is spoken. Very likely
field studies would demonstrate extension of this language well into
western Nepal, without sharp demarcation between it and Eastern
Pahari. Grierson recognizes two component languages, Kumauni and
Garhwali, and he refers to several subdialects of Garhwali including
"Gangapariya" (language of the country beyond the Ganges) or Tehri
Garhwali, which, with some variations, is evidently that spoken in
Sirkanda and surrounding villages. Estimating the total number of
speakers of these languages is difficult because of the nature of records
available and because of the lack of accurate definition of the linguistic
designations. Grierson gave the Pahari-speaking population of India,
excluding the large but undetermined number of speakers of Eastern
Pahari in Nepal, as 2,067,514 in 1891. Of these over half, 1,107,612,
were recorded as speaking Central Pahari, and over half of the Central
Pahari speakers, 670,824, were recorded as speaking Garhwali. The
1931 census is the most recent one for which reasonably complete fig-
ures on Pahari speakers are available. At that time slightly over
4,200,000 Pahari speakers were reported in India, distributed in sig-
nificant numbers from Muzaffarabad District in Kashmir (164,000) on
the west to the Nepal border on the east, with some Nepali or Eastern
Pahari speakers in Darjeeling. A total of 1,725,000, all in the Kumaon
hills of Uttar Pradesh, spoke Central Pahari. Whether or not this fig-
ure included some 300,000 in Tehri Garhwal is unclear.

The 1951 census of India shows just over 4,500,000 residents of the
"Western Himalayan Sub-Region" exclusive of Jammu and Kashmir.
This area plus Jammu and Kashmir, which contains well over 600,000
Pahari speakers, corresponds approximately to the Pahari-speaking area
of India, though it would include a few Tibeto-Burmese speakers and
the non-Pahari residents of several hill stations. We may estimate,
therefore, that there are now around 5,000,000 Pahari speakers in India,
over one-third of whom would be classed with Sirkanda residents as
Central Pahari speakers by Grierson and about 1,000,000 of whom he
would classify as Garhwali speakers (cf. volumes of Census of India
listed in the Selective Bibliography). There is no information upon
which to base an estimate of Pahari speakers in Nepal.

Grierson estimated speakers of Tehri Garhwali at 240,281. In the
1951 census, 408,000 speakers of Garhwali, presumably Tehri Garhwali,
were reported out of a total population of 412,000 in Tehri Garhwal.
However, the administrative boundary probably does not correspond
to a significant linguistic boundary on its eastern border.

Since culture and language are to a large extent correlated in these

THE SETTING 23

hill regions, an idea of the population of the culture area represented
by the subjects of this study can be derived from the above figures.
From this can be inferred something of the representativeness of what
will follow.

Our discussion turns now from the culture area of which this study
is broadly representative to the specific location of the research.

Sirkanda Village

Northeast of Dehra Dun lies a hill area roughly triangular in
shape and seven miles on each side, known as BhatbaiR (sheep den)
in recognition of its use by Bhotiyas of the higher Himalayas as a
winter pasture for their sheep (see map 2). This region comprises pri-
marily a single spur of mountains projecting southwest from the first
ridge of the Himalayas. It lies within Dehra Dun District but is
bordered on the north and east by Tehri Garhwal District, the crest
of the Mussoorie Hills forming the northern boundary and the Bandal
River forming the eastern. To the south and west it is bordered by the
Dehra Dun valley and the Baldi River. Bhatbair has a total population
of around 1,700 people living in about 10 villages and at least 15
smaller settlements. These have been combined, for administrative
convenience, into seven revenue villages including Sirkanda. This is
now the official definition of "Bhatbair."

In local usage the term "Bhatbair" has not been very precise. It is
sometimes used by Sirkanda villagers to refer to the immediate Pahari
area familiar to them and almost entirely within a four-mile radius of
Sirkanda, including the side of the ridge facing them to the east, in
Tehri Garhwal. This larger region forms a relatively independent
economic and social unit of about 60 villages and settlements with a
total population of almost 5,000. In another publication (Berreman,
1960b) I have referred to this larger area as "Bhatbair" for convenience.
In the present account I will use the term in its more restricted sense,
closely approximating the official definition. The larger area can
simply be designated Bhatbair and vicinity.

Probably the earliest historical reference to Bhatbair is that by
Williams (1874, pp. 92 ff.), who refers to a "Rajpoot Princess Ranee
Kurnavutee" who had a palace near the present site of Dehra Dun
before the arrival of the town's founder Guru Ram Rae (that is, before
1700). "Under her fostering care the valley smiled, and many flourishing
villages sprang up such as . . . [among others] Bhat Beer. . . ." Al-
though he has confused the name of a local region and administrative

1 to 4 marriage partners
5 to 8
9 or more
more Sirkanda Chans

Sirkanda village and lands

valley

Bhatbair

Notes:
Circle around Sirkanda indicates area in
which 80% of Sirkanda marriages have
been contracted

General areas and more distant locations
where marriages have been contracted;
precise village unidentified or unidentifiable:

Jaunpur 7 Sirmor(dist. to west). .1

Saklana 6 Punjab(state to west).. .2

Tehri Garhwal 12 Delhi 1

Map 2. Sirkanda marriage network and chans.

-iwro gV

£ 5? i>

£ I CD

o .* E

~-\

'Z.

o
o

□

■-

(/>

"> "3

a> £

c/>

Q CO

-O

•o

*v-

a> .c

—

— 1

c/5

a.
o

-*:
o
o

o
o

-C

■o

en
c

t: ^,

'a.

c .e .e

< (0 O ffl < t- <o _ii-

<05 12[]OI|OO

26 THE SETTING

unit containing several villages with that of a single village, there can
be little doubt that the reference is to the area of which we are speak-
ing.

For some time before the Nepalese conquest in 1804 and the British
conquest in 1815, Sirkanda was included with the rest of Bhatbair,
the Dehra Dun valley, Tehri Garhwal, and Garhwal as part of the
kingdom of the Garhwal Raja. Villagers retain a good deal of pride in
this raja as contrasted to other rajas, albeit they can also recount his
tyranny and injustices. There are several tales told which relate his
special interest in the tall, powerful, and honest residents of Sirkanda
and the recognition given them in his court. In the later days of the
Garhwal Rajas, Sirkanda was one of five villages located on this spur
of hills. According to villagers these five and adjacent lands were listed
in the Raja's records as being the revenue responsibility of Matthu, a
hill Brahmin of Kanda, one of the five villages. As saycma, or tax col-
lector of the villages, he was allowed to keep a considerable portion of
his collections after turning over an annual sum to the Raja. Bhatbair
was apparently

one of those curious taluqs or clusters of several villages, so prevalent in the
. . . hills, which are cultivated by a numerous community of zamindars
[landowners], all enjoying separate and independent proprietary right but
at the same time all bound together by joint responsibility for the revenue
assessed on the whole mahal. (Walton, 1911a, p. 109)

Each taluq (called khat in Jaunsar-Bawar and some other Pahari areas)
was under a sayana who collected land revenue for the Raja from the
landholders in each component village. At one time the landowners
may have been individually responsible directly to the Raja for
revenue, as was the case over most of the Kumaon Himalaya (cf. Stowell,
1907, p. iv; Baden-Powell, 1892, II, p. 313). Whether cultivators were
individually or collectively responsible for revenue, ultimate rights in
the land were vested in the Raja, who not only collected revenue but
was free to alienate any land in his dominion to his own purposes —
most often to cede it to people he wished to reward. When this oc-
curred, former owners suddenly became tenants (cf. Stowell, 1907, pp.
ix, 1-6, 31 f.).

When the British took over in 1815, Bhatbair was joined administra-
tively with the Dehra Dun valley. Due to the accident that it is a very
small hill tract in a district made up primarily of people of the plains,
special conditions in Bhatbair have been ignored in the various revenue
settlements. For revenue purposes Bhatbair villages have usually been
treated like villages of the valley. Initially the revenue settlement for

THE SETTING 27

Bhatbair was probably made with the sayana, Matthu. There was much
vacillation in the policies of early revenue officers dealing with Dehra
Dun District, some favoring development of a landed aristocracy and
acting accordingly, and others favoring owner-cultivators (cf. Williams,
1874, pp. 200 ff.). However, under the British a Rajput of Kanda is
said to have become headman of Bhatbair, which had come to include
seven villages. He wielded considerable influence with the British, and
is credited with having pressed for the rights of cultivators and having
influenced the government finally to allot proprietary rights to the
lands in Bhatbair in equal portions to the principal cultivating fami-
lies, who thereafter paid reduced taxes directly to the British through
a local tax collector. Matthu and his descendants then lost all connec-
tion with the lands they did not cultivate.

At that time Sirkanda had 16 high-caste cultivating families. Cul-
tivated village lands were divided into 16 equal parts, and equal por-
tions of each type of land and of land in each area were assigned to
each of the 16 families. As new land came under cultivation, it too was
divided in this fashion. In continuation of local tradition, uncultivated
lands, trails, and certain open places in the village settlement area
were deemed to be the common property of the village. This allot-
ment, with alterations resulting from family division, inheritance, and
purchase, has remained to the present. This system is referred to as
the "bhdichdra" (custom of brothers) system of land tenure, a British
introduction to the area which Sirkanda people compare most favor-
ably with the landlord-tenancy system retained until Independence in
nearby Tehri Garhwal under the Raja.

Another portion of Bhatbair and lands west of it had been granted
by the Raja to the Sikh religious leader who founded Dehra Dun. He
exacted rent from the people living on this land, people who had until
that time been owner-cultivators. The British honored his proprietor-
ship, and his successors continued to enjoy this revenue until well after
Independence; in fact, they still hold rights to some of the land. Sir-
kanda lay outside this estate, though some Sirkanda-owned lands did
not.

Independence was never an important issue in Sirkanda, as the
British government was looked upon without disfavor. No government
is really approved of by these people, but according to them the British
government was not meddlesome and was preferable to that of the
Raja. After Independence there were revisions in administrative pro-
cedures and in taxation policies and other laws affecting Sirkanda. The
forests and other uncultivated lands adjacent to the village, which
provide a substantial part of the means of livelihood in the hills and

28 THE SETTING

which formerly could be utilized quite freely, were nationalized. Access
to their products and the right to cultivate previously uncultivated
lands were sharply curtailed. This was the most resented and probably
least obeyed of the many unpopular post-Independence reforms in
these hills. Other innovations, such as village panchayat (council) rule,
community development work, and attempts to set up village credit
cooperatives, were largely ignored or actively opposed. Where the ap-
pearance of conformity seemed desirable it was provided, but nothing
more. The government-supported village school, established in 1950,
met with a somewhat more favorable reception. The role of the govern-
ment and reactions to it will be discussed in detail in chapter 9.

Sirkanda is populated by Paharis who identify themselves as such
and more specifically as Garhwalis. In this book the generic term
"Pahari" will be used to refer to them in preference to the more spe-
cific term "Garhwali," except where a contrast is intended. This is
done simply because that is the term most used by the people themselves
and by their neighbors on the plains and because on present evidence
it seems to be a more defensible term culturally and linguistically than
"Garhwali." There is no intention of minimizing differences among
Pahari areas, most of which have yet to be studied. Sirkanda is rep-
resentative of the following areas, in order of decreasing cultural homo-
geneity: Tehri Garhwal, Garhwal, the Central Pahari area, and the
Pahari area.

The people of Sirkanda are closely attached culturally, linguistically,
and historically to their relatives in Tehri Garhwal, though for 150
years they have been administratively separated as a result of British
rule. Theirs is one of the three largest villages within Bhatbair. It is
situated on the crest of the spur of mountains comprising Bhatbair,
5,300 feet above sea level, facing southeast, 2,300 feet above the Bandal
River which separates it from Tehri Garhwal at a distance of one mile,
and 3,100 feet above Dehra Dun, which is ten miles to the southwest.

The history of Sirkanda is vague in the minds of its residents and is
unrecorded elsewhere. Two large stone fortifications in the village
overlook the steep valley of the Bandal and face the hills of Tehri
Garhwal. They are thirty feet square and constructed of stones so large
and heavy that it is believed no modern mortals could have placed
them. According to village tradition they have been in place for many
millennia, and were built in a former age when men were nine yards
tall, lived for a thousand years, and were supremely intelligent. The
use of the structures is unknown, though they are thought to have been
either houses or forts. Bricks and charcoal have been found around
them in recent excavations for housebuilding. Such fortifications occur

THE SETTING 2 9

throughout Garhwal and are responsible for its name. Fifty-two major
forts are frequently cited in descriptions of Garhwal (Sirkanda's not
among them) including one in the Dehra Dun valley. They are thought
by historians to be the work of the various petty rajas who ruled small
independent states in these hills before the seventeenth century. One
writer attributes them to the pre- Aryan indigenes (Bahadur, 1916, p. 70).

Local tradition states that Kedarkhand or Garhwal, of which Bhat-
bair is a part, was first occupied in modern times by hermits, ascetics,
and sages who were attracted by the serenity and beauty of the area
to spend their meditative years there seeking enlightenment. The springs
which provide water for many of the villages, including that for Sir-
kanda, are thought to have been created by sages who did so miracu-
lously in order to provide water at their stopping places. Villagers say
that some of these ascetics failed to keep to the spiritual life, married,
produced children, and so populated these hill regions.

In view of Williams' mention of Bhatbair as having been in existence
before 1700, it is probable that Sirkanda has existed for at least 300
years. Villagers' estimates run to 1,000 years. Among its current resi-
dents there is general agreement that Sirkanda was founded by the
ancestors of its single Brahmin family, who came from Genogi, a vil-
lage in Tehri Garhwal, a long day's trek east of Sirkanda. The Brahmin
family soon thereafter brought a family of barber caste from the valley
below to provide them with agricultural labor and barbering service.
Since barbers are not a caste indigenous to this section of the hills,
their functions are performed by other low castes in neighboring vil-
lages. These were, then, the first two families in Sirkanda. Thereafter
the Khasiya Rajputs, who today form 87 per cent of the village popula-
tion, came in. Each of the two largest Rajput sibs in Sirkanda claims
to have preceded the other.2 Both are said to have come from villages
to the east, in Tehri Garhwal. The drum-playing Bajgi family's an-
cestors also came from Tehri, shortly after the Rajputs. The black-
smith family was the most recent arrival, having come from a village
35 trail miles to the northeast, in Tehri Garhwal, about 90 years ago.

It is thought by all villagers that their ancestors came ultimately from
"Kumaon," the present Almora and Naini Tal districts adjacent to
Nepal. From there they moved to Tehri Garhwal, probably to escape
political and economic difficulties, and then on to Sirkanda to find
new land or to escape the pressures of local rajas. Their origin prior to
Kumaon is unknown, though some believe their ancestors must have
come there from the plains — perhaps to escape the Moghul rulers —
while others claim they have always lived in the hills. High-caste
villagers bear the names of sibs and phratries (gotras) found on the

go THE SETTING

plains. This may reflect plains origin or merely adoption of names of
Rajputs who fled the plains to live among the hill-residing Khasiyas.
Both explanations may be correct in view of the nature of plains-hills
contacts. Two and perhaps three of the four Sirkanda Rajput sib names
are among those listed by Tod in his Annals and Antiquities of
Rajast'han (1829, p. 120) as being among the "eighty-four mercantile
tribes" of Rajasthan: "Pilliwal," "Khandailwal," "Kakulea," corre-
sponding to Palidl, Khandial, and KukhalwalQ) in Sirkanda. Two and
possibly three (JawaRl, Palialp], Khandial) appear among the 116
Rajput groups listed by Raturi (1928, pp. 167 ff.) for Garhwal. Pos-
sibly one (Palial) appears among 102 listed by Bahadur (1916, pp. 96 ff.).
The Brahmin family of Sirkanda does not belong to any of the 68
Brahmin subgroups listed by Bahadur. The barber is of plains origin.
Bajgis and blacksmiths both are populous groups in Garhwal and ap-
pear among the 15 to 30 Pahari Dom groups listed by Bahadur and
Raturi, respectively.

In physical appearance Sirkanda villagers differ little from plains
people. There is perhaps a tendency toward lighter complexion and
narrower features among the high-caste hill peoples, but individual
variation virtually obscures this. Much more distinctive are cultural
traits, and these are often mistaken by other Indians for racial traits.
Paharis are noted by people of the Dehra Dun valley and the plains
for their peculiarities of speech and dress and for general rusticity,
much as are hillbillies in America. They are also known for honesty
and bravery:

Honesty and valour are possessed in ample measure by the Khasas. Their
honesty is beyond question. A verbal bargain is seldom repudiated and theft
is almost unknown. . . . The military exploits of the Khasas in modern
times are enshrined in the records of the 39th Royal Garhwal Rifles, and
we find that the descendants of the ancient Khasas [reported in the Mahab-
harata] are endued with great courage, unyielding and obstinate in battle.
(Joshi, 1929, pp. 23 f.)3

As a group, Paharis are readily identifiable outside their own area
by their dress, speech, and manners. The badge of the Garhwali man
is a black cap and cane. Now the "fit pajama" (tight from the knee
down), shirt, dark vest, and black coat are in vogue for special occa-
sions, with a wool blanket as wrap. Dhoti (loincloth), loose pajama,
shorts or abbreviated loincloth, shirt, and sometimes coat or vest are
daily wear. Formerly a cap, a small loincloth secured by a string, and,
for cold weather, a blanket held in place with wooden skewers sufficed.
Women are noted for their massive silver jewelry, and the large gold

THE SETTING 31

nose ornaments of married women. They do not wear the red mark at
the part of the hair which designates the married woman of the ad-
jacent plains, nor do they wear the red beauty spot on their forehead.
Their hair is usually braided in one piece down the back, extended by
a black or colored artificial hairpiece. The calves of their legs are gen-
erally decorated with tattoos. Men and women alike often have tattoos
on hands and arms. Women's dress consists of a colorful print skirt, a
long black fitted jacket (or sometimes a shirt and vest), and invariably
a head scarf. In earlier times they wore a long, loose blanket-like gar-
ment, the fitted jacket, and the head scarf. They are not secluded in
purdah as are many high-caste plains women. To one familiar with
the area it is possible to determine quite precisely the locality from
which a Pahari woman comes by observing the details of her dress and
jewelry (Berreman, 1960b). Regarding Pahari dress, Traill's comment is
as applicable today as it was in 1828:

It may be observed, generally, of the hill people, that they are extremely
indifferent in regard to the state of their every-day apparel, and continue
to wear their clothes till reduced to mere shreds and tatters, but on holy-
days and festivals, individuals of either sex prefer absenting themselves from
the festivities, to appearing in a worn out garment. (Traill, 1828, p. 212)

Sirkanda is strategically located in the sense that it is at the junction
of two important trails. Several villages of Bhatbair and the adjacent
section of Tehri Garhwal are accessible to Dehra Dun only via one of
these trails, and the other is a good and frequently used pack trail
built in 1914 by the Tehri Raja to enable his subjects to bypass the
toll-tax collectors of Mussoorie on their way to and from the markets
of the valley. This trail is still a trade route of some importance. It
connects valley markets with the heavily used trade and pilgrimage
route between Mussoorie and Tehri, nine trail miles from Sirkanda.

Until recent years these trails were used by Paharis exclusively for
foot traffic. Governmental employees and some traders used pack
animals on them. Today some of the more prosperous high-caste fam-
ilies of Sirkanda and other villages utilize horses or mules for trans-
porting goods. However, goods of all types, from grain to roofing ma-
terials, are still carried primarily by people — men carrying the loads
on their backs and shoulders, women balancing the burdens on their
heads.

Since 1930 a motor road has connected Mussoorie to the valley, but
this does not directly affect Sirkanda villagers because it lies in a dif-
ferent direction. Before 1930, Rajpur, which is a small town seven
miles north of Dehra Dun at the foot of the mountains (seven miles

32 THE SETTING

south of Mussoorie and nine trail miles west of Sirkanda), was the end
of the road. Today there is frequent and inexpensive bus service from
Rajpur to Dehra Dun, a facility used by Sirkanda villagers which cuts
the walk to Dehra Dun from sixteen to nine miles. Recently a motor-
able road has been built into Sahas Dhara, a small market and place of
pilgrimage five miles from Sirkanda on the trail to Dehra Dun. So far
its use has been limited largely to the trucks of a limestone quarrying
company for whom it was built, and there is no bus service on it. Oc-
casionally villagers can hire empty trucks belonging to the quarrying
company to carry heavy loads for them from Dehra Dun to this point.

The village has features in common with others of the Sub-Himalayan
area, and is reasonably typical of those in Tehri Garhwal. Like most
Pahari villages and unlike many villages bordering the hills in the
Dehra Dun valley, Sirkanda is nucleated, with its terraced lands oc-
cupying the steep surrounding hillsides. Some two-hundred acres of
terraced, cultivated lands lie below and beside the village, while 1,200
acres of uncultivated forest and scrub land administratively attached
to Sirkanda surround the cultivated and occupied area. The village
houses are scattered for almost half a mile along the southeast contour
of a hillside overlooking the mountains and valleys of Tehri Garhwal
and part of the Dehra Dun valley. There are nearly sixty of the char-
acteristic stone houses with gabled slate, thatch, or corrugated iron
roofs supported by heavy wooden beams. Most of them are two-storied
with an outside stone stairway and a narrow, shelflike porch extending
the length of the house at the second floor level. Open, arched verandas
with ornamentally carved columns occupy the center of the upper level,
with closed rooms on either end and often in back as well. The rooms
are entered by doorways off the veranda and have small, barred win-
dows for ventilation and light. Verandas, doors and windows open on
the front, which is almost invariably the downhill side. The backs, and
often the ends, of houses are entirely closed. The family occupies the
second floor, its animals the first.

Part or all of the ceiling is boarded over to provide a storage garret,
reached by a ladder. Here are kept boards and implements. Every family
has the necessary agricultural implements or tools of its trade. Two
families have phonographs, one in operating condition. The Bajgis
have their drums and harmonium plus a sewing machine and iron
used in tailoring. In general, high-caste houses are larger and better
equipped than others. Wealth is to some extent reflected in the house
and its contents, but the difference between houses of the well-off and
of the poor is not as great as in most non-Pahari areas. This is partly

THE SETTING 33

because differences in wealth are less, but is also due to consumption
patterns which inhibit conspicuous display of wealth.

Household furnishings are few and simple. One or two of the inner
rooms contain hearths for cooking. Brass and iron cooking utensils, a
wooden churn, wooden vessels for storage of liquids, and large baskets
and wooden cupboards for storage of dry goods are also found in these
rooms. Pottery vessels are used hardly at all because none are made in
this area, transportation is difficult, and they break easily. Light is
given by small oil lamps. Many houses have string cots, the typical
Indian bed, but this is a relatively recent innovation and there are
never enough to go around. Bedding consists of quilts, blankets, and
rugs which are used on the floor if beds are not available. People usually
sit or squat on the dung-plastered floor, but small wooden slabs and
pieces of leather or animal pelts may also be used for this purpose and
are always offered guests. Every house has its water smoking pipe for
the use of all its members and guests.

Circular threshing platforms and adjacent storehouses surround the
village, which is composed of three main settlement areas, the southern-
most being a recent addition and the site of the school. Two small
shops and a blacksmith's workshop are out of sight of the village, over
a ridge and at the junction of the main trails leading into Sirkanda.
The trail leading to the village water supply goes past them, and nearby
is a rain pond which is dry half the year. Smaller trails emanate in all
directions from the village to the fields and to neighboring houses
and villages. The trails are for the most part rough, steep, and often
dangerous. Every year mules, horses, and cattle plunge off, and oc-
casionally men suffer this fate. Travel is difficult and casual visitors are
a rarity.

Although fields surround the village, there are no irrigated lands
near Sirkanda. The village water supply is a small but reliable spring
one-fourth to one-third of a mile from the village on the opposite and
uncultivable side of the ridge. The spring's proverbial purity and re-
liability apparently compensate for its inconvenience, though an el-
derly wife of Sirkanda was once heard to mutter, as she lifted her heavy
water pot to her head and prepared to negotiate again the difficult trail
to the village: "I'd like to set fire to the beard of the man that founded
this village here." Villagers want to pipe the water into the village, but
the ridge forms a barrier. As a result of one abortive attempt to improve
the situation, a small pipe brings water somewhat closer to the village
but this water is considered inferior and is used mainly for animals.

Sirkanda is primarily a crop-raising village with an important sec-

34 THE SETTING

ondary investment in animal husbandry. Normally it produces more
grain than its residents eat in a year, and famine is virtually unknown.
Since the establishment of the hill station and military cantonment of
Mussoorie in 1829-1835, a six-hour trek from Sirkanda (and only half
that from some Sirkanda cattle sheds), and with the growth of Dehra
Dun and adjacent areas, milk-selling has been a profitable enterprise
with which Sirkanda villagers supplement the traditional agriculture.
The introduction of potatoes as a cash crop in the hills above Sirkanda
has afforded further income and additional traffic past the village.

The poverty of Paharis is proverbial, but this reputation is based on
their frugality, the simple clothing and equipment they possess by
plains standards, and their inelegant and unvaried diet wherein
coarse millets substitute for rice. They do not share the precarious life
of many plains people in food-deficit areas. They have sufficient lands,
regular rainfall, and a tradition of maintaining the productivity of
their fields by crop rotation and fertilizers, so that they have a con-
sistently adequate food supply.

Not all Sirkanda lands are near the village, and not all people known
as Sirkanda villagers live in Sirkanda. As population has increased,
new lands have been opened up and cultivated on the hills north and
south of the village. Today villagers cultivate land up to eight trail
miles north and seven trail miles south of Sirkanda, on the same spur
of hills, at altitudes ranging from 7,000 feet to 2,000 feet above sea
level. In order to tend fields and livestock at these distances, it is
necessary to build dwellings near them. Such field houses or cattle
sheds are called chdns. A chan is defined in terms of location and
construction. It is never in a village, though it may be very near one.
It is generally only one story high, unpartitioned, and is less finished
inside than a house (ghar). Usually livestock are quartered in the chan
among its human inhabitants, whereas in a house livestock are always
kept downstairs or in an adjoining, but partitioned, room. Eleven of
the 71 Sirkanda-owned dwellings outside of Sirkanda qualify as
"houses" rather than "chans" on the basis of location or construction.
Chans are occupied by their owners during planting and harvest times,
and some members of the family live year-round in many of the chans
or migrate seasonally between low-altitude winter chans and high-
altitude summer chans. Formerly chans were only temporary dwellings,
but, as they have been improved and as families have grown, many
have become permanent. The 71 Sirkanda chans and second houses
are owned by a total of 45 joint families and are to be found in 31 dis-
tinct locations (see map 2). In most cases the village affiliation of chan
dwellers is strong, and they are readily identified as Sirkanda villagers

THE SETTING

by themselves, by Sirkanda people, and by residents of other villages
and chans. This in spite of the facts that they no longer reside in the
village, that other villages may intervene between them and Sirkanda,
and that they may be surrounded by the chans of other villages. While
some Sirkanda chans are at considerable distances from Sirkanda, chans
of other villages may be found adjacent to Sirkanda. On the other hand,
some Sirkanda chans are so close to the village that a nonvillager might
not distinguish them from village houses. This makes village census
reporting and land records a complex and difficult matter. The prob-
lem is solved in official reports by ignoring it. For administrative
purposes village membership is assigned by geographical proximity,
based on administrative boundaries.

Total 384
Males 196 M Females 188

8765432 1012345678
Total population, per cent

Usually resident in Sirkanda | [Usually resident in chan
Fig. 1. Sirkanda population pyramid, 1958.

Of a total Sirkanda population of 384, only 178 people are usually
resident in Sirkanda, 43 stay there at regular intervals to work land,
and 163 rarely or never reside there (though 38 of these occupy chans
close enough to afford almost daily contact). On festive or ceremonial
occasions many people who normally live away return temporarily to
the village. Women move from chans to the village to have their babies,
and elderly people usually return to live their final years in the village.

Chan sites may develop into village sites, and in fact this is a means
by which new villages are often formed. Residence in a cluster of chans

g6 THE SETTING

can gradually lead to new village affiliation. The former chan status
of many present-day villages is revealed in the village names. In some
of these locations part of the residents consider themselves residents of
that village, while others maintain their ties to some antecedent village
affiliation. About one-fifth of the living persons (excluding married
daughters) recorded in Sirkanda genealogical materials as having been
born of Sirkanda families are no longer considered to be affiliated with
Sirkanda, though their origin is remembered. Nearly all these people
now live in locations once considered to be chans. There are undoubt-
edly others who have been forgotten or are no longer even mentioned
in this context. People who belong to joint families with a dwelling in
the settlement area of Sirkanda are considered by all to be Sirkanda
villagers regardless of where they live. Anyone who separates from his
joint family, taking a chan and the lands near it as his share of the
property and giving up all rights in the village, may cease to be con-
sidered a village member; certainly his children will not be considered
members of the ancestral village.

Other important extra-village ties are those of marriage. Over 80
per cent of Sirkanda marriages take place outside the village; that is,
the bride comes from or goes to an alien village. This makes for close
ties and contacts with almost 100 villages; most of the ties are with
villages within 8 trail miles of Sirkanda, and virtually all are with
Garhwali villages (see map 2).

Immigration or emigration of men is rare. One Bajgi has come to
live in Sirkanda in the past 25 years. Other immigrants are the five
agricultural servants who are considered, and who consider themselves,
to be temporary residents. In addition there is a shopkeeper from
Rajpur who spends considerable time in his shop near the village. In
recent years a schoolteacher and periodically a Village Level Worker
and Economic Cooperative Supervisor, all government employees, have
lived in the village school. In the past ten years one man has emigrated
to an urban area, and another joined him during 1957. Most villagers
who have left the village have returned. A few have disappeared.

Garhwali villages are so constituted that no one village is likely to
have the range of castes necessary to provide the occupational variety
required for maintenance of community life. This brings about fur-
ther intervillage ties of economic and ritual interdependence, to say
nothing of the social ties of an intra-caste, intervillage nature. All
such factors make an area such as Bhatbair (combined with an equally
large adjoining area in Tehri Garhwal) a more nearly self-contained
economic, social, and religious unit than the village.

Sirkanda is predominantly a Rajput village; 87 per cent of its mem-

THE SETTING 37

bers are of this caste. The one Brahmin family adds another 3 per cent,
making 90 per cent of the village of high caste. The 10 per cent who are
of the Dom castes include two households of blacksmiths, one of bar-
bers, and four of musician-tailors (Bajgis). The proportion of high
castes to low castes is almost identical to that found throughout Bhat-
bair and Tehri Garhwal. However, particular villages may be pre-
dominantly or even exclusively Brahmin. In some, relatively large
proportions of low-caste people are found, while in many, few or no
low-caste people live.

In population Sirkanda is above the average for Tehri Garhwal
villages. In caste composition it is not unusual. It is as representative
of Tehri Garhwal villages as one village could be expected to be ex-
cept that it is somewhat less isolated than most, and in the past 150
years its administrative history has been at variance with the Tehri
Garhwal norm and similar to that of British Garhwal and other dis-
tricts under direct British rule. The effects of its relative lack of isola-
tion and its subjection to British rule have been less than might be
expected because of its distance from Dehra Dun, the difficulty of
access to that administrative center, Sirkanda's insignificance in the
view of administrators, and the fact that its members' attention and
identification have been directed throughout its history primarily to-
ward the northeast — into Tehri Garhwal.

THE ECONOMIC CONTEXT

The subsistence base in Sirkanda, as in most Indian villages, is agri-
culture. Of second importance and closely related to agriculture is
animal husbandry. Both of these activities are primarily the province
of high-caste villagers but are open to and practiced by some members
of all castes. The functional unit in these as in almost all economic
matters is the joint family, often referred to as the household (chula,
cooking hearth). Goods are produced, distributed, and consumed pri-
marily by the household unit, and any relevant economic decisions are
made there. In Sirkanda there are 45 such household units or joint
families, averaging eight or nine individuals each but ranging in mem-
bership from one to twenty-five. The household is ideally a patrilocal
extended family under the leadership of the eldest active male. In prac-
tice we find that 32 (71 per cent) of the households in Sirkanda are of
this type, including seven which consist of only a nuclear family and
one parent of the husband. Thirteen (29 per cent) consist of strictly
nuclear family units. Eighty-four per cent of the population lives in
extended family groups. Nonagricultural castes and families tend to be
divided into more and smaller independent units than are agricul-
turists.

Improved land and livestock, the main forms of capital in this area,
are owned and manipulated by the joint family, and their products are
shared within it. Ideally the joint family occupies one large house with
separate rooms for each nuclear family when they are in the village,
and shares one or more chans when they are out. In reality some fam-
ilies own more than one house in Sirkanda, so that all their members

THE ECONOMIC CONTEXT 39

do not live under one roof, while others are crowded with more than
one nuclear family per room. Nearly all have chans, many of which
are almost continuously occupied by some nuclear families of the
household. Members of the joint family who live under a single roof
eat from the same cooking fire (the literal meaning of chula), and all
members, regardless of living arrangements, eat from the same store
of food.

Land

Since agriculture is the basis of livelihood in Sirkanda, land is of
utmost importance. At least one popular Pahari song concerns a hero
who fought to the death rather than lose land which was rightfully
his. Sirkanda families own 298 acres of cultivated land, according to
records of the village accountant. These include 176 acres in the im-
mediate vicinity of the village and 122 acres in outlying areas, tended
from chans or other houses. As previously noted, about half the cur-
rent village population usually lives in the 71 Sirkanda chans and other
houses in 31 separate locations extending over fifteen trail miles north
and south of the village settlement area (see map 2). Some idea of the
fragmentation of village lands can be gathered from the following
figures. Two hundred and five acres are recorded as cultivated "Sir-
kanda lands." x The land is officially divided into 745 separate plots
varying in size from less than one-tenth of an acre to seven acres.
These figures are based on long-standing recorded divisions of land.
In reality the plots have been subdivided into smaller parts as the old
extended families have broken up. Today there are 1,144 separate land
claims recorded for the 745 plots, and unrecorded but recognized di-
visions probably bring the number of separately cultivated plots to
nearly 1,500.

Cultivated land is categorized by villagers in several ways relevant
to its productivity. Probably the most important distinction is between
land which is irrigated and that which is not. Most Sirkanda land is
unirrigated. Only three acres out of 205 are recorded as being irrigated
and hence unusually valuable. Sirkanda-owned land in lower-lying
areas is often irrigated. While records for these lands were unobtain-
able, crop yields for rice indicate that 16 Sirkanda households own 25
to 30 acres of such valuable land, while 29 households own none. Some
Bhatbair villages are located in the river valley and have considerably
more irrigated land. Another valuable type of land is that at higher
altitudes and on north slopes suitable for growing potatoes. Twenty-
one Sirkanda families have such lands, totaling around 15 acres.

40 THE ECONOMIC CONTEXT

Although irrigated land and potato land are the most valuable,
neither is necessary for livelihood. The traditional economy of Sir-
kanda is based upon dry-land grain agriculture, and dietary staples
are grown on dry lands in amounts exceeding consumption rates.

Villagers recognize and terminologically distinguish several types of
land other than those based on irrigation. The two most important of
these classifications are by soil type and exposure to the sun. The best
soil is classed as "loamy," while inferior soil is "rocky." Fields on the
sunny (generally south) side of slopes are preferred. Fields on the
shady (north) slope, called "place of moisture," are considered less
productive of the staple crops. Consequently they are chosen for cul-
tivation when no more sunny-side slopes are available, although some
villages are so situated that all their fields are of this type. For certain
crops, such as potatoes, and in exceptionally dry seasons shady fields
are superior. All Sirkanda lands near the village are on the sunny slope,
and more than half are on loamy soil. Fields are also classified in terms
of terrain; level ones are preferred to those on slopes. All Sirkanda
lands near the village are on slopes and are, of necessity, terraced.
Vertical stone walls three to eight feet high follow the contour of the
hills to retain the soil of these terraces. Entire hillsides of scores of
acres are so terraced. Fields located below the village, as are most
Sirkanda fields, are considered preferable to those above the village
because of the greater ease of carrying fertilizer to them.

As a result of constant fertilization of fields with animal manure and
the burned and raw stalks of harvested crops, combined with system-
atic crop rotation, most fields produce consistently good crops. Certain
distant fields which are too inaccessible for easy fertilization are simply
used until they become exhausted, after which they are left, perma-
nently or until they have regained their vitality.

Land ownership by households shows the expectable caste-based
variation. The ten per cent of the population who make up the low
castes own only two per cent of the land. In the days of the Garhwal
Raja's dominion, low castes were not allowed to own land or houses at
all. This was the situation in nearby Tehri Garhwal until after Inde-
pendence in 1947. In Sirkanda it ceased to be the rule when the British
took over in 1815, but it has continued to influence practice in this
area to the present. It was possible under the Raja for high-caste land-
owners in the village to grant land or buildings to low-caste members
for use rent-free, and land was regularly granted as a reward for services
or to enable a service-caste family to make a living in the village and
provide their services to villagers. However, the tenant was not allowed
to sell or rent the land, and at his death or at the owner's discretion

THE ECONOMIC CONTEXT 41

the property reverted to the owner. During British rule each of the
low-caste families of Sirkanda managed to get some such lands re-
corded in their names.

Brief reference has already been made to the history of land tenures
in Bhatbair. The pattern of land ownership which exists today in
Sirkanda dates from the early days of British rule, when the bhaichara
system of land allotment in equal portions to cultivating families was
established. At the time there were sixteen cultivating households in
Sirkanda, one family of Brahmins and the others of Rajputs. The
sixteen divisions of cultivated land averaged about eleven acres each.
As new lands came under cultivation they were similarly apportioned.
Since that time, as lands have been divided, bought, and sold, they
have retained the designation of the caste or sib of the original owner
as patti Brahmin (Brahmin section), patti Palial (Palial Rajput sec-
tion), and so on, for identification purposes. Uncultivated lands were
held by the village in common.

Under the bhaichara system low-caste members did not get shares
in village lands, but they were assigned the small plots that had been
previously granted them by the high castes or that were subsequently
given to them. In this fashion the blacksmiths, for example, acquired
slightly over two acres. Half this land had been granted by the village
tax-free; that is, it remained in the names of former owners, who con-
tinued to pay its taxes. The other half had been given to the black-
smiths outright. The lands given outright were not assigned by name
but instead were recorded as having been allotted to "blacksmiths."
Later a friendly village accountant recorded all this land in the indi-
vidual names of the blacksmith owners, thus protecting their interests.
The other low castes were granted land in similar fashion, so that the
barber family had nearly 1% acres and the Bajgi family had nearly 3
acres. Virtually all low-caste lands are on steep, rocky, unproductive
soil, some of such poor quality that it is hardly worth cultivating and
would be spurned by high-caste agriculturists. This reflects the nature
of high-caste generosity to the service castes.

Under this system there are said to have been only two classes of
landowners in Sirkanda, in contrast to the many types recognized else-
where: (1) the original sixteen owners, who paid the government tax
on the land they acquired under the bhaichara division, and cultivated
it or let it out as they wished; (2) those who cultivated lands not al-
lotted to them under the bhaichara system and to which they therefore
did not have title. In Sirkanda most agriculturists were in the former
category, although some families who had more land than they could
use let part of it out to those who had too little from their bhaichara

42 THE ECONOMIC CONTEXT

shares to support their enlarged families. Under bhaichara these non-
owning cultivators (tenants or kachcha khaikars, cf. Stowell, 1907, p. ix)
paid the amount of the taxes to the owners but in most cases paid no
rental beyond that amount. This is in striking contrast to lands not
included under bhaichara, such as some in Bhatbair that were owned
by the Sikh founder of Dehra Dun and his successors. In such cases
the landlord collected an amount several times the government tax and
kept all except the tax money for himself.

With Independence bhaichara lands went to the cultivators. Non-
cultivating owners were reimbursed by the government in the amount
of ten years' tax on the land they were to relinquish. Nonowner cul-
tivating tenants were assessed an amount equal to ten years' tax on the
land they occupied and then were recorded as its owners. Each owner
thereafter paid the taxes on his own land, direct to the government.
Taxation on owner-cultivated lands in Sirkanda has become slightly
higher since Independence (from around Rs. 1.50 per acre before to
Rs. 1.75 after).2 For non-bhaichara land, the cultivators' taxes have
greatly decreased since Independence because the landlord middleman
has been eliminated. This decrease in the affected portions of Bhatbair
and nearby Tehri Garhwal has been from about Rs. 4.50 to Rs. 1.75 per
acre. The precise amount of tax is a function of the quality of the land
as recorded by the government. In an official threefold categorization
of Sirkanda-owned lands, none around Sirkanda is of first quality, and
most is third and therefore least heavily taxed.

Under bhaichara, when a joint family dissolved, land was to be
divided equally among sons and each son's portion was to be divided
equally among his sons if they separated. These "ancestral fractional
shares" were to be officially recorded at the time of division. Each
holder would then pay his share of the taxes, depending upon the
proportion of village lands he held. In practice the division was gen-
erally done informally and went unrecorded; hence the chaotic state
of land records at present. Today three families — two of the Rajputs
and the one Brahmin family — retain their original yi6 portion of vil-
lage lands undivided. This retention has been possible because the
families have remained relatively small. Consequently they are among
the best-off of villagers. In 1958 the Brahmin family was preparing
to divide its lands among three brothers, so that only two of the orig-
inal shares would remain intact. Some bhaichara portions have been
divided into four or more parts, though often these have been supple-
mented (in at least one case to an extent exceeding the original share)
by lands subsequently purchased outside the village.

Land is the basis of livelihood, and those who control it, primarily

THE ECONOMIC CONTEXT 43

the high castes, control the economy. Ownership of land, and especially
of good land, is highly correlated with wealth. It is not, however,
sufficient in itself to ensure wealth. There must also be manpower to
work the land. Land that is unused is of no benefit to anyone. One of
the wealthiest families in Sirkanda at present — the possessors of one
of the original undivided shares of village lands — was poverty-stricken
two generations ago when the sole member of the family was a teen-age
boy who could not work the land properly and so had to hire out as a
servant to others to make a living. Years later, when he and a son were
both able to farm and to hire a servant as well, he made profitable use
of the same lands. Acquisition of extra wives, servants, and adopted
sons are means used to increase the labor force in a household and
thereby to derive more benefit from the land.

Although there is variation in wealth and land ownership by family
and by caste, there are no big landholders in the sense of the "zamin-
dars" of the plains, and there are no tenant farmers in Sirkanda. Sir-
kanda agriculturists are owner-cultivators; they live by their own work
on their own lands. The biggest landholding recorded in Sirkanda is
20.4 acres, owned by a joint family of 22 individuals (nine adults,
thirteen children below 16 years of age). The smallest holding by an
agricultural family (that is, one of high caste, though some of the low-
caste families own none) is 2.3 acres, supporting six people (two adults,
four children). The average size of recorded holdings in the agricul-
tural castes is 4.5 acres per eight- or nine-member joint family. Most
agricultural families have about one-half acre per adult, counting two
children as equivalent to one adult. Some have less, and the most pros-
perous have one acre per adult.

There is little rental or sharecropping of lands in Sirkanda. One man
has a share in some lands owned by his wife's relatives in a distant
village. Another lets out on shares some lands he owns that are too far
away for him to look after. One man, whose lands were too scattered
for him to look after, effected a mutually advantageous exchange of
lands with another farmer so that lands of each were somewhat con-
solidated. Occasionally some elderly or dependent person who cannot
tend his or her land will rent it out to others on a share basis, as one
Sirkanda widow does. Others sell land they can no longer work. More
commonly, the owner gives the land to whoever is slated to inherit it
and joins that household for the remainder of his lifetime. Several
instances of the latter arrangement are in evidence in Sirkanda.

Until Independence it was possible for any agriculturist to extend
his holdings by cultivating previously uncultivated land. The primary
requisite was labor — it is hard and time-consuming work to clear and

44 THE ECONOMIC CONTEXT

terrace new land. The availability of land prevented the pressure on
land common in much of North India. An expanding Pahari agri-
cultural family in this area could usually, with effort, expand its hold-
ings.

For low castes the story is different. In pre-British times they were
prohibited from owning land and from cultivating new land without
permission from high-caste villagers. Under the British they were pro-
hibited by informal but effective village sanctions from seeking new
lands. They were prevented from becoming economically independent
rather than from farming at all. This situation has continued to the
present when, as will be discussed in the context of intercaste relations,
a blacksmith has been denied access to unused land in Sirkanda by the
high-caste village council. Thus, low castes have been kept dependent
and poor while both cultivated lands and population have increased.

Throughout the period since 1815, the main settlement of Sirkanda
is said to have remained about the same in population. Records
of land within the village boundaries show very little change since
1904. However, in this time lands outside the village have come
under cultivation or have been purchased until the village cultivates
twice as much land as formerly. Also, the lands have been increasingly
fragmented. The resident population of Sirkanda has remained about
150 or 175, but the population of Bhatbair and vicinity has doubled
or tripled. Additional village population has been drained off to chans
and second houses. The number of villages in Bhatbair has doubled
since 1815. Chans, which were a rarity then, have been built and fields
cleared where there was previously only wilderness. The trend is
evident even in the period since Independence with the addition of new
lands and new chans in several locations. Within Sirkanda the entire
southern section of the village, comprising five houses, is a post-
Independence phenomenon. As old houses in the village have become
decrepit or crowded, new ones have been built in the new location. New
lands have been opened up on the southern periphery of the village
surrounding the new houses. It is one of the ironies of the post-
Independence era that governmental restrictions have been imposed
which discourage expansion of agricultural lands into hitherto un-
cultivated areas. This is a source of keen resentment among high-caste
Paharis of this region because, until Independence, uncultivated lands
had been considered the property of the villages nearest them, and
available to their members for cultivation. Now they have been made
subject to stringent governmental regulation. As one Sirkanda farmer
pointed out: "The government asked us to increase crop production
to help the nation. When we did so in the only way possible, by open-

THE ECONOMIC CONTEXT 45

ing up new lands, they accused us of infringing on government property
and brought a suit against us."

Agriculture

Agriculture is a family enterprise — those who share the hearth
share in the work and the harvest. There are two annual crops and
harvests which govern the work of villagers and influence the yearly
cycle of all activities in Sirkanda. These are "millet harvest" in early
winter (mid-September to October) just after the rainy season, and
"barley-wheat harvest" in mid-summer (April) before the hot, dry pe-
riod. These correspond to the kharif and rabi harvests of the plains,
respectively, but are not called by these terms. Each planting season
follows closely after the previous harvest.

Winter Crop

While the millet harvest is in progress, men of each family begin
plowing the harvested fields preparatory to planting the principal
winter crops, wheat and barley. Stalks of the previous crops are burned
and the ashes are plowed into the ground along with manure, which
has been spread on the fields. The plow, similar to those used on the
plains, is wooden with a wooden moldboard and an iron bit and is
pulled by a yoke of bullocks. A week or two after plowing men sow
the fields by hand broadcast. Immediately thereafter the fields are
again plowed and a leveler is dragged over them. The fields are plowed
and seeded between mid-October and the end of November. Ideally,
approximately equal proportions of land are devoted to wheat and
barley. In reality, wheat, the preferred crop, may be sown in two-
thirds of the available 200 acres of land in the vicinity of Sirkanda.
Masitr (Ervum lens), a pulse, is also planted at this time but in much
smaller quantities.3 Some land is left fallow. Once planting is com-
pleted the main agricultural activity is transporting animal manure
and spreading it on the fields. This is done two or three times after
planting, at intervals of about a month.

In February and March, over a month before harvest time, the
fields that have been left fallow over the winter are plowed for the
first time in preparation for planting janghora (Oplismenus frumen-
taceus), a millet which is one of the principal crops of the rainy grow-
ing season. The plowing continues, twice per field, over a period of
six weeks, overlapping with the barley harvest, since barley fields as
well as fallow ones will be planted in janghora. Toward the end of
March and in early April barley and masur are harvested, and the

46 THE ECONOMIC CONTEXT

fields are plowed after the stubble or stalks have been burned. Wheat
matures shortly after the other winter crops and is harvested in mid-
April. As wheat is harvested threshing begins. This is done by driving
cattle over the grain heads, which have been spread on a hard, dung-
plastered circular threshing floor. The cattle, two to ten in number,
are hitched to a central post and driven around by people of almost
any age and either sex. It is an eight- to ten-hour job to thresh one
batch of about two maunds (160 pounds) of unthreshed grain. There-
after the grain is winnowed by women in traditional North Indian
fashion with a winnowing tray and the aid of the breeze. Finally ashes
are mixed with the grain "to prevent worms from infesting it." Women
do this by trampling the grain and ashes on a threshing platform. It
is then stored in the home in large baskets sealed with dung-plaster or
in wooden storage boxes. Each family spends about two weeks at
threshing, winnowing, and storing. Since not every family owns a
threshing platform (there are 21 platforms in the village), they are
borrowed for further use as soon as the owners are finished. Such
borrowing occurs freely across sib and caste lines. The chaff is burned
and the straw is saved and stored in tall stacks supported by central
poles or saplings, to be used later as thatch or fodder.

Rainy-Season Crop

Before and during the wheat and barley threshing period, janghora
is planted in the same manner as was wheat the previous season. Con-
currently, the harvested wheat fields are burned over and plowed.
After plowing they are left for about two weeks and then planted with
rainy-season crops other than janghora, primary among which is a
millet called khoda or mandud (Eleusine coracana). Janghora and
khoda are the principal rainy-season crops, forming about 75 per cent
of the grain product of the season. They are planted in roughly equal
amounts. At the same time are planted kulat (Dolichos biflorus), a
pulse, and chauldi {Amaranthus polygamous, Amaranthus blitum), an
amaranth which supplies a grain (forming about 20 per cent of the
grain harvest of this season) and also leaves which are used as vege-
tables. Other crops of the season include dry and wet rice, taro, pump-
kins, beans, corn, ginger, chili, cucumbers, leafy vegetables, and to-
bacco. Wet rice is planted at this time in seedbeds in areas where
irrigation is possible. At the beginning of the rains (late June) wet
rice is transplanted from seedbeds to irrigated fields, and chili plants
are transplanted to Sirkanda gardens from irrigated seedbeds else-
where.

The fields are manured during the rainy season just as they were

THE ECONOMIC CONTEXT 47

in the winter. From shortly after the rains begin, in July, until the
first of September, harrowing, followed by weeding of the fields, is a
full-time job. First the fields are harrowed from one to three times
each with a long-toothed harrow pulled by bullocks. This has the
effect of thinning the fields, which were purposely overseeded to pre-
vent erosion and uprooting of the seedlings by the heavy rains. It is
also alleged to kill weeds, the belief being that grain withstands the
disturbance better than weeds do. Some incidental transplanting may
be done at the same time to even out plant distribution in the fields.
Thereafter the fields are repeatedly and painstakingly weeded by hand
by the entire families, using small pick-like tools. By the first of Sep-
tember weeding is no longer necessary, and the harvest begins in late
September. First janghora, corn, and beans are harvested, followed in
about two weeks by khoda, chaulai, kulat, and the rainy-season crops.
The harvest continues to the end of October. Threshing of the millets
extends throughout October and overlaps with plowing and sowing
for the winter crops.

Maximizing Productivity

In the planting of crops there is an ideal pattern of crop rotation
said to enhance the productivity of the soil. The sequence in an in-
dividual field is as follows: (1) barley, followed by (2) janghora or
dry rice, followed by (3) wheat, followed by (4) khoda, followed by
(1) barley, and so on. Although the sequence is not adhered to re-
ligiously, it is a traditional pattern that is approximated by most
farmers in Sirkanda. Periodically crops not included in the rotating
sequence are planted in some of the fields. Any field that appears to
be losing its vitality as evidenced by smaller crops is left fallow for a
season or two. The above pattern applies only to fertilized fields. Those
which cannot be adequately fertilized or are poor to begin with are
left fallow every other season. Often entire hillsides are planted in a
single crop and are rotated in the same sequence, a result of informal
consultation among the various people who own the many plots in-
volved. Crop rotation is practiced in most Pahari areas but in differ-
ent patterns (cf. Stowell, 1907, p. 11; Walton, 1910, pp. 26 ff.; 1911a,
p. 55; Pant, 1935, pp. 136 ff.).

Few fields can be irrigated in this terrain. Stowell reported in 1907
(p. 9) that only 3 per cent of all cultivated land in Garhwal was irri-
gated, and 8 per cent in Almora. Only three of the 205 acres within
Sirkanda's boundaries are irrigated, plus 25 to 30 of the 93 acres which
lie outside of Sirkanda and are owned by Sirkanda people. These 25
to 30 irrigated acres are nearly all low, flat lands bordering the valley.

48 THE ECONOMIC CONTEXT

Irrigation of fields in these hills is an engineering feat of consider-
able complexity. Water is obtained from streams or springs which are
blocked with stone dams at distances varying from a few hundred
yards to more than a mile upstream from the fields to be watered. From
the dams narrow channels are constructed following the contour of
the land at sufficient gradient to keep a steady flow of water. Where
these canals skirt bare rock cliffs or bridge small ravines, they are
made of wood lined with clay and are held in place by wooden or
stone supports wedged in crevices. The canals empty into the highest
of the fields to be irrigated, and outlets are so arranged as to provide
maximum subsequent use of the water in lower fields. Irrigation canals
are built and maintained cooperatively by those who use them — usually
high-caste people of a single village.

The nature and extent of cultivation of some crops is limited by
the type of land available; that is, irrigated lands are necessary for
wet rice; shady, loamy lands are required for potatoes. Another im-
portant factor is animal molestation. Corn is grown only in small
quantities and in fields adjacent to the village because bears and por-
cupines destroy more distant stands. Potatoes are grown only in the
vicinity of chans in most areas because wild pigs destroy unsupervised
fields. Monkeys menace several crops. No specific measures are taken
to ward off these pests other than to plant their favorite crops where
they are unlikely to come and, occasionally, to build a shelter near a
field where someone can stay or a dog can be tied to scare off the
marauders. Persistent animal pests may be ambushed and shot.

Thus far only the "practical" means by which agriculture is car-
ried out and good crops are ensured have been discussed. To enhance
success there are also supernatural means, ranging from an annual
ceremony to deities who influence crops and weather (described in
Appendix IB) to beliefs as to the appropriate day of the week for be-
ginning planting and harvest. Ceremonials and beliefs of this nature
are not extensive and seem to be of decreasing importance, perhaps
reflecting the relatively reliable and adequate nature of Pahari agri-
culture. Difficulties with crops are usually treated, as are illness, finan-
cial reversal, or other adversities, as manifestations of the displeasure
of specific household or village deities. These matters will be discussed
in the following chapter.

Cash Crops

Growing crops primarily for sale is a relatively recent innovation
in the village. The most important such crop now is potatoes. Several
varieties with different growing seasons are cultivated, so that some are

THE ECONOMIC CONTEXT 49

in the ground at almost any time of year. Potatoes are grown in chan
locations north of Sirkanda and usually above it in altitude. Less than
half of the villagers own land suitable for this crop.

Other cash crops grown on Sirkanda-owned lands include ginger,
onions, garlic, koreander, and the surpluses of wheat, rice, and taro.

Preparation and Consumption of Food

Year-around daily staples eaten in Sirkanda are the millets jang-
hora and khoda (mandua). Wheat and barley are often mixed with the
latter variety of millet. In addition, dal (pulses) and vegetables, pri-
marily taro and potatoes, are eaten regularly.

Grain must, of course, be milled before use. Normally milling is
done at intervals after harvest so that an adequate supply is on hand
for use, but at any one time most of the stored grain is unmilled. Two
water-powered mills are patronized, payment in flour being made to
their owners at each milling. The mills are four and five miles distant
in opposite directions, the farther and more patronized being en route
to points in the valley and owned by a member of a Sirkanda Rajput
household. In addition, some families have small stone hand mills
which are used occasionally. The daily supply of rice and janghora
is husked each morning in a stone mortar with a long wooden pestle.
There are several community mortars in convenient locations around
the village and individual ones at several houses. Pounding in them
is regular early-morning women's work.

Cooking is done by women in an inner room of the house, over a
small wood fire in an open, dung-plastered stone fireplace. Brass and
iron utensils are utilized as in the plains, and the style of cooking is
similar to that in the plains.

Two large meals and one or two small ones are prepared daily.
At 6:30 or 7:00 a.m., about an hour after rising, there is a light meal
consisting of leftover chapaties (unleavened bread) eaten with salt and
chilis or milk or one of several milk products. Between 10:30 a.m.
and noon a large meal is prepared consisting of janghora boiled in
water and served with boiled dal. Boiled vegetables (potatoes, taro,
pumpkin), and occasionally curds, may also be served. At 4:00 to
7:00 p.m. another light meal may be served, consisting of leftovers from
the noon or previous evening's meal. This meal is sometimes omitted.
Between 8:00 and 10:00 p.m., one to two hours before retiring, another
complete meal is prepared. This consists of chapaties made of khoda,
often mixed with wheat or barley, and a vegetable mixture, usually of
potatoes or taro cooked with spices and chilis. Water and rarely milk
or buttermilk are drunk after each meal. This fare is considered very

gO THE ECONOMIC CONTEXT

plain and inferior to that which is served guests, in which the chapaties
are made of wheat, rice is substituted for janghora, and milk, ghee
(clarified butter), and other fancier foods are served. Paharis are self-
conscious about their food before outsiders and frequently make depre-
catory remarks about the poverty of Paharis and Pahari food. "Jang-
hora is Pahari basmati [the finest grade of rice]," is a frequent wry
comment in Sirkanda.

When guests are present tea is served regardless of time of day,
though it is rarely used otherwise. Ceremonial feasts are elaborations
of the daily fare with a variety of pulses, ghee, curds, vegetables, pickles,
sweetened rice, and halwa (a sweet dish) generally added to it. In
any case the food is almost exclusively of locally grown products, so
that cash expenditures on food are minimal.

Animal Husbandry

Second to crops in the economy of Sirkanda is livestock. A man's
first investment after land is in livestock. Most important of his animals
are his bullocks or oxen which pull the plow. The usual way to desig-
nate how much land a man owns is by how many oxen he uses to till it.
In Sirkanda there are twenty-five "two-bullock" (one-plow) house-
holds, twelve "four-bullock" households, three "six-bullock" house-
holds, and one "eight-bullock" household. A total of 124 bullocks
are owned by 41 households — a sizable investment in view of the
fact that a good pair of oxen are worth 250 to 350 rupees. No family
owns more oxen than it uses, and any that are not in good condition
are quickly sold.

Almost equally important are cows and buffaloes. These are kept
for their two valuable products, milk and manure. Animal manure is
considered necessary for agriculture and is often sufficient reason for
keeping an animal. Milk is a cash product. Most of it is sold to shop-
keepers in the form of milk, butter, ghee (clarified butter), or a boiled-
down solid product used in the making of sweets. Relatively small
amounts are used in the village.

Most cows and many buffaloes are kept at chans and are moved
seasonally from low altitudes in winter to higher altitudes in summer.
This protects them from climatic extremes. It also gives them an
optimal diet, and brings them close to the milk markets of Dehra
Dun or smaller, closer markets in the valley during the winter, and
to Mussoorie (with its big milk trade during the vacation season) or
closer hill markets in the summer. Sirkanda villagers own about 725
cows worth Rs. 70 to 150 each, and 140 buffaloes worth Rs. 250 to

THE ECONOMIC CONTEXT 51

400 each. Buffaloes are more valuable as milk producers than cows,
but are harder to care for since they are too heavy and awkward to be
allowed out to pasture or to get water in this rugged terrain.

In the markets buffalo milk is generally preferred to that of cows,
but in the village cow's milk is often preferred. It is admitted that
buffalo milk is "better for the body" and makes men virile (any ap-
preciative wife or lover will feed her man buffalo milk and its products
frequently), but it does not help the brain. One villager commented
that, "If you drink too much buffalo milk you will get buffalo wis-
dom," that is, stupidity. The difference in quality of milk from these
two sources is locally ascribed to diet — a cow has a more varied diet
since it grazes for itself, and therefore its milk is superior. On the same
basis goat's milk, though it is not used (see below), is better than that of
cows, and woman's milk is best of all.

Buffaloes are occasionally killed as offerings to gods in Bhatbair.
Their flesh is said to be eaten only by people of the shoemaker caste,
but there is some evidence that others, even in Sirkanda, may occa-
sionally eat buffalo meat. The sanctity of cows is as strongly felt in
the hills as in the plains, and they are never killed nor is their meat
eaten.

About 350 goats and 10 sheep are kept by a total of 24 families in
Sirkanda. They are raised primarily for sale in the markets, where they
bring Rs. 15 to 30 each. In a year 10 to 20 goats may be sold by a
single household. In addition, their manure is considered more valu-
able as fertilizer than that of other animals. Their flesh is eaten in
the village — usually after they have been used as religious sacrifices.
In the entire village not over 20 or 30 goats are used yearly for sacrifice.
Occasionally a goat is killed solely for eating, but this fact is not
readily admitted. Villagers imply that a sacrifice may sometimes be an
excuse to eat meat, but in no case is meat eaten as part of a regular
meal. Instead, it is treated as a special feast and is generally consumed
with liquor. Goat's milk is not used. The proffered explanation for
this is that one cannot take the milk of an animal whose flesh he eats,
nor vice versa, because an animal that gives milk is in some respects
like a mother. This is also a reason for not eating buffalo meat.

Horses and mules are used as beasts of burden and for human
transportation. There are now 15 horses and two mules in the village,
owned by 16 of the larger high-caste households and representing an
investment of Rs. 200 to 1,000 each. They are a relatively recent in-
novation in the village; previously all loads were carried by the men
and women of the village.

It is worth noting that livestock, like land, is predominantly high-

£2 THE ECONOMIC CONTEXT

caste property. The 10 per cent of the population that is of low caste
owns 2 per cent of the cattle, 5 per cent of the bullocks, 6 per cent
of the buffaloes, and 8 per cent of the goats. They, more than many of
the high-caste families, find it worthwhile to supplement their income
by goat breeding, because they need the money and have the time to
devote to this activity. They own no horses or mules.

Other Economic Activities

No animal fodder as such is grown by villagers. The straw of
grains is often kept for use when other fodder gets scarce, but the
major dependence is upon wild grasses and leaves collected daily the
year around, green or dry as the season permits. This is a major oc-
cupation for women of the village. In addition, all animals except
buffaloes are pastured daily, and all of these except horses and mules
are attended at all times. No one goes to the forests to shepherd animals
without bringing in a headload of fodder or firewood. In the months
of May and June grass fires are lit by villagers in order to burn over
the hillsides and ensure a superior grass crop after the rains. These
fires — the bane of the government forestry officers — dot the hills during
this season. Often, and not surprisingly, they leave burned chans, and
occasionally dead livestock, in their wake.

For two or three months before the rains the year-round activity of
wood collecting is intensified in order to prepare a stockpile of dry
fuel for the rainy season. It is the availability of wood for fuel which
makes possible the extensive manuring of fields, in contrast to the
plains custom and necessity of utilizing all manure for fuel.

Housebuilding and repair are winter and summer activities. Con-
siderable frantic roof repair takes place after the rains begin, revealing
its need. Agricultural terraces and stone fences are built and repaired
when the need arises and when other work permits, especially before
and after the rains, which do considerable damage to terraced fields
and retaining walls.

The agricultural slack season comes in December and January, when
only daily routine work and building has to be done. During this time
men engage in minor activities such as rope-making and repairing
equipment and houses, and the whole family contributes to yarn-
making, knitting, and the like in idle moments. Somewhat the same
situation exists in the midst of the rains.

Beekeeping is an incidental enterprise conducted by many house-
holds in the village and in chans. A small hole on the outside wall of a
house leads to a cubical space about a foot square in the wall with a

THE ECONOMIC CONTEXT 53

removable board at the back of the space, inside the house. Wild bees
nest in this space. Twice a year (in April and October) the bees are
smoked out and the honey is removed and sold in the markets. Hunting
and gathering play a very small part in the Sirkanda economy. Although
most high-caste households own guns, they are used primarily to
frighten or kill marauding animals. Occasionally native wild chickens,
partridges, wild pigs, wild goats, or deer are shot for meat. If a freshly
killed prey of some carnivorous animal is found it is eaten by villagers,
though they admit it reluctantly if at all. Fish are obtained from the
rivers by low-caste people of villages near the rivers who net them to
sell, or by young men of Sirkanda who make a picnic excursion of the
fishing and a party of the ensuing feast. In late summer several varieties
of wild berries and wild apricots and peaches may be collected. Except
for the apricots, which are sold in urban markets, there is relatively
little enthusiasm for these products.

Production and Income

Crops and livestock provide the high castes with most of their
livelihood. Estimated annual crop yields were obtained in detail for
20 (just over half) of the high-caste households. The estimates indicated
that almost 150 per cent of the household's annual grain needs were
produced in the fields during a good year. An average household
(eight or nine persons) consumes about 50 to 60 maunds (4,000 to
4,800 pounds) of grain per year and produces 75 to 90 maunds.4 Each
harvest is customarily stored and used throughout the year for sub-
sistence and for payments to artisans and priests. The surplus is sold
only when the succeeding harvest is in. Sale of surplus grain and cash
crops together were reported to yield about Rs. 600 in a good year for
a nine-member family. This would amount to around Rs. 20,000 for
the whole village.

The main source of cash and credit is dairy products. Estimates of
milk production and sale in 39 households indicated an annual cash
income of around Rs. 800 per family from this source. Total yearly
cash expenditures for a nine-member family were roughly calculated at
Rs. i,ooo.5 Thus total income was estimated to exceed total expenditure
by an average of Rs. 400 (equal to about 27 maunds of grain) per fam-
ily per year among land- and cattle-owning families. The estimates are
extremely rough, and it seems probable that both income and expen-
ditures are exaggerated. Moreover, the sample is slightly biased in favor
of the more prosperous joint families. The estimates indicate relative
orders of magnitude rather than precise quantities. There is a range,
with some families exceeding the average and others falling well below

54 THE ECONOMIC CONTEXT

it. Some years bring larger or smaller than average crops, greater or
smaller expenditures, on family and village levels. Some families spend
more than the average on ceremonies and other approved means of
conspicuous consumption, while others economize. There are significant
caste differences, with the service castes having considerably less to live
on than the landowning castes. Low-caste incomes are discussed below
under the description of their occupational specialties. These figures
do serve to suggest, however, the relative prosperity of this village com-
pared to many Indian villages. Moreover, they check with separate in-
quiries, such as that into the village accountant's official estimates of
crop yields per acre multiplied by acreage planted.

Only one "famine" year is recalled by Sirkanda residents, and it
occurred some fifteen years previous to the research. Crops were in-
sufficient to meet food requirements that year, and a number of families
had to go to Dehra Dun to buy grain. The significant fact is that they
were able to buy it; no one went hungry, and no one went deeply in
debt.

It is apparent that some families will accumulate wealth under these
conditions. In Sirkanda, when income exceeds that which is necessary
for family welfare and livelihood, a man first sees to his lands and then
to his livestock. If these are sufficient he is most likely to use his money
for house- or chan-building. If these too are sufficient he may buy
household utensils and jewelry. Beyond this, with the simple needs of
life in the hills met, extra money is stored in concealment or part of it
is loaned on interest (15 to 25 per cent) to other villagers or reliable
acquaintances. Money is never kept in banks. At least six Sirkanda
families regularly or occasionally lend money. These include the three
families who retain their original undivided lands and three others
with higher than average land-to-person ratios. The wealthiest man of
the group is variously estimated to have Rs. 8,000 to Rs. 50,000, the latter
doubtless a greatly exaggerated estimate. Borrowing is done to finance
such major outlays as marriage (which averages from Rs. 2,000 to
3,000 for a son and, depending upon the type of marriage, from nothing
to Rs. 1,500 for a daughter); house construction (which ranges from
Rs. 2,000 to 4,000); purchase of animals, purchase of land, costs of legal
cases. Several Sirkanda families have been chronically in debt from
such expenditures in recent years, but the relative wealth of families
shifts with time. Thus, of the four wealthiest families in the village,
two have become so recently — one in the generation just past, the other
since Independence. Both of the others have increased their wealth
during the past generation. Several other families are said to be less

THE ECONOMIC CONTEXT 55

well off than in previous generations. This type of shift has in all cases
but one been linked with a change in the man-land ratio. The more
land per man, up to a point, the more wealth is accrued. With too little
manpower, even land is useless, and in one of these cases wealth came
when a man finally acquired a son and a servant to help him work his
more than ample lands. In one case sudden wealth came from an un-
expected windfall in the form of money acquired by questionable
means from fleeing Muslims at the time of the partition of India and
Pakistan.

There are few people in or out of the hills who will not attest to the
poverty of Paharis. As applied to Sirkanda and neighboring villages
this stereotype is a fiction, based largely upon the simple material pos-
sessions and hard work that characterize Pahari life. While it is true
that by plains standards Paharis do not acquire great wealth and that
what wealth they do acquire is not displayed, it is also true that they do
not live the marginal existence led by many of their plains brethren, nor
have they faced periodic famines and droughts as have people of Bihar
and Bengal.

Trade and Markets

Neither Sirkanda nor the interdependent and relatively isolated
group of villages which make up Bhatbair and the adjacent area in
Tehri Garhwal are entirely self-sufficient. They are and perhaps always
have been dependent upon larger centers. This relationship is an im-
portant prerequisite for classifying these Pahari villages as an example
of peasant society. A peasant society is one composed primarily of
people who make their living by agriculture and who live in symbiotic
interdependence with market towns or urban areas though living away
from them (cf. Kroeber, 1948, p. 284; Redfield, 1957, pp. 31 ff.).

The peasant has some product which the city consumes, and there are products
of the city — metal tools, guns, patent medicines, or electric flashlights —
which the peasant takes from the manufacturers in the city. Since the coming
of money into the world, the peasant village has come in great degree to define
its economic affairs in terms of this measure. (Redfield, 1957, pp. 31 f.)

Having been included in the "economic nexus of civilized society," the
peasant is imbued with "a spirit of pecuniary advantage" (ibid.).

The dependence of Sirkanda peasants on urban centers is increasing.
Sixty years ago one trip per year to a major market sufficed to exchange
the agricultural surplus directly or indirectly for major items not
locally available, and to transact necessary business with official agents

56 THE ECONOMIC CONTEXT

of authority. Several trips to intermediary markets and occasional
visits from peddlers provided those items needing more frequent re-
plenishment, such as salt, sugar, and glass bangles. As transportation
has improved, urban areas such as Dehra Dun have increased their
direct influence on the village. Some men go every month, and most
go several times a year. Those living in chans may go weekly or oftener
because they are close to town and have milk products to sell. Inter-
mediary markets are visited with proportionately increased frequency.
The range of products sought and of agencies providing them has
increased in the area as money has become more readily available.
Milk and agricultural surpluses are sold to get money to buy an in-
creasing array of goods, to pay taxes and court costs, and even to pay
village artisans.

Reliance upon products and services deriving from urban areas is
therefore a traditional feature of the Sirkanda economy which is be-
coming more prominent. Today the urban economy reaches into the
village in many ways.

There are two shops in Sirkanda. One is owned and operated by a
large household of Rajputs in Sirkanda. It sells mostly staples such as
cigarettes, matches, kerosene, cooking oil, sugar, salt, spices, and tea to
villagers, and hot tea to passing travelers, on a cash basis. The other is
operated by a merchant from Rajpur, who was asked to come by a group
of villagers who did not want to trade in the other shop because of
personal rivalries. He has run this shop for some 25 years.6 His trade
includes selling staples, often on credit, but his main business is buying
some of the villagers' produce, notably the solidified milk product,
pumpkins, apricots, lemons, and anything else he can resell in Dehra
Dun at a profit.

Lower prices and greater selection encourage villagers to buy at
larger centers except in emergencies. Likewise, they can sell their
produce to greater advantage in market towns. Dehra Dun and Mus-
soorie are the large trading centers, but smaller, closer ones are fre-
quently patronized. Suakholi, ten miles from Sirkanda and six from
Mussoorie on the Mussoorie-Tehri trail, is much used by hillmen
throughout this area. Similarly Sahas Dhara, Nagal, and Rajpur are
traditional markets below Sirkanda, five to nine miles away. Supplies
for such special occasions as marriage, and items not produced locally,
such as salt, sugar, oil, jewelry, trinkets, cloth, utensils, and some tools,
are often obtained from these places. In each market are particular
merchants with whom Sirkanda farmers regularly trade. The relation-
ship becomes somewhat personal, so that villagers may stay overnight
in the trader's shop when they visit town even if they have no business

THE ECONOMIC CONTEXT 57

to transact with the shopkeeper. The shopkeeper may also serve as an
informal adviser on matters unfamiliar to the villagers.

Itinerant peddlers come through Sirkanda selling bangles and other
small items of decoration and convenience two or three times a year.
One who came and stayed three days while I was in the village sold
trinkets worth seven maunds (560 pounds) of grain, for a gross of about
Rs. 100 and a profit of at least Rs. 70, in Sirkanda alone. Even with an
extra rented mule, he had to make a return trip to carry off his pro-
ceeds. Interestingly enough, he arrived shortly after the annual fair
at which the people had already spent a good deal of money, much of it
on items which this very peddler had been selling at the fair.

The annual fair (taulu) of Bhatbair is held a few miles from Sirkanda
and is by far the most important event of its kind for the residents of
Bhatbair, but it is too provincial to attract outsiders other than two or
three merchants. Other fairs attended by Sirkanda villagers are those
held annually at Rajpur, Mussoorie, and at a well-known temple on
the Mussoorie-Tehri trail. At all of these there is ample opportunity
to spend money on tea, refreshments, sweets, trinkets, entertainment,
liquor, and women. A number of Sirkanda villagers set up concessions
at the annual Bhatbair fair, but aside from the Ferris wheel operated
by Bajgis of Sirkanda and a nearby village, these rarely show much of a
profit. The profiteers are the two or three outside merchants who sell
trinkets, the clandestine sellers of liquor, and perhaps some of those
who sell sweets, which are the medium of exchange for obtaining sexual
favors from women, one of the major activities of this fair. The in-
experience of the amateur village entrepreneurs who set up booths at
this fair generally dooms them to financial failure. They join the spirit
and activities of the festival and forget to collect cash for their wares
or are distracted from their shop-tending by other activities, while
their stores are good-naturedly depleted.

The Jajmani System

A number of essential activities in the village, especially those
requiring particular skills or knowledge, are performed by specialists.
Most of these occupations are thought of as caste monopolies, and the
arrangements for work and payment are standardized in some form of
traditional exchange, known widely in North India as the jajmani
system (cf. W. H. Wiser, 1936; Beidelman, 1959; Berreman, 1962b, in
which part of the following has appeared).

When Sirkanda villagers use the term jajman they refer to one kind
of exchange: that of the Brahmin's ritual services to his clients (jajmans)
in exchange for traditional "gifts" paid in grain or other goods. This

^8 THE ECONOMIC CONTEXT

is in accordance with widespread usage of the term among villagers in
North India and with its etymological meaning: one who asks another
to perform worship and offers a gift to him in return.

Sirkanda villagers also understand, though they do not themselves
normally use, the term jajman to refer to the traditional arrangement
whereby an artisan serves the needs of an agriculturist in his specialty.
In this relationship the artisan or service caste member is paid a fixed
portion of grain at each harvest, the amount depending upon the size
of the household or landholding of the agriculturist and the type of
service performed.

The Pahari term which Sirkanda villagers normally use in reference
to an artisan's clients, parallel to their use of "jajman" in reference to
a Brahmin's clients, is gaikh, one who purchases the service of another.
Ideally the relationship between artisan and gaikh is a permanent one
with standard traditional payments, but in practice there is a good deal
of shifting, especially where there is more than one local artisan avail-
able.

A third kind of traditional economic exchange in the village is that
of service among artisans. This is not usually included in either the
term "jajman" or "gaikh" by villagers. Blacksmiths, for example, see to
the ironwork needs of the local drummers, who in return drum for the
blacksmiths as needed. Agreements to this sort of exchange are about
as stable as those in the gaikh arrangement.

Finally, many services are performed, and have been traditionally
performed, on a piecework or daily wage basis, with cash or grain used
for payment. These include the sporadic services of ceremonial cooks,
stonemasons, wood turners, and others, who are not resident in or near
most villages, as are blacksmiths and drummers. They include also the
nontraditional or non-caste-specific activities performed by local ar-
tisans. Tailoring and basketmaking, for example, are in this village
done by drummers. Payment for these services is on a piecework basis.

Any low-caste person may be called upon to help a high-caste villager
in emergency jobs or in tasks which the high-caste person is unable or
unwilling to do himself. Most often the high-caste person calls upon
one of the artisans who serve him, that is, for whom he is a client. The
low-caste person cannot refuse to help without a good reason such as
physical incapacity or an urgent prior commitment to another client.
Otherwise he risks economic or physical sanctions. This is not the case
with either Brahmin-jajman relations or intraservice caste relations.

Although each caste in and around Sirkanda is identified with a par-
ticular occupational role, there is considerable flexibility in the system.

THE ECONOMIC CONTEXT 59

Important in the jajmani or gaikh system is the fact that service castes,
unlike Brahmins, have a difficult time maintaining a monopoly on their
services and often have no other reliable source of income such as
farming. The traditional occupation of each caste may remain caste-
specific, but its members' livelihood may depend largely or entirely on
specialties which are not restricted to a single caste. Under conditions
of necessity or even convenience, many occupations are interchange-
able among low castes. Occasionally a high-caste person will perform an
artisan's work. Therefore, if an artisan quits working for a client there
are others, even outside his caste, who can and will take his place.

Another feature of significance in economic organization is the ab-
solutely small number of individuals who make up each local artisan
caste group, often only one nuclear family or two closely related and
recently separated families. Probably partly as a result of this fact the
artisan castes are not very cohesive groups. Unlike the high castes, they
have little or nothing in the way of group organization and discipline
beyond local kin-connected groups. Moreover, in many areas, as popu-
lation has increased, so has access to bazaar-made goods. Agriculturists
have become less dependent upon artisans' services and less inclined
to make full payment to them. There is increasing competition among
artisans for a decreasingly remunerative clientele. As a result there is
considerable friction among them. In Sirkanda the local blacksmith
and drummer castes are both divided within themselves between two
competing and often hostile brothers, neither able to make an adequate
living at his traditional occupation and therefore supplementing it by
brewing illicit liquor, raising goats, practicing medicine and divination,
tailoring, or carrying on agriculture if he has land. Thus gaikh rela-
tionships are unstable, with much jockeying for the better-paying cli-
ents on the one hand, and for the better-performing artisans on the
other. The effect of this situation is shown by the fact that only among
service castes do brothers frequently dissolve the joint family and di-
vide the common patrimony.

Exploitation in the sense of arbitrary and self-seeking control over
the behavior of others is characteristic of the relationship of agricul-
turists to artisans in Sirkanda. While many clients are responsible in-
dividuals who do not often overtly exercise their power over the low
castes, still, as in the village of Karimpur described by the Wisers, "let
there be any move toward independence or even indifference among
them, and the paternal touch becomes a strangle-hold" (C. V. and
W. H. Wiser, 1951, p. 18).

Exploitation is not characteristic of the ritual relationships between

60 THE ECONOMIC CONTEXT

Brahmins and their clients, nor in the exchange of services among the
artisan castes. The reason for this pattern seems to lie largely in the
distribution of power — the control that local castes exert relative to
one another.

In Sirkanda low castes are dependent upon high castes for livelihood
— almost absolutely so. Moreover, they have little leverage on the high
castes because of the absence of caste cohesion among themselves and
the absence of caste monopolies on the essential goods and services
which they provide for the high castes. The high castes, in contrast,
are generally well organized and able to present a united front to the
artisans of any local area. Moreover, potential clients are sufficiently
restricted in numbers, and villages are sufficiently isolated from one
another, that village artisans often find it difficult if not impossible to
seek new clients. In the village context, therefore, they subsist at the
discretion of their high-caste patrons.

Low-caste people who have become agriculturists no longer depend-
ent upon their craft skills are, like Brahmins, free of obligation to, and
potential exploitation by, other agriculturists. They have in most cases
moved out of their traditional villages to nonvillage locations where
they do not have to deal frequently with the high castes and where they
perform most of their own labor.

In the exchange of services among low castes in Sirkanda there is no
opportunity for exploitation, since none controls the livelihood of
others nor performs unique and essential services.

A feature of the artisan-gaikh relationship that should not be over-
looked is the part played by high-caste clients in the work of the low-
caste artisan. A scene repeated daily for several months during the
building season of the writer's year in Sirkanda took place at the
construction site of the Brahmin family's imposing new house. The
two carpenter-masons were engaged in their skilled but relatively lei-
surely work of placing and mortaring stones, making window frames,
or carving columns. Meanwhile the Brahmins, women and men,
struggled with immense loads of stone from the quarrying place %
mile distant. They quarried the stone, dug, transported, and mixed
mortar clay, felled and transported trees, carried corrugated iron sheet-
ing from the valley, handed stones up to the masons (who might airily
reject any particular stone), and generally did all the heavy, dirty work
of the construction. Their exhausted and begrimed countenances were
in striking contrast to those of the masons, who often sat and waited
for more materials to be brought. Lesser manifestations of the same
phenomenon were to be seen when Rajputs operated the bellows or
wielded the sledge under the direction of the blacksmith on jobs too

THE ECONOMIC CONTEXT 6l

pressing or too big for him to handle alone. Rajputs were also to be
seen sewing buttons on their newly made shirts while the tailor worked
on more complicated phases of his craft in order to get a rush order
finished.

No incongruity was seen in these aspects of the gaikh relationship,
and they did not embarrass those involved. The superior skill of the
artisan in his specialty was recognized, and his client got the most for
his money by arranging to let the artisan concentrate on the skilled
aspects of his work. The low-castes were still expected to respond at
once without remuneration, when not otherwise employed, to a call
for assistance in thatching a roof, mixing tobacco and molasses, killing
a goat, plastering a floor, carrying a load, or assisting in other ways
when a high-caste member needed help. The high castes' relative ritual
and social status was never ignored or compromised.

Occupational Specialties

In Sirkanda four castes of specialists are represented: Brahmins,
the high-caste priests, and three low-caste artisan groups: blacksmiths
(Lohar), barbers (Nai), and drummer-tailors (Bajgi). Each of these
groups is tied in some form of the jajmani or gaikh arrangement to
exclusively agricultural families in the village, and usually to other
specialist families as well. If there is more than one household which
follows a particular specialty in the village, each normally has its own
circle of traditional clients, just as each agricultural household has its
own fields. In each instance the household remains the important
economic unit. The proceeds of traditional specialties, like those of
agriculture, are shared within the household. The economic roles of
the occupational specialists will be considered in turn.

Brahmins

Sirkanda Brahmins are primarily agriculturists, most of whose
economic pursuits are indistinguishable from those of Rajputs. Their
income from agriculture is equivalent to that of the more prosperous
Rajputs. Priestly work is done only by Jairam, one of three brothers in
the family, and it does not interfere with his agricultural work. In his
own words and in obvious reference to his competitor, a Brahmin
from Tehri Garhwal who gets most of the high-caste business from
Sirkanda, "Some Brahmins haven't anything else to do but spend all
of their time gallivanting around doing their priestly work. I am too
busy for that."
Jairam officiates at weddings, funerals, and various ceremonies in

62 THE ECONOMIC CONTEXT

Sirkanda but usually only when his competitor is not available or needs
assistance, or when a request comes from low-caste people, whom his
competitor will not serve. He is too close to people in Sirkanda to be
held in awe or any great respect. The Tehri Brahmin from fifteen miles
away is used whenever possible and is considered to be the family priest
by most Rajputs of Sirkanda. He is credited with more priestly learning
than Jairam, and his rates are higher. He serves a restricted clientele of
high-caste people who can afford him. Either Brahmin is paid for the
work he does, for example, around Rs. 20 for a wedding or funeral,
Rs. 1 to 5 for simple ceremonies. These Brahmins are also the recipients
of occasional extra charity on certain religious days and occasions when
people want to acquire merit in this manner. The two Brahmins refer
to those families they serve regularly as their "jajman." They can de-
pend upon being called to serve any ritual needs these families may
have and to receive charity as well as payment from them. However,
neither Brahmin is the hereditary priest or purohit for Sirkanda people
in the true sense of the term.

The traditional Sirkanda Brahmin or purohit is Tula, who lives in
another village two hours' walk from Sirkanda. All Sirkanda families
as well as those in several neighboring villages are in his jajman — the
traditional, hereditary client-practitioner relationship. The office has
been in his family for many generations and continues in full force
although Tula, like his father before him, practices very little and al-
legedly knows even less of priestly duties. Some generations back his
family were highly regarded in Bhatbair as learned priests. Now they
are not so regarded and make no pretense in this direction, limiting
their priestly participation to minor roles in ceremonies, to astrology,
and so on. Nevertheless, as traditional priest or purohit in this and
surrounding villages, Tula attends and is given the purohit's charity
and payment at every marriage, funeral, or other important ceremonial
event and at every occasion calling for the granting of charity to Brah-
mins. Thus he receives many times the amounts that the officiating
Brahmins do. He is given due honor and respect as the traditional and
rightful recipient. As a result he is one of the wealthiest men in the
area, with three houses, three wives, much land, and many children,
some of whom have married into prominent families of Garhwali
Brahmin ancestry in Dehru Dun. He is also the only obese Pahari I
have ever seen and is addressed and referred to almost exclusively as
"mota Brahmin" (fat Brahmin). The jajmani system has worked to his
advantage.

The term jajman is used loosely by all three of the above-mentioned

THE ECONOMIC CONTEXT 63

Brahmins to refer to the people they regularly serve (as, "Family X is in
my jajman"). However, only Tula has the long-standing, hereditary
relationship with its system of traditional payment that is generally
associated with the term "jajman."

Artisans

All the artisan (shilpkar) castes in Sirkanda, commonly referred to
as Doms, depend upon their crafts for their living and are to some ex-
tent prevented from acquiring other means of livelihood. An artisan's
clients are referred to as his gaikh or, commonly, simply "farmer"
(kisari), terms which are roughly parallel in usage to the Brahmin's
use of "jajman." "Jajman" is also used sometimes by the artisans,
though it is not thought to be entirely appropriate. The artisan is
known as the "worker" (kam karnewala) by his clients. The client
usually pays the artisan who works for him a fixed amount of grain at
each harvest, that is, twice a year. There is not the rigidly hereditary
association between worker and client reported for the plains (W. H.
Wiser, 1936). However, no farmer will be denied service by an artisan
if he demands it. Threat of physical punishment can be used if neces-
sary, with the backing of the rest of the high-caste majority. The artisan
may use more subtle pressures, such as inferior or slow service, to en-
courage prompt and adequate payment by a delinquent client. If an
artisan does not perform satisfactorily, his clients may patronize another
craftsman in the same or a nearby village.

BLACKSMITHS

The blacksmith's main job is making and repairing iron tools,
horseshoes, bells, occasional religious images, and so on. Agricultural
implements, axes, and grass knives make up most of the blacksmith's
work. Today there are two blacksmiths, brothers, who work and live
separately in separate households. One of the two has 20 households
in his clientele, including "all the big people" — those who pay reg-
ularly and well — in Sirkanda and nearby villages. He prides himself
on having maneuvered himself into the favor of the more desirable
customers by personal charm rather than superior work. His brother
has 29 of the "lesser" households including 12 in Sirkanda and 17 in
neighboring villages. Though he recognizes the inferiority of his clien-
tele to his brother's, he philosophizes that "God will provide for all."
Some other households patronize blacksmiths near their chans and
only occasionally utilize one of the Sirkanda ones, on a job-payment
basis. Neither of the Sirkanda blacksmiths is kept very busy by his

64 THE ECONOMIC CONTEXT

work, and it provides only a marginal income. When the father of these
two men was the village smith he had a good income; he is said to
have loaned money to Rajputs, and to have left a cash nestegg when
he died. Now the business is divided in half. Moreover, the smiths claim
that payment has fallen off as people have come to use ready-made tools
from the bazaar more frequently and as households have increased in
membership or split up. Where formerly there was one household
there are now often two or three, or membership has doubled or
tripled, while the members continue to pay a total of only one house-
hold's traditional payment. Meanwhile other households have passed
out of existence or have moved to chans, so that their business is lost.
Equally important is the fact that formerly grants of grain, livestock,
or even land, as well as clothing, tobacco, tools, and so on, were given to
all the low-castes as gratuities. Now these things are rarely given. The
attitude of the higher-castes has ceased to be as paternalistic as it was
in former days — partly, according to the high-caste people, because
the low-castes do not now conform to the submissive role expected in
a paternalistic relationship.

The blacksmiths are the poorest of Sirkanda villagers and are the
only ones aside from two Bajgis who now own virtually no land, having
sold what little remained of their holdings some time ago when the
extended family divided. Payment for their traditional work is in
five-seer (ten-pound) measures of grain. At each harvest they are sup-
posed to get two to three measures from each two-oxen household they
serve, three from four-oxen households, and four measures from a six-
or eight-oxen household. This scale is not rigidly adhered to, and the
blacksmiths claim that they are consistently underpaid. They get a
total of 16 to 18 maunds of grain per year — about half their minimum
requirement for livelihood. One of the blacksmiths supplements his
income by distilling and selling home-made liquor — a brisk business
yielding 25 to 35 maunds of grain per year. The other has tried cloth-
selling and other schemes to augment his income, all unsuccessful
largely because of his illiteracy and softheartedness, which have pre-
vented him from recording and collecting debts or refusing credit. He
now makes ends meet by raising goats, from which he nets 25 maunds
of grain per year. He also does odd jobs around the village and gener-
ally remains on such good terms with all villagers that free meals and
occasional gifts come his way.

Blacksmiths live a marginal existence, their income barely exceeding
the requirements for livelihood. However, they do not fear starvation
because they know that other villagers will not allow them to go hungry
if they are really in need so long as they continue to perform their

THE ECONOMIC CONTEXT 65

functions in the village without offending their caste superiors. Such
assistance comes bit by bit as needed rather than in an advance,
security-providing amount. Opportunities exist for blacksmiths to work
where large-scale stone-quarrying operations are in progress some six
to ten miles away. There a cousin of the Sirkanda blacksmiths makes
up to Rs. 20 per day blacksmithing. When asked about this, one Sir-
kanda blacksmith commented that such work, while profitable, is un-
certain. Here he has lifetime security which he would lose if he left,
and the stone quarry might close down at any time. Moreover, he says,
"I've never worked in my life." The latter point has a large element
of truth. No one in Sirkanda leads a life as leisurely as that of the
blacksmiths. This man volunteers that if he worked harder he could
make more money right in Sirkanda — by getting work from customers
in nearby villages and doing it promptly and well. He attributes his
father's prosperity partly to his industry.

BARBER

The Sirkanda barber is a descendant of a barber brought from
the valley at the time of the founding of the village. Elsewhere in this
area the barber's functions are performed by other low-caste members,
usually Bajgis and sometimes blacksmiths. When the Sirkanda barber
left the village for three years once, his duties were taken over by one
of the blacksmiths. Whoever performs the barber's functions receives
annual payment of four seers (eight pounds) of grain at each harvest
for each bewhiskered man in each household and two seers per whis-
kerless boy in each household he serves. At this rate the Sirkanda barber
gets about 25 maunds (2,000 pounds) of grain per year for his small
family from his rather extensive practice, including all of Sirkanda
and several neighboring settlements. This meets his subsistence require-
ment for grain. In addition he receives payment for his traditional
ritual services at weddings, funerals, initial hair-cutting ceremonies,
and one annual festival day, ranging from Rs. 20 to a few annas (one
anna = % of one rupee) each. This barber owns some lands in Sir-
kanda and more land and bullocks in the village of relatives in the
valley, which give him a modest but comfortable income beyond that
of his barbering.

BAJGIS

The work of Bajgis in Sirkanda is varied. With the exception of
barbering, they perform the functions common to their caste in most
Pahari villages: drum-playing, tailoring, and basketmaking.

Of these activities, only drum-playing is done in the traditional

66 THE ECONOMIC CONTEXT

artisan-gaikh relationship. It is done for the entire village by two
households working jointly. They play drums at all ceremonies and
festivals. In return for their drumming each village household gives
the drummers 16 seers (32 pounds) of grain at each harvest and oc-
casional other gratuities in the form of grain. This income is divided
between the two households who do this work in Sirkanda. Formerly
it was one household, but the two brothers have divided their property.
Since the drumming requires a team of two, they have retained this as
a joint enterprise and they divide the proceeds. In former days Bajgi
women danced on special occasions as part of the caste's traditional
duties, but they have given up this defiling practice.

Tailoring is an important activity of the Bajgi caste — so much so
that Bajgis are commonly referred to as Darzi (tailors), whether or not
they follow this profession. In Sirkanda tailoring is done on a piece-
work basis for cash or its grain equivalent. Rates range from six rupees
for a coat down to six annas for a cap. One man does all the tailoring
in Sirkanda, and the payment is kept in his household, one of the
two drum-playing households. This arrangement was the basis for a
major but short-lived dispute when the household which shares the
proceeds of the drumming activity demanded, but did not get, half
the income from tailoring as well.

Baskets are made on a similar basis at rates ranging from three rupees
for a large grain-storage basket down to four annas for a small basket
or winnowing tray. Three Bajgi households make baskets.

In addition, one of the two non-drum-playing Bajgis makes and sells
liquor. The other, an immigrant of some twenty-five years' residence in
Sirkanda, lives entirely on his income as a practitioner of a kind of
curative magic. The two large, drum-playing households own land and
bullocks, which provide a substantial addition to their other income.
They get about 18 maunds of grain apiece annually for drumming,
about one-third of their grain requirements. They make another 12
maunds in their other specialized work, 20 maunds from milk, goats,
and so on, and 40 maunds from their fields. They are therefore reason-
ably well off. The two smaller households, who have no land and no
income from drumming, are about as poor as the blacksmiths.

Blacksmiths and Bajgis exchange traditional services in lieu of pay-
ment. The barber is, however, paid in grain by all who have it and in
cash equivalent by those (like the blacksmiths) who do not.

NONRESIDENT SPECIALISTS

A number of Pahari occupational specialists contribute to the Sir-
kanda economy, although they do not reside in the village. All of these
described below are represented within the eight-trail-mile (four-air-

THE ECONOMIC CONTEXT 67

mile) radius which is the area of face-to-face social and economic inter-
action for Sirkanda villagers.

Sarola is the name of a subcaste of Brahmins who specialize in cook-
ing for Brahmins and Rajputs on ceremonial occasions. They are
ritually superior to other Brahmins and therefore pure enough that
anyone can eat from their hands. Payment is Rs. 16 to 20 for the job
or R. 1 per maund (80 pounds) of food cooked — whichever is greater
— plus their board and shelter while the job lasts. In addition they get
a traditional payment of 2% seers of gur (sugar) and R. 1 at the time of
the ceremonial lighting of the cooking fire.

Sonars are makers of silver and gold jewelry. In these hills there are
two castes of these goldsmiths — one equivalent to Rajputs in every way,
the other a Dom caste equivalent to blacksmiths and carpenters. The
Rajput goldsmiths are called "Sonar" as an occupational term, just as
are Dom goldsmiths, but they are Rajputs and intermarry freely with
other Rajputs. Similarly Dom goldsmiths intermarry freely with black-
smiths and carpenters, with whom they are included under the broader
occupational classification, Mistri. Most of the Sonars of this area are
of the Rajput group, and they include some of the wealthiest of Paharis.
They are paid in cash or grain by the job. They have shops in towns
such as Mussoorie and in small trading centers such as Suakholi. Some
travel from village to village taking orders and doing their work as they
go.

Carpenter-masons (BaRhai) intermarry with blacksmiths and Dom
goldsmiths, but their occupational specialty is usually handed down
patrilineally. They make wooden tools and utensils and construct
houses. In Sirkanda most men make their own wooden tools, but car-
penters are called in to build houses. Carpenters live in the village
while they work, receiving their food and shelter plus Rs. 4 per day.
They may also be paid by the job, for example, Rs. 80 for sawing out
100 boards. There is occupational specialization among carpenters;
some, called Rengdlda, produce lathe-turned wooden containers and
churns which are sold by volume. There is no traditional economic
relationship of Sirkanda households with carpenters, though certain
ones are usually called in because of their skill and proximity.

Kholls are weavers of wool blankets and heavy wool cloth. Formerly
their work was more in demand than it is today because their cloth was
used for coats and they were virtually the only source for blankets and
cloaks. Now cloth from the bazaar is much used and the Kholis' main
trade is in a distinctive type of blanket used as a wrap. They are paid
a piece rate, an arm and hand length of loom width being the unit and
costing Rs. 2.50 if the customer furnishes the wool.

There are several terms for people whose profession is singing and

68 THE ECONOMIC CONTEXT

dancing. The only Paharis of this profession in Bhatbair are called
Beda. One family of this caste has long been associated with Bhatbair,
and, though they have moved from their traditional home in a village
in the heart of Bhatbair, they still live within walking distance in a
Pahari village bordering the valley. They (two men and one or two
women) visit Sirkanda and other Bhatbair villages about once a year,
staying three or four days in each village and going from house to house
entertaining with songs and dances. For this they receive 2 to 4 seers
of grain from each house. People of this caste are thought to possess
unusual spiritual powers. They were formerly the performers of the
rope-sliding feat, a spectacular form of worship which, when performed,
brought them a large lump sum of cash and goods as well as a great
deal of honor (Berreman, 1961b). Since Beda women dance and sing,
they are thought of as prostitutes. Today the Bedas are primarily ag-
riculturists and are trying to give up their hereditary occupation and
the stigma attached to it. They have ceased to give evening command
performances while in the village, apparently because these were linked
to prostitution. Instead they perform only in the daytime at each house
and they give a public performance in the evening in one place. Ac-
cording to the present practitioners of the trade they would give it up
altogether as, indeed, they intend to, if it were not for pressure from
their father to carry on in the family tradition.

Mochis or Chamars are shoemakers and removers of dead animals.
They are not indigenous to this Pahari region, although farther back
in Tehri Garhwal there are local people of this caste. Those who live
and work near Bhatbair come from the Kangra valley, a Pahari area
some 150 miles to the west. They have been in the region for 20 or 30
years or more. Before they came, villagers assert, people here went bare-
foot, as many of them do yet. They remain culturally distinct and are
readily identifiable by dress, house type, and speech, although not by
physical type. Their work is more defiling and their caste rank lower
than that of any other group in these hills. They sell shoes by the pair,
or by an arrangement in which a household gives them 5 seers of grain
per harvest in return for which they provide all the shoes needed in
that household at R. 1 per pair. In addition they are called to remove
any large domestic animal which dies. In return they keep the skin and
give the owner a pair of shoes unless the animal is a cow or an ox, in
which case the owner takes nothing. A few years ago a family of this
caste moved to the outskirts of Sirkanda but found it unprofitable and
moved on to Sahas Dhara, a larger center below.

In addition there are ten to fifteen other Garhwali service castes,
most of them never seen in Sirkanda. Some, such as rope bridge-makers,
may be called upon if a special need arises.

THE ECONOMIC CONTEXT 69

NONCASTE OCCUPATIONS

Several important occupations are not closely or rigidly bound to
caste. Farming is foremost among these, as evidenced by the fact that
currently some members of every caste in Bhatbair make a substantial
part of their living by farming.

Special occupations in this caste-neutral category are primarily those
associated with divining, curing, and dealing with the supernatural.
Whatever their caste, practitioners of these arts are paid in cash or grain
for their work. One Bajgi in Sirkanda makes his entire living from his
work as a practitioner of curing through exorcising spirits. Another
makes some income as a diviner. Specialists in the treatment of various
maladies (chickenpox, mumps, boils, snakebite) are found among all
castes. Jairam, the Sirkanda Brahmin, carries on a brisk business in
simple Vedic medical practice, that is, curing based on ancient Hindu
prescriptions. Shamans are found in several villages in and near Bhat-
bair and are frequently consulted to learn the cause of difficulties be-
setting families or individuals. In the process a god enters the shaman's
body, speaks through his mouth, and provides the desired informa-
tion. The shaman, who may be of any caste, is paid a small fee for
each consultation. Pujarls specialize in the worship of household gods,
playing drums and reciting praise of these gods to induce them to
dance in the bodies of the worshipers and frequently to speak through
them. These practitioners come from all castes, with a tendency toward
the lower castes. They too are paid by the job.

There is no Pahari Vaishya or merchant caste. Shopkeepers and
peddlers come from the plains, from Tibet and Nepal, and from all
Pahari castes. Of three Pahari tradesmen in Bhatbair, two are Rajputs
and one is a Brahmin. As mentioned previously, a Sirkanda blacksmith
tried his hand at selling cloth throughout the region for several years,
and people of all castes may become temporary shopkeepers at the
annual fair.

SERVANTS

Some high-caste agriculturists of Sirkanda employ servants to help
tend their fields and livestock. Those who do the hiring are people
who have more land, or land in more scattered locations, than the men
in the family can handle. In a small family with considerable land a
good servant pays his way many times over.

Servants live with their employers and are paid a small amount
monthly (Rs. 10 to 30). Because of their close association with their
employers they are chosen from high castes to avoid problems of rit-
ual pollution. Thus, of the three permanent servants in Sirkanda, two

JO THE ECONOMIC CONTEXT

are Rajputs and one a Brahmin — all Paharis, two from Nepal and one
from Tehri-Garhwal. Servants are given responsibilities equivalent to
those of a family member, often being left for long periods to look after
a chan, and frequently one of its owner's wives and her children as well.
A lifelong servant is generally given land and a house as a reward for
faithful service. The oldest resident of Sirkanda is the widow of a man
whose father came to Sirkanda as a servant and remained there after
having thus acquired lands.

Temporary servants in the person of carpenter-masons engaged in
house construction, or indebted high-caste men (often relatives of their
employers) working as laborers, are generally to be found in the village
during the winter housebuilding season. As many as eight of these
were resident in the village at once during 1957-1958.

Occupational Variation

Occupational variation within caste is found throughout India
(cf. J. M. Mahar, 1958; Sharma, 1961). Paharis are tolerant and flexible
in this respect. As mentioned above, in addition to the traditionally
caste-neutral occupation of agriculture, there are a variety of magical-
medical practitioners of all castes, Rajput goldsmiths, merchants of
blacksmith, Rajput, and Brahmin castes, and high-caste agricultural
servants. Barber functions are shared among several low-caste groups.
In general, low castes share the artisan occupations, and not uncom-
monly a member of one low caste will be found doing the work of
another, especially where economy demands it and population does
not provide the appropriate caste specialist. Thus either blacksmiths
or drummers may serve as barbers, either may serve as tailors, and it
is not unheard-of for a blacksmith to play drums. Such occupational
flexibility is characteristic of caste and economic organization in the
hills. Just as status differences among Dom groups are somewhat am-
biguous, so are their occupational specializations. For both purposes,
Doms are often treated by the high castes as an undifferentiated group.
Even high castes perform the work of artisans on occasion. They
exhibit many of the skills of artisans while assisting them in their work
in order to hurry things along or to save payment. Since there was no
carpenter in Sirkanda, Rajputs and Brahmins did their own routine
carpentry, calling in an outside specialist only for major building jobs
or tasks requiring unusual skill. In addition high-caste people are not
entirely barred from becoming professional artisans. An example that
has already been mentioned is that of the Rajput goldsmiths. An ex-
ample on the individual level is that of a skilled carpenter of Bhatbair
who is a Rajput and is accepted as such in every respect, although he

THE ECONOMIC CONTEXT 71

makes his living in precisely the fashion of low-caste carpenters and in
their company. Another highly respected Rajput of Sirkanda worked
for a Vaishya shopkeeper as a servant in the trading village of Nagal.

Outside Employment

Residents of Sirkanda and Bhatbair have not been attracted in
large numbers to outside employment. Most of those who have, have
been gone only a short time. A frequent pattern of adolescent male
behavior is running away from the village. Boys often go to the plains
to wander around, see the sights, engage in sporadic employment, and
in a few months return to remain in the village. At least this has been
the pattern until the present. Perhaps with improved preparation
through the village school and increased job opportunities outside,
the high return rate will decrease.

Six recent cases of the kind concerned boys 13 to 15 years of age.
Two, a Brahmin and a Rajput, ran away together to Delhi, where they
worked for three months in a tea stall washing dishes and delivering
tea at Rs. 15 per month plus board and room. They came back after
they had had enough of city life and had been unable to find more in-
spiring or remunerative labor. As one said, "You could spend all your
life washing glasses in a tea stall and never get anywhere." A black-
smith boy had run away to Dehra Dun twice, staying three or four
months each time and working once in a tea stall and once in an em-
broidery shop, at Rs. 15 per month. One boy was taken by an employer
from Dehra Dun to Naini Tal, 100 miles to the southeast, where he
worked as a kitchen boy for seven months before coming home. Another
worked in a dairy south of Delhi for five months and then returned.
Two boys left during the winter of 1957-1958 and had not yet been
heard from seven months later. More than once people from neighbor-
ing villages came through Sirkanda in quest of runaway sons. At least
five older men in Sirkanda recounted similar youthful adventures.
That not all return is indicated by the fact that three or four men were
listed in the genealogical survey as having disappeared from the village
in their youth and having never been heard from again.

Needless to say, such short-lived escapades result in little economic
gain for those involved, and by the time worried relatives have gone
to look for and bring back the truants, as is frequently done, the finan-
cial loss to the household is appreciable.

Not all outside employment is of this youthful and transitory nature.
One young Sirkanda Rajput left the village as a result of a family dis-
pute (there is nearly always some such motivating incident), and got a

72 THE ECONOMIC CONTEXT

job in a textile mill in an industrial suburb of Delhi. He has been at
it six years, has been promoted, and has no intention of leaving the
job. He occasionally visits Sirkanda, but does not keep in close touch
with his family and sends little or nothing in the way of financial aid — ■
partly because he spends most of what he makes (Rs. 120 per month)
and partly because the family dispute has alienated him from his
household. In the minds of villagers, he is the single example of a
local boy who has made good on the outside. Another young man left
the village early in the period of my residence to join the millworker.
He got a job at Rs. 70 per month at which he was still working a year
later, and he gave no indications of quitting. He had hoped to send
money home but had as yet failed to make enough to enable him to do
so. His departure — in secrecy — worked a real hardship on his family,
as he was one of only two adult males available to work the land upon
which ten people depended for livelihood.

One Rajput had left the village several years previously to escape
some outside creditors, and ended up running a tea stall for over two
years in Kalsi, a town in the Dehra Dun valley 20 miles to the west.
He returned at the urging of relatives, stayed three years, and again
left for an unknown destination. Two Sirkanda Rajput men had been
in the army during World War II, though neither left India. Both
returned to the village after demobilization. One, however, neglected
to procure a discharge and was later picked up and forced to return
to the army where he is today, having served about ten years in two
periods. Both these men sent money to their families sufficient to con-
tribute substantially to the support of their dependents, though in both
cases they and others felt that economic as well as emotional ties
would have been better served by their presence in the village. Several
other men of Bhatbair, but outside Sirkanda, served in World War II,
including men of Rajput, blacksmith, carpenter, and shoemaker castes.
A carpenter and a shoemaker who work for Sirkanda families had each
served several years in both the African and Pacific theaters of opera-
tions, one with American and the other with British troops. All these
men were able to make sufficient income so that their families did not
suffer. One deceased Sirkanda Rajput served in World War I, as did a
Rajput shopkeeper from a nearby village.

Dehra Dun is not an industrial city and does not offer many em-
ployment opportunities for people who are uneducated, as are most
Paharis of this region. This situation probably contributes to the low
rate of outside employment. Two educated young men of villages on
the edge of Bhatbair near Dehra Dun (a Brahmin and a Rajput whose
wives are from Sirkanda) have gotten jobs in that city — one with a

THE ECONOMIC CONTEXT 73

small government ordnance factory employing primarily skilled per-
sonnel, and the other with the office staff of a limestone quarrying com-
pany which operates near Bhatbair. Both make about Rs. 100 per
month and live at home, so that they contribute a substantial amount
to supplement the agricultural income of their families.

The total number of Paharis in the ordnance factory is only ten or
twelve, and the others are all from a neighboring region in Tehri
Garhwal. The husband of another Sirkanda woman who lives in the
Bhatbair village nearest Dehra Dun has become wealthy as a result
of having been given the contract to maintain the Rajpur canal, which
supplies much of the water used in Dehra Dun.

One Sirkanda Rajput moved to Sahas Dhara, a small trading center
and site of a popular religious shrine five miles below Sirkanda, where
he made his living for many years as a shopkeeper and where he is
now retired. A Sirkanda-born blacksmith makes his living in the same
place, largely from the work he gets from commercial limestone quar-
ries in the area.

These make up the sum total of Sirkanda men, plus a sampling of
others from Bhatbair, who have engaged in outside employment. The
importance of such employment to village economy is small, though
possibly it presages things to come. To the present, no Sirkanda vil-
lager has relatives resident in Dehra Dun. No Sirkanda couple
or larger family unit has migrated from the village to a larger center.
One of the men who was in the army took his wife with him for a
period of about six months and then brought her back to Sirkanda.
The man who recently went to work in the cloth mill sent without
success for his wife, but would not come back after her. Rarely is out-
side employment considered more than a temporary activity. Rarely
is the type of employment available spoken of by others with approval
or admiration. Nearly every person who has left the Bhatbair area to
work has done so surreptitiously and without advance warning. Other
aspects and implications of these defections, and of outside experiences
and contacts in general, will be discussed in chapter 9.

Illegal Economic Activities

Liquor is manufactured independently by two men in Sirkanda,
a Bajgi and a blacksmith. In other villages other castes engage in this
activity, but most are nonagricultural castes, probably because they
have more time to devote to such side-line enterprises and more need
for the additional income. Since to sell untaxed liquor is illegal, there
is an element of risk, and therefore of secrecy. Usually the distilling

74 THE ECONOMIC CONTEXT

apparatus is so designed as to be easily and quickly disassembled, and
always it is well concealed. Any particular distillery may close down sud-
denly and indefinitely if the owner fears detection by outside authorities,
whose approach is usually, of course, relayed well in advance. Detection
can lead to fine and imprisonment or demands for large bribes. The liq-
uor is made by soaking any of several grains in sugar and water for about
a week, and then distilling it. The product has a high alcohol content.
It is drunk for effect rather than taste and is in constant demand at
R. 1 per bottle. No celebration is complete without liquor. Liquor
manufacture and selling is a profitable business, especially at festival
times, when the demand generally exceeds the supply.

Marijuana is also produced in the village. Its addicts mix it with
locally grown tobacco and smoke it to produce a euphoric effect. Its
sale is illegal, and the two or three Sirkanda men who use it prepare
their own or buy it illegally from neighboring villagers.

Another much publicized and criticized illegal activity of hillmen is
the traffic in women. Although its prevalence has been exaggerated and
it is evidently on the decrease, it has at times been an important ac-
tivity among certain people of all castes. The procedure is simple:
Pahari women of any caste but especially of the low castes, and almost
invariably married women, who are unhappy with the hard and simple
life of the hills and who have heard and have been convinced of the
luxuries and pleasures of plains life, are helped by a Pahari, often of
their acquaintance, to travel to Dehra Dun or some other center. There
they are turned over as wife to a contact man. The woman may live
as the wife or concubine of this person permanently or temporarily,
she may be turned over to someone else seeking a wife or concubine,
or she may be utilized in prostitution. The Pahari's profit in the trans-
action is in the initial payment made by the man who received the
woman to the one who delivered her. One transaction may net the
"abductor" several hundred or a thousand rupees, which is clear profit,
as the wronged husband is unable to collect compensation unless he
can find the person who was responsible, effect his arrest, and press
court action. Occasionally a husband or other family member may
cooperate in the transaction.

Sisters and wives of people now living in Sirkanda have been sub-
ject to this experience. One Bajgi woman currently resident in Sirkanda
had been abducted and sold in her youth, but she returned after a few
years. More commonly such women are not heard from again. Tula,
the hereditary Brahmin of the village, was one of a trio, together with
a Rajput and a Bajgi, who carried on this traffic some 25 years ago.
They were apprehended, convicted, and imprisoned. Only Tula re-

THE ECONOMIC CONTEXT 75

turned, one man having died in prison and the other having escaped
and disappeared. Tula's priestly standing seems to have been unaf-
fected by the incident. The men of one whole Rajput household in the
village were at one time actively engaged in this trade; several were
eventually caught, and others left the region to escape prosecution.
Several Sirkanda men have occasionally profited from participation in
such transactions, as recently as the past ten years. Most of these were
not members of a professional gang, as were Tula and the household
mentioned above, but rather were men who took advantage of an op-
portunity when it arose. Several factional disputes in Sirkanda can
be traced to incidents of this nature.

Throughout North India the hills are thought of as a source of pros-
titutes, sought after for their beauty and lack of inhibition. Paharis and
plains people alike assert the truth of this statement, at least as ap-
plied to the states bordering the hills. This is not to say that there are
prostitutes in abundance in the hill areas, but rather that many pros-
titutes come originally from the area. Apparently the process described
above accounts for the transportation of a good number of women to
the plains. The custom of "bride-price" marriage in the hills facilitates
such transactions, as it is customary to take money for women in per-
fectly valid marriage arrangements. It is not uncommon for men from
the plains to come to the hills looking for brides from poor families
who may not be too inquisitive or particular about their credentials.
Some of these men are genuine seekers of wives. At least two low-caste
Sirkanda women are today in distant places as the wives of such men,
and are apparently happy. Some men have less noble motives. On the
other hand prostitution is not despised by the lowest castes, from which
these girls (at least those who are heard from again) usually come. It
is an occupation providing a good income by Pahari standards and it
has a certain glamour as compared with village life, because of its
association with an urban setting. While not marriageable, these girls
are not outcastes, and they visit their homes occasionally. A famous
prostitute of Dehra Dun is the sister-in-law of two Sirkanda Bajgis, and
prostitutes of Mussoorie and Dehra Dun who come from villages in
the vicinity of Bhatbair are known and patronized, at a discount, by
Sirkanda men.

Division of Labor by Sex

Paharis work hard to make their living and in return they are
able to lead a fairly secure existence. Men, women, and children
above the age of about eight years all make their contribution to

76 THE ECONOMIC CONTEXT

family economy. The men are the heads of the families and make the
decisions in economic matters. Men do all the plowing and other
cultivation of fields that is done with animals. They also generally
sow the fields. Men market the produce and do the family's trading.
They deal with all outsiders and outside agencies, though women,
too, participate in buying from peddlers. Men see to the construction
and repair of houses and fields; they make and maintain household
equipment.

Women cook and care for children. They collect most of the fodder
for animals and, with the children, tend the animals most of the time.
They take care of the manure, dry it, store it, and are primarily respon-
sible for seeing that the fields are fertilized with it. They carry most
of the water for the family and animals. They winnow the grain
after it is threshed and prepare it for storage. When the time comes,
they prepare the food for cooking. They do unskilled labor in assist-
ing the men in constructing houses, terracing fields, and clearing land.

Men, women, and children all take part in weeding crops, harvest-
ing, and threshing, though men are often engaged in plowing by the
time the harvesting and threshing are well under way, and so they
participate relatively little in them. Everyone collects firewood. All
spin wool into yarn and knit yarn into garments in their free mo-
ments.

In the artisan and priestly castes men do the specialized work and
women perform nearly the same tasks as do women in exclusively agri-
cultural families. In the tailoring and basketmaking castes they may
occasionally help their menfolk do some of the stitching or weaving.

One of the most striking features of Pahari life is how hard the
women, in particular, work. Even a person who has been accustomed
to plains villages, where women, especially those of the lower castes,
are far from idle, cannot help being struck by this feature of Pahari
life. Women are almost constantly at work carrying headloads of
fodder, firewood, manure, water, grain, flour, and, in building season,
rocks and clay — often herding cattle or goats at the same time. They
are frequently gone from the village most of the day collecting grass,
leaves, or firewood and tending animals in the forest. On moonlit
nights at harvest time they often come out to work after the evening
meal, from ten to one at night. They are rarely idle. Men too work
hard, but the winter and rainy seasons are periods of relative inactivity,
and in all seasons except plowing and planting they spend some time
sitting, talking, and smoking. Children from an early age help their
mothers, engage in herding goats or cattle, look after younger children,

THE ECONOMIC CONTEXT 77

bring water, fodder, and firewood, and in the specialist castes learn
the traditional trade.

Cooperative Labor

Most work in Sirkanda is done on a family basis or with hired
assistance. There is, however, a tradition of cooperative group labor
within the community which cuts across lines of kinship and caste.
Such voluntary help (called madat dena, to give help) is undertaken
when someone announces that he needs help on a task. On the ap-
pointed day all those who are free to do so assemble at the place of
work. The only people likely not to participate or to be asked to
participate are those who have had a dispute with the person to be
helped or with others closely identified with him. The principal oc-
casion for cooperative labor is the transplanting of rice. Occasionally
weeding of rice or of other rainy-season crops is done cooperatively.
Special songs, vigorous, rhythmic, and heroic in theme, are sung by
those who participate in such cooperative agricultural work. When-
ever there is a great deal of carrying to be done over considerable
distances, cooperative labor may be employed, as when wood, slate,
or sheet iron is to be brought for house construction. Roof beams are
lifted into place on a new house in this cooperative manner. After the
work is completed the "host" serves some refreshment, but no formal
obligation to repay is incurred.

Village property, such as the water supply and the trails, are re-
paired cooperatively when their condition demands. Since Inde-
pendence there has been a tendency to feel that the government should
look after the trails. Governmental attempts at enlisting the voluntary
cooperative labor of villagers have met with little or no success, the
common complaint being that the government servants who make
the requests are paid employees, so they should do the work rather
than asking villagers to do it. "What right has a salaried official to ask
for voluntary work from others?" "Pay us sixty rupees a month and
we'll do voluntary labor willingly." This attitude relates to a larger
area of attitudes toward the government to be discussed in chapter 9.

Inheritance

Property is normally passed down within the patrilocal extended
family in this area. When a man dies his property goes to his sons
as a group. If his wife is living it stays in her custody until her

y8 THE ECONOMIC CONTEXT

death or remarriage, at which time it goes to his sons. If a man leaves
no sons and designates no son surrogate, his nearest male relatives
(in order: brothers, father's brothers, father's brother's sons) take their
place in the line of inheritance. This is at variance with Hindu law,
according to Joshi, but it is in accordance with Khasa customary
law:

The distinct feature of Khasa agnatic succession is that the inheritance
does not go to an individual, but to a group, which may consist of the male
descendants of the propositus himself or of those of one of his ascendants.
There is no rule of the nearer agnate excluding the more remote such as is
found in Hindu law. The sons of a deceased brother take the share which
their father would have received if he had been alive when the inheritance
opens. . . . The distinction between undivided brothers or between full
blood and half blood . . . has no place in Khasa customary law. . . . The
rules of inheritance are based on the theory that agnates alone are entitled
to the estate left by a deceased person, and that the ancestral land held by
a person who has no male descendants reverts to the immediate parent stock
and is distributable accordingly. (Joshi, 1929, pp. 296 f.)

If there is no son, an adopted son, a daughter's son, or a son-in-law
can be designated to substitute for a son, both in inheritance of property
and in performance of the father's funeral rites. In each of these cases
the heir must come to live in the house of the one from whom he will
inherit (Berreman, 1962c). A son-in-law who assumes this function
lives with his wife's father and is known as a "house son-in-law." Ac-
tually the inheritance in such a case remains in the hands of the woman
and ultimately passes on to her sons (or lacking sons, to her lineage
mates) unless a bequest has been made to her husband. However, the
husband has use of the land during his lifetime. This status is em-
barrassing for the son-in-law in a patrilocal society, but it is sufficiently
advantageous to all concerned that it occurs quite often. Since sib
membership is patrilineal, this arrangement often results in a transfer
of property to another sib. Houses and land in Sirkanda have been
transferred in this manner to sibs other than those in which they were
originally held.7 If no lineage claimants exist, property reverts to the
village, not the caste. A woman's jewelry is treated as her husband's
property for purposes of inheritance.

Incorporeal property such as jajman or gaikh clients and the he-
reditary title of mukhia — a village leader — is inherited similarly.
Where, as with a hereditary title, there is a single, discrete entity to
be passed along, it goes to the eldest son. Where no son is left, in-
corporeal property, like other property, may go to a near male agnate
or may be assigned to the husband of a daughter who lives in her

THE ECONOMIC CONTEXT 79

father's house, and thence to the daughter's son, or it may go directly
to a daughter's son who comes to live in his mother's father's house.
Thus the Sirkanda barber inherited his clients from his mother's
father. As with corporeal property, the arrangement often results in
transfer of the property to another sib.

In this patrilocal society the levirate is practiced: a wife is inherited
by her husband's eldest surviving brother or equivalent relative. At
least, first claim to her is so inherited and the decision as to whether
she will become his real wife, or the wife of another brother, or not,
is left to him. If she wishes to marry a nonfamily member (as she
often does) she is generally allowed to do so but, unless she runs away
(as she also often does) or receives special permission, payment has
to be made by the new husband to the family of the deceased husband.
When a woman leaves or is left by her husband, her children generally
remain with their father's family. In this way property is kept in the
patrilocal extended family.

5 THE RELIGIOUS CONTEXT

THE SUPERNATURAL

Most Paharis are Hindus, as evidenced by their own profession of
faith and by application of any realistic definition of that term to
observation of the behavior they exhibit and the beliefs they profess
relating to the supernatural world.1 D. N. Majumdar (1944, pp. 139 ff.)
supports this statement in his discussion of the people of Jaunsar-
Bawar: "The Khasas are Hindus; their customary rites in temples,
the manner and mode of offering sacrifices . . . periodical festivals
... all indicate their Hindu origin . . ."

They are not orthodox Hindus. That is, they are not highly
Sanskritized or Brahmanical in that they do not adhere closely to
written prescriptions and proscriptions of post-Vedic Hinduism (cf.
Srinivas, 1952, p. 30; 1956). This unorthodoxy is evidently confusing
to many who are familiar with Khasas. Majumdar, a few pages be-
yond the above-quoted passage wherein he verifies the Hinduism of
the Khasas, says:

While the Khasas claim to be Hindus . . . their social life as well as their
beliefs and practices connected with their religion do not identify them with
the Hindus of the plains. They re-marry their widows, practise levirate,
sororate and polyandry, recognize divorce as legal, while inter-marriage be-
tween the various Khasa groups is not tabooed and children born of such
marriages do not suffer any social stigma. While they worship Hindu gods
and goddesses, they have a partiality for ancestor spirits, queer and fan-
tastic demons and gods and for the worship of stones, weapons, dyed rags and

THE SUPERNATURAL Ol

symbols. The sun, the moon and the constellations are their gods. (Majumdar,
1944. P- 15°)

Paharis themselves are well aware of these deviations from high-caste
orthodoxy of the plains, and in fact they often actively try to emulate
plains rituals in order to raise their status in the eyes of other Hindus.
That their unorthodoxy makes them any the less Hindu is contra-
dicted by a comparison of their own traditions, practices, and beliefs
with the range of equivalent traits among village Hindus elsewhere
in India (cf. Cohn, 1954, pp. 174 ff.; Dube, 1955, pp. 88 ff.; Lewis,
1958, pp. 197 ff., 249 ff.; P. M. Mahar, 1957, i960; Marriott, 1955a;
Opler, 1958; Planalp, 1956; Srinivas, 1952).

Hinduism in Sirkanda shares virtually all of its forms with that
in the rest of Bhatbair, most of its forms with that in Tehri Garhwal,
many with that in other Himalayan hill areas, some with that in
North India, and fewer with all-India Hinduism.

In this context it is useful to utilize the concept of "spread" of
Hinduism and the terms "local Hinduism," "regional Hinduism,"
and "national Hinduism" (Srinivas, 1952, p. 213). "Spread" refers to
the area (horizontal spread) or social groupings in an area (vertical
spread) within which traits associated with Hinduism are distributed.
". . . As the area of spread decreases, the number of ritual or cultural
forms shared in common increases, as the area increases, the common
forms decrease" (Srinivas, 1952, pp. 213 f.).

Morgan (1953, pp. 3 ff.) notes that Hinduism is "ethnic rather than
creedal," pointing out the difficulty in distinguishing its essential fea-
tures in view of its tremendous diversity. Dube indicates something of
this diversity and the complexity of beliefs which make up village
Hinduism:

A text-book knowledge of the religious lore of India, and an acquaintance
with her ancient classics and their modern expositions will hardly give us
a true picture of the actual religious beliefs, thoughts, feelings and practices
of the people now living in the countryside. A classification of their religious
beliefs and rituals is not an easy task. Folklore and myths, religious teachings
of saint-poets, and contacts with persons having knowledge of scriptures and
popular religious books have all influenced their religious ideology, and
consequently their religion is a mixture of animism, animatism and polytheism,
with the occasional appearance of monotheism also. To these must be added
a living faith in spirits, ghosts, demons, witches and magic. The complex of
all these diverse factors constitutes the picture of the supernatural world as
it is understood by the people in the countryside. Tenets of classical Hinduism
having an all-India spread are mingled with the regional religious beliefs
and forms of worship current among the Hindus of the [particular area].

82 THE SUPERNATURAL

Several cults and worships of a purely local nature add furdier to the
complexity of the beliefs and ritual system of the community. A wide
variety of cults is observed by the family, some by the village as a whole;
and still others by individual caste groups. (Dube, 1955, p. 88)

Despite its diversity, there is a basic unity to Hinduism by which it
may be recognized. Morgan (1953, pp. 6 f.) identifies this as lying in
"common scripture, common deities, common ideals, common beliefs,
and common practices." Srinivas makes a similar statement and then
characterizes the elements of Hinduism according to their spread. He
describes all-India Hinduism as "chiefly Sanskritic" in contrast to the
more limited spread of most non-Sanskritic elements in regional and
local Hinduism. By "Sanskritic," Srinivas means those elements of all-
India Hinduism which are recorded in the classic religious literature
and are often called Brahmanical or post-Vedic Hinduism by other
writers. They are frequently identified as the "great tradition" of Hin-
duism, while local, regional, and even universal Hindu beliefs and
practices which are not included in the "literate religious tradition"
are identified as the "little tradition" (Redfield, 1955; Marriott, 1955a).

It is apparent to any student of village religious practices and beliefs
that there is a considerable body of non-Sanskritic elements of all-
India spread (cf. Berreman, 1961b). Care must be taken not to confuse
"little tradition" with "local spread." Many elements of "little tradi-
tion" as conventionally defined are regional, national, or greater in
extent.

In this chapter religion will be presented in terms of beliefs and
practices regarding the supernatural as they were observed and re-
ported in and around Sirkanda. Part of the purpose is to illustrate one
variety of the genus village Hinduism. Another is to point out specific
examples of the great and little traditions as they manifest themselves
in this village and to throw some light on the relevance and limitations
of these concepts in the village context. In so doing, the evidence for
and the nature of Sanskritization, or change from adherence to the
elements of the little tradition to recognition and practice of the
elements of the literary tradition of Hinduism will be mentioned. The
more immediate purpose, as in the preceding chapters, will be to
furnish prefatory material for the discussion of social organization to
follow, and to illustrate the functions of kinship, caste, and community
ties in the religious sphere of life.

Religious Beliefs

The supernatural is almost as pervasive in the minds of Sirkanda
villagers as is the natural, though to an observer it may be less readily

THE SUPERNATURAL 83

apparent. Difficulty of any kind — crop failure, ailing animals, economic
reversal, mysterious loss of property, persistent family troubles, disease,
sterility, stillbirth, hysteria, death — is attributed ultimately to fate and
more immediately to the machination of one or another of a host of
supernatural beings. A sizable amount of time, effort, and money is
invested in activities designed to influence these beings to tread lightly
on the people of Sirkanda. Most of these activities are carried on at the
joint-family level, some at the community level, and, in this area, prac-
tically none at the caste level.

The supernatural beings who affect humans range from capricious
sprites, malevolent ghosts, and ancestral spirits to household, village,
and regional gods. On another level is a general belief in inevitable fate
(mukadar, bhdg) controlled by a remote, impersonal, and ultimately
supreme deity, Bhagwdn or Nardyan. This deity is neither personified
nor worshiped. To it is attributed almost any event or circumstance
worthy of notice and beyond immediate human control. Such com-
ments as "It is God's will," "God only knows," and "It is in the hands
of God," are frequently made by villagers. There is no effective means
by which to deal with Bhagwan or his manifestation in fate, although
villagers may occasionally direct an appeal for mercy or help to him in
a general way, as "God help me." Nevertheless there is lively interest
and activity in influencing events by propitiating the many specific
deities and other supernatural beings which are thought to control or
influence daily life. When the inconsistency which an outsider may see,
between belief in inevitable fate and simultaneous efforts to influence
or control events, is pointed out to villagers their answers are vague
but not defensive because their belief is not threatened. Such incon-
sistencies are considered irrelevant. One informant affirmed that fate is
inevitable, that Bhagwan controls fate and cannot be influenced, but
that worship of specific deities is still necessary "because gods are
closely associated with fate." And there the matter rests. Evidently it
is through gods that fate is accomplished.

The effects of predestined and unalterable fate are everywhere ap-
parent. To demonstrate its importance a villager pointed out that a
potter makes many pots that look alike, and yet each has a different
subsequent history. That is fate. Similarly, men are born essentially
alike but no two lead the same life. The classic example was that of
myself. "There are millions of Americans and none of them have ever
heard of this small and distant village. There are thousands of villages
in India to which no American has ever come. Now you have come to
live here. Is not that evidence enough of fate?" Death, disease, dis-
ability, poverty, wealth, beauty, travel, marriage — all are attributed to
fate or God's will. Weather too is predetermined. Yet these are pre-

84 THE SUPERNATURAL

cisely the things which villagers seek to control through their worship
of various gods. Late in the dry season I asked a villager when it would
rain. His characteristic reply was: "I do not sit with Bhagwan, nor am
I his brother." But before the week was out he was one of many who
participated in a ceremony calculated to please specific deities who
control the weather and thereby to bring rain. Of course, since every-
thing is controlled by fate, my informant could retreat to the argument
that it was the villagers' fate to carry out such worship with the pre-
ordained results.

Closely allied to belief in fate is the belief in reincarnation. It is
believed that the present condition of any being is largely determined
by his deeds in previous lives. "As you sow, so will you reap." Present
deeds influence fate after death and in future lives, but they have no
effect on the present life. An informant commented on the point,

Fate in life is determined before birth and nothing can alter it. If a man
has given much gold in charity in a previous life he will in the next life
be a raja, if he has given grain he will be a money-lender, if he has taught
others he will be a great leader like Nehru, if he has killed a female or
thirsty wild animal he will be a leper or blind, if he was a slanderer he
will be an idiot, if he did not give ritu dan [seasonal charity — man's obliga-
tion to have sexual relations with his wife after her menses each month]
he will starve sexually.

Disappointments are nearly always rationalized in terms of fate as
determined by misdeeds in previous lives. A Brahmin from the village
nearest to Sirkanda had such an explanation for the nagging, unat-
tractive wife with whom he had been saddled. When his friends asked
him why he was so unfortunate he reflected at length and finally re-
ported that in a previous life he had been a crow and his wife a camel
on whose back he had sat, pecking and worrying her. Now it was her
turn to get revenge. His friends urged him to leave her, but he replied
that he could not as this was his fate and it could not be escaped by
running away.

Low-caste status is always attributed to misdeeds in previous lives,
while people who commit misdeeds in the present are thought to be
ensuring punishment in future lives. The idea of life after death (and
particularly the time between death and reincarnation) is a vague and
confused one to villagers. Many have no systematic conception of what
occurs, and others hold firmly to mutually contradictory views. But all
agree that acts in this life determine the nature of the next life. Punish-
ment for misdeeds may include a long period after death in which
the soul wanders in suffering as a ghost before finding a body in

THE SUPERNATURAL 85

which to be reborn. Then it may be reborn as a crawling worm or
insect, or in an undesirable human condition. A good life is rewarded
by rapid transit of the soul to another body so that the cycle of 184
births through which it must pass may be quickly accomplished and
the ultimate goal attained — heaven or, according to learned Brahmins,
unity with the infinite. The new body will be, in such cases, one that
is desirable: if animal, a bird that can fly and enjoy itself; if human,
of a high caste, in a pleasant locale with access to good food. Light
skin color, beauty, health, wealth, many sons, and happiness are also
considered rewards.

Station of birth is, therefore, no accident. A low-caste informant who
had probably pondered the matter more than most Sirkanda people,
explained:

America and England are pleasant places where everyone is wealthy and
comfortable. To deserve such good fortune the citizens of these countries must
have been very pious Hindus in their previous lives. On the other hand,
in those countries people do many evil things in their lifetimes. They eat
meat, especially beef, they eat eggs, they kill people in great wars. It is a
very sinful life. After death they must be punished by being reborn in filth,
poverty and sorrow — namely, as low-caste people in India.

He felt that this must certainly be the case, though he admitted the
circularity of the sequence and had not solved the problem of how one
could escape the circle.

The same informant considered birth as a Pahari to be punishment,
partly evident in the hard work, poor food, and primitive living condi-
tions which Paharis must endure. Another reason was also given: since
nothing in life is chance — all is part of the larger design called fate —
to be born in circumstances where sin punishable by low birth is in-
evitable is in itself a punishment. Thus, to be born a Pahari, where the
sin of taking the life of animals in sacrifice is part of regular religious
practice demanded by local deities, is not an accident; it is an indirect
kind of punishment in that it will lead to further punishment in future
lives. God or fate uses it as an excuse for further punishment. The
informant referred to this as an "excuse of fate." Perhaps it is considera-
tion of such factors that accounts for the widespread Pahari joke,
"When we Paharis die we are reborn on the plains as donkeys." This
logic might also explain the peculiar fate of Americans as deduced by
the informant. Awareness of punishment in future lives does not lead
Paharis to give up animal sacrifice, nor would it prevent them from
behaving like other wicked Americans were they to be reborn in Amer-
ica. It is all fate and cannot be altered. Paharis accept their fate phil-

86 THE SUPERNATURAL

osophically. A blacksmith who had learned from a Brahmin who read
his horoscope, that he had been a raja in his previous life and was
destined to be a merchant in the next shrugged when asked what he
thought of the prospect, and said simply, "I am adaptable."

After a birth a Brahmin is called to read the horoscope of the new-
born. This tells with perfect accuracy all that has befallen an individual
in past lives and all that will befall him in the present and future lives.
Everything pertaining to the individual's past and future is there to
be read. Where inaccuracies occur the Brahmin is either incompetent
or misinformed. The latter is most often the explanation, as accuracy
depends upon precise reporting of time of birth. As the villagers
point out, in a community with no clock this is a difficult feat.

Deeds in life which lead to punishment in future lives include fail-
ing to give charity, physically hurting others, exploiting others or
taking their property, and taking to evil ways — robbery, killing, steal-
ing, cheating, lying, covetousness. Breaking caste rules was not in-
cluded by informants. Upon inquiry, low-caste informants held that
this is a social, not a spiritual, matter, and is to be enforced by man,
not God. This is in contrast to traditional Hindu doctrines of caste
duty and destiny, doctrines which high-caste people profess. All groups
relegated rules regarding incest, exogamy, endogamy, and other sexual
behavior to the secular realm.

The things Pahari men aspire for in life regardless of caste include:
sufficient money to satisfy wants and a surplus to lend out on interest,
good land, good crops, good cattle, hard-working and obedient sons,
and profit in whatever transactions are entered into. Stories and anec-
dotes illustrating the importance of good deeds to reap desired re-
wards in future lives and, conversely, the dire consequences of evil-
doing in this life, are many, vivid, and well-known in Sirkanda.

It is within this general context of belief in an impersonal, all-
powerful supernatural force called Bhagwan, evidenced in unalterable
fate, that other beliefs regarding the supernatural are held in Sirkanda.
This is in conformance with the great tradition of Hinduism.

In everyday life in Sirkanda people think about and deal with re-
ligion in terms of immediate problems of their welfare and that of
their families. The supernatural agents who are closely involved in
these matters are personal, personified beings whose behavior influ-
ences and can be influenced by people. The effective social unit for
dealing with most of these beings is the household. As in economic
activity, so in religious worship, the extended family is the most sig-
nificant element of social organization.

Village religious life in Sirkanda, as elsewhere in India, is primarily

THE SUPERNATURAL 87

concerned with the maintenance of proper relations with supernatural
beings who have power over the members of the family and of the
village. Their displeasure is easily aroused by neglect and is quickly
evident in the several kinds of difficulties, notably illness, which beset
their negligent worshipers. Fortunately, they can usually be placated.
Their form, origin, and affinities are of less significance to villagers
than are their effects and the means to placate them. Some of these
beings are gods or goddesses which affect the entire village, or which
affect only particular households. Others are ancestral spirits. Other
categories of powerful supernaturals are ghosts and sprites. Some of
the gods can be traced to the great gods of Hinduism, but others cannot.
Some spirits are ghosts of dead relatives or of known types of individuals,
while others cannot be traced to specific people or to people at all.
These facts are of interest, and are often known, but are of little im-
mediate relevance to the villager. All such beings are active forces
which must be recognized and dealt with in specified and often similar
ways, or else their subjects will suffer well-known and dreaded conse-
quences.

Before describing the patterns of worship and the exorcism of these
beings, it will be well to comment briefly on the phenomena usually
described as supernatural possession. Possession is a common occurrence
in Sirkanda in the cause, diagnosis, and alleviation of difficulties of
many kinds. But it is not a uniform process. Its nature varies with the
type of supernatural being involved.

A shaman voluntarily induces his personal familiar spirit to possess
him and speak through him to his clients. He does this by chanting
certain phrases, playing a drum, and performing other acts pleasing to
the spirit. When the spirit leaves it does so without ill effect. Its good
will is maintained by the shaman's worship.

Anyone may be possessed by a household deity attracted by the
drumming of pujaris during worship of a household god. Such posses-
sion results in the deity's dancing in the worshiper's body, and some-
times in his speaking through the worshiper. No one knows who will
be possessed, but possession is expected to result from the activities of
the pujaris on these occasions.

At village-wide worship, individuals often become possessed by
village gods, just as they do by household gods in household worship.
Generally the person whom a particular god will possess is known as
the traditional vehicle through which that god dances and, more rarely,
talks.

In all the above cases the god is attracted by drumming, the rhythm
of which induces him to engage in the pleasurable activity of dancing.

88 THE SUPERNATURAL

One god, Memendia, possesses people unexpectedly and then calls for
music to dance by. A god's presence is indicated by the behavior of the
one he possesses — especially by trembling, rolling of the eyes or fixed
staring, insensitivity to touch or pain, incomprehensible speech, and
generally uncontrolled behavior. Once a god is present, he is honored
with incense, gestures of devotion, and so on.

Such possession is described by a phrase meaning that a god has
"come to the head" of his devotee. It is never harmful to the one
possessed, and the deity leaves voluntarily. In fact, the possessed person
is said to be immune to pain or lasting harm inflicted while he or she
is possessed. Gods do harm individuals, but not as a result of possession.
A god who is angry will possess his victims only briefly in order to tell
them what they must do to appease his anger and to relieve themselves
of his punishment.

Harmful possession is of a different order, described as "adherence"
of a ghost. It occurs unexpectedly. Usually it can be terminated only
by strenuous exorcism by a specialist. Ghosts are not worshiped. Only
by forcing the ghost to leave can the harmful effects of such possession
be alleviated. The fact of ghost possession is indicated by the harmful
effects of the possession — great pain or other inner torment, illness,
misfortune, barrenness, stillbirths, mental derangement, physical im-
pairment, or even death. Ghost possession usually follows severe stress
on the individual. Calamity, frightening illness, the death of a friend
or relative, interpersonal strife, and physical exhaustion immediately
preceded cases of such possession in Sirkanda. Possession is a satisfactory
explanation or excuse for almost any behavior. The one possessed is an
object of solicitude rather than condemnation. Possession therefore
appears to be a psychological mechanism used to alleviate stress, to ex-
plain otherwise incomprehensible and often taboo behavior. While
women are more often subject to this kind of possession than men, it is
by no means limited to women.

Some spirits can harm people without possessing them, merely by
"striking" them. In the case of one type of sprite, such harm is not ma-
licious nor intended, but once it has occurred it is irreparable. Its only
symptom is the harm done, usually a sudden physical, sensory, or mental
incapacity.

Pattern of Worship

The details of worship of various supernatural beings in Sirkanda
vary. There is, however, a basic pattern underlying most worship in
the village. In order to make subsequent descriptions clear and varia-

THE SUPERNATURAL 89

tions more obvious, the general pattern of worship will be presented.
In outline it bears many similarities to worship by plains villagers.

Most supernatural beings make their presence felt by imposing diffi-
culties or troubles upon people — usually disease or death to people or
animals, and sometimes other troubles such as hysteria, faithless
spouses, sterility, poor crops, dry cows, financial loss, or mysterious
disappearance of belongings. "Above all a prevailing health anxiety is
suggested by the data regarding ghost and spirit possession" (Opler,
1958, p. 566). In such cases the householder usually repairs at once to
his favorite shaman (baki), or sometimes to a less prominent practi-
tioner, to find out what is causing the trouble (cf. Berreman, 1961a).
The only supernaturals regularly worshiped without illness or other
difficulty as a signal that worship is demanded, and therefore without
the advice of a shaman or other practitioner, are village gods (to be
defined below). Such worship follows the pattern to be described here
except that consultation with the shaman is omitted. It is relatively
infrequent as compared to the worship recommended by shamans,
which occurs somewhere in the village every few weeks.

A shaman is a man who may be of any caste and who is devoted to a
particular deity for whom he acts as medium in the diagnosis of diffi-
culties. It is from this practice that the shaman makes his living. He
generally holds regular sessions which his clients attend. First he con-
ducts a short pujd (ceremony), and to the beat of a small drum he sings
mantras (prayers or incantations) in honor of the god to whom he is
personally devoted. Gradually, as the god takes possession of him,
the shaman becomes impervious to pain, often demonstrating this by
touching red-hot metal or by some similar action. The god when in
complete charge speaks and acts in the body of the shaman. The god
then singles out the various waiting clients one at a time and tells each
what troubles he has had and what the cause is, that is, what super-
natural being has been tormenting him, and what should be done to
alleviate the trouble. The god may also identify human thieves or
other culprits and point out objects that have deleterious magical
effects. If the victim is not satisfied with the diagnosis he will merely
say "The god knows better than I," and go elsewhere for advice until
he finds a shaman who seems to have a more accurate god.

The treatment recommended by a shaman is almost invariably per-
formance of a puja in honor of the offending supernatural being, or
exorcism if it is a ghost. In cases of theft the shaman will merely identify
the guilty party. In cases of magical affliction he may identify the of-
fending object and recommend its removal. Treatment may also be a
pilgrimage or other specific action designed to please the god. In some

THE SUPERNATURAL

very difficult afflictions, such as apparently incurable insanity, an im-
possible recommendation may be made, such as sacrifice of a cow
which, as Hindus, these people cannot carry out. The necessary puja
is usually a stereotyped one for the particular deity to be honored.
Sometimes the shaman may have specific recommendations as to how,
when, or where the worship should be conducted. The shaman's main
functions are to identify the cause of the trouble and to specify what
action must be taken to alleviate it or to contact the being responsible
in order to hear his demands. From then on the family of the victim
takes steps to carry out the shaman's recommendations, for which
special practitioners are required. In this respect the Pahari shaman
differs from the shaman of the plains reported by Opler (1958) and by
Planalp (1956). The plains shaman, with the help of his personal
spirit, induces a god or spirit which has entered the body of its victim
to make known its demands. Among Paharis the shaman does not take
part in exorcism or in ceremonies in which the god speaks through
its victim and is ultimately appeased. The Pahari shaman is primarily
a diagnostician who is able to call upon his personal god at will, be-
come possessed by him, and then diagnose the difficulties of his clients
through the wisdom of the god. The Pahari sequence is virtually identi-
cal with the shamanism of a Mysore village reported by Harper (1957,
pp. 268 ff.).

Some exorcists of ghosts (not bakis) function in a fashion similar to
that of the plains shamans in that they perform both the diagnosis and
exorcism, sometimes with the help of their own god. However, such
practitioners are less frequently consulted and their advice is less val-
ued. As often as not a shaman is consulted first, and he may direct
his client to one of these practitioners to perform the exorcism. Ma-
levolent ghosts are quite distinct from gods and ancestors and will be
described later. The discussion which follows applies to gods and an-
cestors only.

Sometimes worship of a god or ancestor is held without first con-
sulting a shaman. A less powerful diviner may be consulted because
he is more easily available or has a reputation for accuracy. Occasion-
ally no specialist is consulted at all, when the victims feel certain they
know which god is responsible for their difficulty, either because of
experience with similar past affliction or because of an unfulfilled vow.
Worship may also be held to fulfill a vow before trouble has come to
compel its performance.

Vows are often made to gods and are a frequent antecedent to wor-
ship. If a puja or other propitiatory act which has been recommended
by a shaman cannot be performed at once for reasons of economy or

THE SUPERNATURAL 91

conflicting obligations, a vow may be taken to perform it within a
certain period. The vow has the same efficacy as performance of the
puja itself provided that it is fulfilled. Trouble of any kind may be met
by such a vow even, in some cases, without a shaman's advice. Whether
or not a shaman is consulted, the vow itself is made without the services
of a specialist.

In order to carry out the pujas, most of which are undertaken on the
advice of a shaman, one or more often two practitioners who specialize
in performing worship for particular gods or classes of gods must be
called. These pujaris (or a pujari and his assistant) arrange and per-
form the puja at the request of the family who is to sponsor it.2 The
purpose of the puja is to ensure that the god is pleased or at least has
a chance to possess one of his victims, dance in the victim's body, and
through that person make known any further demands. It is a family
member, never a specialist, who becomes possessed during the per-
formance. These specialists may be of any caste, but usually they come
from the low castes. They play percussion accompaniment for the
dancing that is a part of every such performance and that brings the
god to possess his worshipers. In fact, particular types of pujaris are
known by the term for the kind of drum they play, which in turn is
determined by the god they worship. For some village gods, Brahmin
priests act as pujaris in that they perform the worship, and low-caste
people merely play the drums for dancing.

The ceremonial proceedings in such worship are in three major
parts: the dance, the puja, and the offering. The entire sequence
usually takes place before the shrine of the deity to whom it is dedi-
cated, which is either in the house or somewhere in or around the
village.

Preparing the shrine is a simple process because, though it is ne-
glected in the months or years intervening between occasions of wor-
ship, it is so simple as to require little arranging. The shrine generally
consists of one to four simple iron tridents (tirsul) of varying shapes,
about eight inches in height, made by the local blacksmith. These are
placed in a niche in the wall of an inner room if the shrine is a house-
hold one, or at the base of a large stone or in a stone enclosure if it is
outdoors. During the ceremonies the shrine is illuminated by a small
oil lamp, and often a container of grain and a few coins are placed by
it. Sometimes a small bag filled with rice and coins is hung near the
shrine. Such accouterments are in the nature of offerings to the deity.

The first stage of worship is the dance (kalratra, literally "black
night" or "night for Kali"). It is intended to attract the gods or an-
cestors who like to dance and who can do so in the bodies of humans.

g2 THE SUPERNATURAL

This in itself is pleasing to the gods. It also induces them to speak and
air their complaints and demands if they wish to do so. Usually dancing
occurs during an evening and again the following day. The puja to
one god requires seventeen dances over a period of nine days.

A kalratra or dance begins late in the evening of the appointed day,
when householders, onlookers, and specialists assemble by the shrine.
The pujaris begin to play their instruments, and one of them chants or
sings sacred mantras (prayers or incantations) honoring the deity.
During this time the god in whose honor the worship is being held,
and often other gods and ancestral spirits as well, come and gradually
possess people. The room is filled with onlookers, smoke, and heat, and
it reverberates with the compelling rhythm of the drums. Gradually
one or more members of the household and sometimes others as well
begin to move in time to the drums, then to jerk, shout, and finally to
dance, first gently and then wildly as they become possessed by a god.
This period is called "awakening." The possessed person is honored
with incense and religious gestures and is fed boiled rice because he or
she is a manifestation of the god at that time. The dancing continues,
with occasional breaks for smoking and talking, far into the night.
Some people dance who are not possessed, or, if so, only temporarily.
The same god may possess several people in sequence in one evening.

Village gods are worshiped outdoors, usually at the dancing ground
adjacent to the Pandavas' temple (to be discussed below), but some-
times near Devi's temple. At that time any village god or gods may
possess individual villagers. Household gods do not possess people on
these occasions. In contrast to the nature of possession by household
gods, there is one particular individual whom each village god usually
possesses. Most often village gods utilize the state of possession merely
to dance, but they may speak if they wish to.

After the god has danced his fill, usually on the last night of the
kalratra, he may leave or may choose to speak. If the latter, the dancer
becomes immobile, speaks unintelligibly ("It sounds like English,"
asserted one informant), and then begins to answer questions put to
him. This stage is simply called "questioning." The god tells the cause
of his anger and the action necessary by the victimized household to
appease that anger. Usually he demands a goat or other sacrifice, but
sometimes he asks for a more elaborate puja. The god also gives advice,
solves dilemmas, and issues warnings and ultimatums. When he is
through, he so states, the drums of the pujaris beat briefly, the god de-
parts, and the session is over. The possessed person shows no after-
effects and does not remember the period of possession, although I

THE SUPERNATURAL 93

noted that some showed a remarkable ability to recall the number and
subject matter of photographs taken during possession.

Shortly after the god has left, the pujari and the victimized members
perform a short ceremony or puja in the god's honor, and then make
an offering. The offering to be made, usually a young male goat, is
placed before the shrine and rice is thrown on its back by the household
members as the pujari chants mantras. When the goat shakes itself,
this is taken as evidence that the god has accepted the offering. There-
upon the animal is taken outside and beheaded — usually but not al-
ways by some low-caste person, as high-caste people consider this some-
what defiling and prefer not to do it. Then a foot and the head of the
animal and some delicacies such as bread and sweet rice are placed
before the shrine as offerings for the god. Afterwards the pujari takes
these items along with his fee of Rs. 1.25. The rest of the animal is
divided among the participating householders and observers.

The efficacy of worshiping gods to combat difficulties is not doubted
by most villagers, and many cases of miraculous results are cited in
support of their belief. One teen-age boy professed disbelief until his
father, sick with pneumonia, began to breathe more easily as soon as
the practitioner began his mantras. In the boy's words, "I never believed
until I saw it work on my father. A dying man was saved by mantras
to the god. That is proof." Another informant replied to my question
as to the effect of such a puja in his household: "Of course it worked.
We satisfied the god's every demand, didn't we?" On the occasion of an
elaborate and difficult puja, an informant commented: "Why not go
to all this trouble? It gets results. Gods are like lawyers; the more you
give them, the more they will do in your behalf."

The intended result of these observances is to alleviate difficulties
attributed to deities, or in the case of worship of some village gods, to
ensure their continued good will. An important latent function is that
of providing entertainment, relaxation, and social intercourse for the
spectators. At the same time they reinforce ties of kinship and village
solidarity. There are few recreational activities for these hard-working,
isolated people, and the occurrence of a puja of some sort every few
weeks is a welcome break in routine. It is also one of the few occasions
upon which there is widespread involvement in a common activity.
Every such event involves participation of a group of kinfolk or of the
entire village, and it plays to a full house of all castes including a large
proportion of women and children who have less access to varied ex-
perience and entertainment than do men. The behavior of the audi-
ences at such exercises is very similar to that of cinema audiences in the

g4 THE SUPERNATURAL

big towns of the valley. They smoke, talk, wander in and out, and
generally enjoy the show.

More than once informants implied that a particular puja was held
by someone (not of their own household) largely as an excuse to kill
a goat and have a feast. Generally when a puja is to be held people lay
in a supply of liquor, the perennial accompaniment of meat at Sir-
kanda "parties." A kalratra for village gods, as distinguished from
household gods, is largely a social and recreational occasion for par-
ticipants and spectators alike, and only occasionally does a god seize
the opportunity to talk or make demands of villagers. More often the
gods merely enjoy this opportunity to dance, and many of the dancers
participate without benefit of supernatural possession.

The nature of worship of the gods honored in most of these pujas is
definitely placative. It consists of efforts to appease angry deities, to
cater to their cravings for worship and sacrifice.

The truth is that popular religion in these hills is a worship of fear. . . .
When famine and pestilence stalks abroad, village temples are crowded and
promises of oblations are made; if the evil be averted these promises are ful-
filled, if not the deity is frequently abused and his shrine is neglected. The
efforts of all are directed to appease the malevolence of these spirits who
are supposed to lie in wait to take advantage of any error willingly or un-
willingly committed. . . . (Atkinson, 1884a, p. 839)

Villagers themselves sometimes describe their worship as motivated
by fear. One man asserted, "If a god can do no harm, it has no power
and need not be worshiped." Another commented, "All gods are bad,
but Nor Singh and Agorndth are the most terrible of all." Most gods
are feared for the power they have to punish men who incur their
wrath. The exceptions are certain village gods, to be discussed below.
Even they have the power to punish, but they are primarily benevolent.
The prevalence of disease, untimely death, and unpredictable mis-
fortune stands as a constant reminder of this power. Fear of gods is
continually reinforced by attribution to them, especially by shamans
and diviners, of almost every misfortune that occurs. If a god whose
worship depends on fear fails to make his power felt by punishing
people, he is rarely or never worshiped and is soon forgotten. Anxiety
is maintained by the fact that misfortune often strikes where it is least
expected. However, though fear is an important motive, it is not help-
less fear. Gods do not punish randomly or capriciously. They punish
when punishment is due. The offense is generally unintended and is
often unknown or unrecognized, but a shaman can usually detect it.
Most often the offense has been that of neglecting the god — failing to

THE SUPERNATURAL 95

worship adequately. The offense may have been one against the god
or against one of his devotees; the essential thing is that the god has
interpreted it as an action or oversight deserving of punishment. The
victim is not helpless. In fact, he has at his disposal extremely effective
means of appeasement, which are sometimes difficult and often ex-
pensive, but are rarely beyond the realm of possibility. The efficacy of
these means is attested by the many spectacular successes people recall,
the routine reliance upon them, and the readily available alternative
explanations for apparent failures.

Therefore, Sirkanda villagers do not lead a life dominated by helpless
terror of the gods they worship. Observation of worship reveals a range
in individual attitudes from relative indifference to enthusiasm, from
apprehension to confidence. It seems that gods are treated much as are
powerful people. They are feared insofar as it is well known that their
anger can lead to serious consequences. On the other hand, it is known
that the means to prevent or assuage their anger are at hand and are
often easily put into play, as is not often the case in dealing with
powerful secular forces. The prevailing attitude is one of respectful
awareness of what are conceived to be the realities of life in a world
where ultimate powers are in the hands of divine beings, whose anger
is easily and often unwittingly aroused, but whose demands are com-
municated to their subjects and can usually be met. Worship is the
means by which these dangerous beings are controlled.

Gods

By far the most active class of supernatural beings in Sirkanda are
the gods and goddesses (devta, devl) or, as they are sometimes referred
to in English, the godlings. Gods indigenous to Sirkanda are referred
to in the village as ghar ka devta (household gods) or kul devta (family
gods) and are thereby distinguished from gods of other villages and
other regions. Any god to which local people are devoted is isht devta.
These terms are used loosely and often interchangeably in the village.
However, an analytic distinction can be drawn to categorize deities in
Sirkanda as household gods and village gods. Household gods will here
be defined as those gods worshiped consistently by the members of a
particular household as a group, usually within the house, where the
shrines are kept, and not worshiped jointly with the members of other
households, nor with the aid of Brahmin priests. Devotion to these
gods is passed down in the lineage. Village gods are those gods wor-
shiped jointly by all or nearly all villagers at some central shrine
called a temple (mandir) in or near the village. The worship is usually

g6 THE SUPERNATURAL

under the supervision of a local Brahmin priest. One worships these
particular village gods because he is a Sirkanda villager. Other dif-
ferences between these two categories of gods will be mentioned in the
course of discussion of each class. This dichotomy is not commonly, if
at all, noticed by villagers, though it is evident to the observer of re-
ligious behavior in the village. Neither is it a rigid distinction. Some
household gods have gradually shifted to become essentially village
gods. Some village gods have been adopted as household gods by one or
two families. Occasionally particular households worship village gods
individually at their village shrines. Certain village gods are not wor-
shiped by every household. Some household gods in Sirkanda are vil-
lage gods in other areas, and vice versa. A distinction is, however, use-
ful for purposes of presentation. In the beginning it is usually the god
himself, speaking through a shaman or some other person, who directs
the type of shrine and worship to be given him.

Household Gods

Six major household gods are worshiped in Sirkanda homes: Nar
Singh, Manglla, Gauril, Nag Raja, Agorndth, and Dhagbairu.3 Each
of these gods is worshiped by certain households and not by others.
Thus, of the 45 Sirkanda households, 41, none of whom are Brahmins
or Bajgis, worship Nar Singh; 28, all Rajputs, worship Manglia; 13 (1
Brahmin, 7 Rajputs, 1 blacksmith, and 4 Bajgis) worship Gauril; 13,
all Rajputs, worship Nag Raja; 3, all Bajgis, worship Agornath; 2
Rajput households worship Dhagbairu. These gods appear in villages
throughout the area. Not every village contains households worshiping
each of them, and some have households worshiping other gods, but
the pattern is the same and none of these deities is unique to a par-
ticular village. Every household which worships one of these gods,
except some of those worshiping Nar Singh and Nag Raja, has a
shrine, called "god's place," dedicated to that god, and usually the
shrine is inside the house. In those houses worshiping more than one
god, which includes all but five of the 45 in Sirkanda, separate shrines
are kept for each god. When joint families divide, each new household
unit sets up its own shrines to the family gods.

Household gods are the most acutely feared gods worshiped in
Sirkanda. They are worshiped as long as, and to the extent that, they
demonstrate their power and interest by tormenting household mem-
bers and placing demands upon them. Each is usually worshiped in
ceremonies directed exclusively to him and calculated to alleviate
trouble attributed to him by a shaman. The god has generally been
displeased by neglect of worship or unfulfilled vows. However, trouble

THE SUPERNATURAL 97

may also stem from a plea to the god by one of his devotees who wishes
to inflict harm. A god can be induced, by worship, sacrifice, or vows,
to attack one's enemy if the enemy is either a nonworshiper or a
less faithful worshiper of that god. The victim learns the cause of his
trouble by consulting a shaman. To alleviate the attack, the victim
must worship the god. If he is not already a devotee of the god, he
must worship that god in some house where there is a shrine — and
generally no other household wants to become involved with the god's
wrath — or else he must set up a shrine of his own, in his own house.
Then he becomes the god's devotee and must continue to worship him
because the god thereafter frequents his house.

Three recent cases were reported in Sirkanda, in each of which a
wife found that her husband was having illicit sexual relations with
another woman. Thereupon the aggrieved wife verbally abused the
other woman in public. In response to this abuse the other woman set
her household god's wrath upon the wife. A puja to that god therefore
had to be performed in the house of the wife, and her husband thereby
acquired a new household god. In another case a policeman was set
upon by the god of a household whose members he had apprehended
for selling diluted milk. He had to worship their god, thereby acquiring
him as a household god, and to reimburse them. These examples, in
addition to their relevance to an understanding of village religion,
provide an interesting indication of attitudes on sex and interpersonal
relations (to be discussed in chapter 5).

The most common source of a new god in a household is a bride,
who usually comes from another village. Her household god frequently
goes or is sent with the girl to her new home and protects her in-
terests there. Such a god is known as mathzva devta, as distinguished
from the ghar ka devta that already frequents the household. If the
bride is mistreated in her new home or if her family is slighted at
the wedding or in subsequent situations, the god will attack the family
of the groom. This may happen at any time after marriage. Then it
is necessary for the groom's family to placate the god by worship. If
the god is one not already worshiped in the house, a new shrine must
be set up for him and thereafter he is a household god, although at
kalratra he will usually possess only the bride and will voice his de-
mands through her. Later he may possess her daughters-in-law or other
female relatives.

A household god may also be acquired if a shaman diagnoses some
misfortune as the work of a god who has not previously been wor-
shiped or even known in the household he has attacked, and who is
not acting on behalf or at the bidding of a worshiper. The god simply

98 THE SUPERNATURAL

demands attention and utilizes this means to acquire it. In any event,
the result is the same in that the god must be worshiped to be ap-
peased. He then acquires a shrine, bringing occasional trouble and
demanding occasional worship.

Therefore, although most of the household gods currently wor-
shiped in Sirkanda have been in the families of their worshipers for
several generations, the acquisition of new gods is not an unusual
event. Several informants remembered when their families began to
worship current household gods. Likewise, as interest in old gods is
lost, they lose significance and are finally forgotten. When new houses
are built or families separate, the members will not bother to devote
shrines to ineffectual gods, while in the old family home those shrines
may be ignored. Some Sirkanda villagers remember that their im-
mediate ancestors worshiped none or only some of the present gods,
but most are unable to specify which ones were worshiped previously.

A special household ceremony is performed to protect a new house
and its occupants from alien gods and spirits. It is similar in purpose
to the village-protection rite to be mentioned below, but it protects
the household group rather than the village. It is called literally, "house
pot." In the ceremony an all-night puja is performed in which a pot
containing certain sacred items is sealed into the house wall, where it
remains to protect the household from disease, accident, violent death,
financial reversal, and crop failure.

Other ceremonies associated with the house and household members
are those performed by carpenters when they place doorframes and
the ornamental archways in a new house, and those performed by
Brahmins to purify a house after pollution by violent death or by
occupation by low-caste people, anthropologists, and the like. All assure
the future well-being of those who occupy the dwelling, and inhibit
the depredations of alien supernaturals. The only occasion upon which
Brahmin priests regularly participate in worship of household gods
is during the marriage ceremony, when the good will of the gods is
secured by invoking a brief blessing.

Village Gods

There are seven shrines in Sirkanda devoted to what are here
termed village gods. All are in the immediate vicinity of Sirkanda, the
most distant being that to Matri a. half-mile away on the crest of the
bill which dominates the village (map 3).

All village gods are worshiped at village temples (mandir) rather
than in the home, and usually a Brahmin conducts the worship rather
than a low-caste pujari. Village gods are distinguished from household

THE SUPERNATURAL 99

gods primarily by this greater honor and by the fact that their relevance
and worship is not household-specific. Some may have begun as house-
hold gods, but they have not remained as such. All village gods can be
and occasionally are worshiped by all or nearly all villagers. A few
households, lineages, or sibs are partial to certain village gods and
not to others, but there is not the pattern of household devotion or
non-devotion found with household gods. Several temples are identi-
fied with particular lineages whose members are temple keepers, main-
taining the temple, supervising worship there, and receiving offerings
on behalf of the god, but worship at a village temple is not private or
limited to particular households.

Worship of village gods does not follow a uniform pattern to the
extent found among household gods, though the same elements of
worship occur — kalratra, puja, sacrifice. The worship of village gods
tends not to be as specific or mutually exclusive as that of household
gods. Often several village gods are worshiped together in a single
ceremony. Each village god possesses some villager at ceremonies in
his honor and often at other village-wide ceremonies. Most village gods
possess the same individual each time they appear. When that individ-
ual dies they usually shift to another member of the same lineage, sib,
or clan, but sometimes they shift even to another caste. The god's
possession of an individual is thought to be auspicious for the whole
village, not merely or even especially for that individual's lineage or
household as in the case of household gods. Village gods speak to their
worshipers less frequently than household gods and torment them less,
usually being content to dance occasionally in their bodies.

Village gods can inflict troubles on their worshipers, but they are
less cantankerous and hence less feared than household gods. Worship
of them does not depend upon fear of the consequences of failure to
worship to the extent that is true of household gods. While most village
gods are sometimes worshiped to alleviate specific difficulties which
according to a shaman's diagnosis they have imposed or can counter-
act, they are often worshiped simply to maintain the status quo of
village life. Thus, they figure in periodic worship and that performed
in stereotyped situations, as well as in worship indicated by a shaman's
diagnosis. Finally, they are more permanent than household gods in
the village pantheon. Household gods come and go; village gods rarely
do so, partly because they are not so dependent on a shaman's advice
for their continued relevance to village life, and partly because they
enjoy village-wide interest and support which is less fickle and less
subject to alteration by happenstance and petty disagreements.

As with worship of other supernatural beings, worship of village gods

100 THE SUPERNATURAL

is influenced by shamans because they can attribute difficulty to these
gods. However, most troubles are attributed to household gods — per-
haps because most difficulties strike household units — and so shamans
are less vital in trends of worship of village gods.

Among village gods there are differences of form and function that
are worth pointing out and using as classificatory criteria.4 The classifi-
cation used here is arbitrary, its recommendation being that it permits
orderly presentation and comparison. Two village gods, Memendla
and Bhartwali, are comparable to household gods in several respects,
and so they can be grouped to form a class of "household-like" village
gods, on the borderline between these two categories, just as Nar Singh
and Nag Raja are household gods bearing similarities to village gods.
These two village gods are worshiped to avoid or counteract difficulties
which they may inflict on the village. Like household gods they are
respected to the extent that they are thought capable of inflicting pun-
ishment. Bhartwali is said by some to be Bharat, described in the
Ramayana as the brother of Ram, who is an incarnation of Vishnu.
The temple keeper for Bhartwali is the local Brahmin priest. Memen-
dia, like several village gods, has no specific temple keeper.

A similar god is Raghunath, the only god to whom a real temple
building is dedicated in the village. He is worshiped by all village brides
and grooms, at the time of their wedding, under the supervision of a
Brahmin priest and with the aid of the Rajput temple keeper whose
family built the temple three or four generations ago. The god is also
worshiped by the entire village at the crop ceremony and at almost any
ceremony honoring village gods. Like the paramount gods described
below, he is likely to dance in the body of a particular person — in his
case a Rajput woman — at village-wide kalratras. In addition, Raghu-
nath inflicts punishment on particular households. Two Rajput house-
holds worship him at the temple in a fashion and under circumstances
similar to those under which a household god is worshiped. He is the
most dangerous village god and is among the most powerful. It appears
that he is becoming one of the paramount gods of the village. A few
villagers identify him with Shiva, but in the region he is usually iden-
tified as the grandfather of Ram, an incarnation of Vishnu. Another of
Vishnu's incarnations (Parasii Ram) is said to be separately depicted
in the Raghunath temple. Bhairu, a god associated with Shiva, is also
said to be represented in this temple.

One village god, Lhesania, is specific to an age group, children. He is
associated with the coming of spring and is honored at that time by
children. He is not feared because he does not inflict punishment.
Two village gods are primarily associated with agriculture, crops,

THE SUPERNATURAL 101

and weather. These are Matri, evidently a form of mother goddess, and
Bhula, god of the soil. They are worshiped to ensure success in agricul-
ture, often in ceremonies including offerings to both of them as well
as to Devi or other village gods.

Two village gods, Devi and the Pandavas, are paramount in Sir-
kanda, in that they are distinctly more highly regarded by most villagers
than are other gods. They are worshiped on a wide variety of occasions
for many purposes, and their worship is of more general interest than
that of other gods. Together they receive village-wide worship at most
important ceremonies. They are appealed to in times of village-wide
trouble and are honored to ensure success or good fortune in the village.
They are thought of primarily as protectors who deserve honor rather
than tormentors who must be placated. These gods are thought to take
a special interest in Sirkanda and its people, and to have had a par-
ticularly intimate association with it in the past. Consequently,
the villagers feel closely identified with them. Part of being a Sir-
kanda villager, as distinguished from being a member of another vil-
lage, is this identification which cuts across all other divisions within
the village. At the same time, these gods' relationships to individuals
and families in the village are usually not as specific and personal as
those of household gods. They are, to a greater extent than other gods,
identified with the welfare of the entire village. In some contexts they
are even identified with broader entities such as the residents of Bhat-
bair, or Paharis, or even mankind. They are more closely identifiable
with "all-India Hinduism" than are most gods of this region.

Devi is a local version of the Hindu mother goddess, Durga. A temple
to her, 15 miles to the northeast, is the most important temple in west-
ern Tehri Garhwal. She is honored in a small temple in Sirkanda as a
sort of village patroness. She dances in the body of a village woman at
nearly all occasions of worship of village gods and assumes special
significance in ceremonies having to do with the crops and weather.
Such ceremonies are carried out at her temple. She therefore forms
something of a link between "agricultural" and "paramount" village
gods. She traditionally possesses the barber's wife at public ceremonies.
The barber's family now worships her as a household god, in their
home.

Sltala Devi, the smallpox goddess of the plains, is known to Sir-
kanda villagers, but no shrine or worship in her honor is found in the
area, perhaps because smallpox is rare and of little interest to people
in these hills. Apparently she is assimilated with Devi in the minds of
villagers.

The most honored deities or complex of deities in Sirkanda, and as

102 THE SUPERNATURAL

important as Devi in Bhatbair and the surrounding Pahari area, are
the five Pandava brothers and their common wife Draupadi (locally
Dropt'i), whose story is told in the Hindu religious epic Mahabharata
where they are described as having lived in these very hills. Every
village of the area has its shrine to the Pandavas. Adjacent to this shrine
is the village's ceremonial center, a small open area called maDan
(in Hindi, maidan, field) but frequently described by village folk
etymology as mdndn from the verb "to please," "because that is where
we please the gods." Here most village dances and public ceremonies
are held and here all village gods take the opportunity to dance in the
bodies of their worshipers, but the Pandavas are the gods most directly
honored at the madan. The Pandavas' puja is the most important vil-
lage ceremony. It comprises part of almost all village religious activities,
including the all-important village protection and boundary rite known
as Mundklle. This ceremony is devoted to worship of the Pandavas
and a number of other village gods, prominent among whom is Devi.
Its immediate aim is to protect the village as an entity by excluding
malevolent ghosts and alien gods from its boundaries. (For a more de-
tailed description see Appendix IIIB.) It is usually financed by village
subscription, though in some circumstances a well-to-do family may
sponsor it. In any case participation and benefit is village-wide. Al-
though most village gods play a part in the ceremony, it is considered
by villagers to be primarily for the Pandavas. In it we find evidence of
the interrelation of village gods and of their distinctness from house-
hold gods. Household gods do not appear during this ceremony. In
face, they do not appear at all during worship of village gods, although
the reverse occasionally happens. Not infrequently one household god
appears during the worship of another, just as village gods appear at
one another's pujas.

A god of importance comparable to Devi and the Pandavas in this
area, but not in Sirkanda village, is Mahdsii. This god, often identified
as Mahddev or Shiva, but distinct from Shiva in the minds of most
villagers, is the paramount god in some neighboring villages (cf. Sak-
sena, 1955, pp. 40 ff.). Mahadev, whether equated to Mahasu or not, is
consistently identified with Shiva, and like Mahasu is a major god in
the area. He is worshiped in a number of ceremonies, the most elaborate
of which is the spectacular rope-sliding feat peculiar to the Himalayas
and known as beddRat (cf. Berreman, 1961b). When this is performed
in the region, Sirkanda villagers attend and seek the blessing of the
god (and his wife Pdrbatl), even though they do not worship him.

It is important to note that there is wide variation among villagers

THE SUPERNATURAL IO3

in interpretation, practice, and reported practice in all ceremonies,
the more so the more complex they are. The ceremonial sequence,
number of sacrifices, and alleged objectives are not consistent. This
variation was especially evident in the Mundkile ceremony, where
even the participants differed in their explanations of what was going
on (cf. Marriott, 1955a).

Pilgrimage

Pilgrimages are frequently undertaken for reasons similar to those
motivating worship of village or household gods. A person suffers from
some trouble or disease such as, in one recent case, weeping ulcers. A
shaman is consulted and advises the sufferer to worship at a particular
holy place, in this case Kedarnath. The person vows to do so and sac-
rifices a goat to his god, whereupon he recovers. At the first opportu-
nity he undertakes the journey to fulfill the vow. People also go merely
to see the temple or to give charity there in order to attain credit to-
ward the next life. A number of middle-aged high-caste men of Sir-
kanda have been on pilgrimages to Hardwar, Kedarnath, Badrinath,
and a few to distant Gaya. At least one woman and one low-caste man
have been to Badrinath. Many tales of the wonders of these holy places
and the miracles which regularly occur at them are current in the vil-
lage as the result of reports brought back by pilgrims or traveling
ascetics and priests. The fewer villagers who have seen a place, the more
numerous and amazing are the stories. Many villagers — about twenty-
five men — have been to Hardwar to bathe in the Ganges or attend
funeral rites, and there is relatively little lore about it, but there are
many stories of the wonders of remote Kedarnath.

Equally current, among low-caste people, are the stories of discrim-
ination, disappointment, and denial of access to these places experi-
enced by their caste-fellows who have made the trip, despite official
denial of such practices. The one village blacksmith who had gone to
Kedarnath nonplussed the priests by his presence (fellow pilgrims had
complained) and was not allowed access to the temple. Eventually he
was shown the temple padlock and told to worship that.

Closer and less prominent temples such as that to Devi at Kuddu
Kal and several in nearby sections of the valley are occasionally visited
in order to fulfill vows or contribute charity. Caste discrimination at
these, as in fact at the Raghunath temple in Sirkanda, is as strict as at
the large and famous shrines. Post-Independence laws and pronounce-
ments regarding nondiscrimination in access to temples have been un-
enforced and totally ignored in this part of India.

104

THE SUPERNATURAL

Gods of the Great Tradition

Concepts and deities contained in the great religious literature of
India are known to Sirkanda villagers as a result of their own tra-
ditions and of outside contacts. Traveling priests and ascetics occasion-
ally stop in the village and recite religious works and stories. Residents
of the village occasionally attend religious celebrations such as Ram
Lila, the annual dramatization of the religious epic, Ramayana, which
is held in towns and villages of the valley. They also attend fairs,
moving pictures, and ceremonies held in the larger centers. Much of
the knowledge so acquired does not readily penetrate into the daily
thought and action of these people because of the imperfect manner
in which it is communicated and the alien context in which it is ac-
quired. Among Paharis, plains (desi) beliefs and practices are often
considered sophisticated and even worthy of emulation, but, as they
are observed in casual contacts with plains people, they usually seem
alien and inapplicable to the Pahari context. More potent sources of
such knowledge are the local schoolteacher, villagers who have gone
to school outside the village, and especially practicing Pahari Brahmins,
all of whom communicate plains Hinduism in a Pahari idiom.

There are devout and learned Brahmins in Tehri Garhwal, among
them Har Nam, who counts most high-caste Sirkanda households
among his clients. Such Brahmins read standard Sanskritic Hindu
literature and utilize the standard procedures described therein in their
ceremonial duties. In some areas they have reportedly contributed to a
high degree of emulation of plains Hinduism. Har Nam, on his visits
to Sirkanda, is without doubt the most vocal advocate of plains Hindu-
ism in the village. The influence of such people in Sirkanda has been
insufficient to alter significantly many of the basic beliefs or practices
of villagers, but they have not been without effect. Superficial famil-
iarity with certain deities and concepts identifiable with the "great
tradition" but alien to the hills has resulted from the influence of these
Brahmins, among others. Its superficiality is indicated by the fact that
many of these deities and concepts are not integrated into the religious
life of the village or region.

On the other hand, some aspects of the great tradition are an old
and integral part of Pahari Hinduism. The story of the Mahabharata,
including especially the Pandava brothers and their wife, their allies,
and their enemies, is an example. The god Shiva, relatively recent in
Hinduism but now universal, may even have originated as a Pahari
god. Belief in fate and reincarnation, practice of certain annual and
life cycle ceremonies, and many other religious features could extend

THE SUPERNATURAL IO5

this list. Many of these great traditional elements are as old as Pahari
culture, which like all cultures is constantly changing and adding ele-
ments. In any case they confirm the Hinduism of these people and
their ancestors, but they do not alter the fact of its regional distinctive-
ness nor the significance of its deviations from plains orthodoxy.

Beliefs and knowledge of such aspects of the great tradition as cos-
mology and the origin of man and the universe vary with the experience
of individuals. Broadly they conform to ideas held by village Hindus
elsewhere in India. Some village Brahmins have read or heard religious
works and the explanations given therein. People who have attended
schools have heard the teachers' explanations, which are likewise based
upon the great tradition, with a smattering of Western science. Every-
one has heard Brahmins or others give various explanations along these
lines. Most villagers, however, simply have no consistent or fixed opin-
ions on these subjects. If asked they will repeat stories they have heard
which are widespread in India, such as that the earth was created when
an egg broke in half forming the earth from one half and the sky from
the other, and that the earth is held on the horns of an ox. The prob-
lem is apparently not relevant to most villagers.

It is possible to identify among the deities of local significance in
Sirkanda evidence of relationship to prominent gods in the literary
tradition of Hinduism. These relationships are often vague, sometimes
not extending beyond the similarity of name or form. To many villagers
such relationships are largely or totally unknown, while to others,
though known, they are ignored or their significance is virtually nil.
To educated people and especially to practicing priests these relation-
ships are known and often magnified. The trend toward emulation of
plains Hindus and the attendant self-consciousness about local tra-
dition on the part of educated people leads to the adoption of explana-
tions for local beliefs and practices in terms of all-India and regional
Hinduism where these probably did not previously exist and are un-
recognized by most of the people.

Four or five high-caste families in Sirkanda are alleged to own copies
of one or more classic religious works, but they read these rarely if at
all and none of them holds regular readings or worship. No Sirkanda
family has religious pictures or objects other than shrines to local gods.
The traditional Brahmin or purohit for the village (who lives in
another village) does possess religious pictures representing gods of
the great literary tradition, but he pays little attention to them or to
their meaning.

The effect of educated Brahmins is greatest in the conspicuous as-
pects of plains Hinduism — in getting villagers to Sanskritize their cere-

106 THE SUPERNATURAL

monies, particularly the marriage ceremony, and to observe some of
the periodic all-India festivals. At the same time they have achieved
some success in getting these Paharis to conceal some of their more
flagrantly non-Sanskritic practices from outsiders, by communicating
to them most of the responses and behaviors necessary to gain the ap-
proval of plains Hindus. There is a general self-conscious reluctance
among many Paharis, especially those of high caste, to discuss their
ritual and religious practices, their marital regulations, and so on, and
a tendency to present these as being closer to Hindu orthodoxy than
they actually are. Here acceptability by high-caste non-Paharis and
educated Paharis has been a major goal, and sophisticated Brahmins
have been among its most active advocates. They have attempted to
counteract the derisive connotation which the term "Pahari" and es-
pecially "Pahari Brahmin" holds for many non-Pahari Hindus. It is
toward the same goal that educated and prominent Paharis are striving
when they recall and extol the sacred heritage of the high-caste hill
people of this area. For these purposes conspicuous aspects of practice
are more important than subtle aspects of belief. The religious and
social changes in Sirkanda have been in the direction of the efforts of
those who wish to bring about more of these outward changes.
Crooke refers to the all-India spread of this process:

If the chief of a forest tribe becomes for the sake of respectability an
orthodox Hindu, he brings with him his tribal or village god, who becomes
an incarnation of Vishnu or a manifestation of Siva. If a village shrine gains
a reputation for miraculous cures of spirit diseases ... by and by a Brahman
or an ascetic takes possession of it as a working concern, and develops it ac-
cording to orthodox rule. (Crooke, 1926, p. 28)

Srinivas (1956) is among those who have made the same point:

Each region has its own body of folklore about the heroes of the Ramayana
and Mahabharata and not infrequently, epic incidents and characters are
related to outstanding features of local geography. And in every part of India
are to be found Brahmins who worship the local deities which preside over
epidemics, cattle, children's lives, and crops, besides the great gods of all-
India Hinduism. . . . Throughout Indian history Sanskritic Hinduism has
absorbed local and folk elements and their presence makes easier the further
absorption of similar elements. The absorption is done in such a way that
there is a continuity between the folk and the theological or philosophical
levels, and this makes possible both the gradual transformation of the folk
layer as well as the "vulgarization" of the theological layer. (Srinivas, 1956,
P- 494)

Here some of the apparent ties between beliefs and practices in Sir-
kanda as compared to those found over part or all of India and in the

THE SUPERNATURAL 107

literary tradition will be mentioned. The village gods of Sirkanda show
closer affinities to gods of the great tradition than do household gods,
and so they will be discussed first in this context.

The Bhartwali shrine in Sirkanda is said by some informants to be
in honor of Bharat, brother of Rama in the Ramayana epic. Kali, who
possesses a low-caste woman during kalratras to village gods in Sirkanda,
is the same in name as the Kali who is prominent in Hinduism as a
deification of femaleness. In Sirkanda she is not so identified. Devi is
apparently also a manifestation of the mother goddess and particularly
the goddess known as Durga, who is popular in the hills as well as
elsewhere. She is represented by Sirkanda informants as riding a tiger,
the characteristic vehicle of Durga.

Raghunath is identified by some in Sirkanda as Mahasu or Mahadev,
generally described as other names for Shiva, one of the three great gods
of Hinduism. In Kumaon, Raghunath is identified as Vishnu, another
of this trinity (Atkinson, 1884a, p. 813). In Sirkanda this deity is as-
sociated with Parasu Ram, an incarnation of Vishnu whose image is
in the Raghunath temple. Shiva as represented in prominent temples
is occasionally worshiped by Sirkanda people, and the commemoration
of his birth is one of the Hindu festivals recognized in Sirkanda. Some
historians of Garhwal believe that Shiva as an all-India deity originated
in the beliefs of the aboriginal inhabitants of this area. (Cf. Bahadur,
1916, pp. 73 f.) These are deities worshiped in Sirkanda which are di-
rectly identifiable with gods of the great literary tradition.

Two household gods have tenuous ties of this type: Nar Singh is the
name of an incarnation of Vishnu, and temples are devoted to him in
Garhwal. In Sirkanda, however, he is a household god. As such he is
found throughout this region and is also reported for the hill region
of Chamba (Crooke, 1926, p. 201). Nag Raja, the snake king, has a
Shiva temple dedicated to him elsewhere in Garhwal (Atkinson, 1884a,
p. 811). ". . . We have numerous traces of Naga worship in these hills,
but now chiefly connected with the special cult of Vishnu or [less fre-
quently] Siva" (Atkinson, 1884a, p. 835). Worship of this deity is non-
Brahmanical in origin and is thought to have derived from the religion
of the Nagas, a non-Aryan tribe of uncertain affinities and wide spread
who have apparently long been displaced to the east or have been ab-
sorbed in the later populations in Garhwal (cf. Atkinson, 1884a, pp.
373 f.; Crooke, 1926, pp. 383 ff.). In Sirkanda, Nag Raja is a household,
or at most a village, god.

Paharis are generally described as Shiva worshipers, but in Sirkanda
deities identifiable with Vishnu are more prominent. To Sirkanda
residents, however, these affiliations are irrelevant and largely unknown.

108 THE SUPERNATURAL

To them each is a deity in its own right, and none overshadows house-
hold gods in their relevance to daily life.

Notwithstanding the number and importance of the more orthodox forms
of Vishnu and Siva in this portion of the Himalaya the non-Brahmanical
deities . . . have far more worshippers and are more constantly addressed.
. . . The common resort in times of trouble or distress is Goril . . . and the
other village gods. (Atkinson, 1884a, p. 839)

The general belief of Sirkanda villagers relative to the great tradi-
tion of Hinduism, and the essence of their responses to inquiries de-
signed to bring out relationships between their brand of village Hindu-
ism and all-India Hinduism, were summed up by an informant who
said, "The learned people know all those things, we don't. They don't
concern us."

The term "village gods" or "local gods" as applied to Indian villages
should not be interpreted as meaning that these gods are unique to the
particular village in which they are found. As a matter of fact, like most
other traits of culture they generally occur over cultural-linguistic
regions of fairly broad extent. Those that originate in a single village,
say as a deified ancestor, are likely to spread to others as their fame
spreads and as they accompany brides to other villages, until eventually
their origin is forgotten or in dispute. Most of the non-Sanskritized
deities, village and household, worshiped in Sirkanda are of this super-
local character. "Goril ... if we judge from his general repute and
the number of temples to his name, is the most popular of all deities
worshipped by the lower classes in Kumaon" (Atkinson, 1884a, p. 821).
A birth story recorded by Atkinson for Gauril is different but vaguely
similar to that told in Sirkanda. Kali, who is a manifestation of the
"mother goddess" in great traditional Hinduism and is associated
with the Pandavas in Sirkanda, is reported by Atkinson (1884a, pp.
821 ff.) in that story as the mother of "Goril." This relationship did not
appear in my Sirkanda materials. Kalua, brother of Gardevi and as-
sociate of Gauril in Sirkanda, is the name of an independent deity who,
according to Atkinson, originated in a village in Garhwal as the spirit
of a murdered man, and Masdn (Appendix I), is identified as a demon
who inhabits cremation grounds. Bhuia is well known in villages
throughout northern India as the "tutelary god of fields and boundaries,
... a beneficent deity who does not as a rule force his worship on any-
one by possessing them or injuring them or their crops" (Atkinson,
1884a, p. 825; cf. Crooke, 1926, pp. 87, 92).

Mahasu is a name applied to a group of four regional deities in
Jaunsar-Bawar adjacent to Garhwal (Saksena, 1955, pp. 40 ff.; Atkinson,

THE SUPERNATURAL 10g

1884a, pp. 836 ff.). As mentioned above, the same term refers to a single
deity sometimes equated to Shiva in Sirkanda. Reference has already
been made to various regional interpretations of Nar Singh and Nag
Raja. Devi shows similar variation. Such similarities, variations, dis-
crepancies, and identities could be repeated many times over for the
other deities of Sirkanda, and in fact for those of villages throughout
India.

Caste and Deities

Religious belief and practice do not vary significantly by caste
in Sirkanda. No caste gods, as such, are worshiped. Low castes do not
engage in unique worship or hold beliefs that offer a clue to their
presumed early cultural distinctiveness from the higher castes. In this
Sirkanda differs markedly from plains villages, where caste differences
in religious practice are pronounced (cf. P. M. Mahar, 1957; Cohn,
1954, pp. 174 ff.; Planalp, 1956). Perhaps the absence of caste distinc-
tions in Sirkanda's religion can be largely attributed to the relative
lack of physical and social isolation of castes. We have noted, however,
that certain gods tend to be identified with certain castes in Sirkanda.
Informants point out that such distinctions are most likely to appear
during village dances, when certain gods tend to possess members of
certain castes. Thus, Bhartwali, if he were to appear, would allegedly
possess a Brahmin, Memendia would possess a Rajput (though he has
on occasion possessed members of all castes). A blacksmith (Lohar)
would be possessed by a god called Kaludl Lohar if it appeared, and a
Bajgi would be possessed by Kali Das. In practice in Sirkanda the last
of these appears regularly, the first two rarely, and Kalual Lohar not
at all.

One low-caste informant commented that the gods do not recognize
caste distinctions and they never punish people for breaking caste
rules. As he said, "If gods were caste-conscious would Devi, who is such
an important deity, possess the barber's wife?" The fact that high
castes are apparently somewhat more aware of Sanskritic belief and
ritual than low castes seems to be a result of their greater contacts
with education and educated people rather than of any caste heritage
of belief.

Caste discrimination in worship has been mentioned in the dis-
cussion of pilgrimages. Discrimination occurs in the village temples
as well as in distant ones. At festivals and marriages low-caste members
receive their blessings indirectly from the Brahmin, who flicks the
vermilion paste onto some object as, for example, the drum of a Bajgi,
and it is then applied to the Bajgis' foreheads by the owner of the drum.

HO THE SUPERNATURAL

Low castes perform the same ceremonies and worship as high castes,
but they do so separately or indirectly so as not to pollute the high-caste
worshipers. They stand outside the temple to worship Raghunath, but
they worship him as devoutly as do Brahmins or Rajputs.

Ancestors

The spirits of deceased people may evolve into gods as was re-
portedly the case with Kalua in Kumaon mentioned above, with Happy
Eye in Kishan Garhi (Marriott, 1955a, pp. 212 f.), and Ghatal Deo in
Rampur (Lewis, 1958, pp. 204 f.). In Sirkanda such spirits occur in a
form closely resembling household gods in behavior and demands.
Some of these might conceivably become minor gods — the first step in
"universalization" or elevation to super-local status (cf. Marriott,
!955a> PP- 207 ff.).

Such spirits occur in Sirkanda as rati, the spirit of a deceased male
who returns to torment the household in which he had lived, and
hanthia, the spirit of a female. These beings are identified by shamans
and are worshiped as household deities by the households they afflict
(see Appendix IC). The difference is that only one household worships
each of these, and generally after a generation or two they are dropped.

In Sirkanda there are currently thirteen of these ancestor spirits
which afflict a total of eleven households, nine of which are Rajputs,
one a barber, and one a Bajgi. Eight of the spirits are female. The re-
lationship of these to living household members is as follows: four,
and possibly five, are wives who do not want to be forgotten by their
living husbands, two are mothers, and one is a co-wife. Of the five male
spirits, three are father's younger brothers, one is a father, and one a
brother, of current householders. They are worshiped in a manner very
similar to that for household gods, and under similar circumstances.

An ancestral spirit that has virtually become a devi or goddess and
is, in fact, sometimes referred to by that term, is that of a village wife
who committed satl (immolation on husband's funeral pyre). After the
event no stone or monument was erected in her honor. People of the
village gradually forgot about her noble act and failed to honor her.
After some time the family of the woman began to suffer with boils
and fevers, and their cows went dry. A shaman diagnosed the trouble
as being due to neglect of the sati. Therefore the householders built a
shrine to the shaman's specifications. This shrine is now the place of
worship for the sati spirit or devi. So far its use has been limited almost
entirely to the Rajput sib of the household who built the shrine, but it

THE SUPERNATURAL 111

has extended within the sib beyond that particular household and
lineage. It seems a likely candidate for a village goddess.

All recent ancestors in each household are worshiped in an annual
ceremony {kanagat, described in Appendix II), even though they have
in no way afflicted their survivors.

Ghosts

A variety of ghosts or malevolent spirits inhabit the countryside,
unassociated with any one family or even any one village. They are
usually the wandering spirits of deceased people who are wont to at-
tack living people with results ranging from death to luckless fishing.
In general these ghosts (bhuts) or demons (shetanths) attack at night
and in dark places. Strange or unaccountable sounds, rock slides, and
the like are attributed to them. None inhabit Sirkanda, but some are
known to inhabit deep, jungle-covered ravines on trails leading into
Sirkanda. Whistling, lights, and dislodged stones attributed to them
are frequently encountered at night. Fear of them tends to keep people
from traveling at night.

The worst of the ghosts come to people and enter their bodies be-
cause they are miserable disembodied souls seeking a body in which to
dwell. Others are jealous of human beings, or merely cantankerous.
However, possession by a ghost differs from that by a god or ancestor.
The term for ghost possession is bhut lag gea (ghost has adhered) or
chat patana (to cause torment as evidenced by trembling), as con-
trasted to "sir a gea" (has come to a head) for possession by a god.
Ghost possession is qualitatively different from possession by a god.
Once a ghost has "caught hold" it will usually not leave unless forced
out by exorcism. During the time it is in possession, it causes illness,
unusual behavior, and even death. Insensitivity to pain is the usual
diagnostic symptom of possession by a ghost. Other afflictions are: eyes
rolled back, fever, inability to talk coherently, melancholia, some sen-
sory disability or aphasia, catatonic adherence to a particular posture,
illness, barrenness, or repeated stillbirths. Sometimes these afflictions
may prove incurable because the ghost cannot be exorcised. Inexpli-
cable sudden death is often attributed to ghosts. The very fact of pos-
session by a ghost is harmful to the one possessed. A god, in contrast,
bothers his victim without possessing him. He possesses only to dance or
to speak. He does so temporarily without harm to the one possessed,
and he leaves voluntarily.

Ghosts do not attack as punishment for misdeeds or oversights, and

112 THE SUPERNATURAL

they are not usually propitiated. They are avoided by means of charms,
magical acts, or use of objects inimical to them. Once possession has
occurred, they are driven out with the help of a professional exorcist
who knows what will send them away. A few require sacrifice as ran-
som before departing, but this is not considered to be propitiatory
sacrifice in the sense of an offering such as is given to the gods. People
simply pay off an evil spirit. There is no feeling of guilt or laxity, as
they cannot avoid such spirits by worship in advance.

Usually a shaman diagnoses possession by a ghost and refers the pa-
tient to an expert in exorcism. Practitioners who exorcise ghosts can
often do so simply by chanting appropriate mantras and placing a
sacred mark on the victim's forehead. The ghost cannot face such sa-
credness and must leave. Some ghosts leave under physical duress such
as the application of a scorpion plant to the body, a mild beating, or
blessed grain thrown in the face of its victim. Some require very com-
plex rites of exorcism. Some can be exorcised only at certain places,
for example, by running water, at a cremation ground, at a crossroads
where a goat's or sheep's head has been buried. Some can be warded
off by specific objects, words, or written formulas. Iron, black color, and
scorpion plant are anathema to most ghosts.

In appearance and effect as well as susceptibility to various types of
exorcism, ghosts vary. All or nearly all have backward-turned feet and
long fangs if they have human form. Some take the form of snakes or
other animals; one looks like a white pillar, another like a bright light.
They are adept at disguise. Some are prone to attack women, one at-
tacks new mothers, another attacks men wearing blankets. Cremation
grounds are populated by many ghosts — those of the recently dead,
those which torment the souls of the dead, and those which feed on
corpses. It is often remarked in Sirkanda that it is really ghosts rather
than flames which consume the corpse.

Special precautions are taken at cremation to ensure that nothing of
the body remains for ghosts to use or torment sexually or otherwise,
and to ensure that the spirit of the deceased will not be bothered by
other ghosts. The spirit of a dead person wanders for at least thirteen
days after cremation. No matter how well-loved the person was in life,
his spirit is feared in the village until after the thirteenth-day ceremony,
which allows it to settle in some body or at least to pass into a less rest-
less condition. A new mother and infant are especially vulnerable to the
jealous ghosts of women who died in childbirth. As protection, mothers
carry iron sickles and black blankets when they go out and leave scor-
pion plant and the sickle by their child or in the doorway when they
are at home.

THE SUPERNATURAL 1 1 3

In discussing ghosts of the dead and ancestors, it is worth noting that
the dead who return to torment the living are those who are thought
to have reason to be angry or jealous. They are usually those who died
prematurely or are improperly mourned. A deceased wife torments her
successor; a person who died prematurely, jealously torments the liv-
ing; a woman who died in childbirth attacks young mothers; a father
whose sons failed to give proper charity on his behalf at his funeral
suffers in the afterlife and therefore plagues his sons; a man whose
family neglected to perform the thirteenth-day ceremony after death
cannot enter a new life, and so disturbs his family; a woman who
brought honor to her family by immolating herself on her husband's
funeral pyre torments her neglectful descendants.

Ghosts are often categorized terminologically by their origin. Thus
chuRail is the ghost of a woman who died in childbirth, and shaiad is
the ghost of a Muslim. Eight named ghosts plus a number of unnamed
ones were described by Sirkanda men. Some of these are well known in
other areas. Crooke (1926, pp. 134, 171), for example, describes "churel"
and "shahid" on the plains, corresponding to the two types mentioned
immediately above. Others are doubtless of very limited distribution.

Sprites

One category of supernatural beings differs from others in that
their effects on people are harmful but not malicious. These are the
matriya (matris) or sprites associated with the goddess Matri, mentioned
above and representing the ghosts of little girls. They live around the
shrine dedicated to Matri, atop the hill overlooking Sirkanda, and their
affinity to her may be based primarily on that fact (see Appendix IB).
Sirkanda villagers think of them as a band of fast-moving, invisible,
playful fairies who occasionally leave their normal abode on the hill-
top to flit about over the countryside like a flock of birds or a swarm of
insects. If a person happens to get in their way while they are traveling
or cavorting, he will be struck with fatal or debilitating maladies such
as insanity, blindness, deafness, muteness, or aphasia of various sorts.
Sudden occurrences of such disabilities, without warning symptoms,
are often attributed to these beings. The effects are considered incur-
able, but the sprites have not inflicted them as punishment, nor is it
done in anger or spite. It is simply an inevitable result of happening
into the midst of matris at play. The matris are capricious and without
malice. They strike people in passing rather than possessing them.
Therefore, they cannot be propitiated or exorcised. They are a hazard
to human beings — one that cannot be anticipated, avoided, or cured.

114 THE SUPERNATURAL

"He just got in the way of the matris; it was his fate," is the way their
afflictions are explained. Fortunately, the chances that one will be
struck by them are slim, for like lightning and wild animals they strike
rarely and over an area so large as to make direct contact with them
unlikely. Perhaps one out of each thousand Bhatbair residents has been
struck by matris.

Their nature and activities are very similar to those of the acheri
or fairies described by Traill. "These reside on the tops of mountains,
but descend at dusk to hold their revels in more convenient spots. To
fall in with the train, at the time, is fatal . . . : they occasionally also
molest those who may cross the sites of their abodes during the
day . . ." (Traill, 1828, p. 221; cf. Oakley, 1905, pp. 212 f.).

It should be remembered that the categories of supernatural beings,
such as sprites, ghosts, household gods, and village gods, as well as the
subdivisions of these categories, are not entirely distinct and in practice
constitute a continuum or spectrum of intergrading types of super-
natural beings with similarities running through all. When informants
recount memorable or recent instances of the machinations of super-
natural beings, they often jump from one type to another and back
again without pause, explanation, or even identification of the type of
supernatural responsible. On the other hand, inquiry can usually pro-
duce such identification.

Supernatural Functionaries

In a world beset by such a variety of supernatural troublemakers
there are a variety of professional people whose job it is to deal with
them. In Bhatbair these can be categorized according to the functions
they perform, as priests, temple keepers, shamans, pujaris, diviners,
exorcists, and curers. These categories are not necessarily mutually
exclusive.

Priests maintain the system of Hinduism in ways prescribed in the
great tradition. It is they who help people achieve a favorable afterlife
and rebirth. Their primary responsibility is the long-range welfare of
their clients. They perform the annual ceremonies and life-cycle
rites as well as many special ceremonies. In addition they perform
pujas to village gods, and they practice astrology and some forms
of divination based on written prescriptions of Hinduism. They
are the local experts on the great tradition, though their beliefs and
daily practices in Bhatbair do not differ materially from those of other
castes.

Temple keepers maintain particular shrines to honor gods, receive
gifts on their behalf, and often lead or assist in their worship.

THE SUPERNATURAL 1 15

The other supernatural functionaries deal with the exigencies of
daily life relative to the supernatural. Their primary responsibility is
the immediate welfare of their clients in this life. The shaman who
diagnoses difficulties through the offices of his own personal deity, and
the pujari who performs worship for household gods and plays drums
to let the gods dance, have been discussed sufficiently above so that no
elaboration is required here (cf. Berreman, 1961a). These two types of
functionaries enable people to find out what deities are troubling them
and how to appease them. The pujaris help in satisfying the deities'
demands.

With regard to gods worshiped locally we can say that Brahmin
priests are to village gods as pujaris are to household gods; that is, they
perform comparable functions. By the same token, temple keepers are
to the gods of their temples as household heads are to the gods of their
households; they maintain the shrines and supervise the worship,
though often specialists actually perform the worship.

Aside from these there are a variety of specialists in several kinds of
supernatural activities. These practitioners range from the lowliest
shoemakers to Brahmins, but there is a heavy predominance of low
castes. Their methods vary with their specialties, their place of origin,
and in some cases their personal predilections. Although the specialties
are not hereditary, the practitioner usually derives the necessary knowl-
edge and skill from a parent or other relative who practiced before
him and taught him the necessary lore.

In Sirkanda there are two such practitioners, both Bajgis. Each rep-
resents a common type of practitioner in the area. One is a diviner.
He performs much the same function as a shaman in that he diagnoses
trouble and divines the past and future, but he is without the shaman's
personal god as his source of information. In this practice he uses at
least two techniques of gaining insight, one involving a system of
counting grains of rice brought to him by a client, and the other a
system of reading significance in patterns of drawn lines. The second
Sirkanda practitioner is an exorciser of ghosts and spirits. This man is
credited with several impressive successes. He is especially adept at
exorcism of ghosts which cause stillbirths. One of his techniques is
that of jhdrd tdrd karnd (to do brushing away of evil), a term applied
to the magic used to treat cases of spiritually caused diseases and other
difficulties. He is sometimes called to diagnose and treat a patient as-
sumed to be possessed by a ghost. More often he is called as a result of
diagnosis by a shaman in order to carry out exorcism.

There are many other practitioners of similar arts in neighboring
areas. In a distant but accessible village lives a Nepalese Rajput ex-
orciser of considerable renown and spectacular methods who brings

Il6 THE SUPERNATURAL

ghosts physically into his clients' terrified presence in their natural
form and habitat (at the cremation grounds) and browbeats them into
submission to his will. Like well-known specialists in many cultures,
his services do not come cheaply and he is therefore a last resort in
most cases of ghost possession.

All these practitioners serve over a wide geographical area, and their
practice is a direct reflection of their reputation for success. Rarely are
their services as highly valued in their own villages as in distant ones
— an instance of the general rule in these hills that respect for a prac-
titioner who deals with the supernatural varies inversely with physical
(and sometimes social) proximity to him. Typical is the Sirkanda Bajgi
exorciser, one of the few of his profession able to make a living without
supplementing his income by following a trade, making liquor, or
engaging in agriculture. He is patently in great demand in other vil-
lages and lives as well as many other villagers, but he is rarely con-
sulted in Sirkanda. A villager summed up local opinion when asked
his estimate of this practitioner's skill: "If he were any good would
he live here in poverty as he does?" Another commented, "Everyone
has to make a living; some work, some beg, some steal; he does exor-
cising." Similar attitudes are expressed regarding the local Brahmin
priest as compared to more distant and hence more respected priests.

Diseases are usually considered to be supernaturally caused. Diagnosis
is most often performed by a shaman, who recomends the appropriate
type of practitioner and treatment. As we have seen, shamanism is a
major focus of religion in Sirkanda. Some common diseases are thought
to be amenable to direct treatment either with herbs and diets or with
incantation by minor specialists. Those who treat such diseases tend
to specialize in certain diseases or groups of diseases. Other practi-
tioners, like the Sirkanda Brahmin, will treat almost any ailment.
Afflictions characterized by some external, visible symptom such as
sores or swelling (for example, boils, gout, mumps, chicken pox, sore
eyes, snakebite, and spider bite) are usually treated by the class of
specialists called jharnewala (practitioner of jhama, stroking or brush-
ing). In Bhatbair there is a specialist for each of the above-named ail-
ments, and for most of them there is a specialist in Sirkanda itself.
Such practitioners may be men of any caste who perform this duty
when called upon but do not make a major occupation of it. Treat-
ment varies with the disease, but in general it involves recitation of
appropriate mantras by the practitioner while he repeatedly touches
the afflicted member and the ground (or water, lump of sugar, animal,
or other object) alternately with a hawk's feather, thus transferring the
trouble from the patient to the ground or object. Smallpox is rare in

THE SUPERNATURAL 1 17

the area, but if it occurred would be treated by worshiping the deity
to whose displeasure it was attributed, probably Devi.

As previously noted, the Sirkanda Brahmin practices elementary
Vedic medicine. In this practice he follows a book of instructions,
based upon Hindu religious works, which directs the practitioner in
herbal, dietary, and hot and cold treatments, as well as others. He also
practices jharna.

General debility or lingering illness may be treated by giving charity
to relieve the misery. The treatment, under the direction of a Brahmin,
is called "weighing." It is a rather elaborate process the essence of which
is that the patient is balanced in a large scales against a counterweight of
several items, primary among which is grain. A puja is performed, the
counterweight goes to the Brahmin as charity, and thereafter the ill-
ness is cured.

Occasionally Vedic doctors in Mussoorie or Dehra Dun are consulted
to set bones or prescribe for diseases such as tuberculosis, malaria,
venereal disease, chronic dysentery, general weakness, and male sexual
difficulties. Women rarely go outside the village for treatment. In
serious or lingering cases a variety of treatments and practitioners may
be tried. Some minor ailments are treated in the home without the ad-
vice of specialists, either with traditional folk remedies or with local
interpretations of Vedic prescriptions.

Spell-Casting, Witchcraft, and Evil Eye

Some practitioners of curative magic and occasionally others have
learned to cast spells (jaddu) into people that will cause either slow
suffering and eventual death or sudden death, depending upon the
type of spell. The latter is more difficult and the method is a closely
guarded secret, while the former is well known. The same practitioners
are proficient in the process of lautand, "to send back" or counteract
jaddu, whereby the evil spell can be removed and returned to the
sender or put in some harmless place. Such practitioners are supported
by the god Bhairu. A goat is sacrificed to the god as part of the prepa-
ration for sending or counteracting a spell. There have been Bajgi,
Rajput, and Brahmin practitioners of this magic in Sirkanda. In ad-
dition, there is a lesser form of jaddu which takes the form of practical
jokes which inconvenience or make ridiculous their victims. This type
is especially common among Bajgis, who use it to interfere with the
drum-playing of their rivals.

Casting of spells and counterspells is a very precise and difficult art.
Slight mistakes in the ritual can result in the death of the practitioner,

Il8 THE SUPERNATURAL

and instances are readily cited. In one local case the magically treated
pot by which the spell was to be conveyed turned upon the errant
spell-caster who had prepared it and dispatched him with a blow on
the skull. Similarly, misuse can be the doom of the practitioner, ac-
cording to a belief that fatal jaddu must be used only on those who de-
serve death. It is often stated that anyone who learns this art has to
pledge his teacher never to use it irresponsibly under pain of dire
punishment. One death and a case of affliction with leprosy in the fam-
ily of Tula, the traditional Brahmin of Sirkanda, are attributed to
injudicious, impulsive use of fatal jaddu by a member of that family
on an innocent Bajgi victim.

Jaddu is accomplished by a combination of imitative and contagious
magic. A flour-and-water image of the intended victim is prepared and
placed in a pot along with various rare and esoteric, magically treated
items. Incantations are recited over this, and pulses (legumes) are
thrown over the image as its intended fate is spelled out. Then the pot
is secretly buried in the home of the intended victim, who begins to
suffer the consequences. The source of the victim's trouble is usually
diagnosed for him by a shaman. Then a practitioner — preferably the
one who cast the spell in the first place — is called in at considerable ex-
pense to nullify the spell by a special performance with a similar con-
tainer, or redirect it by finding the buried container so that it can be
reburied elsewhere.

A related technique of spell-casting is magically sending flour
and water images of the victim and of a sheep (the latter as an
offering "to the art of jaddu") to the victim after magical preparation
similar to that exercised upon the pot in the above description. The
images are made to walk in upon the victim by force of magic, where-
upon he is expected to drop dead of fright. This spell can be counter-
acted by a practitioner if the victim survives long enough to summon
one.

Mantras can theoretically be recited by a practitioner over the nail
and hair clippings of a potential victim to drive him insane by jaddu,
but fear of this danger is not sufficient to motivate Sirkanda people
to conceal clippings. Shamans rarely designate clippings as the source
of a person's trouble; ghost possession or contact with matris are the
usual diagnoses of such symptoms. Nail and hair clippings automati-
cally cause bad dreams if left in a house and will cause creaking roof
beams if placed among the rafters of a house. In such cases a shaman
can diagnose the trouble and specify where the offending objects can
be found.

Dag, witchcraft, is practiced secretly by certain women. Other hill

I I

THE SUPERNATURAL 1 1 9

areas, including Jaunpur and Jaunsar-Bawar, are thought to abound
in witches, so that girls are not sent there as wives or taken from there
without careful advance investigation. Three of five Sirkanda girls
who were married in Jaunpur died early, allegedly as a result of witch-
craft. There has never been a witch in Sirkanda and there are no witches
in Bhatbair now, according to Sirkanda people, though there are re-
ported to have been some who came as brides from time to time in the
distant past. The nearest witch now lives in a village on the farthest
periphery of the area within which Sirkanda people regularly marry.
Witchcraft is associated with people who are different, and hence mys-
terious and therefore dangerous. Witchcraft is a reason for staying away
from alien people and areas and for avoiding unnecessary interaction,
especially marriage, with distant and unknown people.

Witches are very covetous. They insidiously destroy any person,
animal, or object which they admire. They can disguise themselves as
cats or other animals to gain access to the homes of their victims.
There they may measure the victim with a thread or eat his liver or
replace it with another object without leaving a trace. As a result the
victim, whether human or animal, will die a miserable death. Some-
times shoemakers report that dead cattle are found to have had no
liver — a sure sign of dag. There is no countermagic for this dread
practice.

The "evil eye" (nazar lagnd, to look covetously) is well known over
North India and bears some superficial resemblance to dag. In Sir-
kanda, as elsewhere, the evil eye consists of a covetous, greedy, admir-
ing, or envious look cast by certain persons at any person, animal, or
object. The result is that the object of the evil eye is harmed — the
child or adult becomes weak or sickly, the cow goes dry, the artifact
breaks. The possessor of the evil eye is usually unaware of the effect
he produces. Not all people possess the evil eye, and those who do
cannot control it. It is not restricted to any particular type of person.
A shaman generally diagnoses the trouble. Specialists prescribe amulets
or short ceremonies to dispel its effects. However, it is not very much
feared in this area, and is rarely blamed for trouble. There is not, as a
result of this belief, reluctance to have a child admired or complimented
in any way, as is often the case in the plains. Neither are such precau-
tions as eye-blackening commonly used to protect children. As one in-
formant said,

If we have many children we don't fear evil eye at all. If we have only
one or two we fear it and we protect the child by giving extra charity and,
if necessary, with amulets. We like to have our children admired and we

120 THE SUPERNATURAL

don't think admiration increases the likelihood of evil eye. It is counteracted
by charity and charms, not by disguise.

A somewhat related concept is that of dux lakshirii, a person of evil
portent. The term refers especially to a person who utters curses or
dire predictions which have a tendency to come true. If a person uses
a common curse such as "I hope you fall and skin your knees," and it
comes true regularly, he is referred to by this term and is feared.

Certain social relationships into which people enter prove to be in-
herently fortunate or unfortunate. Most frequently, marriage is of this
nature. Some marriages bring a marked change in the luck or fortune
of a family. The change is attributed to the new relationship and more
particularly to the bride since, in this patrilocal society, she is the new
element in the situation. There are many examples of Sirkanda house-
holds whose luck changed from bad to good or good to bad immediately
after a marriage. Bad luck is attributed to misread or mismatched
horoscopes of the married couple because, if properly read, the horo-
scopes would have predicted the difficulty and the marriage would
not have been performed.

Similarly the birth of a particular child may prove lucky or unlucky
to a family. Children born under one astrological condition (mul) are
so dangerous to their parents that they are usually adopted by others,
and they may even prove to be harbingers of bad luck to their foster
parents.

There is in Sirkanda a whole series of auspicious and inauspicious
omens, days, acts, and objects with corroborative stories about people
who have felt their effects or narrowly missed feeling their effects.
These need not be enumerated here. Some of the portents are based on
religious writings and are known primarily by the Brahmins who
consult religious works and astrological tables to detect them. Others
are known to diviners and similar practitioners. Many are the common
knowledge of every villager.

THE RELIGIOUS CONTEXT
CALENDRICAL AND
LIFE-CYCLE CEREMONIES

A number of recurrent ceremonies in which household and village gods
play important roles have been mentioned. In this chapter two cyclical
patterns of ceremonial observance will be discussed. The first of these is
the annual cycle of festivals, and the second is the lifetime cycle of rites
of passage — the ritual observance of changes in status of the individ-
ual. No attempt will be made to describe these observances in the
vast detail that would be necessary in order to reproduce them. Rather,
their main outlines, the social composition of groups which participate
in them, and their points of difference and similarity in comparison
with equivalent ceremonies of all-India Hinduism will be described.
It will be noted that no daily cycle of ritual or worship is described
for Sirkanda. There is no tradition of such practice in this village
among the members of any caste, although in other areas of the hills
high-caste people do perform daily ceremonies much like those ob-
served by some plains groups (cf. Atkinson, 1884b, p. 65). In Sirkanda
one middle-aged Rajput who has had eight years of schooling and
operates the village shop is the only person who lays claim to any such
practice, and he says only a brief prayer at the time of lighting his lamp
in the evening. A few high-caste households possess religious books
but do not read them regularly. There is little in the way of private,
individual worship of any sort in the village.

122 CALENDRICAL AND LIFE-CYCLE CEREMONIES

Calendrical ceremonies in Sirkanda are observed by all or nearly
all community members, but specific ceremonial activities which make
up the calendrical observances are carried out primarily within the
household. Life-cycle ceremonies are performed by the kin group
(usually the lineage) but are carried out in the household with house-
hold members playing the most important roles. The household there-
fore is the basic ritual unit.

Annual Ceremonial Cycle

Twelve ritual observances or religious ceremonies are celebrated
annually in Sirkanda.

Nine of these (numbers 4-12 in table 1) are local manifestations of
widespread Hindu festivals, two of which (numbers 5 and 12) are given
little more than lip service in Sirkanda. Another (number 9) is observed
in a manner very different from its observance on the plains. Three
festivals (numbers 1, 2, and 3) are apparently of only regional signifi-
cance.

This sequence can be compared with the calendrical ceremonies re-
ported for various villages of the plains: 12 are reported for the Chamar
caste of Senapur (Cohn, 1954, pp. 200 ff.), and 35 are recorded for the
entire village of Senapur (Planalp, 1956, pp. 249 ff.). Nineteen each
are reported for Kishan Garhi (Marriott, 1955a, pp. 191 ff.) and Ram-
pur (Lewis, 1958, pp. 197 ff.).

Marriott notes that half the festivals of Kishan Garhi are identifiable
with those of the great tradition. In making this comparison he makes
certain qualifications that must be borne in mind. These apply equally
well to other North Indian Hindu villages, including Sirkanda.

. . . The presumption that the festivals of Kishan Garhi are approximately
identical with those of the great tradition needs to be qualified in at least
four ways. These four qualifications bring us to confront the little tradition
of Kishan Garhi.

First, there are four festivals which have no evident Sanskritic rationales.
. . . Second, those festivals of Kishan Garhi which do have Sanskritic ra-
tionales represent only a small selection of the total annual cycle of festivals
which finds sanction in the great tradition. . . . Third, between the festivals
of Kishan Garhi and those sanctioned by the great tradition, connections
are often loosened, confused, or mistaken because of a multiplicity of com-
peting meanings for each special day within the great tradition itself. . . .
Accustomed to an interminable variety of over-lapping Sanskritic mythology,
villagers have ceased to be much concerned with distinguishing the "right"
great-traditional explanation of a festival from such Sanskritic-sounding and
possibly newly invented ones as may be convenient.

CALENDRICAL AND LIFE-CYCLE CEREMONIES

123

Fourth, behind their Sanskritic names and multiple great-traditional ra-
tionales, the festivals of Kishan Garhi contain much ritual which has no
evident connection with the great tradition. (Marriott, 1955a, pp. 193 ff.)

Srinivas has pointed out regional variety in festivals:

Festivals such as the Dasara, Deepavali, and Holi have no doubt certain
common features all over the country, but they have also important regional
peculiarities. In the case of some festivals only the name is common all over
India and everything else is different — the same name connotes different
things to peoples in different regions. (Srinivas, 1956, p. 494)

TABLE 1

Calendrical Ceremonies in Sirkanda»

Name

1. Phul Dalna

2. Poprla Sakrant

Meaning Month

Flower placing; rite of spring Chait (March-April)

First of month when poppers Baisak (April-May)
(type of bread) are eaten; chil-
dren's god worshiped

First of month when image of Jet (May-June)
wild goat is eaten

(Secular fair, following spring Jet (May-June)
harvest)

Wrist charms Bhado (August-

September)

5. Jenem Ashtml Barat Krishna's birthday fast Bhado (August-

September)

Worship of household dead Asoj (September-
October)

Nine nights (worship of Durga) Asoj (September-
October)

Cow honoring Katik (October-

November)

Festival of lights; follows fall Magsir (November-
harvest, one month after December)
plains Diwali

3. GhoRlia Sakrant
(Taulu)

4. Rakrl

6. Kanagat

7. Nauratra.

8. Gaojiman

9. Pahari Diwali

10. Kicherl Sakrant

11. Panchmi Basant

12. Shiv Ratri Barat

First of month when rice and Mau (January-
dal mixture is eaten February)

Spring fifth

Mau (January-
February)

Shiva's (birthday) night fast Phagun (February-
March)
For descriptions of these ceremonies, see Appendix II.

124 CALENDRICAL AND LIFE-CYCLE CEREMONIES

Marriott's first point has been applied to Sirkanda in the discussion
above, and his third and fourth points will be elaborated in the de-
scriptions to follow. The second point, that the festivals in any partic-
ular locality are but a small sample of the universe of Hindu festivals,
can be documented here. Of the thirteen major festivals of Hindu re-
ligion listed by Hindu scholars in Morgan's book, three are celebrated
in Sirkanda as compared to seven in Kishan Garhi (Morgan, 1953, p.
423; Marriott, 1955a, p. 194). "Of the 35 presumably all-Indian Hindu
festivals listed by Swami Sivananda only 9 occur in Kishan Garhi"
(Marriott, 1955a, p. 194), and only seven occur in Sirkanda (Sivananda,
1947, pp. 1-57). In comparing local with regional festivals we note that
among the twelve annual festivals of Sirkanda are nine of the nineteen
festivals of Rampur, seven of the nineteen at Kishan Garhi, nine of
the thirty-five calendrical festivals at Senapur, and five of the twelve
among Senapur Chamars. Obviously some of these are North Indian
regional festivals or festivals which, like Kanagat, are practically uni-
versal in Indian Hinduism, but are not included in the literary "great
tradition."

Sirkanda is within the broad culture area of North India but more
specifically in that of the sub-Himalayan hills. It would therefore be
expected to share religious observance with other people of the area.
Atkinson reports a variety of local, regional, and all-India ceremonial
and festival observances which occur among hill people of Garhwal
and Kumaon, totaling over 60 (Atkinson, 1884a, pp. 847-874). Ten of
these are among Sirkanda's twelve, leaving only two Sirkanda ob-
servances of such limited significance or distribution that they are not
reported for neighboring regions of the same cultural-linguistic area.

The outstanding feature of Sirkanda's ceremonial cycle is its basic
similarity to that of other Indian villages. Its greatest differences seem
to lie in its sparseness and the weakness of its ties to the great tradition,
even where great-traditional affinities can be detected.

Conspicuous by their absence in Sirkanda are the great festivals of
the plains, Holi, and Dashera. Occasionally Sirkanda villagers witness
such festivals in towns or trading villages in the valley. The Ram Lila,
a dramatic presentation of the Ramayana preceding Dashera, is at-
tended in valley villages by a few Sirkanda villagers, and Holi can
hardly be avoided by any who may be traveling outside the hills on
that day. Such events are, however, regarded as alien, just as are simi-
larly recognized Sikh and Muslim festivals.

The social composition of participating groups in calendrical cere-
monies in Sirkanda can be briefly indicated here, although descriptions
of the nature of the ceremonies is confined to Appendix II.

CALENDRICAL AND LIFE-CYCLE CEREMONIES 125

In all but one ceremony, the ceremonial observance takes place
primarily within the household unit, by household members. The
exception, Nauratra (number 7 on the list in table 1), is celebrated by
a village dance and worship of Devi. There is similar worship on
Diwali (number 9), accompanied by household worship and worship by
all devotees of one god, Nag Raja. All village children together worship
the children's god on Popria Sakrant (number 2), and there is house-
hold worship as well on that occasion. On all other calendrical cere-
monies household units worship separately.

Jenem Ashtmi Barat (number 5) and Shiv Ratri Barat (number 12),
the two fasts, honoring Vishnu and Shiva respectively, are observed by
only a few of the more sophisticated (educated, high-caste) families,
who fast in accordance with plains custom. On the other ceremonial
occasions there is enthusiastic participation by nearly everyone, on a
household basis.

Brahmins are prominently involved as priests in three of the cere-
monies, Rakri, Kanagat, and Kicheri Sakrant, and less prominently in
two others, Navratra and Diwali. In the first they tie wrist bands on
household members and receive charity in return. In the third they
are the recipients of charity given for general merit but not in return
for any specific service. In the last two they perform worship to village
gods on behalf of the village.

Bajgis, as drummers, are prominent in all three Sakrants, and in
the celebration of Panchmi Basant, for which they play their drums
and receive gifts of sugar, grain, or cooked food in return. They per-
form similar service for like remuneration on other periodic and
life-cycle ceremonial occasions.

Finally, the barber is the main religious functionary on Panchmi
Basant, when he performs a simple ceremony welcoming spring.

In the rest of the ceremonial observances cited here, the family per-
forms its own ritual functions without assistance.

One annual secular event is important in Sirkanda. This is the fair
(taulu) held in the middle of Jet (May-June), on a hilltop in the cen-
ter of Bhatbair about four miles from Sirkanda. It is widely attended
by Bhatbair residents including many people from Sirkanda, but at-
tracts no non-Paharis other than two or three merchants. This fair
corresponds in its temporal relationship to the spring harvest, precisely
as Diwali, the greatest annual festival of Sirkanda, does in its relation-
ship to autumn harvest. Both occur just after the hard work is over.
The fair bears many resemblances to Diwali, but the resemblances are
not related to religion, for it is not a religious event. As on Diwali,
there is general blossoming out in good clothes, accompanied by gaiety,

126 CALENDRICAL AND LIFE-CYCLE CEREMONIES

sociability, dancing, singing, drinking, and general lowering of in-
hibitions. The fair and Diwali are the two big holidays in the hills, and
they are anticipated with equal relish. Only an occasional wedding
matches them in public appeal. They may well be derived from the
two non-Brahmanical but "really popular festivals . . . held at the
two harvests," mentioned by Atkinson (1884b, p. 64).

Three other fairs are attended regularly by some Sirkanda people.
All have some religious significance that is vague or unimportant to
villagers. One is held at the Kuddu Kal temple in mid-AsaRh (June-
July), one is at Mussoorie in early Asoj (September), and the third is
at Rajpur in mid-Baisdk (April-May). The first two are primarily
Pahari fairs, and the last is heavily attended by Paharis. Sirkanda
people do not care to attend festivities of non-Paharis simply because
they feel out of place.

Life-Cycle Ceremonies

Three events in the lives of all individuals in Sirkanda are rein-
forced by the family and community with ritual performances. These
are: birth, marriage, and death. In addition high-caste males are ad-
mitted as adult members of their caste in the sacred thread ceremony,
generally associated with marriage. These rituals are universal in Hindu
India. Those in Sirkanda, like those in villages throughout India,
show certain differences from, as well as resemblances to, the rites
prescribed in the literature of the great traditions of Hinduism. In
Sirkanda they are based on written prescriptions used in the area by
Pahari Brahmin priests.

Birth and Childhood

Birth ceremonies in Sirkanda are much like those in the plains.
The birth of a son is announced by distributing sugar lumps to friends
and relatives on the day of birth. A girl's birth is not formally pro-
claimed. The first ceremony in the child's life is that of das sotan — the
tenth day after birth. On this day a Brahmin is called to prepare the
child's horoscope, a process which may take many days, during which
time he lives at the house of his client. The Brahmin gives the child a
name at this time, but he is not known by this name. Many people do
not know the names written in their own horoscopes or those of their
children. In the first weeks, months, and years a child acquires names
given by its parents or other family members. One informant had five
such names. Literate men often later adopt yet another name for signa-
tures which, in contrast to their village name, is usually a conventional

CALENDRICAL AND LIFE-CYCLE CEREMONIES 127

Sanskritic Hindu name. One man ridiculed a current trend toward
fancy Sanskritic names: "If you call a jackal 'lion,' it is still a jackal."

One year after birth a Brahmin conducts a ceremony, the "taste of
boiled rice," which is supposed to be at the time of first giving the
child solid food. Boys have their first haircut during a ceremony per-
formed by a Brahmin and a barber, held on the third, fifth, or seventh
birthday, depending upon the Brahmin's reading of the horoscope.

Special ceremonies related to birth include those performed to bring
about conception in childless couples, and those held to eliminate the
danger which two new mothers (those who have given birth in the
same month) pose to one another if they are in contact.1

Marriage

Marriage in any Hindu family is a complex process. Lewis dis-
tinguishes five major phases in what he terms the "marriage cycle" of
Rampur, and within these he notes twenty distinct "ceremonial steps"
(Lewis, 1958, pp. 157-190). He also distinguishes two less frequent
types of marriage, and remarriage of a deserted woman.

In Sirkanda there is greater variation in marriage practices than is

described or implied for Rampur. Ceremonies are used in Sirkanda to

celebrate only the first marriage of a woman — about 65 per cent of all

unions established. Until the post-Independence era at least one-third

of initial marriages of high-caste women were unmarked by ceremonies

simply because of the expense and effort involved. Marriages without

ceremonies still occur, but less frequently. Thus, among the residents

of Sirkanda no more than 50 per cent of current high-caste unions were

marked by wedding ceremonies. Among low castes the proportion is

considerably smaller. Among the 50 per cent of all marriages which

were marked by ceremony, there have been three major varieties of

ceremony: (1) the traditional Pahari wedding (bid), a manifestation of

regional tradition, (2) the traditional Hindu wedding of the plains

modified by bride price and related practices, called tako-ka bid

(money marriage) or paisd (money) wedding, (3) the traditional plains

wedding complete with dowry, called kanniaddn (daughter charity) or

pun (gift) wedding. The first of these was the only type of wedding

ceremony performed in Sirkanda until five or ten years before Indian

Independence, but it has since been dropped altogether. In it, as in

all Pahari weddings but unlike plains weddings, the groom's family

takes the initiative in arranging the marriage. A unique feature of the

traditional Pahari marriage is that the final ceremonies are conducted

at the home of the groom. In the other forms of marriage in this area,

as in the plains, the ceremonies occur in the bride's house.

128 CALENDRICAL AND LIFE-CYCLE CEREMONIES

The plains-type dowry marriage has long occurred among educated
and wealthy Paharis of neighboring areas, but it has occurred in Sir-
kanda only since shortly before Independence. The second type, a
syncretism, is an equally recent innovation and is currently the most
frequently used. The latter two types, and particularly the orthodox
plains Hindu dowry marriage, are indicative of the trend toward emu-
lation of plains culture (Sanskritization) advocated by educated
Paharis.2 The percentage of initial marriages unmarked by ceremony
has dropped in recent years, but has not disappeared.

At weddings a considerable number of interested people participate
— in preparation, caring for guests, assisting the ceremonial cooks and
the priest, assisting the principals in the wedding, and so on. The
people involved in order of decreasing intensity of involvement are:
the extended-family members, the lineage, the kindred (including
people outside the lineage who are recognized as relatives of the bride
or groom), and often friends, especially sib-fellows and caste-fellows in
the village. The kindred comes nearer to visibility as a group at this
time than at any other.

DOWRY AND BRIDE PRICE

The traditional form of Pahari marriage involves a bride price;
that is, the family of the groom pays an agreed-upon sum in cash to the
family of the bride. In the early 1900's this amount was generally under
Rs. 50; by World War II it had risen to between Rs. 200 and 400; and
now it varies from Rs. 500 to 1,500 with an average around Rs. 1,000.
In addition to the bride price the groom's family spends around Rs.
1,000 on food and other materials for a good-sized wedding, Rs. 750 on
jewelry and other gifts, and Rs. 250 for the services of specialists and
other miscellaneous expenses. The bride's family spends an equal
amount on entertaining guests (about Rs. 1,000) and perhaps another
Rs. 500 on clothing, dowry, gifts, and so on. Therefore, the bride price
is used in the marriage by the bride's family. Rarely is any profit
realized. If no bride price is given, as is the case with some well-to-do
families, the expenses on both sides are about equal. A dowry is nearly
always given by the bride's family to the couple, but in bride-price
marriages it is a token gift only.

Everyone knows that bride-price marriage is contrary to high-caste
plains custom.3 Therefore there is considerable striving toward dowry
marriages on the part of some of those who can afford it. The Brahmin
who performs most high-caste ceremonies is a vociferous advocate of
such marriages. Prestige is the goal of these ceremonies. Statements
such as the following from a well-to-do Rajput are common: "We

CALENDRICAL AND LIFE-CYCLE CEREMONIES 129

don't like bride-price marriage. How can you sell a daughter you love?"
Or another: "You can't enjoy money you receive for a daughter."
Usually this is merely repetition of platitudes, as both the men quoted
had given and received bride price in their day. On the other hand,
there are many families who will not give or take a bride without pay-
ment and are proud of it. As one Rajput elder said, "It is not proper
for Rajputs to accept charity. Only a Brahmin can do that. When we
take a girl, we pay for her." There is also a widespread belief that
dowry brides do not live nor produce children. It is only in the past
fifteen years that a real trend toward dowry marriages has begun even
among families who can afford them. Still at least 80 per cent of all
marriages include a bride price. Sometimes when no bride price is
given, the groom's family promises to bear all the wedding expenses. If
so, it is a case of the bride's family trying to have the cake and eat it
too. In at least one recent situation a dowry marriage was held but a
bride price was demanded later.

In former days it was common among all castes to effect a bride-price
transaction without ceremony of any kind beyond agreement between
the families. Meals might be exchanged at the time of agreement and
of bringing the bride. A puja to the household god was frequently in-
cluded. Thus the expenses of marriage ceremonies were avoided. The
practice still exists, especially among the lower castes. Joshi cites a legal
case in which validity of a Pahari marriage was sought to be proved
on grounds that the bride was delivered to her husband, those who
brought her were feasted, and a puja was performed. Although this
failed in the courts where Pahari custom was unrecognized, Joshi
asserts that even these three acts are not required for a valid marriage
under Khasa customary law. "The payment of bride-price and formal
entry as wife in the husband's house are enough for the purpose"
(Joshi, 1929, p. 40).

SECOND MARRIAGE

Some families of all castes now have Sanskritized marriage rituals.
Second or subsequent marriages of women, and often of men, have
always been effected without ceremony. Usually a divorced woman's
parents are then paid the bride price and they pass it on to her former
husband, though occasionally direct payment is made to the husband.
In the case of widow remarriage, a universal practice in this area, pay-
ment is generally made to the first husband's family if the widow
marries outside it, unless she elopes.

Second marriages for men may be carried out with ceremonies just
like the first if it is the bride's first marriage and especially if the man

l^O CALENDRICAL AND LIFE-CYCLE CEREMONIES

is young and has no children. Such marriages are most often bride-
price marriages because a girl's family who wants a big dowry marriage
will usually find an unmarried husband for their daughter. Moreover,
many second marriages are contracted to produce children, and a bride
price is considered more likely to bring that result.

Elopement and inheritance of a wife from a brother are both pub-
licly recognized forms of marriage wherein ceremony and payment are
avoided, though payment is exacted by the aggrieved husband in an
elopement if possible. Poor families of all castes often effect marriage
without ceremony or payment simply by common agreement.

Divorce proceedings will be discussed in chapter 5.

Sacred Thread

Rajput and Brahmin males are initiated into adulthood with the
ceremony of investiture with the sacred thread (jainu or bartbhandan)
indicating their spiritual rebirth (cf. Atkinson, 1884b, pp. 92 ff.). They
are then said to be "twice born" and therefore are possessed of the
knowledge necessary to an adult man of high caste. The ceremony is
evidently of relatively recent origin in these hills, adopted along with
Sanskritic marriage forms in an effort toward plains emulation or
Sanskritization. Normally this ceremony takes place in the groom's
village as part of the pre-marriage ceremonies. If a man is not married
and is an adult, or if he has not gone through a marriage ceremony,
he may call for the sacred thread ceremony at any time after he is about
20 years old.

Death Ceremonies

Ceremonies surrounding death are as complex as those surround-
ing marriage (see Appendix IIIC). It is believed in Sirkanda, as in
Hindu India generally, that a person's life after death will be greatly
influenced by the amount of charity that he has given in this life and
that is given in the ceremonies associated with his death. Specifically
it is thought that only what has been given in life and at these cere-
monies will be available to the spirit for livelihood in the afterlife.
Sirkanda villagers tell an eye-witness account of a woman who came
back to life shortly after death and begged that more charity be given,
since she had found that there was nothing for her to live on and no
comforts in the other world because of her niggardliness in life. There-
fore a person who has no descendants upon whom he can depend to
carry out the death ceremonies properly will usually sponsor pujas and
give the necessary charity in advance. This charity is given to the tra-
ditional village Brahmin or purohit, although in Sirkanda the cere-

CALENDRICAL AND LIFE-CYCLE CEREMONIES 131

monies are performed by more learned Brahmins. The persons in-
volved in the mourning and ritual pollution which follows a death are
primarily the male lineage members — those who trace descent to a
common ancestor — and to a lesser extent the local sib members and
those closely associated with them, that is, the clan, for wives of these
people play a minor role. The eldest son or an equivalent male relative
(sometimes an adopted son) performs the duties of chief mourner.
During the ceremonies the Brahmin performs pujas, directs the chief
mourner in his duties, and receives charity. The barber provides leaf
packets used in the ceremony, prepares the pipe for the Brahmin's use,
and prepares the leaf plates for the feast at the end of the mourning
period.

Widows are not restricted in activity or dress, as they often are on
the plains. An unremarried widow does not, however, take an active
part in birth or marriage ceremonies. Also, she does not wear gold
nose ornaments (the symbols of marriage) for at least a year after her
husband's death or until she remarries, whichever comes first. Widow
remarriage is universal here except when the widow is elderly.

If a villager dies far from home the postdeath ceremonies are per-
formed in his village home. A married woman's death is observed only
in the village of her husband and by the family of her husband. Her
family of origin takes no part in it except as observers if they are in the
same village.

Children, and often unmarried adolescents, who die are buried
without ceremony in unmarked graves near the village. Strangers who
die in the vicinity are usually buried rather than cremated so that
proof of death can be established if necessary later. Adults who die of
epidemic diseases are buried and then exhumed and cremated after
the epidemic has passed, usually three to six months later.

Kin, Caste, and Community in Religion and Ceremonies

The household, comprising the joint family, is the most significant
social unit in religious matters in Sirkanda.4 Deities and other super-
natural beings generally attack the household and must be dealt with
by its members collectively. This is true even when only a single mem-
ber feels the wrath of the gods, since they remain with the household
once they have entered it. The worship of household gods is the most
frequent expression of religious behavior in the village. Life-cycle rites
are likewise observed in the household unit, generally with participa-
tion by local sib fellows, their spouses, and adopted sons (that is, the
clan); other villagers are observers or guests at best. Local lineage is

132 CALENDRICAL AND LIFE-CYCLE CEREMONIES

most significant in death ceremonies for a man, and the local clan runs a
close second. For a woman the corresponding groups are the husband's
local lineage and the clan. The kindred looms large in marriage. Most
annual festivals are performed in household groups. There are special
ceremonies for the well-being of the household. Lineage, clan, kindred,
and sib are significant in such household ceremonies in that order.

No specific caste deities are worshiped in this village, although cer-
tain ones tend in fact to be associated with certain castes. Household
observances, whether worship or life-cycle rites, remain within the
caste in that those primarily concerned are relatives, and all relatives
are caste fellows. For castes separated by little social or ritual distance,
attendance at ceremonies extends across caste boundaries, as, for ex-
ample, between Rajputs and Brahmins, or between barbers, black-
smiths, and carpenters. Members of ritually and socially more distant
castes are not often explicitly barred from attendance, but they are
barred from participation in many activities, such as group eating,
which are central to the celebration and which would be polluting to,
or polluted by, these castes. Usually there is no desire for cross-caste
participation in such ceremonies, as was made explicit by a low-caste
informant who commented that, "There is no enjoyment for me at a
Rajput wedding. We can only have fun at weddings in our own caste."

There are no caste festivals or holidays among the castes represented
in Sirkanda. Caste enters the religious and ceremonial picture primarily
in the caste-specific roles of participants, as in the barber's duties in life-
cycle rites, the Bajgi's role in worship of deities and certain annual
and life-cycle ceremonies, the carpenter's ritual function in blessing
the doorways of a new house, and the Brahmin's priestly role.

Caste differences in religious belief and usage are virtually absent —
a notable contrast to the situation reported for other parts of India.
Where they do occur in Sirkanda, they represent differences in edu-
cation, wealth, and enforced restrictions on low-caste behavior more
than differences in beliefs or aspirations. Educated people know more
of the Sanskritic tradition than do others, and they tend to be of the
high castes, but uneducated Brahmins and Rajputs differ little from
Doms in their knowledge and beliefs. Even among Brahmins there is
little concern with ritual observance or daily worship. They maintain
ritual distance from lower castes as do Rajputs and some of the low
castes, but their ritual life does not differ significantly from that of
other castes. Diet of high-caste members and Doms differs not at all
except as dictated by differential income. Bathing is as foreign to most
Brahmins as to Bajgis.

The lack of caste differential in such matters is apparently attribu-

CALENDRICAL AND LIFE-CYCLE CEREMONIES 133

table in part to the frequent and intense interaction among members
of all castes in Sirkanda. In such villages the entire population is largely
isolated from frequent contact with people other than their fellow
villagers. Since most castes are very small in numbers, their daily
social interaction must be largely with members of other castes. No
caste remains isolated from any other. The various castes have more in
common and interact more frequently than would groups separated
by greater social barriers or with caste communities of their own, and
consequently they have less opportunity to maintain differences in
beliefs and practices (cf. Berreman, 1960b).

The community is the unit which observes some annual festivals,
notably Diwali. It is also the traditional unit for observance of agri-
cultural ceremonies and worship of the village gods. Two specifically
village-directed ceremonies are the Mundkile village-protection cere-
mony and the rope-sliding event to alleviate village and regional
troubles.

Supervillage observances are limited to annual regional fairs and
the rope-sliding ceremony. In Bhatbair the one such fair is evidently
secular in nature, while those in larger centers often have religious
meaning.

Religion is thus an important feature at every level in Sirkanda. Its
place is similar to that of religion in other Indian villages. Opler's
summary statement applies as well to Sirkanda as it does to Senapur,
the plains village of which it was written:

To live a very full and estimable life, a villager has to participate in the
religious round. Religion justifies the existence of his line, the tie between
his ancestors and his sons. It holds his kin together in family rituals. It
provides travel, adventure, and new experience and connects his village with
others. . . . The presence of the protective godlings of the village strengthens
group consciousness. The agricultural rites, the worship of the disease god-
desses, and the life-cycle ceremonies awake courage and hope in areas of life
where uncertainty and anxiety are most prevalent. (Opler, 1959b, p. 226)

Brahmins and Shamans

One of the most striking features of traditional religious organiza-
tion in Sirkanda and vicinity is the strategic importance of the shaman
(baki) (cf. Berreman, 1961a). In the sphere of religion he is a "cultural
policy maker," to use Singer's term (Singer, 1955, p. 30) — a cultural
"gatekeeper." He is a man, who may be of any caste, whose prestige and
power are dependent upon success in his practice of diagnosing super-
naturally caused difficulties through the use of a personal deity who

134 CALENDRICAL AND LIFE-CYCLE CEREMONIES

provides the insights and information, using the practitioner as a ve-
hicle. He is the key man in virtually every instance of traditional re-
ligious worship. He determines which supernatural being is to be
worshiped or placated, be it household, village, regional, or all-India
Hindu god, or be it ancestor, ghost, spirit, or witch. He often determines
which puja will be performed, which sacrifices will be offered, which
pilgrimages undertaken, which new gods will be worshiped, and, in the
long run, which ones will fall by the wayside. Styles and fads of wor-
ship, means of correcting troubles and treating disease, are largely in
the shaman's hands. Certain lesser practitioners share a part of these
important functions. To be sure, such people work within their cul-
ture and are probably largely unaware of their key importance. They
may well be unaware of the extent to which they could manipulate
their clients if they wished. This does not alter the fact of their crucial
role. Traditionally they have undoubtedly been the most important
individuals in influencing the nature of religious practice.

Brahmins have played an important but less significant religious
role from the point of view of dynamics of culture in the traditional
setting. They operate primarily within rigidly prescribed limits of
ritual with little opportunity for initiative or innovation despite their
high status. When they are called and what they will do are often
prescribed by a shaman. They are, in fact, ritual technicians or engi-
neers. The shaman makes most of the decisions and therefore is the
policy maker.

Plains emulation has been effectively advocated by some Pahari
Brahmins, but these advocates have been priests with a plains orienta-
tion seeking to establish or enlarge their clientele, not traditional fam-
ily priests or purohits whose clientele is fixed and assured. It seems
likely that the change would come about more quickly in this area if
it were advocated by shamans. Shamans could diagnose difficulties of
all types as being attributable to failure to conform to plains religious
and social standards. They could presumably prescribe Sanskritic
worship of great traditional gods in many circumstances and people
would be likely to follow their advice.

However, shamans, being uneducated and with a stake in the tra-
ditional Pahari religious system, have remained traditional in their
attitudes and consequently have impeded plains emulation in the
religious sphere. They have nothing to gain and everything to lose by
emulation of plains Hindu orthodoxy, for that orthodoxy undermines
their religious roles. They therefore generally use their considerable
influence to encourage adherence to traditional religious usages and to
stave off plains emulation.

CALENDRICAL AND LIFE-CYCLE CEREMONIES 1 35

Family priests (purohits), with their assured clientele and their
status in the traditional Pahari religious system are in much the same
position. They, too, stand to lose by plains emulation, since villagers
would come to depend upon other priests more learned in plains
Hinduism. Plains emulation, therefore, threatens the purohit, as it
does the shaman, with loss of religious importance and ultimately of
income. For other Brahmin priests, however, the reverse is evidently
true; it is a means to economic and status enhancement (cf. Berreman,
1961a). Plains emulation offers new clients and prestige to priests who
have few clients or no traditional clientele and who have learned plains
Hinduism.

Increasing contacts with plains people have led many Paharis —
especially those with education, and these are mostly from the high
castes — to adopt high-caste people of the plains as a reference group
in many contexts. With this aspiration for plains emulation they come
to demand services which the purohit and shaman cannot provide.
Sophisticated, ambitious priests actively compete for clients in this
sphere, whereas they could not — at least not so overtly — in the tra-
ditional situation. As a result, in recent times some Pahari Brahmins
have been cast in the role of religious innovators or policy makers.
Relative to them, shamans and purohits have been religious conserva-
tives. The effect of the changes advocated by these enterprising atradi-
tional Brahmins in any given locality is more spectacular and perhaps
of more fundamental structural significance in Pahari religion than
that effected by shamans in the traditional setting over a comparable
time span. It involves new religious and social conceptions, whereas
the influence of shamans has been felt primarily in introducing varia-
tions on traditional Pahari religious themes and in invoking sanctions
on individuals and groups. The sanctions can even be invoked against
Brahmin priests, who have traditionally consulted shamans as avidly
as anyone else. It can be argued that through use of such sanctions
shamans in the long run have had the potential for structural effect
as great as that of atraditional Brahmins. It can also be argued that
priests who advocate plains emulation are often merely reflecting or
accommodating to a trend among sophisticated Paharis rather than
initiating it.

That shamans have not become a powerful elite in their own right
is perhaps partly attributable to the openness of their profession. Being
neither hereditary nor caste-bound, it is highly competitive. There are
many such practitioners and they are approached by their clients on a
very pragmatic basis, success being the criterion for patronage. There
is no traditional practitioner-client arrangement comparable to the

136 CALENDRICAL AND LIFE-CYCLE CEREMONIES

jajmani system. Thus the clientele shifts easily, public opinion is a
limiting factor, and no shaman has a monopoly on the market. Sha-
mans do have in common their opposition to plains emulation.

Competition among atraditional Brahmin priests is tempered by
jajmani-like loyalties on the part of their clients, whom they in fact
refer to as "jajmans." Such Brahmins share a vested interest in the
advocacy of plains emulation.

Plains-Pahari Religious Differences

In discussing religious life in Sirkanda emphasis has been placed
on its essential similarity to that found in other Indian villages, and
particularly those of the Gangetic plain. Implicit throughout, however,
has been an emphasis upon differences, upon the unique aspects of
Hinduism as it is practiced in Sirkanda. Such differences have been,
for the most part, differences common to the region — differences which
help define the Sub-Himalayan region as a distinct culture area, that
define the Central Pahari-speaking peoples as a subcultural group,
that distinguish the residents of Tehri Garhwal or even Bhatbair from
their neighbors. The differences have been most frequently differences
of emphasis rather than differences of kind. Paharis do not do many
things that are unknown on the plains nor are many plains practices
entirely foreign to them, but they do more of some things and less of
others and that is where the real difference lies.

As has been noted above, perhaps the most striking religious differ-
ences are those relating to caste. In Sirkanda there is remarkably little
variation in religious belief and practice from caste to caste, whereas
in the plains caste differences are prominent features of village religion.
In Sirkanda, Brahmins and Bajgis carry out very similar rituals, ob-
serve the same festivals, and express virtually the same beliefs. They
have distinct ritual functions and are characterized by the expectable
differences in ritual purity, but they do not possess significantly dif-
ferent religious subcultures.

The Brahmin perhaps plays a less crucial role in religion here than
on the plains. There are fewer events requiring his presence. There is
apparently a more important and complex system of non-Brahmanical
practitioners — of shamans, diviners, pujaris, exorcists, curers, and so
on, than among most plains groups. The services of these practitioners
are available to, and in demand by, all castes, whereas in the plains
they are often caste-specific (cf. P. M. Mahar, 1957).

There is a casualness about matters of ritual purity, marriage regu-

CALENDRICAL AND LIFE-CYCLE CEREMONIES 137

lations, and similar religion-based social practices, which is not char-
acteristic of high-caste plains groups.

Gods of the literate tradition of Hinduism are less widely recognized
and less honored in this area than in the plains. Similarly there is less
observance of the ceremonies and festivals of all-India spread than is
common in the plains. Rituals that are observed are adapted to local
convenience and taste to a degree perhaps exceeding that in most areas.
This is especially noticeable in the celebration of Diwali and in mar-
riage ceremonies. There is virtually no daily or private worship. How-
ever, the people have a lively interest in religion and do not lack for
ceremonies arranged to appease their traditional gods. Worship is
aimed at controlling powerful and dangerous supernatural beings who
plague them. Such worship is a group matter rather than an individ-
ual one — a family or village undertaking. Probably the most distinctive
aspect of worship is the incorporation of animal sacrifice as an integral
part of virtually every ceremony. The life of an animal is required to
please Pahari gods. The gods must also be given the opportunity to
dance in the bodies of their devotees if they are to remain favorably
disposed.

In a low-caste plains community such beliefs and practices would
not seem unusual. The striking feature is that Sirkanda and other
Pahari villages are predominately high-caste communities which
closely resemble low-caste communities of other areas in the religious
life of their members. It is for this reason that Paharis are considered
ritually inferior by their plains-dwelling caste-fellows. And it is for
this reason that plains emulation or Sanskritization is becoming in-
creasingly evident among informed Paharis, who more and more fre-
quently come into contact with critical plainsmen in positions of
authority or influence.

Sirkanda and the Great Tradition

The discussion of religion has demonstrated that Sirkanda is within
the range of variation of other Hindu villages in the practice of Hindu-
ism. While it has perhaps less in the way of all-India Hindu-
ism than many Indian villages — especially in comparison to other
largely high-caste villages of the plains of northern India — it is not
atypical in this respect in comparison to other Pahari villages. This
leads to the observation that, while elements of the all-India Hindu
tradition as represented in sacred writings and India-wide precepts
and practices are prominent enough in the village to make it recog-

138 CALENDRICAL AND LIFE-CYCLE CEREMONIES

nizably and undeniably a part of that tradition, a large proportion of
village religious traits do not fit into that tradition. To suppose that
these other traits are purely local would be, however, fallacious. Many
are characteristic of villages throughout all or much of Hindu India.
Others are characteristic of Pahari villages. "Regional Hinduism" is
perhaps the best term to apply to those traits of less than all-India
spread, since "local Hinduism" is likely to imply that the traits re-
ferred to are unique to the village or small locality, when in reality
they are to be found over a fairly large cultural area (cf. Cohn and
Marriott, 1958).5 Finally, there are purely local elements — if not in
pattern at least in specific content. Such terms as "local," "regional,"
and "national" Hinduism are merely labels along a continuum. There
are no sharp boundaries and the continuum is a moving one.

The terms "great tradition" and "little tradition" present further
difficulties. They imply considerably more than simply geographical,
cultural, or social "spread." The "great tradition" generally denotes
the "literate religious tradition" (Marriott, 1955a, p. 191). However, it
is often used in the context of "all-India Hinduism" in discussions of
Indian religion. Conversely, the "little tradition" is generally implicitly
defined as the nonliterate, vernacular religious tradition but is often
used as though it meant local Hinduism. In India these are two entirely
different dimensions. In village India it is possible to identify literate
religious traditions based upon the great religious writings and em-
bodying the philosophical foundations of the religion. These can be
contrasted, as Marriott has done, with the folk practices and beliefs
which have not (or not yet) been incorporated into the literate tradi-
tion. However, the nonliterate tradition includes many elements which
are as widespread geographically as elements of the literate tradition.
Some of these folk elements may be more widespread than many literate
ones in terms of the population which understands or practices them.
In discussing traits of village Hinduism it is useful, therefore, to discuss
spread in terms of geography, culture, social organization, and so on,
as well as to attempt to apply the literate-nonliterate or great-little tra-
dition dichotomy.

It should be clear that Sirkanda is not unique in its deviation from
the literate tradition as evidenced in the religious beliefs and practices
of its members. Researchers have found similar circumstances in other
Indian villages (for example, Opler, 1958, pp. 553 f.).

Historically, it is suggested by many writers from the time of Manu,
the Hindu lawgiver, to the present, that Paharis of the high castes
(that is, Khasas) came from a culture that was once more highly San-
skritized than at present, but which fell away from the practices and

CALENDRICAL AND LIFE-CYCLE CEREMONIES 139

beliefs that define this term. The decline is attributed to a combina-
tion of the exigencies of life in the hills and the intimate contact which
the Khasas are supposed to have had with the subjugated indigenes
and their "inferior" religion. It is interesting to note the extent to
which local historians credit the lowly Doms, who were presumably
so readily overpowered by the Khasas, with having altered or even
revolutionized the way of life and religion of these allegedly pristine
Hindus. Another view is that these high-caste hill people represent a
relatively untouched and unchanged survival of antediluvian Aryan
culture. It seems probable that in reality the Paharis are the product
of mixture between the early Khasas and Doms (if indeed they were
different cultural groups) and more recent immigrants. Pahari religion,
like the rest of their culture, is the product of gradual change in an
area of relative isolation where they have been out of direct contact
with many of the influences important in shaping Hinduism on the
plains. Beliefs and practices have been influenced by their cultural
heritage and by contacts with peoples of the plains and the higher
Himalayas.

There is an increasing trend toward religious change on the model
of high-caste plains Hinduism at all caste levels in and around Sir-
kanda. There has long been awareness of plains ways through contacts
with educated Brahmins, merchants, government officials, and others
knowledgeable in the great traditions of Hinduism. Recently with
improved means of communication, increased movement of people
between the hills and plains, more easily available schooling, and in-
creased financial capabilities, this awareness has increased, resulting
in an active tendency toward emulation. The motivation is simply to
be respected by plains people, for Paharis increasingly feel the effects
of their unorthodox religious and social practices as a result of their
increasing contacts with people adhering to plains values, especially
with people in positions of authority or influence. Paharis are con-
sidered to be rustic, degraded Hindus by most plainsmen. Plains
Brahmins and Rajputs often reject the caste status claims of their
Pahari caste-fellows, largely because of their unorthodoxy. By adopting
some of the symbols of plains culture, Paharis hope to improve their
status in the eyes of the plains people. Instrumental in this change have
been those Paharis who have had the most extensive and intensive ex-
posure to orthodox plains viewpoints and who feel their status most
threatened by invidious comparisons. These are the high castes and
especially the Brahmins. They have been active in effecting the most
obvious changes in this trend: Sanskritization of marriage customs,
adoption of other Sanskritic rituals and festivals, and attribution of

140 CALENDRICAL AND LIFE-CYCLE CEREMONIES

Sanskritic interpretations to traditional Pahari religious beliefs and
practices. The trend is similar to the process of Sanskritization or status
emulation widely documented among tribal and low-caste groups else-
where in India.

Few Paharis have adopted the modern values and ideas associated
with education, urban living, and cash economy that have been adopted
by some plains people — especially those of the higher castes — and that
have been termed "Westernization" by Srinivas (1956; cf. Cohn, 1955).
The reasons are very similar to those which account for the fact that,
when high castes adopt modern aspirations and values, low-caste plains
groups adopt Sanskritic behaviors formerly denied them rather than
following the high castes in the new life. These groups have differential
experience and different reference groups — in short, different sources
of values and aspirations. High-caste people of the plains have bene-
fited from the results of increased educational and employment oppor-
tunities in the new context, while Paharis, like low castes of the plains,
have benefited less. In fact, the latter groups have had little opportunity
even to learn the modern, Western culture within which such benefits
can be realized. It is a basic axiom of reference group theory that, if
one group is to identify with another so that its members adopt the
outlook, and judge themselves by the standards, of the other group,
there must be both knowledge of the identification group and some
perceived similarity or equivalence to it (Merton and Kitt, 1950, p. 61).
Among many high-caste plains people, travel, education, and employ-
ment have resulted in familiarity with modern ways and competence in
them which make possible perceived similarities and subsequent iden-
tification with extra-traditional groups. Adoption of new values and
aspirations to the relative neglect of traditional Sanskritic values and
aspirations has followed.

Meanwhile Paharis, like low-caste groups of the plains, have lacked
equivalent facilities and experience for acquiring the knowledge and
competence prerequisite to adoption of nontraditional reference groups.
In Sirkanda cultural, physical, and intellectual isolation have in all
castes militated against adoption of new alien reference groups to any
significant extent. Plains people have long served as something of a
reference group for Paharis and have long been known to them. In-
creased contact has resulted in increased knowledge of their view-
points. This, combined with increased wealth, has enabled Paharis to
push toward higher status in their eyes by emulating them. Since Pahari
experience has been for the most part with plains people who advocate
a traditional Hindu world view, emulation has been in this direction.
For low-caste plains people the enabling feature for upward mobility

CALENDRICAL AND LIFE-CYCLE CEREMONIES 141

has been decreased downward pressure from the high castes, as well as
increased financial capability and social justice. For both the Paharis
and the plains low castes, it is the high castes of the plains in their
traditional role who form the reference groups, for they are sufficiently
known, understood, and envied for the other groups to identify with
them and seek to emulate them. From the Pahari point of view, modern,
urban, or Westernized society is simply too alien to be emulated. Tra-
ditional plains culture is a familiar and increasingly attainable ref-
erence point.

There has been little caste differential in the "plainsward mobility"
of Paharis. This is partly because of the cross-caste cultural homogene-
ity of Paharis of this area and the fact that there has been relatively
little difference from caste to caste in opportunity to acquire outside
reference groups. High castes have had only slightly better opportu-
nities for education and close contact with educated Brahmins than
have low castes. Low castes have had as much or more per capita out-
side contact as they travel to trade and procure the tools and materials
of their trades. High castes have had significantly greater financial
capability to carry out Sanskritization by performing expensive rituals
under the supervision of educated Brahmins, and it is in this respect
that high and low castes differ most. But due to the intensive nature
of intercaste interaction in the Pahari context, low castes have been
fully aware of the practices of their caste superiors and follow them
insofar as they are financially able. It could perhaps be asserted that
low castes emulate their high-caste neighbors while the latter are emu-
lating a plains model. If so, the time lag is so short as to be unnotice-
able; it is as if both were emulating the plains model. Both groups are
explicit in attributing their practices to the plains model.

Another factor in the cross-caste nature of Pahari Sanskritization
may be the fact that high castes have not felt a threat of imminently
successful low-caste upward mobility. They have therefore not felt the
necessity of seeking Westernization as an alternative source of hier-
archical supremacy, a motivation for Westernization suggested by
Gould (1961a).

Therefore, while the picture of changing caste status on the plains
is often one of the low castes moving up in the Sanskritic caste-status
hierarchy as the higher, more advantaged castes move out of this hier-
archy into a non-Sanskritic milieu, in Sirkanda the picture is of the
entire Pahari community attempting to move toward the ways of their
plains reference group while castes within the community retain their
relative status positions. This is not to deny some cases of new non-
Sanskritic aspirations in Sirkanda. Neither is it to deny ambitions of

I«li!

142 CALENDRICAL AND LIFE-CYCLE CEREMONIES

upward mobility and competitive scrambling for status among the
lower castes in this area. These occur, for example, among the Bajgis,
whose women have given up dancing for the public in order to raise
their status, and among the Bedas, who, in the present generation,
would like to follow suit. It can be asserted, however, that the dominant
trend is a society-wide movement toward what is viewed as the re-
ligious context of the plains in order to win the respect of members of
that dominant group.

KIN GROUPS AND KINSHIP

Previous chapters have described the relationship of men to their
natural environment in Sirkanda and to the supernatural world which
impinges upon them. This and the following three chapters will de-
scribe and analyze the interrelationships among people in and around
Sirkanda. An attempt will be made to describe and distinguish ideal
patterns of social behavior as expressed by informants and actual pat-
terns of behavior as observed and reported. Similarly, an attempt will
be made to distinguish between what is believed and done, and what
people would like outsiders to think is believed and done (cf. Berre-
man, 1962c).

Here, as in preceding chapters, the way of life of the residents of
this village and the culture area it represents is compared with that of
villages in other parts of India.

From the point of view of social life, the whole of the cis-Himalayan
region behaves as a culture area, as there is a homogeneous social code to
which both the higher and lower groups subscribe. . . . But the hill culture
differs from that of the plains and all cultures that surround them. . . .
(Majumdar, 1944, p. 139)

It will become evident that basic similarities accompany important
differences, and it is essential that these be presented for comparative
purposes. A basic aim, throughout this account, is to throw light on
the role of kin groups, castes, and community organization in the lives
of the people under discussion.

144 KIN GROUPS AND KINSHIP

Basically Sirkanda shares the social structure of Hindu Indian so-
ciety. Family, caste, and community are the most significant social
units.

The discussion will begin with the extended family or household,
the economically cooperating residential kin group and the most inti-
mate, immediate social unit in Sirkanda. From the residential kin
group the discussion will proceed to its extensions in consanguineal and
affinal kin groups, to castes, and finally to the community and its ex-
tensions. It is impossible to understand the functioning of Sirkanda
society at any of these levels without reference to the others. The
system is an interdependent one which does not operate on independ-
ent subcircuits. Choice of family as a starting point for presentation is,
therefore, largely arbitrary and perhaps stems from an inductive bias.

A further qualification should be inserted. In Sirkanda, as in most
Indian villages, there are differences in internal social organization
among various castes. They are less in the Pahari area than in many
others and by comparison they may seem almost insignificant, but
they do exist. Since 87 per cent of the population of Sirkanda is of one
caste — Rajputs — more reliable data were obtained regarding that caste
than any other. However, caste differences were a special focus of the
research and were sought out whenever possible. Such differences as
were revealed have been reported here. Differences not mentioned did
not exist or were not obtained in investigation aimed at detecting
them.

Family

As the Wisers (1951, p. 160) have commented, "no villager thinks
of himself apart from his family." Ideally the basic residential, social,
religious, and economic unit in Sirkanda is the patrilocal extended
family. This consists of a man, his wife, his sons, and their wives and
children plus any unmarried daughters. In Sirkanda this unit is sup-
posed to occupy a house, preferably with a separate sleeping room for
each nuclear family consisting of man, wife, and children. It is an
economic unit which includes fathers, sons, and brothers and their
wives and unmarried daughters, and therefore may be termed a "joint
family." All members share in the family occupation and in the product
obtained. All eat from the same hearth — a distinguishing feature rec-
ognized by the people themselves, who refer to this family unit or
household as the chula (cooking hearth). The eldest active male is the
family head and bears final responsibility and authority for family
well-being. The wife of the eldest male, whether he is living or not, is
the head of the female component (wives and unmarried daughters) of

KIN GROUPS AND KINSHIP 145

the household in domestic matters. She becomes the titular head of the
household upon her husband's death if there are no brothers to take
over, but generally a son acts in her stead. Within the component nu-
clear families the age and sex hierarchy is the same.

I. Largest household in Sirkanda (Rajput): Four dwellings

chan

---*"" "r^ 7^1 ix\

0=A \ . u / 0=A=0 0=A \

\ chan chan / , ,

f 1 ' 1 — 1 — w — n-/ 1 \ \ n /

lp=A O 0=A 0=A \|0=A fO=Al \ O A /

I I H / / ^_ ^

rn 1 — I — 1 1
^^ o A a o a;

\
\
v.

E. Medium (Rajput): Single dwelling m. Small (Rajput): Single dwelling

^ I v / I I

/ r Lr i 1 o=A/

/ 0=A 0=A A/ \ /

\0 A O A /

12. Atypical (Rajput): Two dwellings

I T^H

•=A ,-" 0=A chan

/ \\ s' ~^

•=A/ 0 = A-0 (A) Hired servant

( A" ^ov^

^ec» / ^. ^y

Adeceased ancestor Aliving household member

Fig. 2. Household composition and residence. Residence in a single dwelling
is indicated by a broken line.

In Sirkanda there are several types of deviation from this ideal
pattern (see fig. 2). As has been mentioned previously, most households
(37 of 45, or 82 per cent) own houses or chans outside of Sirkanda.
Twenty-five of these joint families, 55 per cent of those in the village,
are continually or usually split into two or more residential units be-

146 KIN GROUPS AND KINSHIP

cause of regular occupancy of chans or second houses away from the
village. Therefore, in over half the joint families the basic family
economic unit is not a residential unit, as villagers say it ideally
should be. The extended family may become permanently divided,
and more distant households may sever connections with the village.
As has been noted, approximately one-fifth of the living adult males
who appear in Sirkanda genealogical materials do not appear as mem-
bers of Sirkanda households at present. They have severed village and
joint-family ties.

Between one-third and one-fourth of the family units (13 of 45, or
29 per cent) consists only of a nuclear family or a single individual.
Seven more consist only of a nuclear family and a parent of the hus-
band (a "minimal extended family"). In these 20 small family units
there are 99 individuals, while among the 25 larger family units are
the remaining 285 Sirkanda residents. Thus, over one-third of all
Sirkanda residents live in family units much smaller than the ideal.
Ten of the large units are fraternal joint families, consisting of brothers
and their dependents, while 15 include one or both of an elder man-
wife couple and their sons, daughters-in-law, unmarried daughters, and
grandchildren.

Household membership ranges from 1 to 25 individuals with an
average between 8 and 9. Thirteen households have 10 or over; thir-
teen have 5 or under. Household membership spans four generations in
1 case, three generations in 25 cases, two generations in 14 cases, and
one generation in 5 cases. In only 4 cases have real brothers divided
into separate households. In 1 case the only half-brothers currently
in the village without a living father did so, and in 2 cases father-son
divisions have occurred. In the latter instances the sons with their
fathers' approval took advantage of chances for adoption into heirless
families. Half-brothers quite consistently divide their father's prop-
erty upon his death and establish separate joint-family households.
Real brothers of the same mother rarely divide joint property, but
the relatively landless low castes do so more frequently than the high
castes. Of the four current cases, two are in low-caste families — the only
low-caste brothers, with father deceased, in Sirkanda — while only two
are in the high castes, of twelve possibilities in that group.

One of the most prominent differences between high and low castes
in Sirkanda is in joint family size and composition. Not only do low-
caste brothers usually divide their property when their father dies
whereas high-caste brothers retain it intact, but low-caste people char-
acteristically live in smaller joint family units. This is indicated in
table 2. The differences are perhaps attributable more to economic

KIN GROUPS AND KINSHIP 147

conditions than to caste. All three of the low-caste extended family
households are dependent in large part on agriculture, while none of
the low-caste nuclear households has land to till. Agriculture requires
many hands to make it productive, and it will feed many mouths. Two
low-caste brothers, each the head of a good-sized extended family, have
divided their lands and clients, but they perform one economic func-
tion jointly where teamwork is requisite, namely drumming. Thus it
seems that larger joint families are usually retained when it is ad-
vantageous or necessary to do so, and they break down more readily
when they perform no useful or necessary function.

TABLE 2

Sirkanda Population by Family Types

Households Population

Extended Nuclear Extended Nuclear

families families families families

High castes 30 9 306 40

Low castes 3 4 22 16

Division of a household generally takes place among first cousins
(brothers' sons). In no current cases do such first cousins belong to the
same household unless their parents are living. Thus, the division of
the extended family into smaller units occurs quite consistently after
the third generation; that is, brothers do not often divide their father's
property, but when the brothers die their offspring (first cousins) do
divide it. This conforms to the ideal pattern.

It is worth noting in this connection that in nine of the 13 nuclear
family household units there is no other nuclear family related to the
household head more closely than at the first cousin level, the level
at which division of the family is to be expected. Therefore, in only
four cases can the nuclear family be called a voluntarily or prematurely
segmented social and economic unit. Two of these households are
those of blacksmith brothers and one is that of a Rajput who moved
to his wife's house to acquire her inheritance. In the other nine cases,
segmentation has resulted from necessity brought about either by dif-
ferential birth and death rates or by the custom of dividing property
in the third generation. Likewise, only one of the seven minimal ex-
tended families is minimal by choice. One Rajput household head
expressed a common attitude:

These days in some of the families when boys become young men and are
married they think of breaking away from the joint family. In my family
my younger brother and I never thought of separating. Of course, now that

148 KIN GROUPS AND KINSHIP

our father is dead we don't know how things will take shape. But we like
the joint family system. I sometimes feel overburdened being in charge of
such a big family, but the household work, agricultural work, and tending
the cattle is done very smoothly as there are enough people to take care of
it all. If there are only two or three people in a family they just don't know
how to take care of all of the agricultural and other work. They don't know
where to leave their children when they go to work in the fields or to bring
firewood and grass.

In the above figures on household size and composition I have in-
cluded "polygamous families" with nuclear families. A polygamous
family is made up of more than one nuclear family linked by a com-
mon spouse (Murdock, 1949, p. 2). In Sirkanda these are polygynous
— multiple wives and a single husband — and generally with only two
wives. Murdock calls both extended and polygamous families "com-
posite" families, as distinguished from nuclear ones (Murdock, 1949,
pp. 23 ff.). I have lumped the polygynous and nuclear types simply
because in Sirkanda only a man-woman or man-women combination
is a family. A man and woman are potentially an independent family
unit; an additional wife is not. An additional wife does not alter the
family make-up in the same sense that another nuclear family does.
In any event, there are currently twelve polygynous families in Sir-
kanda, ten of which are included in larger joint families. Of these ten,
two have only a dependent parent in addition. The other two con-
stitute households in themselves. All are Rajputs and are therefore
landholding families. Seven of these families occupy two separate
houses, and five families occupy single houses. Size of the extended
family is apparently the most important factor influencing the living
arrangement in the polygynous families. A small family often has to
separate wives into different locations in order to occupy outlying
dwellings, whereas larger families send out an entire nuclear unit. In
all five cases in which the polygynous family constitutes virtually the
entire family unit, the co-wives and their children occupy separate
dwellings. In all five cases in which they occupy a single house the ex-
tended family is large enough to tend its chans with other nuclear
units. This leaves two cases in which wives are separated purely out of
choice — a frequently cited means of minimizing co-wife friction.

Marriage Regulations

The nuclear family is established as a result of marriage customs
described in the preceding chapter. Joshi summarizes the traits of
Khasa marriage which distinguish it from Brahamanical Hindu mar-

KIN GROUPS AND KINSHIP 149

riage: existence of levirate, marriage as a secular transaction involving
bride price, religious ceremony unessential for marriage, marriage
dissolvable by mutual consent, remarriage of widows and divorcees
recognized, sacred thread ceremony not deemed essential. These are
characteristics of traditional marriage in Sirkanda. The religious cere-
mony and sacred thread ceremony are gaining prominence with the
increasing trend toward Sanskritization (Joshi, 1929, pp. 50 f.). Here
I will use the term "marriage" very loosely to designate any instance
in which a man and woman live openly together so that any child born
of the woman will be acknowledged to be that of the man as well.

As has been indicated, the ideal marriage is one between a previously
unmarried man and woman with appropriate and compatible genea-
logical and astrological credentials. It is arranged by the parents of
the principals without their direct participation and at the initiative
of the groom's family.

In order to understand marriage regulations it will be necessary to
jump ahead of the discussion and comment briefly upon two impor-
tant social units more inclusive than the family. The largest of these
is the caste or jdt (jati in Hindi), which has frequently been mentioned
in earlier chapters and which functions much as it does over the rest
of India. It is the endogamous unit — the unit within which marriages
should always be contracted. In fact, its extent may be defined by the
extent of marriage ties. The other unit relevant to marriage is also
called jat or jati by villagers but is here termed sib (cf. Lowie, 1947,
p. 111; Murdock, 1949, p. 47). This is a subgroup of the caste and is
composed of assumed consanguineal relatives. It is exogamous — its
members are not potential mates.1 Neither, according to the rules
followed in this area, are people potential mates if their mothers (or,
ideally, other maternal ancestors for several generations back) are of
the same sib. Caste is virtually universal in India, while sibs occur
primarily among the higher castes. In Sirkanda the low castes claim
to have a sib structure, although their sibs are somewhat less consist-
ently and uniformly defined than those of the high castes. Another
feature of marriage in Sirkanda, as in most of India, is that it is nor-
mally patrilocal — the bride goes to live at the home of her husband
and his family.

In practice it is found that these rules are followed quite rigidly in
first marriages, arranged by the parents with the advice of Brahmins.
The rules are less consistently obeyed in subsequent marriages, which
are often informally contracted, and may even omit bride-price.

A total of 471 marriages were recorded in genealogical materials
collected in Sirkanda — 390 high-caste and 81 low-caste marriages. Of

150 KIN GROUPS AND KINSHIP

this number, 300 were marriages of village men (that is, sons of the
village) and 143 of these were marriages of men who are living and
currently identified as Sirkanda villagers.2 For most purposes the 300
total is most useful in that it is large and is associated with accurate
data. Data on village women's marriages, except those on location of
spouse's village, are not complete enough to warrant their use.

Multiple Marriages

Ninety-six of 300 unions, 32 per cent, were second or subsequent
marriages. These were almost evenly divided between polygynous
unions and nonpolygynous or sequential unions (that is, in which first
wife died or left before the second marriage). It is probable that in
this accounting the number of nonpolygynous plural marriages are
underreported because there is a tendency to overlook first wives, es-
pecially if they were childless and either left or died early in marriage.
This is indicated by the fact that in the figures for living Sirkanda men
such cases outnumber polygynous ones by one-third, while in the total
figures there is a slight preponderance of polygynous marriages. It
therefore seems likely that around 40 per cent of all marital unions are
second or subsequent ones for the man. Polygynous marriages consist-
ently form about 15 per cent of all unions.

Polygyny involving more than two wives is rare. Only four cases are
reported in the genealogical materials, although it is likely that some
cases stemming from inheritance of additional wives have gone un-
reported. Polygyny and nonpolygynous plural marriages have occurred
in all castes represented in Sirkanda. No significant caste differences
appear, but there is a tendency toward more polygyny among the high
castes than among the low castes. This tendency may relate to the
greater usefulness of extra wives in agricultural households than in
others and the fact that extra wives can evidently pay their way in
labor more effectively in agriculture than in the specialized occupa-
tions.

Since the people of the area are widely reputed in India to be polyg-
amous, some special inquiry was directed toward determining the
extent and nature of such marital arrangements. In the process some
detailed information about marriage in general was obtained. I have
already indicated that polyandry is not at all practiced here as it is in
neighboring Jaunsar-Bawar. In the discussion of intrafamily relations
it will become evident, however, that there is not as much difference
as might be expected in the sexual arrangements within the family in
these two areas (cf. Berreman, 1962a).

I have shown in the above discussion that polygynous unions are

KIN GROUPS AND KINSHIP 151

permitted and occur with a frequency of about 15 per cent. Twelve
established polygynous families and one de facto case on the verge of
public recognition were observed in some detail in Sirkanda. They were
the total current cases wherein all the principals were alive and par-
ticipating in the marriage. All were Rajputs.

Reasons given for polygyny in particular cases were four: (1) to
produce children when the first wife has been barren, (2) to help with
the work, (3) for "pleasure," (4) inheritance of an additional wife from
a deceased brother.

In eight cases of polygyny in Sirkanda the first two reasons were
given in combination. The work contributed by the second wife was
stressed in each of these cases, but sterility of the first wife was appar-
ently the primary motivation. In one of the cases the woman herself
asked that another wife be brought for these two reasons, and it was
her father who made the necessary arrangements to secure a distant
classificatory sister as her co-wife. In another case the husband was in-
clined to repeated marriages — he had had four previous wives — but
had produced no children and claimed to need two women at a time
to take care of the work.

One villager commented,

A wife is a valuable asset here. Here the wife takes care of her husband in
many ways, and she does much of the work of the household. Therefore, two
wives are better than one. In your country and in the plains the husband
has to support the woman, so a second wife is a hardship and a luxury.

Atkinson (1886, p. 255) says that in Garhwal "the custom probably
arose from the great difficulty there was in cultivating the large amount
of waste land available. Wives were procured to help in field
work. . . ."

In three cases in Sirkanda "pleasure" was given as the reason for
taking a second wife; that is, there was no question of need for more
workers or more children in the family. In two cases wives were in-
herited from elder brothers. In one of these cases a man inherited two
wives from his brother and already had one of his own. He had not
declared that the widows were his wives, though in fact they were,
and one of the inherited wives had borne a son by him. It was ex-
pected that the relationship would soon be publicly acknowledged,
especially in view of the fact that one of the inherited wives had
moved from her deceased husband's house to that of his brother
shortly before my departure. In two additional cases a man got his
second wife from a living brother. In one case the husband acqui-
esced when his wife declared her intention to leave and expressed

152 KIN GROUPS AND KINSHIP

her desire to marry her husband's brother. In the other case the man
simply took over the wife of his younger brother, lived with her and
his first wife in a chan, and fathered a son by her. After the birth
of the son and an announcement by the household head (a still older
brother), she became publicly known as his wife. At that time the
younger brother was promised a new wife.

Fifteen of the 27 women in these polygynous marriages were virgins
(that is, previously unmarried) at marriage. The twelve nonvirgins
included the five procured from brothers (three inherited, two taken)
and seven divorcees. There is a significant difference in the incidence
of virgin brides among first, as compared to second, wives in these
polygynous unions. Ten first wives were virgins and three were non-
virgins, while five second wives were virgins and nine were nonvirgins.
Although three cases of inheritance account for three of the nonvirgin
second wives, there is still a marked tendency to accept previously
married brides as second wives and to reject them as first wives. There
is also a preference for marrying a virgin daughter to a single man.

In 22 of the 27 marriages in these 13 families, bride-price payment
was made at the time of marriage either to the bride's family or to a
former husband, while in only one case, a first marriage, was a dowry
given instead. The remaining four cases were those of wives obtained
from a brother and involved no exchange of money.

Marriage ceremonies were performed in seven of the 13 initial mar-
riages which established families destined to become polygynous. All
seven were among the 10 with virgin brides. This conforms to the
village ratio: about 50 per cent of all marriages have been without
benefit of ceremony, and ceremonies are performed only when the
bride is previously unmarried.

Second marriages are less often ceremonialized than first ones. Of
the 14 second marriages (that is, those which made the family polyg-
ynous), only two were ceremonialized, and both of these were among
the five such marriages which involved virgin brides. Intervals be-
tween marriages ranged from two to 20 years. In six marriages the
co-wives normally shared a house or chan as residence, and in seven
marriages they were separated.3

Three of the thirteen instances of polygyny were sororal — sisters
married to the same man. Two pairs of co-wives were real sisters and
the other pair were classificatory sisters (daughters of first cousins re-
lated through the male line).

Levirate, or inheritance of wives from a brother, is standard pro-
cedure here. Either an elder or younger brother may inherit at the dis-
cretion of the household head, who is usually the eldest surviving

KIN GROUPS AND KINSHIP 153

brother. Although no preference for junior levirate was expressed in
Sirkanda, it is most frequent, probably because elder brothers tend
to die first. If an outsider wishes to marry the widow, he must secure
permission and reimburse her husband's family. This, too, is frequently
done. Elopement is another frequent means by which a widow marries
the man of her choice. A boy who has been betrothed and then dies
may be replaced by his brother to fulfill the marital contract. There is
no evidence of the systematic practice of sororate, wherein a deceased
woman is replaced by her single sister in the marital union. Two or three
isolated instances occur in the data. Similarly, there are isolated instances
of brothers marrying women who are sisters, but this is an unusual ar-
rangement neither encouraged nor discouraged.

In this society it is important to contract proper marriages for one's
children. Once this has been done, regardless of what may happen
subsequently, the honor of the family is intact. Danger of an inter-
caste or intra-sib union is precluded because a proper match has been
made. If, later, a proscribed alliance develops, the individuals in-
volved will take the blame; the family has done its duty. This explains
the fact that no deviations from rules of caste endogamy and
sib exogamy were found or reported for first marriages in Sirkanda.
Maintenance of family honor is also offered as an explanation for
early marriage. Early marriage prevents the disgrace of an unwed
mother or nonvirginal bride because no girl old enough for child-
bearing or adult sex activity is unwed. That she may have relations
with various men and perhaps even bear children by them in the
absence of her husband is irrelevant — she is a properly married woman.
Virginity at first marriage is important, while sex behavior thereafter
is unimportant. As one man put it, "We disobey the law [which sets
the minimum marriage age for girls at fourteen], but we protect our
honor." As a result, unwed pregnant girls are a rarity. When they are
found out they are immediately married to their lover if possible, or to
one of their sisters' husbands, or to some boy whose family is willing
— usually a family hard pressed to find a bride for financial or other
reasons. Unwed mothers are virtually unknown. The nearest approach
to unwed motherhood that can be cited in the village is the occasional
un-rewed widow who bears a child, to the consternation of her family.

First marriages for women, as described in the preceding chapter,
are likely to be accompanied by ceremony and public acclaim. Sub-
sequent marriages for women are contracted without ceremony and
receive little attention, though they do receive public recognition and
approval. Second marriages for men are less likely to be ceremonialized
than first ones. In second marriages which run counter to rules of

154 KIN GROUPS AND KINSHIP

endogamy or exogamy, and in other "love marriages," the couple
usually elopes or the woman comes secretly to the home of her new
husband and the marriage is accepted by others as a fait accompli.
Numerous cases of this kind have occurred among all castes. They are
frequent bases for disputes, sometimes resulting in long-standing fac-
tional splits involving considerable numbers of people.

Endogamy

Paharis have a reputation for disregarding the rules of caste
endogamy and other orthodox Hindu marriage regulations. My data
indicate that in this region the reputation is undeserved so far as
initial marriages are concerned but finds corroboration in subsequent
marriages. While no deviations occurred among first marriages in Sir-
kanda, some did occur in second or subsequent marriages. The num-
bers of such deviations are not large, but the fact that they exist is
important.

In Sirkanda there is at present one established and recognized inter-
caste union — that of a Rajput man and a Brahmin woman who eloped.
The woman came to the man's village and took up residence with him,
and he then reimbursed her former husband, a Brahmin. The only
long-range effect in Sirkanda was estrangement between the Rajput
husband and his brothers, who resented his bringing a Brahmin woman
into the family. To this day his elder brother will not let the Brahmin
woman into his house, nor allow her to address him as "husband's
elder brother." In the woman's village the Brahmins were angry but
did nothing beyond lodging a prompt protest and threatening to
punish physically the Rajput who took this woman. There is a sim-
ilarly accepted union of a Brahmin man and a Rajput woman in an-
other Bhatbair village.

A second case involving a Rajput man and a Brahmin woman in a
neighboring village was terminated by public pressure after a council
met and directed that the woman be returned to her former husband
and the abductor pay a fine (which he avoided by leaving the village).
A child had been conceived in this union, and there was talk of in-
ducing an abortion. However, when the father left and the woman
settled back with her former husband the idea was dropped. The child
was born and was accepted as a Brahmin — as the offspring of the man
and wife in whose household it was born — despite its well-known and
undoubted parentage. The Rajput involved was a well-known "loafer"
or philanderer who had left more than one wife and had established
unions with several women, including a low-caste woman. Moreover,
he had taken the Brahmin woman far away, to Delhi, on the strength of

KIN GROUPS AND KINSHIP 1 55

promises which were never fulfilled. He failed to provide for her prop-
erly and allegedly mistreated her and sold her jewelry for his own
profit. Had he been a stable member of the community, it is likely that
little sentiment would have been aroused over the incident. Normally
feeling does not run high in such cases. As a Brahmin commented in
the council meeting to resolve this case, "There is not much differ-
ence between Rajputs and Brahmins, so it doesn't matter very much
anyway." On the other hand a Rajput was apparently incensed and
made the comment, "A Brahmin woman is like a mother to Rajputs."
The latter attitude is evidently a statement of ideal rather than actual
behavior.

Unions across high- and low-caste lines meet with greater public
disapproval. They are nearly always unions of high-caste men with
low-caste women; the reverse is violently condemned and could not be
continued if it were known. In Sirkanda two Rajput men have taken
low-caste wives from outside the village — one of barber caste and one
of a charcoal-making group. Both unions were terminated under pub-
lic pressure when the facts became known. One Rajput man related to
a Sirkanda family but resident in a neighboring village took three
low-caste wives, a weaver, a carpenter, and a blacksmith, and lived with
all three at once. In the case of the last of these the couple underwent a
ceremony at the insistence of the girl's parents, for she had not been
previously married. A Brahmin was found to perform it, but no one
other than the girl's relatives attended. The man remained with these
three wives throughout his life but was ostracized by his own caste.
His one son is of the caste of his mother but has inherited the property
of his father. A life-long Rajput-weaver union existed in a nearby
village, and in another village a Rajput had three blacksmith wives.
Both the blacksmiths and Bajgis of Sirkanda trace their ancestry to
high-caste male ancestors, a Rajput and a Brahmin, respectively, who
married women of low caste with the result that their children were
assigned to the same caste as their mothers. Whether this ancestry is
fact or fancy, it is believed and indicates the possibility of such unions
and their results.

An extreme case of intercaste marriage in Bhatbair involved the
elopement of a boy of shoemaker caste and a Brahmin girl — the lowest
and highest castes and the reverse of the expected sex-caste affiliation.
This was a universally disapproved case. When such a union is hypothe-
sized to villagers in a question, the questioner is promptly assured that
the result would be death to the man or perhaps to both partners, and
lifelong ostracism should either survive. In reality, the couple ran off
to Dehra Dun, where they lived together for some time. There fellow

156 KIN GROUPS AND KINSHIP

shoemakers finally put pressure on the boy to return the girl to her
people, and he too went home. Now both live in their respective vil-
lages, unmarriageable but otherwise unimpeded in carrying on their
normal lives. Rumor has it that they are still in love and perhaps in
surreptitious contact and that they have no desire to remarry even were
this allowed.

Among the low castes, blacksmiths and carpenters form a single
endogamous group while weavers and the immigrant Sirkanda bar-
bers are separated from them and from one another by little social or
ritual distance. These groups intermarry frequently and without public
disapproval (see discussion of "Relations Among Low Castes" in
chap. 7).

Bajgis and shoemakers stay within their respective castes more con-
sistently as they are more distant from one another and from the above
group than are the castes within the above group. Exceptions cutting
across these lines probably occur with about the same frequency and
results as the above-listed high-caste deviations. No examples of ex-
ceptions were found in Bhatbair, where only a small sample of low-
caste marriages could be obtained because of their small numbers.

Married women of any caste may be "sold" to outsiders, as described
in the section on "Illegal Economic Activities" in chapter 2. In such
cases caste is not a significant consideration. There are Sirkanda women
who have been married in this way to plains merchants (Banias), Nepa-
lese Rajput military officers, and a Sikh religious leader. There is no
caste in Bhatbair whose women have not on occasion been involved in
such transactions, though all, and especially the high castes, are careful
to conceal the fact. In terms of frequency, more low-caste women have
been sold. Usually the woman's family and her husband lack prior
knowledge and do not consent, although one Rajput husband appar-
ently connived in such a transaction in order to make way for a new
wife for himself. Among the low castes an unmarried girl is occasionally
"sold" by her parents to outsiders in this fashion. Two such instances
were recorded as initial marriages in Sirkanda. These were two of 32
recorded marriages of low-caste village girls, so the practice is not
especially rare.

One current case of cross-caste marriage exists in Sirkanda and three
others have been reported in recent years, constituting a total of about
1 per cent of all unions. Marriage of women to non-Paharis would not
raise this above 2 per cent at the maximum. Of the local intercaste
unions only Rajput-Brahmin and some inter-low-caste marriages sur-
vived public disapproval. In all reported cases marriages which crossed
the boundary between high and low castes resulted either in ostracism

KIN GROUPS AND KINSHIP 157

of the high-caste partner by his caste-fellows or dissolution of the mar-
riage under public pressure. No union would be allowed to persist be-
tween a high-caste woman and a low-caste man, and none would be
openly attempted in the area. There is an explicit tradition that, while
intercaste marriage is not permissible, if it does in fact occur and if it
is repeated for seven generations, a new caste is formed and recognized.
No example of this was known, and the mechanics of its operation were
hazy in the minds of villagers. Normally the children of an intercaste
union belong to the caste of the lower-caste parent, usually the mother.

Before closing this discussion of rules of endogamy it may be rele-
vant to mention that the most popular Pahari song in Sirkanda during
1957-1958 glorified an intercaste elopement and widow remarriage.
This song, sung primarily by young people as they worked in the jungle,
had several versions. In essence, it was the story of a young widow of
weaver caste who was directed by her own and her husband's family
to remain in the family of her deceased husband. Unable to face the
prospect of such a lonely life, she eloped to the plains with a Brahmin
man. Tracked down by her relatives, the couple were taken to court,
where the magistrate ruled in favor of the couple, saying that it was
their right to do as they felt best. He fined the relatives for causing
such inconvenience to all concerned. The couple went off to live hap-
pily ever after, but not before the woman's relatives had secured a
measure of revenge by branding her on the forehead with a red-hot
coin. The song is allegedly based on a true story of recent origin in a
neighboring area of Tehri Garhwal. It was popular with all castes, and
the heroine was known affectionately to all by her pet name.

Thus, marriage across caste lines is not abhorred in this society to the
extent that it is in many areas of India. Rules of endogamy are not
rigidly adhered to, but they are not violated as frequently as the Pahari
reputation would lead one to expect.4 Intercaste marriage is tolerated
if it is not the initial marriage and if the castes of the individuals in-
volved are of the same general economic and social level. As will be-
come evident in later discussion, this is one aspect of a generally looser
definition of appropriate caste behavior in the hills as compared to the
plains.

Exogamy

Besides caste endogamy, marriages are also regulated by sib exog-
amy. In this region no person is allowed to marry within his own sib
or that of his mother. To do so would be to commit incest, since it
would be to marry classificatory siblings or other relatives. All of the
mother's sib is included within the kindred (discussed below). Children

158 KIN GROUPS AND KINSHIP

of mother's siblings (real and classificatory) are classified as siblings of
ego, even though they are of different sib affiliation, just as are children
of father's siblings. In addition, among the high castes the sibs of di-
rect female ancestors are supposed to be excluded from eligibility for
marriage for five to ten generations back, as reported variously by
Brahmin informants, who are the ultimate authorities on such matters.
This is evidently an expression of the Hindu sapinda rule of exogamy
prohibiting marriage within the bilaterally defined kindred:

The marriage rules as regards Sapinda relationship or consanguinity de-
fine that a man should not marry a girl who is related to him through a com-
mon male ancestor up to the 7th generation in the father's line and up to
the 5th generation in the mother's line. Different law books give different
rules. (Karve, 1953, p. 55)

Unfortunately in Sirkanda sib affiliation of village wives proved to
be a difficult and unreliable type of data to obtain for generations
preceding the present, although marriage does not alter a woman's
sib affiliation. However, the village contains two large and two smaller
Rajput sibs. The data on these are good and throw light on the matter
of inter-sib marriage arrangements, since these sibs can and do inter-
marry frequently in the village. Fifty-three Rajput marriages between
people of the sibs found in Sirkanda were recorded, and they revealed
no cases of marriage within own or mother's sib on first or subsequent
marriages.5 Four cases of marriage into paternal grandmother's sib
occurred. One of these four was a first marriage, and two comprised an
instance of sororal polygyny. This indicates that the effective exoga-
mous units are own and mother's sib only, and accords with information
given by Joshi (1929, p. 75) and with the testimony of uneducated
Rajput informants, who mentioned only these as prohibited groups.

A single instance of intra-sib marriage was reported. The wife of a
Sirkanda Rajput left him shortly after marriage to live as wife of a
man in another village who stood in the relationship of father's brother's
son's son to her and was therefore her sib-fellow and nephew. Since this
was a second marriage, objection was mild and the union has endured.
Probably the fact that the relationship was not particularly close and
that the woman was not older than the man added to its acceptability.
No other instances of marriages within incest boundaries appeared.
Perhaps this is partly because when they occur among the most distant
prohibited relationships they are not as conspicuous as the more fre-
quently reported intercaste marriages. However, these data and the
expressed attitudes of villagers suggest that rules of exogamy are less
frequently broken than rules of endogamy in this society.

KIN GROUPS AND KINSHIP 159

Marriage Networks

There is no rule of village exogamy in this area such as is reported
for other parts of North India, including nearby Jaunsar-Bawar
(Saksena, 1955, p. 28; Berreman, ig62d). Neither is there any reluctance
to give and take brides in the same village as is true, for example,
among the Noniyas of Senapur (Rowe, 1960b). As will be mentioned
again, high-caste sibs in Sirkanda may derive from formerly exogamous
community affiliation, but this does not affect the present situation in
and around Sirkanda. In this region, if a man and woman of the same
caste are not within one another's kindred, if they are of different sibs
and their mothers are of different sibs, they are potential mates re-
gardless of village membership. Local exogamy often results from the
fact that these conditions cannot be met within the village, but this is
local exogamy in effect, not in intent. Local exogamy occurs in all
castes of Sirkanda except Rajputs, because all but they are single-sib
local segments of castes. Some Bhatbair villages are entirely single-sib
villages, and hence their members must marry outside the village.

Of 471 recorded marriages contracted by Sirkanda people, 377 were
Rajput marriages, and of these 77, or 20 per cent, were contracted
within the village.6 The remaining 394 Sirkanda marriages were con-
tracted with people of 92 identified villages and four general areas at
greater distances — roughly 400 marriages in 100 localities (see map 2).
The numbers ranged from 45 marriages in one village (Kanda) to one
each in 36 villages. Thirty per cent of these marriages were contracted
in seven villages, and 80 per cent were contracted in 50 villages (in-
cluding the seven just mentioned but excluding Sirkanda), all within
a four air-mile radius of the village, that is, eight trail miles, an easy
half-day's trip. This leaves 20 per cent of the nonviilage marriages
spread over 42 villages and four broader areas up to 18 air miles dis-
tant (about 35 trail miles, a two-day trip each way). Thus, of all Sir-
kanda marriages, 83 per cent are within a radius of four air miles. Six-
teen per cent (all Rajput marriages) are within Sirkanda itself.7 A
villager remarked, "On the plains it is easy to travel, and people there
go great distances for brides. Here it is very difficult to get around so
we have to find ours closer to our own village. It is as hard to go one
mile here as it is to go five miles in the plains on foot, and many places
there they can go by motor bus or at least by cart."

Virtually all Sirkanda marriages are contracted in Pahari villages,
although non-Pahari villages are well-known and easily accessible
within five air miles of the village, in the valley en route to Dehra Dun.
Marriages at distances this great or greater comprise 17 per cent of all

l6o KIN GROUPS AND KINSHIP

village unions, but they are all in Tehri Garhwal, which is culturally
more similar, though physically less accessible, than the valley. The
only exceptions in Sirkanda have been six barber-caste marriages and
the sale of two low-caste women to outsiders.8

Caste differences in distribution of spouses' villages revealed differ-
ential extent in marriage networks (cf. Rowe, 1960b) in the various
castes. Eighty-five per cent of high-caste marriages occurred within the
four-mile radius, and 73 per cent of Bajgi, 53 per cent of blacksmith,
and 35 per cent of barber marriages were that close. This is a direct
reflection of the relative population of these groups and of the outside
origin of the barbers. The fewer the potential mates for a group in the
area, the farther they have to go to find mates. The overwhelming ma-
jority of Bhatbair residents are Rajputs and Brahmins. Likewise, in-
dividual villages reflect their caste and sib composition in the fre-
quency of intermarriage with Sirkanda villagers. One village provided
45 mates, all Rajputs, while another provided 14 mates, 12 of low
castes, and two Brahmins. There were no discernible patterned differ-
ences between distribution of bride-giving and bride-taking villages;
that is, giving and taking of brides appeared to occur randomly among
villages in the marriage networks of each caste, except where sib ex-
ogamy prevented it.

The marriage network of each caste roughly defines the limits of its
informal social interaction outside the village. Most visiting is done
with relatives, and most relatives outside the village are affines. How-
ever, the village community remains the social unit of most frequent
and important interaction. It is not surpassed, even among low castes
(as is the case among Senapur Noniyas) by the intervillage marriage
network. Low-caste people of Sirkanda have not experienced the new-
found freedom of the Noniyas and, even if they had, demography and
topography might have precluded development of transcendent social
and political functions characteristic of the marriage network of Sen-
apur (cf. Rowe, 1960b, p. 310). Sirkanda villagers are still closely tied
to the multi-caste village dominated by the local Rajputs.

New Wives for Old

Divorce and remarriage are frequent among the people of Sir-
kanda, although perhaps not so common as in Jaunsar-Bawar, where
Majumdar (1944, p. 162) reports that barrenness results in divorce and
Saksena (1955, p. 36) reports that the "slightest disloyalty or the slight-
est slip" on the part of a wife may result in divorce.9

A couple who do not get along well together or who have specific
grievances either on their own part or that of their families — for ex-

KIN GROUPS AND KINSHIP l6l

ample, the wife refusing to stay with her husband, or the husband's
family refusing to make good the bride price — may go through a pro-
cedure called chUt, divorce, which breaks the marriage bond. Either
party may initiate this action. In such cases an ad hoc panchayat, or
council, of friends of both parties acts as intermediary to achieve a sat-
isfactory settlement. A written agreement is signed by the father of
the wife and by the husband or his father. The agreement specifies the
amount to be paid by the family of the wife and states that thereafter
the parties are to be free of mutual obligation. In addition, a sum is
paid to the panchayat. If the girl plans to live with someone else, that
man may pay the former husband's family either directly or indirectly.
One such formal dissolution occurred in a Rajput family in the year

1957~1958.

A much more common type of divorce, and virtually the only type
among low castes, occurs when a wife goes to her parents' home or to
another man and refuses to return, or a husband sends her home. In
the former case the husband will try to exact reimbursement but may
not be successful, especially if the wife runs some distance away. Vil-
lage or caste panchayats may intervene to secure a just settlement in
such cases. In any elopement the panchayat is called by the family with
which the girl was affiliated at the time of elopement (usually the hus-
band's family, but sometimes parents), as they are the aggrieved party.

Threats of running away or unannounced short-term retreats to the
home village are used by women to secure better treatment from their
husbands. Husbands may deprive errant wives of things they want
(trinkets, new clothes, attendance at a fair) or physically punish them.
No accurate data were obtained on frequency with which the husband,
as compared to the wife, instigated divorce, but it appears to be about
equal. Although precise data are lacking, it seems likely that at least
20 per cent of all marriages are dissolved by formal or informal divorce.
This estimate is based primarily on figures for high castes. Unlike the
plains situation, where low-caste divorce is frequent but high-caste
divorce is infrequent or concealed, there are no apparent differences
in divorce rates among the various hill castes. Divorce is taken as a
matter of course.

Causes of divorce are many and complex. Failure to fulfill the formal
and informal obligations of marriage are overt reasons given for di-
vorce. Among reasons given in specific instances were: nonpayment of
bride price, nonvirginity of allegedly virginal bride, mental or physi-
cal defect in one partner, bride's failure to perform her duties in the
house, bride's inability to get along with in-laws, bride's refusal to stay
in the household or refusal to return after postmarriage visit to her

162 KIN GROUPS AND KINSHIP

home, husband's mistreatment of wife, husband's failure to provide
adequately for wife, husband's departure from village leaving wife
stranded there, father-in-law's molestation of bride, and persistent
adultery or philandering on part of wife or husband. Occasional adul-
tery is not normally a ground for divorce and is, in fact, expected. How-
ever, when it is practiced openly, when lasting attachments are made,
when a person acquires a reputation for excessive indulgence — es-
pecially if it involves cross-caste relations — or when a person neglects
his or her spouse in favor of a lover, divorce is likely to follow. If either
partner falls in love with someone else, he or she is likely to leave or
purposely bring about divorce. In at least one case in Sirkanda a wife
left when her husband took a second wife, although his intent had
been to establish a polygynous relationship. A Brahmin man in a nearby
village lost his first wife, a Brahmin, when he brought in a Rajput
woman as second wife. Remarriage after divorce is almost universal.
Some men and women have had a succession of marriages. One Sirkanda
man voluntarily parted with three successive wives. One nineteen-year-
old girl is living with her third successive husband, a young Sirkanda
man. On the other hand, one village daughter in Sirkanda has returned
home after a particularly unhappy marriage, resolved to live out her
life in the large extended family of her parents and brothers and never
to remarry.

Remarriage after a spouse's death is expected. High-caste plains
customs in this regard are known but not envied. This is one instance
in which orthodox, Sanskritic custom is not emulated. Even educated
Pahari Brahmins do not press for it, saying that in these hills it is im-
practical to keep dependent widows unmarried. There is no reluctance
to discuss the subject and no inferiority feeling about it, as there is
about some non-Sanskritic practices. One Rajput man said,

Forbidding widow remarriage is a stupid custom. What is a widow to do
with herself if she does not remarry? She is a burden to her family and to
herself. Anyway, it would not work here. Every man needs a woman and
every woman needs a man. If our widows didn't remarry we would have de-
pendent widows and unmarried men who couldn't do all of their work in
the village. It is taken care of easily because the woman often goes to her
husband's brother.

If a widow remarries outside her husband's family, she usually gives
up her rights to her husband's property, though her children by him
will receive it when they reach maturity, even if she keeps them with
her. Occasionally a widow remains in her husband's house and re-
mains a clan member, even though she does not become the wife of

KIN GROUPS AND KINSHIP 163

one of his brothers. Rarely such a woman will take a husband from
outside the family and yet remain in the house. As in the unusual case
of a matrilocal initial marriage, this occurs only if there are no eligible
men in the family and most often when there are no other adults in
the family at all. The new husband then becomes a member of the
clan of his new wife and her former husband. He is derisively said to
have "gone to sit" at the woman's house, and he is likely to be ridiculed
if he is an outsider in the village. The arrangement is advantageous
if the woman has property from her first husband which she can in this
way retain and still be remarried. One such case had occurred in a
Sirkanda chan in recent years.

Adoption

Having children, and especially sons, is very important in Sir-
kanda families, as it is throughout Hindu India. The son not only
helps with the work and carries on the family line but inherits the
father's property and performs the necessary rites associated with the
death and postdeath welfare of his parents. If no son is born to a man
he often designates his son-in-law to fulfill these roles. The son-in-law
must then come to live in the father's house, and he manages the lands
of his deceased father-in-law, ultimately passing them on to his sons
by this wife, or to an adopted son. If he has no sons, the land does not
revert to others in his lineage, but stays in the lineage of his wife.

If a man has no son or daughter, he may designate some other rela-
tive to be his heir and to perform the death rites. More often, however,
he will adopt a son. Girls are not adopted. There are currently four
cases of adoption in Sirkanda. An adopted son is usually a relative who
stands to gain more in the way of property by the adoption than if he
stayed with his real parents. Often he is a daughter's son. He may be a
boy born under an inauspicious asterism, an orphan or semi-orphan,
or one of many brothers in a poor family. Like marriage, adoption
does not alter sib affiliation, but it does alter clan identification. Adop-
tion is frequently resented by relatives of the adopter, who would in-
herit his property were it not for the adopted heir. Sometimes a family
offers a son for adoption, especially when he is near adulthood already,
for purely mercenary reasons: they hope to get additional property into
the family.

Adoption is publicly recognized but is informally effected without
ceremony. The adopted child simply comes to live with the adopting
family, and it is announced that this boy will inherit the property of
his new father. From that time on the boy is treated as a son by his

164 KIN GROUPS AND KINSHIP

new parents. "An adopted son is just like a real son to his father. He
serves and honors his father even more than a real son would. Since a
foster father has no other sons, he showers more affection on the adopted
boy than he would if he had several of his own. He won't adopt a boy
he doesn't like." If adoption occurs in childhood, the boy grows up in
the family as a son. If, as is often the case, adoption occurs later in life
— after the foster parents have given up all hope of offspring of their
own — the boy moves into the household of his new parents even if he
is already married. He begins to share in that economic unit and ceases
to share in that of his real father. The parents by adoption will make
wedding arrangements and payments if the boy is unmarried or takes
a second wife. In such cases the boy may be entitled to claim inheritance
from his real parents as well as the foster ones. Adoption is generally
arranged with an advance understanding on this matter, and usually
the boy surrenders his claim to his real parents' property.

Sometimes an orphaned or otherwise disadvantaged child, usually a
relative, is taken into a family where there are already several children.
There is less interest in such a child, probably because he is thrust
upon the family and will perform no important function for it, and
often he is exploited and deprived in comparison to the real offspring
of the family. Two current instances of this type were found among
Sirkanda families.

Child Rearing

An important function of the family is, of course, child rearing.
Children are usually carefully cared for. If there are already four or
more in the family, a child may tend to suffer from neglect. Infant
mortality rates are high, but accurate data are unavailable. Three
children two years of age or under died out of 31 in the village during
1957-1958. One of five born during the year died shortly after birth.
Two families reported twice as many births as living children; some
others had had few or no infant deaths. Probably infant and child
mortality runs over 20 per cent.

Children are not given solid foods until a year after birth. They
continue to nurse until another child is born, and sporadically there-
after. Occasionally a youngest child may be allowed to suck for six
years or longer.

The atmosphere of child rearing is indulgent and permissive. The
child is allowed to handle anything within reach, and its parents or
relatives and friends are usually around the house, sitting on the floor
where the child can crawl over them as it pleases and where they can

KIN GROUPS AND KINSHIP 165

show it whatever attention it demands. Almost anything a small child
does is accepted. No special effort is made to encourage the child to
walk. Toilet training is gradual and not intensive; the mother or a
sibling simply begins taking the child out of the living area of the
house when defecation or urination seems imminent, and accidents are
tolerated without comment for at least three years. The small child,
until it can walk competently, spends most of the time in the house or
on the hip of an elder sibling. Since dwelling rooms are on the second
story, doorways and verandas are barricaded with boards to protect the
infant from falling. Often the child is tied around the waist with a
leash attached to a bedpost. In most extended families someone, often
an elderly person, is always around to look after the child, and this is
cited as an advantage of the extended family system. In families where
there is no extra person about to watch the child, it may simply be
locked in the house when the parents have to be away to work in the
fields or forest. One infant was burned to death in 1957 when it rolled
into the fireplace after having been left alone while its parents were
working in the fields. As children grow old enough to get about easily
and to take care of themselves, they are free to roam the village with
siblings or other children, though they are encouraged to stay near the
house when unaccompanied.

About half the boys resident in the village and an occasional girl
attend the local school with varying degrees of regularity from age 6
to 1 1 or 12. The children who attend are those whose parents want them
to, and this is largely a matter of whether they are needed to help with
the household work. By age 8 to 12 boys begin to accompany their
fathers in their work, while girls continue to help their mothers. By
age 14 a girl is ready to live with her husband, whom she has married
1 to 3 years earlier. Ideally she should have her first menstruation in
her husband's household, but some marry considerably later. A boy is
ready to take a wife by 16 or 17, though some do not do so until later.

Throughout childhood the child is rarely disciplined. He is repri-
manded, commanded, and threatened, but these words are not often
enforced by physical means, and the child soon learns this. Lackadaisi-
cal compliance is the typical reaction to adult direction. The child is
generally in the company of other children but is rarely excluded from
adult company. There is little a child cannot see and attempt to imi-
tate, although sex activity is (not very successfully) concealed from
children. Caste consciousness and discrimination are learned from
childhood through instructions received, references overheard, and
behavior observed. However, caste barriers do not enter into children's
interaction among themselves until after puberty. One case of mutual

l66 KIN GROUPS AND KINSHIP

"puppy love" during my stay in the village was between a 15-year-old
blacksmith boy and the 14-year-old daughter of the village headman, a
Rajput and the wealthiest man in the village. During this period the
girl became engaged to her future husband, but this did not dismay her
nor her friend. Their relationship was concealed from adults but not
for reasons of caste.

Intrafamily Relations

Patterns of interaction among members of the patrilocal extended
family depend in large part upon interplay of age, sex, and relationship
roles. Males take precedence over females, age over youth, consanguin-
eal over affinal relationship.

As has been stated, the eldest active male is the household head. This
man is in the relationship of father or elder brother to other adult males
in the family. He is responsible for all decisions in the family, and his
is the final authority, whether it be in matters of allocating lands to
crops, performing worship, or arranging a marriage. The father is not
always the authoritarian family head as ideally described. He may
hesitate to rebuke an adult son. The father of one young Sirkanda man
was disturbed at his son's continual absence from home in pursuit of
an illicit love affair in another village. He was not doing his share of the
family work (they were Bajgis). But the father had not the courage to
face the boy on the matter. Finally he went to the husband of his son's
lover and complained to him that his wife was ruining the boy and
taking his money, and that the affair must stop. The husband, who had
known of the affair but had not interfered (probably because his wife
was getting material rewards for her service), then spoke to the young
man and told him to leave the woman alone. It worked, for a time
at least, with a minimum of hard feeling and no intrafamily tension
in the Sirkanda family.

In most instances the family head acts on family matters after con-
sulting with other males and often his wife as well, but this is not
necessary nor prescribed. When an old man becomes senile or inactive
or when he dies, he is replaced by the next in line. If conflict occurs
over succession, the joint family is likely to divide. An old and inactive
man tends to be ignored or actively resented by other family members
and he himself often resents their attitudes toward him, so that every-
one looks forward to his death. Old men, like old women, find occupa-
tion around the house in caring for children or doing domestic tasks.
After the death of the household head his widow may be referred to as

KIN GROUPS AND KINSHIP 167

the head, but in reality the authority usually passes to the next eldest
male if he is an adult.

Among brothers in a family, age takes precedence. Younger brothers
are expected to obey and respect their elders. This rule applies to par-
allel cousins (father's brother's sons) as well as to real brothers. Sisters
are expected to respect and obey brothers regardless of relative age,
though of course they often care for and wield authority over much
younger brothers in childhood. Among sisters, age takes precedence.
The mother has the honor and respect of sons and daughters. She re-
tains authority over daughters till they marry, and loses authority over
sons as they reach maturity.

Before proceeding further it will be well to emphasize that Pahari
women of all castes enjoy a degree of freedom unknown among any
but the low-caste women of the plains. They work alone or in groups
without male accompaniment. They come and go as they please around
the village and talk to whomever they please except strangers. They are
not restricted to separate living quarters, and they are not subject to
purdah (seclusion). Therefore in the ensuing discussion, their sub-
ordination to men must be recognized as subordination within a con-
text of relative freedom. It is not the same order of subordination
found among many plains groups. One indication of this is the ease
with which divorce and remarriage are effected and the freedom allowed
women in sexual matters. If a woman is unhappy she can always turn
elsewhere or go home.

The relationship between the patrilineal family and the wives who
have joined it from outside is ideally one of a cohesive group taking in
a stranger. The bride comes in to be critically appraised by the extant
in-group. She must prove herself by her good works. The new wife
finds herself at the bottom of a well-established hierarchy. She often
does not know her husband, and in any event his loyalties and respon-
sibilities are first to the family. She is under the direct authority of her
mother-in-law and the wives of her husband's elder brothers. She must
show obedience, respect, and deference to all of her elders. She shows
respect by never using the names of the males and elder females in the
family, by never smoking in their presence or laughing to their faces.
Often she is accepted and even comforted by all or most of the women
in the household. Sometimes she is not. Her fate in this respect is in
the unpredictable hands of others, and therefore it is an uncertain and
potentially unpleasant one. Consequently the tears of a bride at leaving
her family of origin are not entirely conventional. As time passes and
a wife proves her value as a contributing member of the family through

l68 KIN GROUPS AND KINSHIP

her industry and skill and especially by producing children, her po-
sition becomes increasingly secure. As younger sisters-in-law come in,
her authority increases. She learns which family members are her
friends and which are not; to whom she can turn for consolation and
from whom she must keep her secrets. Often a woman's husband be-
comes her strongest ally and will mediate on her behalf if necessary.
The husband's younger brothers are traditionally her friends, and often
it is only with them that a young wife is able to establish an easy, in-
formal friendship. There is something of the "joking relationship"
between these two that is traditional over much of North India. Sisters-
in-law may become friends and so may other village wives, especially
if they have previous ties, as when they are related to one another or
are from the same village. In such cases a woman may receive moral
support outside the family, but her loyalties and responsibilities must
remain with her husband's family. This is true even when her own fam-
ily lives in the same village. In such a case a woman is in frequent inter-
action with her family of origin, but she must be careful not to let
such interaction interfere with her responsibilities to her husband's
family.

One day in Sirkanda a young village wife who lived a stone's throw
from her parents' house stopped to talk to her mother. The mother
and she requested that she be photographed with her infant daughter.
When I offered to do so at once, they quickly declined, saying that the
girl's mother-in-law had sent her to collect fodder and would not like
it if she delayed to be photographed at her mother's request. I was
advised to come to the mother-in-law some time and offer to take the
photograph without mentioning this prior arrangement.

A woman's natal household gods may intervene in her behalf if she
is mistreated by her husband's family. If worst comes to worst, it is
relatively easy for her to escape to her home or to another potential
husband. This happens frequently in the village, though usually a
girl's parents return her to her husband, often after a conference with
her husband's family. In happier circumstances a wife from outside the
village can expect to visit her home about once a year for a few days
or weeks and can expect to see relatives and friends from her own village
occasionally on visits.

In visits to her village of origin, among her parents and siblings and
childhood friends, a woman is very free and relaxed. She is under the
authority of parents and brothers, but it is normally an indulgent
authority. She has few responsibilities at home. Her brothers' wives do
the work, and she is the guest of honor. There is no one to watch her
moves and report or criticize her behavior. She is more likely to joke,

an hi

KIN GROUPS AND KINSHIP 169

talk, or flirt with men and dance if a dance is held, than in her hus-
band's village. Often she uses the opportunity to carry on clandestine
love affairs with the men of the village whom she knew in her youth.
There is something of the "double standard of morality" for a woman
in the house of her parents and that of her in-laws, which is more pro-
nounced in Jaunsar-Bawar, where Majumdar reports: "A woman has
two standards of morality to conform to, one in her parents' house, the
other in her husband's. In her parents' house she is allowed every kind
of liberty and licence and nothing is an offence unless specifically pro-
hibited" (Majumdar, 1944, p. 163; cf. Saksena, 1955, p. 36).

Marriage is an important event in a man's life, too, but it is not as
potentially traumatic as it is for a woman. He remains in his own fam-
ily and village where there are friends and allies on every side as con-
trasted to the strangers and critics who usually greet a bride. A man's
social situation changes relatively little. That his responsibilities are
somewhat increased is recognized in the vows he makes at marriage. A
village joke runs to the effect that an unmarried man is a free man, he
can run about as he pleases "on two legs"; a married man is like a
quadruped, he must spend most of his time foraging for food; a man
with children is like the eight-legged spider, he must weave a net and
be ever ready at its center to seize any food that comes his way.

A husband who visits his wife's home is in the position of all in-laws.
He is the honored and respected guest, but neither he nor his hosts
are likely to feel at ease and the visit is usually made as short as pos-
sible. Of course, in many instances, the husband and his in-laws know
one another well and may visit frequently and easily, but the ideal
pattern is one of respect and distance. Most visiting outside the village
is with affinal relatives, especially the wife's brothers. When a man's
in-laws live in his own village, he is often on close terms with them and
frequently visits with, and works with, his father-in-law and brothers-
in-law. However, an element of mutual respect is maintained.

The relationship of wife to husband is ideally one of devotion, serv-
ice, and respect. The husband is referred to by his wife as mdlik, owner,
or simply "man." He makes the decisions and gives the orders, though
in reality a wife may exercise as much influence as their personalities
allow, and the henpecked husband is a familiar concept.

A woman shows her respect and devotion by catering to and antici-
pating her husband's wishes. When he comes back from a trip or from
working, even if she is tired herself, a wife will massage his limbs with
oil, feed him, prepare his pipe for him, and make him comfortable.
She will not refuse his sexual advances and afterwards will "replenish
his virility" by feeding him ghee and other milk products. She will bear

1»70 KIN GROUPS AND KINSHIP

and care for his children, keep his house, cook his meals, and do the
necessary household work in the fields and forest. In the words of an
informant,

A wife should have three qualities: (1) she should be beautiful, (2) she
should keep the house in order, (3) she should be able to cook good food
when guests come. When she has none of these qualities it is very sad; she
cannot expect much from her husband, and he may leave her. My wife has
no such virtues, but I cannot leave her. The chains of flesh are too strong —
I love my children.

As has been noted in the section on death ceremonies, widows are
not expected to retire from public life as is the ideal among high-caste
plains groups. Pahari widows mourn their husbands during the thirteen-
day mourning period, and many do so for a year. They take off their
nose ring and do not wear it again until they have remarried or until
they no longer wish to display their respectful grief for the dead hus-
band. The practice of sati, wherein a wife threw herself in anguish
upon her husband's funeral pyre, was popular among high castes in
the plains in the seventeenth and eighteenth centuries, but was ap-
parently never widely practiced in this area, perhaps partly because a
widow here had no unpleasant future to dread. One sati did occur,
however, in Sirkanda a few generations ago, and at least two popular
Pahari songs celebrate the stories of brave men whose wives destroyed
themselves after their husbands' deaths. In general, the Pahari attitude
is that a wife should respect, honor, and obey her husband in life but
that she has little obligation to him after the ceremonies immediately
following death.

In return for his wife's fealty, a husband is expected to provide for
her physical well-being, and is supposed to fulfill his obligation to have
sexual intercourse with her at least once a month, and preferably
oftener, as a woman is believed to have seven times the sexual energies
that a man has.

If a man has regular sex relations with his wife, she will be happy and do
anything for him. If he neglects her for a while she gets suspicious and rest-
less and may start an argument and accuse him of loving another. Divorce
is always due to a man's failure to satisfy his wife in this way. If he kept
her satisfied she would never look elsewhere and would have eyes for no
one but him.

Also, a husband may go out of his way to intercede for his wife in
intrafamily matters, to see that she has a chance to visit her family
periodically, and to provide her with things she wants or take her to
some fair or market outside the village occasionally.

KIN GROUPS AND KINSHIP 171

When a man has more than one wife he usually takes the second
with the approval, and sometimes with the encouragement, of the first.
If the wives are sisters it is expected that trouble between them will
automatically be at a minimum. If not, they may live in separate dwell-
ings, one in the village and one in a chan. In former days they more
frequently lived together, as chans were fewer, less productive, and
more isolated than today and were therefore less suitable for year-
round occupation. Whatever the arrangement, polygyny is a potentially
explosive situation. The first wife is traditionally the dominant one,
and her attitude toward the second is much like that of a mother-in-law
or elder sister-in-law toward a younger brother's wife. Skill and tact may
win more power to a second wife, but tradition is on the side of the
first. The first wife is especially likely to be jealous or vindictive if the
second wife has been taken to make up for her own failure to bear
children, for in such a situation she may feel threatened. The mother
of a man's children is likely to take precedence over other wives in the
eyes of both the husband and the extended family. However, wives are
usually separated if they do not get along together, and many who
live together do so congenially. They may even cooperate to press de-
mands on their common husband. A husband must always be careful
not to favor one wife conspicuously over the other. One man com-
mented in this context, "If you get two cups of tea from the teapot and
I get only one I will naturally be jealous. So it is with wives sharing a
husband." If a co-wife feels neglected she will react just as does a single
wife who suspects her husband of having another lover. She will sulk,
do her work poorly or not at all, and try to provoke an argument in
which to express her charges against her husband. If he is unable or
unwilling to make amends, she may run away.

The relationship between a wife and her husband's brothers is one
of considerable sexual freedom. Though polyandry is publicly ridi-
culed and abhorred by people in this area, covert sexual relationships
within the family are not greatly divergent from those among frater-
nally polyandrous families such as those of Jaunsar-Bawar (cf . Berreman,
1962a). Paharis are well aware of standard Hindu attitudes on poly-
andry and familial sexual freedom, so that these practices are generally
concealed from outsiders. Despite such precautions, they have attained
a reputation for deviation from orthodox Hindu behavior. A change
toward orthodoxy or modernity in both attitudes and behavior is
taking place as some of the younger and educated people object to tra-
ditional patterns of intrafamily sexual behavior and make their feelings
known.

The traditional view here is that a wife's sexuality may be freely

172 KIN GROUPS AND KINSHIP

used not only by her husband but by his brothers as well. One low-
caste elder drew laughter from listeners when, in providing genealogi-
cal materials, he listed his brothers' wives as his own in addition to his
own wives. His response was, "What's the difference? They are all
like wives to me." In discussing the matter other informants said, "A
woman would never refuse herself to her husband's brother because
he is in the same relationship as her husband and she would not like
to create discord in the family."

An elder brother has the right to make sexual use of his younger brothers'
wives. Younger brothers do the same so long as the age difference is not too
great. There is no quarreling or jealousy on this score. It is their right, and
they are expected to do it. They don't do it in the husband's presence. They
wait till he has gone out, if only to fetch water, and then approach his
wife. The wife will yield readily if she likes the brother, but even if she
doesn't he can insist. If she is angry she may tell the man's wife, and the
wife will scold him for forcing himself on the unwilling woman. Usually
such things are kept secret only from the wife who may be jealous. A brother
should not be jealous on these grounds. If a woman wishes she may send
her sister-in-law on errands in order that she may have access to the girl's
husband.

A woman is obligated to satisfy her husband's sexual wants and her
husband always takes precedence, but she should not deny the re-
quests of his brothers either. Informants were incredulous that rela-
tions between a man and his brother's wife could be expected to lead
to divorce, if discovered, in some cultures. A husband has no justifiable
grounds for complaint if his wife has relations with his brothers as
long as he is given precedence and is not denied his own sexual rights
and as long as the relations are carried on discreetly, away from the
house or in his absence. In reality husbands do occasionally complain
on this score. If a man's brother is overly attentive to the man's wife
he may abuse the brother by saying, "You have been eating my feces"
— a circumlocution which serves as a warning to the brother to pay
less attention to his sister-in-law.

If a wife is suspected of extrafamilial affairs or is going to work in
the jungle where such liaisons can easily be made, her mother-in-law or
elder sister-in-law may send one of her husband's brothers — often an
unmarried one — to accompany her or to work in the same area. This
protects the wife from the advances of nonfamily members and en-
courages relationships between her and her brother-in-law. "It pro-
tects our wives' honor from men outside the family." It is assumed that
every wife has at least occasional relations with her husband's brothers,

KIN GROUPS AND KINSHIP 173

and this is a typical way in which boys are initiated into adult sexual
activities.

Such freedom does not cross generational lines, and if it did it would
be considered a serious and reprehensible deviation. Generation is an
important boundary in all sex relations. This is reflected in the preju-
dice against relations between a very young man and a much older
sister-in-law. Brothers' and parallel cousins' wives are available as part-
ners; wives of sons, of brother's sons, of father's brothers are not. The
kinship terminology for a younger brother's wife, who is available
sexually, and for a son's wife, who is not, is the same, but this fact ap-
parently does not bear on the current behavioral situation. One case
was reported in Sirkanda in which a man made advances to his
daughter-in-law, but she was promptly recalled to her parents' home
and was not returned to her husband. A middle-aged and much-married
man who took as wife the wife of his young classificatory uncle was
widely criticized for crossing the generational boundary.

Great sexual freedom is practiced outside the extended family by
both men and women. It is assumed to be virtually a foregone conclu-
sion that if a man and woman meet alone in the jungle, intercourse
will ensue. The nature of the terrain and of the work engaged in by
both men and women make such meetings, planned or not, frequent.
Clandestine liaisons occur regularly, often with the connivance of the
female age-mates of the woman. There are well-known signals and
other means of arranging these. Few women are considered unap-
proachable in such matters, and it is thought to be unlikely that any
woman would refuse an insistent request. As in the case of the Lepchas,
". . . casual sexual relationships are so unimportant emotionally
. . . that few women would think it worth while making a fuss"
(Gorer, 1938, p. 160). While villagers may overestimate the frequency
of casual sexual contacts, it seems likely that most men and women of
the same general age and caste status in the village, outside of the
exogamous sibs, have had relations with one another. Some lasting
liaisons have been established. Instances of cross-caste and in-sib re-
lationships are also known.

The difference between all such relations and those between brothers
and their wives is that in the former cases legitimate grounds for com-
plaint exist, whereas within the family there is no such sanction. If a
husband wishes to press a charge of persistent adultery outside the
family he can usually win the support of the community to bring about
divorce, punishment, or a promise to desist. Even a wife may be sup-
ported in presenting such a charge. A similar charge against a brother

174 KIN GROUPS AND KINSHIP

would be laughed off or hushed up in the family as mere bickering.
In line with the shift toward adoption of plains behavior and San-
skritization in general, there is apparently a tendency away from the
pattern of sharing a wife's sexuality among brothers. Some high-caste
families, perhaps under the influence of the outside Brahmin who per-
forms their rituals and is the most vocal advocate of Sanskritization in
the village, express an orthodox Hindu view on this matter. They state
that an elder brother's wife should be respected like a mother, a younger
brother's wife should be treated like a sister or daughter. This view is
found in association with disapproval of bride price (also a Sanskritic
attitude), but both are often practiced by their detractors. One young
man in the village was quite upset by the fact that his wife had lived
openly, for over a year, with his elder brother in a chan. When she
returned to the family house to bear her husband's brother's child, to
be known officially as her husband's child, intrafamily tension grew
until the young wife suddenly discarded her nose-ring (symbolic of
marriage) and ran away to her own village. It was later found that she
was possessed by a ghost who impelled her to take this action, and
she nearly died of its affliction before she was brought back to her
husband's village, where the ghost was exorcised. After she gave birth
the problem was solved by a familial announcement that the baby was
the elder brother's. Thenceforth the baby's parents were considered
man and wife and a new wife was to be procured for the dispossessed
younger brother. It was largely at the insistence of the younger brother
that this solution was reached.

The relationship among nuclear families in the joint family is above
all a combination of the various specific relationships mentioned above.
Nuclear families or larger subdivisions of the extended family are the
units which remain together when the extended family breaks up.
Despite the frequency of divorce, there is a unity within the nuclear
family which in some cases is stronger than that which runs through
the extended family. The break-up of extended families follows mutual
agreement among the separating units that they can get along better
separately than they could together. Most often the break-up is among
cousins (sons of brothers) and is blamed on their wives. It is said that
wives drive a wedge between brothers (classificatory as well as "real"
brothers). A wife complains to her husband that she and he are doing
more than their fair share of work in the family and perhaps that
other nuclear units are expending more than their share of the in-
come. Behind this is often sister-in-law jealousy, and particularly re-
sentment of the domination of an elder sister-in-law. The husband
finally becomes convinced or willing and expresses his desire to separate.

UN GROUPS AND KINSHIP 175

Eventually a division may be effected. Wives are often blamed for what
s really fraternal strife. Brothers are supposed to get along with one
mother harmoniously and to respect their elders. Therefore, if they
:annot get along in the joint family, it is socially more acceptable to
)lame their strife on disputing wives who, as outsiders, do not reflect
m the family reputation as directly as do brothers.

Division of the joint family may thus be brought about by disputes
tmong brothers or cousins, though for sake of appearance it may be \

attributed to friction among wives. The Brahmin family in Sirkanda
vas on the verge of dividing its lands into three parts in 1958, as a re-
ult of the feeling by each of the three Brahmin brothers that he was
:ontributing more than his share and getting less in the joint economic
irrangement. This split was to occur despite the fact that their elderly
nother was still living. As a first step in the dissolution of this joint
amily, the fields were being marked off into three equal parts and a
arge house containing three separate dwelling units was being built
o replace the less spacious one the joint family had shared. In this
nstance the old house furnished part of the materials for the new one.
n other cases the old house has remained in use by one of the new
amily units, has been given to a Dom for occupancy, or has been
bandoned. As has been pointed out, low-caste brothers more frequently
livide their property than do those of high castes. This difference is
vidently related to their occupations. In the agricultural high castes
and can best be worked by large joint families, but the occupations
»f artisans can be efficiently performed by smaller family units.

While Sirkanda and its component extended families have increased
n population, there has apparently been no commensurate increase
n the frequency of breakdowns into smaller family economic units
is is reported for other parts of India. Some informants, in fact, re-
>orted that proportionately fewer splits occur now than in former
lays; that joint families have become larger rather than smaller. This
rend, which is not in the direction one might expect, is attributed to
he fact that in former days the entire joint family occupied a single
touse where its members, and particularly the wives, were constantly
n contact with one another and disputes were frequent. Disputes led
o disintegration of the joint family because disputing groups within
t could only avoid conflict by moving out, and moving out meant
etting up a new household. Few families had chans in those days, and
:hans were isolated and surrounded by relatively little cultivated land,
o that they were not suitable for year-round living. Now most families
lave chans or second houses. There are several chans at most locations,
ind the lands have been developed so that year-round occupancy is

176 KIN GROUPS AND KINSHIP

possible and even necessary in terms of the crops and animals to be
cared for. Many nuclear family units are now separated from others
within the same joint family. They thus avoid the friction that is
inevitable in the close daily contact of a shared dwelling. With less
frequent disputes and with the possibility that a disgruntled nuclear
family can live in another dwelling, there is less pressure to divide
family property. The joint family remains intact as an economic unit
while it no longer remains a residential unit.

Between half-siblings of polygynous marriages there is often no love
lost. This probably reflects the relationship between the co-wives who
were their mothers. Such rivalry is common knowledge among villagers
and finds expression in folklore. Frequently in adult life the dislike
takes the form of merely ignoring the half-sibling or having little or
nothing to do with him. Sometimes it takes the form of continual dis-
puting or unfriendly rivalry. In one family in Sirkanda it is the basis
for a four-generational factional split. In most of the reported cases
half-brothers have divided their father's property into separate eco-
nomic units after his death.

Lineage

The patrilocal extended family is the most important social, eco-
nomic, and ritual unit in most phases of life in Sirkanda. There are,
however, more inclusive kin groups which are relevant in certain con-
texts. The smallest of these may best be termed the "lineage" — a con-
sanguineal kin group that traces common descent through known an-
cestors, in this case male ancestors and therefore a patrilineal group
(see fig. 3). In practice in Sirkanda this unit usually excludes married
sisters and daughters and includes wives of the male lineage members.
As such it resembles a "compromise kin group" in Murdock's formu-
lation, combining unilineal descent with residential unity, so that
lineage members who live away from the group are excluded from it
and nonlineage members who live with the group are included in it.
I will here refer to this unit in Sirkanda as the lineage because the uni-
linear rule of descent is paramount; lineage ties do not disappear in
fact or theory at marriage. This is indicated by the fact that women are
accepted heartily as lineage members when they visit their parents'
homes. Lineages, as segments of sibs, are strictly exogamous. A woman
has sex relations with her husband's brothers, real and classificatory,
but never with people of her own lineage or sib. When a woman is
widowed or divorced she either marries her husband's brother or is
accepted back into her parents' household, where subsequent marriages

KIN GROUPS AND KINSHIP 177

are arranged with reference to her parents' lineage and sib rather than
that of her husband, that is, for considerations of exogamy. The lineage
of her erstwhile husband is disregarded except that it is reimbursed if
she leaves voluntarily. It should be remembered, however, that resi-
dence, in combination with lineage, is important in daily life. As long

Jawari Rajput Lineages

Lineag

• =
1

e I

A = •
1

= •
1

Lineage

=A
1

• = A

•=i • = !:

A A A «|i

1
•=A

•=A

1
•=A

1 1
L -A

1
A

Minimal Lineages A,B,C
(Relationship to one anothe
A-""

Lineage III (Maximal

funcertain) \

NB
•=A

Lineage)

~~~~~~— - C

1 1

•=A

1 1

1
1

l 1

l l
•=A *=A

1
•=A

1 1
A 0=A

1
A

1 1

A A

1
A

•=A

Low-Caste
1

Lineages
•=A

, 1

1
•=A

A A(l

1 1
A A

1
A

mmigrant)

Blacksmiths

3ajgis

A deceased ancestor

A living household head

Fig. 3. Sample lineage structures.

as a woman's husband lives, she participates in the ritual, social, and
economic activities of his lineage and sib. If she dies, her funeral is
performed by her husband's family, not her family of origin. But she
retains her identification with her lineage of origin. In short, she be-
comes a member of her husband's clan, the compromise kin group de-

178 KIN GROUPS AND KINSHIP

termined by descent, residential unity, and social integration (Mur-
dock, 1949, p. 68) but not of his sib.

The lineage is termed khandan locally, and its component house-
holds are called "hearths." In Sirkanda there are few households which
do not belong to lineages containing several households. As genera-
tions pass, lineage connections in the larger village castes grow dim
and eventually are forgotten as subsidiary divisions assume lineage
proportions. The male Sirkanda members of each caste except Rajputs
— the Brahmins, blacksmiths, barbers, and all of the Bajgis except the
immigrant one — belong to single lineages as well as single sibs within
their respective castes. There are at least nine Rajput lineages.

Three of the Rajput lineages in Sirkanda are named. They have
acquired the names of honored ancestors from whom their members
trace common descent. One of these lineages is well-known for its
devotion to the worship of the Pandavas and for the special attention
the Pandavas show its members in return. Corroborative stories are
told of this relationship. In two of these lineages the common descent
of members can be accurately traced. In the third there are three
named sublineages or minimal lineages within which common descent
is easily traced but between which the relationship has grown hazy.
If the names persist and the relationship of the sublineages is for-
gotten, the sublineages may be elevated to lineage status and the present
lineage would then become a kind of sub-sib and might conceivably
assume sib significance, just as sibs (jatis, see discussion below) may
have succeeded gotras as sib units in this area. At present the three
sublineages together resemble a maximal lineage in that together they
are considered a lineage relative to others, while each of the three
component sublineages is considered to be a lineage relative to the
other two.

Four other Rajput lineages are distinct, while the remaining two
are known to be related to one another but have drifted apart as the
ties of relationship have been forgotten. None of these is a named
lineage.

The typical lineage contains three or four households. The largest
contains eleven, but it is the one which is now reduced in reality to
three sublineages. The largest intact lineage has seven component
households. Beyond the household it is the members of the lineage
unit who are most likely to participate actively in the life-cycle cere-
monies of their members. They tend to support one another in social
situations, to worship the same gods, and, of course, to honor the same
ancestors. In short, they share a feeling of common identification and
a somewhat more homogeneous culture than occurs across lineage

KIN GROUPS AND KINSHIP 179

lines, though less homogeneous than that within the extended family.
There is also a tendency toward residential proximity within the
lineage, based apparently on common origin. The lineage has in-
herited from people who once held property in common, and it is part
of this property upon which houses have been built. Factions tend to
follow lineage lines, but here the exceptions almost equal the examples
in their frequency.

Kindred

Not all the people significantly related to one another in Sirkanda
are related through the male line. There is a circle of kinfolk recog-
nized as relatives of the individual, who are related to him through
either one or the other parent, that is, bilaterally. They will be termed
the kindred, following Murdock (1949, pp. 46, 56, 61). The kindred
form a group only from the point of view of the individual to whom all
are related. They do not comprise a self-conscious identification group.
The kindred's common interest is focused specifically on an individual
who ties it together on certain occasions, notably life-cycle ceremonies.
Such bilateral relationship is widely recognized in Northern India. The
common name for the kindred is sapinda, mentioned above (p. 158) in
the discussion of marriage regulations. Although this term is not com-
monly used in Sirkanda, the concept is prevalent.

Those who are born of one body are Sapinda. A child is a Sapinda of his
father and mother. He is also the Sapinda of his father's brothers, sisters,
father and grand-father as also of his mother's brothers and sisters, father and
grand-father. He is Sapinda to his cousins (father's sister's children, mother's
sister's children, father's brother's children and mother's brother's children)
as he shares with them common body-particles either from his mother's side
or from his father's side. In this meaning of the word Pinda, a common
kinship of blood is established with both the father's and mother's side.
(Karve, 1953, p. 55)

As has been noted, one cannot marry within the sapinda group,
though its boundaries are variously defined even for this important
function. In other situations the kindred is variable in its extent and
inclusiveness. As in its function of regulating choice of marriage part-
ner, the function of the kindred in ceremonial activities is parallel to
that described by Murdock (1949, pp. 56 f.) for American society
where ". . . its members are collectively called 'kinfolk' or 'relations,'
[and] it includes that group of near kinsmen who may be expected to
be present and participate on important ceremonial occasions such as
weddings, christenings, funerals. . . ." Kindred serve many of the

l8o KIN GROUPS AND KINSHIP

same functions as lineage members, and in fact by definition kindred
membership overlaps greatly with that patrilineal group. Participation
of kindred in ceremonial functions is likely to be less intensive than
that of the lineage simply because kindred are residentially separated
and include relatives whose relationship to one another is traditionally
one of mutual respect and even avoidance. They do not identify them-
selves as a group. Kindred are not bound by diffuse, interconnected
group ties. It is in this crucial respect that the kindred differs from the
lineage group, which is closely bound by ties of consanguinity and com-
mon identification to which any particular individual is incidental.

Sib

Above the lineage in degree of assumed patrilineal relationship
is the sib (jat or more often jati in local parlance). In this named kin
group, relationship in the patrilineal line is assumed but cannot be
traced. It is the effective exogamous unit within the caste, and it ex-
tends across village boundaries. It occurs most prominently among
Brahmins and Rajputs. Judging by sib names and local tradition, some
sib names may have derived from a former place of residence of the
group — indicating, perhaps, that local exogamy was there the rule. If
some sibs did originate in this way, local exogamy ceased to be the rule
when people emigrated from the villages in which it was practiced to
new localities, such as Sirkanda, where people of other villages or areas
settled. The other migrants were potential mates because they were
from other villages, and their common residence in Sirkanda was
deemed irrelevant for marriage purposes. That is, the ancestral vil-
lages of some Sirkanda Rajputs may have been exogamous. If so, when
they settled in the present location, they may have excluded from
marriageability their former village affiliates but not the members of
their new local group. In that way, former village affiliation would
have become important so that the names of those villages were used
to identify the exogamous group (hence sib). Other groups evidently
carried their sib names with them from the plains or adopted those of
plains groups. The latter explanation seems the most probable in light
of available evidence.

As has been mentioned in chapter 1, some 116 Rajput sibs have been
listed for Garhwal. These include three of the four found in Sirkanda
(Raturi, 1928, pp. 167 ff.). The names of these three (Jawari, Palial,
Khandial) are attributed in that listing to the names of the villages in
which their ancestors settled. Whether the sib got its name from the
village or vice versa is a moot question. The village name was likely

KIN GROUPS AND KINSHIP 101

derived from its early settlers or founders, though it may then have
been adopted by subsequent immigrants. Tod (1829, p. 120) lists
"eighty-four mercantile tribes" of Rajasthan among which are "Pilli-
wal," "Khandailwal," and "Kakulea," which are apparently three of
the Rajput sib names of Sirkanda. He suggests that the Palial (Pilliwal,
Palliwal) group originated in Palli, the great commercial market of
western Rajasthan. "A community of Brahmins then held Palli in
grant . . . whence comes a numerous class, termed Palliwal, who fol-
low mercantile pursuits" (Tod, 1829, p. 700). Bahadur (1916) mentions
a "Pyal" Rajput group (an alternative pronunciation, according to a
Sirkanda informant) which is said to have come to Garhwal from Delhi.
This could well be the same group or a branch of it. Villagers them-
selves retain no traditions on the matter beyond assertions that their
people came from various locations in the Kumaon Hills to the east
and that sib names indicate the village of origin. The sib names of
two village wives were said to be the names of their villages of origin
in Garhwal. One village man said, "Jati is named for the place where
a member of the caste went and settled. His descendants carry this
name." Another attributed it to the name of an ancestral leader who
brought his people to some new locality. A glance at the village map
(map 3) will show that sibs retain a pattern of residential proximity,
though it is not rigidly followed and most villagers are not even aware
of it. It probably reflects the fact that sib ancestors settled in one loca-
tion when they first came to the village as a group, and that this loca-
tion has remained somewhat intact through patrilineal inheritance of
land and houses.

The low castes claim such kin groups and utilize them as exogamous
units just as do the high castes. Among low castes the units are ap-
parently of varied origin — some are the jati, gotra, or caste names of
alleged high-caste ancestors. Some are the personal names of particular
ancestors, and some may be place names. Some may well be lineage
names of fairly recent origin. The low castes may often have adopted
names in imitation of high-caste practice, as, in fact, Khasiya Rajputs
may also have done in earlier days. In any event, the named groups are
in effect sibs, and they are referred to as jatis or gotras. Insufficient
data were collected on low-caste marriages of any one caste to determine
precisely the extent to which their "sibs" are important or to verify
their functioning. The data that were collected and the testimony of
informants showed no consistent differences at this level between high-
caste and low-caste sibs in marriage regulation. Both conform equally
well to the definition of sib. Differences are apparently of origin and
extent rather than of local functioning.

182 KIN GROUPS AND KINSHIP

Since all non-Rajput castes in Sirkanda contain only single lineages
(except for the single immigrant Bajgi), they are also made up of single
sibs; for example, the local Brahmins are of the Kotari sib. Among
Rajputs four sibs are represented: Jawari, comprising 20 households
and three main lineages, including the one which is in effect three
sublineages at present; Palial, comprising 11 households and three
lineages; Kukhalwal, comprising five households and two lineages; and
Khandial, comprising a single household and lineage. Within these
sibs the lineage divisions are sufficiently recent that the tradition of
interlineage relationship is strong, strong enough that the term khandan
(lineage) is applied to each of the four local Rajput sibs as wholes as
well as to their component lineages. A Rajput elder's comment on the
subject was typical: "Our Khandan divided seven generations back,
but now no one remembers who the people were who divided it." In
other contexts people distinguish khandan as lineage from jati as sib.
The term biradari (brotherhood) is also used to designate the sib.
Among low castes this term is applied to the entire caste group.

The significance of sib groupings is primarily in establishing bound-
aries of marriage eligibility. Members of the same sib are never po-
tential marriage partners. Neither are people whose mothers were in
the same sib. A secondary function is participation in ritual. The entire
local sib unit is supposed to observe a degree of ritual pollution after
the death of a member, and to a lesser extent the sib functions as does
the lineage in other life-cycle ceremonies. Actually, since the group
involved in ritual is localized and for some purposes includes wives
of sib members, it is perhaps more properly termed the clan. It appears
likely that the lineages may eventually replace sibs or clans in these
functions, as sib and clan responsibilities in this context seem to be
decreasing in importance in the minds of villagers.

Sibs and clans also enter into faction formation. Although there are
many factions within sibs, and factions which separate sibs, relatively
few cut across sib lines allying members of different sibs against their
sib-fellows. That sib identification is still a real factor in the village is
indicated by sib loyalties. In reference to the founding of Sirkanda,
each of the larger sibs claims precedence. A member of either one is
likely to disparage the members of the other, especially to compare
them unfavorably in terms of wealth, honesty, or Sanskritization of
ritual observances and marriage regulations. An old Jawari, head of
the largest household in his sib, said, "We Jawaris were the first Raj-
puts here. Then those Palials showed up, and our forefathers let them
settle here too. We don't know where they came from or who they
are. Hell, we don't even know if they're Rajputs — we just took their

KIN GROUPS AND KINSHIP 183

word for it!" The speaker's wife of 40 years, mother of his five adult
sons, is a Palial. There is considerable rivalry between the sibs, as
evidenced in all their relationships and recorded in disputes, court
cases, and recently in voting patterns in elections of village officers.
The divisions are not rigid, but they do show up.

There is no concept of hierarchical ranking of sibs in the village —
all are equally prestigeful. Some sibs or larger subcaste groupings out-
side the village are thought of as more or less prestigeful. Among the
Brahmins and low castes prestige of subcastes is often related to oc-
cupational specialization. Ceremonial cooks (Sarola) are ritually purer
than family priests. Carpenters often consider themselves higher than
blacksmiths. Among Rajputs and some Brahmin groups the place of
origin is important. Plains groups are more prestigeful than Pahari
ones; that is, those admitting the appellation "Khasa" are lower than
some of those denying it. Among some low castes prestige is related to
alleged high-caste ancestry. These distinctions do not come into play
in the village and, except in the case of Sarola Brahmins, they rarely
crop up in Bhatbair. With the exception, again, of Sarola Brahmins,
they are unimportant in determining marriage relationships within
this area, although they do assume significance in some outside ar-
rangements. No evidence of subcaste hypergamy appeared in the
research.

Clan

The clan in Sirkanda is made up basically of the local members
of one sib together with their wives.10 As such it is very closely iden-
tified with the sib, and its functioning often cannot readily be dis-
tinguished from that of the sib or the lineage. It is an unnamed, overtly
unrecognized group which can be of considerable analytic utility. Its
core is the local portion of the patrilineal sib. It is distinguished from
the sib by the fact that it is a compromise kin group (Murdock, 1949,
p. 66), that is, it includes some people on the basis of their local
residence who are not tied to it by consanguinity, and it excludes
some on the basis of their outside residence who do have consanguineal
ties to its members. It includes the wives of local members of the sib,
and it excludes their married sisters and daughters. It also excludes
sib-fellows of other villages and areas. It functions to include the
occasional adopted son who is of another area or another sib, and the
rare man who marries into the local sib matrilocally. Conversely, it
excludes the local boy of the sib adopted elsewhere and the rare
local man of the sib who marries outside and lives in his wife's village.

184 KIN GROUPS AND KINSHIP

The third criterion of a clan (in addition to a uniiineally related
core and residential unity) is social integration. This is a feature of
Sirkanda clans which becomes most apparent at times of ceremonial
participation. Confusion easily arises at this point, since villagers
themselves do not recognize the clan as a group. When describing
life-cycle and other ceremonies, they describe the participants as mem-
bers of the family, lineage, sib, kindred, caste, or community, depend-
ing upon the nature of the participation. In their descriptions of sib
participation, however, it is often apparent that they are including
local members of the sib and their spouses and excluding nonlocal
members of the sib, that is, they are describing participation of the
clan, as that term is defined by Murdock. Even "lineage" participation
often proves to include wives of lineage members and to exclude
lineage members who live elsewhere; it includes the most closely asso-
ciated clan group centering on the lineage, rather than the lineage
itself.

The concept of the clan, as distinguished from the sib, provides a
consistent explanation for a number of otherwise perplexing features
of Pahari social organization. First, of course, is the composition of the
groups which participate in ceremonies. While family, kindred, caste,
and community are social units which can account for many of the
non-unilineally related participants in ceremonies, the clan more suc-
cinctly defines the local, socially integrated group centered on a
uniiineally related sib, but including other individuals, which par-
ticipates in several kinds of ceremonies.

Second, the clan concept clarifies the complex matter of changes in
group affiliation, identification, and inheritance which accompany
changes in residence at the times of marriage and adoption (in the
relatively rare cases where adoption occurs across sib lines). Without
the concept of clan this has proved difficult to understand. Wives,
adopted sons, and matrilocally resident men stoutly deny that they
have altered their sib affiliations when they go to live among people of
another sib. Yet some informants claim that they have done so. Cer-
tainly, for some purposes, they have severed ties with their natal groups
and have established ties with new groups. To explain this in terms
of complex alterations in sib affiliation and responsibilities, as is
usually done, is both confusing and inaccurate. Sib ties have not been
altered. What has happened is that clan affiliation has changed. The
individual has ceased to be a member of one localized clan and has
become a member of another. If the relationship is terminated and
the person returns to his natal group (as when an adopted son leaves
his foster family) or joins yet another group (as when a divorced

KIN GROUPS AND KINSHIP 185

woman remarries) then clan affiliation is changed again without diffi-
culty. Clan affiliation is as easily changed as is residence, but sib
affiliation is inherited and unalterable. This is the affiliation which is
important (by delimiting the exogamous group) in determining mar-
riage. Thus, a woman of sib A and clan a who marries a man of sib
B becomes a member of clan b when she goes to live with him, but
she remains in sib A. She has sexual relations with her brothers-in-
law in sib B as well as with her husband, and upon being widowed
or divorced she may marry someone else in sib B. If it were claimed
that upon marriage she had become a member of sib B, these relation-
ships would be incestuous and therefore prohibited.

When a man dies, the wives of his local sib-fellows participate in
the mourning insofar as women participate at all. His married
daughters and nieces do not. When a wife dies, her major funeral
ceremonies are performed by the clan-fellows she acquired by mar-
riage, including those extending beyond the extended family, but are
not performed by her sib-fellows. A lingrd stone (see Appendix IIIC),
symbolizing that the deceased woman is one with Shiva, is placed with
those of the rest of the dead of the local clan group. It is not sent
back to rest with those of her sib-fellows. A woman participates in
the life-cycle ceremonies of her clan-fellows much more consistently
and actively than in those of her sib-fellows. Moreover, an adopted
son or a son-in-law resident in his wife's father's house can perform
the father's funeral rites and can inherit his property. In this manner
property can pass from one sib to another. The lingra stones of
adopted sons and resident sons-in-law are placed with those of the
family with which they reside. The stones are not sent back to rest
with those of their family of origin. This practice, like inheritance
by an adopted son, does not depend upon a change in sib affiliation;
rather it depends primarily upon place of residence and therefore
upon clanship. A sib-fellow, lineally related, is preferred to perform
funeral rites and to inherit property, but failing such a person, a
clan-fellow can substitute. One who lacks both sib and clan ties (for
example, a son-in-law resident in his natal household) is not eligible
under normal circumstances. These facts emphasize the social struc-
tural importance of the residence group. If funeral participation is
described as primarily involving the household, lineage, and clan, with
lesser participation by the sib, kindred, caste, and community; if
patrilocal marriage is viewed as entailing a change in clan affiliation
for the woman, and if adoption and matrilocal residence are viewed
as entailing a change in clan affiliation for the man, then these prac-
tices are consistent and more readily understood than if one were to

i86

KIN GROUPS AND KINSHIP

assume that there is no significant change in group affiliation at mar-
riage or adoption or that sib affiliation is changed at these times.11
Therefore, the local compromise kin group, the clan, assumes greater
significance in ritual and ceremonial participation (especially that
surrounding life-cycle ceremonies) than does the sib. The unilineal

V, Sib
'x-'Clan

KEY

Head of household
Ultimate recipient of
property of household head

"Normal" household (patrilocal)

Matrilocal son-in-law (x)

,'<g>0

I
I

IO=A(x) i

Matrilocal son (x) Adopted son (x)

Fig. 4. Sib and clan affiliations in Sirkanda families. From Berreman (1962c).

descent group, the sib, determines eligibility for marriage and is the
preferred means for determining inheritance. It is the core around
which the clan is formed. Sirkanda villagers lump these analytically
distinguishable groups under a single term, jati, which literally refers
to the sib. They tend to think of wives as members of their husbands'
sibs in the ceremonial context and as members of their natal sibs

KIN GROUPS AND KINSHIP 187

in most other contexts. They think of adopted sons as members of the
family of adoption for purposes of ritual and inheritance, and as
members of their natal families for purposes of marriage. In most
cases (other than those involving determination of limits of exogamy)
they think of the sib in terms of its local membership, that is, those
sib members who live in one village or in closely adjacent villages.
A clear distinction between clan and sib, as such, is not made by
villagers, but it is implicit in their own conceptualization of their
social organization. This, combined with its analytical utility, make it
worth distinguishing. The "clan" is somewhat less sharply defined —
less a corporate group — than one would wish if use of the term
"clan" were to be fully justified.

Gotra (Phratry)

The exogamous unit among high castes over much of India is
called the gotra. In many areas this unit amounts to a patrilineal sib
(cf. Lewis, 1958, p. 23; Dube, 1955, pp. 42 f.). Gotras appear in
Sirkanda and are said by educated villagers to derive from the names
of twenty-four great religious teachers or ascetics who founded vast
lineages whose descendants in the male line retain the name of
the founder as the gotra name. A local Brahmin explained:

Gotras are called by the names of great rishis who were founders of the
original families. Bharadwaj was a great rishi. Some of his children went
into religious work, and their descendants are Bharadwaj Brahmins. Others
among his children went into the work of governing, and their descendants
are Bharadwaj Rajputs. Jawari [a sib name] was the name of the headman of
a group of Bharadwaj Rajputs who came and settled in one place in these
hills. Now his descendants are Jawari Bharadwaj Rajputs.

Tod (1829, P- 27) explains gotra origin in a similar way. Others
describe Brahmin gotras as deriving from a lineal ancestor's name,
while Rajput gotras derive from the name of the religious preceptor
(guru) of a lineal ancestor. In any event, the same gotras are found
among both Brahmins and Rajputs in this area, but among no other
castes, and remote common ancestry for Brahmins and Rajputs is
assumed by some.

On the basis of his research in "Kalapur," a plains village, Hitch-
cock (1956, pp. 43 ff.) refers to this usage of the term "gotra" as
"Brahmanical." "Nothing in our data shows that the gotra concept
in the Brahmanical sense now has any functional significance for
marriage . . ." (ibid., p. 45). However, he reports that it is used in

1 88 KIN GROUPS AND KINSHIP

a second sense (the sense common on the plains) to refer to the sib —
a usage parallel to the Pahari use of the term "jati" as described here.

In Sirkanda the low-caste sibs are sometimes termed "gotra," and
sometimes "jati." Only the high castes have both units consistently
and distinguish them precisely. Even among them "gotra" does not
mean what it means among many plains groups. The term "gotra,"
as Hitchcock's comment would suggest, does not mean the same thing
throughout India to those who use it. "Sometimes a caste is divided
into exogamous groups called Gotra with no further division. Some-
times a caste is divided into exogamous Gotras or endogamous Gotras
with further smaller divisions" (Karve, 1953, p. 118).

In Sirkanda there are two gotras among the Rajputs: Bharadwaj,
which includes the sibs Jawari and Kukhalwal, and Kasib, which
includes the sibs Palial and Khandial. The Brahmin gotra in Sirkanda
is also Kasib. Each gotra includes many sibs — an estimated 20 to 50
in each gotra — but only two of each are represented among Sirkanda
Rajputs. In addition, there are at least ten other gotras in this
general area.

Both the Sirkanda gotras receive mention by Tod as having been
present in Rajasthan. He refers to an ancient lineage in which, in
the fifteenth recorded generation, nine brothers " '. . . took to the
office of religion, and established the Causika Gotra or tribe of
Brahmins.' From the twenty-fourth prince in lineal descent from
Yayat, by the name of Bhardwaja, originated a celebrated sect, who
still bear his name and are the spiritual teachers of several Rajpoot
tribes" (Tod, 1829, p. 27). Rajputs and others took the gotra names
of their priests when they adopted the gotra system (Karve, 1953,
p. 115). This constituted an early step in Brahmanization or San-
skritization.

Gotra, like sib, is an inherited affiliation passed down patrilineally.
Since sibs are subdivisions of gotras, the gotra contains many sibs
but one sib never spans two gotras. Gotras therefore fit the definition
of phratry, wherein ". . . two or more sibs recognize a purely con-
ventional unilinear bond of kinship . . ." (Murdock, 1949, p. 47).
Among Rajputs in Sirkanda this would amount to a moiety division
because there is a division into two phratries. This is fortuitous:'
neighboring villages have one, three, or more phratries.

In Sirkanda and the surrounding region, gotra is not now a func-
tional unit in marriage regulations. Joshi (1929, p. 272) states, "The
Khasas have no real gotras. Mr. Atkinson noted that they all stated
themselves to belong to the Bharadwaj gotra, and had no idea of what
'gotra' meant." As a Sirkanda Rajput said, "People must marry in

KIN GROUPS AND KINSHIP l8g

their own caste and they cannot marry in their own sib or their
mother's sib. That is all that matters in marriage. Gotra is just a
name, it doesn't mean anything." In this respect these people are
not unique, as indicated by Hitchcock in the discussion noted above.
Karve reports that among the Rajputs of Rajasthan and among the
Kayasthas of Uttar Pradesh, gotras do not seem to have any function
in marriage relations (Karve, 1953, pp. 119, 141). On the other hand,
ultimate common ancestry in the gotra is admitted. This is consistent
with Murdock's point (1949, p. 47) that the larger the kin group,
the less likely it is to be exogamous.

Among Sirkanda Rajputs there are six possible sib combinations
for marriage within the village based on the division into four sibs.
All six have occurred among the 53 marital unions that were recorded
as having been contracted among the four village sibs. Of the six
possibilities, two are in-gotra combinations and four are cross-gotra
combinations, based upon the division of the four sibs into two gotra
groups (phratries or moieties). Since the two large sibs are in opposite
gotras and since four of the six possible sib combinations in marriage
are gross-gotra ones, cross-gotra marriages would be expected to pre-
dominate among those contracted within the village. This proves to
be true, with 39 of 53 marriages across gotra lines. However, 14 in-
gotra marriages were also recorded, virtually all of which were first
marriages. This is actually greater than the expectable proportion.
Half (5 of 10) of the in-village marriages of members of the smallest
sib (Khandial), for example, were in-gotra ones, whereas well under
half of all potential mates in the village for these people have been
in their own gotra (at present 38 per cent of the village population
is in this gotra and out of this sib).

Almost two-thirds (9 of 15) of the in-village marriages of the other
small sib (Kukhalwal) were in-gotra ones, although potential mates
in the village for these people have been about equally split between
the same and opposite gotras (at present 58 per cent of the population
is in this gotra and out of this sib). The relatively high rates of in-
village and in-gotra marriage in small sibs reflect the larger number
of potential mates in the village for members of small sibs (those with
small population in the village) as contrasted to members of large sibs.

I do not suggest that in-gotra marriage is preferred. There is a
tendency in this direction within the village, but it is not based upon
a large enough sample to establish significance, and in any event when
extravillage marriages are included it is an even less impressive trend.
What I have demonstrated, however, is that gotra does not enter
into regulation of marriage contracts in this area. There is no prefer-

igO KIN GROUPS AND KINSHIP

ence for cross-gotra marriages over in-gotra ones. This substantiates
verbal testimony. No Sirkanda informant expressed a feeling that
gotras should play a part in marriage arrangements, and no case was
reported in which they did. Few villagers know their gotras and fewer
yet know those of their wives, whereas everyone knows his own sib,
and many know the sibs of their wives. Precisely the same situation
obtains for Brahmins, except among those who are better-informed
because they deal in horoscopes and life-cycle rites where gotras are
mentioned. Interestingly enough, the "gotra puja" is an integral part
of every Sanskritized Pahari marriage ceremony, which includes most
first marriages these days. In this puja, taken directly from the religious
prescriptions of the "great tradition," the gotra affiliations of the bride
and groom are repeated many times over in the recitation of the
sacred verses. The history and honor of the gotras that are being
joined by marriage is reiterated. In Sirkanda the gotras so described
and honored are sometimes one and the same. This is perplexing to
many high-caste non-Paharis who learn of it. However, even the most
sophisticated of the Brahmins who function in Bhatbair see no con-
tradiction in the practice. It may be expected that plains emulation
will result in the attaching of greater importance to gotra in the future,
as has happened among upwardly mobile castes of the plains.

The gotra unit was found to be virtually nonfunctional among
Sirkanda Rajputs, in both theory and practice, in every sphere of
life. The only suggestive evidence that the gotra is of any importance
is the fact of residential proximity of several households of the two
sibs comprising the Bharadwaj gotra in the easternmost of the two
Sirkanda settlement areas. This is probably the original village site
(map 3). If, as is likely, this is more than a chance occurrence, it is not
recognized as such in the village and it has no detectable ramifications
beyond those of residential proximity found elsewhere in the village.
I would speculate that it represents early common association or com-
mon origin of the two sibs, so that when they settled here they built
houses near one another. More recent building has not followed this
pattern and it is not noticeable in other sections of the village, al-
though, as has been noted previously, residential unity of lineages
and sibs is found fairly consistently.

Among the lower castes there are apparently no gotras in the sense
of phratries such as those found in the high castes. Some claim to
have them, but these usually prove to be sibs. Sometimes non-sib names
are given in addition to sib names but they are usually lineage names,
names of ancestors, names of localities, or subcaste occupational terms
and are recognized only over a very limited area by fellow caste-

KIN GROUPS AND KINSHIP 1C)1

members. Sometimes the names apply to precisely the same group
as the sib name, even though they are given as gotra and are dis-
tinguished from jati. It is not unlikely that real phratries do exist
among some low castes in this area, but none was found in Sirkanda.
Such claims are apparently attempts at achieving Sanskritic respect-
ability. Said a Bajgi, "We are just like Rajputs and Brahmins in that
respect. We have the same divisions and marriage rules. Some other
low castes don't have that, but we do." The high castes do not concur
in this: "The low castes have regulations, but they don't have the
same kind we do and they are lax about these matters."

Kinship

In order to understand the aggregates of related individuals that
have been referred to above as kin groups, it is necessary to understand
the kinship system, the system of relationships defined in terms of
consanguinity and affinity which exists among people. This is codified
in the system of terms used by people in addressing and referring to
one another. The terms are given in Appendix IV. In Sirkanda, terms
of reference and address are generally identical, although the latter
are sometimes contractions of the former (didi jl becomes dijl). The
most prominent exceptions are cases in which terms of reference are
avoided in address between persons who stand in a relationship of
respect or conventional social distance. Sometimes in reference de-
scriptive terminology is used for clarity, especially when a term applies
to more than one kind of relative. Specific exceptions are noted below
and in Appendix IV.

In order to characterize Pahari kinship relative to other systems
that have been reported, we may refer to the criteria by which kin
are classified suggested by Murdock (1949, pp. 100-106).

In Pahari kinship, generation is specified with these exceptions: the
term for wife (reference) and for junior female affine (address) and
that for husband (reference) and young male affine (address), apply
in all generations. Terms for father's brothers apply also for spouse's
sisters' husbands. Sex is specified in all terms. Affinity is specified in
most terms of reference. That is, consanguineal relatives are not
referred to by the same terms as affinal relatives. The primary ex-
ception to this is a pattern of equating close affinal relatives within
a generation to consanguineal relatives of that generation. A man
refers to his wife's brothers' wives as he would to his own sister, and
a woman refers to her husband's sisters' husbands as she would to
her own brothers. A person's mother's brother is terminologically

10,2 KIN GROUPS AND KINSHIP

equated to his father's sister's husband, and his father's brother is
equated to his mother's sister's husband. The same equation is made
in the grandparental generation. This resembles a moiety distinction.
It is elaborated to the point that a man refers to his wife's sisters'
husbands as "co-brother" or "half-brother." He addresses and refers to
his mother's sisters as "elder mother" or "younger mother," the same
terms by which he addresses his father's co-wives. There is a special
term of reference for father's co-wives. A woman addresses her hus-
band's brothers' wives as her sisters, although they are distinguished by
terms of reference. Her husband's sister's husbands are referred to
as her own brothers. Thus, members of one's own patrilineage are
equated to affines of one's mother's or spouse's patrilineage, and mem-
bers of one's mother's or spouse's patrilineage are equated to affines
of one's own lineage. This might suggest that there is or was a norm
of sororal polygyny or a custom of brothers of one family marrying
sisters of another. Both types of marriage occasionally occur, but there
is no independent evidence that either has ever been common or pre-
ferred. Therefore, we can safely say only that there is a tendency to
identify relatives of a single generation terminologically as members
of a single household, which potentially they are. It is here that the
criterion of affinity is ignored in kinship terminology.

Collaterality, that is, the degree of relationship in the consanguineal
line, is ignored except in the distinction between father and father's
brothers and in that between mother and mother's sisters. In the
latter case the terms are derivatives of that for mother, meaning "elder
mother" and "junior mother." Otherwise, merging is the rule. Bifur-
cation, distinguishing whether the connecting relative is male or
female, is an important criterion in Pahari kinship. Whether the
person who links two relatives is a mother or a father, a brother or
a sister, a wife or a husband, is specified terminologically. Relatives
linked by sons and by daughters are not distinguished from one an-
other, however. Polarity, use of the same kinship term reciprocally by
two relatives, is found only in affinal relationships between people of
the same generation where the affinal link is in the first descending
generation from the relatives using the reciprocal term. Relative age,
whether the person referred to is elder or younger, is always specified
among consanguineal relatives of the same generation. In address,
relative age of cousins and siblings is sometimes ignored. Among
affinal relatives, it is the relative ages of the connecting relatives that
are crucial. For example, whether a wife's brother's wife will be
referred to by the term which means elder sister or that for younger
sister depends upon whether that woman's husband is an elder or

KIN GROUPS AND KINSHIP 193

younger brother of ego's wife. Generation always takes precedence
over age in determining seniority. An uncle who is younger than his
nephew should still be accorded the respect due his generation by
the nephew. Speaker's sex is not a factor in Pahari kinship, nor is
decedence (whether the connecting relative between two relatives is
living or dead).

Pahari kinship can now be analyzed according to Murdock's
typologies of kinship terminology, based primarily upon cousin terms
(Murdock, 1949, pp. 223 ff.).

In Sirkanda siblings and all cousins, cross and parallel, related
through mother and father, are called by terms which vary only by
sex and relative age. The terms are: elder brother or male cousin,
dlda; younger brother or male cousin, bhula; elder sister or female
cousin, didl; younger sister or female cousin, bhuli. The same pattern
appears in kin terms recorded for the Western Pahari dialect (Karve,
1953, pp. 98 ff.). This is the standard "Hawaiian" type of cousin
terminology in Murdock's classification. Since in Sirkanda this termi-
nology is associated with exogamous patrilineal kin groups, the system
conforms by definition to Murdock's "Guinea" type of social organ-
ization, and since these groups are patrilocal, it belongs in the "normal
Guinea" subtype. Murdock (1949, pp. 225 f.) includes ten criteria in
his discussion of types of social organization. These will be briefly
presented for Sirkanda: (1) descent: patrilineal; 12 (2) cousin terms:
Hawaiian type; (3) residence: patrilocal; (4) clans-demes: groups
closely resembling patri-clans are analytically discernible, and demes
are absent; (5) other kin groups: bilateral kindreds present; (6)
exogamy: patrilineal exogamy and bilateral extension of incest taboos
(all second cousins ineligible for marriage, though deviations occur);
(7) marriage: polygyny permitted, incidence around 15 per cent;
sororal polygyny permitted, incidence around 20 per cent of polyg-
ynous marriages; (8) family: patrilocal extended family; (9) aunt
terms: bifurcate collateral, that is, separate terms for mother, mother's
sister, and father's sister (the mother's sister's terms are, however,
derivatives of the mother term, which indicates a tendency toward
bifurcate merging, and Karve [1953, pp. 98, 100] reports terms in-
dicating bifurcate collateral terminology for both mother-aunts and
daughter-nieces in the Western Pahari language); (10) niece terms:
bifurcate merging, that is, one term for daughter and brother's
daughter, another for sister's daughter. All these data were obtained
from men, but so far as could be determined, the same terminology
is used regardless of whether a man or woman is speaking. This con-
forms to usage elsewhere in North India (cf. Karve, 1953, pp. 99 f., 104).

ig4 KIN GROUPS AND KINSHIP

In a patrilocal, patrilineal society such as this, patrilineal relation-
ship and male relatives would be expected to be more prominent than
matrilineal relationship and female relatives. This is true in that
the kindred extends farther on the paternal side than on the maternal
one. But in accounting for greater refinement of terminological dis-
tinctions, residence seems to be a more important factor. Those rela-
tives among whom there is or might be repeated interaction are mem-
bers of the residential group — the patrilocal extended family. They
are the ones among whom the most refined terminological distinctions
are made, especially those having to do with relative age. Those who
are not in the patrilocal group (mother's brothers) and those who leave
it (father's sisters) are in less frequent interaction and are not dis-
tinguished as to relative age. Father's brothers are in this group, and
mother's sisters potentially could be (in line with the Pahari kinship
terminological pattern, they are called by terms indicating that they
are). Both of these categories are distinguished by age relative to the
connecting relative. Seniority is an important matter to be kept in
mind among those who interact frequently, for it has significant be-
havioral concomitants.

Affinal relatives outside one's own generation are classified under
fewer terms than are consanguineal relatives, suggesting the lesser
emphasis placed on affinal relatives and the greater social (and in the
case of a man, physical) distance between a person and his affinal
relatives.

Something has already been said of the behavior of various relatives
toward one another. Husbands and wives never address one another
by name nor by the term of reference. Rather, teknonymy is generally
resorted to ("father of X," "mother of X"). Often a wife is addressed
in the household as "girl" (chori). In reference by the spouse, gharwall
(one of the house, housewife), or "woman," is used most often for a
wife, and malik (owner), or "man," for a husband. Similarly the terms
for mother-in-law and father-in-law are never used in address. They
are avoided by teknonymy or other circumlocutions and indirections
because they are relationships of respect. All younger relatives or those
who wish to show respect add the honorific jl to the relationship
terms, especially in address. Minor phonetic changes often accompany
this addition. In the table in Appendix IV this information has been
omitted except where it always occurs in both reference and address.

There is a tendency in Sirkanda to adopt some kinship terms com-
mon among Hindi speakers of the adjacent plains and distinct from
Pahari terms. Most often these appear as variants rather than as sub-
stitute terms. Those Hindi terms most frequently used simplify or

KIN GROUPS AND KINSHIP 195

make more symmetrical the terminological structure. Thus the Hindi
term mausa, husband of a senior woman (mother's sister's husband,
wife's elder sister's husband, husband's elder sister's husband) tends
to be used in place of the more complex set of Pahari terms for these
rather distant and seldom-seen relatives. In three cases distinct Pahari
terms for affines are replaced with Hindi terms derived from the term
for the consanguineal relative connecting ego and the affine, making
a pair of terms for a man and wife that are identical except in ending.
For example, Hindi pupha replaces Pahari mama as husband of puphu.
A trend toward consistency seems to explain the changes shown in
table 3.

TABLE 3
Some Variations in Kinship Terminology a

dldi esh: bahena

bhull ysh: said

dldi web: bahena

bhull wyb: said

Traditional local terms

Newer, H

es: dldi esh: jijd
ys: bhull ysh: jawai

Wife of web: didi web: jeThu
Wife of wyb: bhull wyb: syalu

es:

ys:

wife of web:
wife of wyb:

See Appendix IV for meanings of the abbreviations.

Use of Kin Terms for Non-Kin

In a number of situations kinship terminology is applied outside
the kin group.

Most common in this respect is the general tendency to address
anyone of roughly one's own age in friendly and respectful greeting
by the term "brother" or, in the case of a woman, "sister." This
usage occurs often in addressing strangers and is common among
Hindi speakers throughout Northern India. "Puphu" (father's sister)
may be used respectfully to address any older woman. There is no
general practice of referring to fellow villagers (as distinct from mem-
bers of other villages) by kin terms as is done in the plains. This may
reflect the fact that Sirkanda is not an exogamous village, and so all
villagers are not thought of as relatives.

In Sirkanda close friendship between two men sometimes results in
their consistently addressing and referring to one another's wives as
"sister," and treating them as sisters in social relationships. It is then
said that the offspring of the two couples would not be potential mates
even if sib affiliations and other factors were favorable.

If two women who are close friends have infant sons of approxi-
mately the same age they may suckle one another's child. In such cases

196 KIN GROUPS AND KINSHIP

the children are treated as brothers and will later so address one
another. They are referred to as "milk brothers" and are expected
to be especially close and mutually attached, much as would real
brothers. Probably the sister of each would not be a potential mate
for the other, though in the cases recorded they were ineligible on
other grounds.

Across caste lines relationship terms are also used in address. A
blacksmith in Sirkanda is consistently called "uncle" (mother's brother,
mama) by high-caste people younger than himself, and his wife is
called "aunt" (father's sister, puphu). To address a stranger as mother's
brother would be an insult implying that one's father has had rela-
tions with the stranger's sister. It has no such connotation in this case,
however. The blacksmith addresses younger high-caste women as
"niece" (sister's daughter, bahanji). He addresses high-caste women of
his own age as "sister," and they sometimes address him as "brother,"
whereas he addresses older women as "father's sister." All these terms
used to address high-caste women are terms of respect and honor preclud-
ing sexual interest, an important fact for low-caste men's relationship
with high-caste women. High-caste men are usually addressed by low-
caste people in honorific terms, usually by their caste title (thakurjl,
landlord; panditjl, priest).

Terms of reference across caste lines are never kinship terms. Names,
caste names, or occupational terms are used. An elderly person, regard-
less of caste or relationship, is likely to be addressed and referred to
simply as "old man" or "old woman." Descriptive terms are frequently
used in reference, as "black man," "white man," "fat man," "tall one,"
"crippled one."

CASTE

The caste system has probably attracted more attention from soci-
ologists and anthropologists than has any other feature of Indian
society. The emphasis is not unwarranted, for it is one of the dominant
social and cultural facts of life in India. In this chapter caste will be
described as it functions in Sirkanda. Some of the differences and
similarities of the Sirkanda system as compared to other examples
that have been reported in India and elsewhere will be pointed out.

Caste has been defined variously or not at all by the many writers
who have discussed it.1 Many have been concerned exclusively with
India and have therefore dealt with caste in terms so specific as to
apply only to India, the area of their interest or experience (for ex-
ample, Gilbert, 1948, pp. 31 f., and Leach, i960). Such narrow defini-
tions are useful because caste in India is unique and has distinctive
regional variations. The same is true of religion, family organization,
economy, and most other phases of life. To understand these things
thoroughly it is well to make explicit their unique characteristics.
Srinivas emphasizes a prominent and unique aspect of caste in India
when he says, "The concept of pollution governs relations between
different castes. This concept is absolutely fundamental to the caste
system and along with the concepts of karma and dharma it contributes
to make caste the unique institution it is" (Srinivas, 1952, p. 28).

However, to define caste in terms of its uniquely Indian attributes
eliminates or at least diminishes its use as a cross-culturally com-

1 98 CASTE

parable phenomenon (Bailey, 1959; Berreman, 1960a). This in itself
is not a criticism. Cultures do have unique traits. I prefer, however,
to define caste more broadly in order to include, for purposes of com-
parison, similar systems of social stratification which occur in other
cultures. More is lost, from the point of view of social science (though
not, perhaps, from that of Indology) by emphasizing its unique aspects
at the expense of comparability than is gained by the added precision.
The definition of castes which will be followed here seems to be
valid both in India and elsewhere: Castes are ranked endogamous divi-
sions of society in which membership is hereditary and permanent.2
Implicit in this minimal definition are additional criteria: castes are
recognized as groups (i.e., they are usually named); they are in some
ways interdependent; barriers to free social intercourse exist between
castes; there are cultural differences between castes; there are differ-
ential degrees of power and privilege between castes. Associated with
caste in many and perhaps all instances is a degree of occupational
specialization. While all the members of a caste are not often com-
mitted to one line of work, there is a particular occupation or range
of occupations which is considered to be appropriate to each caste.
Passin (1955, p. 41) notes that among the "untouchable" groups of
India, Japan, Korea, and Tibet, ". . . What is particularly striking
is the detailed similarity of their occupations. In principle, they were
restricted to the despised and menial functions of the community
which were, however, essential. Someone had to do them." Occu-
pational specialization may be evidenced in the occupations a caste
cannot or does not practice as well as in those it does practice. In
India caste groups are specifically characterized by (1) a common
traditional occupation and / or a claim to common origin, and (2)
a ritual status which must be maintained and which can be defiled
by specified types of behavior and contacts with other groups.

The concepts of ritual status and pollution, as mentioned above in
the quotation from Srinivas, are important in understanding the func-
tioning of caste in India. Stevenson specifies that ". . . it is from
ritual rather than secular status and from group rather than personal
status, that the caste system derives it unique consistency and viability."
"The ritual status relationship between individuals and groups and
between groups of different categories [castes] rest wholly upon be-
havior-patterns linked with mystical beliefs in general, and mainly
upon behavior linked with a particular corpus of beliefs concerning
purity and pollution, . . . the Hindu Pollution Concept" (Stevenson,
1954, p. 46).

Indian caste distinctions are explained in terms of elaborate reli-

CASTE 1 99

gious, mythological, and historical rationalizations. It is sometimes
asserted that as a result of such rationalizations the caste system is
made acceptable to all those who are included within it. This may
be true, but the implication that therefore all who are included within
the system are content with their assigned position in the hierarchy
either as individuals or as groups is contradicted by the present re-
search and by the many examples of caste-group mobility and indi-
vidual aspirations of low-caste people that have been reported (Berre-
man, 1960a).

In India castes are generally divided into five major hierarchically
ranked groupings or levels. The top four of these are the four varna
or categories described by Manu, the Hindu lawgiver, comprising the
three "twice-born" groups (Brahmins or priests, Kshatriya or rulers
and warriors, Vaishya or traders and farmers) and the Shudra artisans.
At the bottom, outside the varna system, are the untouchables. This
system is well known and is discussed at length elsewhere (cf. Basham,
1954, pp. 137 ff.; Dube, 1955, pp. 34 ff.). Also well known is the division
of these major caste levels into many endogamous regional subgroups,
called jati in much of North India, the "castes" of ordinary parlance.
The many details of ritual and other barriers restricting social relations
between castes have been explained at length by other authors (cf.
Srinivas, 1952, pp. 24 ft.; O'Malley, 1932, pp. 1-31). One of the best
concise discussions of these matters is that of Stevenson (1954). Recently
attention has been directed toward objective and precise determina-
tion of caste functions, characteristics, and especially relative status
or ranking of castes (cf. Gough, 1959; P. M. Mahar, 1959; and Marriott,
1959, i960). It is unnecessary to go further into these matters here.

Caste Functions

Caste (jat or jati) is the ultimate extension of the kin group. Its
members are considered to be remotely related to one another, and
it is the furthest extent of potential or actual affinal kin ties. In fact,
the geographical and social limits of a caste may be defined most
satisfactorily in terms of the marriage network. As a result, caste
functions are in many respects similar to family, sib, clan, and kindred
functions, although caste is a less intimate group. It is a unit of social
control in that its members or their representatives determine col-
lectively what the membership may and may not do, and what shall
be done to enforce these rules. It is the unit of ritual and social
equality. It is the endogamous unit, the commensal unit, the unit
within which there are no ritual prohibitions on contact and outside

200 CASTE

of which such prohibitions must, theoretically, be observed. It is an
economic unit of importance in that it tends to be the largest com-
petitive unit — the largest unit whose members engage in the same
traditional occupation. Caste is an important reference group for its
members. Matters of importance to individuals, families, sibs, clans,
and kindreds are of some concern to the caste as a whole and are likely
to engage the attention of caste-fellows. Ritual and ceremonial occa-
sions reveal this. Cliques and factions show the influence of caste
boundaries. People interact socially and ritually with their caste-fel-
lows more often than with members of other castes.

As a result, caste members share a culture which differs slightly
from that of other castes. Although caste cultures vary less in Sirkanda
than on the plains, differences do become apparent in matters pertain-
ing to traditional occupation, to social organization, to caste prefer-
ences in worship, and to interaction with other castes.

Caste in the Sub-Himalayan Region

There has been no systematic study of Pahari caste organization.
There are, however, enough references to the population and customs
of the Pahari areas to give a general picture of caste structure there
and to verify much of what came to light in Sirkanda. The Pahari
caste system is similar to that in other parts of India and is well within
the range of regional variations.

All over the cis-Himalayan region, the Simla states, the Doon valley, Kulu
and Kangra valleys there exists a hierarchy of social status, though the rigidity
of the caste system as in the plains does not exist. The upper class consists
of Brahmins and Rajputs (Kshatriyas) . . . the lower strata is composed of
innumerable social groups who form the artisan elements in the population
of these parts. . . . These suffer from a number of disabilities and are treated
as serfs or dependents and thus provide a dual organization of economic
classes in these hills. (Majumdar, 1944, pp. 137 f.)

It has been noted before that the component castes of the "upper
class," the Khasiyas, are not separated by great social or ritual distance
in some Pahari areas and that, in fact, intermarriage between them is
not rare. The "lower strata" are called Doms, a term which applies to
all the artisan castes of the hills. "The Doms . . . are as far as can be
asserted the aborigines of the country. They are found wherever the
Khasiyas are found, living with them in a state even now not far re-
moved from serfdom" (Walton, 1910, p. 62). Saksena reports that
while the Brahmins and Rajputs of Jaunsar-Bawar form a single en-
dogamous group, "each group among the Doms constitutes a single

CASTE 20 1

endogamous group, arranged in hierarchical order" (Saksena, 1955,
p. 28).

There can be no doubt that we are dealing with a caste society.
It is not as rigid, perhaps, as that found on the plains, but the bases
for status evaluation of castes, the barriers limiting contacts between
them, and the nature of interaction among them are in general similar
to those characteristic of the Hindu caste system throughout India. It
cannot be accurately described as a class system, as Majumdar wishes
to do, because legitimate individual mobility within the system is
impossible.

Most writers attest to the physical and cultural distinctiveness of
the Doms as contrasted to the high castes or Khasas. The difference is
traced to their alleged separate origins as early aborigines and Indo-
Aryan invaders, respectively. These views have been discussed at some
length in chapter 1. Whatever may have been the earlier situation,
it is evident that these distinctions now exist primarily in the minds
of their advocates and to a minor extent if at all in the physical and
cultural make-up of the Paharis. Today Khasas and Doms are phys-
ically virtually indistinguishable and culturally similar.

There is a homogeneous social code to which both the higher and lower
groups subscribe. The disabilities that obtain in these parts among the lower
castes in the matter of dress, food and drink, in the restrictions to marriage
and inter-dining, are mostly superficial, and they have not affected the social
relationships to any appreciable extent. (Majumdar, 1944, p. 139)

Thus Pahari caste stucture is characterized by a twofold division
into high-caste groups (Brahmin and Rajput or Khasiya) and low-caste
artisan groups (Dom). The latter are accorded the status of untouch-
ables and include most of the occupational groups found among both
the clean Shudras and the untouchables of the plains. Within each
of these classifications there are status distinctions, but not of the same
order as that of the main division. Even intermarriage is tolerated
between castes of the same general social and economic level. As will
be made clear below, the high-low or pure-polluted caste dichotomy
is a striking feature of caste structure in Sirkanda as distinguished from
the multiple divisions of the plains. On the other hand, this is not
a qualitative difference — plains people too distinguish "clean" castes
(the four varnas) from untouchables, and "twice-born" castes (the top
three varnas) from those which are not twice-born (the Shudras and
untouchables). Cohn (1954, p. 116) notes that, in the plains village he
studied, "it would appear that, although the caste system is a graduated
hierarchical one, the difference between the Thakurs and all the others

202 CASTE

is greater than any differences among the low-caste people." Bailey
(1957, p. 8) emphasizes the importance of the "line of pollution"
between untouchables and others in a South Indian village. Hocart
(1950, p. 4) comments on the distinction in Ceylon between the three
leading castes or "good people" as ". . . opposed to the 'low-castes,'
who comprise fishermen, smiths, washermen, . . . tailors, potters,
weavers. . . ." Marriott (i960, pp. 43 ff.) cites comparable systems of
caste status ranking in the middle Indus and Bengal delta areas. In
much of South India Brahmins, Shudras, and untouchables are the
major groupings.

In plains villages the most important division seems to be at least
threefold (twice-born, Shudra, untouchable) if not fivefold (Brahmin,
Kshatriya, Vaishya, Shudra, untouchable) rather than twofold as in
Sirkanda. Also, in the plains there is evidently greater social distance
between groups which would in Sirkanda fall within the high- or low-
caste divisions. Another difference is that in the hill area there are
few ranked subcaste divisions such as those characteristic of plains
castes. Among the indigenous hill castes in and around Bhatbair, the
subdivisions (sib, gotra) are of equal rank and purity. The only excep-
tion is in the superior ritual purity of ceremonial cooks over other
Brahmins.

The range of castes in the hills, as in any other particular locality
in India, is but a small segment of the total of endogamous castes
and subcastes found in India. Ghurye (1952, p. 23) estimates that at
the beginning of the nineteenth century there were 500 to 2,000 sub-
castes in each linguistic area of India. In comparison to the adjacent
plains, the absolute number of castes in the Pahari area and in each
Pahari village is small. For Garhwal, Raturi (1928, pp. 196 ff.) lists
Brahmins, Rajputs, and 30 untouchable castes. Bahadur (1916, pp.
101 f.) lists 15 "professional castes" belonging to the Dom group. It is
probable that both these listings of artisans include occupational
groups which cannot accurately be described as castes in that they are
not endogamous, but the pattern of two high castes and a variety of
low, ritually impure artisan castes of roughly comparable status is
verified. Conspicuous by their absence in the Pahari region are in-
digenous Vaishya (members of the merchant caste) and Shudras (clean,
but not twice-born, castes, primarily artisans).

Caste in Sirkanda

There are five castes represented in Sirkanda: Brahmins and
Rajputs in the twice-born, Khasa category, and blacksmiths, Bajgis

CASTE 203

(drummers), and barbers in the untouchable, artisan Dom category.
Although the barber caste is not an indigenous Pahari caste, the
Sirkanda barbers were brought into the village by its Brahmin
founders and have therefore been an integral part of Sirkanda since
it has existed as a village. Blacksmiths are the most recent immigrant
group, having come about 90 years ago. One Bajgi man settled in
Sirkanda only 25 years ago, but the other Bajgis are long-time resi-
dents. At one time or another, in the past, shoemakers and weavers
have lived temporarily in or near the village, but, like the Vaishya
merchant who operates one of the village shops, they were never
identified by themselves or others as villagers. Within a half-day's
walk of Sirkanda are a few other service groups who are occasionally
utilized in the village, notably shoemakers, weavers, and carpenters.
Traditional caste occupations, occupational deviation, and the role of
caste in the local economy have been discussed in chapter 2.

Caste organization in this area is extremely loose and informal.
There are no organized caste governments or councils. When an in-
fraction of caste rules occurs in a high caste, a group of male high-
caste members may be called together as a council to take action. If
so, they discuss the case, listen to any evidence or pleas that are to be
given, and come to a decision. They may fine or otherwise punish the
offender or, in extreme cases, ostracize him from the caste, that is,
outcaste him by refusing to let him take further part in caste functions
and refusing to interact socially with him. The only remembered occa-
sions upon which such councils have been called have been for con-
sideration of breaches of rules of caste endogamy. Theoretically they
might be called for other cases of disobedience of rules of ritual purity.
Often no meeting is called and the offender is merely ostracized by
informal agreement.

Generally councils are made up of Rajputs and Brahmins. It is
thought proper by both Rajputs and Brahmins to invite representa-
tives of both castes to any meeting that may be held by either of them
because they are interdependent as priests and protectors and there-
fore have common interests.

Members of low castes usually try to settle their problems informally
among the individuals or kin groups involved. This course is most
frequently taken as numbers are too small to make a caste council
feasible, and the members wish to attract as little attention and inter-
ference as possible. If agreement cannot be reached on that level, high-
caste people will intervene whether requested to do so or not. As the
masters and possessors of superior judgment, and as people with a stake
in smooth social relations and docility among their subordinates, high

204 CASTE

castes do not hesitate to assume their traditional authoritarian,
paternalistic role even in the internal affairs of low castes. Often a
meeting to decide a dispute or punish a misdemeanor involving only
low-caste people will be made up entirely of Rajputs and Brahmins.
It may even be called by the high castes when the caste involved has
no intention of acting in the matter. Individual low-caste people often
turn in times of trouble to high-caste patrons — usually people to whom
they are committed by employment — who are expected to intercede on
their behalf. The dominance of high castes (in this area, Rajputs) is
readily apparent in such matters, as they can control the behavior not
only of their own caste members, but to a significant extent that of
other castes as well.

The real source of day-to-day control within the caste is social pres-
sure applied informally, but often relentlessly, by the group.

Caste Identification

Caste identification or loyalty is prominent despite the informality
of caste organization. It is to be expected that in a caste society in
an isolated area caste identification will be strong. From childhood
the individual learns that only among caste-fellows is he among equals.
He hears the stereotypes about other castes and sees the conventional
behavior between castes. He is taught that he cannot deviate from
caste rules. If he is of high caste he gradually learns the advantages
which accrue to him as a result of his position. If he is of low caste he
learns of his dependent status and the sanctions which can be applied
to keep him in his place. He learns that to leave the village is to leave
all that is familiar, all that provides security. Most people therefore
come to accept their position, though not necessarily to like it. Inter-
caste conflicts are frequent enough, and rivalries are persistent enough,
that other castes are seen as rival groups in many contexts. Sympathy
for caste-fellows generally overrides local loyalties. In the interests
of village harmony or personal survival, caste loyalties may be sup-
pressed but they do not disappear. Thus, initial inquiry among low-
caste people invariably brought the response that in a dispute they
would side with their high-caste village-mates against aliens regardless
of caste. "Our village must stand together." Later, as they came to trust
the inquirer they stated emphatically that their sympathies invariably
lay with their caste-fellows regardless of village affiliation. A blacksmith
informant said, "If there were a dispute between a blacksmith of
some other village and a Sirkanda Rajput my whole sympathy would
be with the blacksmith, but I would speak out for the Rajput, if I
had to speak, for fear of the punishment of Rajputs. This is always

CASTE 205

the case. I think it is true of all castes." A Bajgi commented, "If there
is an intercaste dispute involving a Bajgi we are on his side whether
he is from a neighboring village or from Bombay. If they knew of it,
every Bajgi in India would be on his side."

That such attitudes do not seriously affect village stability is largely
attributable to the concentration of numbers and power in the high
castes. Organized opposition to them is hopeless, and overriding caste
loyalties among low castes are restricted largely to private attitudes.

Caste loyalties appear to be somewhat less strong in Rajput-Brahmin
relations, where village unity or in-caste factions often override inter-
caste rivalry. In a real power struggle, however, a high-caste man can
usually count on the support of his caste-fellows, or demand it.

High castes frequently intervene or support one side or the other
in inter-low-caste disputes. Low castes try to avoid involvement in
inter-high-caste disputes within the village and enter into such dis-
putes across village lines only at the urging of high-caste villagers.

The status identification of low-caste people is recognized by all
castes. A radio program devoted to "Harijan uplift" (untouchable
uplift) heard on the anthropologist's radio, attracted the immediate
attention of low-caste people and was the subject of pointed remarks
and jokes by high-caste villagers. "Well, you will soon be a big man,
Ram Lai. The government will teach you to read and make you a
cabinet minister." Low-caste people are expected to be interested in
talk of their position and government efforts to improve it, but not
to take it seriously. They are watched closely by high-caste people, and
a wrong response or attitude can bring abuse. I described to a group of
Rajputs the visit to India of United States Congressman Dr. Singh
Saund, and mentioned that he was of Punjabi Indian origin and,
according to the papers, had been a blacksmith. My listeners shouted
at once for the blacksmith, who was working in his shop. The story
was repeated for his benefit, and all expectantly awaited his reaction.
Having been put on the spot, he mulled the idea over briefly and then
said evenly, "I want to go to America," and returned to his shop.

Dominant Caste

A study of the locally dominant caste and the kind of dominance it enjoys
is essential to the understanding of rural society in India. Numerical strength,
economic and political power, ritual status and Western education and oc-
cupations, are the most important elements of dominance. Usually the differ-
ent elements of dominance are distributed among different castes in a village.
When a caste enjoys all or most of the elements of dominance, it may be said
to have decisive dominance. (Srinivas, 1959, p. 15)

206 CASTE

Of 384 current residents of Sirkanda, 90 per cent are in the high
castes (87 per cent Rajputs, 3 per cent Brahmins) and 10 per cent
are in the low castes (5 per cent Bajgi, 3 per cent blacksmiths, 2 per
cent barbers). Of 45 households or joint families, 37 are Rajputs, one
Brahmin, two blacksmith, one barber, and four Bajgi. Therefore
Rajputs are numerically the dominant caste in this village.

Rajputs own 94 per cent of the arable land surrounding Sirkanda,
while the Brahmin household owns another 4 per cent, leaving but 2
per cent for the low castes. Similarly the high castes own 97 per cent
of the cows, buffaloes, and oxen, all the horses and mules, and 92 per
cent of the sheep and goats of Sirkanda. Of the livestock, the Brahmin
family owns its proportional share as a larger-than-average high-caste
joint family — about 4 per cent. A Rajput family has the only locally
owned and operated shop, and another Rajput family owns the build-
ing in which the outside merchant keeps his shop. Low-caste income
is derived primarily from high-caste villagers in return for services
as artisans. It is therefore apparent that Rajputs are economically
the dominant caste in Sirkanda.

Formal village government and political affiliation count for little
in Sirkanda. However, of the eleven Sirkanda members of the village
council, 10 are Rajputs and one a blacksmith (by law one seat is
reserved for low castes). The president and vice president of the council,
as well as their opponents in the last election and their predecessors,
all are Rajputs. The five Sirkanda members of more inclusive govern-
ing bodies (judicial councils) have all been Rajputs, including the
president of one of these councils. The two officers of the Sirkanda
credit cooperative, although they have never had occasion to function,
are Rajputs. In the past various official and semiofficial offices were
given to appointed villagers by the British for administrative pur-
poses. One of these, village accountant, was held by a Rajput; another,
village tax collector (Lambarddr), was held successively by two Rajputs
of Sirkanda and before that by a Brahmin of a neighboring village. The
hereditary title of Mukhia (a local administrative officer supposed to
be a keeper of the peace) was held in a Rajput family, and a son-in-law
of that family retains the title today.

No official position within the village, except the recently designated
low caste seat on the village council, has been held by other than a
Rajput. A frequent criticism which low-caste people voice against
the village council is that it is dominated by Rajputs and therefore
dispenses "Rajput justice." A blacksmith commented:

The president of the village council is a Rajput. He was put into office
by Rajputs. Even if he wanted to be fair to us and help us get some land

CASTE 207

to till he could not. There would be too much pressure on him from other
Rajputs. So everything he does is for the benefit of Rajputs.

Another low-caste man said:

They run the village to their own advantage. They never think of us. They
fine us if we bring a dispute before them and do not fine Rajputs in the
same circumstances. Then they use our money for their enjoyment. They
prey on our vulnerability.

In the more important realm of traditional leadership and power
in the village, Rajputs play the dominant role. When meetings or
councils are called to decide disputes or infractions of rules involving
villagers, those called are high-caste people, so that Rajputs are always
represented and dominate even though their caste-fellows may not
be involved in the issue at hand. They combine the traditional roles
of landowners, rulers, and protectors of the village, and so they are
its leaders.

It is therefore evident that Rajputs are politically the dominant
caste in Sirkanda.

Nontraditional education is as yet not a powerful influence in
Sirkanda. The village school, established in 1950, is open to, and
attended by, all castes. However, of the twelve villagers who went
outside Sirkanda for schooling before establishment of the local school
or for higher education than that offered in the village, all were
Rajputs.

Of Sirkanda villagers only three, all Rajputs, have been in military
service. Rajputs have no monopoly on casual outside contacts or
temporary sojourns in urban or non-Pahari areas. Members of all castes
have had such experience, and low castes, in particular, travel —
primarily to get materials for their crafts. However, such travel seems
to have relatively little lasting effect. The two villagers who have
moved out to take apparently permanent employment in urban areas
are both Rajputs. On the whole, then, Rajputs have had greater
access to nontraditional education and experience than have other
castes in Sirkanda.

Ritually, Rajputs are second to Brahmins in status, but the differ-
ence is relatively unimportant in this area. As will become evident, the
difference in ritual status is not as great between these two groups
as it is in the plains; intermarriage is tolerated (although it is dis-
approved and occurs infrequently), and a high degree of social in-
timacy is practiced. Moreover, these two castes are far above the other
Pahari castes in ritual status.

Thus Rajputs enjoy nearly all the elements of dominance in Sirkanda

208 CASTE

and are therefore the possessors of "decisive dominance" as defined by
Srinivas. In Sirkanda, as throughout the hills, Rajputs and Brahmins
are what Mayer (1958) terms "allied castes" in that they act together
on certain occasions, notably in relation to the low castes, to present
a united front. There are disputes and jealousies between the two
high castes, but they are rarely divided in their relations with the
low castes, to whom they appear as a single dominant group.

In some villages of the area Brahmins are approximately equal in
numbers to Rajputs, and in a few they outnumber Rajputs. I would
estimate the Brahmin population of Bhatbair and vicinity at one-
fourth to one-third that of Rajputs. According to the census of 1872,
there were 81,000 Brahmins and 152,000 Rajputs in Garhwal (Atkin-
son, 1886, p. 277; see also table 4 below).

TABLE 4

Population by Caste

Sirkanda

Caste

Sibs

Lineages

Households

Individuals

Number Percentage

. , J Brahmin
lg 1 Rajput

1
11

11 3 ]
335 87 i

f Blacksmith
dw J Barber
I Bajgi

Totals

1
1

2
9

45"

12 3 I

5 2
21 5 J
384 100

90 %

10%

26 Villages Which Each Supplied at Least 1% of Sirkanda Mates*

High-caste individuals 4332 (93%)

Low-caste individuals 339 ( 7%)

Total 4671

India b

Harijans 51,000,000 (13%)

Others 333,000,000 (87%)

Population of Garhwal, 1872 c

Brahmins 81,000 (28%)

Rajputs 152,000 (53%)

Doms 52,000 (18%)

a Census of India, 1951.

b Planning Commission (1958), p. 380.

c Atkinson (1886), p. 277.

Relative size of population is decisive in determining which of
the two high castes will exercise practical dominance in any given

CASTE 209

village, because they differ little in the other criteria of dominance.
The importance of numbers is as readily apparent to villagers as it is
to the outside observer. Even among low castes a frequent rational-
ization for the power of high-caste villagers is in terms of their superior
numbers. A blacksmith described the discrimination practiced by
Rajputs in the village against his own caste, and then commented,
"The reason Rajputs can do those things is that they outnumber us
so much. If blacksmiths were in the majority, the Rajputs would be
untouchables." He also recognized the importance of land to power:

Because of our skill at blacksmithing, our ancestors spurned land — they
wanted to make their living by their craft. What they did not realize was
that people with land are people with power. So now we blacksmiths are
suffering the consequences. We are just like servants now. We have to do
what the landowners tell us to do. Since we do not grow our own food, we
must come begging to them for payment.

Another low-caste man explained:

Caste is a matter of wealth and numbers. A wealthy untouchable can have
the District Magistrate and others to his house. A poor untouchable cannot
even draw water from the public well. In this village there are many Rajputs,
so they can tell the low castes where to sit. If the numbers were reversed they
couldn't do this. As an example, there are few shoemakers and sweepers
[the lowest castes] in Dehra Dun and they are poor. They are not allowed
to touch anyone of higher caste. On the other hand, before partition there
were many Muslims in Dehra Dun and many of them were well-to-do. The
Muslims were fed and entertained in the homes of high-caste Hindus and
vice versa. A Muslim kills the sacred cow and eats its flesh, while a shoemaker
merely removes the hide of a dead cow and with it makes shoes for people
— a necessary service. But Muslims were numerous and wealthy, so in spite
of their defiling practices they were treated well. Shoemakers are few and
poor, so they must suffer as untouchables. Such is the nature of caste.

Although Rajputs are the dominant caste they are dependent upon
other castes in and around Sirkanda for economic and religious serv-
ices, as discussed in previous chapters. Before we proceed to an examina-
tion of intercaste relationships in these and other contexts, it will be
well to consider the important matter of relative ranking of castes.

Caste Ranking in Sirkanda

Determination of the relative rank of castes is not as easy as it
might at first appear. To simply ask a person may or may not produce
a coherent account of rank order. To ask several people is most likely
to produce several rank orders.

2 1 0 CASTE

Individuals and even groups may be given one rank order in a particular
context and a different ranking in another context. . . . What is quite con-
stant is a set of criteria for ranking; what varies is the interpretation given
in a specific instance to a particular combination of characteristics. (Mandel-
baum, 1955, p. 241)

Recently a good deal of attention has been directed toward this
problem. Among the most interesting studies has been Pauline Mahar's
use of a multiple scaling technique to get at caste ranking from the
point of view of ritual purity and pollution by investigating the "norms
relevant to the interactions symbolizing inequality of ritual status in
dyadic inter-caste relations" (P. M. Mahar, 1959, p. 128). Through
this technique she was able to show quite precisely the relative ranks
of 18 of 22 castes in Kalapur and to specify areas of disagreement in
caste ranking.

In Sirkanda the situation was less complex than that with which
Mahar and most other researchers have worked in Indian villages, in
that a relatively small number of castes were to be dealt with. The
pattern of intercaste relationships was, however, essentially the same
as that found by other researchers. In view of the limited number of
informants and castes, I was able to question extensively on this
subject and to derive a fairly consistent body of data. In analyzing
these data I compared a range of statements of what should be done
and what is done with the observations and reports of actual inter-
action among members of the various castes. As a result, I obtained a
quite complete picture of caste as it functions in the village and as
it is seen from the various caste levels.

In the research an attempt was made to pay equal attention to
"ritual" and "secular" criteria of caste ranking. It soon became evident
that "secular status" is significant to caste ranking primarily as it is
reflected in "ritual status," but not in and of itself. The two cannot be
separated. Traditional occupation, which is an important factor in
caste status and might be assumed to be a secular consideration, is in
fact not secular. Inherent in it is attribution of ritual status. That is,
occupation is accorded ritual significance. A shoemaker or blacksmith
is, by the nature of his traditional occupation, ritually impure. If a
group gives up or adopts a ritually impure occupation, its ritual status
will usually vary accordingly. Thus it was reported that some Bajgis
further in Garhwal had, in the absence of available shoemakers and
at the insistence of high castes, adopted the occupation of skinning and
disposing of dead animals. They had thereupon become untouchable
to other Bajgis. On the other hand, a group of Bedas near Bhatbair
were trying to raise their status by giving up their traditional occupa-

1. Sirkanda village and neighboring hills. View eastward from the school
(see map 3).

2. Settlement area of Sirkanda. The Brahmins' house in the foreground
is being rebuilt by carpenter-masons while two members of the family and
two elderly Rajputs watch.

3. The respected head of a
prosperous Rajput household.
He wears gold earrings to
honor a household deity.

4. Wife and mother in a Rajput household. The gold
nose ring indicates marriage; the heavy silver neck-
lace was given at her engagement. The other jewelry
is simply for "fashion."

5. Three young Rajput girls
and a little boy. The eldest
girl is approaching marriage-
able age.

6. These boys, like men, carry loads on their shoulders
or backs rather than on their heads as women do. The
boy on the left is a Bajgi; the other two are Rajputs.
They show physical differences which fit the stereo-
types of their castes.

7. Women such as these Rajputs spend considerable time every day
carrying water more than a quarter of a mile from the village spring to
their homes.

8. Residents of two good-sized villages farm this terraced hillside. One of the
villages is discernible in the lower foreground; the other is near the crest of
the hillside on the right.

9. A Rajput harrows his plot of millet on a large terrace early in the rainy
season.

10. Children and bullocks here perform die tedious task of threshing wheat.

11. Bringing fodder from the forest is a year-round job for women
and children. This Rajput boy is bringing dry fodder in the dry
season.

12. Women of all castes devote much of the day to housekeeping. This Rajput
woman grinds spices as she watches the children.

13. The barber is of a caste alien to these hills, but his people have lived in
Sirkanda since its founding and have intermarried with Pahari service castes.
Here be performs his traditional duty for the village schoolteacher.

14. A blacksmith does all the iron work for his clients in return for traditional
payments in grain at each harvest. This man is repairing a grass knife.

"J' WMkMittMi

15. This Bajgi earns his liv-
ing by making baskets to or-
der. His relatives are farmers,
musicians, and tailors, and
one operates an illegal still.

8r»»

*}|

urn r h

I II I !:

•r"

*»iii.
Mi,

N II .,

-:^0

16. A gala wedding party (barat) sets out for the bride's village. In this group
the musicians and palanquin bearers are Bajgis while the groom and guests
are Rajputs. The groom will walk most of the way because the trail is rough
and the way is long, but during the triumphal exit from his village and the
entry into the bride's village he will be in the palanquin.

17. The Brahmin priest (book in hand) calls for a recess in the
night-long wedding ceremony. It is being performed in a chart
where the bride's family lives, sharing this room with three buffalo
even during the ceremony.

18. While the wedding
ceremony is in prog-
ress, some of the men
amuse themselves by
singing and dancing
out-of-doors.

19. A least is part oi
every major life-cycle
rite. Women do not
travel to ceremonies
but local women eat
together, adjacent to
their menfolk and
the male guests.

m Ml

20. The annual Pahari fair (taulft) of the area is a secular event that is greatly
enjoyed, much as are weddings and the religious festival, Diwali. Here gaily
dressed men and women take a ride on a Ferris wheel.

21. A Bajgi woman dances while possessed by a household god whose shrine
is visible in die rear wall. The pujaris who have induced die god to dance
sit on the right. The female victim of the god's wrath, whose illness has neces-
sitated the ceremony, sits behind the dancer.

22. A Bajgi diviner analyzes the
troubles of a Rajput client. In
the foreground is a typical wa-
ter pipe.

23. Taking milk products to
market to sell for cash or credit
is a regular activity of young
men such as these Rajputs. These
men's ideas of stylish dress differ
considerably. The one on the
left is a villager with little ex-
perience outside. His companion
has been to school in the valley
and has acquired a taste for
plains styles.

*^*i«£jfci~

24. Occasionally an outside merchant brings his wares to Sirkanda. Here an
itinerant Tibetan bangle seller, his son, and his employee are enjoying a lively
market for their sjoods.

25. The school is a recent and relatively successful source of governmental
influence in Sirkanda. Note that two students are spinning wool thread while
reciting their lessons.

26. An even more recent outside intrusion is All India Radio, brought to
Sirkanda for the first time via the anthropologist's radio, shown here. The
listeners on the cot are Rajputs, the others are Doms. If a village panchayat
house is built, a free radio and loudspeaker will be provided by the govern-
ment.

27. These members of the older
generation of Rajputs look back
contentedlv on their lives. Thev
have four obedient sons and good
land. They do not foresee rapid
or unsettling changes in the vil-
lage, but diev do expect voting
men to take advantage of new
opportunities.

28. This young Rajput is considered to be the out-
standing pupil in the village school. From a pros-
perous family, he expects to go to high school and
perhaps to college. As a fourth-grader, he is the first
member of his family to be educated beyond the first
three grades. He would like to get a job "in service,"
that is, a white-collar job.

CASTE 2 1 1

tion of singing and dancing. It is important to keep in mind that ritual
status is a group phenomenon. An individual cannot raise his status
by changing his occupation, and he is unlikely to lower it in this
fashion unless he is ostracized by his own group. The discussion in
chapter 2 of caste occupations, occupational variation within castes,
and caste-neutral occupations is relevant in this context. It seems likely
that occupational variation within castes is more acceptable and has
wider limits here than in the plains.

In Sirkanda the details of caste ranking van' with the status level
from which the individual views his fellow villagers. In general,
ritual and social distance (in one direction) between the caste of an
observer and two castes being observed varies inversely with the
perceived ritual and social distance between those two castes. How-
ever, pattern of ranking is remarkably consistent. In the discussion of
caste ranking below, ideal patterns will be presented first and there-
after some of the exceptions and deviations will be mentioned.

Ritually purest of all castes are the Brahmins — traditionally the
priestly caste. Regardless of whether or not individuals follow a
priestly occupation they are accorded their traditional ritual primacy.
Their touch or presence defiles no one. All castes will take water and
boiled (kachcha) food from them.3 One group of Brahmins is ritually
even purer than other Brahmins. They are the ceremonial cooks (cf.
p. 67) who accept kachcha food from no one but their own subcaste-
fellows. Brahmins in general take kachcha food from no other caste,
will let no other caste enter their cooking and eating area, and when
smoking the water pipe will share the wooden pipestem with no other
caste.

Rajputs are a close second to Brahmins in ritual status. They are
approximately as far from other castes in ritual distance as are
Brahmins but are slightly inferior to Brahmins. Brahmins will not
accept kachcha food from them, share the wooden pipestem with them,
nor let them eat in a Brahmin kitchen. When a Rajput eats in a
Brahmin house he rinses the utensils he has used and leaves them out-
side where they can be cleaned before being used again in the Brahmin
household. Rajputs and Brahmins do, however, freely share the brass
pipestem or, lacking this, the brass base of the water pipe, and ciga-
rettes. In most situations they interact as equals.

The Brahmins and Rajputs are the two "twice-born," ritually clean
castes. From their point of view all other castes are achut (literally "un-
touchable") or defiling and are referred to as Dom. Distinctions among
Doms are known but irrelevant to these high-castes. They will accept
from Doms no drink except liquor and no cooked food except parched

2 1 2 CASTE

grain, and potatoes or meat roasted directly on the coals of a fire. Doms
must not touch the water pots or other utensils of high-caste people.
If they do, the contents are discarded and the utensil, if it is brass
(as is usual), is cleaned. If it is of the alloy known as kansi, which
absorbs pollution easily, it cannot be cleaned and must be discarded.
Bodily untouchability is not observed — it is not polluting for a high-
caste person to be touched by a Dom. Cohn found the same to be true
in Senapur. "When a Thakur uses the term achut in reference to a
Camar he does not mean that he cannot touch or be touched by him,
but rather that he cannot take water or food from him and that the
Camar's touch of utensils or of cooked food is defiling" (Cohn, 1954,
p. 120).

Doms are not supposed to sit on the same string cot with high-
caste people or, in fact, on a string cot at all in the presence of high-
caste people. Doms are not allowed to enter or sit in the houses of
high castes. When visiting high-caste homes Doms are expected to sit
on the narrow porch outside the house or on the steps leading up to
the house. On the other hand, high-caste individuals can enter freely
into the homes of low-caste people and are given the seat of honor
when they do so. Doms cannot share either the pipestem or the base
of the water pipe with high-caste people. When smoking in a group
they are given only the clay pipe bowl to smoke. They are not allowed
to share cigarettes with high-caste people. They cannot worship in
the Raghunath temple or the larger temples outside the village. When
being blessed by a Brahmin they are not touched by him. Instead he
throws a small amount of vermilion substance onto a convenient
surface and a Dom applies it to the foreheads of his caste-fellows.
Sexual contacts between Dom men and high-caste women are taboo.

When a Dom and a high-caste person meet one another, the Dom
is expected to bow or nod and respectfully address the high-caste person
with the traditional Pahari greeting, "Samarii Thakur" (greetings,
landlord).4 This is an enforceable rule.

There is thus a great social and ritual barrier between Rajputs and
Brahmins, on the one hand, and Doms on the other. Strengthening
the barrier, in the minds of villagers, is the traditional occupation
associated with each caste. Brahmins are traditionally priests and Raj-
puts are traditionally administrators and warriors, but both are thought
of, in the present context, primarily as landowning agriculturists.
The high castes are also thought to be ultimately related in that sons
of the great ascetics became Brahmins and Rajputs depending upon
their choice of occupation. On the other hand, Doms are identified
as artisans — people who work with their hands. A word which is some-

CASTE 2 1 3

times used in reference to Doms, and which carries the same connota-
tions of pollution and inferiority implied in that term, is shilpkar,
artisan. In this category fall the craftsmen and specialists who would
be classified as Shudras and untouchables on the plains. The same sorts
of disabilities are imposed upon them as upon untouchables of the
plains.

The Doms of Sirkanda are blacksmiths, barbers, and Bajgis. Car-
penters, weavers, and shoemakers will be included in this discussion,
since these groups are continually in contact with Sirkanda villagers
and are residents of the area. Mention will also be made of Bedas (the
rope sliders and musicians described in previous chapters), who are
indigenous to the area and are still in occasional contact with Sirkanda.
All these groups are lumped together under the term "Dom" by high-
caste villagers, who often assert that "Doms are all alike." A blacksmith
remarked, "They [high castes] treat all low castes alike. We are just
Doms to them even though there are great differences among low
castes." 5 High-caste people know the distinctions among low castes
but consider them unimportant and irrelevant, as indeed they are to
high-caste people. All low castes are considered by them to be approxi-
mately equally polluting.

Blacksmiths and carpenters actually form one endogamous group,
although there is a strong tendency for males of the families to remain
in one or the other occupational specialty. There are no formal ritual
distinctions between them, and for most purposes they can be con-
sidered a single caste. Dom goldsmiths also fall in this group. These
three occupational specialties are collectively and exclusively termed
Mistri. This group and barbers and weavers are roughly equivalent
in status and consistently rank above other Dom castes in the ritual
hierarchy among Doms. The barbers, being of alien origin, have been
in a somewhat anomalous position. They share a common marriage
pool with Mistris, although both barbers and Mistris claim ritual
superiority and often will not allow one another in their kitchens.
Weavers are generally considered to be slightly lower in status than
Mistris and they, on their part, claim to be higher. Both observe
ritual distance in that they do not normally go into one another's
kitchens. Intermarriage is allowed but is not encouraged and is in-
frequent. Weavers are in nearly all respects equivalent in status to
the Sirkanda barbers and, like Mistris, intermarry freely with them.
Thus, the barbers form a link between Mistris and weavers. This
entire group is of nearly equivalent status, so that a rigid hierarchy
cannot be outlined. All claim to be highest; all claim to be polluted
by others in the group; all occasionally intermarry.

214 CASTE

Below the above-listed group of artisan castes are the Bajgis,
musicians who are also tailors and basket weavers and in this area
usually perform the functions of barber (although Mistris, too, may
perform these duties). Restrictions placed upon Bajgis in their relation-
ships with Mistris are, according to the latter, the same as those
Rajputs place on all Doms. In practice, however, the restrictions are
not as far-reaching or carefully observed as those between Doms and
Rajputs. Bajgis frequently challenge the superiority of Mistris, but
if forced to a showdown will concede it. Roughly equivalent in rank
to Bajgis are other musician castes such as Beda. The most defiling
characteristic of these people is apparently their occupation as musi-
cians and dancers, with the assumed correlate that their women are
prostitutes. This interpretation is based both upon observation of
caste ranking and upon local explanations, which usually include men-
tion not only of the occupation of Bajgi men but also their women's
role as prostitutes. To explain the significance of this in terms of the
pollution concept, we might turn to an interesting point made by
Stevenson, who distinguishes between "external pollution" and "in-
ternal pollution," the latter being much more defiling and difficult to
counteract (Stevenson, 1954, p. 52). Stevenson points out that bodily
emissions are ritually polluting to Hindus. He speculates that this
may explain the fact (true also in Sirkanda) that intercaste sexual
congress is much more serious for a woman than for a man. A lower-
caste partner defiles a woman permanently and irreparably but affects
a man only slightly. Stevenson hypothesizes that, ". . . since in sexual
intercourse it is the man who emits the polluting secretion, and the
woman who receives it internally, the man is exposed only to external
pollution, which can be removed by a bath, whereas the woman is
internally polluted [so that she cannot be purified] . . ." (Stevenson,
1954, p. 57). Perhaps the low ritual status of the musician castes might
be accounted for by interpreting the sexual availability and assumed
prostitution of their women in these terms — their women are con-
tinually and irreparably polluted by sexual contacts with members of
many castes. More simply, they are occupationally associated with
human emissions which are polluting, and therefore they are polluted.
Bajgi women in Sirkanda have given up dancing, and the Beda women
in this area dance less than formerly in an effort to overcome this
stigma.

Lowest among the Dom castes are the shoemakers and skinners of
dead animals (Mochis, Chamars). They are often musicians as well.
These people are treated by Bajgis and all higher castes much as
Doms in general are treated by Rajputs and Brahmins. They are the

CASTE 215

scapegoats for Bajgis, who are the scapegoats for Mistris, just as all
Doms are scapegoats for higher castes. Mochis and Chamars are con-
sistently ranked lower by all castes than other Doms and are reportedly
even more restricted in their contacts with high castes. Their rank
is apparently due primarily to their uniquely defiling occupation. It
is often asserted or suspected by others that they occasionally eat not
only dead buffaloes, but even dead cows or oxen. Their alien origin
and persistent cultural and residential separateness in the area prob-
ably contribute to the great social distance between them and other
castes. Much of what Bhatbair people "know" about them is rumor.
This added factor makes assessment of their caste status relative to
other Doms especially difficult. While they are definitely the lowest of
the low castes, it is difficult to determine to what extent high-caste
attitudes toward them differ from those toward indigenous Doms in
terms of pollution, and to what extent cultural factors are accountable.

Among low castes, then, there are three hierarchically ranked en-
dogamous groups whose ritual status is linked to their traditional
occupational specialties: all are artisans and therefore, by Pahari stand-
ards, polluted. On this basis alone the clean castes treat them as a group
and largely ignore their differences. Among artisans the purest are
those who practice only a craft (metalwork, woodworking, weaving,
barbering), whose women are not suspected of prostitution, and who
do not handle dead animals. Inferior to them are those castes who,
in addition to whatever other occupation they may follow, are mu-
sicians and whose women are dancers and therefore suspected of prosti-
tution. Lowest of all are those whose work is with dead animals. There
is a consistent progression of pollution: All low castes are artisans,
those who are also musicians are lower than those who are not, and
those who also deal with dead animals are lower than those who do
not (see table 5).6

The details of this hierarchy differ from those reported by Atkinson
for Garhwal and Kumaon. Speaking of the Doms of this area he says:

According to popular estimation, they are divided into four grades, all
equally impure and outside ordinary caste life, but furnishing certain dis-
tinctions from occupation and the like which bring up the first grade very
close to the lower forms of Rajput clans and these again connect with
Brahmans, so that no link in the chain of social distinction between the
highest and the lowest is wanting. (Atkinson, 1886, p. 444)

Atkinson then lists the four grades or status levels of Doms, start-
ing at the top, as: (1) smiths, carpenter-masons, weavers, and "Khasiyas
degraded for caste offences"; (2) basketmakers, wood turners, oil

2l6 CASTE

pressers, messengers, miners; (3) leather workers; (4) "vagrant tribes of
musicians, dancers, jugglers, acrobats, etc." The differences between
that account and the situation in Bhatbair are that the second category
is missing in Bhatbair, since wood turners are categorized with other
carpenters, and basket makers are musicians. The castes which approx-
imate the third and fourth are reversed in status level. Perhaps the
facts that leather workers are not indigenous to the Bhatbair area
and that in Garhwal and Kumaon their women are not thought of
as prostitutes account for the reversal of their status.

TABLE 5
Caste Ranking in Sirkanda

Brahmins
1 ' Rajputs

Blacksmith-carpenters
(Mistris)

Barbers, weavers

Bajgis Bedas

I 4 J Shoemakers

1. Clean castes (twice-born) — landowning agriculturists.

2. Unclean castes (achut) — artisans.

3. Castes in which men play musical instruments and women dance.

4. Castes that handle dead animals and/or scavenge.

In this schematic representation, a solid line indicates strongly enforced
prohibition of intermarriage; the broken line indicates ideal prohibition of
marriage but rare toleration of it; the dotted line indicates some claims to
ritual distance, including marriage restrictions, but little distance in fact.
Those caste names which appear on the same horizontal plane are of approxi-
mately equal status.

In general, a caste's members may freely engage in any activity
less defiling than that appropriate to their caste status or irrelevant
to caste status, but not in anything more defiling. Caste terminology
is instructive in this context. Mention has already been made of the
use of the traditional Pahari greeting as indicating relative status,
and that clean castes refer to all the lower castes as "Doms" and con-
sider them "achut." Among low castes the term Dom is resented and
is applied only to castes lower than one's own. Blacksmiths refer to
Bajgis and shoemakers as Doms but deny that the term is properly
applied to themselves. Bajgis, in turn, apply the term only to shoe-
makers, and shoemakers resent the appellation, preferring a more

CASTE 2 1 7

specific caste term. The government listed all these low-caste groups
as "scheduled castes" in the 1951 census. The census taker who came
to Sirkanda termed them Harijans, Gandhi's term for untouchables —
literally "children of god." As one Bajgi said, "In the census they
listed all who are not Brahmins or Rajputs as Harijans. Of course that
is absurd. There aren't any shoemakers or sweepers in this village,
and they are the only real Harijans." Low-caste people resent general
terms which lump them with other low castes. They prefer specific
occupational terms. Some low-caste groups have alternative terms of
reference which they prefer to the usual one. Thus Bajgis often refer
to themselves as Darzi or Auji (tailor). Shoemakers are often called
Chamar by other castes, but prefer to be called "shoemaker."

Class Differences

Within castes there are differential advantages and status based
in part on wealth. In general, as has been shown previously, high-
caste villagers control the wealth of Sirkanda. Wealth correlates
positively with caste but is not rigidly bound to it. Although great
differences in land ownership and income do not exist among high-
caste families, this does not mean that no differences are found. Some
families are in debt; others have money to lend. Some are financially
better off than they were a few years or generations ago; others have
less than formerly. Changes in fortune are a prime topic of gossip
in the village. One Rajput family came into a large and unexpected
sum of money obtained by dubious means from Muslims at the time
of India's partition. They now possess the best house in Sirkanda,
display their wealth, and affect airs of superiority — behavior which is
repulsive to the entire village. Villagers commented that they are
little people who got too much money and do not know how to live
with it. "A big person is unchanged by a change in fortune; a small
person is overcome with it. Money is like a rainstorm which affects
the ocean hardly at all, but makes a mountain stream into a raging
torrent that tries to carry away the mountain with it."

Low castes are, on the whole, much less secure financially than are
the high castes. However, there are differences among them. At present
the blacksmiths and the two small Bajgi households are the poorest
villagers. They have no land to cultivate, they own few animals, and
their traditional income is scanty. The barber and the two larger
Bajgi families are better off. They have animals and some culti-
vable land in addition to income from their traditional occupations.
They are almost as well off as the poorer Rajputs. The shoemakers of

2 1 8 CASTE

this area are as well off as the Bajgis. They too own land and have
a sizable traditional income. Several artisans of the Mistri group who
live in areas adjacent to Sirkanda have become very well-to-do, and
live in a manner superior to that of Sirkanda Rajputs.

Thus, while "class" differences are correlated with caste status,
there are exceptions and there is differential wealth within castes.
Such deviations and differences do not affect caste status (that is, ritual
purity), but they are relevant to individual status, standard of living,
and to interpersonal and intercaste relations. Blacksmith-Bajgi rivalry
in Sirkanda is based partly upon the disparity between their tradi-
tional ritual status and their present relative economic status (cf. Rowe,
1960a, pp. 70 ff.). According to reports, blacksmiths were formerly
relatively well off, and friction between them and the Bajgis was
minimal. Sirkanda Brahmins hold their own in the village, despite
their small numbers, largely because their wealth backs up their
ritual status.

Low castes hesitate to press for their rights partly because of their
economic inferiority to, and dependence upon, high castes. A black-
smith said, "We cannot press charges against a Rajput. He and his
caste-fellows would pay bribes and hire lawyers such as we could
never afford. Twenty Rajputs could raise Rs. 100 by collecting Rs. 5
each. What could two poor blacksmiths do against that?" A Bajgi
does not hesitate to press charges against a blacksmith because he
knows he can bring at least as much money (and therefore influence),
to bear as can the blacksmith.

Although there are well-known differences in income within the
village, display of wealth is tightly controlled by custom. To display
wealth in disapproved or unconventional ways is to invite public
criticism, social pressure, and even open hostility, as has the newly
rich Rajput family mentioned above. This is true at all caste levels
but especially at the lower levels, where it would be considered not
only bad taste but arrogance. The objection is not to living beyond
one's means, but to living beyond the approved level of expenditure,
which is considerably below the means of a number of families.

Wealth is legitimately displayed among high-caste villagers in ex-
pensive weddings and, to a lesser extent, in funerals and other cere-
monies, and in the giving of charity to Brahmins. This ritualistic
sphere is really the only one in which "conspicuous consumption" is
practiced without disapproval. A family's wealth is shown in other
ways: in the jewelry worn on special occasions by its women; in large,
well-built houses (here the number of ornately carved pillars on the
veranda is a publicly recognized criterion of wealth, as is the type

CASTE 2 1 9

of roof, thatch being least expensive, corrugated iron next, and slate
most expensive); in the number and quality of cattle owned; in the
number and quality of household utensils (brass trays, tumblers, and
so on) used; and in the quality of food served in the house. Educa-
tion of a son outside the village is coming to indicate wealth, but
its absence is not a sign of lack of wealth. Wealth cannot be legit-
imately displayed by low-caste members in the village. Any display
would bring prompt and painful sanctions.

For the most part, however, wealth is not displayed even by high
castes. Instead, money is stored in concealment or loaned on interest.
Clothing, household furnishings, and most other material possessions,
as well as the work and other activities of well-to-do families, are
not significantly different from those of poor people. Wealth is usually
apparent more in spite of its owner's wishes than because of them.
There is an effort at nonconspicuous consumption, or perhaps more
accurately, conspicuous nonconsumption.

The Wisers have explained an important reason for villagers' reluc-
tance to advertise their wealth — the fear of exploitation by agents of
authority, fear of increased taxes, and so on — by paraphrasing village
sentiment in these terms:

In self-protection we have learned to make it almost impossible for anyone
to tell who is prospering among us. You may guess, and we may guess. But
who is going to tell us if we are right. . . . Some among us are honestly poor.
And the rest of us, excepting the affirmed leaders, have learned to make a
show of poverty. (C. V. and W. H. Wiser, 1951, p. 159)

The wealthiest man in Sirkanda shows his wealth in no perceptible
way except in the very favorable marriages (with educated boys of
prominent, well-to-do families) that he has contracted for his daughters.
Quality of marriage arrangements in the family is the ultimate crite-
rion of wealth and status within the caste.

The most important factor in "class" or status within the caste
besides wealth and the things correlated with it, is Sanskritization. The
more a family conforms to Sanskritic orthodoxy in life-cycle cere-
monies, the more highly it is thought of. This is not wholly inde-
pendent of wealth, as Sanskritization is expensive, but it is a somewhat
different dimension.

Caste Stereotypes

Associated with conventional criteria of caste ranking in terms
of ritual status are a number of caste stereotypes. Rajputs are said

220 CASTE

by low-caste informants to be proud, haughty, jealous of their power,
authoritarian. Brahmins are supposed to be niggardly with their own
property and exacting in their demands for charity and service from
others. "A Brahmin is never satisfied," is a frequent saying. Low-caste
people are considered by high-caste informants to be lazy, impulsive,
thoughtless, ignorant, irresponsible, childlike, addicted to music, drink,
and narcotics. Low castes are capable of taking advantage of these
stereotypes by using them as excuses for behavior that would other-
wise be inexcusable. If a Bajgi is found holding a drunken songfest
in his home or is away from the village pursuing an illicit love
affair while he is supposed to be preparing a set of clothes for a Rajput
client, he will be cursed but, after all, "what can be expected of such
people?"

Low castes hold particular stereotypes about one another which
are shared to varying degrees by high castes. Blacksmiths are con-
sidered physically powerful, Bajgis are considered hard-working and
clever, and the barber (whose functions are elsewhere performed by
Bajgis) is also considered clever. Shoemakers are considered degraded
in every respect. (Cf. Oakley, 1905, pp. 260 ff.)

Stereotypes tend to justify the system and perpetuate it. The mem-
bers of various castes are often said to be suited to that caste and
no other by their very nature. To give advantages to low castes would
be useless, as they would not be able to make use of them. "Maggots
which live in feces cannot live in grain." On the other hand, in Sirkanda
as elsewhere, this philosophy is not put to the test by making ad-
vantages accessible to low castes. Rather, low castes are actively denied
access to advantages which high castes assert they could not use or
would not want anyway.

Attitudes About Caste

Individual attitudes on caste are shaped from earliest childhood
by observation of behavior of elders and by direct instruction or
admonition. A blacksmith discussed his own feeling toward Bajgis in
response to a query following up apparent inconsistencies between his
behavior and his statements about caste. He had condemned caste
discrimination while stoutly denying that he would treat a lower-caste
person as he wished to be treated by Rajputs.

I don't mingle with Bajgis any more than Rajputs mingle with blacksmiths.
In my heart I cannot bring myself to do it. My conscience will not let me.
I have been taught all my life that they are inferior and unclean. Even if
I wanted to, I could not overcome this feeling. Secondly, it is due to the

CASTE 221

character of their women who become dancers and prostitutes — and the men
play drums. Such people cannot be my equals. Finally, there is pressure from
other people in my caste and the knowledge of what higher castes would
think. If I mixed with Bajgis the high castes would treat me just like a
Bajgi, and it would degrade my whole caste. High castes would expect the
same behavior from blacksmith women as from Bajgi women. My own caste
members would be angry with me.

This particular informant made all the following statements rep-
resenting a variety of different views: "The high castes treat all low
castes alike, even shoemakers." "Caste is an undesirable thing, it divides
people and makes for conflict; it should be abolished." "There are
really only two castes in the world, men and women; all others are
artificial and unnecessary." "I don't mix with Bajgis and shoemakers;
I treat those Doms just as Rajputs treat low castes."

Low-caste people's resentment of their own caste position is con-
stantly evident. Equally evident is their need to rationalize their
status relative to others. Their thoughts along these lines vary from
individual to individual and from time to time. An attitude shared
and expressed by several low-caste villagers was stated by one: "Eng-
lishmen and Muslims are untouchable because they have an alien
religion and they eat beef. This is logical. We are Hindus and we do
not eat beef, yet we too are treated as untouchables. This is not
proper. We should be accorded higher status."

A blacksmith explained, "Long ago Bajgis used to kill and eat
cattle and buffaloes and that is why they are untouchable. Our people
never did that, so we should not be called untouchable." This alleged
inconsistency in occupation and status is usually rationalized in terms
of high-caste numerical and economic dominance. It is commonly
also explained in terms of mythology, which takes away some of the
sting of the negative associations of low status, usually by providing
respectable ancestry for the caste and explaining its decline in terms
of unfortunate circumstances. Atkinson notes that in Kumaon "the
Doms, like all the others, claim an exalted origin and say that they
are the descendants of a Brahman named Gorakhnath and were
turned out of caste for eating forbidden food" (Atkinson, 1886, p. 446).
The blacksmiths and Bajgis of Sirkanda lay claim to relatively recent
but unrecognized Rajput and Brahmin ancestry, respectively.

Q. What is your caste?

A. We are Lohars [blacksmiths]. Sometimes we are called Mistri.
Q. Are you Doms?

A. No, Bajgis and shoemakers are Doms. We are above them. We cannot
take water from them.

222 CASTE

Q. Someone told me you are Mahar Rajputs.

A. Oh yes, that is true, but people here don't recognize that. We are
descended from a Mahar Rajput who lived several generations ago farther
up in the hills and who took a blacksmith wife. Since we are his children
we are of his caste, but people here call us blacksmiths. There are other such
cases. A man should be of the caste of his father, but people don't recognize
that when the mother is of low caste.

Similarly, the Chamars of Senapur trace their ancestry to Rajputs
or Brahmins (Cohn, 1954, pp. 112 ff.).

Besides rationalizations for the low status of their family or caste,
people almost invariably seek rationalizations for individual status;
that is, in addition to denying that his caste deserves low status or
has always been of low status, a man usually denies that, granted the
low status of his caste, he is deserving of having been accorded such
caste affiliation or that he is an ordinary low-caste person. The attitude
is, "I was meant for greater things." There are many stories of low-
caste individuals who were destined for high status because of their
good works in previous lives, but through mischance were assigned
to life in a low caste. The same people who adhere to claims of un-
recognized higher caste status believe (in apparent contradiction) that
they have been assigned low status in some almost accidental way.
Thus, the blacksmith who gave the above responses on his caste
status also recounted the story of his own origin as it had been told
him by a Brahmin who read his horoscope.

The pundit read my horoscope and told me my fate and my history. He
is always right in these matters. He said that in my previous life I was a
wealthy Raja. I was well known for my valour, honor, and piety, and the
fairness of my rule. I was destined for a very high rebirth. One time I de-
cided to give away a great deal of wealth in charity to my Brahmins. In order
to do this I had my servants bake a great many pastries. Instead of filling
them with food, I instructed my attendants and servants to stuff them with
gold and distribute them to the Brahmins. This was done secretly and was
intended as a surprise. This was a very meritorious act. However, my assist-
ants and servants were dishonest, and instead of doing as I had instructed
they stole the gold and filled the pastries with stones. These were distributed
and I was unaware of the deceit. The Brahmins who received them did not
tell me because they did not know my intent, so I never found out. Although
my intent was pure, I was guilty of putting trust in untrustworthy assistants.
Therefore, when I died, I was punished by being born a blacksmith. It is
my fate in this life never to succeed in any enterprise and never to acquire
or retain wealth. This has proved true. I have tried several schemes to get
money and all have failed. When I had some money which I inherited from
my father, it was stolen from me on a trip into Dehra Dun. The only way

CASTE 2 23

I can get anything out o£ money is to spend it at once before I lose it, and
that is my policy. In my next life I am destined to be a merchant — perhaps
I will do better then.

A Bajgi from another village had a comparable tale. He had been
an ascetic destined to become a Raja in the next life. While meditating
on the infinite, he had ignored the plight of a cow entangled in the
brush. As punishment he was reincarnated as a lowly Dom.

Such stories of deserved but unintentionally provoked punishment
apparently account for many instances of low birth. They are dis-
seminated by Brahmins and serve as a means of rationalization and
comfort for low-caste people. As such they are effective means of
control for the high castes. Whether or to what extent Brahmins are
conscious of the usefulness of the stories I did not find out, but their
effect is obvious. As will become evident, the high castes in general
and the Brahmins in particular have a strong vested interest in the
caste system and therefore in keeping the low castes in their place
and relatively content.

No informants were found who said, in effect, "I was a scoundrel
in a previous life and now I am getting my just desserts." Neither
was any caste found whose members said in effect: "We have always
done defiling work. This is what we were created to do and we do it.
Therefore we are untouchable." These (and particularly the first)
seem to be psychologically untenable positions for individuals to
accept. On the other hand, high castes readily admitted that they
were receiving the deserved rewards of exemplary previous lives.

In Sirkanda, as in Hindu society generally, there appears to be an
inherent contradiction between what some would call social structure
and culture. That is, according to the concept of dharma (here definable
as inherent duty), every member of every caste group is enjoined to
behave according to the hereditary station of the caste in the hierarchy.
This is an explicit injunction against social mobility in the context of
the caste system.

At the same time there is upward mobility among low castes in
India. Lipset and Bendix have asserted that

the fact that there is constant striving for upward mobility in the most status-
ridden society in the world, adds considerable weight to the hypothesis derived
from Veblin, according to which a system of stratification is a fundamental
source of mobility motivation in and of itself. Apparently, there are impera-
tives which prompt men to resist and reject an inferior status and these
imperatives persist regardless of the way in which any given society has
legitimated inequality. (Lipset and Bendix, i960, p. 63)

224 CASTE

If this statement is correct then a dilemma faces the low-caste person
who according to his religion should stay in his place, but according
to this "natural law" wants to rise.

In Sirkanda and, I believe, throughout India, this dilemma is often
neatly resolved by exactly the mechanisms described above: accept-
ing cast dharma but denying that one's apparent dharma is his real
dharma. That is, the individual takes the position that a person should
behave according to his caste status but that he or his kin group or
jati is not really of the caste or status to which others ascribe him
(or them). He is of a higher caste. He aspires to live as a member of
his "true" caste in order to fulfill his true dharma. Therefore his
mobility aspirations are legitimized and even made imperative. He
rationalizes his own mobility aspirations without challenging the
system.

The over-all pattern is one of high castes justifying their superior
position in terms of myths and religio-philosophical beliefs. Subor-
dinate castes assert their superiority to some castes while rationalizing
their inferiority to others, and their consequent mobility aspirations,
in terms of unrecognized but deserved higher status and dharma than
that accorded them by society.7

In Sirkanda expressed resentment about caste was usually put in
terms of the way it affects "me" or "my family" or "my caste." It
rarely extended to other groups. Even those low-caste individuals most
piously opposed to upper-caste abuses were likely to be equally abusive
to their caste inferiors. No group admitted to being lower than all
others. Even shoemakers would point to sweepers of the plains or to
beef-eating Muslims or Christians as being lower than themselves.

Thus, objections to the caste system, as such, were not common.
Those voiced in Sirkanda were isolated references apparently derived
from the opinions of a previous low-caste schoolteacher or others who
had had experience with the Arya Samaj (a reform movement), or
Gandhian or government views on caste. They did not reflect a con-
sistent objection in the village to the caste system.

This is a feature of intercaste relations common throughout India.
Gould (1961a, p. 946) attributes it in large part to "repressed hostility
which manifests itself not in the form of rejecting the caste system
but in the form of its victims trying to seize control of it and thereby
expiate their frustrations on the same battle field where they acquired
them." I would suggest that an important factor is simply that all
castes are so imbued with the value of hierarchy that none wants
to associate with those it considers inferior. Should the caste system
be abolished, the opportunity to mingle as equals with superiors would

CASTE 225

be accompanied by the necessity to mingle as equals with inferiors.
The latter would be an especially distasteful prospect to a group
whose only claim to status is a tenuous superiority to one or a few
degraded groups. To seek to undermine the caste system would be to
seek the end of one's own superiority to at least some other groups.
To seek to raise the status of one's own caste within the system gives,
promise of superiority to more groups. The former would put any
caste in an ambivalent position relative to its previous status; the
latter would bring clear-cut advantage. Above all, the Indian ethos
is not one of equality but one of hierarchy, of which caste is the
epitome. The upwardly mobile Indian peasant seeks superiority, not
equality.

Individuals who are the objects of caste discrimination generally
direct their resentment not at the intergroup phenomenon as such,
but at the manifestation of it which affects them personally. They
resent not caste discrimination, but discrimination directed against
themselves. One of the persistent problems of "Harijan uplift" in
India is that, once untouchables become educated and can break away
from many of the restrictions and disabilities imposed by their tradi-
tional caste identification, they tend to disassociate themselves from
their caste fellows and (with notable exceptions) contribute nothing
to further the cause which was at least part of the reason they were
given scholarships or other extra advantages which made possible
their rise in status. The former schoolteacher in Sirkanda, who was of
blacksmith caste, had been popular among all castes in the village,
but was accused by his caste-fellows of having curried favor among
high castes and others to his own advantage while ignoring the wel-
fare of his own caste. This pattern is not surprising, nor is it unique
to the hills or to India. It occurs in intergroup relations everywhere.
The same kinds of accusations are directed toward the "emancipated
Negro" of the United States.

However, the pattern seems to be somewhat more acute and general
in India, partly, perhaps, because in India intergroup relations are
largely carried out on such an intensely personal, individual level.

I sometimes discussed the status of whites, Indians, and Negroes in
Africa with urban people of Dehra Dun. There was great indignation
about the situation, but it invariably boiled down to the statement,
"They treat Indians just like Negroes there — it is a grave injustice."
Curiosity about American race relations was of the same quality, in
essence: "How does an Indian fare in America? Like a white [as he
deserves] or like a Negro?" No one I knew was disturbed about or
interested in racial or national discrimination in principle. This

226 CASTE

attitude reflects the pervasiveness of the theme of hierarchy. People
assume there is everywhere a hierarchy in social relations based on
inherited status, with attendant discrimination, and they want to be
sure that they are at or near the top.

Changes in Caste

Judging by testimony of informants and by the few available
accounts of Pahari castes, caste organization today is not greatly
different from what it was 50 or 150 years ago. It seems probable that
there has been a tendency to shift from a more consistently dual
division of the society into low- and high-caste groups to one in
which, while this division has remained paramount, status differences
within the two groups have become increasingly pronounced. On the
other hand, this may reflect a high-caste conception of the earlier
situation (a conception to which the high castes still adhere) as com-
pared to a more realistic appraisal of the present situation. On the
high-caste side of the barrier, it seems likely that in earlier days (before
increasing outside contact and "Sanskritization") Rajput-Brahmin
distinctness was less emphasized. Intermarriage between the groups
was reportedly more frequent, and very likely other ritual barriers
were much reduced.

Doms were evidently somewhat less differentiated among them-
selves than at present. Bajgis claim that at one time they and the
Mistris were one endogamous group. Under the Raja of Garhwal and
more recently in nearby Tehri Garhwal, Doms of all subgroups were
legally prohibited from owning land or dwellings outright. They
could not own or use eating utensils of a particular alloy. They were
prohibited from wearing gold jewelry. Today one still finds an occa-
sional old Dom woman whose marital nose ring is a silver replica of
the gold one (now worn by all castes) which was formerly restricted,
to the high castes. In many areas Doms were prohibited from living
within the village. Thus, they were treated as a homogeneous group
by the high castes, and the scanty evidence available suggests that
they may have considered themselves more nearly as a single group.

Doms claim to have been in former days the recipients of more
paternalistic concern by the high castes than at present. They were
given land, houses, grain, and gratuities by the high-caste people for
whom they worked. The high castes interceded on their behalf and
protected their interests from outsiders when necessary. In return the
high castes demanded respect and obedience. Physical punishment and
economic sanctions were used by high castes to maintain their status.

CASTE 227

Now high-caste people complain that the low castes are growing dis-
respectful and independent. According to high-caste people, the low
castes no longer want to live by their traditional occupations. Low-
caste people complain that the high castes no longer give them
gratuities or even their proper pay, that high castes will not come to
their aid when they need help. Each blames the other for having
brought about the changes.

Among low castes there has been increased rivalry. Apparently the
situation was relatively stable shortly before Independence — or at
least it seems to informants to have been so in retrospect. Then the
Bajgis, who had rankled under the growing superiority complex of
the Mistris, began to press for equality, encouraged by rumors of the
Congress political party's platform and their economic superiority
and traditional claims of former equality. This has irritated the
Mistris, whose status position is shaky at best. Now each group claims
superiority, although Bajgis often display deference behavior and are
evidently ready to settle for equality. A recent council case resulted
when a Bajgi's sarcastic greeting of "Samani Thakur" to a blacksmith
was accepted and acknowledged by the blacksmith as though it had
been respectfully offered. The Bajgi lodged a complaint at the urging
of high-caste villagers. In typical council fashion the Bajgi was fined
for thus greeting a low-caste man; the blacksmith was fined an equal
amount for accepting the greeting. The council members, all high-
caste men who had pressed for the suit in the first place, enjoyed a
liquor party from the proceeds — a result which low-caste people not
involved in the case claimed was predictable and planned in advance.

In general the trend in behavior associated with caste status is now
toward adoption of orthodox Hindu behavior in some contexts. As
has been noted in previous chapters, the trend is especially evident
in life-cycle and annual ceremonies. In other spheres of life the trend
is limited largely to concealing unorthodox practices from potentially
critical outsiders, while in some matters there is little or no self-con-
sciousness about such practices. Thus sacrifice of buffaloes, consump-
tion of buffalo meat, and eating of carcasses of recently dead wild
animals found in the jungle are carefully guarded secrets. That goats
are killed purely for their meat is also denied. Sexual freedom within
the caste and sharing of wives among brothers are concealed. Cross-
caste and polygynous marriages are not readily admitted. Liquor and
meat consumption are admitted with little concern, but excuses are
occasionally offered. Widow remarriage and lack of seclusion of women
cause no embarrassment at all. There is surprisingly little caste differ-
ence in attitudes and behavior on these matters, although low castes

228 CASTE

cannot afford to carry out Sanskritic rituals to the extent that high
castes can. They are somewhat less inhibited about their non-San-
skritic behavior than are the high castes, who have a greater prestige
stake and who are also in somewhat closer contact with educated
Brahmins and other advocates of Sanskritization. When high-caste
Paharis are accused of behavior proscribed in plains Hinduism,
especially that involving ritual purity, they often deny the behavior
themselves and attribute it to low-caste Paharis.

INTERCASTE RELATIONS

Intercaste relations, as practiced and enforced in Sirkanda, differ
significantly from ideal norms. Analysis of intercaste behavior as it
compares to the normative model will contribute to an understanding
of the dynamics of stability and change in the caste system. The
previous chapter dealt primarily with caste organization, and with
attitudes and behavior which are publicly recognized. The discussion
will turn now to the nature of caste and the conduct of intercaste
relations as revealed by observation of behavior to which caste is
relevant, both behavior which is "relatively easy for the individual
to manipulate at will," and that "in regard to which he seems to have
little concern or control" (Goffman, 1959, p. 7; cf. Berreman, 1962c).
As a result, the functioning of the system will be made more com-
prehensible.

Relations Among High Castes

As has been noted, intermarriage can and does occur between
Rajputs and Brahmins in and around Sirkanda. Although such mar-
riages are disapproved and are never arranged as initial marriages,
relatively little indignation is aroused by them once they have occurred.
Brahmins engage in considerable informal social interaction with Raj-
puts, as is inevitable in a village where Brahmins are in such a small
minority. This is also the case, however, in nearby villages where
the two castes are more evenly distributed. They can be seen sharing

23O INTERCASTE RELATIONS

freely even the wooden pipestem, although this is verbally denied.
Brahmins and Rajputs do not usually distinguish rank by deferential
behavior, seating arrangements, greetings, or the like, although an
honored Brahmin from outside may receive such deference from Raj-
puts.

Extramarital sex relations between Brahmins and Rajputs are
treated in no appreciably different light than those within each of
these castes. The most celebrated beauty of the Sirkanda area was
a Brahmin girl of a neighboring village who was sexually available to
Rajputs and Brahmins alike. What little critical gossip circulated
about her was concerned with the frequency and openness of her
contacts, not their intercaste character. In fact, her reputation and
the identity of several of her lovers (in Sirkanda all were Rajputs)
were revealed to me some time before I learned that she was a Brahmin.
Her family voiced no objection, and even her husband kept quiet. The
analysis given by villagers was, "He can't say anything— if he did she
might leave him and then he would have nothing. It is better to share
something good than to lose it altogether."

Relations between Rajputs and Brahmins are not always amicable,
and the power distribution between the two groups often comes to
light when conflict arises. In Sirkanda Rajput-Brahmin relations have
been somewhat strained because, while the Brahmins are said to have
been the founders of the village and are one of the more prosperous
families, they are far outnumbered not only as a caste but as a sib
and lineage, so that they are outnumbered in interfamily and inter-
sib disputes as well as in intercaste ones. A dispute over ownership
of a valuable tree on the border between Brahmin property and a
Rajput family's property created a tense situation in which the Rajput
family head threatened the Brahmin with physical punishment if he
carried out his stated intent to cut the tree. It was well known by all
that the Rajput could make good his threat with the help of sib-
fellows if necessary. Cooler heads prevailed when the Brahmin braved
the Rajput's threat and felled the tree. Had there not been Rajputs
who valued peace over power, the Brahmin would probably not have
dared risk such action.

A further irritant in relations between the high castes is that the
Rajputs depend upon outside Brahmins for most of their ritual needs
but expect the local Brahmin to be available in emergencies, while
at the same time they disparage his capabilities in religious matters.
The outside Brahmin is respected partly because he restricts his prac-
tice to high castes and partly because he devotes full time to his
religious duties. The local Brahmin serves the low castes also and

INTERCASTE RELATIONS 23 1

spends much of his time on agriculture because he does not get enough
high-caste clients to make a living by his traditional occupation alone.
It is a circular situation and causes resentment on both sides. More-
over, as a result of daily contact and competition, it is evident to
villagers that the local Brahmin is a very ordinary person while the
alien Brahmin has an aura of purity, wisdom, and infallibility about
him which is perhaps due more to the lack of intimate contact with
these Rajputs than to any inherent priestly superiority. The effect of
distance is confirmed by the Sirkanda Brahmin, who receives more
respect in alien villages than in his own. Any religious practitioner
in these hills would confirm the adage that familiarity breeds con-
tempt. Upreti (1894, p. 378) quotes a Pahari proverb to this effect. The
relative position of the alien Brahmin is suggested not only by his
practice in Sirkanda but by his reception there. He invariably stays
at the homes of Rajputs, who compete for the honor, rather than
at that of his Sirkanda caste-fellows, and he himself speaks disparag-
ingly of the local Brahmin's abilities and merits.

As a result, the Brahmin family of Sirkanda is somewhat isolated
socially from Rajputs, not because of ritual barriers, but because of
a combination of circumstances among which ritual differences are
minor. However, Tula, the purohit or traditional Brahmin for
Sirkanda, participates intimately with Rajputs in every type of inter-
action including drinking and meat-eating parties. He is also known
as a great ladies' man. Important in this context is the fact that
Tula retains his position by tradition — he will not lose his clientele
because of his behavior. Moreover, in his position he is neither re-
quired nor expected to have special knowledge or special virtue. The
Sirkanda Brahmin, on the other hand, has a voluntary clientele and
risks losing clients or failing to get more if he does not retain what
esteem he has in their eyes. Since he is responsible for ceremonies, he
is expected to know more than other people and perhaps be more
pure than they — expectations which are not verified in daily contact
with him in the village. Also, the charity given the village Brahmin
can as meritoriously be given to any other Brahmin, whereas Tula
receives traditional charity which must go to him or his family regard-
less of circumstances. As a Bajgi said of the local Brahmin, "He works
for Rajputs just like we do; he can't afford to displease them."

Therefore the local Brahmin is sensitive to public opinion. He
participated little in Rajput drinking parties. He was extremely reluc-
tant to discuss the village or his work with me if there was a chance
others might find out, for fear that villagers would accuse him of
divulging secrets and would bring pressure to bear upon him. Similar

232 INTERCASTE RELATIONS

charges were frequently made against other villagers and were usually
ignored or dismissed, but the Brahmin was especially sensitive to them.
He was ever aware of his minority status, his vulnerability, and their
implications for his place in the village which was, after all, his home.

Relations Among Low Castes

Among low castes there is also considerable deviation from stated
norms. Although blacksmiths and carpenters form an endogamous
group, it is not uncommon for one to offer the other only the bowl of
his pipe, not the stem, just as though there were a great difference in
status between the two. Weavers and barbers keep themselves ritually
separated from Mistris in some contexts. There is evidently a reciprocal
feeling of superiority among all these groups. On the other hand, mar-
riage is acceptable among them and inter-sex contact is very free. More-
over, unfriendly rivalry among them appears to be at a minimum,
partly because not all are represented in the village and so they do not
have close contact with one another.

In contrast, Bajgis are said by all the above-listed low-caste groups
to be inferior in ritual status. This is made explicit in a number of
disabilities imposed on Bajgis. The pipestem is never shared with
them. Restrictions on eating and drinking are observed about as
carefully as those across the high-low caste boundary. Intermarriage is
denied, and no cases were reported. Bajgis often yield the seat of
honor to blacksmiths in the presence of outsiders. Bajgis occasionally
strike back by proclaiming their own superiority and imposing the
same restrictions upon other low castes which those castes inflict on
Bajgis. The attitude of the Bajgis is, "If they can do it to us, we
can do it to them too." The Bajgis' reaction toward the other groups
is often one of resentment that people of approximately their own
status should turn against them and affect superiority. In the presence
of outsiders, or even of some high-status villagers, blacksmiths assume
an air of superiority which is not characteristic of their normal rela-
tionship with Bajgis and which is therefore irritating to the latter.
Relations between the two groups are not helped by the fact that
blacksmiths are less prosperous than Bajgis and yet claim ritual and
social superiority. The discrepancy constantly rankles both parties.
Evidently the distance between them is neither too great nor too little
to discourage rivalry.

Bajgis consistently claim equality even when not claiming superior-
ity. In reality there is considerable freedom of interaction between
these groups and also considerable rivalry. Sex relations are frequent,

INTERCASTE RELATIONS 2 33

with Bajgi women being more openly pursued by men of the higher
status group than blacksmith women are pursued by Bajgi men. Friend-
ship is common across this line, and there is a much greater air of
camaraderie in the relations between men of these groups than across
the high-low caste boundary. Deference behavior is not practiced be-
tween these groups, although both claim to expect it of the other.
Cigarettes are shared freely, quite in contrast to blacksmiths' private
assurances that this is not done. Despite such evidence of relatively
close relations, Bajgis recognize that their claims to equality are un-
likely to be recognized and they get small comfort from knowledge
that the government of India supports their rights.

Contacts between Mistris and Bajgis, on the one hand, and shoe-
makers, on the other, are few. The former groups are anxious to
disavow any equivalence between themselves and the lowly shoemakers.
Geographical and cultural distance prevent much contact, but social
and ritual considerations would probably be sufficient in themselves to
accomplish the same thing. Shoemakers are the true Doms or Harijans
from the point of view of Bajgis, though high castes recognize rela-
tively little status difference between them. With regard to sex con-
tact and social interaction, the relationship between the shoemakers
and the Bajgis is comparable to that between Rajputs and Doms.

Shoemakers themselves adjust to their status as aliens who are lowest
of the low by keeping away from other people to a considerable extent.
They live by themselves, and thereby avoid many of the discriminatory
acts directed toward their caste. They retain the cultural distinctive-
ness derived from their alien origin in the Kangra Valley area far to
the west. They cling to their assertions of superiority to sweepers of
the plains and to non-Hindus.

High-Low Caste Relations

Relationships between high and low castes do not always conform
to the ideal patterns of paternalistic control and maintenance of ritual
distance by the high castes. The most conspicuous deviations are the
examples of intercaste marriage and elopement discussed in chapter 5.
Although instances of this type which cross the high-low caste boundary
are widely disapproved and result in dissolution of the relationship
or ostracism of the high-caste member (who is usually the man) by
his caste-fellows, the fact that they occur is significant. Much more
common than marriage or elopement are instances of informal liaisons
and sex relations between high- and low-caste individuals. Most often
high-caste men take advantage of the vulnerability and traditional

234 INTERCASTE RELATIONS

receptivity of low-caste women. Such situations attract little or no
attention and, in fact, are routine. They become the subject of gossip
only if they are flagrantly pursued, if a particular union becomes well
established, or if a man is openly accused by his wife of indulging too
freely in such escapades. Low-caste people overtly accept the situation
with a shrug and perhaps a bitter comment: "What can we do about
it? We are at their mercy." They harbor strong resentment, however,
and express it privately.

Relations between low-caste men and high-caste women are strongly
condemned and severely punished if detected by high-caste people. A
low-caste informant who had an apparently well-earned reputation
as a ladies' man in his own and other low castes commented, "There
is plenty of opportunity for sex relations in all castes, but I would not
risk an affair with a high-caste woman. I fear for my head." It was
universally agreed that a man caught in such a relationship would
be beaten, probably to death, or chased out of the village. Low-caste
informants asserted that such affairs were occasionally pursued success-
fully but that the man involved had to be very sure of the cooperation
of the woman because she could easily bring about his downfall by
complaining to others, and if the relationship were discovered she
might put on a show of indignation and shift the blame to her lover.
The man takes most of the risk. A low-caste informant said:

One of us would approach a Rajput or Brahmin woman only if we were
sure she had her eye upon us. It is, after all, the woman who takes the
initiative in sexual matters by making her wishes known through looks and
signals. Contact would have to be made secretly in a secluded spot unknown
to others. If the woman were willing, the relationship might be carried out
and continued successfully. If the woman were unwilling and she were a
gracious person who wished to avoid trouble, her reaction to such an ap-
proach would be to reply, "Brother, you have asked me once and I excuse
you. I respect you as my brother, but never ask me again. I am your sister."
This would be the end of the matter and no one would ever know. However,
if the low-caste man were foolish enough to persist or if the woman were
touchy, she might at once go to the men of her family and complain and
then the low-caste man would be in grave trouble. For this reason we avoid
such dangers.

That such relationships do occur is widely known, although they are
concealed by high castes. Their results are less drastic than verbal
testimony would lead one to believe, as evidenced by the shoemaker-
Brahmin elopement described in chapter 5. A recent case in Sirkanda
had been that of a Bajgi man and a Rajput woman, the first of two wives
of a prominent Rajput householder. The lovers had evidently been

INTERCASTE RELATIONS 235

carrying on a secret affair for some time when they were accidentally
surprised in the jungle by the village Brahmin. The Brahmin was in-
different or unable to identify the man, who beat a hasty retreat, and
he told no one. The Rajput woman was, however, afraid that the
Brahmin would spread the word through the village. She revealed her
fear in confidence to a trusted friend. Later the two women had a falling
out, and the erstwhile friend exposed the illicit relationship to others.
The Brahmin, who up to this point had kept his knowledge to him-
self, corroborated the story. The cuckolded husband did nothing to
punish the low-caste man until another Rajput man had occasion to
beat the offender for a different reason. At that time the aggrieved
husband loudly encouraged the beating and shouted, "Beat him to
death!" The end result was that the beaten man and his father went
to court and on the basis of testimony by witnesses to the beating,
including a Rajput of a rival faction, received a judgment in their
favor by which the two Rajput men each had to pay a fine. The hus-
band held his tongue throughout. To admit publicly to having been
cuckolded by a Bajgi was apparently harder than to ignore it publicly.
His wife continued to be his first wife, and her lover was sufficiently
frightened to break off the relationship. This was certainly not the only
affair of its kind, but it was the most recent one to have come to public
notice.

In intercaste sex relations in Sirkanda, as in Negro-white sex rela-
tions in a town of the southern United States reported by Dollard, "It
would seem . . . that the taboo falls heaviest on social acknowledge-
ment of such relations rather than on the fact of their occurrence"
(Dollard, 1957, p. 151).

Persistent resentment is harbored by low-caste people toward high-
caste people in Sirkanda, especially by blacksmiths, the most deprived
group in the village. Most of the Bajgis and the barber make an
adequate income, have some land and animals of their own to supple-
ment their income, and therefore have less cause for complaint. They
share the blacksmiths' resentment of the indignity of their position
and the injustice of their treatment, but economic well-being softens
their feelings. Only the blacksmiths feel acutely underpaid and under-
privileged. They have been blocked by Rajputs in their recent efforts
to acquire land in the village. When a strike was described to a black-
smith he responded by saying,

It would never work here. The Rajputs would just beat up the ones who
refused to work and throw them out of the village. Then they would find
someone else to take their place — at twice the pay if necessary. We are small
in numbers and therefore weak. The Rajputs hold the key to the low castes

236 INTERCASTE RELATIONS

and can manipulate us as they wish, just as you manipulate the radio dial.
Rajputs don't want us to have land because they want us to have to come
begging for our grain payment. When we ask for land, they laugh at us.
The Bajgis and barber don't join with us in our efforts to get land because
they make a good income from their work, and anyway they already have
some land. We, on the other hand, have neither land nor a good income
from our work. We are alone in our desire to get land, and it seems hopeless.

High- and low-caste men alike affirmed the obedience accorded
high-caste men by low-caste men. However, subtle countermeasures
were admitted by low-castes and complained of by high-castes. Thus,
in performing his craft a blacksmith claimed to do work quickly and
well for those who paid him promptly and justly and who treated him
civilly (as many, in fact, did) and to do it poorly and slowly for
delinquent or arrogant clients. An incident of this nature occurred
in the writer's presence. While it conformed to no reported pattern
of intercaste behavior, it aroused no comment among witnesses, so
it was apparently not out of the range of acceptable behavior. A
young Rajput man of a large household known by village artisans
as a bad credit risk and not a particularly desirable client, came to the
blacksmith with an axe he wanted sharpened. The blacksmith, who
was listening to the anthropologist's radio, took the axe, inspected it
with evident distaste, and announced, "This axe is worth eight annas,
[to cents]. My file is worth 15 rupees [three dollars]. It would spoil
my valuable file to sharpen this worthless axe. Go find a flat rock and
sharpen it yourself." Further feeble entreaties brought nothing but
refusal from the blacksmith, and the Rajput left, presumably in search
of a flat rock.

Needless to say, the blacksmith did not frequently practice this
pattern of behavior, as he could not have done it with impunity in
other circumstances. He would not have tried it, for instance, with
an older or more prestigeful man. However, in an occasional well-
chosen situation he could get by with it, and it obviously gave him
considerable satisfaction. Members of all low castes relished tales of
moral victories by low-castes over high-caste people.

High-caste men do not observe rules of pollution carefully when
they go to the larger towns. There they eat in public places with
people whose caste they do not know, and even with low-caste people
of their own village. They patronize the same prostitutes as are
patronized by Doms (and by Brahmins from the plains). They would
even eat at the homes of the writer and his Muslim assistant, although
they would not eat in their house in the village. A 15-year-old Rajput
boy expressed the attitude of most men when he politely refused to

INTERCASTE RELATIONS 237

share the writer's dinner in the village with the comment, "I would
gladly eat if we were elsewhere, but we are in my own village. Here
everyone knows my caste and I must be careful what I do. In town no
one knows me nor I them, and I do as I please without fear of
consequences."

The influence of the caste-equality, anti-discrimination stand of
the dominant Congress party in India has been felt even in Sirkanda.
The president of the village council occasionally has to attend a
meeting or workshop for council presidents in the valley. There he
not only hears the official policy of the party to which he nominally
belongs but has to practice it by eating with fellow officials and civil
servants of all castes. The food is prepared and served by people of
unknown (and highly suspect) caste status. He makes no effort to
conceal this when he returns to the village, but it does not alter the
fact that in the village he is as caste-conscious and discriminatory as
anyone else.

Anti-discrimination talk led to the nearest thing to a test case of
caste discrimination that has ever occurred in the village — in itself an
indication of new or changing attitudes. The blacksmith who holds
the village council seat reserved for untouchables planned to brew
tea at his home and serve it at a village council meeting which was to
be attended by the teacher, village level worker, accountant, tax col-
lector, forestry officer, economic cooperative supervisor, and the anthro-
pologist in addition to the local council members. The blacksmith's
intent was to press charges of caste discrimination if anyone refused the
tea, and he counted on the outsiders to form an august body of im-
partial witnesses. He made his intent known only to the teacher,
village level worker, and anthropologist. However, when the time
came, no tea appeared. The advocate of the test case lamely claimed
to have forgotten, but in reality he had apparently not felt like risking
the probable consequences of such a defiant act in his vulnerable
position.

In discussing caste relations in the village a young Rajput man
complained:

I would like to spend more time at your house talking with you and
listening to your radio. I would share tea and food with you. I could learn
many things and have a good time. But too many low-caste people come
there. I do not care so much about that for myself, but people here are very
strict. On the plains caste rules are broken frequently and everything is
breaking down, but not here. We cannot share food, drink, utensils, or
cigarettes with Doms. I must live my life in this village. If I associate with
those low people my people will be angry with me. I could be your friend

238 INTERCASTE RELATIONS

and associate with Doms at your house, but I would lose my friends in the
village. There would be no comfort in the friendship of a few Doms when
you are gone.

People of all castes denied that there were individual differences
in the strictness of caste observance among high-caste individuals.
Everyone maintained that in the village all were equally rigid. Ob-
servation proved this to be not entirely the case. While most high-
caste men shared cigarettes only in their group, a few shared them
with low-caste men as well. Some interacted socially with low-caste
men regularly; others did so rarely or not at all. Age and position
within the family were evidently important factors in this context.
Old men and young men were noticeably less concerned with
caste rules governing social interaction than were middle-aged men.
Middle-aged men (roughly 35 to 55 years of age) were usually also
the effective family heads. They were the leaders of their castes. They
felt responsible for maintaining the status of their families and their
castes. They were the most suspicious of outsiders, the most proud,
the most arrogant in their relationship to low castes, the most authori-
tarian in all their relationships.

In a village of the size and caste composition of Sirkanda, it is
almost inevitable that a good deal of informal social interaction will
take place across caste lines, including the high-low caste boundary.
There are very few low-caste people in a predominantly Rajput village,
and the village is isolated from other villages. If low-caste people
are to have social life at all, it must be to a large extent with Rajputs.
In Sirkanda this is especially true because the two blacksmith house-
holds are not on good terms with one another, nor are the three main
Bajgi households, because of intrafamilial strife. Moreover, the black-
smiths and Bajgis are on somewhat strained terms, largely because of
the claims to higher status held by the blacksmiths and the reality
of greater prosperity of the Bajgis. Bajgis have a number of relatives
within easy walking distance of the village, and much of their social
life is with these caste-fellows. In the village, however, they, like
the blacksmiths who have few relatives, must find friends among the
Rajputs. The amount of friendly interaction between high and low
castes is therefore considerable — evidently more than would be found
on the plains and certainly more than one might expect from the
expressed attitudes of these groups toward one another and the formal
restrictions placed upon interaction between them. That such inter-
action is allowed by the high castes does not mean that caste status is
ambiguous but rather that it is so secure that it is not jeopardized by
interaction of this kind.

INTERCASTE RELATIONS 239

The place of work of each artisan is a gathering place for men
of all castes to sit and talk. The tailor's veranda is rarely without one
or more high-caste people watching the craftsman at his work, talking
to him and his relatives and to other visitors. The same is true of the
blacksmith's workshop, the basketmaker's porch, the sites of house
construction, and the carpenters' work areas. The village shops and
school serve also as meeting places for informal social interaction
across caste boundaries. High-caste houses are the locus of high-caste
interaction, but not uncommonly a low-caste person may participate,
sitting on the steps or doorsill or standing outside.

Intercaste groups form in various circumstances. Intercaste work
groups are common, including both cooperative labor and independ-
ent labor performed in groups (as tending goats and gathering wood
or fodder). In times of trouble or need, caste boundaries may be subor-
dinated. A Rajput family of Sirkanda owes its prosperity to loans
granted it a generation ago by a sympathetic Brahmin when the
family was not producing enough to survive. A nearby Brahmin family
took in a homeless Rajput widow and her invalid son when they had
nowhere to turn. Rajputs did not hesitate to borrow money from a
blacksmith two generations ago. When the Brahmins were in need
of a place to stay during part of the time their house was being re-
built in 1958, they stayed with a blacksmith family although they ate
separately. Those who have to travel away from the village seek travel-
ing companions, and no caste is excluded from such a group. Inter-
caste groups often go to town to trade, to do business at government
offices, or even to seek entertainment together. In the village and out-
side, drinking and gambling groups are often intercaste in composition.
The famous illicit woman-selling gang of Bhatbair was made up of
a Brahmin, a Rajput, and a Bajgi who worked as a team, sharing the
risks and profits with little caste distinction.

Perhaps the most frequent occasion for intimate social interaction
across caste lines in the village is at drinking parties. There caste
barriers are largely ignored. Low-caste people may be invited to a
high-caste house where a party is to be held, and there they are allowed
to participate fully in it. Of course such parties are held on verandas
of houses, not in cooking areas, and boiled food is not served, so it is
not potentially a very polluting situation. Although low castes may
contribute liquor, they do not furnish the site for such parties. In
Sirkanda one blacksmith and one Bajgi are inveterate participants in
high-caste drinking parties.

In general, activities which are illegal, overtly disapproved, or non-
Sanskritic are much more likely to be intercaste in nature than are

240 INTERCASTE RELATIONS

those which are entirely legal, approved, and orthodox. Thus inter-
caste drinking, meat-eating, dancing, and trips to town are indulged
in often with little regard for caste differences. Even illicit sex activity
may be pursued by intercaste groups who together go to a house of
prostitution or approach low-caste girls outside the village. A Rajput
and blacksmith both told a story on themselves which occurred at
the Pahari fair. The blacksmith was paying for Ferris wheel rides
and sweets for two Muslim girls of easy virtue when the Rajput came
and joined in. Both were confident of their reward until the girls
got off the Ferris wheel and were spirited away by two husky strangers,
to the dismay of the girls' erstwhile benefactors.

Among the most colorful personalities in Sirkanda is a blacksmith
man. His company is sought and enjoyed by all. No party or discussion
is complete without him. His wit and good judgment combine to make
him popular despite his caste status. He is the greatest talker and gossip
in the village. He spreads news and helps formulate opinion. He is
the repository of knowledge not possessed by others; he remembers
things others forget. More than once Rajputs turned to him when
questioned for details of their own genealogies or family histories.
Although in many contexts he plays the role of the joker, his opinion
on serious matters is highly valued but rarely, to his disgust, acknowl-
edged. When unusual circumstances arise, he is often sent forth to
appraise the situation or express village sentiments. Numerous ex-
amples could be cited. When a horse trader came to Sirkanda, the
blacksmith was sent to look over the horses and sound out their owner
before high-caste potential customers made an appearance. He passed
judgment on the trader's honesty, his willingness to bargain, and the
value of his horses. As he himself noted, he received no thanks for his
efforts, although several villagers relied heavily on his advice in sub-
sequent purchases.

When the Brahmin family wanted to move a large rock which
they thought endangered their house but which was on village prop-
erty, Rajputs objected. The blacksmith looked it over, said "This rock
is a hazard to the Brahmin's house and should be moved," and began
to decide the best method of moving it. Soon the Rajputs were helping.
This blacksmith is the usual choice for making contact with and
appraising strangers — a role which a low-caste man can perform well
without committing the village. He is the informal channel for com-
munication with outsiders and sometimes with rival factions. When
the schoolteacher was new, the blacksmith communicated to him village
attitudes on schooling and the role of the teacher. When the anthro-
pologist considered moving into a house in the most crowded section
of the village, the blacksmith was sent to voice the objections of some

INTERCASTE RELATIONS 24 1

influential villagers (stating them as his own), who were thus able to
have their opinions voiced while denying any responsibility and, in
fact, condemning the blacksmith for inhospitality.

This is not to assert that the blacksmith is the most important man
in the village or even that his opinion was decisive in all the above
cases. It does, however, point out that one low-caste man, at least, is
important far beyond the admission or realization of his caste superiors
and far beyond their ideal of the complaisant, subservient, know-
nothing Dom. He has achieved importance largely as a result of
personal characteristics which override his caste status. It is significant
that his importance is not admitted or even realized by most villagers,
and his caste status is never forgotten or ignored.

He is apparently an atypical Dom, but he is not beyond the range
of permissible behavior in Doms. Some individuals in each of the
other Dom castes were to a lesser extent influential in the village.

High-caste jealousy of this blacksmith reveals implicit recognition
of his role. He was often accused of not knowing his place, of having
big ideas, or of being a troublemaker. His wit was sometimes dis-
paraged by those who could not equal it. His love of liquor, women,
music, laughter, and leisure were frequently criticized. That he made
fun of himself on these very grounds only served to exasperate his
detractors. As he said,

People here are very jealous. I have to be careful lest I suffer their wrath.
Even if I wanted and could afford some comforts and a better house or better
clothes I could not have them because Rajputs would accuse me of putting
on airs. I must always remain humble to them if I am to survive in this village.

This man frequently mentioned wealthy caste-fellows of his, resident
in other villages, who surpass local Rajputs in wealth and sophisti-
cation but are still untouchable in their eyes. This he considered to
be an example of unreasoning pride on the part of Rajputs. He as-
serted that they would not tolerate action on the part of low-caste
people which they consider inconsistent with caste status.

I could put these Rajputs to shame by getting a table and serving you
with food and drink at my house. They would be envious and angry and
after you are gone they would make it hard on me. Therefore I must show
that I know my station and not be unduly close to you in public social re-
lationships. They will be jealous when you give me your radio. For that
reason I will have to tell them that I bought it from you. They are jealous
to see me associating with big people.

On another occasion he remarked upon his relationship with
Rajputs:

242 INTERCASTE RELATIONS

No matter how friendly a Rajput may be at one time, he will turn against
a low-caste person the next time. All are proud and jealous of their caste
position. None are true friends to us. They always resent my presence in
social situations. They will eat with me in Dehra Dun but never in the
village. They will drink liquor with me and often invite me to sit on their
veranda to do so. If they are having something dry like parched gram [chick
peas], I can eat it with them. If they have something cooked, I am given a
separate plate. They are very strict on such matters. I am with them a great
deal, but I cannot say they are my true friends. I am always a Dom and they
are big people. Here big people mingle with other big people as friends.
I call that man truly big who mingles with high and low alike. Such are
not to be found in this village.

Vested Interests in Caste

The functioning of the caste system in an Indian village can be
assessed on several different levels. From the point of view of the
community or of the society as a system, the Indian caste system is
remarkably efficient. Ideally, it assures a stable division of labor with
a constant supply of specialists in all occupations. In return it pro-
vides the individual with an assured occupation, an assured income,
and a body of people who share his interests. It provides a religious
and philosophical rationale for differences in status and standard of
living which minimizes discontent and subversion. It is preordained
and static, so that status change is, ideally, impossible. It reduces am-
biguity by the recognition of rules and symbols segregating social
groups. It minimizes intergroup frictions. It provides stable group
identification and affiliation for individuals at all status levels and
minimizes the chances of disparity between reference group and mem-
bership group and the potentially disintegrative results of such dis-
parity.

Not only does everyone have some place within the Hindu system, but
it is significant that every group, from the Brahmin to the Chamar caste, has
been somehow integrated into the social and ceremonial round of the com-
munity and has been given some opportunity to feel indispensable and proud.
(Opler and R. Singh, 1948, p. 496)

That the system has not been completely successful; that change,
discontent, and subversion occur in spite of the system, does not belie
its relative efficiency as a system. Breakdowns in discipline, changes
in caste status, and the like, are either suppressed or rationalized.
Rationalizations become part of the system and are not remembered as
deviations from it. What these facts do reveal is that there is more

INTERCASTE RELATIONS 243

to caste than its ideal structure. Human beings are involved, and the
effects of the system on the individuals who live in it must be under-
stood if its functioning in reality is to be understood. Despite pious
statements to the contrary in India and elsewhere, no group of people
has been reported which relishes a life of deprivation and subjection
to other groups. That people submit to depressed status does not
mean that they feel it is justified nor that they would not like to see
it changed, nor, in fact, that they would not do everything in their
power to change it if given the opportunity. The rationalizations for
caste status which are consistent and convincing to those who benefit
from them or are unaffected by them seem much less so to those whose
deprivation they are expected to justify or explain. Adherence to a
religion or a religious principle may not significantly affect attitudes
and behavior to which logic would seem to tie it. It will be well,
therefore, to look briefly at caste as it affects people. As John Dollard
has said in studying caste and class structure in a town of the southern
United States,

We should like to know something not only about the class structure but
also about the differential advantages and disadvantages of membership in
any particular caste or class; and, in particular, we wish to state these ad-
vantages and disadvantages from the standpoint of the types of direct, per-
sonal, ultimately organic, gratification derived. (Dollard, 1957, pp. 97 f.)

Following Dollard, I will state the advantages which the caste
system in Sirkanda provides for its high-caste participants. Inherent
in most of these advantages are disadvantages which automatically
fall to the low castes. An effort will be made to point out as well those
ways in which low castes may benefit from the caste system. "In using
the concept of 'gains' we are not leaving the 'social' plane of percep-
tion; we will merely . . . look for a moment at the individuals rather
than at the society" (Dollard, 1957, p. 98). In listing gains the three
broad categories suggested by Dollard will be used: the economic,
sexual, and prestige gains. In addition, a fourth category, here called
"ultimate rewards," will be mentioned. Use of Dollard's categorization
for this portion of the Sirkanda data is based not upon the pressing
of data into alien molds but upon my belief in real similarities in the
situations analyzed — a position elaborated in another publication
(Berreman, 1960a).

The discussion need not be lengthy because most of the evidence has
been presented in this and other chapters. An effort will be made
not only to assert that a particular gain is achieved but to show that
gain by particular individuals is a result of the caste hierarchy. Al-

2^4 INTERCASTE RELATIONS

though the discussion is based primarily upon Sirkanda data, the same
sort of analysis, leading to essentially the same conclusions, could no
doubt be made in other Indian villages.

It seems likely that, in villages where caste dominance is less clear-
cut than in Sirkanda, our present approach would produce less con-
sistent evidence. An economically deprived high-caste or a wealthy
low-caste person might well upset the statements about economic
gain, but they would have done so in spite of caste affiliation. Caste
affiliation would doubtless still be a force in the expected direction
even if it were overcome by other factors. Such a case, looked at from
this point of view, would be an interesting study. The obvious con-
clusion would probably be that caste operates to the best advantage
of high castes when their rank is correlated positively with decisive
dominance.

Obtaining reliable data on intercaste relations and attitudes is a
difficult task. It is especially difficult in India, where any evidently
educated or urbanized person is automatically classed by low-caste
villagers as a member of the elite regardless of his professed aims and
affiliations, and by the high-caste villagers as a potential threat to
the status quo. The low-caste informant is likely to be wary of saying
or doing anything which might conceivably be held against him at
a later date. Many high-caste people, on the other hand, are aware
of official doctrines of equalitarianism and may respond verbally in
ways quite different from those which represent their true feelings
on the subject (cf. Berreman, 1962c).

In Sirkanda these were problems in obtaining data, but information
was obtained from enough different people and observations were
made in so many different situations, that a body of reliable and
apparently valid information was obtained. Where inconsistencies
occurred, they were often closely linked to situational factors. Re-
sponses of both low- and high-caste people varied predictably in the
presence of the opposite status group, while responses of some in-
dividuals varied predictably in the presence of their caste-fellows.
Since these were predictable variations, they apparently did not con-
taminate the data.

In this discussion the only status distinction to be considered will
be that between nigh and low, or clean and untouchable, castes. This
is a valid distinction in the Pahari area, as preceding discussion has
shown, in terms of both ritual and secular status attributes. Subsidiary
caste distinctions will not be dealt with because in Sirkanda they are
minor by comparison and would only serve to provide lesser examples

INTERCASTE RELATIONS 245

of the points made more clearly in an examination of the major status
distinction.

The Economic Gain

It has been pointed out that land is wealth in Sirkanda. The
high castes, by restricting land ownership to themselves before British
dominion, ensured their own economic dominance. The paramount
importance of land is recognized by Sirkanda villagers, as has been
illustrated by examples and quotations in previous chapters. It is
further illustrated by the history of the struggle which the villagers
have waged and are still waging for land. The high-caste villagers
considered it a victory when, under the British, the bhaichara land
ownership pattern was established, eliminating an intermediary land-
lord and giving the cultivators direct ownership of their lands. The
low castes considered it an even greater victory when lands formerly
allotted to them only for use (that is, they did not own these lands)
were assigned by law to their castes and later to them as individuals.
Until that time high castes had control over low castes in that lands
and dwellings alike were in the names of high-caste owners. An un-
ruly, disobedient, or disrespectful artisan was readily dispossessed of
house and livelihood. Post-Independence restriction on further de-
velopment of uncultivated lands by villagers has brought bitter resent-
ment of the government by all castes.

Efforts by low castes to acquire more land have been fought at
every turn by the high castes, not because they stand to lose land
as a result (they do not), but because they stand to lose a measure of
economic dominance. Thus, the village blacksmiths have needed only
the support of the village council president or the accountant to put
in a request to open up heretofore uncultivated village lands. The
council and its president have stated, "If you want land, you will have
to leave this village to get it." They have used their influence to
prevent the village accountant (a government official alien to the
village) from submitting a favorable report on behalf of the black-
smiths. In the closest village to Sirkanda a blacksmith had been
granted a share of village lands under the British, the only low-caste
person of that village who was so fortunate. A few years ago he decided
to sell part of this land. Over high-caste protests he sold it to a local,
landless Bajgi who was his friend. The high-caste villagers were in-
furiated. They retaliated by calling the police, making out a false
accusation of theft against the hapless blacksmith, and bribing the
police to ensure that he would be punished. As a result he served

246 INTERCASTE RELATIONS

seven months in jail. As one of his caste-fellows said, "You can't go
against the wishes of high-caste people and get away with it around
here." Low-caste people occasionally openly accuse Rajputs of "wanting
to keep us in slavery," and not wanting to see them economically
secure or self-sufficient.

To keep the low castes dependent not only assures greater income
for the higher castes; it also assures a ready supply of cheap labor.
For, as long as the low castes are dependent for their livelihood upon
high castes, the latter can call upon them for all sorts of services, under
implicit or explicit threat of economic sanctions. Low-caste members
are available to carry loads, thatch houses, pound tobacco, clean up
debris, plaster floors, run errands, or help in any work where another
person is needed. They also brew liquor, kill animals, and perform
other unpleasant, risky, or polluting tasks at high-caste bidding. As
long as they are landless or own insufficient land for survival, the high
castes can use them almost at will. Thus the high castes have a strong
economic interest in maintaining the caste system in its traditional
form, which includes economic dominance over the low castes. This
is perhaps the key enabling factor in sexual and prestige gains as well.

An economic advantage of the caste system which is specific to
Brahmins is their role as recipients of charity — charity which is neces-
sary to ensure a favorable future for the soul of the giver (cf. Weber,
1958, p. 60). This is an India-wide phenomenon and the degree to
which individual Brahmins exploit it varies, but it is apparently no
accident that the formulators and primary agents of communication
of the religio-philosophical tradition are its greatest benefactors.
Stories illustrating the necessity of giving ample charity are widely
repeated in Sirkanda and elsewhere. Such stories are told primarily
by Brahmins and are often included in their professional services of
reading the past and future fate of villagers in horoscopes. Villagers
often refer to the avariciousness of Brahmins and sarcastically hint
at pecuniary motives in their visits or scheduling of ceremonies.

All economic gain is not on the side of the high castes; there are
compensatory gains for low castes. Specific gains which accrue to low
castes include exclusive access to a number of foods and other goods.
Shoemakers, for example, get the carcasses of dead animals, which
provide both materials for their trade and meat for consumption. Such
advantages are more pronounced on the plains, where diet differs by
caste more than it does in Sirkanda. Under the traditional system the
low-caste individual was assured of work and of payment for it as long
as he did not offend his employer. He also expected and received a
paternalistic sort of care and protection much like that accorded the

INTERCASTE RELATIONS 247

Negro who knew his place in the American "Old South." These were
notable compensations for dependent status. However, they often did
not satisfy the low-caste person and were in themselves aspects of the
exploitative economic situation perpetuated by the upper castes. They
were the kinds of compensations an authoritarian system often offers
its subjects.

Srinivas notes in the Mysore village he studied that:

While the Governments of India and Mysore want to abolish Untouch-
ability, and the Untouchables themselves want to improve their position, the
locally dominant caste stands in its way: its members want the Untouchables
to supply them with cheap labor and perform degrading tasks. . . . They
have the twin sanction of physical force and boycott at their disposal.
(Srinivas, 1959, p. 4)

K. Singh found that, out of five frequently cited grounds for conflict
leading to tension between landowners and low castes in a plains
village, two were directly economic and two others indirectly so
(Singh, 1957, p. 185). Lewis makes much of the economic advantages
which accrue to high-caste landlords at the expense of the low castes
that serve them in Rampur (Lewis, 1958, pp. 55 ff., especially 79-84).
P. M. Mahar (1958), interviewing in a plains village, found that a
Rajput elder ". . . disapproves of educational advancement for un-
touchables not only because of sacred precedent, but also because it
will lead to a depletion of the Rajputs' labor supply." She quotes the
informant:

Now the government says all should become one. We are afraid of that.
Now we can't get anyone to work for us. If they are all clerks and gentlemen,
who will plow our fields? Even now it is hard to get laborers. It used to be
you could get a man for five rupees to work for you. Now they ask thirty
or forty. (P. M. Mahar, 1958, p. 56)

This reasoning explains in part the emotional reaction by high-
caste people in Sirkanda to low-caste attempts to acquire land and
achieve economic security. Such attempts are a threat to their own
economic dominance and ultimately to their status.

The Sexual Gain -

The sexual advantages of high-caste status in Sirkanda are not
inconsiderable, but they are less pronounced and are probably less
important motives for maintenance of the status quo than is the case
in many plains areas. The reason is simply that in Sirkanda there is
a relatively high degree of permissiveness in matters of extramarital
sex relations within and between castes. Men and women rarely lack

248 INTERCASTE RELATIONS

for variety in sexual partners if that is their desire. On the plains, in
contrast, there is close supervision of high-caste women.

This does not, however, alter the basic fact that in Sirkanda the
low castes provide a constant source of available women for high-
caste men, and the men are not reluctant to make use of this advantage.
It is the one relationship in which a man can find a sex partner to
whom he owes nothing and from whom he needs fear no trouble.
While he may make sexual use of women of his own caste, he will
be expected to supply them with trinkets, cigarettes, sweets, or other
favors to show his appreciation. He also runs the risk of a fight with
an irate husband or his own irate wife. A low-caste woman expects
and usually gets no more than a cigarette for her favors; her husband is
in no position to exact revenge; the chances of the news reaching the
high-caste man's wife are less across caste boundaries than within a
caste, and even if it does the wife is unlikely to be particularly upset
or able to win sympathy from others if she is. A low-caste woman is
not considered a serious threat to a high-caste wife. Such liaisons are
expected and are usually lightly dismissed. An example of this attitude
is found in the case of a prominent Rajput, whose short-lived liaison
with a low-caste girl has become a family joke told by his wife, whereas
his liaison with a Rajput widow caused a serious and never-again
mentioned family fight. Low-caste men's resentment of free use of
their women by high-caste men is suppressed but comes to the surface
readily when the subject is discussed out of the hearing of high castes.
Low-caste women, too, resent the advantage taken of them by high-
caste men.

Relations between low-caste men and high-caste women, when they
do occur, are extremely risky for both parties in view of the attitudes
of high-caste men. Informants of all castes affirm the dire nature of the
punishment which would befall the participants if caught. Exagger-
ated fear is also revealed in precautions taken by high-caste men to
prevent such contacts, for example, often prohibiting their women-
folk from attending intercaste occasions where drinking and dancing
occur, such as fairs, celebrations, and so on.

Some idea of the sexual gain enjoyed by high-caste men can be
inferred from the fact that eleven recognized unions of some duration
between high-caste men and low-caste women were reported for
Sirkanda and neighboring villages. Only one recognized union between
a low-caste man and a high-caste woman occurred, and it was in a
more distant village than any of the others (thus, more unusual, and
so reported and remembered more widely). It involved an elopement
out of the area, unlike the other eleven, all of which were carried on

INTERCASTE RELATIONS 249

in Bhatbair. Similarly, casual affairs and extramarital sex relationships
were frequently reported between high-caste men and low-caste
women, while extremely few cases of the reverse situation were re-
ported.1 Whether these reports reflect sexual behavior accurately or
not (and it is likely that they do), they certainly reflect attitudes
accurately.

There is little in the way of compensatory sexual gain for low-
caste men in Sirkanda. Looking at the situation from the low-caste
woman's point of view, it could be argued that she derives a gain
comparable to that of the high-caste man and that the high-caste
woman is deprived as is the low-caste man. Actually, the high-caste
man has the advantage of choice which the low-caste woman is denied,
though both have access to a greater variety of partners than have
their spouses. In other areas of India greater freedom of sexual ex-
pression is allowed low-caste people within their status group than is
allowed high-caste people within theirs, but there is little caste differ-
ence on this score in Sirkanda. An economic gain may accrue to low-
caste people from the high-caste sexual gain. Occasionally a low-caste
woman succeeds in getting money, clothing, or other goods from a
high-caste lover, and in at least one instance in a village near Sirkanda
the low-caste husband encouraged his wife in an affair that bordered
between prostitution and concubinage.

Sexual gain derived by high castes as a result of their status position
in the system is frequently implicit in accounts from other areas of
India and seems, in fact, to be a universal aspect of caste in India.
Majumdar (1944, pp. 175 f.) remarks that Rajput men cannot marry
or have social intercourse with Dom women but can have sex relations
with them or keep them as mistresses. Lewis (1958, p. 257) notes that
". . . lower-caste women are more vulnerable than the women of other
castes to the sexual advances of higher-caste men." Cohn (1955, p. 68)
says that "it was a commonplace a generation ago for a Thakur man
to have sexual relations with a Camar woman. This still occurs . . ."
Stevenson asserts:

A man may keep as a lover or a concubine a lower status woman, from
whose hand he would not take either food or water, without requiring
further purification than a bath after contact. A high status woman con-
ducting a liaison with a lower status man, however, would be expelled from
her status group. (Stevenson, 1954, p. 57)

A standard joke told in the plains concerns two Chamar women watch-
ing the funeral procession of an old landlord. As the body is carried
past, one hand falls out from under the shroud and flops about. One

■ '

25O INTERCASTE RELATIONS

Chamar woman turns to the other and says, "You see, Thakur Singh
is dead, but he still beckons to us."

In general the situation conforms to that in Southerntown as re-
ported by Dollard. In this and the following quotations from Dollard,
I have substituted the words in brackets: "high-caste" for "white," and
"low-caste" for "Negro."

. . . [High-caste] men, by virtue of their caste position, have access to two
classes of women, those of the [high] and [low] castes. The same condition
is somewhat true of the [low-caste] women, except that they are rather the
objects of the gain than the choosers, though it is a fact that they have some
degree of access to [high-caste] men as well as to the men of their own caste.
[Low-caste] men and [high-caste] women, on the other hand, are limited to
their own castes in sexual choices. (Dollard, 1957, p. 135)

The Prestige Gain

High-caste people gain, by virtue of their caste status alone, def-
erence from others, constant reinforcement of a feeling of superiority,
and a permanent scapegoat group in the form of the lower castes.

The gain here is very simple. It consists in the fact that a member of the
[high] caste has an automatic right to demand forms of behavior from [low-
caste people] which serve to increase his own self-esteem.

It must always be remembered that in the end this deference is demanded
and not merely independently given. (Dollard, 1957, p. 174)

In Sirkanda relative prestige and the attendant symbols conform
well to these statements. The honorific greeting accorded the high
castes by low castes has been mentioned previously. It is enforced if
it is not volunteered. A respectful form of address and reference is always
expected from low castes by high castes. The continual inflation of
high-caste ego is emphasized not only by verbal adulation but by
other symbolic acts. Low castes sit lower than high castes. Normally
they also remain outside high-caste houses, squatting on a step or
doorsill rather than entering. On the other hand, the high-caste person
can enter freely into the house of the low-caste person. Low castes step
out of the way when high castes pass on the trail. They perform small
services for high-caste members upon demand, including especially
those that are inconvenient, risky, dirty, or defiling. They are expected
to live in inferior dwellings, use inferior utensils, wear inferior clothes
and ornaments, and generally play the role demanded by their in-
ferior status. They must follow the leadership of high-caste people
and refrain from pressing complaints. They must accept judgments
handed down from the high castes, even in their private affairs, in-

INTERCASTE RELATIONS 25 1

eluding the punishment or abuse that often accompanies them. They
must endure quietly many kinds of impositions upon them, including
sexual impositions upon their women. They must often beg payment
for their labor and receive it humbly. They must carefully avoid
high-caste temples and be ever on their guard not to defile the persons
or possessions of high-caste people according to the traditional rules
governing intercaste contacts. Powerful sanctions — economic, social,
and physical — are at hand to enforce these rules and are not used
reluctantly. Gould (1961a, pp. 946, 948) cites examples, from the plains
village of Sherupur, of fear of "rule of the lower orders" as high-caste
villagers saw their position of respect being undermined.

The whole tenor of intercaste relations is prescribed by custom.
The high-caste person is paternalistic, authoritarian. The low-caste
person is submissive, subservient. He must not pay undue attention
to high-caste women. He should be indulgent and friendly to high-
caste children. He is expected to be respectful, agreeable, mildly
humorous in the presence of his superiors. He must laugh at jokes
at his expense and conceal resentment if he feels it. He must know
how to respond with just the right note of humor or self-deprecation
when intercaste relations are discussed in a mixed-caste group. A light
remark can bring an ominous response.

In Sirkanda the pride of high-caste people is proverbial. They are
quick to censure the low-caste person who steps out of line. The rules
are known by everyone, but a misjudgment by a low-caste person can
lead to a rebuke or physical punishment. When a Rajput asked me
the time, using an honorific form of address, a low-caste man made
bold to look at his tattoo-watch (a common form of male adornment)
and say "4:30 by my watch." The questioner shot back edgily,
"Watches sometimes get broken."

In general the impression I gained, not only from my own experi-
ence but from that of the schoolteacher and other outsiders in the
village, was that, while high-caste villagers feel relatively secure in
their prestigeful status position in the village, they feel very insecure
in the presence of outsiders, especially non-Paharis. This is doubtless
related to self-consciousness about their own unorthodox Hindu prac-
tices and their resultant low status in the eyes of plains people, com-
bined with awareness of their relatively simple or primitive living
conditions, clothing, and foods. Their insecurity in this context is
revealed in self-conscious jokes about their being "wild men" (un-
domesticated), being destined to be reborn as donkeys on the plains,
being readily identifiable in town by their boorish manners and rustic
appearance despite efforts to appear cosmopolitan, and the like. An-

252 INTERCASTE RELATIONS

other factor may be their awareness of government efforts to raise
the status and living conditions of low-caste people. For these reasons,
among others, the high castes show extreme reluctance to have any-
thing to do with outsiders — to feed them, talk to them, or even offer
them a seat or a smoke. After he had been in the village three months,
the schoolteacher complained that no one had invited him to eat and
no high-caste person had even inquired as to his origin or family
status. He was himself a Pahari Rajput from a neighboring area. Stories
of the suspicious and inhospitable nature of Paharis are many and
graphic. To the extent that they are true, they seem to be based
largely on a general insecurity in the presence of the status threat
posed by strangers. Any stranger is an unknown quantity who may be
of low-caste or alien religion and hence defiling. He may be an advocate
of intercaste equality and interaction. Worse yet, he might try to put
some such ideology into practice, thereby defiling high castes or putting
ideas into the minds of low castes. On the other hand, he may be a
high-caste Hindu and hence be critical or contemptuous of unorthodox
claimants to high status.

Also of fundamental significance is the fact that this is an isolated
and relatively closed social system where kin, caste, and community
ties are extremely important. Anyone who lacks these familiar ties
is outside the system and cannot be placed within it readily if at all.
Such a person poses an inherent threat to the community. Therefore,
the best way to handle him is to get rid of him, or, if that cannot
be done quickly, to ignore him. If he remains he will gradually come
to be accepted as an outsider who is resident in the village, tolerated
but ignored except in the context of his legitimate function in the
village. Fear of outside agents of authority also leads to rejection or
avoidance of strangers.

Low-caste people are often more relaxed than are those of high
caste in the presence of outsiders. Evidently they have little to fear
and nothing to lose in terms of status. They are often used by high-
caste people, perhaps for this reason, to deal with strangers. This
aggravates intercaste tensions because low-caste people come to know
outsiders (the schoolteacher, the village level worker, the anthro-
pologist) better than do high-caste people. They learn more of the
ways of outsiders, become accustomed to being with them, and even
acquire habits and ideas from them. The high-caste people fear what
may be passing between the local untouchables and the potentially
threatening outsider, and dread that they will be "found out," ridi-
culed, and hence lose status. They therefore try to get rid of outsiders
if possible, or keep them away from close contacts with villagers.

NTERCASTE RELATIONS 253

Middle-aged high-caste men are especially sensitive on this point.
3nly when they are secure in the knowledge that their status and
mportance are recognized and properly appreciated do they relax
heir pose. On the other hand, it is not a matter of consciously striving
'or prestige and deference. These are considered to be natural and
ust concomitants of caste status. When infractions of rules of inter-
:aste relations are brought to the notice of these men, they usually
ittribute them to outside influence or general deterioration from the
golden age of amicable intercaste relations — a period in the indefinite
3ast when the low castes knew their place and the high castes were
ible to play their paternalistic role to best advantage.

Compensatory gains of low-caste people are few in the social sphere,
rhey strive to maintain status superiority to one or more castes which
hey consider lower than themselves. Their only direct gain is in
dative freedom from ritual and status restrictions. In the hills, this
loes not result in important advantages for low castes because all
:astes are relatively unorthodox. In the plains the demands of ritual
jurity and social distance on high castes are considerably stricter,
particularly in such matters as the seclusion of women, continence
)f widows, abstinence from liquor and meat, and so on. There low
:astes can ignore these prohibitions while high castes must adhere to
hem, at least in public. In Sirkanda the primary advantage to low-
:aste people is in their prerogative to largely ignore prestige con-
iderations in their daily life and live with relative freedom from fear
)f loss of a respect they do not have.

Ultimate Rewards as Gains

High-caste people feel that they are justified in demanding respect
ind service from their caste inferiors, as the direct result of their own
neritorious acts in previous lives. But the matter does not end there.
\ccording to the conventional Hindu view they are destined, by the
'act that they play their high-caste role well, to reap further and even
nore desirable rewards in subsequent lives. In turn, low-caste people
:an hope to improve their lot by submitting to their fate as disad-
vantaged people. Weber (1958, p. 122) points out that "the neglect of
me's caste duties out of high pretensions unfailingly is disadvantageous
n the present or future life." In the orthodox Hindu view, high castes
:an increase their chances for ultimate rewards by increasing the
economic and prestige advantages they seek in this life, while low
:astes can increase their chances for ultimate rewards by subordinating
economic and prestige gains in this life to the cause of pursuing their
:aste duty, including the serving and honoring of high castes. Thus,

254 INTERCASTE RELATIONS

immediate and ultimate rewards are consistent in the behavior they
require of high-caste people but contradictory in the behavior they
require of low-caste people.

In Sirkanda this view is held primarily by educated, high-caste men.
Orthodox Hindu views of caste duty, as such, are not held by most
villagers. However, it is believed that by living a good life and giving
charity a person can enhance his caste status in the next life. The
opportunities for low castes in the next life are less than those for
high castes simply because the next step for them is not as high as
that for high castes. Moreover, the economic advantages possessed by
high castes enable them to give more charity at less personal sacrifice
than can low castes, thus furthering their prospects for ultimate
rewards.

Maintenance of the System

The caste system is maintained by caste stereotypes and by religio-
philosophical beliefs relating to fate and proper behavior. Many vivid
stories are told of the wonderful results of living a good life and the
dire results of improper behavior. Low- and high-caste status is rational-
ized and justified, both on caste and individual levels, as described
above. More immediately and practically, proper caste behavior is
enforced on the low castes by high castes and on all people by their
caste-fellows, through social and economic pressure and physical force.
The actions of individuals are likely to have results which affect the
entire caste. A high-caste person who pollutes himself endangers
the status of others of his caste; a low-caste person who angers a high-
caste person may bring retribution to his caste. Sanctions are most
conspicuously applied, however, across caste lines. A recalcitrant Dom
can be readily taken care of through economic pressures (such as non-
payment), by physical violence or threats thereof, by expulsion from
the village, or by legal action (wherein high castes can control the
decisions through superior wealth and judicious use of bribes).

To the present, low castes in Sirkanda have made no concerted effort
to break out of their status as a group, nor is it likely that they will
do so. Particular castes or individuals have made brief sallies in this
direction but nothing more. Relative caste status remains quite stable.
A certain flexibility among younger and older high-caste men, com-
bined with a fair degree of realistic tact on the part of low-caste
people, has prevented open conflict. Many low-caste members have
thought out rationalizations which make acceptance of their status
easier. One Bajgi said:

INTERCASTE RELATIONS 255

We actually are better off than the high castes in some ways. When they
want work done, they have to come ask us and we can refuse if we are too
busy. They have to give us grain or money periodically. When we need help
in doing our work we get high-caste people to do it and they follow our
instructions.

A blacksmith said:

Anyone who serves another is a slave regardless of his position because
he does not determine his own action. In that respect we are no worse off
than many people who are wealthy and respected. The district magistrate,
the schoolteacher, the surveyor — they are servants just as surely as are their
peons or we blacksmiths. The district magistrate has a great deal of power.
He can collect fines and send people to prison whether they are poor or
wealthy, strong or weak. Therefore he is greatly feared. However, even he
can do only what the law says. He is not to be feared as a person because
he only implements the law. He has no power beyond what the law gives
him and he can harm no man who has not disobeyed the law. He can give
no punishment not written in the law. Therefore he is just a servant of the
law like any other servant of a master. This is true of many positions. All
of us who work for others are servants.

The methods which low-caste people adopt in accommodating to
their depressed status are similar to those reported by Dollard (1957, p.
253) for Negroes in Southerntown. In Sirkanda the most common reac-
tion is adoption of attitudes of passive accommodation and acceptance
of gratifications commensurate with low-caste status. Overt aggression
is rare, though not unknown. More characteristic of the plains and
particularly of urban areas than of the hills is competition by low-
caste members for high-caste values and increased status (cf. Bailey,
1957, pp. 186 ff.; Cohn, 1955; Opler and R. Singh, 1948, p. 476; Rowe,
1960a, pp. 58 ff., 298 ff.). In Sirkanda low-caste members characteris-
tically turn their aggressive impulses toward members of their own
group. They are explicitly encouraged in this by high-caste members.
Overt aggression by low castes in Sirkanda (fights, cases referred to
council action, and even legal suits) is almost exclusively directed
toward caste-fellows or members of castes of roughly equivalent status,
often with high-caste support.

Tension between high and low castes is not lacking but is usually
kept on the covert level. It is based on the dominance of high castes
in all spheres of life, which results in differential availability of ad-
vantages, primarily economic and prestige advantages. K. Singh, in a
study of intercaste tensions in two plains villages, found essentially
the same situation as that in Sirkanda, but with overt tensions and

256 INTERCASTE RELATIONS

resultant conflict evidently considerably further developed. Speaking
of these tensions he states:

The major conflict in both villages was found to be between the Thakurs
on one side and the low castes on the other. The conflict between castes in
the two villages seems to lie in the attempts of the low-caste people to estab-
lish autonomy from Thakur domination and to reduce the extent of their
dependence upon them. Almost in its entirety, low-caste tension may be seen
in terms of their desire to assert their independence, on the one hand, and
their inability to do so on the other.

... As dependence decreases tension is also seen to decrease. (K. Singh,
1957, pp. 183 f.)

If the low castes are to decrease their dependence in Sirkanda, they
must have access to more land. At present this is an issue on which the
high castes are in no mood to conciliate and on which the low castes
are in no position to press demands.

Caste Trends

Sanskritization in relation to caste has been discussed in chapter
4. There it was noted that a trend toward adoption of the orthodox
Hindu usage of the plains is occurring among all castes in Sirkanda
as compared to the dual trend toward low-caste Sanskritization and
high-caste urbanization (or what might be called "atraditionalism")
in the plains reported by Srinivas, Cohn, and others. The high castes
of Sirkanda have adopted somewhat more of the orthodox traits than
have the low castes, but this difference seems to be more a result of
their economic ability than of differential aspirations. In Sirkanda
there is relatively little evidence that any one caste is making an
organized effort to raise its status in the system. The most conspicuous
effort has been that of the Bajgis, whose women have ceased to dance
professionally. A similar effort was underway in the Bhatbair Beda
caste. The status rivalry, and its behavioral ramifications, among black-
smiths and Bajgis is another approach to mobility, but it lacks organ-
ization or consistency, and does not affect other castes.

Therefore, low castes are anxious to rise in status, but they seldom
see any way to do it. They feel that they are subject to the will of
high castes who would never tolerate infringement upon their superior
status. Moreover, the low castes are small in numbers and divided
among themselves by the jealousies inherent in a competitive situ-
ation, such as that with which households in the same occupational
specialty are faced. As Cohn (1955, p. 74) has pointed out, mobility
in a caste system must be a group phenomenon. So far, group effort in

NTERCASTE RELATIONS 257

his direction on the low-caste level has been largely lacking in
iirkanda. Low castes realize that any organized effort to raise their
iwn status, even if it could be undertaken, would meet with bitter
md powerful opposition by the high castes. The high castes prefer to
leal with low-caste people as dependent individuals. Organized op-
)Osition is a much greater threat than individual opposition.

On a broader level, trends in Pahari culture as a whole have brought
:hanges in behavior and perhaps some enhancement in the status of
he people of this culture area relative to those of the neighboring
jlains. This, at least, is the goal of changes in ritual usages among
?aharis, and especially among people of high caste. Elsewhere this
,roal has been referred to as "plainsward mobility" (Berreman, 1961b).

Caste is not a static phenomenon. Not only are relative caste status
ind the attendant caste rules in constant flux, but the caste system
tself assumes new and varying significance under changing circum-
itances. The accommodations, alterations, and new functions of caste
n nontraditional settings have been discussed at some length by several
mthors (cf. Bose, 1958; Gadgil, 1952, pp. 184 ff.; Niehoff, 1959; Olcott,
1944; O'Malley, 1932, pp. 161 ff.; Ryan, 1953, pp. 307 ff.; Srinivas,
1955a, 1957). In these accounts emphasis is placed upon the functions
}f caste in the newly relevant political arena. Political consciousness
ind participation began to spread widely in India only shortly before
Independence. Caste has arisen as a vital unit in national and regional,
is well as local, politics. In Sirkanda these changes have not yet be-
come apparent. While Sirkanda people have voted both in local and
national elections, there has been no caste alignment evident in the
elections. The low castes do not see the vote as a means to obtain
desired ends. To put up a low-caste candidate never occurred to
Sirkanda Doms, and when an outsider suggested it the idea was dis-
missed as useless and likely to cause trouble. Doms are so outnumbered
that their attitude is realistic. They are cynical about the reservation
of a seat for untouchables on the village council. While it is good in
principle, it is an empty gesture in practice, because one low-caste
man has little influence in such a body. Doms are pleased that school-
ing is equally available to all in the village and that scholarships to
higher schools are available for untouchables, but no one plans to
make use of the latter provision simply because it is too far from the
realities of village life. Therefore, caste has not become a divisive
factor in recent years as it reportedly has in other areas. Srinivas, who
has discussed the increasingly divisive nature of caste in the villages
and the nation, has also emphasized the "bonds opposing the divisive-
ness of caste," the "links that bind together the members of different

258 INTERCASTE RELATIONS

castes who inhabit a village or a small local area" (Srinivas, 1955b, pp.
32, 34). In Sirkanda it is true, to a significantly lesser extent than in
the villages studied by Lewis and by Srinivas, that caste is "... a
distinct ethnic group with its own history, traditions, and identifica-
tions, and [that] each caste lives in more or less separate quarters in
the village" (Lewis, 1958, p. 314). In Sirkanda the numbers of in-
dividuals in all castes except Rajputs are too small to function as
separate little communities. Nevertheless, in Sirkanda caste per se is
divisive — it frequently turns the attention and loyalties of everyone
but Rajputs to the world outside the village and toward the caste
brotherhood. Caste creates intergroup friction within the village and
often prevents common purpose. However, caste organization entails
economic and ritual interdependence, and associated with it in the
village context is social interdependence. In the following chapter
factors contributing to village unity and cohesion will be discussed.
These are the factors which override intracaste identification and inter-
caste friction to make the village community a functioning social,
economic, and religious unit within the larger field of its extensions
in surrounding villages and its relationship to larger social, economic,
and administrative units.

THE VILLAGE COMMUNITY

"From time immemorial the village has been a basic and important
unit in the organization of Indian social polity" (Dube, 1955, p. 1). As
such, villages are useful units for anthropological analysis (cf. Bailey,
1959). In fact, I would maintain that in the area in which this study
was carried out, the community is the most relevant manageable unit
of analysis if one's aim is to achieve an over-all understanding of the
way of life of the people in a limited time. Detailed studies of an
entire region would be impossible. Studies of kin and caste groups
would be fruitful, but in a different direction.

Like all Indian villages, Pahari villages are not static, isolated, or
autonomous. To be understood they must be viewed in historical
perspective and in their relationship to other social, political, eco-
nomic, and religious systems of which they or some of their members
are a part. The nature and extent of these larger systems must be
determined empirically. "Like any unit in a segmentary social system,
the Indian village has to be examined to determine in what respects
it stands alone and parallels for its members the advantages and pur-
poses of similar units, and in what respects it combines in various ways
with other units to serve wider purposes" (Opler, 1956, p. 10).

SlRKANDA AS A COMMUNITY

Sirkanda is readily identifiable as a village community. It is named,
and it is called a village (gabri) by its residents and by others familiar

26o THE VILLAGE COMMUNITY

with it. Village affiliation of individuals is widely recognized in this
area by people of other villages as well as by fellow villagers. There
is village loyalty and, in some contexts, rivalry between villages. There
are stereotypes about villages as entities, as in the local saying, "Two
things to be wary of are Sirkanda's rocks and Kanda's women."
Travelers on the trail invariably inquire as to one another's village
of origin and the villages of their departure and destination. People
are often identified in conversation by personal name, caste, and
village. The villages of origin of brides brought in from outside are
long remembered.

As has been mentioned, slightly over half of all Sirkanda villagers
(206 of 384) live outside the nucleated settlement area which com-
prises the village, and yet their affiliation with it — resulting from their
origin, family ties, and property owned in the village — is not ques-
tioned. The focus of community identification is, therefore, the well-
defined nucleated settlement area containing dwellings and sur-
rounded by cultivated lands. People who identify with this settlement
rather than with other similar settlements are members of this village.
Whether or not they usually live in the settlement, it is the focus of
their religious activity and much of their social and economic activity,
and it is the location of traditional family lands and dwellings. The
village is therefore unquestionably a real, functioning social entity
of great importance to its members and to others who come into
contact with its members.

Economic and religious functions of the community and its rela-
tionship to other communities, towns, and regions in these matters
have been presented in chapters 2, 3, and 4. Kinship and caste as they
function within and beyond community boundaries and as they affect
community organization have been discussed in chapters 5, 6, and 7. In
this and the following chapters it will be my purpose to comment
further on the village community of Sirkanda as a social and ad-
ministrative entity. The village will be examined in terms of both
its internal organization — the relations of various groups and cate-
gories within the village — and its relations with, and inclusion within,
entities external to it.

Interaction by Age and Sex

Relationships between individuals within kin groups as influenced
by age, generation, and sex have been discussed in chapter 5. Age and
sex have a similar importance in the relationship of individuals un-
related by kin ties: Males almost always dominate over females; age

THE VILLAGE COMMUNITY 26 1

dominates youth. Women are usually more relaxed and sociable among
women than in the presence of men; young peoples' behavior is in-
hibited by the presence of elders. Women and young people are
especially careful of their behavior in the presence of outsiders or
members of other villages.

A recurrent situation in which these relationships were clearly mani-
fested was that presented early in the period of this research when
the anthropologist wished to secure photographs. Young men were
almost invariably willing to be photographed when alone or in the
presence of age-mates. In the presence of male elders they would
usually await an indication of approval from the elders. With women
the situation was even clearer. A young woman would refuse to be
photographed if she were in the presence of either men or older women
unless she received specific encouragement from them. With a word
of encouragement from males or elders (or in their absence) a woman
was as eager to be photographed as was a man.

Ideally a male dominates even an older female, but this does not
always work out in practice. An elderly woman can persuade a young
man to allow his wife to be photographed. At the Bhatbair fair, an
intervillage function, a group of gaily dressed young women of various
villages agreed to be photographed until a proud, middle-aged, and
inebriated Brahmin man stormed up and ordered them to refuse. An
elderly, high-status Rajput woman of Sirkanda who knew the photog-
rapher, as the Brahmin did not, and who had been observing the
proceedings, was obviously irritated by the arrogant intervention.
She interceded by stepping to the center of the group of girls. Eyeing
the Brahmin evenly, she said, "Please take my photograph." Her age,
status, and composure were sufficient to abash the Brahmin, who left,
and to encourage the women, who posed willingly for the picture.

Within the village there is great freedom in the relations among
men and women. They work, travel, and participate in village func-
tions together. Women do not leave the village in marriage parties or
other such functions, but in the village they participate freely. No-
where is this more evident than in drinking and dancing parties
where, especially if outsiders are not present, women often participate
with their menfolk. At public feasts women often eat with men.
Evidently in villages more remote than Sirkanda such participation
is even freer; Sirkanda has felt, to some extent, the effects of plains
customs on this score. Such behavior is significant as one aspect of
informal community organization.

Freedom in sexual behavior has been discussed in chapter 5, where
it was emphasized that both men and women commonly engage in

262 THE VILLAGE COMMUNITY

extramarital sexual relations. Men have especially free access to
women of lower caste than themselves and to the wives of their
brothers. Women visiting in their natal villages are also unusually
free. Much of the conversation of young men is devoted to their
sexual exploits, a topic upon which immoderate boasting is not un-
common. At fairs and other celebrations sexual activity accompanies
drinking and dancing as a source of diversion. An ideal of romantic
love pervades the culture, expressed in stories and songs, and realized,
with greater frequency than seems usual in most of India, through
elopement and extramarital liaisons. Pahari songs generally recount
tales of romantic love or of heroism. The most popular type of song
can best be described by quoting Traill (1828, p. 219):

The Byri, or Bhagnaol, is a species of duet, sung commonly by a male and
a female, who respond to each other in extemporary stanzas alternately. The
subject has commonly reference to the situation or actual occupation of the
parties, clothed in numerous metaphors and similes, drawn chiefly from
vegetable products; where the parties are skillful, the Byri is made the
vehicle of personal praise or satire: this style of singing is highly popular
in the Kamaon pergunnahs, and it is there a common saying, that no female
heart can withstand the seductions of an accomplished Byri singer. The
measure is slow and plaintive.

The haunting melodies of these songs, as verses are exchanged be-
tween young men and women working out of sight but within earshot
of one another on the steep, forest-covered hillsides, no passer-by can
easily forget. Such singing is a means of maintaining contact and
finding reassurance in lonely work in isolated areas, as well as a vehicle
for indicating romantic interest and exercising creative wit.

Caste and Community

Caste has been presented in the preceding chapter as a primarily
divisive force in the village. However, it is worth emphasizing again
that this is only part of the picture. The nature of caste in this area
creates economic and religious interdependence in the village. Every
local caste is essential to every other, and, to the extent that people
depend upon members of other castes within the village, a strong
cohesive bond is formed. From such interdependence and from the
crucial fact of residential proximity, intercaste social bonds also grow
to be effective cohesive forces. There is certain lore about the village,
its locale and people, which is shared almost exclusively among vil-

THE VILLAGE COMMUNITY 263

lagers without regard to castes and cliques. There are common atti-
tudes on many subjects. There is an essentially common body of
religious belief and behavior. These common understandings help
tie village members together.

Participation in common enterprises, ownership of common prop-
erty, and preoccupation with common problems and common antag-
onisms further bind the community together despite caste, sib, and
clique alignments. Community members participate in annual cere-
monies, ritual observances, and informal drinking, dancing, and sing-
ing parties. Cooperative work on village-owned trails, on the water
source, and in certain phases of housebuilding and agriculture also
contribute to community identification. Indicative of a degree of
village unity and interdependence is ownership by the village in com-
mon of large cooking vessels and a few large tools available to all
community members as needed. Community government (discussed
below) and informal councils are enterprises cutting across divisive
loyalties, though they also provide a setting for engaging in disputes.
Common loyalties are shared by villagers in their identification of
themselves as Sirkanda villagers. They share pride in village history,
its former relationship with the Garhwal Raja, and in him and his
kingdom.

Common problems and antagonisms are found in uniform dislike
of the present government, its programs and personnel. No faction
deviates on this matter. Rivalry with other villages and areas, and
feelings of strangeness in alien settings, serve also to bind people to-
gether before strangers. Village-wide suspicion of outsiders has a
similar effect. Certain village secrets are kept by all factions. Caste
may have a counterfactional function in that cross-caste antagonisms
make for occasional cross-factional alliances within castes, especially
in disputes with caste or status implications.

Therefore, in spite of the many cleavages and potential cleavages
in the village, there are counterforces which make it a real community
— an identification group as well as a membership group. Srinivas
(1955b, p. 35) has said that in Mysore, "the village is a community
which commands loyalty from all who live in it, irrespective of caste
affiliation. Some are first-class members of the village community, and
others are second-class members, but all are members." This is true
too in Sirkanda. The village is thus enabled to survive as a unity, to
maintain conventional understandings about behavior, and to deal
with deviants effectively.

264 THE VILLAGE COMMUNITY

Village Exogamy

An outstanding feature of Sirkanda as a community, in contrast
to many plains communities, is the absence of village exogamy (Berre-
man, ig62d; cf. Gould, 1960, 1961b). This is important in community
consensus and cohesion in that intravillage marriages create strong
ties across cleavages of sib, clan, and clique. Lewis (1958, p. 325) has
emphasized the importance of rules of village endogamy and exogamy
in the world view of those who live under them; they are "so im-
portant that it might be useful to add endogamy and exogamy as
crucial universal variables in our models of the folk society and peasant
society."

Despite absence of formal exogamy in Sirkanda, in 84 per cent of
all marriages the wife comes from outside the village. In the minority
castes (that is, all but Rajputs) all wives come from other villages
due to sib exogamy. However, the "region whose limits are determined
by kinship bonds" is relatively small in the hills as compared with
the plains. In Sirkanda, 83 per cent of all spouses come from within
four air miles of Sirkanda.

To the extent that Sirkanda marriages are contracted within a
small area and with relatively few families, most of whose members
are personally known to Sirkanda villagers, Sirkanda presents a con-
trast to Rampur, as reported by Lewis, and to similar plains villages.
The area of "rural cosmopolitanism" surrounding the village, which
Lewis found to be so extensive around Rampur, is relatively restricted
in the hills. There is a very real "isolationist" and "inward-looking"
tendency among residents of Sirkanda and neighboring communities.
This isolationism occurs not at the village level, as it would in en-
dogamous villages, but at the level of the immediate cultural area —
the area of the marriage network. There is not, therefore, sufficient
contrast between exogamous Rampur and nonexogamous Sirkanda to
test Lewis' ideas rigorously.

A contrast between Sirkanda and Rampur, which may be partially
attributable to the difference in marriage rules, is the apparently greater
flexibility, changeability, and ambiguity in factional alignment in
Sirkanda (Lewis, 1958, p. 114). This probably stems in part from the
incidence of in-village, cross-sib, and hence often cross-faction marriage
in Sirkanda. The extreme suspiciousness which Sirkanda residents
display toward outsiders mav be partlv attributable to the relatively
small, well-known group of people with whom marriages are con-

THE VILLAGE COMMUNITY 265

traded and other intercommunity relations occur. However, most dif-
ferences which appear between villages of the hills and of the plains
are more plausibly attributable to other cultural and ecological factors.

Cliques and Factions

The most prominent social units within Sirkanda are those based
upon considerations of kin and caste ties. These have been discussed
in previous chapters. However, there are other relationships among
individuals and groups which are significant to life in the community
although they are not coextensive with these units. Primary among
these are informal alliances and friendships which go to make up
cliques or factions of individuals who tend to help one another in
economic and religious undertakings, to interact with one another
more frequently than with others in social relations, and to support
one another in rivalry or disputes with others in or out of the village.
In Sirkanda these are not powerful, stable, well-delineated factions
which consistently oppose one another and command the loyalty of
their members, as are those reported by Lewis (1958, p. 114) for
Rampur. Rather they are often vague, shifting, uncoordinated groups
of people. It therefore seems preferable in most contexts to call such
alliances in Sirkanda "cliques" in translation of the villagers' own
term, gutt. The term "faction" will be used in this presentation to
refer to groups involved in disputes. No rigid terminological distinc-
tion is advocated here. Friendship groups and other groups distinguish-
able on grounds other than disputes will be termed cliques. Actively
disputing groups will be called factions. Often these groups are the
same, in which case context will determine usage.

In Sirkanda such groups are found primarily in the Rajput caste,
it being the only group large enough to contain many such alignments,
just as it is the only caste large enough to contain several different
phratries, sibs, clans, and lineages within the village.1 However, not
all Rajputs are aligned with cliques nor are all non-Rajputs unaligned.
Some members of other castes are tied into Rajput cliques, sometimes
despite their own efforts to remain neutral. In addition, as has been
noted, there are disputes within and among joint families, between
castes, and between high- and low-caste groups. Examples of these have
been discussed in previous chapters in connection with the social units
in which they occur.

In Sirkanda there are no named groups which correspond to cliques.
The nearest thing are some kin groups which happen to be coterminus

266 THE VILLAGE COMMUNITY

with certain cliques. Most cliques, in fact, are not generally recognized
social entities. There are no special social centers or regular occasions
for intra-clique interaction. Even informants who wish to cooperate
in giving information on this subject and who understand the concept
well are hard-pressed to delineate cliques or to name clique members
in any but the most obvious alliances. Clique divisions and membership
can in many cases only be inferred by observation of interaction in
economic, religious, and social contexts. Important sources of data in
this respect are patterns of visiting and friendship, and disputes among
villagers.

As a result of inquiries along these lines it was found that in Sirkanda
cliques tend to follow caste, sib, clan, lineage, and family lines, in
order of increasing relevance. The core of a clique is usually a large
or well-to-do family, and those influential in the clique are usually
family heads. But around this core are other individuals who side
with the core group. To consider cliques independently of family and
larger social divisions would be to misunderstand them. If they deviate
from, or cut across, these boundaries, that fact in itself is important,
since it runs against traditional ideals of caste and kin-group unity.
Another important dimension was found to be that of age. People
of roughly the same age tend to participate on the same level of
interaction in inter-clique behavior, while those of other age groups
have somewhat independent alignments. Sex is also a factor in that
women have their own cliques, apparently somewhat independent of
those of their husbands and based upon ties of relationship and com-
mon village origin as well as upon jealousies and disputes among
themselves. Women, as family and clan members, often follow in the
clique alignments of their husbands. The wives of men who are at
odds generally also avoid one another unless they have close ties of
relationship or common village origin, and sometimes even these ties
are subordinated to male factionalization. In this study only male
cliques were studied sufficiently to draw many conclusions.

Friendship and Visiting Patterns

One important manifestation of clique membership is that of
patterns of friendship and visiting in the village. These are not easily
delineated. One villager commented:

There are no special friends here. People talk occasionally to everyone —
whoever they meet at the water source, at work, or around the village. They
don't talk a lot with anyone outside of their joint family. An exception is
the young men of the village, who often sit around and talk togedier. Cer-
tain people avoid one another because of disputes they have had.

THE VILLAGE COMMUNITY 267

There is not the custom, such as is found on the plains, of men
sitting in a living area reserved for them and passing time with their
friends. There is no area in Pahari houses where men can gather,
undisturbed by women and children. Informal visiting by small groups
of men (two to four) does occasionally take place on the veranda of the
house, but many men rarely participate. Principal locations for social
interaction among groups of men are the village shops and the places
of work of artisans. These represent "neutral ground," and an indi-
vidual does not compromise himself by stopping there regardless of
who else may be there or may later come. Since everyone is welcome
at these places and is, in fact, expected to stop at least briefly if he is
passing by, patterns of friendship and clique composition are not as
clearly revealed as they would be in individual houses where only
friends are welcome. Other informal groups relevant to an under-
standing of friendship and clique formation are those of men working
or traveling together.

On the basis of observations of friendly interaction in Sirkanda, six
main cliques were found among Rajputs. One of these consisted of
two lineages of the Palial sib, the second consisted of the third lineage
in that sib, and the third consisted of most of the members of the
largest lineage in the Jawari sib. The fourth and fifth were formerly
one clique consisting of another lineage in the Jawari sib, but it has
split so that now there are essentially two family isolates. The sixth is
another small Jawari clique (see fig. 5). The importance of kin ties
in these alliances is obvious. Two of the cliques follow lineage bound-
aries (one includes two and the other one lineage), and the other four
follow family boundaries (in each there are two family cliques within
a lineage). In only one case are brothers in different cliques.

In addition to these main groups, there are individuals and groups
whose interaction does not reveal any clique alignments or, more
importantly, whose interaction spans clique boundaries or makes cross-
clique groups. Four main factors seem to be basic to such individuals:
(1) age ties, (2) affinal ties, (3) ties of geographic proximity, (4) lack of
particular kinship ties which would tend to commit the individuals to
specific cliques. In the first category, age ties, is the group of young
men (20 to 35 years in age) who generally ignore the factional affiliations
of their elders and associate quite freely with one another. They even
cut across caste barriers to some extent, with a blacksmith and a Bajgi
often being included in the group. Children are another such age
group. In the second category, affinal ties, are the relationships be-
tween men who have married within the village, and their male in-
laws. There are several of these relationships, which take the form

268

THE VILLAGE COMMUNITY

of frequent informal social interaction at the home of the father- or
brother-in-law. In the third category, geographical proximity, fall the
friendship and economic and ceremonial cooperation of the occupants
to this category is the interaction of neighbors, especially the greater
of four chans in a small area % mile distant from the village. Related
frequency of interaction among people living within each of the three
main settlement areas of Sirkanda than occurs between these sections,

Jawari lineages

Palial lineages

1 1

F j G ! H

\
\
\

•

Reported
early

stage

Time

Present

household

NEUTRAL 1

Kukhalwal Rajput „ i .

Brahmin
Blacksmith ,.. .. i .

Khandial Rajput
Barber ■

One Blacksmith household
Bajgi

Fig. 5. Principal Rajput cliques through time.

Cliques are enclosed in solid-line boxes. Broken lines indicate lineage bound-
aries within cliques. Solid arrows indicate development through time. Broken
arrows indicate casual alliances. Asterisked lineages are at present small in
numbers and largely unaffiliated.

other things being equal. In the last category, lack of kinship ties
which commit individuals to cliques, fall the members of three rela-
tively small Jawari Rajput lineages, the members of one small sib
in the village (Kukhalwal), and the members of other castes. The
remaining Rajput sib, Khandial, consists of one family which is com-
mitted by marriage to the first Palial clique listed above.

Clique alignments and deviations from them are evident not only
in visiting patterns but in cooperation in ritual and economic under-

THE VILLAGE COMMUNITY 269

takings as well. No family, lineage, sib, or clan is large enough to
perform an important marriage without outside assistance, and this
is usually not available from members of opposing cliques. Drinking
parties are another activity in which cliques and clique deviations are
readily apparent.

Disputes and Dispute Resolution

Dispute reveal clique divisions sharply. They are, in fact, usually
the bases for the avoidance between individuals and groups which
becomes stereotyped into clique behavior. A study of disputes, there-
fore, leads to an understanding not only of what the clique structure
is, but of how it got that way, that is, how cliques are formed and how
they change through time. In addition, such a study reveals important
information about the means and dimensions of social control and
conflict resolution in the community.

In discussing disputes a major difficulty is that of deciding relative
significance. Disputes of all intensities occur, from the minor irritation
that is quickly forgotten to the major power struggle that is remem-
bered for generations. Some disputes bring about formation of cliques
or factions; most are symptomatic of already existing factionalization,
and some are unrelated to clique affiliation. Here disputes that have
aroused strong feelings among the participants and that are apparently
associated in some way with clique alignments will be considered. They
include all those upon which sufficient relevant information was avail-
able for analysis — the recent and the well-remembered. A few occurred
during the writer's residence in Sirkanda; most occurred within the
previous 10 years; one took place 40 years ago. In all cases, however,
relevant and apparently reliable data were obtained from informants
who were present at the time of the dispute.

That this is a biased sample is indicated by the fact that of the
32 in- village disputes analyzed, 19 resulted in estrangement of the
parties to the dispute although no apparent estrangement had existed
previously, and only 13 perpetuated or were symptomatic of old
animosities, so far as could be determined. Obviously the latter would
be far more frequent than the former in a random sample, but they
are less well remembered.

Once a major factional split has occurred, further disputes usually
follow at frequent intervals which sustain and often deepen the rift.
Disputes without significance either as causes or symptoms of factional-
ization do not appear here, not because they do not occur, but because
they are frequent, usually minor, and not germane to this discussion.
In this category would fall most such temporary and recurrent disputes

270 THE VILLAGE COMMUNITY

as those which occur frequently among family members, between in-
dividual women in the village, and between village occupational
specialists and the clients they serve.

Fifty-one disputes are analyzed here. Of these, 44 involved Sirkanda
people either as both parties in the dispute (32 cases) or as one party,
the other being from a neighboring village (12 cases). The seven re-
maining cases involved primarily people of neighboring villages, but
Sirkanda people were involved as relatives, council members, or
witnesses.

The disputes have been analyzed in terms of the nature of the
grievance and the action taken as a result. Grievances were categorized
as follows: (1) sex, arising from illicit sex relations or alienation of
affection; (2) property, arising from conflicting claims to land, houses,
or income; (3) status, arising from nonadherence to caste rules, or
inappropriate caste or intercaste behavior; (4) legal testimony, arising
from a person's refusal to testify for, or his testimony against, another
in a court case in which he was to have appeared as a witness and
in which some connivance had been planned; (5) beating or physical
assault on another (in all cases this occurred as a result of grievances
falling under one of the other categories listed here); (6) wife-stealing
(transporting a woman out of the village for profit rather than for
personal sexual reasons); (7) malpractice of a curing rite.

Actions taken to resolve or win a dispute have been categorized
as follows: (1) threats, avoidance, insults, family division, and so on,
which involved none of the other types of action listed below but which
sometimes eventuated in a modus vivendi or solution acceptable to
both; (2) beating or physical assault on another; (3) revenge by com-
mitting a like offense, the only such case in this sample being one in
which a man whose wife was regularly having sexual relations with
another man helped steal that man's wife; (4) council (referral to an
ad hoc council of villagers); (5) intervention by gods; (6) intervention
by officials (referral to police, courts, or government officials outside
the village); (7) "frame-up" (revenge by bringing a false legal charge
against an opponent after arranging matters so that he will be found
guilty). An example of the last befell a man whose prosperity was
envied by his caste-fellows. They had once tried unsuccessfully to get
some of his land by bribing the government records officer to testify
that he had acquired the land illegally. Failing in this, they obtained
revenge by hiding liquor in his house and calling the police to say
that he was dealing in illicit liquor.

The frequencies of these types of grievances and the actions taken
are presented in table 6. Not shown in the figures in table 6 is the

THE VILLAGE COMMUNITY 2 J 1

fact that beatings, and to a lesser extent all types of action taken in
disputes, were preceded and accompanied by threats, avoidance, in-
sults, and so on, and that a fight was always a possibility. As one
villager remarked when a crippled man verbally abused another man,
"A person should not use abuse unless he is strong enough to fight."
Also not shown here is the fact that the council case for woman-
stealing was subsequently prosecuted by the government in the courts,
but not as a result of action by villagers.

TABLE 6
Sirkanda Disputes by Type and Resultant Action

Action Taken
Threats, Officials

avoid- Like and Frame-

ance Beating offense Council Gods courts up Totals
Grievance

Sex 6 5 13 2 2 1 20(39%)
Property 8 1 — — — 4 1 14(27%)
Status 3 — — 2 — — 1 6(12%)
Testimony 3 — — — — _ _ 3 (7%)
Beating — ____ 4_4 \%%)
Woman-
stealing 2 — — 1 — — — 3 (6%)
Malpractice — — — 1 — — — 1(2%)
Totals 22 6~~ ~~T 7 2 ~7u 3 5T~

(43%) (12%) (2%) (14%) (4%) (20%) (6%) (100%)

These figures and the relationships among them show a number
of things about the nature of Sirkanda disputes. Sex and property are
the main causes for disputes which get into the public eye and are
felt to be important by all concerned. Together they account for two-
thirds of all disputes recorded. Sex disputes are consistently resolved
in the local context. The two cases that went to court involved obvious
breaches of law. Property disputes, on the other hand, go much more
frequently to official, outside agencies for resolution. Beatings are
usually cause for a lawsuit. The reasons here are very obvious: you
do not go to an official, the police, or a court of law unless you have
good reason to think you will win your case. These people have found
that beating can be easily proved and prosecution is sure, and that
in property disputes the law courts often decide cases on the basis
of evidence which is ignored in the village context. Consequently, it
is worthwhile to take such cases to official agencies. Sex, status, mal-
practice in curing, failures to keep a promise to testify in court, and

272 THE VILLAGE COMMUNITY

woman-stealing are areas in which court decisions are either unavail-
able or unlikely to be useful. Woman-stealing, which might seem to
be amenable to court decision, is rarely referred to the courts because
the woman has invariably cooperated so that evidence against the
"abductor" is usually lacking. Those who have engaged in woman-
stealing as a regular business rarely participate in the abduction of
local women, and those who do it only once can always plead innocence
and support their plea by evidence of good character.

Further insight into the nature of appeal to courts is found in
the three "frame-ups." In each case grievances which could not be
successfully brought to a court of law were resolved by manufacturing
a case which could be decided in the courts. The three actual grievances
were an illicit sexual affair, jealousy over property, and anger caused
by low-caste insolence. The grievances manufactured and brought to
court were, respectively, assault and rape, liquor-selling, and theft.
Incidentally, this was the only case of alleged rape known in the area,
and one of the very few allegations of theft.

One other fact worth noting is that the village council is not used
for property disputes. Villagers feel that justice cannot be obtained
because the council is made up of people with heavy vested interests
in local property, who would decide solely in terms of self-interest or
bribes.

Disputes are not submitted for arbitration by courts (in cases of
alleged breaches of legal statutes) or by the village council (in cases
of alleged breaches of custom) unless those who present them are con-
fident that they will win. Often those with grievances are challenged
to take them to court or a council. As one Rajput informant said,
"If there is a legitimate complaint the case will go to a panchayat or
to a court. If not, it is just bickering."

Gods intervened in only two disputes, both hinging on sex. Sex
disputes are the most ambiguous as well as the most frequent kind
of grievance. Gods can give satisfaction to one party when no one else
can. They intervene much more frequently than this analysis in-
dicates, because they often enter a sex dispute between two women
or between family members which never reaches the public eye or
never enters into factional alliances. Likewise gods frequently inter-
vene on behalf of brides from other villages in matters unrelated to
sex.

The functions of various kinds of groups in the village can be made
clearer by investigating the nature of social groups which engage in
various types of disputes and employ various types of action to resolve
them.

THE VILLAGE COMMUNITY 273

Forty of the disputes analyzed here took place between people of
the same caste. Thirty-two of these were Rajput disputes, and eight
were blacksmith or Bajgi disputes. Eleven were intercaste disputes,
three between Rajputs and Brahmins, two between blacksmiths and
Bajgis, and six between particular high castes and low castes. These
figures are largely a reflection of population and social structure in
Sirkanda. There is only one household of Brahmins. Disputes, as
defined here, do not occur within households except when they result
in a lasting division of the household unit. Bajgis and Lohars are few
in numbers, and each of these castes in Sirkanda is composed of a
closely related group of people in a single lineage. The eight intra-
caste disputes within these two groups are a reflection of the fact
that low castes are more likely to divide the household and compete
for livelihood than are the other castes. Rajputs are numerous and com-
prise four different sibs; consequently, most disputes in the village
involve them. Most disputes which villagers would describe as inter-
sib disputes involve local sib members, and usually wives side with
their husbands. Therefore such disputes will here be termed inter-
clan disputes.

In the 40 intracaste disputes, 22 were between clans, five were within
clans but between lineages, and 15 were within lineages. One dispute
involved both inter-clan and interlineage disputants, and another in-
volved both inter-clan and intralineage disputants; hence the disparity
in totals.

In table 7, grievances and action taken are tabulated according to
their occurrence in intercaste disputes, disputes which are within
the caste but between clans (inter-clan), disputes which are within
the clan but between lineages (interlineage), and those which are
within the lineage (intralineage). This table shows that inter-clan dis-
putes are predominately concerned with alleged sex offenses, while
intralineage disputes are most often centered on property. This can be
easily explained in retrospect. Only within the lineage is property
derived from a common estate, and only there is property held in
common by several members of the group. That is where one would
expect to find disagreements over property — in its management and
division. Intense sexual activity occurs across clan lines because clans
are essentially sibs, the exogamous units. Since all legitimate mates
are found across sib lines, the potential for this type of dispute among
these groups is high. Within the clan, sib, and lineage, wives are quite
freely available to their husbands' brothers, real and classificatory,
while other women associated with the lineage are exempt by incest
rules. Therefore, disputes of this kind are less likely within these

274 THE VILLAGE COMMUNITY

largely exogamous units. The sex disputes which did occur within
lineages nearly always involved one man's taking over another's wife
or carrying on a conspicuous affair with her.

TABLE 7
Incidence of Disputes by Social Groups Involved a

Intercaste Inter-clan Interlineage Intralineage

Grievance

Sex

Property
Status

2
3

6
3

Testimony

Beating

Woman-stealing

2
1
1

2*
2

1
1

Malpractice

—

Action Taken

Threats

Beating
Like offense

—

1
1

Council

—

Gods

—

Officials

Frame-up

—

a The two totals marked with an asterisk which appear on the same horizontal
row include one case in common; for example, in the property disputes, there
was one with inter-clan and interlineage disputants, and so it was included in both
columns although it was a single dispute. This accounts for the fact that the total
number of cases listed here is 53, whereas only 51 cases were analyzed.

Disputes in general are more common among status equals than
across status boundaries. Large differences in status are associated
with differences in power, so that disputes across status boundaries
are infrequently essayed and are likely to be nipped in the bud if
begun. Of the six recorded high-caste-low-caste disputes, one involved
woman-stealing and included a Brahmin and Rajput as well as a
Bajgi among the culprits, so that it was hardly a case of high caste
versus low caste. Two of the others involved low-caste insolence, one
resulting in a "frame-up" and the other, which involved behavior of
children, resulting in threats and verbal abuse. Another was a low-
caste grievance against a Brahmin who failed to pay for a cow taken
on agreement to purchase. The final two stemmed from a sexual liaison
between a low-caste man and an acquiescent Rajput woman. The man
was beaten and then took his assailants to court.

The other five intercaste disputes were between castes adjacent in

THE VILLAGE COMMUNITY 275

the status hierarchy, and did not differ in type or distribution from
what would be expected within castes.

Status disputes occurred either between castes or within lineages.
The former were matters of low-caste insolence prosecuted by the high
castes. The intralineage status disputes were cases of individuals failing
to conform to requirements of caste or family honor. Lineage mem-
bers, as those most immediately involved, were the group which sought
to rectify matters.

Figures on action taken show that lineage solidarity is effective in
keeping most intralineage disputes on a verbal level, with a minimum
of public display and outside intervention. Beatings, council action,
and court cases occurred frequently in inter-clan and intercaste cases.
Apparently group controls across these lines are relatively weak.

Physical assault as a means of resolving a difference occurred most
often among status equals unrelated to one another, that is, at the
inter-sib or inter-clan level. Relatives are reluctant to fight, and low-
caste people are reluctant to get themselves into a position where
powerful high castes will have an excuse to attack them. The only
intercaste beating was the one described above, in which a low-caste
sex offender was beaten.

The distribution of cases in which there was resort to outside officials
reflects closely the distribution of disputes over property, since it is
in property disputes that official agencies are most useful.

When a council is appealed to voluntarily it is almost always as
a result of an intercaste dispute. Within the caste, less conspicuous
means of arbitration are preferred. Three of the four cases involving
single castes (that is, inter-clan) which went to councils were intra-
low-caste disputes into which councils intervened without the request
of those involved. Low castes try to keep their disputes out of councils,
where they say low castes always lose and only the council wins.

"Frame-ups" occurred only in intercaste and inter-clan disputes,
probably because such plots are a drastic action, usually resulting in
imprisonment, which relatives would rarely inflict on one another.

Clique and Faction Formation

The clique or faction structure revealed by disputes corresponds
closely to that shown in friendship and visiting patterns, and explains
the formation of current alignments.

The Jawari sib of Rajputs is characterized by several factional splits
and antagonisms toward other groups as a result of a number of
important disputes. It was allegedly once undivided by major disputes.
Then two factions, A and B, split off a few generations ago, as the

276 THE VILLAGE COMMUNITY

result of a serious sexual dispute (see fig. 5). Later each of these
factions split into two groups along family lines (Ai, A2; Bi, B2) as
a result of property and status disputes. One of these divisions also
involved a half-brother relationship, a relationship which quite con-
sistently results in social distance between the parties. Most recently,
one family in Ai was alienated from that clique as a result of a status
dispute, and so it joined A2, its current alignment. Similar histories
could probably be traced for other cliques in the village.

In general there is considerable disputing between the two large
sibs in Sirkanda, and they have each acted as factions in and of
themselves. Within these groups lineages and sometimes households
have formed cliques. The importance of kin ties can be readily seen
in the diagram, although they are not equally important in all cliques.
Other groups which might prefer to remain neutral tend to become
aligned with cliques through friendship, work relationships, affinal
relations, proximity, and so on.

In correlating friendship groups with disputant groups it becomes
evident that not all friendship groups are antagonistic to one another.
Between the Palial cliques there is little or no animosity (H, FG).
The same holds for the relationship of B2 to H and FG. These are,
therefore, cliques that are on good terms with one another. On the
other hand, Ai, A2, Bi, and B2 are antagonistic to one another, and
Ai, A2, and Bi are antagonistic to FG and, to a lesser extent, to H.
The household comprising clique Bi is generally disliked by everyone
in the village except one Bajgi household, with which it is in a patron-
like relationship, and lineage C, which has sided with it occasionally.
Clique FG is the most generally liked group in the village. It has had
few disputes with others and comprises a loose-knit, open group.

The attitudes between cliques are usually related directly to the
types and intensity of disputes they have had with one another. The
Jawari sib is permeated with jealousy, animosity, and disputes, while
the Palial sib is at present relatively congenial. As has been mentioned
before, all but the most serious disputes tend to be ignored after one
or two generations unless they are reinforced by new ones. Therefore,
young men of Ai and A2 associate freely with those of H and FG, and
even the youngest man of the isolated and generally disliked clique,
Bi, has some informal social contact with these groups.

The effects of disputes and factionalization on the community are
important and far-reaching. They sometimes seriously affect village
cohesion just as they do lineage and clan unity. The complexity of
disputes, the means of dealing with them, and their effects may be
illustrated by picking two examples and describing them briefly.

THE VILLAGE COMMUNITY 277

Case I

The most serious dispute in recent history was over land. About the time
of World War I the army purchased some Sirkanda lands a little over a mile
from the village, to be used as a small summer encampment for Ghurka
troops. Two stone houses and some other improvements were made there
by the army. For nearly twenty years the camp was often occupied during
the summer months. The army was rigidly segregated from the villagers by
army orders, with one exception — one village family (A in the diagram of
cliques) sold milk to the army and became friendly with its personnel. When
the army left in 1932, the milk-selling family bought the desirable army land,
complete with its buildings, for a ridiculously small sum. Other villagers were
incensed when they found out because many of them had formerly owned
shares in the land and they had been given no opportunity to buy their
shares back. According to them, the sale should have been publicly announced
and all should have had a chance to buy some of the land.

A nine-year legal suit resulted which was eventually won by the milk-selling
family (A). Although the law seemed to be on the side of the other villagers,
the village accountant, whose records were crucial to the case, was a member
of the winning family. Also the army officers had liked that family, had
cooperated with them in the sale, and had testified in their behalf in court.
The appellants were of the opposite clan (clique FG) but were actively and
openly supported by members of a rival faction (Bi) in the milk-sellers' clan
who hoped to get land, although they had not been among the original
owners of the land. It was essentially the village versus the milk-sellers
(cliques Bi, B2, H, FG vs. Ai, A2).

During and after the trial, relations between the disputants and their
respective supporters deteriorated. An outside merchant was brought in when
the suit began, so that villagers would not have to buy from the only village-
run shop, owned and operated by the milk-sellers' family. These two shops
have remained in the village, in full view of one another, jealous of one
another's patronage, and a constant reminder of the dispute. A member of
the milk-selling family was replaced as village tax collector under pressure
occasioned by animosity resulting from this case. Villagers blame subsequent
lack of village unity largely on the case. Attempts to improve the water supply
have failed, and money granted by the government to repair the school and
to build a community center has gone unused. Villagers say that animosities
kindled and fanned by the old dispute prevent agreement and cooperation
on the new projects and lead individuals to purposely subvert plans that are
made. Even the lineage that won the case has been split because one member
family (A2) could not pay its full share of the heavy legal expenses and so
was given none of the lands it had been promised for its support. It there-
fore broke completely away from the rest of the lineage.

Case II

A Rajput woman, who lived with her husband in a chan, fell in love with
a relative of her husband who lived in a neighboring chan. When the hus-

2^8 THE VILLAGE COMMUNITY

band came to know he moved away from the chan to the village, whereupon
his wife went to her parents, refusing to return to her husband. After two
years the wife wanted to rejoin her husband, but he then refused to take her.
Her family insisted, but to no avail. A council meeting was suggested, but
the husband's family refused in highly insulting terms. At the urging of
another relative, Sohan, they wrote the woman's family a letter saying that
the marriage was void and they would have nothing to do with her. This
angered the girl's family because, since no money had been demanded as
compensation, the divorce implied that she was immoral and worthless. They
especially resented this because no public charges had been placed against
the girl. Moreover, the unstated charges involved the husband's own relative;
therefore, they felt, the girl was not at fault. In their anger the woman's
relatives caught and beat her husband and his brother. Thereupon Sohan
urged the beaten men to go to court about the beating. This they did upon
the assurance that Sohan would stand as witness in their behalf. However,
when the case came up, Sohan did not appear as promised. The husband's
family spent a good deal of money on the case, which lasted a year and was
then dropped. Meanwhile, Sohan had shifted his loyalties to the other side
and helped them with advice and encouragement. When the case was dropped,
the ex-wife was turned over as wife to her former lover at the chan, who had
been paying part of the legal expenses on her behalf.

As a result, the disputants (and Sohan, on the side of the wife) refused
to ever again attend social or ritual occasions where they would meet and
especially where eating was to take place. This avoidance was practiced for
over five years until a prominent and well-liked village woman died. Her
husband had been in one of the two large Rajput sibs, and she was a member
of the other, a daughter of the village as well as a village wife and mother.
The entire village was united in love and respect for her, and yet the two
disputing factions refused to attend the death ceremonies together. A council
meeting was held to come to a new agreement, as it was felt by those not
involved in the dispute that the estrangement should not continue with such
disruptive effect. A compromise was reached which is still followed today. The
two parties agreed to attend all village functions but not to eat in the kitchen
area on such occasions. They eat outside and leave as soon as they have
eaten. They never eat in the kitchen area of the same house on the same day.

Villagers generally admit that the husband was in the right in this case,
but because of the arrogant rejection of council arbitration and the generally
haughty attitude of the husband and his family, village sympathies were with
the wife — to teach the husband a lesson. General dislike of the husband's
family also entered in. In this case the husband's family comprised clique Bi;
Sohan's family was clique Ai; the wife's family was in clique FG.

A villager commented, in reference to this case, "Village disputes are of
this silly nature."

It is evident from these somewhat simplified accounts that disputes
are complex, that alignments are not firm nor entirely consistent and

THE VILLAGE COMMUNITY 279

predictable, and that ramifications may be far-reaching and disruptive.
The motivations determining sympathies and alignments among non-
disputants are usually based in previous grievances and are often
extremely complex. Nurturing and compounding these animosities
by planning revenge is a major activity among some villagers. On the
other hand, in daily activities the animosities are kept under control
so that they come into the open relatively infrequently in view of the
amount of interpersonal contact which is inevitable in the village.

The village settlement pattern is suggestive in this regard. While
it conforms in general to sib and lineage distribution, it conforms
even more closely to clique alignments (see map 3, p. 25). Thus,
clique Bi, the most isolated and antagonistic clique, is physically
isolated in a single house on the western edge of the village; clique
Ai is isolated in the house nearest the Devi shrine; clique A2 is in
two adjacent houses on the edge of the eastern settlement area; clique
H is in the line of houses nearest the school. Clique FG, the most open
of all, is in the center of the village. In general, the most congenial
households are those which occupy the central portion of the village,
while those harboring animosities and rivalries and those which are
simply asocial are found on the peripheries, where they have built
houses in recent years; for example, all houses in the southwestern
settlement area of the village have been built in the past ten years.
A possible earlier pattern is suggested by older houses in the village.
Among them there is a general separation of the Palial sib (lineages
H, FG) and Jawari lineages B, D, E (all located in the central settle-
ment area, although clustered by sib) on the one hand, and Jawari
lineages A and C and the Kukhalwal sib (located in the eastern settle-
ment area) on the other.

Cliques have no formal organization. The elder men of the house-
holds involved tend to be the policymakers for the cliques and to
enforce, when possible, loyalty and conformity within the group. The
clique is an extremely loose relationship when more than one house-
hold is involved.

Clique alignments serve as barriers to communication and common
action among villagers, and therefore they are divisive factors. On the
other hand, they facilitate interaction among members of the same
clique, thus forming a close tie that is not always coextensive with
ties of kinship and caste. Insofar as they do that they help bind the
community together across natural lines of cleavage. Moreover, it
should be emphasized that cliques operate primarily in the community
context. Outside the community they have little relevance. None of
the cliques in Sirkanda has close ties in other villages which divide

28o THE VILLAGE COMMUNITY

them against other groups in the community in intervillage disputes.
In relations with other villages, caste is often a divisive force but
cliques are not. An outside threat brings unity to the village unless
there are conflicting caste or kin-group loyalties. Also, clique differ-
ences are largely disregarded in dealings with the urban and official
worlds. In town and in relations with officials, Sirkanda villagers
usually support one another regardless of clique differences in the
village.

An important source of cross-clique communication which leads to
a weakening of clique loyalties and a lessening of the divisiveness of
cliques is that of in-village marriage. The importance of in-law rela-
tionships among village men in cutting across clique lines has already
been noted. Perhaps even more important is the role of the local
girl who marries in the village and is at once a wife in one clique
(and sib) and a daughter or sister in another. Seventeen per cent of all
marriages (20 per cent of Rajput marriages, the relevant ones in this
context) occur within the village. Such marriages always occur across
sib lines, all sibs are connected by such affinal ties, and nearly half
of all intracaste disputes occur across sib lines. Therefore, in-village
marriage is an important factor in village cohesion which is absent
in exogamous villages.

Councils (Panchayats)

The ad hoc councils (panchayats) which were summoned in eight
of the disputes discussed above, to arbitrate in cases concerning sex,
property, and status, are the locus of formalized traditional community
government insofar as this exists at all (cf. Newell, 1954). These coun-
cils act primarily in cases involving traditional rules of behavior.
Atkinson notes that in Garhwal:

Panchayats for the settlement of social disputes have long been known
... in the hills. . . . They are now usually assembled for the settling of
cases of abduction or seduction of women, or offenses against caste. Witnesses
are heard on each side, and the award given is usually submitted to. A fine
is often imposed and a feast given to the assembled brethren at the expense
of die offending party. (Atkinson, 1886, pp. 265 f.)

Panchayats are ad hoc rather than permanently constituted bodies.
Their membership is recruited by invitations sent out by elders of the
households involved to other interested households, who then send
representatives. The membership varies with circumstances, but it is
traditionally confined to the high castes, and women are never council

THE VILLAGE COMMUNITY 201

members. In intervillage disputes councils are intervillage in com-
position, generally with representatives from a number of neighboring
villages. In an intravillage dispute they may include representatives
of neighboring villages if the matter is considered to have implications
which extend beyond the village, such as a breach of proper intercaste
relations. For purely local matters panchayats are made up of local
men, many or few depending upon interest. A case of woman-stealing
will usually involve a large meeting; a mutually desired divorce will
involve a council of only three persons. Since councils consist of high-
caste people, Rajputs dominate them in Sirkanda. However, except
in purely internal matters within a high caste, both Brahmins and
Rajputs are represented in most council meetings. Such councils decide
those low-caste disputes which reach a council as well as high-caste
disputes and disputes between high and low castes. Within the council,
discussion and decision making are quite democratic, with the expected
deference to age and generation. Wealth, ability to speak and reason
convincingly, and an even temper are other qualities which lend
influence to a man in these situations. Clique affiliation becomes im-
portant in that those aligned with small, unpopular, hostile groups are
able to wield less influence than those affiliated with large, popular
groups or those who are known to be objective or little committed to
any particular group.

Councils or panchayats serve mainly to coordinate and express
public opinion. Usually the members of the council are well aware in
advance of the facts and opinions relevant to any dispute put before
them. Their decision is often a means of making official that upon
which there is already general agreement. Most often they attempt
to come to a mutually agreeable compromise. In more serious cases,
or those in which the parties are unable or unwilling to compromise,
the council can impose sanctions under threat of physical punishment,
social ostracism, or legal action. It may order a man to leave the
village, to make payment or some other concession to another party,
to make payment to the council itself, or to refrain from certain types
of social interaction. Public humiliation or a public beating may also
be inflicted, usually upon flagrant sexual offenders, primarily wife-
stealers. In two such disputes among those analyzed here, the offender
was led through the village wearing a garland of worn-out shoes
around his neck — an extremely humiliating punishment because of
the defiling nature of shoes as being made of cow leather. In one case
the offender was then beaten with shoes.

There is little confidence in the objectivity of councils. They are
thought (quite correctly) to be heavily influenced by caste, clique, and

282 THE VILLAGE COMMUNITY

kin-group loyalties and to be easily swayed by money and favors. As
a Rajput said, "Whoever treats the council to the best feast wins the
case." Low-caste people, in particular, distrust councils, because in
disputes among themselves as well as those between themselves and
high castes they are often victimized by the councils. Stories of such
cases are numerous. An example follows:

A blacksmith girl of a neighboring village was married to an old blacksmith
man. She was unhappy with this arrangement and soon returned to her
parents. From there she later eloped with a blacksmith boy. None of the
parties involved were concerned enough to try to alter the situation. How-
ever, Rajputs of the village found out about it. They knew (according to
blacksmiths) that the boy involved had Rs. 600 in his possession. They there-
fore called a council meeting of Rajputs and went to the girl's father and
extracted from him, by threats and promises, a complaint in the case. They
then caught the boy, brought him before the council, and fined him Rs. 600.
The terrified boy paid this sum willingly, glad to escape without a beating.
Two hundred rupees of the fine went to the girl's father, and the remainder
was "swallowed up" by the council.

Even high-caste people are occasionally victimized in this way. A
council meeting is always expensive for those whose dispute is to be
decided, as they are obligated to feed and house the council members
and must do it well or risk losing the case.

For these reasons cases are rarely referred to councils by the dis-
putants. People prefer to carry on private feuds, as witness the fact
that well over half the cases discussed here were kept on this level
although they were of quite major importance. Councils usually inter-
vene upon the insistence of people not parties to the dispute and
usually when the dispute is such that it affects others not immediately
involved in it. In extreme cases, and especially in the case of a beating,
woman-selling, or illegal alienation of land, when it is felt that a law
has been broken, people take their grievance to the courts — again
an expensive undertaking where money counts above all else, but
where the decisions are firm. Informants state that in earlier days the
main resort in all cases was to gods, who would punish a person's
rival or bring a favorable conclusion to a dispute in response to worship
and sacrifice. As education and money have increased, courts and coun-
cils have been increasingly used, often in addition to appeal to the gods.

In these hills there are still relatively few civil and criminal cases,
especially the latter. Litigation in the courts is a less important activity
than it is in many plains areas, while crimes of theft and violence (other
than occasional retributional beatings) are very rare. Murder occurs
rarely — never in the history of Sirkanda, and four or five times in the

THE VILLAGE COMMUNITY 283

history of Bhatbair. The cases in Bhatbair were apparently carefully
planned retributional killings. In each case the victim was widely
known as a scoundrel. Those who had grievances against him caught
him, tied him up, told him their intent and the reasons for it, and
then methodically beat him to death. Less severe beatings are also
sometimes administered in this way.

Suicide is virtually unknown in this area. In the past a woman
occasionally immolated herself on her husband's funeral pyre. Only
one such case is known for Sirkanda. Rape has never occurred in
Sirkanda; villagers have difficulty imagining it in the Pahari context.
Woman-selling, illegal sale of liquor, and illegal use of government
forest land are the most common offenses, and they are not offenses in
Pahari customary law.

Leadership and Influence

Outside the family, in which the eldest active male is the recog-
nized authority, there is no generally recognized leadership in the
village beyond that of elder, landowning males. Individuals hold power
in the village primarily to the extent that they are influential as a result
of personal qualities in important cliques, or are members (especially
household heads) of large and wealthy families. Wealth, verbal ability,
an even temper, modesty, and simple living are virtues which win a
man respect. But no man is in a position to tell others outside his
family what to do. Villagers found it hard to conceive of an influential
leader, and seemed opposed to the notion. They recognized a former
condition when a "sayana" held rights to collect taxes and as superior
"landlord" had control over those on his lands. But since all high
castes now own land, they feel that none should have power over others.
Their attitude parallels that reported by Baden-Powell (1892, I, p. 153),
who noted the absence of a headman in "landlord" villages where
"the proprietary families were too jealous of the equal rights to allow
of any degree of authority residing in one head." Sirkanda villagers
do not pay much attention to the official headman (council president),
nor is the position sought after. They resent the powers officially in-
vested in this office. They neither want them for themselves, nor will
they grant them to others. One man summarized:

Here all [high-caste] men are the same. They do what they want without
regard for others. They decide on the basis of what is best for them. If
several people agree, that is good, but it will not lead others to agree unless
they come independently to the same conclusion. Every man does what is
best for his household without regard for what others think or want to do.

284 THE VILLAGE COMMUNITY

This does not mean that power and influence are evenly distributed,
but it does mean that they are ideally so and that in reality they will
approximate this ideal state.

At one time Sirkanda lands were included under those granted by
the Raja to a Brahmin of Kanda, as discussed previously. This man,
as "sayana," collected taxes and theoretically had some control over
those who occupied his lands. The term "sayana" still carries strong
connotations of status and power among people of the area. The first
remembered formalized village government in Sirkanda was that of
the hereditary mukhia or keeper of the peace, recognized by the British.
He decided petty local disputes and transgressions brought before him
and referred more important cases to a council of five mukhias in the
vicinity. It was his responsibility to serve as the administrative link
between the British government and the villagers. Above him were
the courts and the district government.

The village lambarddr or appointed tax collector was also a man
of influence under the British and was considered to be a village leader
much like the mukhia. Apparently the lambardar is a more recent
innovation in this village.

Today there is only one paid government employee who is normally
a villager. He is the village chaukldar or watchman. His duties are to
report births and deaths in the village and to act as a government mes-
senger when needed. In this area it is a position of little importance
and no prestige. The man who fills this position for Sirkanda and
several surrounding villages is a resident of Chamba, a nearby village.
He is rarely seen in Sirkanda and is dependent upon the Sirkanda
village president for the information he is required to supply to the
government.

Formal Village Government Today

Today formal village government is in the hands of the village
council as specified in the U. P. Panchayat Raj Act of 1947 and as
amended thereafter (Dwivedi, 1957). According to this act, a village
or group of villages is under the jurisdiction of a "village government"
or assembly (Gdbn Sab ha), consisting of all sane adults (that is,
people 21 years of age or over) within the area for which it is estab-
lished. Presiding over this body is a president (pradhan) elected from
among the membership for a three-year term of office, and a vice-
president (up-pradhan) elected for a one-year period. This body is
supposed to hold two annual meetings as well as others upon request
of the membership.

From among its members the village assembly elects a "village

THE VILLAGE COMMUNITY 285

council" (Gdon Panchdyat) to function as an executive committee.
This, too, is presided over by the village president and vice-president.
Membership in this body is for the same period as the president's term
in office (now three years). The powers, duties, and functions of village
councils are wide, covering all aspects of village welfare, upkeep, and
control and serving as the local agency for higher governmental author-
ity as sanctioned by that authority. Moreover, it is empowered to im-
pose taxes and hold property. It is required to report to the village
assembly at that body's semiannual meeting.

In Sirkanda these provisions are unknown to many villagers and are
imperfectly understood by those who know of them. The village as-
sembly is thought by some to consist only of men, and in practice this
is apparently the case. It did not meet in 1957-1958, nor had it met
since the election of the village council two years earlier. A meeting
was called by the teacher and village level worker to celebrate Republic
Day in 1958, but despite advance publicity, not a single individual
appeared at the appointed hour. This body is significant in village
government in Sirkanda only in that it is the electoral body for the
council. The area which is included with Sirkanda in a common
village government includes four other, smaller villages, from 1% to
8 trail miles from Sirkanda. They have little in common beyond the
fact that they are the only villages in a large, relatively unpopulated
geographic area which the government has chosen, for administrative
purposes, to consider a unity. Two of the villages are trading centers
on the Mussoorie-Tehri trail; one (Suakholi) is very active; the other
is virtually deserted. The other two villages are similar to Sirkanda.
The local governmental seat for this area is Sirkanda.

The village council is composed of 16 members at present. Eleven
of these are from Sirkanda (including one low-caste man in a seat
reserved for "scheduled castes"); two are from the largest of the other
villages, a trading center; and one is from each of the other three
villages. The president and vice-president are Sirkanda men, though
they need not be. All are Rajputs except for the occupant of the
reserved scheduled-caste seat, a blacksmith, and a Vaishya merchant
from the trading center included in this area.

This system of village government was established in Sirkanda in
1951 when Sirkanda was joined, for administrative purposes, with
five other villages in the immediate vicinity, all with common interests
and problems. At that time a widely respected Sirkanda Rajput was
elected president (for the then-prescribed five-year term) without op-
position. At the end of his term, in 1956, the laws had been changed
so that the term was reduced to three years and the area had been
redivided administratively in its present unrealistic manner — ap-

286 THE VILLAGE COMMUNITY

parently partly to reduce the population included under one village
government. At that time the former president declined to run again
and gave his approval to another man in the village — the wealthiest
Sirkanda villager, but a man well known for his modesty, simplicity,
and hard work, and one unaligned with any antagonistic clique (he
is in clique H in figure 5; his predecessor was in clique FG). Against
him ran the most generally disliked villager (head of clique Bi), a man
who apparently wanted the power and did not realize his own un-
popularity. Needless to say, the latter lost the election, receiving only
a few votes from some of his sib-fellows and apparently none from
other sibs or castes or the other three major cliques in his own sib.

It was difficult to get enough candidates for the village council;
there was, therefore, no competition. Some villagers on the village
council are under the impression that the president of the village
helps select the candidates for the council and that he must approve
them before they take office. They evidently acquired this idea be-
cause, in the elections held thus far, the village president has had to
use his influence to get people to accept nomination for that office. A
factor which has discouraged candidates is that each had to pay a fee of
Rs. 6 before assuming office. This appears to be the fee for membership
in the Congress party, which many villagers evidently assumed to be
a prerequisite to election. On this point there is confusion, as some
villagers claim to have had to pay Rs. 12, Rs. 6 for nomination and
6 to become Congress members. In any case, there are now 17 villagers
who joined the Congress party at the time of elections, including those
who became council members and a few others. They paid Rs. 6 each
to become members, while the council president reportedly had to
pay Rs. 12. The villagers joined the Congress party, according to one
of their number, under the impression that they would thereupon
become members of a "ruling group" — the power elite of Sirkanda.
They apparently gained this idea from government officials who helped
set up the village government and who recruited members. One man re-
marked ruefully, "If I had spent that money on ghee, I could have
been big and strong by now; this way it was wasted."

The defeated candidate for president of the council refused to join
the Congress party, and so the election was officially a Congress party
victory — grass-roots support for the party. In fact, no villager under-
stands the Congress party or even the significance of party affiliation.
Moreover, to a man, the village is opposed to the present (Congress)
government of India and to virtually all its policies, as will be made
clear below. Worst of all, from the party's point of view, the villagers
are opposed to Independence.

THE VILLAGE COMMUNITY 287

Village Government in Action

The village council of Sirkanda met once during 1957-1958, in
a meeting called by the village level worker, a government employee
stationed in the village. This meeting was announced well in advance
and was scheduled to coincide with the annual visit of the government
tax collector and the even rarer presence of the village accountant,
with both of whom all landholders had to deal. The village level
worker was motivated to call the meeting primarily because it was
time for him to turn in certain reports, but it was announced as a
meeting to discuss village improvements. No agenda was circulated in
advance. The meeting was held in the open near the two village shops
and beside the trail to the village water supply — a pleasant, convenient,
and conspicuous location, with the shops serving as comfortable re-
treats for bored participants and observers. Even in these favorable
circumstances attendance was poor. Most council members did their
business with the government officials and left. Most of those who at-
tended the meeting did so out of idle curiosity, drifting in as they felt
inclined and away again as they tired of joining their colleagues in
peering through the village accountant's spectacles, reading the tax
collector's book of blank receipts, stamping one another with the offi-
cial panchayat seal, gossiping with their neighbors, or making sarcastic
remarks about the business of the meeting.

Three of the five outside members attended the meeting, having been
summoned to Sirkanda for that purpose and having nothing else to
do while there. Four Sirkanda members attended throughout most of
the meeting and were a fairly attentive, if passive, group. Four others
were there sporadically, and four or five nonmembers participated
occasionally. One of the nonmembers was among the eight who signed
the minutes — an official act which was the highlight of the meeting.
Two members, including the vice-president, stopped by for this pur-
pose only, having heard none of the prior business, and joined eagerly
in the signing of the minutes (recorded by the village level worker)
without having read them. In addition an occasional member and
various nonmembers of all ages and both sexes stopped briefly by on
their way to or from the water source, joining curious children and
livestock to hear or see what might occur. Following is an account of
the meeting as recorded by the author:

The tax collector and village accountant had been doing their business in
one of the village shops beginning at about 10 a.m.

At 2:30 p.m. the village president and the village level worker [VLW]

200 THE VILLAGE COMMUNITY

arrived with their record books and a tarpaulin and went to a spot near the
shops to prepare for the meeting. The VLW sat down and wrote out an
agenda of the meeting, which should have been circulated to members in
advance. It read as follows:

All members of the Gram Panchayat, Sirkanda, are informed hereby that
a meeting of the Panchayat is to take place on Sunday [in reality it was
Monday], the twenty-third of December, 1957, at Sirkanda School, at 2 p.m.
All members are requested to be present at that time. The following items
will be considered by the members:

1. Consideration of last meeting's proceedings.

2. Consideration of doing the remaining work to improve the drinking
water supply.

3. Consideration of collection of panchayat tax in arrears and of future tax.

4. Consideration of voluntary labor in the village.

5. Consideration of opening a night school for adults.

6. Consideration of sending five people to village training camp.

7. Consideration of starting a Village Credit Cooperative Society in Sirkanda.

After the above agenda had been written, the VLW asked the president to
call the council members and begin the meeting. This was done and seven
people came and sat for the meeting — four Sirkanda members, three outside
members, and one nonmember. One additional member, who had been pay-
ing his tax, signed the agenda to indicate his presence but left at once with
the excuse that he had to arrange for the tax collector's lunch. The village
shopkeeper, who was within earshot, participated occasionally in discussions.

The VLW started the meeting by reading the minutes of the last meeting,
over one year previous.

Then the second item on the agenda, concerning the drinking water
supply, was discussed. The VLW asked the president what he thought should
be done. The latter told him to request the Block Development Officer to
send an overseer to Sirkanda to examine the problem and advise them how
to solve it.

The third item on the agenda was read by the VLW to the members. The
tax arrears were, for every household in the village, R. 1 per year from 1949
through 1955. The VLW read an official circular which said that a Tax Col-
lection Week should be observed by every panchayat for which the president
should appoint a committee of five persons and go from house to house with
them to collect the tax arrears. When the president heard this, he said it
was impossible. First, no one would go with him as a committee member to
collect the tax. Second, even if someone would go with him, no one would
pay the tax. The president said, and others agreed, that the government
should send an official to collect the tax.

At this point a young village man on his way for water stopped by to
hear what the discussion was about. When the president saw him, he told the
VLW that he would ask this man to pay the tax. Everyone thought that the
man would refuse to pay and thus they would be proved correct. But, to

THE VILLAGE COMMUNITY 289

everyone's surprise, the man took out a ten-rupee note, threw it before the
president, and said, "Now what do you think of me!" Everyone laughed. A
receipt for his Rs. 6 and the change were returned to him, and he went on
his way. This incident embarrassed the members, and the VLW immediately
started putting down the names of the committee members — all the men
present except two of the outside members.

The fourth item, consideration of voluntary labor in die village, was taken
up. The VLW suggested that the villagers should repair the trail from
Sirkanda to the valley (five miles) and should work for one week, beginning
the first of January. Members suggested repairing local trails, but had diffi-
culty deciding which to repair. Objection to repairing the trail to the valley
was based on the fact that it is used by people of other villages who would
thus be benefiting from the labor of Sirkanda without contributing to it.
Members also complained that once before they had done voluntary labor
on trails with the understanding that they would be paid for it and, although
they worked hard, no payment was forthcoming. The VLW pointed out that
this was probably a trick some official played on them to get them to work
and, in any event, they should be satisfied with the achievement and not
demand money. The members still objected. They claimed that the residents
of a neighboring village got money for working on their own village trail
once. Finally they agreed to work on two short trails used only by Sirkanda
villagers (and rarely by them) for one week, starting in three days.

The fifth item, opening of a night school, was then read by the VLW.
Everyone said that it was useless to have a night school, since no one would
attend it. It had been tried by a teacher here once and failed. The matter
was dropped.

The VLW then read the sixth item. It concerned sending five villagers to
a training session on better methods of farming to be held in Sirkanda at
some future date. This was met with considerable incredulity and banter.
When one member, who had retired to a nearby shop for tea, was hailed
and told that he would be sent to school to learn to farm, he replied that
he had done that every day of his life. It was suggested that the blacksmith
be sent. He objected that he had never held a plow in his life and owned
not a foot of land. Finally five people, including one not present, were put
down for this program.

Next the VLW turned to a discussion of work to be done according to the
government's second five-year plan for the village. According to that plan,
he said, the village was supposed to construct a community center during the
year. He described the advantages and uses of such a building: meetings
could be held indoors; it could be used for housing visitors; it could be used
at weddings; the village would be entitled to a free public radio to be placed
in it. At this point a member quipped that the blacksmith (one of the most
avid listeners to the anthropologist's radio) would then not need to buy a
radio. This brought laughter. The VLW pointed out that the government
would give the village Rs. 800 for the community center, an amount consti-
tuting 40 per cent of the total cost of such a building, according to estimates

2gO THE VILLAGE COMMUNITY

submitted by this panchayat a few years before. Members requested the VLW
to get them Rs. 1000 and they would construct the building with that money
alone and spend nothing of their own. The VLW insisted that it would not
be difficult to raise the money in the village. The members were dubious.
One man commented, "If people were like that [willing to donate labor
and money], would the water supply now be in bad condition, and would
the village trails be so rough?" Someone asked how much the anthropologist
would contribute. I replied, "As much voluntary labor as anyone else gives,"
a response which evoked considerable merriment. Finally the president said
he would give Rs. 50. Another man said he would contribute only if the
center were to be constructed in the central settlement area where it could
be useful at times of marriages. An argument developed as to the best location,
the local shopkeeper coming out and arguing that it should be outside the
main village in view of the fact that strangers might use it sometimes. [It
later developed that he had a vested interest in this suggestion. He hoped
to get the panchayat to buy the other shop as a community center, thus
eliminating his business competition.] Nothing could be agreed upon, and
the matter was postponed to a future meeting.

Next a letter from the schoolteacher was read, stating that according to
Education Department rules all children between the ages of 6 and 12
should be in school. It urged all villagers to comply with this rule. The
president turned to one member and said, "Why isn't your son in school?"
The reply was, "Those rules are meant for children that don't have anything
else to do and spend their time idly at home; my boy herds cattle." One
villager supported the teacher's letter, saying, "They didn't have such facili-
ties in this village when we were young, so here we are signing these registers
with our thumbprints. We should send the children to school so that they
will be able to do better than we do."

The next matter was an announcement by the VLW that the council
owed Rs. 12.50 to the Block Development Office. The amount was made up of
Rs. 10 as contribution for an Extension Exhibition to be held in a town in
the valley, Rs. 2 for some sort of training school held there, and half a rupee
for a booklet on village councils. The president said it would be paid from
funds on hand, though no one knew anything about these matters.

Next the economic cooperative supervisor, who had joined the meeting,
asked the VLW to take up the seventh item on the agenda: formation of a
credit cooperative society in Sirkanda. This the VLW did, explaining the
advantages of such a society and of cooperatives in general. He told his
listeners that a society had already been registered for Sirkanda on the basis
of work done by a former government employee who had visited this area
and had secured their approval of such a plan. He urged them to buy shares.
He read the names of those who had been recorded by the previous officer
as being interested. No one wanted to buy. Finally two shares each were
recorded in the names of the president and another man, and one share each
in the names of the shopkeeper and the blacksmith, over the latter's loud
protests of insolvency. No payments were made on these shares. A villager

THE VILLAGE COMMUNITY 2gi

who was not present was listed as president of the cooperative, and one of
those present was designated its treasurer. None of the group showed any
interest at all in this item.

The VLW then passed the register around for signatures, the last official
act of the meeting. Two additional members had come for this event, and
two of the original seven had left, and so a total of seven members and one
nonmember signed. The meeting was over at 5 p.m.

As the members left, the VLW made a plea for the purchase of National
Savings Certificates as part of the government's small savings program. By
the time he finished he was talking to the anthropologist and his interpreter
only, all others having departed.

Not one of the suggestions or decisions for positive action coming
out of this meeting was ever acted upon. The lack of success of this
meeting, as of village self-government and government programs in
general in this area, lies in part in the fact that it did not contain
much of real interest to villagers. Of the six items proposed for action,
only item 2, having to do with the water supply, was of any real
interest to members. Of four additional subjects discussed, only one,
that concerning the community center, was of interest to villagers.
Even more important is the uniformly negative attitude of villagers
toward the government and everything associated with it, including
the village level worker who called the meeting, the council itself,
and every item on the agenda. This attitude and related matters are
discussed in the following chapter.

In summary, the meeting was an utter failure at accomplishing its
ostensible purpose — that of contributing to village uplift and improve-
ment in conformance with the goals of the government's second five-
year plan. In accomplishing immediate ends, which were evidently
also the ultimate ends of those who called it, it was an eminent success.
Every item of business was completed with some kind of action re-
corded or with appropriately filled-in lists of names, volunteers, or
pledges of funds. It was a paper success and from the point of view of
the village level worker and council president, that meant a real
success. From the point of view of the villagers, it was a good meeting
too because it satisfied the officials with a minimum of interference in
village affairs. Only the schoolteacher considered it a farce and a
failure, accurately predicting that nothing would come of any of the
discussions held or decisions made. Not one of the decisions to act on
matters raised at the council meeting was followed through as planned.
Discussion of this phase appears on pp. 319-320. Many villagers stayed
away from the meeting, convinced that it would be a meaningless waste
of time, thus helping fulfill their own prophecy. The meeting was a

20,2 THE VILLAGE COMMUNITY

typical example of reactions to government-sponsored local self-govern-
ment and other programs in this area. In fact, there are villages in
Bhatbair which are considerably more hostile to such programs than
is Sirkanda — villages in which the meeting would, in all probability,
have been boycotted entirely.

Some indication of the significance of the ineffectiveness of village
self-government such as that reported here and elsewhere can be de-
rived from an examination of the attitude of the government of India
toward village government:

The Planning Commission clearly acknowledges that rural progress de-
pends entirely on the existence of an active village organization which can
bring all the people, including those at the bottom of the social and eco-
nomic ladder, into common programs and activities, using assistance from
the Government, and their own contributions in labor, cash and kind.

Further, India feels keenly that the nation's strength as a democracy de-
pends to a most important degree on building, in hundreds of thousands of
villages, effective local governments through which the people can, as respon-
sible citizens, plan and participate in their own and the nation's progress.
(Planning Commission, 1958, p. 123)

Judicial Council

Above the village government is a body locally known as "judicial
council" (adalat panchayat, now officially known as Nyaya Panchayat).
This has jurisdiction over a number of villages. In this area one
judicial council has jurisdiction over most of Bhatbair and a few
neighboring villages. The functions of this council are unknown to
most villagers. Many think that when its name is used the speaker is
referring to the former constituency of the village council. The judicial
council is made up of a total of five individuals chosen from among
five (or fewer) individuals elected by each village assembly within its
jurisdiction. Those elected may not also be members of a village coun-
cil. The composition of the judicial council varies on a rotating basis
with each meeting. This council is presided over by a headman from
among its membership. One Sirkanda man has been headman of this
council, and two Sirkanda men are currently members. This council
has the benefit of a petty government official, the "panchayat secretary,"
to advise it, keep records, and so on. Five members are supposed to
meet weekly at its headquarters, a good-sized village on the edge of
the valley. In reality, it meets relatively infrequently because it has
not enough business to warrant such frequent meetings. It has wide
judicial powers in deciding disputes and breaches of law which would

THE VILLAGE COMMUNITY 293

otherwise have to go into the law courts, or which have been referred
down to it by courts of law. It is legally defined as a "court" and its
members are "judges." It can fine those who appear before it, or refer
their cases to higher courts. It is, therefore, a potentially powerful
body. In practice it is avoided when possible by those with grievances
because of their fear of the nonobjectivity of its decisions. One villager
commented:

It has never solved a single case since it began. The members are mostly
dishonest and unscrupulous, and any who aren't are in an ineffectual minor-
ity. They listen to the man who feeds them well or does them favors. They
are not partial to caste so much as to favors. In this respect they are just like
all other panchayats. Panchayats would be all right if they were honestly
and impartially run, but they never are.

THE OUTSIDE WORLD:
URBAN CONTACTS AND
GOVERNMENT PROGRAMS

Mp\\\\\ The Pahari village is always incomplete in the goods, services, and

personnel necessary to its functioning. These are accessible, to a great
extent, within a larger surrounding area which includes several vil-
lages and is characterized by cultural unity, frequent social interaction,
and economic and ritual interdependence. We have seen in previous
chapters that for Sirkanda this unit comprises the surrounding hill area
(Bhatbair and a corresponding area to the east) within a four-air-mile
or eight-trail-mile radius, a half-day's journey (see map 2). This area
contains some 60 villages and settlements and around 5,000 people.
It is the area within which all the castes and occupational specialists
necessary to the economy are found. It is, therefore, the area over
which artisans move to perform their duties and collect their payments.
In it are also merchants who serve as important sources of goods from
the outside and as markets for local goods. All Sirkanda lands are
found in this area. Within the area all the necessary religious practi-
tioners are to be found — Brahmins, ceremonial cooks, shamans, pujaris,
exorcists, rope-sliders. It is the area within which most (83 per cent)
Sirkanda marriages take place, and therefore within which consan-
guineal and affinal kin ties are concentrated. As a result it is the
largest area within which visiting frequently occurs. It is the maximal
area over which most women have traveled in their lives. Within this
area castes which may number only one or a few families in a village

THE OUTSIDE WORLD 295

find caste-fellows for ritual participation and moral support. In this
area are found not only the formal and informal councils for local
self-government and the judicial councils, but the local representatives
of the state and national governments — the teachers, village level
workers, economic cooperative supervisors, and forestry officers. Into
this area come the land records officers, tax collectors, police, and others
with whom villagers must deal. In the past local officials such as the
mukhia, lambardar, and sayana had their jurisdiction within this area.

In short, while the village is a real, functional unit, the immediate
culture area — that which can be reached in half a day — is the more
nearly self-sufficient unit. Every village within this area relies upon
other villages for many of the necessary elements of village life. It is
this area within which we find most of the "extensions" of the village
as discussed by Opler (1956). Here is found the "rural cosmopolitan-
ism" among villages mentioned by Lewis (1958, p. 320). In relation to
the outside there is, among the villages in this area, a corresponding
"rural isolationism."

However, even this immediate cultural area is not isolated nor in-
dependent. It is in constant contact with similar neighboring areas of
the hills. In fact, it is an area primarily from the point of view of
those villages toward its center. Villages on the peripheries are seen
by their residents as the centers of other, overlapping areas, just as
they are the centers of other, overlapping marriage networks unless
important cultural boundaries hem them in. Moreover, such areas,
including the one surrounding Sirkanda, are in constant contact with
the valley and especially with urban centers such as Dehra Dun and
the hill town of Mussoorie. Many goods and services must come to
these hill areas directly or indirectly from outside. Villagers often go
outside to obtain goods and services rather than obtaining them
through intermediaries. Sometimes, as in the licensing of firearms, they
must go outside. Authority over the people of the area rests ultimately
with outside law and outside government located in Dehra Dun and
Mussoorie. Higher education, certain religious activities, and some
kinds of diversion are available only outside the area. People who
live within the area feel the effects of the outside world in the presence
of outsiders in their villages, in communication with outsiders, and
in their own trips to the outside.

Levels of Identification

In discussing outside contacts it is necessary to define the "out-
side," for several levels of contact and identification are pertinent. Sub-

296

THE OUTSIDE WORLD

groups within the community — the family, sib, clan, caste, and village
— have already been discussed at some length, and these are funda-
mental units of individual identification and orientation. In this chap-
ter the discussion will be in broader terms. Bhatbair (including the
adjoining area in Tehri Garhwal) will be considered to be the "inside"
as distinguished from the "outside" world. It is the unit within which

Fig. 6. Levels of identification and interaction, showing relative inclusiveness
of individual's membership-identification groups. Significant groups are en-
closed within numbered areas. Broken lines indicate less clearly delineated
groups.

Key:
1.

3-
4-

7. Sib

Ego

Nuclear family

Extended family (household)
Lineage (or minimal lineage)
Kindred (affinal kin follow a
pattern similar to that of kin-
dred but not coextensive with

it)
6. Maximal lineage (where ap-
plicable)

Clan is indicated by diagonal hatchures. Wives (and other local affiliates) of
ego's sib are indicated by cross hatchures within the clan but outside the sib.

Note that cliques (10) do not divide extended families (3) nor do they extend
out of the village (11), but they do divide lineage (4), kindred (5), maximal
lineage (6), sib (7), and caste (8 and 9). Caste as a marriage network (8) includes
all kin (1-7), and excludes some clique, village, and Bhatbair people (10-12).

9-
10.
11.
12.

13-
14.

15-

Caste as a marriage network

Caste as a named group

Clique

Village

Bhatbair and vicinity

Garhwal

Pahari area

India

THE OUTSIDE WORLD 297

Sirkanda residents feel entirely at home. Beyond it they are, to vary-
ing degrees, "outside" — in alien territory, among people different from
themselves.

Nearest to the residents of Bhatbair, in terms of cultural distance,
are the residents of adjacent Pahari areas to the east — people who speak
essentially the same dialect, dress in much the same way, share many
of the same beliefs and values. These are areas from which marriage
partners are occasionally obtained, in which distant relatives live, to
which local people are sometimes invited for ritual observances and
fairs. These are the areas to which Sirkanda people trace their own
ancestry. Rivalry and animosities occasionally crop up, as when a
council meeting was held to decide a dispute arising out of insults
exchanged between two men of the same caste, one from Bhatbair and
one from an area to the east. Each area supported its man, and the
council had wide representation from villages in each area. More
commonly, good-natured insults and banter are exchanged between the
residents of these areas. Bhatbair people used to call their neighbors in
Tehri Garhwal "slaves" because they were under the Raja's absolute
power. Now the shoe is on the other foot and the taunting goes the
other way, as the former subjects of the Raja allegedly have more
forest and land at their disposal.

More significant differences and loyalties are those associated with
more prominent cultural boundaries. The residents of Jaunpur, a hill
area north of Bhatbair (and not identical in extent to the area which
the government designates by this term), dress somewhat differently,
build houses in a slightly different style, and in general bear cultural
differences noticeable to both groups. Mutual suspicion becomes
prominent at this level. Jaunpur women are thought to be given to
witchcraft by people in Bhatbair, and, although wives are occasionally
obtained or given across this boundary, it is done with some trepidation.

Where linguistic and social traits are significantly different and dis-
tances are too great to permit accurate appraisal of the residents of
two areas, feelings are even stronger. The polyandrous people of
Jaunsar-Bawar, though recognized as fellow Paharis, are considered
immoral, treacherous, and given to evil magic and witchcraft. Wives
are not exchanged with them, travel to the area is avoided, and most
information about them is in the nature of inaccurate gossip. Residents
of a Pahari area some distance to the southeast are thought of in a
somewhat similar light. At this level dialectal differences become im-
portant. Jokes are frequently told which ridicule the alien dialect.
One probably apocryphal story concerns a man of Bhatbair who
allegedly went to a village to the southeast to see about making mar-
riage arrangements for his nephew:

2g8 THE OUTSIDE WORLD

In (he alien village he overheard the brother of the prospective bride
make a statement to his mother which, in the dialect of that region, meant
"The buffalo is ready to be milked," but to the visitor it sounded like, "I
am ready to have sexual intercourse." The mother replied, "Then untie the
calf," which in the visitor's dialect meant, "Then untie your loincloth."
Shocked and frightened to hear a man speak in this unseemly fashion to his
mother and to receive such a reply from her, the visitor ran out of the
village, headed for home. The dismayed family shouted after him, "Stay,
for tomorrow we will kill a goat to feast on," which only served to accelerate
his departure, as in his dialect it meant, "Stay, for tomorrow we will engage
in sodomy."

Despite differences in beliefs and attitudes with distance, there is
recognition of cultural similarities among Paharis. A Pahari is a more
congenial stranger than is a non-Pahari; he is more easily understood
and can be more readily placed in terms of the relatively closed social
system indigenous to this area. Common language, culture, and history
result in a kind of incipient Pahari group consciousness. The fact that
Paharis are considered and treated as a group by non-Paharis may also
be a factor. So far this consciousness is largely in negative terms at
the broader levels; the differences between Paharis and non-Paharis
are more important than the similarities and common interests among
Paharis. However, there is some agitation for a separate hill state in
Pahari regions to the west. Among educated Paharis of Garhwal and
Tehri Garhwal there is considerable pride of origin, with local societies
and even publications in some urban areas stressing the heritage of
Garhwal.

Paharis distinguish themselves terminologically from plains and valley
people who are called desi (native). Foreigners are even farther re-
moved and are called pardesl (foreign). Representatives of either group
are suspect, threatening, and somewhat baffling to Paharis. In an iso-
lated area which stresses kin, caste, and community affiliation they are
simply outside the system. To fit them into it is usually an impossible
task; hence the traditional unfriendliness with which Paharis greet
strangers. The usual reaction to them is avoidance. Inevitable contacts
with them are kept as short and impersonal as possible. Plains people
are, of course, better known and better understood than foreigners,
but in their presence Paharis feel ill at ease. The plains person is
considered to be sophisticated and well-off, to have a superiority com-
plex, and to be arrogant, untrustworthy, avaricious, and immoral.
These attitudes are not born of recent contacts. In 1828 Traill (1828,
pp. 217 f.) noted that "all mountaineers unite in an excessive distrust
of the natives of the low country, whom they regard as a race of

THE OUTSIDE WORLD 299

swindlers and extortioners. . . ." One present-day Pahari who had
been to Delhi commented, "I don't like Delhi or any of those places.
There are too many people there and they make fun of us. I like
Dehra Dun where you can always see lots of Paharis around." Sirkanda
people feel that they can size up a Pahari, but a plains person or a
foreigner is an unknown quantity: "You can't tell how much water
they're standing in."

Concerning Europeans there are conflicting stereotypes. One set of
stereotypes, based upon observation and stories of British soldiers, views
them as immoral scoundrels. Another, based upon the reputation of
the British civil servants, views them as stern, cold, and distant but
honest and fair. A third, based upon closer observation of individuals
(customers for milk, occasional hunters and hikers), views them as well-
meaning but odd. One villager commented,

Englishmen [that is, "white people"] are very strange. They always cooper-
ate with one another and never dispute among themselves, but their anger
is easily aroused. Their customs are strange. If one of them comes to a village
he does not enter it and ask for a place to stay, but pitches his tent alone,
outside the village. They do not care what their food tastes like; they use
no spices but just put a pinch of salt on it and gobble it up. They eat mostly
from cans. They never take off their shoes — even if one were going to climb
a tree he would do it with his shoes on. They also have immoral sexual
customs.

It is usually assumed that foreigners (other than hunters) who come
to the mountains are missionaries, government officials, or spies. All of
these pose a threat. One foreigner came through the village several
years ago riding a horse and carrying a pistol. He stopped for tea and
then passed on. He is assumed to this day to have been a spy. Aside
from occasional military personnel, he is said to have been the only
foreigner to come to Sirkanda.

The Non-Pahari World

The non-Pahari world impinges upon Sirkanda and the surround-
ing villages in many ways. Although the degree to which it does so
has apparently increased since Independence, such contacts are not
new. From earliest times traders, ascetics, pilgrims, and immigrants
have traveled from the plains to this area. Tibetans or Bhotiyas have
come from the high country to the north to pasture their sheep in the
winter. There has been outside authority — the Garhwal Raja, the
Nepalese, the British, the "Congress government." This authority has

gOO THE OUTSIDE WORLD

been represented to villagers largely by its soldiers, police, and tax
collectors.

The British found the hills especially useful as a place for summer
rest areas, military cantonments, encampments, and maneuvers. They
established the famous "hill stations," such as Simla, Mussoorie, Naini
Tal, Darjeeling, and military cantonments, such as Chakrata and
Ranikhet. The "Tommy" (British soldier) gained the same reputation
for arrogance and immorality that he gained over most of India and
that soldiers have had from time immemorial among the civilian
populations near which they have lived.

Before the beginning of World War I a small summer rest camp
was set up by the army a little over a mile from Sirkanda. There small
contingents of Nepalese troops and their British officers came during
the hot season. Although they were rigidly segregated from all vil-
lagers except the family which sold them milk, and had no conflicts
with the villagers, their presence was felt. The army bought village
land, was a customer for village milk, and its presence restricted the
movement of villagers, especially women, in the vicinity of the camp.

World War I found many Garhwalis, including a few from Bhatbair
and one from Sirkanda, enlisting in the Garhwal Rifles to fight for
the British. Sirkanda people still sing the song and proudly recount
the tale of a Garhwali hero who took command of his unit when all
its officers had been killed and led it to victory in a crucial battle.
His widow resides in her husband's village on a comfortable pension
from the government, and this too is a point of pride among Paharis
of the region.

In the 1920's several Sirkanda boys attended school in nearby vil-
lages of the valley, two or three reaching the eighth class. They were
the first from this area to be educated. When the army abandoned the
summer encampment near Sirkanda in the early 1930's, the resulting
intravillage court case over ownership of the land lasted nine years.
It necessitated extended trips by several villagers to courts as far
away as Allahabad. During this period a merchant from the valley
was called in to run a competing shop in the village, and he has re-
mained to the present.

Throughout these times adult male villagers occasionally went on
pilgrimages to the shrines of the high Himalayas, to Hardwar or to
distant Gaya in Bihar. Boys and young men, then as now, occasionally
left home to seek adventure in the plains, usually returning after a
few months.

World War II brought further army recruitment, with two Sirkanda
men among half a dozen Bhatbair volunteers. The Sirkanda volunteers

THE OUTSIDE WORLD 3OI

saw much of northern India in their service but did not go outside
India.

Independence affected Sirkanda relatively slightly. The partition of
India which accompanied it brought bands of fleeing Muslims and
pursuing Hindus through these hills, but for the most part Sirkanda
was not involved. Bhatbair residents refused sanctuary to Muslims of
the valley out of fear of the marauding Sikhs who were pursuing them.
However, they withheld information from the Sikhs about Muslim
movements and helped the Bhatbair Muslim family to remain un-
molested. In nearby Pahari areas fleeing Muslims were looted and
occasionally killed as they passed. One Sirkanda family became wealthy
overnight from money apparently looted from a Muslim chan left
in their care.

After Independence the influence of the government of India was
felt increasingly. Laws regarding use of forest lands were altered and,
worse yet, enforced. A school was built in Sirkanda and various govern-
ment programs had their effects locally, as will be discussed below.
Government representatives in the village have become an increasingly
familiar sight since Independence. A schoolteacher lives in the village
most of the year; two other officers are there much of the time. Govern-
ment survey parties on training exercises have often camped near the
village during the hot season, providing a source of outside companion-
ship for some of the younger men. The postal service now reaches
Sirkanda with a postman coming through about once every two months,
while letters can be mailed whenever a villager or passer-by goes to
the valley.

Throughout this time outside markets have increased in importance.
Milk has long been a cash product to be sold in Dehra Dun, Mussoorie,
and the smaller, intermediate markets such as Nagal, Rajpur, Sahas
Dhara, and Suakholi. Mussoorie has been a market for over 100 years,
Dehra Dun for longer than that. Rajpur was a major market for 80
years until a motor road was built to Mussoorie in 1930 and Rajpur
ceased to be an important way station en route to Mussoorie. As better
trails have been built in the Bhatbair region and beyond (beginning
with the pack trail past Sirkanda built by the Tehri Garhwal Raja
in 1914), and as horses and mules have been acquired, agricultural
products have been sold with increasing frequency. Potatoes have be-
come an important cash crop. At the same time, more goods from the
outside have come into the village — cloth and clothing, utensils and
tools, corrugated iron roofing, tea, cigarettes, and so on. Trips to the
valley and Mussoorie for trade have become commonplace. Dehra Dun
has been increasingly frequented by villagers on official business re-

g02 THE OUTSIDE WORLD

lating to the government. Dehra Dun and Mussoorie have the added
attraction of a full range of entertainment facilities.
An 85-year-old man commented:

In the old days a man wore a loincloth and had a blanket for a wrap.
Now everyone has to have a wardrobe of three coats, three shirts, three
pajamas, and shoes and many other things. We used to go to town once a
year, in a group, to buy staples, and we came home as soon as we were
through. Now someone goes almost every week. They go alone and don't
even tell others they're going. Sometimes they stay for a day or two and
waste money on the cinema or prostitutes. In my day we didn't have time,
money, nor inclination for such things.

Occasionally a Vedic doctor is consulted in town, and once a valu-
able horse was taken to the government veterinary. The number of
boys going to school in town has increased. Two young Sirkanda men
have taken jobs near Delhi, but their families remain in the village.
Two men from a Pahari village nearer Dehra Dun have taken jobs
in Dehra Dun, but they continue to live in their village. In all this
time no one from Bhatbair has moved permanently into Dehra Dun
or Mussoorie. When villagers go to Dehra Dun they have to stay over-
night with the wholesalers to whom they sell their products, or in a
temple compound which is open to all. In Mussoorie they usually
return to chans before nightfall. There are no relatives or fellow
villagers in either place to afford them a place to stay.

Limestone quarrying operations have penetrated to the foot of the
hills, so that trucks and workers operate within five miles of Sirkanda
and are pushing farther constantly. Tourists and worshipers come in-
creasingly to the temple and springs at Sahas Dhara five miles away
now that a dirt road enables cars to reach it. Travel to Dehra Dun is
facilitated by regular and inexpensive bus service from Rajpur to
Dehra Dun, cutting the 16-mile walk from Sirkanda to about 9 miles.
The Mussoorie-Tehri trail is being prepared for bus service so that
before many years have passed the walk to Mussoorie will be cut from
16 to 8 miles.

Therefore, outside contacts have been increasing. In view of the
changes that have come about over a long period, it is perhaps sur-
prising that Sirkanda has not been influenced more than it has by
these contacts. A Rajput householder with an eighth-grade education
commented:

The reason most people here are so backward is that they are uneducated
and they have never been outside and gotten new ideas. You have to go
outside to learn to improve yourself. The trouble with this village is that

THE OUTSIDE WORLD 3°3

people aren't educated — look how few children attend school. The village is
accessible enough to town that it should have improved but it hasn't, while
some other Pahari areas have improved greatly. Even some government
officers are Paharis, but none come from around here. Here people don't
know enough to be able to improve.

Not all villagers share this attitude toward the outside. Some see it
as a contaminating influence. One man remarked:

Contact with towns has been bad for Pahari villages. Farther up in the
hills the people are better off. There they still wear simple clothes and prac-
tice the old customs more than we do. They are freer. Their women are not
ashamed to dance and sing and drink with the men. They have more village
celebrations. They are content with what they have, simple though it is, and
many of them are wealthy because they have nothing to waste their money
on. We are better off in this respect than are people who live near the cities,
but the people farther up in the hills are better off than we are.

One reason for the relative isolation of Sirkanda is that it has
nothing to attract outsiders. It is high on a ridge and is not en route
to any place of interest to plains people. Unless limestone is found in
its immediate vicinity, it is likely to be bypassed for a long time.

Outside influences have been felt in Sirkanda primarily as a result
of direct, personal contact. Villagers who have taken extended trips
outside have returned to tell their stories. Some villagers make frequent
trips to market, where they learn the current news. The alien merchant
and government employees have added their experience. Itinerant
merchants and religious practitioners have contributed, too. From such
contacts villagers have acquired some idea of what goes on in the out-
side world. They are familiar with Gandhi, Nehru, and the Congress
party. They know of India's difficulties with Pakistan and sympathize
with India (or, more accurately, they are anti-Muslim). They know
something of the struggle between Russia and the United States. Cer-
tain domestic policies — land reform, Harijan uplift — are vaguely
known. New developments — the launching of Sputnik, the outlawing
of prostitution in Dehra Dun — cause comment and speculation. Aware-
ness of and interest in foreign countries and other parts of India are
very limited. Discussion with the anthropologist about America arose
relatively infrequently. Villagers usually began with an inquiry as to
how long it would take to walk there and proceeded through discus-
sions of the price of consumer goods, productivity of land and animals,
social customs, and finally sex habits.

Written materials and mass media have been of little importance.
Mail, which is received rarely, consists primarily of instructions to the

304 THE OUTSIDE WORLD

council or of personal and official mail for the teacher and other out-
siders in the village. Villagers who go outside communicate rarely if
at all with their relatives in Sirkanda. There are no periodicals in the
village and no books other than the school materials of the teacher and
a few largely unread religious works. The only current documents
are posters given to the village level worker or the village president,
urging public health and personal hygiene and explaining in graphic
detail the intricacies of the change to the decimal system of coinage.
There has been no functional radio in Bhatbair, although one was
once brought to a Sirkanda chan, where it failed to operate. One
brought by the anthropologist attracted an interested audience. Film
music and news programs were the favorite radio fare, in that order.
Newspapers and magazines brought to the village aroused little in-
terest aside from their illustrations. The low level of literacy even
among those who claim to know how to read is probably largely re-
sponsible for this reception.

Contacts between villagers and representatives of urban culture and
the "new India" will probably continue to take place primarily
through villagers' trips outside and through government representa-
tives who are attempting to initiate various programs in this and
neighboring hill villages. Governmental programs of planned change
and villagers' responses to them will be discussed below.

Urban Contacts

Urban centers are important to Sirkanda villagers as they are to
all peasants. In fact, Redfield (1957, p. 31) defines peasantry with
reference to the presence of cities, stating that "there were no peasants
before the first cities."

In Sirkanda as in other peasant communities, "the account the
peasant takes of the city or town is economic, political, and moral"
(ibid.). That is, he carries on some trade in the city, he pays his taxes
to people from the city, he is subject to control by institutions ema-
nating from the city, he knows something of city ways, and in fact
adopts some values of urban dwellers as his own.

The sources and amounts of urban influence in Sirkanda are varied.
One of the most important kinds of contact is that of direct, personal
experience. This varies from person to person. Women have much less
urban experience than do men. While there are men who have never
been to Dehra Dun, others go every few weeks. Of 116 men in Sirkanda,
3 have never seen Dehra Dun, 62 have seen nothing beyond Dehra
Dun and Mussoorie, 20 have been as far as Hardwar or an equivalent

THE OUTSIDE WORLD 305

place within 40 miles and bordering the hills (such as Rishikesh and
Kalsi), 6 have been into the plains but only as far as one of the smaller
towns or cities within 60 miles of Dehra Dun (Saharanpur, Muzaf-
farnagar, Ambala), 15 have been as far as Delhi (150 miles), 8 have
been on pilgrimage to Gaya (750 miles), 1 has been to Calcutta, and 1
has been over most of North India in the army.

Outside employment, discussed in chapter 2, has given a number of
villagers firsthand experience with city life for months at a time, but
no one has moved his family outside and only one man has taken a
lasting job outside (another who recently joined him may be the second
to do so). The general opinion of city life is that it is easy, entertaining,
sinful, and expensive. For well-to-do plains people it is fine, but it is
not suitable for poor Paharis. One Rajput, apparently referring to
emigration, commented: "Anyone who leaves the village is a fool be-
cause here he is a king but there he is a slave." Another, referring to
short-term visits, objected to this view, saying, "A person can learn
and improve himself only by going away and having other experiences.
In the village he can learn nothing."

The one Sirkanda man who had gone out and made a success of
life on the outside was pointed to with pride by some villagers, al-
though he had apparently not been well thought of at the time of his
departure and he left as the result of a family dispute. Disputes were
involved in many cases of temporary emigration from the village.
Niehoff (1959, p. 501) comments that among factory workers of Kan-
pur, ". . . the push from the village was stronger than the pull from
the city and the factory." The rise of the Sirkanda emigrant was
graphically described by a villager:

Maru used to live here. He was ignorant and dirty. He wore the worst
clothes in the village. He always had manure all over his hair and clothes
from carrying it to the fields, and he never had taken a bath. Even his wife
didn't care for him, and she took up with his cousin. Because of that he ran
off to Delhi, and no one heard from him and no one cared. Then he came
back on a visit. He was clean, well-dressed, and handsome. He talked just
like a sahib. He had a job at Rs. 150 per month in a spinning mill. He had
had one hand mangled in a machine and got Rs. 800 for that. He brought
a radio and good liquor with him. The radio did not work, but everyone
was impressed. He still works there and is doing very well. It was all in his
fate. It would never work for others unless that was in their fate, too. A boy
from here has joined him recently. He may fail, but perhaps it is in his fate
to succeed, too. We shall see.

There was considerable interest on the part of some Rajputs in Sir-
kanda in getting a boy into "service," that is, into a white-collar job.

go6 THE OUTSIDE WORLD

So far there had been no success. Two Rajput boys had recently been
sent through the eighth grade with this end in view, but they had
failed. The father of one complained:

I wish one of the boys from here would get a job in service. We sent my
eldest son and his cousin out to school so that they could succeed. We gave
them everything they needed so that they could study and do well. When
they wanted bicycles, we bought them. We got them good clothes. But all
they did was waste their time going to the cinema and the like. They did
not apply themselves, so they both failed the examinations at the end of the
eighth grade. Now they are useless. There is plenty to be done here in the
village and a good income can be made from it, but they do not want to do
village work any more. They have been spoiled. My next eldest son, whom
I did not educate, is content to work in the village, but not these educated
boys. They are too big for that. They carry a pen and pencil in their pockets
when they go to herd goats. They go off to town when they get a chance
and waste their time and money. My boy says he wants a job there, but what
can he do? Nothing.

I know that young men have new ideas now and I try not to stand in my
boy's way. In my youth we wore only a loincloth. We ran around bare-
footed. Now every young man wants an outfit of 22 clothes and goes to
fetch water in boots costing Rs. 18. One has to move with the times. I cater
to my boy's wants within reason. Last winter I bought wool for coats for our
family. That for myself cost Rs. 8 per yard, that for my father cost Rs. 12
per yard, that for my son cost Rs. 16 per yard. But he did not even appre-
ciate that. This fall the boy sold his coat to the schoolteacher for Rs. 10 so
he could use the money for something else. He claimed the coat wasn't
stylish. Now my youngest boy is in school. I hope he will do better, but I
fear the same fate for him. The boys go wrong because they are not super-
vised in town, but how can we supervise them when we live way up here?

The son in question frequently inquired of me about jobs in town,
but he had totally unrealistic goals in view of his qualifications. He
spoke of wanting to be an office worker or cinema actor. He would
consider no job as lowly as those for which he might have been
qualified. His father commented that "Even if he made Rs. 100 per
month it would not meet his expenses outside. Here in a good season
we make more than that and our expenses are small." This boy's
cousin was the one who ran off to join the millworker, and this boy
was tempted to join him, but ties of wife and joint family were too
strong to allow him to do it secretly and he was not granted permission
by his father to do it. His affectations were the bane of his age-mates,
who had the maximum village education (five years) or less. They, by
and large, were content to remain in the village to live and work with
occasional visits to town for fun. Their requirements for clothes and

THE OUTSIDE WORLD 307

other possessions were patterned on his, but their concepts of what
constituted "fashion" in clothing were less sophisticated than his.

The opinion of most villagers is that cultural distances are too
great to permit their successful adjustment to town life. Despite their
efforts to practice the proverb, "When in Nepal, eat buffalo" (do as the
Nepalese do), they find it difficult to adjust to the unfamiliar way of
life of the city. One young man, who had aspirations toward the bright
lights of the city, explained:

People in this village are mostly content. They have the things they want,
they eat enough, and many of them are well off by local standards. Some-
times they like to go to Dehra Dun. They put on their best clothes and feel
proud of themselves. But when they get to town, everyone there thinks they
are fools. We don't know how to dress or act in town. There even a poor
untouchable puts on a shirt and pajama and looks respectable, but we can't
look like that. Even if we spend Rs. 200 on the finest cloth and have the
best clothes made, we still look like fools in town.

Many stories are told on themselves by villagers, illustrating this
point.

Attractions in towns which draw Paharis are cinemas, "hotels"
(shops which sell sweets, meals, and tea to the accompaniment of loud
radio music), drinking establishments, and prostitutes. Although these
are rarely the sole reason for going to town, they are the side attractions
of trips intended primarily for trade or official business with govern-
ment offices. A Sirkanda blacksmith is often teased about his affinity
for these pleasures. He protests, "It is true, no doubt, but why must
you advertise the matter? There is no act of valor in it." Two examples
of his experiences, as told by himself, will serve to illustrate the adven-
tures of Paharis in town. The first of these concerns his first visit to a
prostitute in Dehra Dun.

I had a little money and nothing to do, so I decided to go to a prostitute.
I had never been before and I didn't know the times when they are avail-
able, but I knew the compound where their houses are. About 12 at night
I went there and entered the gate of the large compound containing the
houses of prostitutes. Just after I went in and was looking around to decide
my next move, the Muslim gatekeeper closed and locked the gate I had just
come through. I was frightened — I wondered how I could ever get out again.
Then I saw the gatekeeper hitting people who tried to leave [apparently
people who wanted to sneak out without paying]. I wanted to get away. I
walked around the compound, keeping to the shadows, but saw no way to
escape. Finally I climbed a stairway to one of the balconies. The girl there
asked me what I would pay. I had Rs. 20, so I gave her that [actually Rs. 2
to 5 would have been plenty]. She took me into her room and brought out

go8 THE OUTSIDE WORLD

betel nuts and a cuspidor. She called the musicians and they played and she
sang for me. I sat and watched and listened. I never knew what would
happen next, but I didn't want to go out and be beaten by the gatekeeper.
The girl asked me several times if I was ready but I always said no, thinking
that would cost extra and I had no more money. Finally, much later, the
girl grew disgusted and demanded, "Why have you come here and paid all
this money and not wanted intercourse? What do you want?" Then I felt
foolish and realized that this was included in the price. After that I went
back many times.

Another occasion was that upon which this man decided to attend
a Western-type restaurant just as he had seen the important towns-
people do.

I went to the Nilam Bar where I had seen the big people go in their
fancy cars — the same people who drive back and forth past the prostitutes'
compound waiting until there are not many people around, and then send
a servant in to make sure there is no one like me widi their favorite girl,
before coming in themselves.

I bought a newspaper to take in with me so that they might not know I
am illiterate. I went in and sat down. Soon a waiter came, all dressed in
white, and handed me the menu, but I just said I would have tea and cakes.
The waiter asked if I wanted single or double. I thought single would be a
cup and double a tumblerful, so I ordered a double. I was amazed when he
brought a great glass pitcher full of tea — enough for an army. The sugar
and milk were in separate containers rather than cooked in the tea — a sur-
prise to me. However, I drank some tea in a very blase manner and ate some
cakes. I left some of each uneaten to show that I am well-bred. The big
people never eat all that is put before them as we villagers do. If I had
eaten all that was served they would have known I was a bumpkin. After
I was through I looked at my paper and the waiter soon brought my bill
on a plate. I then realized that I did not know the cost of my refreshment
and could not read the bill, but I did not want to admit this. Therefore, I
got up absent-mindedly and walked to the door. There I asked the man at
the desk where my bill was, and he told me it was on my table. I said I
hadn't noticed it and now didn't have time to go back to get it as my car
was waiting for me outside, and I asked him to tell me the amount. He did
— it was Rs. 6, which I paid him. That is the most expensive tea I ever had
[normally tea and cakes would cost not over R. 1 at the fanciest of eating
places].

The city is an exciting, but in many ways a mysterious and dangerous,
place for most Paharis. Even the cinemas are not clearly understood
by many of those who attend them. One man, having recently seen
one, was asked about it. Being illiterate, he knew neither the title
nor the cast. The only actor he recognized, he said, was Dr. Rajendra

THE OUTSIDE WORLD 309

Prasad (President of India), who appeared near the beginning of the
film. He had evidently failed to differentiate the preceding newsreel
from the featured film. He was vague about the plot but agreed
vigorously when it was suggested that it was about a boy and girl in
love who, in spite of great obstacles, finally came together (the plot of
go per cent of the films shown). "Yes, yes, that was the one I saw," he
affirmed. Occasionally a villager happens into an English-language
cinema but, having paid his admission, usually stoically sits it out.
He is often rewarded, according to those who have seen these British
and American films, with a display of nudity and sexual aberrations —
the local interpretation of some Western women's wear and love scenes,
respectively.

An important feature of village-urban contacts is that of victimiza-
tion of the naive Pahari villager by urban shopkeepers and others. A
foreigner in India is likely to get the impression that shopkeepers,
service personnel, and officials are "out to get" him. S. J. Perelman
(1947, p. 96) has summed up his impression of India in the phrase,
"It's not the heat, it's the cupidity." However, this is not exploitation
of the foreigner as such; it is exploitation of the ignorant, the naive,
the unwary, in short, the vulnerable. It hits the villager, and partic-
ularly the Pahari villager, hard, for he is just as vulnerable as the
foreigner and much less able to afford the results of his vulnerability
than are most foreigners. Such exploitation affects villagers in most
contacts with outsiders, but especially in the urban areas.

The average villager is at the mercy of shopkeepers when he goes
to town, and his vulnerability is advertised by his dress, speech, and
manners. He is an easy mark. He is the hillbilly come to town, and
everyone knows he has money with him, as otherwise he would not
have made the trip. He buys inferior goods at inflated prices and is
none the wiser. He pays Rs. 2 for a flashlight bulb that costs anyone
else R. 0.25, and feels fortunate to have found a store which sells
such an intricate mechanism. He is victim of the cheat and thief
and the dishonest official as well. The only Sirkanda villager who ever
tried to buy a radio had his pocket picked of the Rs. 350 in a cinema
before he had had a chance to spend it. Money is extorted from Paharis
regularly by corrupt inspectors, toll-tax collectors, and police who
threaten a false report if no bribe is paid. It is a simple matter to
produce a bottle of illicit liquor and accuse a Pahari of having sold
it — better to pay a bribe than risk a heavy fine or imprisonment. The
Wisers have described well the position of residents of the plains
village in which they worked for several years. Paraphrasing villagers'
statements they say:

o10 THE OUTSIDE WORLD

In the cities they devise ways of exploiting us. . . . We are at home in the
wholesale market. But when we get our money and want to take home some
cloth, the shopkeepers get out the pieces which they have been unable to
dispose of, and persuade us to buy them at exorbitant prices. We know that
they are laughing at us. But we want cloth, and the next shopkeeper will
cheat us as badly as the last. Wherever we go in the town, sharp eyes are
watching to tempt our precious rupees from us. And there is no one to ad-
vise us honestly or to help us escape from fraudulent men. When we go to
town to attend the courts, there are men everywhere waiting to take advan-
tage of our ignorance and fear. Our lawyers charge fees which they know
are beyond our means to pay. And then if we win a case they think that
they deserve an extra large gift. Sometimes there is a sincere helper among
them, but we are never sure who is what. (C. V. and W. H. Wiser, 1951, p. 163)

As has been mentioned previously, social pressure is one factor
which keeps people from bringing urban or other alien traits into
the village. The man who had returned to the village after being
in the army in World War II, and who had enjoyed his outside experi-
ence very much, replied to an inquiry:

When I got back to the village I didn't like it at first. I was used to the
comforts of army life. But I gradually became accustomed to the village
again. I would have liked to bring some things home with me, especially
utensils and other conveniences for the house. However, I brought only some
cloth for the family, and what I brought is similar to what other people
here use. I didn't bring any really nice clothes or other things because peo-
ple would have laughed at me.

This attitude has not prevented acquisition of material possessions,
but it has been one reason for the gradual nature of adoption of new
things. Clothing styles, especially those of men, who are in contact
with outsiders more frequently than are women, have changed con-
siderably in recent years. Costume jewelry and other ornaments have
been acquired by women. Household possessions have come to in-
clude more utensils procured from town bazaars. For the most part
these have been relatively minor changes — better artifacts rather than
new ones. A few people have gotten new items, spurred by urban
example. One villager had a functioning phonograph and a few worn
records, and someone else was alleged to have a similar machine, long
inoperative. The only radio in Bhatbair history, brought by the man
employed near Delhi, had failed to operate. When I left, mine was in
great demand but the likelihood is that it was used only until its
battery was exhausted. A few young men in the village had mechanical
pencils or pens, and at least one owned a cheap watch. One purchased
a Japanese cigarette lighter adorned with photographs of nude girls.

THE OUTSIDE WORLD 3 1 1

Many boys owned pocket knives. The village tailor had a hand-
operated sewing machine and a charcoal-heated iron for use in his
craft.

Tools and materials for agriculture have changed little. Corrugated
iron is now often used in roofing, and one new house has a little
cement in its construction. Household furnishings now include string
cots and cotton quilts or rugs in addition to the traditional wool
blanket used for sleeping on the floor. A few kerosene storm lanterns
and one or two flashlights are in the village.

There is an accelerating demand for all these things as they become
increasingly accessible and as villagers become increasingly aware of
them. However, material items in the village, like behavior and beliefs,
are predominantly traditional.

The Government in Sirkanda

Sirkanda is strategically located in many respects. It is the gateway
from the valley to much of Bhatbair and neighboring areas to the east.
It is the most accessible of the large villages of the interior of this
area. As such it has been the focus of recent attention by agencies of
the state and national governments. It has been designated as the seat
of the council, originally for much of Bhatbair and now for a less
populous area on the western edge of Bhatbair. It is the village which
serves as headquarters for the village level worker and the economic
cooperative supervisor. It contains the largest of three schools in Bhat-
bair. On the rare occasions when tax collectors or other officials come
to Bhatbair, they usually do their work in Sirkanda.

In order to understand the relationship of villagers to specific pro-
grams in the village, it is necessary first to understand attitudes toward
the government in general.

The first question asked of the anthropologist in Bhatbair villages
and in all other hill areas visited was invariably, "How can we get
our forests back?" In Bhatbair the remark which was usually associated
with this question was, "When are the British coming back into
power?" These two questions sum up village attitudes toward the
present government. It is felt by villagers that the British government
had two outstanding qualities: (i) it did not meddle in village affairs,
(2) it was an ultimately available source of impartial, if not always
comprehensible, justice. Thus, it is believed, it was the antithesis of
the present government. To the further credit of the British was the
fact that they established the bhaichara system of land ownership in
Sirkanda, giving every high-caste family the land it cultivated, elim-

312 THE OUTSIDE WORLD

inating the intermediate landlord of earlier times, and charging
relatively low taxes.

The current state and national governments (not distinguished
from one another by villagers) are thought by Sirkanda people to be
made up of corrupt troublemakers. Anything advocated by the gov-
ernment or its representatives is automatically suspect and is usually
opposed out of hand. Villagers' attitudes were perfectly exemplified
in their response to Republic Day ceremonies planned for the village
by the teacher, and supported by the other two governmental em-
ployees stationed in Sirkanda. This celebration was announced well
in advance. A full day of activities was scheduled, beginning with
a flag-raising ceremony at the school, followed by a procession of
school children, led through the village by the teacher and his fellow
sponsors, singing patriotic and religious songs. In the afternoon a
village assembly meeting was to be held at which patriotic speeches
would be made and some village business would be conducted. At
the conclusion of this, tea and sweets were to be distributed.

Villagers completely ignored the whole program. When the singing
group went through the village not a soul joined, acknowledged, or
even watched, the procession. Even the usual number of people who
would be expected to be in view were out of sight, consciously avoid-
ing the display. No one came to the scheduled meeting, and only the
school children were on hand to receive sweets. This was enough to
discourage even the conscientious teacher and to convince the other
workers that their job in Sirkanda was hopeless.

Land Reform

One of the first acts by the Independent government of India
to affect Sirkanda was the nationalization of forest lands in 1953.
This affected all uncultivated and unoccupied lands surrounding the
village. In so doing, the government placed restrictions upon cultiva-
tion of new lands and upon access to the products of uncultivated
lands, apparently with a view to reducing floods, erosion, and depletion
of forest resources. By this act alone, most hill people were alienated
from the government because it hit at the basis of Pahari livelihood.
Paharis live in their none-too-hospitable environment largely by full
utilization of forest products. Forest grass and leaves are fodder for
their animals, and forest trees are made into their tools, many of
their utensils, and essential parts of their houses. Most importantly,
the forest provides wood for fuel. Without such fuel the villagers
would have to resort to the plains practice of burning dung, and they
would then be unable to fertilize their fields. Without fertilizer their

THE OUTSIDE WORLD 3 13

fields would be unlikely to produce a subsistence crop. Moreover, as
is readily evident to any observer, such utilization of the forests does
not seriously affect the forests. Where devastating depletion has oc-
curred and is still occurring rapidly and irreparably, is where outside
contractors have been given the right to cut trees in order to make
and sell charcoal. In this manner whole forests go down that have
been unaffected by hundreds of years of village use. Villagers would
gladly see curbs put on this practice, which benefits them not in the
least and depletes their own forest resources.

The government has not forbidden use of forest lands and products
(especially trees) to villagers, but it has placed tighter restrictions upon
such use than existed heretofore. The villagers, however, feel that the
forest is their own land, and they refuse to make even a pretense at
asking permission to use that which they believe to be rightly theirs.
Anyway, the appropriate governmental representatives from whom
permission can be obtained, the forestry officer and the land records
officer, are not easily accessible and not always sympathetic, and the
necessary procedures are ponderously time-consuming and complex,
so that any desire to obey the law is discouraged at the outset. Local
caste and factional alignments are an obstacle because applications for
forest products have to be made through the local council. Therefore,
disobedience of public laws is accepted practice in this sphere, a fact
which may have facilitated its spread to other spheres.

The seriousness of public reaction to these regulations may be in-
ferred from the headline of a story in a newspaper of Dehra Dun
and Mussoorie in August, 1958: "congress men threaten direct
action, Forest Rights Issue Causes Unrest" (Himachal Times, 1958b).
The local political figures' threat came five years after the issue arose
and followed a state government decision, based on a subcommittee's
report, not to make any concessions on the matter. This issue has
been a major factor in support of demands, farther to the west, for a
separate hill state for the "suppressed and exploited" hill people
(Himachal Times, 1958a).

Defiance of these rules (as of the government in general) takes many
forms in the village. It is against the law to burn grass in the forest.
Throughout the lower Himalayas from time immemorial, the forest
grass has been burned every dry season just before the rains to make
a better crop of new grass after the rains. In this process fires occasion-
ally get out of hand. Occasionally a chan and the animals in it are
consumed in flames. However, villagers consider this a calculated and
not unduly heavy risk, inherent in the performance of a necessary
task. The government, apparently worried about forest trees (which

314 THE OUTSIDE WORLD

villagers insist are unaffected by the fires), has outlawed the practice
with notably unimpressive results. In June the lower Himalayan grass-
lands are still blackened and a thick pall of smoke obscures the
Himalayan view. The difference is that now the practice has become
a contest to outwit and frustrate the forest guard — a contest which
villagers invariably win. In 1958, when the forest guard came to fight
some fires around Sirkanda, he was greeted with complete apathy. He
asked for volunteers to help fight the fires and got no one to help.
He was told, "You get paid Rs. 60 per month to fight fires. Why should
we who earn nothing help you? Pay us Rs. 60 per month and we will
gladly join you. Otherwise, leave us alone."

Wherever the forest guard went, new fires would appear over the
ridge or just out of sight, usually with a few idly watching "observers"
to keep the flames away from valuable property such as chans or caches
of firewood. This occurred in spite of the fact that the forest guard
was relatively well liked as an individual and had the villagers' welfare
in mind. He felt compelled to report one particularly large and visible
fire in 1957, and asked villagers to make statements as to its origin.
All he wanted them to say was that it began by a public trail, probably
from the cigarette of an unknown traveler. This report would have
been satisfactory to the authorities and would have precluded further
investigation. Instead, villagers responded with statements to the effect,
"It is our forest, and we will burn it or not as we please." This would
have brought further investigation and a fine to the village had it
gone in the official report. Only by privately working out the report
with the village council president was a sufficiently innocuous state-
ment obtained and submitted to avoid further complications. Such
cases of defiance of forest laws are commonplace in Sirkanda and
throughout the hills. Recently there has been talk of assigning a forest
guard to Sirkanda, making it the center of a new forest district. This
plan is bitterly opposed by villagers. At present Sirkanda is at the
junction of three large forest districts under the jurisdiction of three
distant forest guards, a situation which facilitates evasion of the forest
laws. The boundaries actually meet in the settlement area of the vil-
lage, and the responsible officers are stationed in three widely separated
centers.

As has been stated earlier (chap. 2), nationalization of the forest
lands has made it illegal to cultivate new lands without special per-
mission. Villagers cite nationalization as an example of governmental
stupidity. "The government asked us to increase crop production for
the national welfare. We were happy to comply. However, at the same
time they made illegal the only means to accomplish this." The ruling

THE OUTSIDE WORLD 315

means that, as fields become depleted or inadequate for increasing
family size, new ones cannot be legally prepared without special per-
mission. An understanding village accountant (probably he understood
bribes) looked the other way when villagers disobeyed the law, but a
later accountant, either out of respect for the law or pique at the lack
of bribes or of sufficient bribes, reported the matter. Considerable
rancor was aroused before a reasonably happy solution was reached
in the land records office, where the offenders were assessed back taxes
and allowed to keep the land. This is not a solution for the future,
however. As in the use of forest products, notably trees, there is not
a rigid prohibition against use of new land, but there is a procedure
of application, payment of fees, and so on, which must be followed.
Villagers refuse to abide by these rules just as they do the forest laws.
Their attitude is not softened by the extreme difficulty they encounter,
the necessity for repeated trips to distant headquarters, the arrogant
and often greedy officials with whom they must deal in making such
applications. It is a task to frustrate anyone, most of all a semiliterate
villager. This point will be elaborated below as one element in a
pattern which villagers face in dealing with the government.

Another aspect of land legislation is that of taxation. Unfortunately,
when landlordism was abolished in the state in 1953, taxes were raised
slightly for Sirkanda landowners. Although the increase was slight
and lower taxes resulted for the majority of agriculturists in the
state, it was resented in Sirkanda, where all agriculturists had long
owned the land they tilled. The village comment is, "Congress prom-
ised to lower taxes and instead they raised them." Villagers are
totally unaware of the larger picture on this matter.

The indiscriminate application of general programs of land reform
that are inappropriate to the Pahari context has thus been an im-
portant factor in alienating these villagers from the government. In
itself it would probably have been sufficient to achieve this result.
However, it was not the only factor.

Agents of Authority

For villagers, one of the most frustrating aspects of their relation-
ship to the government is their contact with its official representatives,
and especially those who hold power over them. The Wisers have
discussed the problem in a plains village at some length (C. V. and
W. H. Wiser, 1951, pp. 130 ff.; cf. Beals, 1954, Newell, 1954). The
brutality and dishonesty of the police are proverbial in Sirkanda. The
maxim, "Never trust a policeman," is often quoted and religiously prac-
ticed. Specific instances of police brutality and dishonesty are legion,

gi6 THE OUTSIDE WORLD

ranging from simple demands for bribes upon threat of a false accu-
sation of illegal distilling to unreasonable violence in trying to extract
confessions or information (or, alternatively, heavy bribes) from people
who obviously had no connection with a case.

Similarly, the corruption of village accountants is proverbial. These
officials today are people totally unassociated with the village or even
the area. The accountant for most of Bhatbair in 1957-1958 had not
been seen by villagers since his appointment two years previously,
until suddenly word came to him that a higher official would be coming
on an inspection tour. The accountant appeared in great haste to
prepare records in two weeks that should have been kept over the past
two years. In the process his arrogance, incompetence, and inaccuracy,
combined with his corruption, put the villagers at a disadvantage and
aroused their resentment.

Such examples are not unique to the post-Independence era. They
carry on a long-established tradition. As many examples could be cited
under British rule.

The point, as the villagers see it, is that, compared to British times,
(1) there are now more occasions upon which a person is likely to
confront officials, and (2) there is now no ultimate authority from
which justice can be obtained, and therefore officials are arrogant
to an extreme never before seen. Whether or not the comparison is
accurate, it is believed by villagers.

To take these points in order, it is readily apparent that the amount
of official activity and consequently the number of agents of authority
affecting villagers has increased greatly since Independence with the
institution of land reforms, community development plans, and other
programs. The activities of rationing authorities may be considered
as an example. Villagers often use corrugated iron for roofing. To get
it they must apply through a rationing office whose officials, being in
league with black marketeers, never authorize enough of the material
to fulfill the villagers' needs. The official procedures are such that to
obtain any legal materials at all requires several trips to town. Most
villagers would agree with one who said:

By the time you have gone to Dehra Dun two or three times [a two-day
undertaking necessitating an overnight stay in strange surroundings] and
have been insulted by the officials and then get only one-fourth of the roofing
you need, you learn that it is better to go directly to the black market and
get what you need at the higher price. That is where you will have to go in
the end anyway.

This is just one of several newly created official hazards for villagers.
The biggest complaint of villagers against the present government

THE OUTSIDE WORLD 3 T7

is the lack of ultimate, impartial justice. As they frequently point out,
under the British the police were corrupt and the accountants were
corrupt, but the courts were fair. Examples are told of corrupt officials
of all types who were found guilty in the courts under the British,
and of false court cases exposed by the district magistrate. Officials
were restrained by the knowledge that this authority existed. Tax
collectors were often polite. Milk inspectors were often honest or at
least refrained from pressing their authority too far. Villagers contrast
that idealized (and perhaps unrealistic) picture with the realities of the
present. An example of immediate importance and tangible effect
was recounted by villagers as follows:

It has long been our custom to sell milk in the markets of Dehra Dun and
Mussoorie. The government sometimes inspects this milk now, as it did
under the British. Under the British, if we adulterated the milk with water
and were caught, as sometimes happened, we were fined and were denied
permission to sell for a specified period thereafter. This we knew and could
count upon. Now when we carry our milk to market we may be stopped by
the inspectors who, without even looking at our milk, accuse us of adulter-
ating it. We must pay them a bribe or go to court and be fined and penalized,
for what is the word of a milkman against that of an inspector. For this
reason most of us have quit selling milk since Independence. We now sell
instead a solidified milk product which is less profitable but which cannot
be easily adulterated and which is therefore not inspected.

The courts are held in utter contempt by most villagers. Examples
are given from this area of how even the most serious and flagrant
crimes or civil cases are decided on the basis of bribery. "No crime is
now so serious that money cannot win acquittal; no man is so innocent
that an enemy cannot put him in prison or win a judgment against
him if he has sufficient money for bribes." "Under the British there
was justice — water for water and milk for milk. Now water and milk
are one" (that is, lies were then treated as lies, truth as truth; now
they are not distinguished). It is felt that now even the highest official
is subject to bribes and that a poor villager has no one to whom to turn.

Comments from a range of informants of all castes indicate con-
sistent attitudes toward the government:

Before Independence India was ruled by the British; now it is ruled by
rascals. Law was then just, now anyone with money can avoid the law.

Under the British there was government by men; now it is government
by money.

In days of British rule India was a subject nation, but at least there was an
established law and order. Now India is like a woman with several husbands,
dominated by whichever is present at the moment.

9X8 THE OUTSIDE WORLD

Under the British we were free — we cultivated and worked as we pleased.
Now we are slaves. The government tells us what to do and what not to do.

The Congress promised to circulate gold money when freedom was won.
Instead they are circulating money of alloy that is not even pure silver or
copper. So it is with their promises.

The people, especially the leaders, wanted to get rid of the British. But
in fact, India under British rule was better than it has been since Inde-
pendence under the Congress government.

This government will be able to stay in power by force but not by the
people's will.

In this context of grass-roots rejection the government has attempted
to carry out various specific programs in Bhatbair.

Community Development

On October 2, 1952, the anniversary of Mahatma Gandhi's birthday,
India launched a program whose purpose was no less than a rural revolu-
tion. What it proposed was to transform the social and economic life and
outlook of the rural people, raise farm production and incomes, and create
from stagnant backward villages a vital, progressive rural community. It
proposed to do all this, not by coercion, but, as the "essence of the program,"
by self-help and participation of the people themselves. (Planning Commis-
sion, 1958, p. 168)

This was the Community Development Program. Its aim was to enable
people to raise their standard of living by introducing schools, com-
munity centers, pure water, better seeds, tools, and techniques of
acquiring a livelihood, better transportation, and better public health.
The key person in this program is the village level worker (VLW)
or Gram Sevak (village servant).

Living in one of a "circle" of the eight to ten villages under his care, the
worker goes from village to village, from farmer to farmer, using all the
techniques familiar to extension work in the west — field demonstrations,
individual talks and group discussions, audio-visual teaching, approach to
villagers with help on felt needs in order to awaken new needs and interest
in change.

Trained as a multi-purpose worker, he brings help and information on
improved methods of cultivation, on health care and sanitation, on cattle
diseases and their prevention, and so on. Working through village leaders,
he enlists the interest and participation of the village as a whole in change
and progress. (Planning Commission, 1958, p. 171)

In January, 1955, a Community Development Block was opened in
Eastern Dehra Dun District which brought a total of 88 village ad-
ministrative units (and a population of 62,000 people), including those

THE OUTSIDE WORLD 319

of Bhatbair, into the Community Development program. Shortly there-
after a VLW was stationed in Nagal, a large village in the valley seven
miles from Sirkanda. His area of responsibility was large, including
Bhatbair and a number of villages in the valley. In 1957 this Develop-
ment Block was changed from a National Extension Block to an In-
tensive Development Block, the shift being intended to increase de-
velopment activity. At this time additional VLW's were assigned. In
addition to the one at Nagal, who continued to have responsibility
for a number of valley villages and a few on the edge of Bhatbair, a
new VLW was assigned with Sirkanda as his headquarters and the
interior of Bhatbair as his area of responsibility. This man was in
Sirkanda a year, from the summer of 1957 until the summer of 1958.
Therefore, virtually all his official activities were observed by the
anthropologist. This period overlapped the third and fourth year of
the program in the area, the second year of India's second five-year plan.

Specific goals for Sirkanda in the second five-year plan were the
construction of a community center and repair of thirty miles of trails.
Other projects attempted were those mentioned in the council meeting
described above (chap. 8). In addition the VLW had a box of equip-
ment for first aid and equipment for inoculating and castrating
animals, and he was supposed to encourage the villagers in self-gov-
ernment, patriotic fervor, equalitarianism, and other aspects of the new
national creed.

At the end of his year in Sirkanda, when he finally achieved his goal
of securing a transfer, the VLW had achieved no tangible results in
any of his areas of responsibility in Bhatbair. He had, however, turned
in satisfactory reports. The proceedings of the only council meeting
held during that year are indicative of the programs undertaken and
the way villagers reacted to them. It is relevant here to summarize
briefly the results of the items taken up and decided upon in that
meeting: No one ever came to inspect the problem of the village's
water supply, and nothing was done about it. The panchayat tax was
not collected. No voluntary labor was performed on trails or on any
other project discussed in the meeting; the agreed-upon days for this
came and went without comment from the VLW or villagers. No
attempt was made to set up a night school (but this, unlike the other
items, had been formally rejected in council meeting). No agricultural
training session was held, although the VLW had been specifically
directed to hold a session in which he would explain to villagers the
Japanese method of rice cultivation. Despite considerable talk about
the matter on his part, no such session was ever held. The Village
Credit Cooperative Society of Sirkanda got no paid-up members as a

320 THE OUTSIDE WORLD

result of the efforts of the VLW, the council, or the society itself, al-
though later (as will be explained below) some token success was
achieved on this score. No further action or discussion about a com-
munity center took place. School attendance did rise, but because of
independent efforts of the teacher rather than council action. No
National Savings Certificates were sold. In addition, in the entire year
the interior of the VLW's first-aid kit and the veterinary equipment
were never exposed to the light of day. In short, the VLW was totally
ineffective in this village and throughout his area of responsibility.
As will be explained in the discussion below, this was a result of
several factors, including the training, motivation, and personal char-
acteristics of this VLW; villagers' attitudes toward the government,
toward change, and toward community cooperation; and inappro-
priateness of the program and of its manner of presentation in the
village.

Credit Cooperative Program

A second civil servant assigned to the village for approximately
the same period as the VLW was the economic cooperative supervisor,
locally known as the "supervisor sahib," and here termed ECS. His
job was to set up government-sponsored credit cooperatives in the
various villages of Bhatbair on the strength of a report submitted by
another officer who had previously visited the area and had secured
signatures in several of these villages indicating interest in the plan.
The idea was that villagers were to buy shares at Rs. 20 (payable over
two and one-half years), each of which would entitle the holder to
borrow Rs. 100 at 8.5 per cent annual interest. This plan met with no
success in any village in the area. In the words of the ECS:

All of these villages signed up with the agent saying that they wanted
cooperatives, but they probably did it just to get rid of him. There is no
real interest, and none of the people who signed up have received me
cordially. In no village has anyone paid any dues, and there is no prospect
of setting up cooperatives here.

In the council meeting at Sirkanda the necessary minimum of five
pledges was recorded, including the most unlikely member of all, a
landless blacksmith. A month later the ECS was notified that he was
to lose his job for not producing results. Upon receipt of this intel-
ligence, the teacher and he managed to round up nine people who
made payments of Rs. 2 to 10 each as initial membership fees in
response to a plea to save the job of the ECS. No mention of the
advantages of the society was made in this hurried fund drive. The

THE OUTSIDE WORLD 321

council president, who had been a steadfast foe of the plan, later
commented, "We started a cooperative society so the ECS wouldn't
lose his job." The factors relevant to the ineffectiveness of the ECS
are precisely parallel to those relevant to the case of the VLW.

Failure of the Programs

In analyzing the lack of success of these two men and the pro-
grams they were employed to advocate, it is obvious that social and
cultural factors — the relationship of the social organization and cul-
ture of the villagers to the programs presented — were crucial. It is
important, therefore, to consider both social and cultural factors and
the content of the programs presented. Also significant in the degree
of success or failure of the programs was the manner in which they
were presented and the personnel who presented them. These factors
will be discussed in that order.

SOCIAL AND CULTURAL FACTORS

Dube (1956, 1958) has discussed at some length social and cultural
factors in relation to change in Indian villages:

Agents of rural development projects and of programs of technical as-
sistance are confronted with these factors at almost every step of their work.
The acceptance of the agents of change as well as the effectiveness of the
media through which they endeavor to communicate their innovations are
largely governed by the cultural predispositions, attitudes, and social organi-
zation of the community in which they operate. The acceptance of the pro-
gram itself, or of its constituent parts, is determined to a considerable extent
by a variety of complex cultural factors, ranging from simple habits and
accepted social practices to the intricate patterns of belief, social structure,
world view, and values and attitudes. (Dube, 1956, pp. 19 f.)

Attitudes of suspicion and avoidance of outsiders seem to be char-
acteristic of the relatively closed society of Sirkanda, where familiar
ties of kinship, caste, and community are necessary if an individual
is to be accepted into the group or even to be dealt with on equal
terms. Lacking these, an outsider is an unknown and potentially dan-
gerous quantity best avoided.

The Wisers have paraphrased the attitude of plains villagers toward
change as follows:

To a new-comer we may seem suspicious, obstinate, intolerant, backward
— all that goes with refusal to change. We did not choose qualities for our-
selves. Experience forced them upon our fathers. . . . Refusal to change is
the armour with which we have learned to protect ourselves. . . . We are

^2 2 THE OUTSIDE WORLD

not blind to the advantages of the new, but unless we know just where it
will lead us, we prefer to let it pass us by. (C. V. and W. H. Wiser, 1951,

PP- 153 f0

In Sirkanda the unfamiliar, be it a person or a program of change,
is regarded with suspicion. The reasons are readily apparent. Contacts
with outsiders have been limited largely to contacts with policemen
and tax collectors — two of the most unpopular forms of life in the
Pahari taxonomy. Such officials are despised and feared, not only be-
cause they make trouble for villagers in the line of duty but also
because they extort bribes on the threat of causing further trouble
and often seem to take advantage of their official position to vent their
aggressions on these vulnerable people. Since India's independence,
governmental responsibilities have increased and extended to matters
previously ignored, such as closer supervision of enlarged government
forest lands and rationing of certain goods. The grounds for inter-
fering in village affairs have multiplied as the variety of officials has
proliferated. Any stranger, therefore, may be a government agent and
as such he is potentially troublesome and even dangerous.

Villagers' fears on this score are not groundless. Aside from the
unjust exploitation which such agents are reputed to employ, the
villagers themselves carry on many illegal or semilegal activities which
could be grounds for punishment and are easily used as an excuse
for extortion. In Sirkanda, government forest lands and products have
been illegally appropriated by villagers, taxable land has been under-
reported, liquor is brewed and sold illicitly, women have been il-
legally sold, guns have gone unlicensed, adulterated milk is sold to
outside merchants, marriages of children under legal age are per-
formed, men have fled the army or escaped from jail, and property
has been illegally acquired from fleeing Muslims at the time of parti-
tion. Any of these and similar real and imagined infractions may be
objects of a stranger's curiosity and therefore are reasons for discourag-
ing his presence in the village.

Paharis are thought by people of the plains to be ritually, spiritually,
and morally inferior. They are suspected of witchcraft and evil magic.
In addition, they are considered naive bumpkins; the hillbilly stereo-
type of other cultures is shared by Indians. Paharis try to avoid inter-
action with those who hold these stereotypes. Alien Brahmins may
seek to discredit their Pahari counterparts by finding evidence of their
unorthodoxy; alien traders may seek to relieve Paharis of their hard-
earned cash or produce by sharp business practices; scoundrels may
seek to waylay or abduct village women; thieves may come to steal
their worldly possessions; lawyers or their cohorts may seek evidence

THE OUTSIDE WORLD 323

for trumped-up legal proceedings which a poor Pahari could not hope
to counteract in court. Christians may hope to infringe on their reli-
gious beliefs and practices. Strangers are therefore suspected of having
ulterior motives even if they are not associated with the government.

The only way to feel sure that such dangers do not inhere in a
person is to know who he is, and to know this he must fit somewhere
into the known social system. Only then is he subject to effective local
controls so that if he transgresses, or betrays a trust, he can be brought
to account. The person who is beyond control is beyond trust and is
best hurried on his way.

To take a stranger's advice and change accepted practices would be
foolhardy. In view of past experience with the government, govern-
ment sanction of the advocate and his program merely serves to in-
crease the distrust of villagers.

In considering cultural factors which may be relevant to the con-
sistent failure of Sirkanda villagers to respond to programs of village
self-government and cooperative efforts toward self-improvement, cer-
tain similarities may be noted to the situation found by Banfield in
a village of southern Italy, ". . . the extreme poverty and backward-
ness of which is to be explained largely (but not entirely) by the in-
ability of the villagers to act together for their common good or, in-
deed, for any end transcending the immediate, material interest of the
nuclear family" (Banfield, 1958, p. 10).

This ethos is defined in terms of the hypothesis by Banfield that
the villagers act according to the following implicit rule: "Maximize
the material, short-run advantage of the nuclear family; assume that
all others will do likewise" (ibid., p. 85).

Adherence to this rule Banfield terms an ethos of "amoral familism."
He presents a list of seventeen "logical implications" of this rule (ibid.,
pp. 85-102). Many of these implications as well as other observations
by Banfield are reminiscent of findings reported here and in other
studies of Indian family and village life, with the qualification that in
India the relevant unit is the extended rather than the nuclear family.
There is a temptation to ascribe to this society a characteristic of
"amoral familism" on the extended family level. However, as Ban-
field points out, similar observations could be made about any society.

Amoral familism is a pattern or syndrome; a society exhibiting some of
the constituent elements of the syndrome is decisively different from one
exhibiting all of them together. Moreover, the matter is one of degree: no
matter how selfish or unscrupulous most of its members may be, a society is
not amorally individualistic (or familistic) if there is somewhere a significant
element of public spiritedness or even of "enlightened" self interest. (Ban-
field, 1958, pp. 11 f.)

324 THE OUTSIDE WORLD

In view of this rigorous requirement, Sirkanda villagers are ex-
cluded from "amoral familism" even on the extended family level.
"Amoral factionalism," "amoral casteism," and so on, are also ex-
cluded; there are too many occasions upon which behavior is incon-
sistent with the rule of total commitment to the short-run material
advantage of these groups. The symptoms do not add up to the syn-
drome. Perhaps elsewhere in India there are groups to which the syn-
drome does apply — or more nearly so. This does not mean that Ban-
field's conception is useless for an understanding of social behavior
in Sirkanda or other Indian villages. Its real usefulness is that it points
up the "moral basis" for characteristic social behavior. That is, there
may be a cultural norm or world view, an ethos, a pattern of values
and attitudes, which predisposes the members of a society against
combined action for long-range common goals. Similarly there may be
cultural bases for rejection of change or of certain kinds of change.
Certainly in Sirkanda there is enough "amoral" preoccupation with,
and loyalty to, the extended family that it is a serious obstacle to
community action and leads to most of the ramifications which Ban-
field has listed as "logical implications" of amoral familism.

In Sirkanda this is a relative matter. In interlineage relations the
lineage becomes the focus of self-interest; in inter-sib relations it is
the sib; in intercaste relations it is the caste. In intervillage matters
where caste loyalties are not at issue, there is similar preoccupation
with one's own community, as witness the Sirkanda council's refusal
to volunteer work on trails to be used by other villages. It might be
argued that all these manifestations of group self-interest could be
reduced to "familism," since the family is the unit with which the
individual is most closely identified and of which he is thinking when
he takes a stand on issues at any of these levels. That is, he may take
his stand on the basis of the best interests of the family which, in
certain circumstances, happen to coincide with sib, caste, or commu-
nity interests.
* i /

'Amoral" self-interest seems especially likely to function at the level
of the social unit which acts as an entity in facing insecurity. In any
event, culturally sanctioned self-interest on family and other levels is
a matter which might usefully be considered along with other cultural
factors by anyone attempting to influence people to implement action
programs in India.

One aspect of this feature of community life in Sirkanda is the
absence of an effective tradition of community cooperation in any
but very specific contexts. Another is the absence of village-wide ac-
ceptance of formal leadership. Everyone claims to make his own

THE OUTSIDE WORLD 325

decisions, and those who advocate their viewpoints for others are
viewed with suspicion. Neither is there a precedent for democratic
village government.

Opler has described perhaps the most crucial obstacle to community
development programs as "the social organizational difficulty of ex-
pecting a social structure which was essentially fluid, diffuse and con-
servative to implement programs which demand decision, dispatch,
and an experimental frame of mind" (Opler, i960, p. 197).

The controlling group in Sirkanda has always been the high-caste
landowners, and they remain so today. Although they are not a co-
ordinated body on many matters, they are coordinated in their relation-
ship to the low castes, whom they control in almost every sphere. This
leads to one of the most perplexing problems in community develop-
ment in Sirkanda, the conflicting interests of high and low castes.

Sirkanda has a sharply segmented society with important privileges
granted to high castes and withheld from low castes. High castes
naturally have a heavy stake in maintaining the status quo. The
Congress party and the government of India have proclaimed an
equalitarian, anticaste ideal. To Sirkanda villagers this is one im-
portant characteristic of "government" and of "Congress." This alien-
ates high-caste people from the government and its representatives,
who are assumed to be dangerous radicals who threaten the traditional
system. They feel that ultimately the government will force them to
associate with their caste inferiors and will help low castes to in-
dependence, prosperity, and arrogance, possibly at high-caste expense
and certainly to their detriment. This is an important and explicit
reason why high castes refuse to cooperate with government people
and programs in Sirkanda.

Low-caste people have been hopeful of improved status and liveli-
hood on exactly these grounds. They hope to benefit from the govern-
ment's attitudes and programs. The Community Development Pro-
gram, as it has functioned thus far, has been designed primarily to
benefit agriculturists — the high-caste, well-off landowners. This is to
be expected because, as Lewis (1959, p. 536) has pointed out, the main
aim of the program has been "increases in production rather than
social justice" (cf. Dube, 1958, pp. 82 f.). This has become apparent to
low-caste Sirkanda villagers, who note that the VLW does not talk to
them or consider their problems and obviously has no interest in aid-
ing them. He merely tries to help those who already are well-off and
in control. Low castes have not benefited from the program, and there
has been no action taken to implement the government's equalitarian
pronouncements. Therefore, low-caste people are disillusioned about

326 THE OUTSIDE WORLD

the government's stated interest in them and their welfare. As a result
they are alienated from the Community Development Program and
other governmental activities in the village.

High-caste people are antagonistic to the government, partly be-
cause of its alleged interest in the equalitarian ideal, which they feel
will ultimately result in active championing of low-caste grievances.
The government's clumsy attempts to increase agricultural production
in the village have not succeeded in counteracting this antagonism,
and at the same time have contributed to alienation of the low castes.

Thus, low castes want their positions improved; high castes want
their positions maintained by suppression of the low castes. The gov-
ernment has alienated the low castes by their actions, and the high
castes by their words. This is a dilemma that has not received explicit
recognition but which is very real in Sirkanda and has no readily
apparent solution without a choice between production and social
justice.

Other reasons for the failure of programs, and for the general atti-
tude toward the government in Sirkanda, are to be found in the nature
of the programs, their manner of presentation, and the personnel who
have presented them. Since these are matters directly amenable to
action by those responsible for the implementation of changes, it is
appropriate to consider them here.

THE PROGRAMS

In general, the programs have ignored the traditional orientations
of villagers and have not been in line with the desires or "felt needs"
of the people.

Dumont and Pocock (1957, P- 19> c^- Tinker, 1959) have attributed
the lack of success of village government throughout Uttar Pradesh
to lack of understanding of the culture of those being governed:

They have tried to transfer the idea of the assembly or pancdyat from a
caste-group to the multi-caste village. The result, according to our own ob-
servations confirmed by reports from elsewhere (all in U.P.) is a standstill
since the enterprise comes up against the total disinterest of the elected
judges. The institution of the caste pancdyat rests on the solidarity of the
caste-group, which is highly sensitive to certain kinds of offenses, while the
assumed solidarity of the village is simply nonexistent at that level.

This criticism would seem to be applicable in Sirkanda to a large
extent, although it would have to be modified in view of the essentially
one-caste nature of the Sirkanda council today and the traditional
dominance of that same caste in the village.

THE OUTSIDE WORLD 327

Specific programs in Sirkanda have consistently ignored the wants
of villagers themselves. Of those presented in 1957-1958, only the
water supply project was felt by villagers to be really worthwhile.
An earlier attempt to improve that situation had ended in an expensive
fiasco at the hands of an unscrupulous contractor, so there was great
reluctance to try anything new. The VLW was unable to offer good
advice on the matter and no technically qualified person who could
have done so was made available, although the council requested such
advice. The community center evoked interest but no general agree-
ment. Other programs informally introduced or advocated by the
VLW were equally unsuccessful. Improved seed, seed potatoes, and
chemical fertilizer, which the previous VLW had attempted to intro-
duce, were rejected by villagers largely because local resources were
felt to be sufficient. Adequate natural fertilizer and local seed were
available, and there was a general conviction among villagers that
seeds, fertilizer, and agricultural techniques which were developed or
tested in the plains would be unsuccessful in the hills. Moreover, they
were difficult to obtain. They had to be purchased in a distant village
of the valley, and by the time they were transported to the hills they
were exorbitantly expensive. The government required that the seed
be returned with 25 per cent interest and that it be carefully cleaned,
requirements which discouraged villagers from its use. Seed potatoes
were requested by one villager, but the minimum allotment was more
than twice what he could use, and so the idea was dropped.

In the case of the ECS, there was no desire among villagers for the
benefits a cooperative credit society could offer. People were accus-
tomed to borrowing from one another or to obtaining credit from a
local shop when necessary. They had no desire to risk a new system,
especially one which they did not understand at all.

Thus, a general lack of appropriate programs was an important
factor in the failure of the VLW and the ECS. It is likely that, had
the VLW been able to present the people with a useful solution to the
problem of their distant water supply, he would have been well on
the way to winning their confidence. Had he practiced the first aid and
simple medicine (including veterinary medicine) supposedly at his
command, he certainly would have won favor. Had he been sensitive
to the desires of villagers, he would have increased his chances of
establishing rapport and ultimately of achieving other goals. Among
young men there was a real desire for organized recreation in the
village; interest had been expressed in volleyball (introduced briefly
by a former teacher) and evening singing and music sessions. The
VLW ignored these interests — they were not in his program.

328 THE OUTSIDE WORLD

He was bound by policy to specific programs which did not lead to
the general goals of winning the confidence of the people so that they
might be helped to help themselves. What was needed was not rigid
application of programs, but study of local conditions — cultural,
ecological, agricultural — so that appropriate programs might have
been formulated to satisfy some of the immediate wants of villagers.
In this way their confidence and cooperation might have been won
so that the improved standard of living which is the ultimate goal of
community development could be obtained. The overt goal of helping
or serving people was in practice subordinated, both by the VLW and
by higher administrative levels, to the covert goal of getting minimal
compliance, on paper, with official programs. As a result, the entire
project was an empty facade. In administrative headquarters this
area had achieved a record as a successful unit. In reality it was a
failure.

Had specific local programs been formulated in terms of general
goals of community development and had accomplishment of covert
goals not been substituted for fulfillment of overt goals, success might
have been achieved.

MANNER OF PRESENTATION

Even the programs which might have been successful were not
presented effectively or even comprehensibly to villagers. There was
no effort to discuss planned programs informally in advance with
household heads or others nor to invite discussion in their formulation.
The programs were invariably presented to the village president, who
then relayed the information to villagers as being the VLW's plan,
or they were presented through the village council without advance
warning. The villagers had no chance to think over a project in advance,
much less to ask questions, make suggestions, or suggest alternatives. The
VLW is known by the title "village servant," but his projects were
presented as governmentally decreed programs to be carried out by
villagers under his supervision. Villagers were not led to feel that
they were being helped to help themselves, but rather that they were
being directed to meet the whims of a government which did not under-
stand their problems and with which they were already at odds. As a
result their resentment was compounded.

The government repeatedly demonstrated its ignorance of, and in-
difference to, the problems of the villagers by advocating programs
with no apparent local relevance (chemical fertilizer, Japanese rice
cultivation) and by sending as its representatives personnel with no
familiarity with the way of life of people in the hills. Simple misunder-

THE OUTSIDE WORLD 329

standings were also frequent, based partly on the inability of the VLW
and ECS to speak or understand the Pahari language, and partly on
these officials' failure to explain their plans adequately. Thus, villagers
believed that their investment in a credit cooperative would be
"swallowed up" by the officials of some government agency and that
it could not be refunded. They were totally unaware of some of the
services which a VLW was supposed to be able to provide, especially
first aid and medicine.

PERSONNEL

The VLW and ECS were poorly trained and poorly motivated.
They had no special preparation for work in these rugged and isolated
hills. They were therefore thought of by villagers as complete out-
siders and greenhorns. Villagers ridiculed their inability to get around
easily on the rough mountain trails. When one of these officials made
his first appearance in the village, he was guided part of the way
by a small village boy. Villagers never forgot the boy's tale of how
he had to encourage the man onward at narrow points on the trail and
how the man inched forward on all fours, asking God's mercy as he
went. These officials constantly complained of the unpleasant sur-
roundings, the difficult terrain, and their own loneliness and boredom.
Their meager salaries did not compensate for these hardships. The hill
environment did not provide them with the rewards to which they
felt their education and official position entitled them. They had no
intense commitment or emotional involvement in their jobs, as did
the schoolteacher. They were devoted to their status rather than to
their responsibilities. They did not receive the prestige, public ac-
ceptance, or personal satisfaction that came to the teacher in his
well-known and widely respected profession. As a result they were per-
sonally ineffective workers as well as advocates of inappropriate pro-
grams.

The ECS has been assigned to the village (perhaps as punishment)
after having failed the training program for the post. Both the men
spent much of their energies trying to effect a transfer to more favor-
able circumstances. Worst of all, from the point of view of their
reception by villagers, was their lack of understanding of or interest
in the people with whom they worked. They neither understood nor
sympathized with the way of life and problems of the Paharis. They
considered the villagers dirty, ignorant, and immoral, not capable of
raising their standards and unworthy of close association. Their lack
of knowledge of the Pahari language contributed to this state of affairs.
They could not even communicate easily with villagers. As a result,

330 THE OUTSIDE WORLD

neither of the officials talked informally with villagers or came to know,
even superficially, any of them other than the shopkeeper, the council
president, and one or two men who frequented the school building
where these men lived. Neither of the officials was ever seen to enter
the village proper nor to go to the fields where villagers worked, in
the year that they lived on its outskirts. They won neither the respect
nor the trust of villagers. Although the villagers were not hospitable,
it might be expected that as agents of change these officials would
have made a real effort to win their confidence and understand their
outlook. Instead, they relied entirely upon their official positions in
their attempts to initiate and carry out programs. Gandhi has com-
mented on the fruitlessness of this approach:

When an official becomes a reformer, he must realize that his official posi-
tion is not a help but a hindrance. In spite of his Herculean efforts people will
suspect him and his motives, and they will scent danger where there is none.
And when they do certain things, they often do them more to please the
official than to please themselves. (Gandhi, 1952, p. 103)

Education: A Successful Program

The third government employee in Sirkanda is the schoolteacher.
Education is not new to Sirkanda villagers, but until 1950 they had
to go out of Sirkanda to get it. Usually they went to schools in one
of two or three villages of the valley not far from Bhatbair. Even now
anyone desiring education beyond the fifth class must go to those vil-
lages, and anyone seeking education beyond the eighth class must go
into Dehra Dun. People as old as 60 years acquired some education in
their childhood. Three men in their late forties have had an eighth-
grade education. There are 24 local students currently enrolled in or
attending classes regularly in Sirkanda, and three attend higher classes
outside Sirkanda. In addition to these students, 36 villagers have at-
tended a village or outside school in the past — 33 Rajputs, 2 Brahmins,
and 1 blacksmith. Thirty-three of these are men and three are girls.
The girls attended school in the village after 1950. In addition, one
Brahmin man received some religious training in another hill village.
The average period of attendance of these 36 was three years; five
went to the eighth class, and four completed five years. It is obvious
that this amount of schooling cannot make for a high rate of effective
literacy. While all who have gone to school and a few who have not
can sign their names, many can read or write little if at all. Little
reading or writing is done in the village except in the school, and
so literacy is quickly lost by those who acquire it. At best perhaps half
of those who have been to school are usefully literate, making a literacy

THE OUTSIDE WORLD 33 1

rate among village men of less than 15 per cent and that for women
around 1 per cent — a village rate among adults of less than 8 per cent.

Attitudes toward education in Sirkanda are generally not favorable.
School is thought of as a place for children with nothing better to do.
In most families only children who are not needed elsewhere are
allowed to go. Most families consider education for girls to be en-
tirely useless and actually detrimental. As a result of these attitudes,
a maximum of 20 out of 35 eligible boys in Sirkanda attend the local
elementary school while 4 of 31 girls do so. Most children are in agree-
ment with their parents on this issue, but at least one small girl at-
tended school sporadically despite her father's refusal to enroll her,
and one boy requested private tutoring when his father demanded that
he work during school hours. Often one or two boys of a large family
are educated and others are not. The educated one is usually one who
shows some liking or aptitude for school or, perhaps more often, who
dislikes other work or is unable to perform it.

One of the most successful educated men of Bhatbair is a man of
another village who began school at the age of 14 after an attack of
pneumonia which left him weak and sickly. His father, who had
previously forbidden education in his family, saw the boy's weak con-
dition and said, "Now he's not good for work, he may as well sit in
the school each day." Despite his late start this boy went on through
high school, got a job on the office staff of a limestone quarrying com-
pany and quickly advanced to an excellent position. Now he is edu-
cating a younger sister although his parents insist that it is a waste to
do so.

Family size and land ownership are important factors. A small
family, especially if it has much land, cannot afford to let boys go to
school who might otherwise be helping to farm. A family's wealth is
a primary factor, especially in higher education. Sometimes a wealthy
family can afford to hire a servant to do a boy's work so that he can
go to school if he wishes. Education outside the village is expensive
because the student must board. The only Bhatbair boy to go to college
is a Brahmin from the wealthiest family of Kanda, a neighboring
village. All those in Sirkanda who have gone outside the village for
schooling have been from the wealthiest families, and some had rela-
tives near the valley with whom they could live and from whose homes
they could commute to school.

Establishing schools in Bhatbair was not an easy task. Three one-
teacher schools were authorized in 1949, one in each of the three
largest villages, with a potential enrollment of around 50 students
each. Teachers were assigned by the Dehra Dun District Board in

332 THE OUTSIDE WORLD

1950. They were all well-trained, young but experienced, apparently
quite highly motivated, and, most important, were from other hill
areas so that the living conditions, terrain, language, and culture of
the area were not strange to them. In each village where they have
established schools they have come to know villagers on a personal
level and have to a considerable extent been accepted by them. How-
ever, in the initial stages they were beset by many difficulties. In one
of the three villages originally picked for a school (and it is a village
with a long-standing reputation for inhospitality), the teacher was
unable to get a single pupil to attend school. After several months he
was authorized to move to a village with one-sixth of the potential
student body, but one more favorably disposed toward education. It
is a village made up entirely of relatives of the wealthy Brahmin who
serves as the traditional Sirkanda priest. There the teacher has run
a school on the veranda of a house for six years, with an average of
five pupils in attendance. The other teachers did somewhat better in
Kanda and Sirkanda where eight to fifteen students have attended the
schools over this period.

In 1957, when a new teacher came to Sirkanda, he found an attend-
ing student body of 8 students and an official register of over 20. He
found the standard of achievement extremely low. Being a conscien-
tious person, he set about to rectify both matters and he met with
remarkable success. On the point of attendance, he was faced with
three alternatives: (1) continue to falsify the attendance record as his
predecessor evidently had, under pressure from the District Board to
increase enrollment, though this involved the risk of detection and
punishment; (2) report the truth of the matter, thus provoking an
investigation that would hurt either the previous teacher, himself, or
both; (3) raise the enrollment to match the records, a seemingly hope-
less task. This teacher elected to follow the last course and in a period
of three months he had, by begging, cajoling, and threatening, man-
aged to raise his regular daily attendance from 8 to around 20. He
received no credit for this beyond his own satisfaction and peace of
mind. No one in the village cared, and the increase in attendance did
not show on his records to higher officials.

As to achievement, the teacher undertook to raise standards by
intensive effort, discipline, and some tutoring. His success is indicated
by the fact that the only candidate for the fifth-class examinations, an
exceptionally backward pupil at the beginning of his fifth year, passed
the district examinations in the upper third of the group examined.
For this, too, the teacher received no credit from villagers, and his
efforts were unknown to the school authorities.

THE OUTSIDE WORLD 333

The teacher's efforts were not unappreciated by his students. Of 31
students, roughly 20 of whom were in attendance on any particular
day, there was a core of about 10 who attended consistently. That is,
10 students came regularly, while of the remaining 21 an average of
about 10 appeared daily. The consistent attendees appeared to be
sincerely interested in school and respectful and appreciative of the
teacher. Several of the other students were equally interested but
lacked the support of their families which would have enabled them
to attend regularly. It was the interest displayed by these pupils
which inspired the teacher in his largely thankless task. He was espe-
cially pleased when one of his students, on his own initiative, took
English lessons from the anthropologist's interpreter for a period of
three or four months. However, neither the schoolteacher nor any
villager knew English, and the boy finally gave up the effort.

The school year in Sirkanda consisted of about a 5-hour day of
classwork, 5% days a week for approximately 10 months per year. The
curriculum was similar to that in elementary schools throughout
Uttar Pradesh, consisting of exercises in reading, writing, and arith-
metic and occasional study of stories, religion, history, and geography,
all by rote. There was no attempt to relate school studies to practical
problems of village life.

Official school enrollment in Sirkanda was 25 at its maximum, and
with 6 unenrolled attendees, attendance remained consistently at
around 20 despite daily absentees, except in the harvest periods and
heavy rains, when it dropped to around 10. Seven of these pupils (2
Bajgis, 1 Muslim, and 4 Brahmins) were boys from two other villages
about three miles distant, who came daily over an extremely rough trail.
Twenty-one enrolled and unenrolled Sirkanda students were Rajputs,
2 were blacksmiths and 1 a Brahmin, and only 4 (all Rajputs) were
girls. Grade distribution reflects both the drop-out rate and the suc-
cessful recruitment of beginners by the new teacher: first grade, 1 1
enrolled and 6 in informal attendance; second grade, 5 enrolled; third
grade, 4 enrolled; fourth grade, 3 enrolled; fifth grade, 1 enrolled.
Only one girl attended at all regularly.

Three Sirkanda boys were enrolled in high schools in 1957-1958,
one each in the seventh and eighth grades in Majra, a village border-
ing the valley, and one in high school (ninth grade) in Dehra Dun.
At the end of the year, one Sirkanda student went on to the sixth
class at Majra and the eighth-grader there went on to high school in
Dehra Dun.

It seems likely that when a high school is established in Majra,
where the sixth, seventh, and eighth classes are currently held, more

334 THE OUTSIDE WORLD

Sirkanda boys will pursue their education. Such a school is under
construction and is expected to be opened in a year or two. There,
as villagers point out, the boys can live with relatives and attend
school relatively inexpensively, without being exposed to the tempta-
tions of city life.

The school, then, is a government enterprise that has fared better
in Sirkanda and neighboring villages than have other programs. The
reasons for this success are several. The school is an institution to
which villagers have become accustomed over a long period of time,
first outside the village and more recently in the village. Its potential
value in achieving tangible results in the form of jobs and prestige
is obvious to all, though many hold that these results are not suitable
or attainable for their children. The teacher's role is a familiar one
to villagers, and his purposes and functions, although not always ap-
proved, are at least understood. He may be a bother and a distracting
influence to children who could be better engaged elsewhere, but at
least he does not pose a threat to adult villagers nor is he suspected of
ulterior motives. He is not, in the conventional sense, an agent of
authority. In these respects he is different from other government
employees.

Teachers have been consistently effective as individuals in this area.
They have been relatively well trained in teacher training schools.
They are self-selected, doing what they planned to do in a well-estab-
lished profession with a secure future. They have been intelligently
assigned, with their own cultural background in mind. Indicative of
their appropriateness for their jobs is the length of their tenure in
these jobs despite isolation and frustrating working conditions: 6,
6, and 4 years for the three recent teachers, and the replacement for
the last of these is beginning his second year without yet seeking
transfer. They have consistently established fairly good rapport with
villagers, considering the obstacles which any outsider faces in the
closed society of the area. This did not come easily. One or two years
were required for each to gain acceptance, but from the point of view
of the program the time has probably been well spent. Rapid turn-
over of personnel would have precluded the possibility of achieving
such rapport.

Some of the important factors which have led to failure of other
programs in the village have therefore not characterized the educa-
tional program. The goals have been relatively consistent. In the con-
text of Indian or at least district educational policy covert goals have
been the same as overt goals (the principal disparity being in the
emphasis upon enrollment versus actual attendance), and general goals

THE OUTSIDE WORLD 335

(teaching children facts and skills as outlined in the curriculum) have
been in harmony with specific goals (enabling children to pass the
examinations, or at least to fail them only after a noble try).

In spite of its relative success, education has not been enthusiastically
received in Sirkanda. The teacher's job is a frustrating and thankless
one. This can in part be attributed to villagers' suspicion of outsiders;
in part to antigovernment sentiments; in part to lack of enthusiasm
about education. The last-mentioned factor is conditioned to a large
extent by the lack of tangible benefits to those who have been educated
in the past. To many villagers education is simply irrelevant in tradi-
tional village life, the only way of life accessible to these people. Vil-
lage education is not conducive to orientation toward the outside, nor
sufficient to enable people to make a success of life on the outside, nor
is it of any great practical value in the village. Its demonstrated value
in Sirkanda is only that it brings to the villager the advantages of a
limited literacy (which villagers are coming to appreciate in the light
of increasing necessity to deal with the government) and some general
awareness of geography, history, and nationalism. The villager is
presumably thus helped to become a more responsible citizen in a
republic — an official goal of education of which villagers are unaware.

However, to the extent that education has been successful and is
becoming more so, we have presumptive evidence that inherent cul-
tural factors do not preclude success in governmental programs. Care-
fully planned programs presented intelligently by properly trained
and motivated people can be successful in the village. If they have
the further advantage of fulfilling a "felt need" and bringing a tangible
benefit, their success is even more likely. With initial hard-earned
successes, subsequent accomplishments might be expected to come
more easily as traditional suspicions are allayed.

Balance of External Relations

Relations between Sirkanda and the outside can be summarized
by saying that on the whole, intrusions from the outside into the
village have far outweighed, in numbers and effect, the extensions of
the village into the outside world. That is, while villagers do go out
of the hill area for trade and pilgrimage, and village boys and young
men not infrequently seek adventure in other areas, the excursions
are short-lived and usually result in greater appreciation of the village
and its familiar way of life. Those who have been outside bring back
stories of their adventures and certain material goods obtained there,
but they do not often bring lasting changes to the village. On the other

336 THE OUTSIDE WORLD

hand, government activity in the village, and to a lesser extent outside
merchants and others, have brought into the village a number of alien
ideas, practices, and material goods. Although the activity of outsiders
has recently increased in effect, it is not a recent thing. Under the
British and before that under the Garhwal Raja, outside government
impinged on Sirkanda, and traders have long frequented the village.
And for fifty years schooling has been acquired by some village children.

Within the more restricted area of the "marriage network" sur-
rounding Sirkanda — the immediate culture area — but not extending to
"outside" or urban areas, it is evident that extensions of the village
are many and pervasive, as with the plains village, Senapur, of which
Opler (1956, p. 10) writes, "the involvement of . . . villagers with
organizations, places, and events outside of the village is considerable
and it seems that this has been the case for a very long time."

The fact that this village, like communities everywhere, has such
relations and extensions does not make it any less a functional unit.
This fact merely puts it into perspective as one unit within a larger
complex of systems on several different levels.

In discussing the problem of extensions of village communities to
the extra-community world, R. J. Smith (1956, p. 4) has said:

There are three types of extensions:

1. Intellectual awareness

2. Physical mobility

3. Organizational affiliation (or membership).

These are not, of course, water-tight compartments, and it is unlikely that
any one activity or extension will be entirely one or the other of these.

It is evident from the discussion above that all three of these types
are found, in varying degrees, in the relationship of Sirkanda to the
outside world. However, a fourth type of extension was of interest
in this research and appears to be crucial to community cohesion and
relevant to cultural change. This is individual or group identification
with an outside group. That is, in addition to awareness of, mobility to,
or organizational affiliation with, an outside group, there is also the
possibility of adoption of that group as an identification group or, as
it is commonly known in the sociological literature, a reference group
(Newcomb, 1950, pp. 225 f.; Turner, 1956, p. 328).

One of the aims of the research reported here was to investigate fac-
tors relevant to social cohesion and disintegration in the community.

Unfortunately, comparative data from a culturally similar com-
munity is not at hand. There is, however, some literature of a theoret-
ical nature on the subject, and the writer has made comparable ob-

THE OUTSIDE WORLD 337

servations in a community in the Aleutian Islands of North America
which provide some useful contrasts (Berreman, 1955). In the Aleutian
village, it was found that individuals tended to identify consistently
with an alien group, non-Aleut participants in the urban culture of
the mainland. This attitude was disintegrative to the community be-
cause the norms of the alien group could not be realized in the con-
text of the isolated community. People had to leave the community
to achieve any chance of acceptance by members of the group providing
the frames of reference which influence their attitudes and behavior,
or even to be able to practice this behavior and satisfy their aspirations.
The situation was attributable largely to the influx into the community
of alien agents of socialization, who represented the source of authority
and of rewards in the community and who instilled in its members
values which were not realizable in the community context. As a result,
people began to look outside the community for their standards of
behavior and achievement. Unable to meet these standards in the
village they left, or remained as frustrated "marginal men." Their
plight was evident in the high emigration rate and the many types of
stress experienced by individuals in the village, including stress of
unprecedented kinds for which no adaptive mechanisms were readily
available.

On the basis of that example it was hypothesized that "persistently
attempted projection to an . . . alien context, if it cannot be achieved
by community members within their community, is disintegrative to
that community" (Berreman, 1955, p. 58). That is, to the extent that
members of the community adopt values which cannot be realized in
the community but can be realized outside the community, that
community is likely to disintegrate. Such values are most often ac-
quired from alien reference groups — groups which are the sources of
authority and rewards. Such groups are especially potent when they
are in key positions as agents of socialization of children (for ex-
ample, teachers), as they were in the Aleutian community. To state it
differently, a high positive correlation exists between the extent to
which individuals' reference groups are also their membership groups
and the integration of the community as evidenced by other criteria
of integration.

In attempting to counteract disintegrative trends in a community,
success is as likely to be achieved (and probably more realistically so)
by making new values achievable within the community as by trying
to prevent their acquisition or by trying to reestablish traditional
values.

Sirkanda offers a notable contrast to the Aleutian village in its

338 THE OUTSIDE WORLD

state of cohesion or integration as a community. In Sirkanda there is
little in the way of adoption of alien values or reference groups aside
from the ritual sphere, where plains standards are adopted as a means
of social mobility. Socialization remains primarily in the hands of
villagers. Schooling within the village has been relatively innocuous
from this point of view. It has reached relatively few villagers with
any significant intensity. Moreover, teachers have been of the same
general Pahari cultural background as the villagers. The curriculum
has not taught an alien way of life, nor has it conflicted in many
respects with traditional values of villagers. Village schooling (unlike
that obtained in Dehra Dun) has not turned students' aspirations to
the outside, nor has it lessened their effectiveness as villagers. This
situation is in sharp contrast to that in the Aleutian village (Berreman,

1955» PP- 54 £•)■

In Sirkanda there has as yet been no significant interest in move-
ment to the outside, no significant dissatisfaction with village life.
There are few signs of stress in individuals or groups that are not well
within the traditional and expectable range and are therefore dealt
with effectively in traditional ways. The village is, within the tradi-
tional cultural context, which includes caste and factional divisions,
cohesive and well-integrated by most indexes that could be set forth.
This does not mean that it will remain so. The two recent emigrants
to city jobs may be straws in the new wind which is blowing in India,
generated by increasing awareness of other ways of living and resulting,
as with these two men, in identification with another way of life and
attempts to become part of it.

Traditional ties are strong in India. Even villages near large cities
remain remarkably cohesive, and people who go to cities to work
generally retain close ties in their home villages. Pahari villages very
close to Dehra Dun are tied culturally, socially, and religiously much
more closely to the hills than to the city. Urban emigration in them
is virtually unknown. Attitudes and behavior of their residents seem
to differ surprisingly little from those of their fellows in more isolated
villages. It is a question to what extent education may alter this situ-
ation. So far, the evidence in Sirkanda does not contradict the hypoth-
esis regarding community cohesion derived in the Aleutian village.
These are two extreme cases which follow the expected pattern. Further
work in less clear-cut circumstances is needed for more substantial
verification.

CONCLUSION

The function and relative significance of ties of kinship, caste, and
community in the lives of people in and around Sirkanda have been
a major focus of this research. Implicit in this focus is a belief that
the findings will be relevant to an understanding of other villages
not only within this culture area, but throughout Northern India
and to a lesser extent in India at large.

In the five chapters immediately preceding this one it has been
demonstrated that each of these levels of organization is vital in the
lives of Sirkanda villagers. They are the structural framework for
Pahari culture and social interaction. They come into prominence in
varying degrees in various situations. Of the three, kinship ties are
of most immediate significance in the lives of individuals. The patri-
local extended family is the residential unit, the property-owning and
work group, the group that finds wives for its sons, that participates
most actively in the life-cycle rites of its members, that worships com-
mon gods together, and that applies safeguards and sanctions to its
members to keep the family reputation untarnished. It is, in short,
the basic economic, social, and religious unit. Above it, the lineage,
clan, sib, and kindred are kin groups which function in many contexts
as social and religious units of progressively less relevance to the in-
dividual. Informal organization within the caste and community tends
to follow lines of kinship. Those kin ties are strongest which occur
within the community.

The caste or jati is theoretically an extension of the kin group,

34o CONCLUSION

since its members are supposed to be descended from a common an-
cestor. Their kin relationship is, however, sufficiently remote that it
assumes little practical significance beyond that basic to caste itself,
namely the ritual and social equivalence of its members. Pahari caste
is the hereditary, endogamous unit, ranked with regard to other such
units in terms of ritual and social status. It is associated with a tradi-
tional occupation or range of occupations. Caste is significant in its
regulation of marriage and other social and ritual contacts, in its in-
fluence on the religious and economic activities of its members, and
as an effective identification group for its members. It is an important
tie which extends throughout the immediate culture area across com-
munity boundaries and transcends community loyalties.

The village community is made up of people identified with a par-
ticular nucleated settlement area surrounded by agricultural lands
cultivated primarily by its high-caste residents. The village cannot be
understood without reference to its extensions in surrounding villages,
especially those which can be reached in half a day and with which
its members have most of their marriage and other extravillage social,
religious, and economic ties. However in most circumstances the village
is a more important identification group than is the larger locality.
People feel loyalty to it, keep its secrets, and rally to its support when
confronted by outsiders. It is the place where property is held, where a
living can be made, where one's primary emotional attachments are
focused, where one feels secure at times of crisis and happy at times
of rejoicing. Disputes among villagers are most often resolved within
the community, usually by high-caste pressure or arbitration. Trails,
water supply, and uncultivated lands are held by the village in com-
mon. It functions as an entity in the worship of village gods, its unity
and identity becoming especially explicit in the important village-
protection ceremony.

Interaction among villagers is far more frequent than their inter-
action with members of other villages, except for villagers who live
elsewhere but still identify with the village. The status of nonresident
villagers, who retain their traditional village affiliation although they
live in outlying dwellings where they may interact more frequently
with members of other villages than with members of their own village,
is one of the contrasts between hill villages and those of the plains.
Often nonresident hill villagers represent the first step toward creation
of a new village. They usually retain their original village identifica-
tion as long as they belong to joint families which own a dwelling in
that village.

An individual's identification and commitment is first and always

CONCLUSION

341

with his local kin groups. It is not so easy to determine precedence
in more distant ties within the caste and in cross-caste ties in the
community. The individual is closely identified with, and committed
to, both. In disputes which test these loyalties he usually takes the
side of his caste over that of his village, but such situations are rare.
When there is no such conflict of loyalties, intercaste disputes find
him siding with his caste, and intervillage disputes find him siding with
his village. Village affiliation can be changed, though not easily, while
caste affiliation is inherited and unalterable. If a person emigrates from
his village he may seek new village affiliation, although his village of
origin will not be forgotten. He cannot seek new caste affiliation. He
establishes himself among members of his own caste in his new en-
vironment. Therefore, community ties, though strong, are less stable
than, and in certain respects secondary to, caste ties.

2J Primary orientation
Secondary orientation
Tertiary orientation

1. Kin groups

a. Household

b. Lineage (minimal and maximal)

c. Sib

d. Kindred

2. Caste as named group

3. Local community

4. Other sub-culture areas

Fig. 7. Kin, caste, and community orientations.

The relationship of these three levels of orientation for the in-
dividual— kin, caste, and community — can best be viewed in terms of
the area in which they coincide. A portion of the diagram of "levels
of identification and interaction" (fig. 6) can be adapted as a sche-
matic way of looking at the relationships. It is intended to suggest
reality rather than to define it precisely. Where kin, caste, and com-
munity ties all function at once to tie individuals to one another,
we find the area of the individual's fundamental or primary identifi-

342 CONCLUSION

cation (diagonally hatched portion of figure 7). Where two of the three
coincide, there is less intense or secondary identification (crosshatched
portion of the figure). Where only one of these ties is present, there
is a tertiary level of identification. Of course, kinship ties always occur
within the caste, a fact which assures the local kin group of its pre-
eminent place in the orientation of the individual. In other words,
kinfolk who live in the same village identify most closely with one
another. Other things being equal, as they rarely are, kinfolk who
reside in other villages and non-kin caste-fellows who reside in the same
village are of secondary immediacy in the orientation of the individual.
Non-kin caste-fellows of other villages and local villagers of other castes
are of tertiary immediacy in the orientation of the individual.

These statements are applicable only within the familiar area of
marriage relationships around the village, that is, Bhatbair and vicin-
ity. To go beyond the immediate culture area is to go beyond the
familiar world of villagers and hence to bring in factors of strangeness
and uncertainty. Thus, a caste-fellow from another area is not com-
parable to a resident of Bhatbair in the minds of the people of Bhat-
bair. He is outside the marriage network. Here degrees of difference
are crucial. A resident of a neighboring area is more easily appraised
and is more likely to be acceptable in a marriage arrangement than
one of a more distant and more different area. Any speaker of the
Central Pahari dialect can be fitted into the system more easily than
can speakers of Western and Eastern Pahari, while any Pahari is
less foreign than a non-Pahari. A fourth level of identification might
therefore be defined as non-kin, noncaste, noncommunity members
who are of the same subcultural area.

The status of the stranger emphasizes the relatively closed nature
of Sirkanda society. Ties of kinship, caste, and community are crucial.
If a person has such ties in the culture area he can be incorporated
into the social system. If not, he is an outsider and a stranger. Since
kin and caste ties cannot be acquired, the only hope of acceptance an
outsider has (shopkeeper, schoolteacher, village level worker, servant)
is through community affiliation, and this comes only after continuous
residence and interaction with villagers. The measure of acceptance
granted the anthropologist and his interpreter came after some time,
on the basis of residence in Sirkanda and interaction with its mem-
bers. Redfield (1957, pp. 33 f.) says that "institutionalized forms for
admitting strangers" are characteristic of peasant communities, re-
sulting from the fact that such communities are dependent upon cities
and therefore must admit representatives of necessary urban institu-
tions into their midst. Some of these representatives have to reside

conclusion 343

in the village for extended periods, and the peasant society is prepared
to allow them to do so. Redfield points out that such aliens do not
participate fully in village life (indeed they would not be allowed to
in Sirkanda), and cannot become village members. Like the shop-
keeper and governmental employees in Sirkanda, they become in-
stead "resident strangers," admitted upon sufferance of villagers to
play their limited roles.

In this research attention has been called to the confusion which
often arises between caste conceived of as an abstract, idealized system
and caste viewed as behavior — as attitudes expressed by people and as
interaction among people (cf. Berreman, 1960a). Both are important
for an understanding of the system, but their differences and spheres
of relevance have to be kept clear in the observer's mind if accurate
interpretations are to result.

While this point has been made with special reference to caste, a
similar point is pertinent to discussions of kin group, community, and
extracommunity organization and interaction. What people say should
be done is often quite different from what is done and enforced within
and among these groups. Some of the confusion about the "great"
and "little" religious traditions in India, with the frequent implication
that these are correlated with national and local spread, respectively,
is attributable in large part to failure to make the appropriate distinc-
tions in observation and analysis or to make explicit the context in
which reported religious attitudes and behavior correspond to the facts
of daily life. Similar confusion is evident in much of the comment
concerning the effectiveness, on the community level, of programs of
community development and local self-government. Speculation about
the decline, in some areas, of the village and of the extended family
as effective social and economic units may also be biased by confusion
on these points. Behavior, observed and analyzed in context, is neces-
sary information upon which to base conclusions about culture and
social structure.

This does not mean that such factors as private attitudes and, at
the other extreme, great traditions, should be overlooked. These are
important aspects of reality whose existence and relevance to those who
hold them are detectable only through behavior, often verbal or other
symbolic behavior. To overlook such factors would be to overlook an
aspect of culture as important as any other. However, they should
be analyzed in terms of their relationship to other types of behavior.
When information is obtained from secondary sources or literary
sources, care has to be taken to obtain the behavioral context in which
it functions. Caution is especially necessary when, as in matters of caste

344 CONCLUSION

in India, certain groups have strong vested interests in adhering to
particular and often limited views of reality, particularly before an
audience of outsiders.

The Pahari Culture Area: Sources and Affinities

In comparisons and generalizations the Himalayan foothills from
Kashmir across North India and Nepal have here been referred to as
a "culture area." In some contexts smaller segments of this area as,
for example, the region occupied by Central Pahari-speaking peoples
(roughly between the Jumna and Kali rivers), Tehri Garhwal, or even
Bhatbair and vicinity have been referred to as culture areas. This is
a matter of level of generalization, since a culture area is simply an
area within which the cultures are similar to one another and distinc-
tive relative to cultures outside that area. Sirkanda residents could be
legitimately described as representatives of any or all of the following
culture areas: India, North India, Pahari, Central Pahari, Garhwal,
Tehri Garhwal, Bhatbair and vicinity. Other areas could be delimited,
including some that would crosscut these. For present purposes the
Pahari culture area, an unusually distinct and sharply defined one,
will be discussed.

A brief description of this culture area which has been given else-
where can be repeated here to indicate the nature of Pahari culture:

The distinctiveness of the Paharis as a group is suggested by the fact that
they share a common and distinctive linguistic stock. They also share a num-
ber of other cultural features which distinguish them from the rest of the
North Indian culture area and specifically from the plains-dwellers adjacent
to them. These features, like their language, are not entirely unique or di-
vorced from those of the rest of North India, but are divergent forms grounded
in a common heritage. In emphasizing differences, care must be taken not to
ignore the numerous and basic similarities common to Paharis and other
North Indians. Differences are, however, the primary subject of this analysis.
Among distinguishing Pahari characteristics are:

(1) A somewhat distinctive caste structure wherein there is a major division
between the dominant high or twice-born castes ("big castes" in local par-
lance), made up of Brahmins and Rajputs, and the "untouchable" (achut)
low or "small" castes. The former are the land-owning agriculturalists; the
latter comprise all of the service castes (blacksmiths, carpenters, weavers,
musicians, shoemakers, and others), collectively termed Dom, and make up
only about 10 percent of the population in any area. While there is hierarchi-
cal caste ranking within each of these two major categories, it is of significance
primarily to those within that category. From across the high-low caste pollu-
tion barrier, it appears insignificant. The range of castes found in the hills is

conclusion 345

smaller than in the plains. Conspicuous by their absence are indigenous
Vaisya (merchants) and Sudra (clean caste artisans). On the other hand,
occupational variability within castes is considerable in the hills. . . .

(2) A number of rules pertaining to marriage which would be unacceptable
to many plains groups and especially to those of high caste. These include
bride-price marriage with no necessity for a Sanskritic marriage ceremony,
polyandry in some areas, levirate, divorce by mutual consent, remarriage of
widows and divorcees, toleration of intercaste marriage within the high- or
low-caste group. There is also a good deal of postmarital sexual freedom and
sanctioned relations of brothers with one another's wives. Marriage is uni-
versally prohibited only in own and mother's clan [that is, sib], and village
exogamy is not everywhere the rule.

(3) No seclusion of women and freer participation of women in most
aspects of life than on the plains, including their participation in singing
and dancing at festivals. Relatively free informal contact between the sexes
is usual.

(4) A number of religious and ritual features such as absence of the require-
ment for a Sanskritic marriage ceremony and absence of the requirement for
a sacred thread ceremony for high-caste boys, though such ceremonies are
coming rapidly into vogue in some areas. Distinctive Pahari marriage and
death ceremonies are performed. There is a great reliance upon mediums and
diviners and in some areas the Brahmin priest is relatively less important
than on the plains. Frequent and elaborate ritual purification and other
religiously motivated acts common on the plains are less widespread in the
hills. There are many distinctively Pahari religious beliefs and forms of wor-
ship. Animal sacrifice is a part of most Pahari ceremonies, and buffalo
sacrifice is found in some areas. Paharis are widely known for their devotion to
the Pandavas of Mahabharata fame and to Siva. The unique and spectacular
rope-sliding ceremony is performed in honor of the latter. . . .

(5) Distinctive folklore, songs, dances and festivals.

(6) Consumption of meat and liquor by all castes.

(7) Greater flexibility of intercaste relations and freer intercaste interac-
tion than on the plains. The caste hierarchy is important and caste status
differences are actively enforced, but the rules allow considerably more con-
tact and informal interaction than is usual in India.

(8) In addition to a nucleated settlement adjacent to a concentration of
village lands there are temporary-cum-permanent dwellings on widely scat-
tered and often distant agricultural and grazing lands. These are thought of
as part of the village even when other villages intervene.

(g) Terrace agriculture with primary dependence on millets, wheat, and
barley. Soil productivity is maintained by systematic fertilization, crop rota-
tion, and fallowing. Water is scarce but wherever possible is used for irrigated
rice cultivation.

(10) Dwellings of stone and timbers, often with slate roofs. Distinctive archi-
tecture of two stories with lower floor as barn and upper floor as living area,
often with large open veranda or porch at the upper level.

346 CONCLUSION

(11) A number of artifacts including lathe-turned wooden utensils, elabo-
rately carved wooden porch columns, lintels, windows, etc.; virtual absence of
pottery.

(12) Distinctive women's dress and ornamentation, including full skirt,
fitted jacket, and several types of gold and silver jewelry. Men's dress is not
so distinctive and has rapidly become like that of men of the plains, but now
includes a black or colored cap, a woolen blanket, and a cane as typical
Pahari accoutrements.

This list is suggestive rather than exhaustive. Some items on it may not be
as widespread in the hills as others, especially in the area east of Garhwal,
for which there is little information. It serves to make the point, however,
that this can for some purposes be considered a distinct culture area or
subarea within the greater North Indian area. In view of its geographical
and ecological isolation, its distinctiveness is not surprising; in view of its
common heritage with the rest of North India, its basic similarity thereto
is only what would be expected. (Berreman, 1960b, pp. 775-778)

Several explanations for these and other distinctive Pahari traits
have been offered by the people themselves and by various observers.
Perhaps the most frequent explanation is an environmental one. A
villager commented:

We can't observe all of those rules that plains people do. Our women
have to work, they can't bother with being secluded [purdah] or being out of
circulation when they menstruate or for a long time after they give birth.
We haven't enough water nor enough time to waste bathing all of the time
like some plains people do. If a Brahmin here practiced all the observances
a plains Brahmin does, his family would starve.

Some writers have set forth very explicit, and often curious en-
vironmentally deterministic, explanations. Raturi (1928, p. 207) says
that "it is natural to eat meat and drink wine and wear woolen clothes
to keep warm in cold countries," and Majumdar (1944, p. 128) asserts
that "in cold climates dances form the most important form of
recreation."

Some traits have been explained in terms of economics. Bahadur
(1916, p. 135) attributes polygyny to the need for wives as agricultural
laborers, and many villagers also give this explanation. Walton (1911a,
p. 88) attributes fraternal polyandry, where it occurs, to pressure on
the land and a desire to maintain family lands intact. Majumdar (1944,
p. 127) mentions the expense of a woman's jewelry as contributing to
polyandry. Several writers have attributed polygamy in this area to
disparity in proportions of the sexes (Berreman, 1962a).

A frequently suggested source of the unique configuration of cus-
toms found in this hill area is in the assumed cultural origins of the

conclusion 347

hill people. They are often pictured as modern survivals of an earlier
era. Briffault (1927, p. 671) says that "The highland regions of the
Himalaya are but a residual cultural island which preserves social
customs that had once a far more extensive distribution. The in-
stitutions which are found surviving there were once common through-
out the greater part of Central Asia." Another author has called them
"a fossil of the age of Mahabharata" (Munshi, 1955, p. i). Statements
such as the latter are based on similarities between certain practices
common in the area (polyandry, meat-eating, freedom of women, lack
of caste rigidity) and practices recorded in classic texts of Hinduism.
As has been pointed out in chapter 1, it is frequently asserted that
the high-caste Khasiyas represent a population of Indo-Aryan speak-
ing invaders who came into this area from the northwest, either directly
or via Rajasthan. These people presumably conquered or absorbed
an indigenous non-Aryan population ancestral to present-day Doms.
It is often assumed that the Indo-Aryans were originally of the pure
high-caste stock both genetically and culturally, but that they have
since degenerated as a result of isolation or adoption of the practices
of the people they conquered. Grierson (1916, p. 7) states that the
Khasiyas ". . . were looked upon [by the most ancient Indian au-
thorities] as Kshatriyas of Aryan origin. . . . They were considered
to have lost their claim to consideration as Aryans, and to have become
. . . barbarians due to their non-observance of the rules of eating and
drinking observed by the Sanskritic peoples of India."

The Himalayan area may once have been the home of the ancestors
of today's plains Hindus. Basham states that "It has been reasonably
suggested that the main line of Aryan penetration was not down the
[Ganges] river, the banks of which were then probably thick swampy
jungle, but along the Himalayan foothills" (Basham, 1953, p. 41).
"While the Aryans had by now [later Vedic period, ca. 700 B.C.]
expanded far into India their old home in the Panjab and the North-
West was practically forgotten. Later Vedic literature mentions it
rarely, and then usually with disparagement and contempt, as an
impure land where the Vedic sacraments are not performed. It may
have been once more invaded by Indo-Iranian tribes who did not
follow the orthodox rites" (ibid.).

Majumdar believes that the social and religious life of this culture
area is a result of mixture of the cultural traditions of indigenes and
the Indo-Aryan invaders, and even claims to find evidence that the
Doms were "matriarchal" and the Khasiyas "patriarchal." He assigns
particular distinctive Pahari traits such as "the double standard of
morality practiced by women" (in their home villages as contrasted

348 CONCLUSION

to those of their husbands) to the influence of alleged matriarchal Dom
culture (Majumdar, 1944, pp. 173 £.).

Saksena (1955, p. 30) notes that "it has been suggested by Mayne
that the Indo-Aryans adopted their polyandrous customs from the
aborigines or from a neighboring polyandrous people." Meat-eating,
religious unorthodoxy including worship of household and village
gods, and even worship of the god Shiva, have been attributed by some
writers to the "aboriginal" culture. Saksena disagrees, on grounds that
"when a superior culture imposes itself over an inferior culture, it
is the latter which is affected more" (Saksena, 1955, p. 31). Instead,
following Briffault, he attributes polyandry as well as most distinctive
contemporary practices to survivals of the common Indo-Aryan cul-
tural origin of the high-caste people.

The unique Pahari cultural configuration can also be attributed in
part to the distinctive combination of cultural contacts to which the
area has been subject since the period of settlement of Indo-Aryan
speaking peoples there.

The Pahari area, a long, narrow strip following the southern face
of the Himalayas, comprises the northernmost border of the Indian
and North Indian culture areas, meeting throughout its length the
southern edge of the Tibetan culture area. Off the western end of the
Pahari area lies the Southwest Asian area — the Indus valley and
Afghanistan (cf. Bacon, 1946). To the east the Pahari area seems to
terminate in contact with Tibetans, Tibeto-Burman, and perhaps
other "tribal" peoples (cf. Iijima, i960), and North Indian plains
groups. Paharis are to some extent physically isolated from these other
areas and groups. They are separated from peoples of the Indo-
Gangetic plain by the fact that they occupy rugged hills bordered by
a band of talus slopes, swamps, and jungles. They are separated from
the Tibetan plateau by the high Himalayas, and from Afghanistan
by the mountains of the old Northwest Frontier Province. More im-
portant in separating Paharis from adjacent peoples have been cultural
and ecological factors. Kawakita (1957, 1961) and Iijima (1961) have
commented in some detail on ecological zones and their ethnic cor-
relates in the Nepal Himalaya. Such barriers have contributed to
Pahari isolation, but they have not proved insuperable.

The most intensive outside contact in Garhwal has been with the
people of the plains of North India, who share historical ties with
Paharis and are of the same racial, cultural, and linguistic stock.
People from North India — the Gangetic plain, the Punjab, and
Rajasthan — have come to the Pahari area frequently on pilgrimages,
to trade, to seek refuge from the tribulations of their native areas, or

conclusion 349

to find new lands and subjects. Thus, the Paharis have felt directly
or indirectly the effects of most of the important invasions and other
upheavals of North India. Occasionally South Indians have come, espe-
cially on pilgrimage, and a few important temples in the mountains
are said to have South Indian priests.

Contact with the Southwest Asian area has been extremely limited in
contemporary times because of the short common boundary, the great
distances, intervening mountains and desert, and political and cultural
factors. However, historically this was very likely a region inhabited
at least for a time by the Indo-Aryan populations which ultimately
extended over all of North India including the Himalayan hills, and
which are presumed to have been the source of much of today's
Indian culture. The Paharis share this cultural tradition. Contact with
Kashmiris, in the northwest, continues today as an indirect Southwest
Asian contact.

Observers have frequently attributed Pahari polyandry to contacts
with polyandrous Tibetans. This is an unlikely source in view of the
distribution and functioning of Pahari polyandry. There has been,
however, considerable contact and mutual influence between the two
cultures. There are numerous regular trade routes over the Hima-
layas into Tibet including at least one (Nilang) in Tehri Garhwal.
Trade is carried on in both directions. Moreover, there are peoples
(the Bhotiyas) living in the higher Indian Himalayas, whose cultural
affinities are with Tibet (cf. Kawakita, 1957, 1961; Iijima, 1961; Pant,
1935; Srivastava, 1958). The Bhotiyas are in frequent contact with
adjacent Paharis, and they come each winter to trade and to pasture
goats and sheep in Pahari areas. In fact, as was noted in chapter 1, the
traditional name of the hill area in which this study was carried
out, Bhatbair, means "sheep den," in reference to its yearly use by
Bhotiya shepherds. Pahari culture has been influenced by contacts
with Bhotiyas, especially in Almora District. As a rule in India,
Bhotiyas seem to show more markedly than Paharis a cultural amalgam
or syncretism brought about by a combination of Tibetan affinities
and contacts and close association with Paharis of the Indian culture
area. In the hill area of Nepal (the Eastern Pahari-speaking region)
Tibetan influence has evidently been considerably stronger than it has
in most of the Indian Himalayas and is evidenced today in the greater
numbers of people and cultural traits of Tibetan affinities there (cf.
Kawakita, 1957, 1961; Iijima, 1961).

The Pahari area is, therefore, a relatively isolated one whose resi-
dents share common origins, history, contacts, and environment and
who interact with one another more than with outsiders. They have

350 CONCLUSION

numerous contacts with the residents of the adjacent North Indian
plains area, with whom they share origins, much history, and therefore
much culture. A significant number of Paharis are relatively recent
immigrants from the plains. With the Tibetan area Paharis have long
had occasional contact. They have had historical ties and rare con-
tacts with the Southwest Asian culture area as well. No attempt can
be made here to outline cultural element distributions to enumerate
the debts Pahari culture owes to these other areas. Comparative data
are not available and their collection was beyond the scope of this
research, but it is obvious that Pahari culture owes them much and
has contributed to them as well.

Origins, outside contacts, cultural amalgamation, and environmental
and economic adaptation are therefore all relevant to an explanation
of the distinctive Pahari culture. In conjunction with such explanatory
factors, the fact of isolation and the concept of cultural drift are use-
ful in throwing light on this subject. «,

All cultures change through time as a result of the transmission,
among those who carry them, of the results of experience — cultural
variants, alternatives, and additions and deletions, derived from new
conditions, relationships, contacts, and insights. Common experience
resulting from common environment and from frequent, intensive
interaction results in common culture. However, when two or more
groups of people become isolated so that social interaction and there-
fore cultural transmission is decreased or cut off between them, they
gradually accumulate differential experience and consequently their
cultures change in divergent ways. This cultural drift accounts in
large part for the differences between plains and Pahari cultures (Berre-
man, 1960b). Paharis have had more frequent and intensive contacts
with one another than with peoples of the plains or of other cul-
ture areas. For example, marriage networks apparently interconnect
throughout the Pahari area, but they end abruptly at the plains
boundary. Despite their probable common origin, Paharis and plains
people have long been separated by topographical and sociocultural
barriers, and subjected to different contacts and environments. As a
result they are culturally distinct. Language, social organization, econ-
omy, religion, and material culture all reflect their distinctiveness.

Degree of mutual isolation also explains contrasts between the de-
gree of cultural variability found among social groups in the hills
and that in the plains. Although Pahari culture displays relatively
consistent differences when contrasted with plains culture, and in
that context appears to be relatively homogeneous, small Pahari
localities and regions show a high degree of cultural difference from

CONCLUSION 351

one another. This variability is evidently greater over shorter dis-
tances than in the plains, just as Pahari marriage networks are smaller.
This is largely a result of greater isolation between Pahari localities
with consequent divergent or differential culture change, that is,
cultural drift. Because of the terrain, Pahari settled areas are relatively
inaccessible to one another. A fertile valley or gently sloping ridge
may have a fairly dense population but may be a difficult journey from
another such settled locality. It is such localities which differ culturally
so conspicuously from one another. Rosser (1955) describes an extreme
example of an isolated and culturally distinct Pahari locality.

Caste groups within a Pahari locality show notably less cultural
difference from one another than do castes of the plains — probably
also a function of isolation. Plains castes are socially more isolated
from one another than are those of the hills and so have had an
opportunity to become or remain culturally more distinct than if
they had been in frequent, intensive, and extensive contact as are
hill castes. On the plains, most intensive contacts are among caste-
fellows, often across local boundaries. In the hills, most intensive
contacts are within the locality, often across caste boundaries (Berre-
man, 1960b).

Functional Implications of Plains-Pahari Contrasts

Most generalizations about North Indian society and culture in
the anthropological literature are in reality generalizations about the
Indo-Gangetic plain. They ignore the Himalayan area either through
lack of information or an impression that Paharis are unimportant
marginals or possibly tribal peoples. One aim of the present research
is to provide a basis for correcting such errors or oversights and to
broaden the scope of available information about Indian village life
and its regional variations by describing a Pahari village and its region.

Distinctive Pahari characteristics and their possible sources have
been pointed out above and contrasted to those of the plains of North
India. In conclusion, it is appropriate to comment on the functional
implications of some of the most prominent contrasts between these
two culture areas.

The nature of the Pahari local group — the village — is affected by
its physical environment. Cultivable areas are often small and
scattered. Travel between them is slow and difficult as compared to
that in the plains. As a consequence, villages are small and isolated.
They have an inward-looking, self-contained character as compared
to plains villages, and the area of near self-sufficiency in economic,

352 CONCLUSION

social, and ritual matters is relatively small. Interaction has to be
within local or nearby groups if it is to occur at all.

In many Pahari areas there has always been more potentially cul-
tivable land available than could be put into cultivation under the
rigorous conditions of this sparsely populated region. Therefore,
population increase has proceeded with minimal pressure on the land.
Since much usable land is scattered in small amounts, a system of
residence in field houses (chans) has developed. As a consequence,
population increase has taken place without significant increase in the
size of particular villages. New settlements have arisen instead. The
chans may also have enabled growing joint families to remain intact
as social, economic, and ritual units by permitting them to divide as
residential units, thereby reducing interpersonal frictions which lead
to dissolution of joint families.

The small, inward-looking nature of the Garhwali village is reflected
in the relations among social units within it. The village is not exog-
amous. All members of the locally dominant Rajput caste in Sirkanda
have affinal or consanguineal ties with one another. Marriage net-
works of all castes are small, their size varying inversely with the popu-
lation of the caste in the surrounding area. Between high and low
castes there is more permissiveness or flexibility in rules of interaction
than on the plains, so that within the village they interact frequently
and intensively. Minority castes have to interact with their fellow
villagers of other castes if they are to interact at all, because caste-
fellows are few, far, and difficult of access. Permissiveness of interaction
as compared to plains custom does not indicate decreased importance
of the pollution barrier or ambiguity as to where it lies or what it
means. It merely indicates that there are different behavioral symbols
of status in the hierarchy. The lines are drawn as sharply as in the
plains, but in different places and in different ways.

Because Paharis have relatively little interaction with distant people,
even of their own caste, regional variation in culture is a prominent
feature. As a result of the nature of interaction between high and low
castes, cultural homogeneity within a locality is great. Everyone knows
everyone else's way of life, beliefs, and secrets. To a large extent, these
are common to all groups except where particular behaviors are sym-
bols of status differences or are the means of maintaining advantages for
particular groups in the village. Freedom of association, including
toleration of occasional intermarriage between Rajputs and Brahmins,
seems to indicate a genuinely smaller social and ritual distance at this
level than is found on the plains.

Associated with permissiveness of intercaste relations is a great

conclusion 353

amount of flexibility of occupational specialization among castes. There
are few castes of occupational specialists, and few individuals in these
castes. In any one village or local area, certain specialists may be lack-
ing, or inaccessible. It is therefore important to have artisans who can
adopt any of several occupations, and even to have high castes willing
and able to take over in an emergency. The flexibility of rules of caste
behavior makes this possible without jeopardizing caste status, but it
has the effect of reducing the economic security of artisans, for they
have little in the way of effective occupational monopolies. Conse-
quently the jajmani system of traditional exchange of goods for serv-
ices is less rigid, and is less effective from the artisans' point of view,
than on the plains (Berreman, 1962b).

The fact that villages have been occupied primarily by owner-
cultivators has contributed to self-reliance and an independent attitude
among high-caste Sirkanda residents quite in contrast to the alleged
ma-bapism, or dependence upon paternalistic agents of outside author-
ity, reported for some peoples of India. Paharis owe allegiance to no
landlord. No cultivator in the locality in recent times has been master
of any other. No landholder is free of arduous agricultural labor in
his fields. No one has conveniences or luxuries not available to nearly
everyone else. Pahari agricultural techniques, such as fertilization of
fields and crop rotation, together with ample land, regular rainfall,
and large forest areas to provide fuel and fodder, have made for an
economy sufficient to consistently meet subsistence needs. Famine is
virtually unknown. People work hard, but economic sufficiency is the
result. For the dominant high castes this means a kind of security un-
known to many cultivators of the plains.

Also contributory to an independence of spirit in and around Sir-
kanda has been relative isolation from outside supervision and inter-
vention. Sirkanda people seem to feel less restricted than many plains
people. They have a mild kind of frontier mentality associated with
freedom from authority and freedom from absence of alternatives. For
high-caste Paharis there has always been more land for the taking if
one could muster the labor to prepare and cultivate it. Moreover, one
could move out of his family's house and out of the village to a new
chan location if he wished. If worst came to worst, he could remove to
an entirely new area, as did those who settled Sirkanda over 300 years
ago.

Low-caste people are acutely aware of their insecure economic po-
sition and of high-caste advantage. They feel that they are virtual
puppets of the agriculturists. They do not need to fear starvation as
long as they remain in the good graces of the cultivating high castes.

354 CONCLUSION

What they fear is the rather erratic bestowal and withdrawal of those
good graces. Among themselves they are more competitive and less
cohesive than is usual among equivalent plains castes. They must often
compete for clients because they are dependent on their craftwork for
livelihood and there are others, even of other castes, who will take
their place if the opportunity arises. This competition varies from place
to place and time to time, depending upon the population of artisans
in a particular area.

Because of their isolation from caste-fellows resulting from their
small numbers and the isolation of villages, low-caste people have little
opportunity to coordinate their activities to defend their interests.
Perhaps, in view of factors noted above, they would not do so anyway.
They have little in the way of caste government, so that their position
relative to the high castes is especially weak. The high castes have
maintained the situation by serving as authoritarian and ideally pa-
ternalistic overseers of the low castes, deciding among themselves all
matters pertaining to low-caste people, both internal and intercaste.
Low castes are controlled by being kept economically dependent
through being denied the opportunity to acquire land in amounts
sufficient to make a living. All castes recognize that land means inde-
pendence for those who own it and control over those who depend
upon the owners. This is why agrarian reform is one of the most impor-
tant problems of rural India.

The nature of Pahari economic organization has ramifications in the
position of women, which is one of unusual freedom compared to that
of high-caste plains women. Like low-caste women of the plains, hill
women participate in the economy in a way incompatible with seclu-
sion.1 Their work in fields and forests is essential to the well-being of
their families. If anything, they do more of this kind of work than the
men. Under the circumstances it would not only be difficult for them
to observe the niceties of seclusion and other disabilities common on
the plains, but also impossible for their menfolk to supervise them
closely. This is probably a factor in the sexual freedom allowed Pahari
women.

The Pahari rule of bride-price marriage is combined with easy di-
vorce and universal remarriage for divorcees and widows. The family
of the groom seeks out a bride and arranges the marriage with her
family, whereas in the plains, where dowry is the rule, the bride's fam-
ily performs these functions. The Pahari bride is in a sense considered
the property — and valuable property — of the husband's joint family
who paid for her, who derive benefit from her labor, and who will give
her up only upon return of the bride price. Even after her husband's

conclusion 355

death the family's interest in her remains. Usually she becomes her
husband's brother's wife by the custom of levirate. The idea of a wife
as common property is undoubtedly related to the fact that brothers
share their wives' sexuality and that in several Pahari areas fraternal
polyandry is the rule (Berreman, 1962a).

Hindu unorthodoxy in the area is due in part to the cultural herit-
age of the Paharis and in part to their isolation from areas where
modern Hinduism developed. It is ascribed by Paharis to poverty and
their difficult environment. They have little time or money for the
luxury of the elaborate rituals of the plains. They claim to lack the
water for the frequent purificatory baths of plainsmen, the leisure to
endure fasts and food taboos, the money to hire learned Brahmins, and
the education to appreciate these things. In this, as in the freedom of
women, they are like low-caste people of the plains. Their increasing
awareness of this similarity, resulting from increasing contacts with
plains people, leads high-caste Paharis to aspire to emulate plains be-
havior in order to improve their relative status.

In discussing Pahari culture, the plains have been chosen for com-
parison because the culture of that area is closely associated with that
of the Himalayan hills. The two areas have a common cultural base
upon which differences stand out in high relief. Also, it is among people
of the plains that the most comparable work has been done upon
which comparisons can be based. However, one familiar with the
plains must guard against plains-oriented ethnocentrism, in which the
plains become the yardstick by which Pahari culture is measured. It is
important to bear in mind that Pahari culture is viable in its own right.
The fact that it is marginal to the Indian and North Indian culture
areas and that it has also apparently been influenced from Tibet does
not mean that it is a hodgepodge of borrowings or a distorted reflection
of other cultures. The borrowed traits have been integrated into the
matrix that is Pahari culture.

The observer of Pahari culture is struck not only by the traits which
bear similarities to those found in other culture areas and that have
been either borrowed from them or derived from a common origin, but
also by the unique traits that, so far as can be determined, were de-
veloped by the residents of this region. These range from agricultural
techniques such as complex feats of terracing and irrigation, and
uniquely constructed and styled houses, to the songs, dances, folklore,
and religion of the area.

Moreover, the Pahari area has made important contributions to the
cultures of adjacent areas. The Hindu god Shiva, for example, may
have originated in Pahari culture, and it is not improbable that some

356 CONCLUSION

cultural characteristics commonly attributed to the plains have ema-
nated from the hills.

Perhaps more significant than particular traits and their affinities
is the distinctive over-all configuration of Pahari culture. To those who
live it, Pahari culture is as distinct and as internally consistent as is
the culture of any other group, whether it be in an area of "culture
climax" or in a "marginal" area. Marginality is relative. From the
Pahari's point of view the plains of North India constitute an area as
marginal to his own as do the high Himalayas to the north.

When the Pahari disparages his homeland and his way of life to
outsiders, it does not mean that he thinks them inferior. It merely
indicates his knowledge of outsiders' views of him and his anticipation
of their remarks or thoughts. He may feel that the time is coming when
the modern plains way will be the only way to success in a new world.
He may envy a movie star or a well-to-do contractor or a suave lawyer,
but he does not envy the plains villager or urban peon. He is embar-
rassed by his conspicuous rusticity when he goes to town. He resents
taunts and exploitation by strangers. But as long as he is in his home-
land, among his own people, he sees the Pahari way of life as superior.

The Pahari's rejection of strangers and their ideas is in part an indi-
cation of his satisfaction with his own way of life, a fact which is fur-
ther indicated by the community cohesion characteristic of this area
and by the fact that few have left the area. The Pahari ridicules the
man of the plains who cannot easily negotiate the mountain trails, who
is afraid of the terrain and its fauna, who cannot survive on its re-
sources. He pities the people of the plains who do not dance, sing, and
drink together; whose women are secluded; whose widows cannot re-
marry; who must drink stagnant water and breathe dusty air; who do
not eat meat; who suffer extreme heat, and risk drought and famine in
the hot season; who face flood and malaria in the rainy season. Paharis
work hard, but they know how to enjoy their spare time together, re-
gardless of age, sex, and caste differences, in a way that plains people
might envy. Paharis are proud of their way of life and their environ-
ment, but are sensitive to ridicule of them.

While it is useful to analyze this culture in terms of contacts and out-
side sources of diffusion, the analysis, if it is to be complete, must con-
sider the adjustments, innovations, and distinctive integration which
make of the Pahari region a culture area in its own right. As more data
become available on the way of life of the peoples of surrounding areas,
it will be possible to determine more accurately the culture areas of
this part of the world and to assess the mutual cultural debts and

conclusion 357

affiliations of the Paharis and their neighbors, and the kinds and di-
rections of change they face.

In the meantime, the Himalayan hill peoples, in common with
villagers all over India, are beginning to play an increasingly important
role as contributors and recipients in the development of the national
culture of Independent India. They live in a region which will assume
increasing importance to the nation as a reservoir of rich mineral,
forest, agricultural, and hydraulic resources, natural beauty, climatic
attractiveness, and religious inspiration. The Paharis themselves, as the
human resources, are essential to the development of this area and
important in the development of the nation. They have much to offer
as well as much to learn if India is to realize the potential inherent
in them and in the land they occupy.

Bibliography

Atkinson, Edwin T.

1882, 1884a, 1886 The Himalayan Districts of the North-Western Provinces
of India. Vol. I (1882); Vol. II (1884); Vol. Ill (1886), comprising Vols.
X-XII of the Gazetteer of the North-Western Provinces and Oudh.
Allahabad: North- Western Provinces and Oudh Press.
1884b, 1885 "Notes on the History of Religion in the Himalaya of the
N. W. Provinces." Journal of the Asiatic Society of Bengal, Vol. 53
(1884), 39-103; Vol. 54 (1885), 1-16.
Bacon, Elizabeth

1946 "A Preliminary Attempt to Determine the Culture Areas of
Asia." Southwestern Journal of Anthropology, Vol. 2, pp. 117-
132.
Baden-Powell, B. H.

1892 The Land-System of British India, 3 vols. Oxford: Clarendon Press.
Bahadur, Rai Pati Ram

1916 Garhwal, Ancient and Modern. Simla: Army Press.
Bailey, F. G.

1957 Caste and the Economic Frontier. Manchester: University Press.
1959 "For a Sociology of India?" Contributions to Indian Sociology, Vol.

3, pp. 88-101.
Banfield, Edward C.

1958 The Moral Basis of a Backward Society. Glencoe, 111.: Free Press.
Basham, A. L.

1954 The Wonder That Was India, Evergreen edition. New York: Grove
Press.

360 BIBLIOGRAPHY

Beals, Alan

1954 "The Government and the Indian Village." Economic Development

and Cultural Change, Vol. 2, pp. 397-407.
Beidelman, Thomas O.

1959 A Comparative Analysis of the Jajmani System. Locust Valley, N.Y.:

J. J. Augustin.
Berreman, Gerald D.

ig55 "Inquiry into Community Integration in an Aleutian Village."

American Anthropologist, Vol. 57, pp. 49-59.
1959 "Kin, Caste and Community in a Himalayan Hill Village." Ph.D.

dissertation, Department of Sociology and Anthropology, Cornell

University. Ann Arbor: University Microfilms.
1960a "Caste in India and the United States." American Journal of Soci-
ology, Vol. 66, pp. 120-127.
1960b "Cultural Variability and Drift in the Himalayan Hills." American

Anthropologist, Vol. 62, pp. 774-794.
1961a "Brahmins and Shamans in Pahari Religion." Paper presented to

the Seminar on South Asian Religion, Berkeley, Calif., August 21-

25, 1961.
1961b "Himalayan Rope Sliding and Village Hinduism: An Analysis."

Southwestern Journal of Anthropology, Vol. 17, pp. 326-342.
1962a "Pahari Polyandry: A Comparison." American Anthropologist,

Vol. 64, pp. 60-75.
1962b "Caste and Economy in the Himalayas." Economic Development

and Cultural Change, Vol. 10, pp. 386-394.
1962c Behind Many Masks: Ethnography and Impression Management in

a Himalayan Village. Monograph 4, Ithaca, N. Y.: Society for Ap-
plied Anthropology.
ig62d "Village Exogamy in Northernmost India." Southwestern Journal

of Anthropology, Vol. 18, pp. 55-58.
1962c "Sib and Clan Among the Pahari of North India." Ethnology, Vol.

1, pp. 524-528.
Blunt, E. A. H.

1931 The Caste System of Northern India. London: H. Milford, Oxford

University Press.
Bose, N. K.

1958 "Some Aspects of Caste in Bengal." American Journal of Folklore,

Vol. 71, pp. 397-412.
Bougie, C.

1908 Essais sur le Regime des castes. Paris: Librairies F^lix Alcan et

Guillaumin R^unies.
Briffault, Robert

1927 The Mothers, Vol. I. New York: The Macmillan Company.
Brown, W. Norman

1953 The United States and India and Pakistan. Cambridge, Mass.:

Harvard University Press.

BIBLIOGRAPHY 36 1

Census of India, 1931

1933 Bengal and Sikkim, Vol. 5. Calcutta: Central Publication Branch.

1933 Punjab, Vol. 17 (Parts I and II). Lahore: Civil and Military Gazette
Press.

1933 United Provinces of Agra and Oudh, Vol. 18 (Parts I and II). Alla-
habad: Superintendent, Printing and Stationery, United Provinces.

n.d. Jammu and Kashmir State, Vol. 24. Jammu: Ranbir Government
Press.
Census of India, 1941

1943 Jammu and Kashmir, Vol. 22. Ranbir: Ranbir Government Press.
Census of India, 1951

1952a Paper No. 2, Population Zones, Natural Regions, Sub-Regions and
Divisions. New Delhi: Government of India Photo-Litho Press.

1952b District Population Statistics, Uttar Pradesh: I. Dehra Dun District.
Allahabad: Superintendent, Printing and Stationery, Uttar Pradesh.

1952c Volume II, Uttar Pradesh, Part Ha, General Population Tables.
Allahabad: Superintendent, Printing and Stationery, Uttar Pradesh.

ig53 District Population Statistics, Uttar Pradesh: XXXIX. Tehri-
Garhwal District. Allahabad: Superintendent, Printing and Sta-
tionery, Uttar Pradesh.

1954 District Census Handbook, Uttar Pradesh: I. Dehra Dun District.
Allahabad: Superintendent, Printing and Stationery, Uttar Pradesh.

1955 District Census Handbook, Uttar Pradesh: XXXIX. Tehri-Garhwal
District. Allahabad: Superintendent, Printing and Stationery, Uttar
Pradesh.

Cohn, Bernard S.

1954 "The Camars of Senapur: A Study of the Changing Status of a
Depressed Caste." Ph.D. dissertation, Department of Sociology and
Anthropology, Cornell University. Ann Arbor: University Micro-
films.

1955 "The Changing Status of a Depressed Caste," in Village India, ed.
by McKim Marriott. Memoir 83, American Anthropological Associ-
ation, Vol. 57, pp. 53-77.

Cohn, Bernard S., and McKim Marriott

1958 "Networks and Centres in the Integration of Indian Civilization."
Journal of Social Research (Bihar University), Vol. 1, pp. 1-9.
Corbett, Jim

1946 Man-Eaters of Kumaon. New York: Oxford University Press.
Crooke, William

1926 Religion and Folklore of Northern India. London: Oxford Uni-
versity Press.
Davis, Kingsley

1949 Human Society. New York: The Macmillan Company.
Dollard, John

1957 Caste and Class in a Southern Town, 3rd edition. Garden City,
N. Y.: Doubleday.

362 BIBLIOGRAPHY

Dube, S. C.

!955 Indian Village. Ithaca, N. Y.: Cornell University Press.

1956 "Cultural Factors in Rural Community Development." Journal of
Asian Studies, Vol. 16, pp. 19-30.

1958 India's Changing Villages. London: Routledge and Kegan Paul.
Dumont, Louis, and D. Pocock, eds.

1957 Contributions to Indian Sociology, No. 1. Paris: Mouton.

1958 Contributions to India?! Sociology, No. 2. Paris: Mouton.
Dwivedi, J. N.

i957(?) Practice and Digest of U. P. Panchayat Raj, 2nd edition. Allahabad:
Allahabad Law Agency.
Fraser, James B.

1820 Journal of a Tour Through Part of the Snowy Range of the Himala
Mountains and to the Sources of the Rivers Jumna and Ganges.
London: Rodwell and Martin.
Gadgil, D. R.

1952 Poona, A Socio-Economic Survey, Part II. Poona: Gokhale Institute
of Politics and Economics.
Gandhi, M. K.

1952 Rebuilding Our Villages. Ahmedabad: Navajivan Publishing House.
Ghurye, G. S.

1952 Caste and Class in India. New York: Philosophical Library.
Gilbert, William H.

1948 "Caste in India, a Bibliography." Washington, D.C.: Library of
Congress. (Mimeographed.)
Goffman, Erving

1959 The Presentation of Self in Everyday Life. Garden City, N.Y.:
Doubleday.

Gorer, Geoffrey

1938 Himalayan Village. London: Michael Joseph.
Gough (Aberle), Kathleen

1959 "Criteria of Caste Ranking in South India." Man in India, Vol. 39,
pp. 115-126.

Gould, Harold A.

i960 "The Micro-Demography of Marriages in a North Indian Area."

Southwestern Journal of Anthropology, Vol. 16, pp. 476-491.
ig6ia "Sanskritization and Westernization: A Dynamic View." Economic

Weekly, Vol. 13 (No. 25), pp. 945~95°-
1961b "A Further Note on Village Exogamy in North India." Southwest-
ern Journal of Anthropology, Vol. 17, pp. 297-300.
Grierson, George A.

1916 Linguistic Survey of India, Vol. IX, Part IV. Calcutta: Superintend-
ent of Government Printing.
Harper, Edward B.

1957 "Shamanism in South India." Southiuestern Journal of Anthropol-
ogy, Vol. 13, pp. 267-287.

BIBLIOGRAPHY 363

Himachal Times

1958a "Demand for Separate Hill State." Himachal Times, Mussoorie-

Dehra Dun, Vol. X, No. 21 (June 5, 1958), 1.
1958b "Congress Men Threaten Direct Action." Himachal Times.
Mussoorie-Dehra Dun, Vol. X, No. 31 (August 8, 1958), 1.
Hitchcock, John T.

1956 "The Rajputs of Khaalaapur: A Study of Kinship, Social Stratifi-
cation, and Politics." Ph.D. dissertation, Department of Sociology
and Anthropology, Cornell University. Ann Arbor: University Mi-
crofilms.

Hocart, A. M.

1950 Caste, A Comparative Study. London: Methuen.
Hutton, J. H.

1946 Caste in India. Cambridge, England: Cambridge University Press.
Iijima, Shigeru

i960 "The Thakali — A Central Himalayan Tribe" (Torbo Ethnography
No. 2). Japanese Journal of Ethnology, Vol. 24, No. 3, pp. 1-22. (In
Japanese.)
1961 Agriculture and Land System in Nepal (Research Series 18). Tokyo:
Institute of Asian Economics. (In Japanese.)
Jain, S. C.

1948 "Some Features of Fraternal Polyandry in Jaunsar Bawar." Eastern
Anthropologist, Vol. 1, No. 2, pp. 27-33.
Joshi, L. D.

1929 The Khasa Family Law in the Himalayan Districts of the United
Provinces of India. Allahabad: Superintendent, Government Press,
U. P.

Karve, Irawati

1953 Kinship Organization in India. Deccan College Monograph Series
11. Poona: Deccan College Postgraduate and Research Institute.
Kawakita, Jiro

1957 "Ethno-Geographical Observations on the Nepal Himalaya," in
Scientific Results of the Japanese Expeditions to Nepal Himalaya,
I952~53> Vol. 3 (Peoples of Nepal Himalaya), ed. by Hitoshi Kihara.
Kyoto: Fauna and Flora Research Society. Pp. 1-362.

1961 "Some Ecological Observations in Nepal Himalaya" (Torbo Ethnog-
raphy No. 3). Japanese Journal of Ethnology, Vol. 25, No. 4, pp.
1-42. (In Japanese.)
Kroeber, Alfred L.

1930 "Caste," in Encyclopedia of the Social Sciences, Vol. 3, ed. by E. R.
A. Seligman and Alvin Johnson. New York: The Macmillan Com-
pany. Pp. 254-256.

1948 Anthropology. New York: Harcourt, Brace.
Leach, E. R.

i960 "Introduction: What Should We Mean by Caste," in Aspects of
Caste in South India, Ceylon and North-west Pakistan, ed. by E. R.

364 BIBLIOGRAPHY

Leach. Cambridge Papers in Social Anthropology, No. 2, pp. 1-10.
Cambridge, England: Cambridge University Press.
Lewis, Oscar

1958 Village Life in Northern India. Urbana: University of Illinois Press.

1959 "Review of Pilot Project, India, by Albert Mayer et al." American
Anthropologist, Vol. 61, pp. 534-536.

Lipset, Seymour M., and Reinhard Bendix

i960 Social Mobility in Industrial Society. Berkeley and Los Angeles:
University of California Press.
Lowie, R. H.

1947 Primitive Society. New York: Liveright.
Mahar, James M.

1958 "Occupational Adjustment in a Caste Society." Paper read before
the 57th annual meeting of the American Anthropological Associ-
ation, Washington, D.C., November 20, 1958.
Mahar, Pauline Moller

*957 "The Functional Relations of a Hindu Cult." Paper read before the
meeting of the Society for Scientific Study of Religion, Cambridge,
Mass., November 2, 1957.

1958 "Changing Caste Ideology in a North Indian Village." Journal of
Social Issues, Vol. 14, No. 4, pp. 51-65.

1959 "A Multiple Scaling Technique for Caste Ranking." Man in India,
Vol. 39, pp. 115-126.

i960 "Changing Religious Practices of an Untouchable Caste." Economic
Development and Cultural Change, Vol. 8, pp. 279-287.
Majumdar, D. N.

1944 The Fortunes of Primitive Tribes. Lucknow: Universal Publishers.

1955 "Demographic Structure in a Polyandrous Village." Eastern Anthro-
pologist, Vol. 8, pp. 161-172.
Mandelbaum, David G.

*955 "The World and the World View of the Kota," in Village India, ed.
by McKim Marriott. Memoir 83, American Anthropological Associ-
ation, Vol. 57, pp. 223-254.
Marriott, McKim

*955a "Little Communities in an Indigenous Civilization," in Village In-
dia, ed. by McKim Marriott. Memoir 83, American Anthropological
Association, Vol. 57, pp. 171-222.

1959 "Interactional and Attributional Theories of Caste Ranking." Man
in India, Vol. 39, pp. 92-107.

i960 Caste Ranking and Community Structure in Five Regions of India
and Pakistan. Deccan College Monograph Series 23, Poona: Deccan
College Postgraduate and Research Institute.
Marriott, McKim, ed.

1955b Village India, Memoir 83, American Anthropological Association,
Vol. 57, No. 3, Part 2.

BIBLIOGRAPHY 365

Mayer, Adrian C.

1958 "The Dominant Caste in a Region of Central India." Southwestern
Journal of Anthropology, Vol. 14, pp. 407-427.

Merton, Robert K., and Alice Kitt

1950 "Contributions to the Theory of Reference Group Behavior," in

Continuities in Social Research, ed. by R. K. Merton and P. F.

Lazarsfeld. Glencoe, 111.: Free Press. Pp. 40-105.
Morgan, Kenneth W., ed.

1953 The Religion of the Hindus. New York: Ronald Press.
"Mountaineer" [Mr. Wilson of Mussoorie]

i860 A Summer Ramble in the Himalayas. London: Hurst and Blackett.
Munshi, K. M.

1955 "Forward," in Social Economy of a Polyandrous People, ed. by
R. N. Saksena. Agra: Agra University Press. Pp. i-viii.
Murdock, George P.

1949 Social Structure. New York: The Macmillan Company.
Nadel, S. F.

1954 "Caste and Government in Primitive Society." Journal of the An-
thropological Society of Bombay, Vol. 8, pp. 9-22.

Nevill, H. R.

1904 Naini Tal: A Gazetteer, comprising Vol. XXXIV of the District
Gazetteers of the United Provinces of Agra and Oudh. Allahabad:
Government Press.

Newcomb, Theodore M.

1950 Social Psychology. New York: Dryden Press.
Newell, William H.

1954 "A Note on Village Government in the Indian Northern Hill States
Since Indian Independence." Economic Development and Cultural
Change, Vol. 2, pp. 416-419.

1955 "Goshen, a Gaddi Village in the Himalayas," in India's Villages,
Development Department, West Bengal Government. Calcutta:
West Bengal Government Press. Pp. 51-61.

Niehoff, Arthur

1959 "Caste and Industrial Organization in North India," Administra-
tive Science Quarterly, Vol. 3, pp. 494-508.

Oakley, E. S.

1905 Holy Himalaya. Edinburgh: Oliphant Anderson and Ferrier.
Olcott, Mason

1944 "The Caste System of India." American Sociological Review, Vol.
9, pp. 648-657.
O'Malley, L. S. S.

1932 Indian Caste Customs. New York: Cambridge University Press.
Opler, Morris E.

1956 "The Extensions of an Indian Village." Journal of Asian Studies,
Vol. 16, pp. 5-10.

366 BIBLIOGRAPHY

1958 "Spiritual Possession in a Rural Area of Northern India," in
Reader in Comparative Religion, ed. by W. A. Lessa and E. Z. Vogt.
Evanston: Row, Peterson. Pp. 553-566.
1959a "Family, Anxiety and Religion in a Community of North India,"
in Culture and Mental Health: Cross-Cultural Studies, ed. by Mar-
vin K. Opler. New York: The Macmillan Company. Pp. 273-289.
1959b "The Place of Religion in a North Indian Village." Southwestern

Journal of Anthropology, Vol. 15, pp. 219-226.
i960 "Political Organization and Economic Growth, the Case of Village
India." International Review of Community Development, No. 5,
pp. 187-197.
Opler, Morris E., and R. D. Singh

1948 "The Division of Labor in an Indian Village," in A Reader in Gen-
eral Anthropology, ed. by C. S. Coon. New York: Henry Holt and
Company. Pp. 464-496.
Pant, S. D.

1935 The Social Economy of the Himalayans. London: George Allen and
Unwin.
Pargiter, F. E.

1922 Ancient Indian Historical Tradition. London: Oxford University
Press.
Passin, Herbert

J955 "Untouchability in the Far East." Monumenta Nipponica, Vol. 11,
pp. 27-47.
Perelman, S. J.

1947 Westward Ha! New York: Simon and Schuster.
Planalp, Jack M.

1956 "Religious Life and Values in a North Indian Village." Ph.D. dis-
sertation, Department of Sociology and Anthropology, Cornell Uni-
versity. Ann Arbor: University Microfilms.

Planning Commission, Government of India

1958 The New India: Progress Through Democracy. New York: The
Macmillan Company.
Raturi, Pundit Hari Krishnan

1928 Garhwal ka Ityhas (History of Garhwal). Dehra Dun: Garhwali Press.
(In Hindi.)
Redfield, Robert

x955 "The Social Organization of Tradition." Far Eastern Quarterly,
Vol. 15, pp. 13-22.

1957 The Primitive World and Its Transformations, Great Seal Books.
Ithaca: Cornell University Press.

Rosser, Collin

ig55 "A 'Hermit' Village in Kulu," in India's Villages, Development De-
partment, West Bengal Government. Calcutta: West Bengal Gov-
ernment Press. Pp. 70-81.
n.d. The Valley of the Gods. London: Kimber (in press).

BIBLIOGRAPHY 367

Rowe, William L.

1960a "Social and Economic Mobility in a Low-Caste North Indian Com-
munity." Ph.D. dissertation, Department of Sociology and Anthro-
pology, Cornell University. Ann Arbor: University Microfilms.
1960b "The Marriage Network and Structural Change in a North Indian
Community," Southwestern Journal of Anthropology, Vol. 16,
PP- 299-311.
Ryan, Bryce

1953 Caste in Modern Ceylon. New Brunswick, N. J.: Rutgers University
Press.
Saksena, R. N.

x955 Social Economy of a Polyandrous People. Agra: Agra University
Institute of Social Sciences.
Sankrityayana, Rahul

1958 Kumaon. Banaras: Jana Mandal. (In Hindi.)
Sharma, K. N.

1961 "Occupational Mobility of Castes in a North Indian Village." South-
western Journal of Anthropology, Vol. 17, pp. 146-164.
Singer, Milton

x955 "The Cultural Pattern of Indian Civilization." Far Eastern Quar-
terly, Vol. 15, pp. 23-36.
Singh, Kailash K.

1957 "Inter-caste Tensions in Two Villages in North India." Ph.D. dis-
sertation, Department of Sociology and Anthropology, Cornell Uni-
versity. Ann Arbor: University Microfilms.
Sivananda, Swami

1947 Hindu Fasts and Festivals and Their Philosophy. Rikhikesh: Siva-
nanda Publication League.
Smith, Robert J.

1956 "The Problem of Extensions of a Community." Cornell University
Studies in Culture and Applied Science, Cornell University. (Mime-
ographed.)
Srinivas, M. N.

1952 Religion and Society Among the Coorgs of South India. Oxford:
Clarendon Press.

1955a "Castes: Can They Exist in India of Tomorrow?" Economic Weekly
(Bombay), Vol. 7, pp. 1230-1232.

1955b "The Social System of a Mysore Village," in Village India, ed. by
McKim Marriott. Memoir 83, American Anthropological Associa-
tion, Vol. 57, pp. 1-35.

1956 "A Note on Sanskritization and Westernization." Far Eastern
Quarterly, Vol. 15, pp. 481-496.

*957 "Caste in Modern India." Journal of Asian Studies, Vol. 16, pp.
529-548.

1959 "The Dominant Caste in Rampura." American Anthropologist, Vol.
61, pp. 1-16.

368 BIBLIOGRAPHY

Srivastava, R. P.

1958 "The Bhotia Nomads and Their Indo-Tibetan Trade." Journal of
the University of Saugar, Vol. 7, Part 1, Sec. A, pp. 1-22.

Stevenson, H. N. C.

1954 "Status Evaluation in the Hindu Caste System." Journal of the
Royal Anthropological Institute of Great Britain and Ireland, Vol.
84, pp. 45-65.
Stowell, V. A.

1907 A Manual of the Land Tenures of the Kumaun Division. Allahabad:
United Provinces Government Press.
Tinker, Hugh

1959 "Authority and Community in Village India." Pacific Affairs, Vol.

32, PP- 354-375-
Tod, James

1829-1832 Annals and Antiquities of Rajast'han, 2 vols. London: Smith,
Elder.
Traill, G. W.

1828 "Statistical Sketch of Kamaon." Asiatic Researches, Vol. 16, pp. 137—

234-
Turner, Ralph

1956 "Role-Taking, Role Standpoint, and Reference-Group Behavior."
American Journal of Sociology, Vol. 61, pp. 316-328.
Upreti, Pandit Ganga Datt

1894 Proverbs and Folklore of Kumaun and Garhwal. Lodiana: Lodiana
Mission Press.
Walton, H. G.

1910 British Garhwal: A Gazetteer, comprising Vol. XXXVI of the Dis-
trict Gazetteer of the United Provinces of Agra and Oudh. Alla-
habad: Government Press.
1911a Dehra Dun: A Gazetteer, comprising Vol. I of die District Gazetteer
of the United Provinces of Agra and Oudh. Allahabad: Govern-
ment Press.
1911b Almora: A Gazetteer, comprising Vol. XXXV of the District Gazet-
teer of the United Provinces of Agra and Oudh. Allahabad: Gov-
ernment Press.
Weber, Max

1958 The Religion of India, tr. and ed. by H. H. Gerth and D. Martin-
dale. Glencoe, 111.: Free Press.
Williams, G. R. C.

1874 Historical and Statistical Memoir of Dehra Doon. Roorkee: Thoma-
son Civil Engineering College Press.
Wiser, Charlotte V., and William H.

1951 Behind Mud Walls, 3rd printing. New York: Agricultural Missions.
Wiser, William H.

1936 The Hindu Jajmani System. Lucknow: Lucknow Publishing House.

Appendixes

APPENDIX I: GODS
WORSHIPED IN SIRKANDA

A. Household Gods

The following gods are worshiped separately by particular house-
holds in Sirkanda, before shrines maintained by each household, and
usually without the aid of Brahmin priests.

Nar Singh

The god Nar Singh, though generally referred to as a single god,
has three manifestations which require separate worship: Dudadharl
Nar Singh, Janghoria Nar Singh, and Kerarl Nar Singh. The first of
these is by far the most often worshiped of the three, and he is the
gentlest, his name deriving from the word for milk. The second has the
reputation of being the most dangerous. It is difficult to determine in
any particular household which of the three is worshiped, since they are
identical in representation and in the worship demanded. Many vil-
lagers themselves are not clear as to the distinctions or even which one
is worshiped in their own household. At least two households worship
two of these manifestations and have two separate shrines honoring
them. Two tirsuls are used to represent Nar Singh (v^ P ). During
kalratra, musical accompaniment is provided by a brass tray placed on
a butter churn and beaten by the pujari, and a large drum played by

370 APPENDIX I

a local Bajgi. Those who become possessed by this god often use iron
chains to beat themselves and may also rub ashes on their bodies. These
mortifications are done most vigorously when the most virulent form
of this god is in possession. Informants pointed out that in earlier days
this god was more feared and hence more frequently worshiped than
at present. However, he was worshiped more than once during 1957-
1958. His affinity to the incarnation of Vishnu which appears by the
name Narsingh in the great tradition was unknown to most villagers.
Until recent years most households in Sirkanda contained shrines
to Nar Singh. Then some households became divided among them-
selves on the question of whether or not to worship this god; more
importantly, other households had split up as economic units and no
longer wished to worship the god together as a household. The matter
was solved when members of the households who wished to continue
the worship took their tirsuls to a tree near the village where other
shrines are located, and deposited them there at the base of a large
rock. Subsequently other households have followed suit, and now a
number of families carry out their worship at that spot, though some
still keep shrines in their homes. The shrine at the tree resembles, in
most respects, other shrines to village gods and is in fact very close to
three village shrines. It is not, however, termed a temple (mandir), as
are other village shrines, nor is it worshiped by the village at large.
This deity may be undergoing metamorphosis from a household god
to a village god (see Appendix IB).

Manglia

This god is similar in many ways to Nar Singh. His shrines and
tirsuls are identical. He is, however, mild in his demands and punish-
ments. When he possesses a person, the person dances gently and
shivers and shakes softly as compared to the effect of other gods, and
the body of the dancer is not mortified in any way during the dance.
In Sirkanda only Rajputs worship Manglia, although no one feels
that the god could not also demand worship from others. Reportedly
the pujari for this god must be a Bajgi who plays either of two types
of drums. These types of drums are used in pujas for most household
gods except in those for Gauril. Manglia was last worshiped about a
year before I arrived in Sirkanda. He was alleged to have come orig-
inally from Jaunpur, an area in Tehri Garhwal north of Bhatbair.

Gauril

Gauril is the most prominent household god in Sirkanda from the
point of view of interest and frequency and intensity of worship,

APPENDIX I 371

though not in terms of number of devotees. He is a god whose prom-
inence is not fading, as some others are said to be. He is worshiped by
Brahmins, Rajputs, blacksmiths, and Bajgis. No one who worships
Manglia also worships Gauril.

Worship of Gauril is more complex and varied than that of other
household gods, and in Sirkanda there is more known mythology sur-
rounding him than any other household god. As a result of lore about
this god transmitted by his pujaris, Gauril's believers in Sirkanda know
the rather elaborate tale of the marriage of Gauril's parents, his birth,
his relationship to other gods associated with him, and stories of his
power and accomplishments.

Gauril is never alone. He is in inseparable association with his three
deputies or assistants: the gods Kalua and Kdld Kalua, and the goddess
Gardevi. These three are said to be three of the hundred siblings born
or created as the infant Gauril was thrown one hundred times into
scorpion plants by a jealous co-wife of his mother. The foursome is
referred to collectively under the name Gauril. A single shrine is de-
voted to them as a group, and they are always collectively identified as
Gauril in cases of possession.

The tirsuls which go to make up the Gauril shrine are four in num-
ber (^f^Y P p) representing, respectively, Gauril, Gardevi, Kalua, and
Kala Kalua. During pujas four lamps and a container of grain with a
few coins in it are placed in the shrine. For the dance a special type
of drum, used on no other occasion, is employed by a pujari. During
possession, demands are made by Gauril, but only for his assistants. He
himself never tortures people nor makes demands in his own name. He
is satisfied at receiving a little sujl (farina) fried in ghee and sugar.
The female assistant can usually be appeased by sacrifice of a cock, and
the male assistants are satisfied with sacrifice of a goat. A coconut is also
frequently offered in the name of Gardevi.1 Gardevi is well known else-
where as a form of the mother goddess, unassociated with Gauril.

Two special pujas may be performed in the name of Gauril, usually
in response to demands made during periods of possession. One is
demanded by Gardevi to honor a being identified locally as her con-
sort M asan, who does not otherwise receive notice. Masan is often iden-
tified in other areas as a ghost or a type of ghost which inhabits cre-
mation grounds. The puja in his honor in this area is called Samen
and is usually demanded only when severe forms of torment have been
inflicted on people as a result of the god's extreme anger. This puja
involves a series of seventeen dances over a period of nine days fol-
lowed by an offering at the cremation grounds. The required offerings
include five kinds of animals, water from seven places, flour from seven

372 APPENDIX I

mills made into an image of Shamshan, seven kinds of grain, seven kinds
of thorns, fruit, jewelry (for Gardevi), and sixty-four dough lamps —
altogether a complex and expensive undertaking. The ceremony oc-
curs about once every two years in Sirkanda. It had been given by one
family in 1957 and was promised by another for late 1958.

The other important special puja for Gauril is in honor of his birth,
Gauril ki ashtami, and is commonly referred to by the term for one of
the ceremonial elements in it, hariall (green). It is held upon demand
and allegedly would be held every few years even if not demanded. For
this puja the tirsuls of the three gods are plastered with rice and that
of Gardevi is ornamented with bangles on its two "arms" and a red
cloth around it, a comb, a bindl (red spot worn on forehead by plains
women), and a mirror placed at its foot. Four lamps and a container
of grain with a lamp on top are placed in the shrine. A small plot on
the earth floor in front of the shrine is sown in barley. On the fifth day
after sowing a canopy is made with four special twigs as corner posts,
a red cloth as the canopy, and a white cloth as side curtains. On the
tenth day the barley grass is reaped. A blade or two is placed in front
of each tirsul, and the rest is worn by men who place it behind their
ears, and by women who place it in their hair. Kalratra (comprising
four dances held over two days), puja, and goat sacrifices are held.
This ceremony had last been held about a year before my presence in
the village.

Agornath

The crudest god of all is Agornath. He is described as dirty and
vicious. Only he, of all the gods, would dare bother a new mother and
infant during the traditional forty-day postpartum pollution period.
In Sirkanda at present only Bajgis worship him, though he is wor-
shiped by other castes elsewhere in Bhatbair. His symbol is a single
tirsul ( P) illuminated by one lamp during puja. His kalratra includes
three dances over two days to the accompaniment of a drum and a brass
tray set on the ground and beaten. Possessed persons rub their bodies
with ashes during the dance. A ram is the appropriate sacrifice for this
god. He is reportedly less worshiped now than previously and no puja
had been given him for some time, though one was promised for late
1958.

Nag Raja

The snake king god, Nag Raja, is worshiped by thirteen Rajput
households. This god has taken on many of the characteristics of a

APPENDIX I 373

village god, although its worship is still specific to certain households.
At present only one villager has a household shrine to this god, and it
is in a chan and is no longer used. A blacksmith of Sirkanda origin who
now lives in Sahas Dhara has such a shrine in use in his house. At one
time all who worshiped the god did so in their homes. Now, however,
most Sirkanda devotees of Nag Raja worship at a central shrine or
temple which has been erected near the village and is known as a
mandir, as are the shrines of village gods.

Worship of this god in the home is similar to that for other house-
hold gods. The shrine is unique in that it is placed not in a wall niche
but on a raised platform above the floor. In addition to a single tirsul,
the shrine contains a small flag, a conch-shell horn, a small drum,
chains for mortification during possession, an incense vessel, an oil
lamp, and a brass tray containing rice, vermilion powder, and silver
bangles. Ideally the first day of every month (sakrant or sankranti) is
celebrated by kalratra and puja in the homes of devotees. If the god
possesses anyone at the kalratra, a sacrifice follows. Halwa is distrib-
uted to finish the celebration. Actually this puja is held, like the others,
in time of need only. Some Sirkanda worshipers of Nag Raja wear gold
earrings or heavy silver bracelets decorated with images of snakes'
heads in fulfillment of vows to honor this god.

The village shrine to Nag Raja was built by a village Rajput family
who had probably been directed to do so by the god. The head of this
family remains the keeper of this temple. He presides at every cere-
mony there and receives the head of the sacrificial goat. Worship at the
shrine sometimes follows kalratra at the house of a devotee, or it may
be held without prior dancing. It is performed about every three years
in order to keep the god happy, even if he is not tormenting his be-
lievers. The puja and sacrifice at the temple are the same as at the
household shrine and are a substitute for the latter. At the annual
Diwali festival the worshipers of Nag Raja go in procession to his
temple and deposit there a blazing torch in his honor. He is the only
god to be so honored.

A special ceremony involving the snake is found widely distributed
in Northern India in one form or another. In Sirkanda it is performed
to avoid the displeasure of Nag Raja and of snakes in general. If a
farmer touches, cuts, or kills a snake with his plow while working in the
fields, he is supposed to stop plowing at once and give the oxen, the
plow, and his own clothes to Bajgis. In addition he gives the field away
to them or, more frequently, leaves it fallow for several seasons. There-
after a special snake puja is held in which three metal snakes — one sil-

374 APPENDIX I

ver, one gold, one copper — are worshiped, after which a number of
Brahmins and the female household members are ceremonially fed.
This ceremony has been performed by three or four Sirkanda men.

Dhagbairu

One god, Dhagbairu, has so far appeared in Sirkanda only as the
god of incoming brides. This god now occasionally possesses one Sir-
kanda wife, though it has never caused trouble to her household and
therefore has not had a shrine devoted to it. In another household the
god has caused serious trouble in a distant chan and has begun to be
worshiped there at a shrine. The shrine differs from the pattern de-
scribed above for household gods. It is out-of-doors, under a tree. A
stone rather than a tirsul symbolizes the god, and other differences
exist. Kalratra, puja, and sacrifice are held at this shrine, similar to
those for other household gods. In Sirkanda not much is known of this
god.

One other god, known only as Bard Devta (big god), is worshiped
privately by the immigrant Bajgi in Sirkanda. The god is apparently a
local or regional god of this man's native village which he has brought
with him. He has a small shrine to Bara Devta in his home, but the
god has never bothered other villagers and is of no concern to them.

B. Village Gods

The following gods are worshiped at public temples (mandir) and
by the village at large rather than by particular households. Local
Brahmin priests usually perform the attendant rituals.

Household-Like Village Gods

Two gods, Memendia and Bhartwali, each have shrines located
west of the village. When either is to be worshiped, kalratra, puja, and
goat sacrifice are held.

MEMENDIA

The unique and memorable thing about this god was his mode of
entry into the village. Some thirty or forty years ago there was a great
epidemic in the plains, apparently plague, which reached the hills.
A number of people in Sirkanda were afflicted, and three or four died.
At this time virtually all Sirkanda villagers, and especially the young
adults, suddenly showed typical but exaggerated symptoms of super-
natural possession — they began to laugh hysterically, dance, shout, and

APPENDIX I 375

speak nonsense. Their heads were turned to one side, and they shook
violently. They shouted for drummers to come "make us dance." The
drummers came and the people danced until they were exhausted,
slept where they fell, and danced again. The dancing went on for
three or four days. The unaffected villagers recognized at once that
this was possession by a supernatural being, but they had never before
seen it on such a mass scale nor did they connect it at once with the ill-
ness and fear of illness in the village. When the possessed people were
asked who they were and what they wanted, some answered that they
were possessed by Memendia, a god who had never been in Sirkanda
before. They were told to erect a temple under a tree west of the vil-
lage, place tirsuls there, and perform a puja in his honor. The villagers
promised to carry out his instructions, and at once the god left them.
The temple was then set up and the puja held under the direction of
the local Brahmin. At the same time those suffering from the plague
were cured, and no more died. Apparently this god similarly afflicted
other Pahari villages at about the same time with similar results.

Usually possession under these circumstances affects an individual
rather than a group, and it is attributed to ghosts or spirits rather than
to gods. Memendia is the only god who regularly possesses people be-
fore the drumming. He possesses and then demands the drums, whereas
other gods are attracted by drums to come and possess people. Since his
original appearance, this god has caused very little trouble and has
therefore received little worship. Occasionally an individual villager,
one in particular, has been possessed by him. Then the drummers are
called, the god speaks, and a sacrifice is offered. The last time he was
worshiped was about 15 years ago. His shrine now consists of an un-
distinguished stone at the base of which are three tirsuls and a few
remnants of offerings. The tree under which it was originally placed
has long since disappeared.

BHARTWALI

The temple to Bhartwali consists of a rectangular stone enclosure
open at one end, about four by six feet and three feet high. It was built
in honor of Bharat, brother of Ram in the Ramayana, by the Sirkanda
Brahmin family who are still its temple keepers. The Brahmins offer
a puja and goat sacrifice there about every five years, and occasionally
the village at large does so, allegedly at the Brahmins' urging, "just for
protection." This was last done about three years ago, but interest in
the god has dropped to a very low ebb. He has possessed no one and
troubled no one for a long time. One villager predicted that there will
never again be an offering made to Bhartwali in Sirkanda. In former

376 APPENDIX I

days a goat was sacrificed here, as at most other village temples, after
the rain ceremony, but the practice has long been abandoned. Now if
anyone thinks at all about this god it is the Brahmin keeper of the
temple and his family. Presumably Bhartwali could be on the way to
becoming a household god, after beginning as a village god. More likely
he will disappear from the scene altogether.

Raghunath

The only temple in Sirkanda that is a real building is that to
Raghunath, equated by some villagers to Mahasu, who in turn is often
identified as Mahadev or Shiva. More commonly Raghunath is con-
sidered by educated people to be Ram Chandra, a manifestation of
Vishnu. In this form Raghunath is honored at a well-known temple in
Deoprayag. The Raghunath temple in Sirkanda consists of a small and
simplified reproduction of the standard two-story village house, about
eight by twelve feet, with a door opening to each floor. It is white-
washed and roofed with corrugated iron. The lower floor is used to
store anything belonging to the temple, for example, a goat donated
for sacrifice. The upper floor is the shrine containing the items used
in puja (bells, conch, utensils) and items in honor of the god, including
the ceremonial flyswitch, several chatar (hanging umbrella-shaped ob-
jects of cloth with silver fringe), and two iron staves four or five feet
in length. This temple was built by a Rajput family of Sirkanda about
sixty years ago. They control and maintain it and are its keepers
(pujaris). The Raghunath temple is the most sacred and restrictive of
the Sirkanda temples. Low castes must worship from afar, and aliens
are kept away from it unless they have come to worship. Shoes are re-
moved in its vicinity, and it is opened only for special worship. If a
stranger asks for the village temple, this is the one that will be pointed
out.

Its most important use is at the time of marriage. When a village boy
or girl is married or when any outside wedding party passes Sirkanda,
they detour to this temple where the bride or groom makes an offering
and is blessed by the pujari. If this ceremony is neglected, the wrath of
Raghunath will descend upon the couple, whereas if it is performed,
he will protect them and ensure the success of the marriage.

Other occasions upon which this god is honored are similar to those
of worship of household gods. The household of the temple keeper
and one other family occasionally hold kalratra and puja for Raghu-
nath at this temple to alleviate suffering at his hands. At such times
he possesses people much as do household gods. Other families oc-
casionally make vows to this god, and if their wishes are fulfilled they

APPENDIX I 377

carry out the vows, either with a cash donation to the temple or a puja
and sacrifice at the temple with the pujari officiating. At village-wide
kalratras on any occasion Raghunath is likely to possess a particular
person (in Sirkanda this person is a Rajput woman) much as Devi
does. He may also be honored with a kalratra by the village at large on
almost any occasion when powerful help is felt to be needed. This
temple is the only one which attracts outsiders to worship in Sirkanda.
Occasionally (once in my year of residence) someone of Pahari descent
will come from as far away as the valley to make or fulfill a vow at this
temple. Then the temple keeper keeps the person as his guest and re-
ceives the offerings. More frequently people from neighboring villages
come to use this temple.

In association with Raghunath are two gods, Parasu Ram and Bhairu,
images of which are said to be just inside the door of his temples. Parasu
Ram is an incarnation of Vishnu, according to the great tradition.
Bhairu is identifiable as Bhairo, Shiva's chief executive officer. He is
often worshiped separately in Pahari villages at a shrine identified by
a rounded stone similar to that indicating a Devi temple site. Such a
shrine exists in a village close to Sirkanda. This is further evidence of
the lack of clear-cut distinction between gods identified with Shiva and
those identified with Vishnu in this area. Bhairu is also the patron for
practitioners of the art of casting spells and counterspells. He is wor-
shiped privately by such people to assure his continued protection.

Children's God

Lhesania is a god connected with the local celebration of spring,
known as Phul Dalna (flower placing) and Popria Sakrant (see Appen-
dix II). He is worshiped by children, who are the primary participants
in these ceremonies, and he is worshiped only at the time of the latter
ceremony. He never possesses anyone or makes his presence felt. His
shrine is a stone in a field near the village. Its location may vary from
year to year because it is determined by children, who select it just
before the ceremony each year. At the yearly ceremony the god is
honored with offerings of flowers, flags, and food.

Agricultural Gods

The two following deities are worshiped almost exclusively as an
aid to crop production.

MATRI

The crest of the hill upon the flank of which Sirkanda and its fields
lie is known as Matrl kd ddnda (mother goddess peak). On this peak,

378 APPENDIX I

1,200 feet above Sirkanda and somewhat difficult of access, is a small
U-shaped stone enclosure which is the temple honoring this deity. In
it are a grain-pounding pestle and the remains of many small offerings
appropriate to a goddess, such as coins, bangles, mirrors, combs, and
lamps. There is some confusion as to whether this goddess is singular
or plural. Atkinson (1884b, pp. 75 f.) lists sixteen Matris or "divine
mothers" who are widely worshiped under this and other names in
India, Nepal, and Tibet by the "aboriginal and non-Aryan tribes."
He describes the goddesses as unrelated to the worship of "female
energy" or of the "consort of the great divinities" of contemporary
Hinduism. It is to one of these (or possibly to a personification of all
of them) that the Matri shrine in Sirkanda is evidently devoted. The
Bhatbair village of Kanda has a shrine to a sister goddess of Matri
(perhaps one of the other sixteen Matris), according to Sirkanda resi-
dents, but otherwise they know little about her.

Closely associated with the shrine, and resident on the same hilltop,
is a band of sprites or fairies referred to as matris (matriya) (see p. 113).
Such beings frequently occupy hilltops in the Pahari area. The relation-
ship between the sprites and the Matri goddess or goddesses is unclear
to villagers, possibly because both occupy the same location and have
therefore become identified, whereas there is apparently no necessary
connection between them. Contributing to the ambiguity is the con-
fusion between Matri as a single goddess, as Sirkanda villagers view
her, or as sixteen goddesses, as she is known elsewhere in the Himalayas.
The latter perhaps merges into the idea of the band of fairies in the
view of Sirkanda villagers.

Worship is offered at the Matri shrine in Sirkanda at the beginning
of a variety of undertakings, much as Atkinson has described for Matri
elsewhere in the hills, but especially in association with agricultural
rites. Ideally, as soon as harvesting and threshing are completed each
season, Sirkanda villagers come as a group to this shrine, where a puja
is held under a Brahmin's supervision and a goat is sacrificed in honor
of the goddess, to show appreciation for the crops and to ensure suc-
cessful future crops. In conjunction with this ceremony a similar one
is held at the shrine of the earth god, Bhuia, described below. Neither
of these rites has been held in recent years.

If rains are either too heavy or too light for good crops, a puja may
be held at both of these shrines and often at the shrine of Devi as well.
Usually the Pandavas, Raghunath, and (in former days) Bhartwali re-
ceive goat sacrifices in their honor after the rain puja, though these are
intended to show respect for their power only, and are not an integral
part of the ceremony. The difficulty of reaching Matri's shrine has per-

APPENDIX I 379

haps been a factor in the recent trend to omit her from inclusion in the
rain puja, and may be partially responsible for cessation of the harvest
puja.

Matri is also honored by hunters who want to ensure success in
getting game. A hunter goes to her shrine and leaves a small offering
before setting out.

BHUIA

Immediately above Sirkanda is a large oak tree near which are the
temples to Bhuia (the earth god), Devi, the shrine for Nar Singh, and
the repository for the cremation stones of high-caste dead (map 3).
Bhuia is thought of as the god of the soil and its products. His temple
consists of a stone under which tirsuls have been placed. He is wor-
shiped both independently and in conjunction with Matri at harvest
time. Bhuia is also worshiped in conjunction with either Matri or Devi
in the rain puja. He, like Matri, is never given a puja unless one is also
being given to the other agricultural deity. In the village Bhuia is an
old god whose origin and affiliations are unknown to his worshipers.
As an earth god he is well known on the plains. The hill villagers say
his temple is to be found in many villages, always located under a tree
and often near a Devi temple.

Paramount Village Gods

Two gods, Devi and the Pandavas, are of greater village-wide im-
portance than any others.

DEVI

The goddess known simply as Devi (goddess) is honored in several
contexts in Sirkanda. She is a manifestation of Durga, the divine female
or mother goddess. Her temple is a stone enclosure, roofed over loosely
with branches and located at the base of the large oak tree near the
village which is the site of several village shrines. Her powers are varied,
but important among them is her function as a deity of earth produc-
tivity, along with the two gods just described. When the harvest cere-
mony is held, she is one of the deities honored.

She is the focus of the rain ceremony at present. This ceremony,
known simply as Manana (to please), is held whenever the weather
threatens to harm crops — usually late in a harsh dry season or a harsh
rainy season. It consists of a kalratra held near Devi's temple, a puja
under the direction of the local Brahmin, and a sacrifice to Devi. The
sacrifice is usually a goat, but it may be a coconut instead. Each fam-
ily attending brings a container holding seven grains, which is circled

380 APPENDIX I

over the head of each family member to bless them and remove any
evil that may be in them. Then the grain is thrown over the temple
and surrounding area. Food is then cooked and distributed among the
participants. In former days goats were also sacrificed to Matri and
Bhuia, and pujas were held at their shrines as part of the ceremony;
the next morning additional dances might be held at the Pandava
temple and the Raghunath temple, and goats were often sacrificed there
as well as at Bhartwali's temple. Such rites were in honor of these
powerful gods but were not part of the rain ceremony itself. It has
been the custom to purchase the sacrificial animals and the distributed
food by village-wide subscription. Now it is said that people are not
interested and will not contribute. Only a few households who own
land and chans in one area immediately east of the village are enthusi-
astic about the rites now. They participate as a group and give their
offerings together, whereas others participate only in the enjoyable
kalratra portion of the proceedings. When this rite was performed
during my stay in the village, the Devi temple was the site of the en-
tire evening's proceedings and of the only sacrifice — of a coconut. The
following day dances were held at the Pandava and Raghunath temples.
The ceremony did not produce significant results, but many people
expected this outcome because of the niggardly attitude of most house-
holds and the resultant paucity of sacrifices.

Apparently the harvest ceremony was virtually identical wth the
rain ceremony in the days when it was held. At kalratras associated
with these ceremonies, as at those for the Pandavas and Raghunath,
villagers are likely to be possessed by any of the village gods. Such pos-
session is only for the pleasure of the gods concerned. They do not
usually talk or make demands, nor do they overexert those in whom
they dance.

Devi is typical of the village gods who possess people on such oc-
casions. Possession by Devi, as by other village gods, is of a different
order than that by household gods. She does not possess indiscrimi-
nately. Rather, she possesses only a certain person. Presently in Sirkanda
this person is the barber's wife. For three generations — since Devi first
possessed the barber's grandmother during a cholera epidemic and
saved the village — she has possessed a female member of this family.
The family maintains a household shrine to her as a result. This is the
only instance in the village of a household shrine to a village god.
According to villagers, if no suitable woman of barber caste were in
the village she would possess someone else, not necessarily of low caste.

Everyone in the village honors Devi. Kalratras and pujas may be held
for her at her temple or at the Pandava temple whenever someone

APPENDIX I 381

wishes to ask questions of her. The worship closely resembles that for
household gods, though local Bajgis play the accompanying drums
and the pujari is the highly esteemed Brahmin, Har Nam, who offici-
ates in many ceremonies for high-caste people but not in those of house-
hold gods. Devi also participates in the ceremony dedicated to the
Pandavas, described below. Devi is one of the deities who appears at
almost every village-wide kalratra no matter in whose honor it is being
given. In this respect she is like the Pandavas and Raghunath and
unlike other deities.

Devi is known and honored throughout the Pahari area, but only
one other village in Bhatbair has a temple for her. The largest temple
on the 40-mile Mussoorie-Tehri trail, surmounting the highest peak
along the trail, is a temple to Devi at Kuddu Kal. It is the most honored
temple in this region of Tehri Garhwal. According to some villagers
the Devi of the other temple is a sister to the Sirkanda one, but accord-
ing to others the Kuddu Kal temple is dedicated to her head while
the Sirkanda one is to an arm. Sirkanda villagers often visit the Kuddu
Kal Devi during an annual fair held there. Coconuts are offered to her
there. A former pujari of that temple is said to have objected to the
great slaughter of goats that had taken place, and he proclaimed the
substitution of coconuts as suitable sacrifice.

The story of the founding of the Kuddu Kal temple at the insistence
of Devi herself, who appeared incognito as a young woman in distress,
is well known to villagers. Also it is known that a sacred stone (bindi)
representing Devi determined the precise spot for the temple. Another
popular story is that of a blacksmith (allegedly a maternal relative of
the Sirkanda blacksmiths) who responded to an appeal made by Devi in
a dream, to remove a thorn from the foot of her mount, a tiger. In
reward the blacksmith was granted a large tract of land by the Raja
of Garhwal, and he became wealthy. The goddess attained such renown
that the income at her temple was great enough to finance the building
of the present imposing structure in that remote spot. It is believed
that Devi still rides her tiger in the vicinity of all temples in her honor,
including that at Sirkanda. Thus, Devi is an important goddess in the
Hinduism of this entire region.

PANDAVAS

The most honored deity or complex of deities in Sirkanda, and
as important as Devi and Mahasu in Bhatbair and the surrounding
Pahari region, are the five Pandava brothers, whose story is told in the
Hindu religious epic, the Mahabharata. These deified culture heroes
are reputed to have lived for many years in the hill area of Kedarkhand,

382 APPENDIX I

latterly known as Garhwal. The villagers' feeling of respect and ad-
miration for the Pandavas is derived from a long and close association
quite different from the vague feeling they have developed for most
other Hindu deities and heroes as a result of hearing about them from
traveling priests and mendicants. The Pandavas may well be indigenous
objects of worship in this area who have been universalized to become
a part of the literary tradition of Hinduism. In most other areas they
are not considered to be gods. Every village in the Pahari region has
its shrine to the Pandavas, and every villager honors them. In Sirkanda
the descendants of one man who was an especially devoted follower of
the Pandavas are still identified in that light, though they comprise
eight households now. The Pandava brothers are known in Sirkanda
as Yudarshter (Yudhishthira), Bhim (Bhima), Arjan (Arjuna), Nikul
(Nakula), and Sedev (Sahadeva). They are worshiped as a group. In
addition, their common wife Dropti (Draupadi) is included in wor-
ship, as well as a second wife, not mentioned in the Mahabharata, but
known here as Basudunta.

The Pandavas' temple in Sirkanda and other nearby villages con-
sists of a small stone platform against the side of a village house. On
this platform is a stone altar upon which are kept a tirsul representing
the gods, and the wooden maces, bow, and iron staves representing
the weapons used by these gods in their epic battle, and used during
worship by the dancers they possess. Near the temple or mandir is a
small square house known as the Pandavas' chaurl (umbrella, shelter).
It is used as the place of worship or puja room for the presiding Brah-
min and the storage place for his ceremonial apparatus during im-
portant ceremonies. In many villages a convenient room in a house is
used instead. Finally, there is an open, flat area adjacent to the mandir
and chauri which is called a madan ("stage"). This is the site for
village-wide worship and dancing in honor of the Pandavas. It is also
a gathering place for social occasions not directly related to their wor-
ship. It is the ceremonial center of village life and the site of most
village dances.

The Pandavas are worshiped on five major occasions: (1) when
there is trouble that they may be able to relieve, or when a vow has
been made to them on such an occasion, (2) at the annual Diwali festival,
(3) at the annual Navratra observance, (4) at the occasional performance
of the Mundkile ceremony reinforcing village solidarity and security,
and (5) after the performance of a rain or crop ceremony. In addition,
dances may be held at the madan at the time of weddings or any other
village celebration. No case was recorded in which they were thought
to have punished people for failure to worship.

APPENDIX I 383

Worship of the Pandavas is much like that for other gods: a kalratra
is held, followed by a puja, sacrifice of a goat, and distribution of halwa.
Ideally a number of people become possessed by various village gods at
dances honoring the Pandavas. Possession occurs with such regularity
that villagers admit that feigned cases far outnumber those of "real"
possession. Such possession is almost invariably for the pleasure of the
gods involved and does not produce a session in which people hear
the gods' demands or seek their advice. In any event, five men — not
always the same five but in recent years always Rajputs — become pos-
sessed by the five Pandavas and dance with the symbolic weapons kept
at the shrine. One Rajput woman, who is regularly possessed by Dropti,
their wife, performs a stereotyped dance with a tray of rice revolving
in one hand and a cup in the other. Another Rajput woman becomes
possessed by Basudunta, the second wife of the Pandavas. A Bajgi and
his wife become possessed by Kali Das and Kali, a god and goddess as-
sociated with the forces of evil against which the Pandavas fought,
according to villagers (other interpretations might be given by more
educated Hindus). The possession is not random and shifting, as with
household gods. Other villagers traditionally become possessed by gods
not closely or not at all linked with the Pandavas in mythology but
linked in the minds of villagers to this puja. The barber's wife becomes
possessed by Devi, a Bajgi man becomes possessed by Blr (identified as
a relative of Devi), and a Rajput woman is possessed by Raghunath.
If any other village god possesses people, it is likely to do so at the time
the Pandavas are worshiped. At any one moment some or all of these
gods may be dancing, but ideally sometime during the kalratra each
makes an appearance. In reality it often happens that only a few of
the gods appear, and on some occasions none appear. Depending upon
its purpose, the dancing may or may not be accompanied by a puja
and sacrifice, but in any event it honors the Pandavas and their as-
sociates and, to a lesser extent, other village gods.

The Mundkile ceremony is an elaborate affair, the immediate aim
of which is to protect the village by excluding malevolent ghosts and
alien gods. It is a form of boundary rite (cf. Crooke's discussion of
boundary rites, 1926, pp. 83 ff.). Since it is of such importance to the
community as a whole, in contrast to the rites of household gods, and
since it differs considerably from the routine worship of the Pandavas
in other circumstances, it will be described here.

The ceremony is held every five to eight years or whenever the need
is felt, usually in the period October through December, after the
autumn harvest. It is usually financed by village subscription, though
in some circumstances a well-to-do family may sponsor it.

384 APPENDIX I

The first step is a hariali ceremony — the planting and reaping of
barley. All villagers contribute a half-seer (one pound) of barley to
be planted or used in offering, and when it is reaped all get shoots of
barley to place on the doors of their houses as blessing, and some of the
soil in which it grew to spread on their fields. This is similar to the
ceremony of the same name described above for Gauril.

Each night for eight nights during the growth period of the barley,
a kalratra is held at the madan. On the last night the Mundkile cere-
mony and the Pandavas' puja are held. On the evening of the Mund-
kile ceremony the officiating Brahmin goes with some villagers and
affiixes a peg (to which the name of this ceremony refers) at each of the
four directional corners of the occupied area of the village. He pours
some liquor at each peg and sacrifices a pig, a ram, a cock, and a pump-
kin, one at each of the four corners. As he does so he entreats the
Pandavas and their wives and Devi and Kali to protect the village
from alien spirits and evil of all sorts. Then the villagers return to the
madan, where the final kalratra is held. Late in the night, after Dropti,
wife of the Pandavas, has possessed some village woman, a goat is sac-
rificed at the madan in her honor. Immediately thereafter a second
goat is sacrificed to Basudunta, second wife of the Pandavas. The next
morning, after the reaping and distribution of the barley which com-
pletes the hariali, another goat is sacrificed in the puja room used for
the hariali — this one in honor of Devi. A coconut, which has formed
part of the sacred apparatus in the hariali, is also offered to Devi at
this time. Thereafter goats may be sacrificed at the shrines of Raghu-
nath and Bhairu if these are present in the village (Sirkanda has only
the former).

The hariali portion of the ceremony is evidently performed in honor
of Devi, but she and all the Pandavas are asked to protect the village.
One informant from a nearby village attributed the whole performance
to the honor of Durga, the female deity or mother goddess with whom
Devi is probably best identified. Most villagers do not identify Devi
with the name Durga, but they recognize the significant role of Devi
in this ceremony. However, the over-all village attitude is that this is a
Pando ki puja — a ceremony for the Pandavas. In this ceremony we have
further evidence of the interrelation of these village gods and their
distinctness from household gods, with which they are never confused.
Household gods do not appear during worship of village gods.

In most villages in Bhatbair a male buffalo is sacrificed to climax
the Mundkile ceremony, though Sirkanda residents deny that they do
the killing themselves. The ceremony is dedicated to the Pandavas,
Dropti, Basudunta, Devi, and Kali. The buffalo is led around the vil-

APPENDIX I 385

lage while drums play. Then it is fed fine foods while standing in a
circle drawn on the ground. Rice is thrown on the buffalo to the
accompaniment of mantras chanted by the Brahmin. When the buffalo
shakes, signifying that the gods have accepted it as an offering, a village
elder (usually the sayana) strikes it with a sword in an effort to behead
it. Thus far the details of sacrifice are identical to those of the goat
that has been sacrificed previously in this ceremony. If the buffalo were
felled in a single blow it would be a very good omen; if it were to run
away and escape it would be a very bad omen. Neither actually occurs.
The buffalo is invariably wounded, onlookers join in hacking the
tendons of its legs so that it will not escape, and eventually it is hacked
to death. It is then dragged to the madan, where more mantras are
chanted over it. Thereafter it is reportedly given to shoemakers, who
use the hide and eat the flesh. A goat is then offered to the same gods.
The sayana holds it and then throws it over his shoulder, and someone
else kills it when it hits the ground. Its meat is distributed to all as-
sembled. Thereupon the sayana is possessed, shivering from terror
at having killed a buffalo, apparently a ritually dangerous act. He is
covered with a blanket and taken into the puja room until he is re-
covered. It is not clear what possesses him at this time, but a blanket
is a standard protection against ghosts, and it may be that he is vul-
nerable to their attack after performing this act.

Buffalo sacrifice may also be given the Pandavas in fulfillment of
vows. The most recent buffalo sacrifice in the Bhatbair area occurred
in a Mundkile ceremony three years previous to my stay there. Ac-
tually two were then sacrificed, as two disputing factions were unable
to agree on the matter. The Mundkile and associated ceremonies took
place without buffalo sacrifice in a nearby village during my residence,
and in Sirkanda about three years previously.

MAHASU, MAHADEV

The god Mahasu, who is well known in the area, is described by
some Sirkanda villagers as identical with Mahadev, who in turn is
identical with Shiva. In Jaunsar-Bawar, Mahasu is a distinct god, or
more accurately four gods (Saksena, 1955, pp. 40 ff.).

Mahasu, though sometimes identified with Shiva, is not honored
under this name on Shiva's birthday. Those inclined to such worship
go to a large Shiva temple at Sahas Dhara or elsewhere where a full-
time Brahmin priest is in charge. At one of these distant Shiva temples
Sirkanda villagers occasionally have a Brahmin perform the special
worship in their behalf called japp (repetition) during the sacred month
of Saon (July-August) to secure relief from chronic difficulties.

386 APPENDIX I

Many villages are devoted above all to Mahasu or Mahadev just as
Sirkanda is to Devi and the Pandavas. Mahadev is said to have special
power to return illegally acquired lands or houses to their rightful
owners. The rightful owner simply takes a clod or stone from the land
or house in question, tells Mahasu the circumstances, and asks re-
venge on the illegal possessor. Then the occupant must leave, or he
will die under the power of the god. In the past this god, identified
as Shiva, and his wife Parbati, have been honored in the spectacular
and elaborate rope-sliding ceremony peculiar to the Himalayas. Since
this ceremony has been described in detail elsewhere (Berreman, 1961b),
it can be passed over lightly here.

The feat which forms the focus of this ceremony is performed by a
person of the Beda caste, a caste of dancers and entertainers indigenous
to these hills. Ideally it used to occur about once every ten or fifteen
years in any one area. It was outlawed by the British and has not oc-
curred in Bhatbair for 30 years, though it continues to be performed
in attenuated form in other Pahari regions. The immediate motiva-
tion for the ceremony in a particular village was to alleviate serious
village-wide difficulties such as sickness or poor crops, attributed by a
shaman to the displeasure of this god. In the ceremony itself, after
elaborate preparation sponsored by an entire village and supported
with donations from the surrounding region, the Brahmin priests
would bless the Beda and worship the god. Then, sitting on a small
wooden saddle, the Beda would slide down a rope incline over the
edge of a cliff or down a steep slope. By so honoring the god, not only
the sponsoring village derived benefit but all those who contributed
and all in attendance, which included people for many miles around.

C. Ancestors

Certain ancestors come to be worshiped much like household gods,
and under similar conditions. The spirit of a deceased man who comes
to torment the household in which he had lived is known as rati, and
that of a deceased woman is known as hanthia (see p. 110). They are
symbolized in the household by a hand-sized piece of wood on which

the following figure is drawn: ^\T^ • This is kept in a wall niche, and

when it is to be honored it is placed in a wooden container of grain
and is illuminated with a lamp, just as are the shrines of household
gods. Sometimes the symbol is engraved by a goldsmith on a silver
plaque and is worn by a woman as a neck pendant during worship and
on festival occasions.2 When the spirit torments the family, illness —

APPENDIX I 387

especially skin disease — and dry cows are the usual symptoms. A sha-
man makes the diagnosis. Then a kalratra is held, in which a house-
hold member becomes possessed by the spirit. The spirit may also come
during a kalratra honoring a household deity. The person possessed
dons a cap and carries an umbrella or cane if possessed by a male an-
cestor, or wears a head scarf if possessed by an hanthia, and weeps as
he or she dances. These symbols always identify an ancestor spirit. A
knowing murmur goes around when, during a kalratra, someone be-
gins to dance with these accouterments. No sacrifice is made after
the dance, but halwa and purls (unleavened bread fried in oil) are
served.

APPENDIX II: CALENDRICAL
RITES IN SIRKANDA

Following is a brief description of all the calendrical ceremonies ob-
served in Sirkanda, as they are practiced and understood by the people
who observe them (see table 1, p. 123). These ceremonies can be com-
pared with the detailed account of calendrical observances in Kumaon
presented by Atkinson (1884b, pp. 55-63).

Phul Dalna (Chait, March-April)

Each day during this month the preadolescent children of every
household traditionally pick blossoms of whatever flowers are avail-
able. Each morning before sunrise they place the blossoms on the door-
sills of their houses (hence the name of the observance, "flower plac-
ing"). In reality relatively few engage in this performance throughout
the month with any degree of regularity. The exact significance of
the custom is unknown, and it has no evident Sanskritic affinities.1
It is associated with the coming of spring and leads up to the following
festival, Popria Sakrant.

Popria Sakrant (first day of Baisak, April-May)

The first day of every month — the date upon which the sun is
supposed to pass from one constellation to another — is called Sakrant.
In some areas every such date is marked by festivals; in some, none are
observed; and in some, Sirkanda included, particular Sakrants are

APPENDIX II 389

celebrated. The first of Baisak is the most important of the three
Sakrants celebrated in Sirkanda. It marks the end and climax of the
month-long Phul Dalna. It is known as Popria Sakrant after the thin
rice bread called popper that is served on this day. Doors and other
woodwork of village houses are decorated with spots of the white rice
flour used in preparing the poppers — the only annual occasion for
which houses are decorated in Sirkanda. Children are the primary
participants in this festival. Before sunrise they go to the watering
place, bathe, and pick seven kinds of flowers into small baskets. They
return home and place the flowers at the doors of their own homes.
Their parents refill the baskets with food — poppers, puris, and a
vegetable mixture. Each child then passes in and out of the doorway
of his house seven times, receiving a vermilion mark on his forehead
and back from an elderly household member each time he enters
and exits. Thereafter all the village children go in a group to a place
in the fields which they have selected or where they have placed a
stone to serve as the shrine for worship of the children's god, Lhesania.
They place flower petals and small flags on the shrine and eat their
food offering, and the celebration is over. In former days Bajgi women
danced on this day and were given grain in return. The significance of
the festival beyond its obvious association with spring is unknown. At-
kinson reports a similar ceremony called Min or Haldua Sakrant in
Kumaon and Garhwal. It has no evident Sanskritic affinities.

Ghorlia Sakrant (first of Jet, May-June)

On this day images of a species of wild goat which is called
GhoRl, are made of sweetened dough and are fried in oil. They are
made by the high-caste women and are given to their children, who
play with them, pretend to kill them, and then eat them. They are
also given to low-caste people of all ages. The significance of this
performance is unclear. Perhaps it benefits the recipients of the images.
It is the only annual festival in Sirkanda which fails to appear in
association with any festival mentioned in the literature I have sur-
veyed. It bears some superficial resemblance to Magh Cauth as de-
scribed by Planalp (1956, pp. 355 ff.), but there is no evidence upon
which to base a relationship between the two.

Taulu (Jet, May-June)

This annual fair has no detectable religious implications. It re-
sembles Diwali in its timing (it follows the spring harvest as Diwali

390 APPENDIX II

follows the fall harvest) and in the gaiety of the event. It may be
related to a widespread non-Brahmanical Pahari harvest festival (cf.
Atkinson, 1884b, p. 64). It is described briefly in chapter 4.

Rakri (full moon of Bhado, August-September)

On this date Brahmins go from house to house tying string
bracelets (rakri) on the wrists of household members. In return the
Brahmins receive from an anna to a rupee from each household. A
number of Brahmins come and are never refused, so that each person
usually receives several of these charms. This is supposed to be aus-
picious for the recipient. One informant commented that, though he
did not know the exact purpose, it seemed to him that the performance
benefited the Brahmins more than anyone else. It has no connotation
of brother-sister devotion as it does in some plains areas. It is readily
identifiable with Raksha Bandhan, celebrated in Dehra Dun and else-
where, and with Silono or Saluno, reported by Lewis (1958, pp. 208 f.)
and Marriott (1955a, p. 198), and U pa-karma of Kumaon (Atkinson,
1884a, p. 850). Its origin and exact implications are obscure to Sir-
kanda people.

Jenem Ashtmi Barat (eighth day of Bhado, August-September)

This is the anniversary of Lord Krishna's birth and is an all-
India Hindu festival. In Sirkanda it is said to consist of a day-long fast
followed in the evening by a feast of halwa, kir (sweet rice boiled with
milk), and fruit. It is ignored by most villagers.

Kanagat (first fifteen days of Asoj, September-October)

This is the period in which the spirits of the family dead are hon-
ored. A Brahmin determines the precise day (within the fifteen-day
period during which such worship occurs) upon which each deceased
relative is to be honored, apparently keeping in mind his own schedule,
as he has many such observances over which to preside. On the ap-
pointed day the family prepares kir, which is fed to the Brahmin and
family members. The Brahmin conducts a puja and receives charity of
R. 1, uncooked rice, flour, and sugar. The stated purpose is to comfort
the souls of the deceased and prevent them from becoming restless and
troublesome to the living. Informants stressed that there is consider-
able social pressure to carry out the ceremony to demonstrate respect
for one's ancestors. The ceremony is widespread in India, though it

APPENDIX II 391

is not mentioned by scholars as part of the great tradition of Hinduism.
It is known in some areas as Srdddhd or Shrddh.

Nauratra (first nine days of second half of Asoj,
September-October)

This festival, also known locally in Sirkanda as Nardtd, is the
nine-day Hindu celebration honoring the female form of Shiva as
Durga, Kali, or other manifestation. In Sirkanda the festival is known
but largely ignored. There is a tradition, however, of holding a village-
wide kalratra on one or more of the nights of this period to honor
the village gods. As is usual on such occasions, Devi, who is the local
form of Durga, is among the deities expected to dance, and the local
Brahmin performs worship in her honor. The traditional Hindu
festival, Dashera, climaxing the nine nights' celebration, receives no
recognition at all in Sirkanda.

Gaojiman (fifteenth day of Katik, October-November)

This is the ceremony honoring cattle (the literal translation of
its title), which on the plains follows immediately after Diwali and is
widely known as Gobardhan. In Sirkanda it falls on the day that
elsewhere is Diwali. In Sirkanda on this day, family members feed
their cows and oxen a special boiled grain mixture, followed by fried
delicacies, which on other occasions are reserved for human consump-
tion. Oil is rubbed into the animals' horns, their hooves are washed
with water, and ghee is burned so that the smoke drifts over them.
Finally, family members fold their hands in respect before the animals
and put vermilion marks on the animals' foreheads. On this day
family members, too, eat special delicacies. Two days later (the seven-
teenth day of Katik), called Mavas, the animals are given salt to lick,
and again their horns are oiled, ghee smoke is wafted over them, and
marks are put on their foreheads. This, villagers say, honors the sanc-
tity of the cow.

Pahari Diwali (fifteenth day of Magsir, November-December)

This, the festival of lights, is the greatest festival in Sirkanda
and one of the greatest in India. It is supposed to celebrate Ram's
victorious return from his conquest of Ceylon and to honor Lakshmi,
goddess of wealth. In Sirkanda many people are vaguely aware of
these implications but are little concerned with them.

392 APPENDIX II

In Sirkanda, as in most neighboring villages, Diwali is celebrated
exactly a month later than on the plains. This is a long-standing Pahari
custom, from which some villages in Bhatbair have deviated in recent
years as they have come to emulate plains ways on many matters.
Plains Diwali falls precisely in the midst of the frantic activity of fall
harvest and planting in the hills. By postponing it a month, hill vil-
lagers can fully enjoy it.

Diwali is of two days' duration in Sirkanda, "big Diwali" being held
the day before "little Diwali." This is the reverse of the plains custom
and was explained by an informant: "Here whatever comes first is
biggest." For the Diwali festival many Sirkanda people who normally
live in chans come home to the village, so that this is the time of
greatest population concentration. There is much visiting, drinking,
eating of meat, dancing, singing, some gambling, and general merry-
making. Throughout the harvest and planting season people have been
wearing their oldest, dirtiest, and most ragged clothes. On Diwali they
suddenly appear clean and brightly dressed in new or almost-new
clothes. The carefree atmosphere is in marked contrast to the fatigue
of the harvest and planting just past. Hospitality and camaraderie
are at their height.

On big Diwali delicacies are cooked and served in every household.
Small oil lamps made of dough are placed in windows and doorways of
some houses (many do not do this, and it is apparently an imitation of
the plains custom). Late in the evening those who worship Nag Raja
proceed to his shrine is procession with a large torch consisting of a
long stick with a cucumber on the end, from which protrude many
small burning sticks. The torch is left at the shrine to honor the god.

Many villagers go to the top of a small knoll near the village, where
the men and boys "play with fire." A pitchy pine chip is tied to
the end of a light rope-like creeper about four feet long. The chip
is ignited and then swung around with considerable skill in arcs and
circles by its owner. Several men do this at once, and many villagers
watch the display. Thereafter a kalratra is held at the madan or open
space devoted to worship, next to the Pandavas' temple. This is one
of the occasions when various village gods are likely to dance, but
the main purpose of the dance is entertainment. Both women and men
dance, and several folk-dance styles are displayed.

Little Diwali is a repetition of big Diwali on a reduced scale with-
out the Nag Raja puja and usually with few or no lights. By its timing
and content one is led to suspect that the whole celebration is derived
from a regional tradition, probably a harvest ceremony, that has been
assimilated to some features of Diwali of the great tradition (cf. Atkin-

APPENDIX II 393

son, 1884b, p. 64). This seems especially likely in view of its similarity
to the nonreligious taulu following immediately after the other yearly
harvest.

Kicheri Sakrant (first of Mau, January-February)

This festival is named for the boiled mixture of rice and dal
(kicheri) that is the characteristic dish of the day. It is known on the
plains as Makar Sankrant. On this day Brahmin men are supposed to
take a ritual bath. All high-caste village households cook kicheri with
butter. In the morning Bajgis go from house to house playing their
drums and receiving kicheri in return. On the same day Brahmins go
through the village and are given charity in the form of uncooked
grain, or money. At this time Brahmins appear who are seldom other-
wise seen. It is primarily a day of charity-giving.

Panchmi Basant (fifth day of Mau, January-February)

This day is widely celebrated in Northern India as a heralding
of spring, to which the title refers. In Sirkanda it is too early in the
season to be a fitting spring festival, and other such festivals occur
later as described above. However, Panchmi Basant is celebrated in a
very simple fashion in Sirkanda. The village barber takes barley shoots
from the fields and goes from house to house in the village affixing a
few such sprigs with cow dung at the two upper corners of every
outside door. In return each family gives him one pound of grain. In
addition the Bajgis play their drums and are rewarded with sugar and
grain. A few families prepare special food for themselves on this day.

Shiv Ratri Barat (thirteenth day of Phagun, February-
March)

This fast honors the birth of Shiva, according to villagers. Families
who are interested fast throughout the day. In the evening these
families feast on specially prepared foods. Not many families observe
this occasion and those who do, do so inconspicuously.

Other Annual Observances

This completes the yearly ceremonial round in Sirkanda. In addi-
tion, the month of Saon (July-August) is thought to be especially
sacred or auspicious for worship. During Saon two special types of

394 APPENDIX II

worship are performed by learned outside Brahmins: japp, the repeti-
tion of sacred texts in honor of Shiva at a Shiva temple to alleviate
any persistent trouble, and rudrl patt, recitation of sacred texts in the
home, usually to bring about conception in childless wives. Eclipses
are announced in advance by Brahmins. When they occur they are
generally observed by fasting followed by feasting. Pregnant women
and people whose horoscopes indicate that the particular eclipse is
unfavorable for them, avoid looking at it.

APPENDIX III: LIFE-CYCLE
RITES IN SIRKANDA

Atkinson (1884b, pp. 64 ft.) gives a detailed account of life-cycle rites
as they ideally occur among educated, well-to-do people of Kumaon.
He comments with regard to this category of ritual:

It is followed as a rule only by the educated and orthodox and ... its use
has not yet permeated the masses, nor are its rules, except in a very abbrevi-
ated form, understood by many of them. The ritual for temple use has been
compiled by a class for their own purposes. . . . The village-deities have no
formal ritual committed to writing and in general use, so that the ceremony
is a meagre imitation of that in use in the orthodox temples and varies with
the celebrant, (pp. 64 ff.) With regard to these and all other observances their
length and character would seem to depend on the means and inclination of
the person who causes the ceremony to be performed. The poor man obtains
a very shortened service for his few coins, whilst the wealthy can command
the entire ritual and the services of numerous skilled celebrants, (p. 64)

The interested reader should consult Atkinson for detailed data
on the rituals discussed below. Here are given only the outlines of the
important events and rituals of the life cycle as performed and per-
ceived by Sirkanda villagers.

A. Birth and Childhood Ceremonies (see Atkinson, 1884b,
pp. 79 ff.)

Birth takes place in a room of the family house with the mother
lying on a leaf-mat rain covering. There are no professional midwives,

396 APPENDIX III

but any one of several experienced village women assist at the birth.
These women may be of any caste. At present a Rajput and a Bajgi
are the two who most often assist. The umbilical cord is cut by the
mother herself. The afterbirth is put in a basket and buried by the
husband or other male family member. Afterwards the mother bathes
herself and the baby, with the assistance of other household women,
and puts on different clothes. The room is replastered with dung.
Immediately after birth the mother eats sweetened porridge and a
mixture of milk and ghee to strengthen herself. She continues to do
so daily for a few days. She is supposed to bathe herself and the child
daily for about thirty days, using cow's urine as well as water. Actually
the mother and child should bathe regularly with cow's urine for
several years, but people content themselves with an occasional rinse
of arms and legs in cow's urine. The mother's period of ritual pollution
is supposed to last forty days but, as with menstrual pollution, this
is not observed in Sirkanda. "We can't afford those things — there is
too much work to do." Therefore a woman is back in normal circu-
lation about ten days after giving birth. For those ten days no one
else takes food or drink from her hand, and she does no work and
does not sleep with her husband. She also takes special precautions
to avoid the attack of churail, the ghosts of women who have died in
childbirth.

Das sotan, the child's first ceremony, is held on the tenth day after
birth. Friends and relatives may be invited to attend, and if so, they
often bring small gifts of jewelry and adornment for the child. After-
wards they are fed halwa. On this day a Brahmin is called and he
begins preparation of the child's horoscope, a process which may take
many days, during which time he lives at the house of his client.
Early in the morning of the tenth day he performs a puja, and the
hawan (sacred fire) is worshiped. Then the family is purified in the
following manner. The Brahmin mixes five substances in a cup: Ganges
water, sugar, curds, cow's urine, and a copper coin. The Brahmin
chants mantras as the mother and then other household members
sip a little of this concoction from their hands and then place their
dampened palms on their hair. The remainder of the mixture is
sprinkled about the house. The child receives a name from the Brah-
min, who writes it in his horoscope.

The "taste of boiled rice," held one year after birth, is accompanied
by a puja, and hawan, the sacred fire, is worshiped. Relatives may be
invited.

Kindred are called as guests to witness a boy's first haircut. On
this day the Brahmin performs a puja before and after the haircutting

APPENDIX III 397

(cf. Atkinson, 1884b, p. 91). Coins have been tied in cloth and put in
the child's hair, and the barber (or Dom who performs barbering
services) keeps these as well as various delicacies that are placed in a
tray next to the child during the procedure. The hair clippings are
placed with cow dung, milk, and two coins in a cloth to be later
deposited in a holy river to bless the child and prevent its sacred first-
hair from being defiled. After the haircutting the child is then ritually
cleansed in the well-known purificatory ceremony (ban) involving a
ceremonial bath given by female relatives, just as the bride and groom
are given in preparation for the marriage ceremony. There is feasting
and giving of small gifts to the child after the ceremony.

When two women have babies within the same month, the mothers
are potentially dangerous to one another. If the shadow of one falls
on the other, the child of the latter will become sickly. Infant illness
is frequently diagnosed as stemming from this event. To prevent this
the "colliding" ceremony is performed. In this ceremony a dark place —
usually a room in a house — is chosen where no shadow is formed.
There the two mothers stand on either side of a string cot turned up
on its side, embrace one another over the cot, and exchange necklaces.
Thereafter they need not fear one another.

If a woman has not had children, her family may resort to the
special worship called rudri patt. This is done yearly during the sacred
month of Saon (July-August) until a child is born or the hopeful
couple gives up. A learned Brahmin performs this worship by coming
to the home of his client and there reading scriptures through the day
for four to eight consecutive days. He blesses the household members
each time, and is paid in return. If the worship produces results it is
continued yearly throughout the lifetime of the children so begotten.
At least three Sirkanda families have had success with rudri patt.

B. Marriage Ceremonies (see Atkinson, 1884b, pp. 94 ff.)

Engagement

The father or elder brother of a marriageable boy makes the
initial arrangements for marriage. This procedure is the reverse of the
usual one in the plains, where it is the responsibility of the family of
the bride to make the initial contacts (Lewis, 1958, p. 159; Planalp,
1956, p. 453). The difference may bear some relationship to the
financial obligations of the two parties. In the plains it is the bride's
family who must give a gift (dowry) to the family of the groom, but
here it is the reverse (bride price). In both cases it is the family making

398 APPENDIX III

the investment which shops around and makes the preliminary ar-
rangements. When a prospective bride is found, her father is con-
sulted and, if willing, gives the girl's horoscope to the prospective
groom's father, who takes it to his Brahmin.1 If the horoscopes of
boy and girl are compatible, the prospective groom's father returns to
the girl's parents and gives them pakoras (fried delicacies similar to
fritters) and halwa.2 The girl's family then gives the boy's father a
meal, a "relationship feast." This signifies that the "asking" has been
done and that the union is agreed upon. The groom's father informs
the family of the bride when he will return for sagai (the formal en-
gagement). On the appointed day the father comes in the company
of a few male relatives, bearing the hasli (heavy, solid silver necklace),
which is worn thereafter by the girl as the symbol of engagement. He
also brings a set of clothes for the girl. The Brahmin performs a puja
and puts the hasli and a vermilion mark on the girl. Then her parents
feed the guests and other villagers, and the engagement is formally
established. If it is to be a bride-price wedding, half or all of the
amount is paid at this time. The wedding ceremony follows at the
convenience of the bride's family, who act upon the advice of their
Brahmin in setting a date. Certain months are considered auspicious
for marriages — notably the period December through March, which
is also the agricultural slack season — and most marriages take place
during these months on dates found to be favorable by the Brahmin.
The wedding usually takes place from a month to a year after the
engagement. The engagement normally takes place between the ages
of ten and fifteen for girls, and fourteen and twenty-one for boys. Be-
tween the initial agreement and marriage the couple are not supposed
to talk with one another, but there is no objection to their seeing one
another if they happen to be in the same vicinity.

There is another type of engagement called a "womb engagement,"
in which two pregnant women of the appropriate caste and sib
grouping agree to marry their offspring if they are of opposite sexes.
The two potential mothers put marks on one another's foreheads and
become simdJil (in-laws). The bond is, of course, ignored if the children
are of the same sex or if one dies. If all goes well the marriage may be
celebrated at age eight or nine.

Preparation for the Wedding

Well in advance of the wedding the barber and his family prepare
leaf plates for use at the various feasts associated with it. The barber
is also responsible for inviting the wedding guests. He does this by
taking a lump of gur (brown sugar) to each household head to be

APPENDIX III 399

invited, and announcing the time and place. In the same manner the
Brahmin, too, often invites people whom he sees on his rounds. Shortly
before the ceremonies are to begin, various delicacies may be dis-
tributed to villagers by the families involved in the ceremony.

From this point on the ceremonies begin, and they differ according
to the type of marriage.

The Wedding

For the traditional Pahari ceremony a group of four or five men
from the groom's village and sib go to that of the bride on the day
of the ceremony or the day before, depending upon the distance to
be traveled. There they complete the payment of bride price and are
fed. The bride has been dressed in the one-piece garment worn by un-
married girls, and in a head scarf. She is carried by her own brother
or her mother's brother, on his back, in the company of the groom's
relatives, to the village and house of the groom, where the marriage
ceremony is performed.

The Sanskritized ceremonies are so similar, whether in bride-price
or dowry marriages, that they will here be described together and,
where points of difference occur, the differences will be mentioned.
The greatest difference is in quantity — dowry marriages are larger,
gayer, more elaborate (reflecting the greater wealth of the participating
families) than are bride-price weddings. Participants in a wedding in-
clude primarily the kindred of each of the principals. Members of the
household, lineage, clan, sib, and non-sib kindred participate in that
order. These plus others of the same or ritually similar castes are
guests. Various specialists (Brahmins, barber, Bajgis) perform special
functions as described below.

There is activity for some time in advance of the wedding, preparing
food and accommodations for the guests, preparing the clothing and
other accouterments for the ceremonies, cleaning, whitewashing, and
decorating houses, and so on. Before the groom's party leaves for the
bride's village, where the wedding is to take place, the guests are fed
twice: once on the evening before and once on the morning or noon
of the day of departure. The food is generally cooked by hired Brahmin
cooks of the ritually pure Sarola subcaste. It is the duty of the Bajgis
to play their drums during each meal served. The barber is present
to set out the leaf plates and perform any tasks that may be required
of him throughout the festivities and particularly at mealtime.

Ceremonies in the groom's village begin either the day before the
departure of the marriage party or, more commonly, on the day of
departure. The choice is made by his family in consultation with

400 APPENDIX III

their Brahmin and depends partly on how elaborate the ceremonies
are to be, and partly on the distance of the bride's village. Here I
shall assume that the ceremonies occur on the day of departure.

In the morning the groom has his hair cut and then undergoes the
ban ceremony, in which he is ritually purified. The area of the cere-
mony has first been plastered with cow dung by the barber. Then a
sacred diagram is drawn on the earth with flour by the Brahmin. Over
this diagram is placed a square stool to which are tied certain sacred
objects, and upon which the groom sits throughout the ban ceremony.
A canopy containing ritual items in a tray is held over the groom's
head. The Brahmin begins with a puja. Next, female relatives of the
groom ritually purify him and then ceremonially bathe him in a mix-
ture of oil, water, and turmeric while the Brahmin chants mantras.
This drawn-out ceremony is accompanied by considerable horseplay
and public interest, especially in its later stages. After the bath the
groom is carried by a male relative to a room in his house, where the
sacred thread ceremony is performed if the groom has not already
undergone the ceremony, as he probably has not unless he has been
previously married. Then the groom is dressed in the traditional white
clothes, green sash, yellow turban, and veil or garland which he will
wear throughout the ceremonies to follow. The barber does not assist
in the dressing, as he often does in the plains. Then the Brahmin
performs the Ganesh puja (elephant-headed god ceremony), which en-
sures success in any venture. Thereafter the groom is carried to a large
rug, where he sits with another man who acts as accountant as they
receive and record the traditional gifts of money which relatives and
other guests bring and which must be repaid at future marriage cere-
monies which the groom or his relatives attend as guests. All this is
standard practice in Hindu marriages. Then the party is fed and prep-
arations are made to leave for the bride's village in time to arrive
at about sundown. The entire party, called a bardt, collects at the
groom's house, where they line up, musicians first, then the palanquin
or horse which will carry the groom, followed in turn by the family and
guests. Only men take part in the barat. To the accompaniment of
horns, drums, and waving flags, the barat proceeds to the Raghunath
temple (if the groom is a Sirkanda villager), where the groom leaves
an offering and receives a blessing from the keeper of the temple. Then
the party is off to the bride's village, stopping periodically to rest and
in at least one place to give gratuities to the musicians. Their way
into the bride's village is lighted by bonfires built by the bride's family
if darkness falls before their arrival. As they enter the village the mem-
bers of the barat take pains to make a favorable impression. The

APPENDIX III 401

musicians recommence to play, the groom is in his palanquin (though
he may have walked most of the way), and the guests and relatives
put on their shoes, which were taken off to facilitate walking. They
are given a place to rest and service of tea. The groom undergoes a
short puja in which the Brahmins of both families participate, and all
the guests are blessed. In a dowry marriage the bride's father greets
the groom and ceremonially presents him with a coconut and one
rupee tied with a sacred string, to symbolize that the groom is to
take full responsibility for the bride and is considered, by his father-
in-law, to be a worthy recipient of this "daughter-charity."

Even when the bride lives in the same village as the groom there
is a barat and precisely the same procedure is followed, except that the
procession goes only to a large tree outside the village, circles around
it, and comes back to the bride's house.

Equivalent preparations have taken place in the bride's village. The
bride goes through the same ceremonies as the groom except that there
is no investiture with the sacred thread, and only women are present
during the ceremonial bath. Her dress is not ritually prescribed as is
the groom's, and generally consists of the fanciest clothes available,
now often including the sari (women's garment typical of the valley
and new to these hills). The bride's family receives grain rather than
money as the traditional contribution of relatives. They have had
their own guests to feed and preparations to make for the coming
wedding party.

When the entire wedding party is assembled at the bride's home
dinner is served, the groom eating with the barat and the bride with
her female relatives. Then the bride's Brahmin calls her to dress. He
gives her the jewelry brought by the groom's family and also the silver
paper mask (mor) which she will wear during the ceremony. The
jewelry given by the groom's family always includes the large gold nose
ring (natola) which she will wear throughout her married life in honor
of her husband and as the symbol of her marriage. Generally a nose
pendant (bulaki) of gold, also a symbol of marriage, is given, and
whatever else the groom's family can afford — much more in a dowry
marriage than a bride-price one. In jewelry gifts, as in all phases of
the wedding, prestige is an important factor — the more lavish and con-
spicuous the expenditure, the more prestigeful is the family who made
the expenditure. A bride's family is likely to complain loudly if they
consider the expenditures of the groom's family insufficient, and to be-
little them in any event.

Next the bride goes to the place of the ceremony, escorted by a sister
or other female relative and by her Brahmin. The groom is led by

402 APPENDIX III

his Brahmin. By now it is probably late at night — the exact time is
determined by the Brahmins. Meanwhile the guests have been engaged
in visiting, singing, dancing, and drinking, and they will continue in
this, drifting in and out of the ceremony itself as the spirit and spirits
move them.

SANSKRITIZED CEREMONY

In bride-price marriages the wedding takes place indoors so that
the audience is limited in numbers — usually no handicap since there
are relatively few guests at such marriages. In dowry marriages the
ceremony is held out-of-doors. In the initial stages of the ceremony
the bride and her Brahmin are separated from the groom and the
audience by a black curtain. Sometimes they are in a separate room.
A long puja is conducted (gotra puja, phratry ceremony) in which the
gotra (phratry) affiliations of the couple are recounted, though, as
was shown in chapter 5, these do not have the relevance in this area
that they have in the plains. The puja is followed by the well-known
pherd ceremony, which takes place in a different part of the room
or area. In this the bride and groom circumambulate seven times an
elaborate shrine (bedl) containing, among other things, a sacred fire
— an all-night process. Before doing so they undergo a puja and are
joined by knotting together the sash of the groom and the stole or sari
of the bride, symbolizing their union. The circumambulation proceeds
to the accompaniment of puja and recitation by the Brahmins, with
great reluctance displayed by the bride, who sometimes is offered
bribes to speed her up. Then the bride and groom feed one another
curds as the Brahmin sits between them and recites mantras. There-
after they each eat some of the food that the other has left — an act
symbolizing their intimate relationship, as only members of the same
nuclear family eat one another's food. Then there is a brief ceremony
in which seven lamps and seven coins are placed on a stone grinding
slab, with the stone pestle lying at one end. The groom sits and the
bride stands beside him. The groom extinguishes the lamps by taking
the bride's foot in his hands, placing it on the stone, and with it
pushing the pestle across the slab, knocking away the lamps and
coins.3 Then the couple is led by the Brahmins in making seven
matrimonial promises. Following this, gifts of clothing, and sometimes
a ring and money, are given by the bride's father to the groom. Also,
at this time, dowry is presented by the bride's father. There is always
some dowry presented regardless of whether or not it is a dowry mar-
riage. If it is a bride-price marriage the dowry will be only a brass tray
and pot and perhaps a cow. If it is a dowry marriage, utensils, bedding,

APPENDIX III 403

jewelry, and other things will be given, often in considerable quantity.

Thereafter the couple goes to the bride's house, where they eat
separately. At this time the bride's sister removes the groom's face
garland and gets Rs. 1.25 in return. The groom also exchanges two
rupees for the blessings of his parents-in-law ("in-law offerings"). Then
the groom rejoins the members of the barat and the bride stays in
her house. In a dowry marriage the bride and groom undergo one
more puja in the morning and must sit in the bedi or shrine before
leaving the village.

Before leaving her village the bride joins her parents and the
Brahmin in a puja to their household god or gods. The barat forms
outside the bride's house with the groom mounted on horseback, and
with the palanquin fitted with side curtains to receive the bride. The
bride is carried to it by her brother or her mother's brother, who will
accompany her to her husband's village. Everyone is blessed by the
bride's Brahmin, a gun is fired, and the barat proceeds as tearful fare-
wells are exchanged by the bride and her relatives.

When the barat reaches the groom's village it proceeds to the Rag-
hunath temple or other village temple, where the bride leaves an
offering and is blessed, and the groom leaves another token offering.
Then they proceed to the groom's house, where the couple sits on a
carpet in front of the house and a puja is held honoring the household
god or gods of the groom. The groom's mother greets the couple by
pouring water on all sides of them as a purificatory rite and then
sipping the water. Next the bride enters the house and sits while the
mother-in-law washes her feet and touches the soiled water with her
lips. The bride kneels at the feet of her mother-in-law and sisters-in-
law, placing one rupee at the feet of each. The mother-in-law blesses
the couple with vermilion marks. The women of the house embrace
the bride and she sits with them, the groom with the men. Thereafter
everyone is fed and then retires, the bride with her female in-laws and
the groom with his male relatives.

The following morning the bride accompanies village women to
the water source where she performs a brief ceremony and returns
with the others, carrying a pot of water to her new home. At noon
there is the final feeding of the guests, who then disperse to their homes,
including the bride's relative who accompanied her.

Two to four days later, as the Brahmin decides, the bride and groom
return to the bride's house, where they stay one to four days. This pe-
riod is called donoj. If the bride is considered too young to live with
her husband, he will return home and the donoj period may last a year
or longer. When the donoj is finished, whether it is a few days or a

404 APPENDIX III

year or more, the bride and groom return together to the groom's
village, generally accompanied by male relatives of each, and there
they remain. The bride usually returns home to visit once every one
to three years. The couple generally do not begin to cohabit for
several days, weeks, or months after the donoj, as the groom's female
relatives keep a close eye on the bride and to some extent determine
when she and her husband can be together. There are not, however,
separate women's quarters, as are found among many plains groups,
and it is possible for the couple to arrange early contacts if they so
desire, especially since both men and women work away from the
house part of each day. A story is told of one frightened and naive
bride in Sirkanda who avoided her young husband for over a year.
Finally a group of village wives waylaid her in the jungle, told her to
start sleeping with her husband, and unceremoniously deflowered her
artificially to overcome her reluctance.

TRADITIONAL CEREMONY

In the non-Sanskritized wedding, when the girl arrives on the back
of her brother or maternal uncle at the house of the groom, she and
the groom and his family enter immediately into the puja for the
gods of the household. Then the Ganesh puja is performed. Each guest
puts a rupee on a tray for the Brahmin, after which the Brahmin
blesses the couple and their relatives with marks on their foreheads.
In the evening a feast is held. The next morning a goat is sacrificed
to the household god and another feast is held, ending the festivities.
Some few educated families (none in Sirkanda) are said to have sub-
stituted a reading from scriptures for the puja to household gods, and
in such cases this was done the morning after the wedding and no goat
was sacrificed. The traditional wedding always included a bride price.
When dowry marriages occurred, the Sanskritized wedding ceremony
was followed.

C. Death Ceremonies (see Atkinson, 1885, pp. 1-16)

For descriptive purposes death ceremonies in Sirkanda can be
divided into (1) preparation for death and cremation, (2) cremation,
including trip to and from cremation grounds, (3) post-cremation cere-
monies.

Preparation for Death and Cremation

When a person is thought to be dying, the "cow charity" ceremony
is performed in which a cow is given to the purohit. If death comes

APPENDIX III 405

suddenly, the ceremony is conducted as soon as possible after death.
The cow facilitates the soul's journey across the river barrier to sudrg
(heaven) which takes place on the thirteenth day after death (or,
according to some reports, may not occur until a year after death if
the deceased has not led a good life). When death is imminent, the dy-
ing person is put on the ground by household members and is purified
by putting small bits of gold wrapped in cotton in the nostrils, ears,
and mouth. A calf and various other items are gathered to be given
to the purohit. When death occurs, villagers and nearby friends and
relatives are informed, usually by a low-caste boy or by any passing
person who has heard the news. Those involved most directly in the
mourning and ritual pollution which results from death of a man
are the male lineage members — those who trace descent to a common
ancestor. Those involved for a married woman are members of her
husband's lineage. The sib and clan are also important at this time.
They and non-sib kindred and fellow villagers of the same caste who
wish to, fast for a day as a symbol of mourning.

The body is bathed by family members, and new or at least clean
clothes are put on it. Then the body is put on a stretcher-like rec-
tangular bamboo and cloth frame. It is covered with a white sheet
and a red shawl and tied down. Then (the same day if there is still
daylight, the next if it is evening) it is carried to the cremation place,
which is located near running water. Only men attend the cremation
ceremony, as many as are invited and wish to attend. This group always
includes household and lineage fellows (husband's lineage fellows if
a wife has died), and, depending upon the importance of the individual,
it includes other members of the clan, sib, caste, ritually similar castes,
village, and area, in order of decreasing likelihood of attendance. In
Sirkanda the wealth and importance of the deceased determines where
he or she will be cremated. The locations, in order of increasing de-
sirability, are: one of two places by running water within three trail
miles of Sirkanda; a place eight trail miles away at the junction of
two rivers; and the Ganges at Hardwar some 60 miles away. The last
of these is expensive (requiring a hired motor vehicle) and only a
well-off, highly respected, elderly male would be taken there. En route
to the cremation ground five, or sometimes seven, balls of barley flour
dough are carried with the funeral party. One is deposited near the
sacred pepil tree in Sirkanda, one by each of three places where water
is crossed en route, and one at the cremation ground. This procedure
is called simply "ball placing." If a Brahmin is with the group he
deposits these and chants mantras; otherwise family members place
them. Sometimes no Brahmin is called until the thirteenth-day cere-

406 APPENDIX III

mony, and almost never is one called before the ceremony which
begins on the seventh day. None is needed during cremation, and only
in a lingering death is one called to preside at the time of death. In
Sirkanda the Brahmin called is the same one called for weddings or
other rituals, but the traditional charity associated with death cere-
monies all goes to the traditional Brahmin or purohit, Tula.

Cremation

At the cremation ground the body is placed naked on a layer of
wood and covered with oil and clarified butter and then covered with
more wood. Straw is stuffed in the open places of the pyre.

The eldest son or equivalent relative is the chief mourner, and is
called "the one who sits for the kind (postdeath) ceremony," or simply
"keria." He should be at least eight years old; otherwise a substitute
is generally used. If a child plays this role the ceremonies are abbre-
viated to adjust to the shorter attention span and lesser stamina of the
child. The chief mourner lights the pyre near the head of the body, and
others assist, lighting elsewhere. Then as the fire begins the chief
mourner has his head shaved and is for thirteen days thereafter ritually
unclean. After the fire has burned down and the head of the corpse is
exposed, the chief mourner takes a pole and strikes the skull seven
blows to break it, dipping the pole in the water after each blow. The
most sacred parts of the body, because life leaves them last, are the
heart and lungs, called collectively tis. As the fire burns down an effort
is made to keep the chest cavity from being entirely consumed, and
before it is, the heart and lungs or fragments thereof are removed with
a stick. They are tied in a cloth and buried in the river under a large
stone. The fire is then rekindled to consume the body. Thereafter
the remains are thrown in the river except for a few bones, which are
kept to be deposited in the sacred Ganges at a convenient date, to the
accompaniment of a puja performed by priests who reside there. If
cremation is held at the Ganges, all the remains are at once deposited
in the river. The chief mourner then bathes in the river and the party
returns to the village. Before leaving, the chief mourner picks up a
stone from the site of the cremation and carries it home. This is called a
lingra, and is the symbol of Shiva. It represents the fact that the
deceased is now one with Shiva, though the average villager thinks
of it primarily as representing the deceased.

As already pointed out (p. 112), it is often said that actually ghosts
rather than flames consume the body during cremation. No remains
of the body are left, especially of women, lest the ghosts utilize them
for nefarious purposes or their own sexual enjoyment. There are

APPENDIX III 407

magical means of making these ghosts visible, but it is a dangerous
undertaking. Ghosts are prevented from following the funeral party
home or disturbing the ghost of the deceased by being imprisoned
when a thorn twig is placed under a flat stone on the trail and each
member of the group steps on it as he passes. On the way home the
funeral party is fed by the family of the deceased, who are themselves
fasting.

Post-Cremation Ceremonies

Upon arrival of the funeral party back in the village, all the men
turn their caps inside out and have a lock of hair cut off as a sign of
mourning, which lasts for the next eleven days. The men are also
sprinkled with purifying cow's urine by the women of their families.
The male household and lineage-fellows of the deceased continue to
fast and rest until the next day. They are the primary mourners,
and in funerals of unimportant people they are often the only ones
who attend the cremation. In the case of a woman, the mourning group
is her husband's lineage. That night a little flour is spread near the
door of the house. Early the next morning it is inspected in order that
the fate of the soul may be detected. If there is a bird track in the
flour the soul has entered a bird and there is rejoicing; if an insect
track appears, people are saddened. During the next thirteen days
relatives and friends come to pay their respects to the family of the
deceased. For the thirteen days after cremation, the ghost of the de-
ceased wanders about and is sufficiently feared that villagers are espe-
cially reluctant to go outside at night during this period.

At this juncture the course of ceremonies can vary, depending upon
the wealth and importance of the deceased. For the funeral of a person
of no great import, such as a young adult and especially a woman, the
ceremonies will consist of pujas done in the home on the seventh day
followed by further ceremonies in the home on the thirteenth day,
and that is all. For a more important funeral the four-day ceremony
called das karam is announced and is carried out beginning on the
seventh day and ending on the tenth, and followed by further cere-
monies on the thirteenth day. For the most important funerals of all,
the das karam plus a ceremony on the first annual anniversary of the
death, called barsotl, is announced.

The most elaborate of these possibilities will be described here be-
cause, though it is infrequent, it is the ideal and the lesser ceremonies
are modifications and abbreviations of it.

Das karam consists of elaborate all-day ceremonies for four days
performed by a Brahmin and the chief mourner assisted by the barber.

408 APPENDIX III

The ceremonies are carried out for two days at the village water source
(that is, at running water), then one day at a pepil tree in the village,
and a final day at the water. The Brahmin performs pujas and directs
the chief mourner in his actions and recitations honoring the deceased
and securing his well-being. The barber provides leaf packets used in
the ceremony and prepares the pipe for the Brahmin's use as he works.
There is no need to describe these lengthy ceremonies beyond men-
tioning that they help the soul of the deceased in its progress to the
next stage of existence and include ritual recognition of the 184 stages
of life through which the soul must pass before achieving release. The
rites also involve the giving of additional charity to the Brahmin.
They are performed in accordance with prescriptions contained in the
Brahmin's sacred books as modified by economy, necessity, and local
tradition.4

On the eleventh day the men who attended the funeral are given
haircuts by the village barber, and they put their hats on right side out.
On the following day the chief mourner bathes, is shaved, and has his
hair cut; on this day also, daughters and sisters of the deceased come
bearing puris (fried bread). Food is prepared, and the barber brings
the leaf plates for serving the food.

On the thirteenth day the major postdeath ceremony, term! (thir-
teenth), is held. A puja is performed and the purohit is given more
charity — a cow, a bed, bedding, clothing, an umbrella, and a pipe.
A cow is given to a daughter or sister of the deceased, and scarves are
given to his other female relatives. A feast ("bitter food") is held for the
entire village and guests. With this, the mourning period is over.

Barsoti, the one-year anniversary of death, is an important occasion,
and many guests and many Brahmins are invited from a considerable
distance around the village. A puja is held, charity is given, and a great
feast is held.

When barsoti is to be held, the chief mourner may observe some aus-
terities for the entire year before its performance: perhaps not wear
shoes, not sit on a cot, not use an umbrella, not eat meat or fried food,
nor participate in festivals. The entire family may refrain from ob-
serving festivals, and the entire village may modify or delay festivals
during the year. Any festival which occurs later in the same month as
a death is usually omitted or postponed a day in honor of the deceased.
Diwali was celebrated only one day instead of two by many villagers
in 1957 because of the death of an important man since the last
Diwali. Likewise, Popria Sakrant was delayed a day because of the
death of an old man in the preceding month.

Widows are not restricted in activity or dress, as they often are in

APPENDIX III 409

the plains. An unremarried widow does not, however, take an active
part in birth or marriage ceremonies. Also, she does not wear her gold
nose ornaments (the symbols of marriage) for at least a year after her
husband's death or until she remarries. Some do not again wear these
ornaments at all, or not until their sons are adult. This is not a formally
enforced rule. It is done out of respect for the husband, and the length
of time during which they are not worn is a publicly recognized index
of such respect. Widow remarriage is universal here except when the
widow is elderly. A young, unremarried widow is virtually an unknown
phenomenon.

A ceremony which occurs on the thirteenth day after death is that
of the lingrd dhdrna puja (ceremony in which the lingra are placed).5
For thirteen days after cremation the lingra stone brought from the
cremation ground is kept outside the house in a brass tumbler. Each
morning milk is poured over it as a form of worship.6 On the thirteenth
morning a Brahmin, the chief mourner, and the unmarried girls of the
family go to a place just outside the village where the lingra stones
are traditionally placed. In Sirkanda each local caste, and among the
Rajputs each clan, has a boulder used as a shrine, at the base of which
the lingra stones are deposited. All the high-caste shrines occupy one
small area, and the other castes' shrines are scattered in other areas.
The Brahmin conducts a puja before the shrine, places the stone,
blesses the chief mourner and the young girls, and receives a small
amount of charity. Villagers assert that the lingra stones gradually
disappear from these smaller shrines, one or a few at a time, and fly
to the great temple and place of pilgrimage in the Himalayas,
Badrinath. As a result, a huge pile of these stones is at that temple
and, conversely, there are never very many at the shrine in the village.

APPENDIX IV:
SIRKANDA KIN TERMS

Kinship in Sirkanda has been discussed in some detail in chapter 5.
Terminology used to designate kin is presented here in the form of a
table. A few general statements of explanation regarding the table
are in order.

In the table, terms are grouped relative to ego, horizontally by
generation, and vertically by consanguinity (on the left) and affinity
(on the right). The relationship to ego of the person to whom a par-
ticular term is applied is indicated by initials as follows:

f-father h-husband

m-raother S-son

b-brother d-daughter

s-sister e-elder

w-wife y-younger

Thus, for example, febd means father's elder brother's daughter.
Where a kin term appears followed by an = sign, the term to the
right of the sign is the term used by ego for the spouse of any person
indicated to the left of the sign. Whether the spouse is husband or wife
is indicated by h or w, respectively. Thus, the = sign serves the func-
tion of a horizontal ditto, precluding the necessity of repeating all the
relatives to whose spouses a term refers.

Kin terms are numbered consecutively in parentheses, on the left,

APPENDIX IV 411

for purposes of identification and reference. Numbers in brackets, to
the right, indicate (at the first occurrence of a term) other places where
the same term occurs.

In some cases Hindi terms, or terms apparently derived from Hindi
usage in the nearby plains, are in frequent use, although Pahari terms
are also known and used. The Hindi or Hindi-derived term is in-
dicated in parentheses below the Pahari term. In some cases two Pahari
terms are known or used, evidently stemming from dialectal differences.
Then the alternative usage is indicated in brackets below the usage
deemed most "proper" by informants.

The terms "elder" and "younger" refer to age of the person to
whom the term applies relative to that of ego in the case of con-
sanguineal kin of ego's own generation, and to seniority in the case of
affinal kin (determined by age relative to that of the connecting kin)
and of consanguineal kin of generations other than ego's own (deter-
mined simply by generational seniority). Terms which apply to more
than one of the categories of relatives commonly distinguished in
Pahari kinship (see p. 191) are frequently replaced by descriptive
phrases for the sake of specificity in reference. Some terms are used
in reference only, teknoymy or other indirection serving in address.
These are asterisked in the table. In address, terms are frequently
abbreviated, and not infrequently less precise terminology is used than
in reference. This has been indicated where relevant.

I am indebted to Mr. A. C. Chandola, linguist from Pauri, Garhwal,
and a former graduate student at the University of California, Berkeley,
for checking these kin terms. Despite some dialectal differences be-
tween the terms he knows and those used in Sirkanda, he was able to
clarify a number of points. As spoken by my informants, final vowels
in these terms seemed usually to be long. However, I am informed that
actually they vary freely from long to short. Therefore, I have in-
dicated as long vowels only those which Mr. Chandola and I decided
were consistently long. Any errors in terminology or transcription are
my own.

412

APPENDIX IV

TABLE 8
Sirkanda Kin Terms
I. Ego's Generation

Consanguineal relatives

Affinal relatives

(1) e: b, fbS, fsS, dida = (5) w bau

msS, mbS

(2) y: b, fbS, fsS, bhula [26] =

msS, mbS

(3) e: s, fbd, fsd, dldi [14] =

msd, mbd

(4) y: s, fbd, fsd, bhuli [15] =

msd, mbd

(6) w bwari [9a, 70, 72, 78]

(7) h jija
(bahena) [10a]

(8) h jawai [18a, 71, 73, 79]
(said) [11a]

(9) w bairwdn a
(9a) [bwari]

(10) webjeThu = (14) w dldi
(10a) (bahena)

(11) wyb sydlu = (15) w bhuli
(11a) (so/a)

(12) wes jeR sdsu = (16) h saRu bhai a [17]

(jethdu) [21a] (mausa) [25a, 40a,
40c]

(13) wysyya/j = (17) h same as [16]

(18) h manseru a
(18a) [jawai]

(19) heb jeThd = (23) w jeThdni b

(20) hyb ctyz/r = (24) w dyurdni c

(21) hes pwp/ze sdsu = (25) h 6Aa£ a
(21a) (jethdu) = (25a) (maMja)

(22) hys naA^d = (26) h bhula d

(nawdo)

(27) Swm, dhm, Swfs samdhlN e
Swms, dhfs, dhms [5amrf/?aN']

(28) Swf, dhf, Swfb samdhi e
Swmb, dhfb, dhmb

a Term of reference only.

b Term of reference; term of address is dldi.

c Term of reference; term of address is bhuli.

d Term of address; term of reference is descriptive.

e Reciprocal term.

APPENDIX IV

413

II. First Ascending (Parental) Generation

Consanguineal relatives

Affinal relatives

(29) f baba

(30) feb baDa [40b]

(31) fyb chacha [40]

(32) fs(eandy) puphu

(36) m bdi

[bwe] (ma)

(37) mb (e and y) mama

(38) mys
(38a) mes

nanbai
[kansi ma]

nanbai (?)
[jeThl ma]

(33) w baDl

(34) w chachl

(35) h mama [37]

(puphd)

(39) w mat

(mami)

(40) h chacha
(40a) (inausd)

(40b) h baDa
(40c) (mama)

(41) wm, wms, wfs, hm, = (44) h sasura a
hms, hfs sasu a [46]

(42) wmb, hmb mdsdsura a = (45) w masdsu a

[mamya sasura a] — [mamya sasu a]

(43) wf, wfb, hf, hfb = (46) w sasu a
sasura a [44]

(47) stepmother, mausyaN f
father's co-wives

III. Second Ascending (Grandparental) Generation

Consantmineal relatives

Affinal relatives

(48) ff, ffb (e and y) ffsh, fmsh ddda

(49) fm, fms, ffbw

dddi

(54) wmf, hmf, hmfb, buDha sasura
wmfb, wff, hff, hffb,

wffb

(55) wfm, hfm, wmm, buDhi sasu
hmm

(50) fmb buDha

(51) ffs, fmbw buDhi

(52) mm, mms, mfbw, mmbw, ndni

mfs(?)

(53) mf, mfb, mmb, mmsh nana

1 Term of reference; term of address is the term for elder mother or younger
mother, that is, the same as terms for mother's sisters (38, 38a).

4*4 APPENDIX IV

IV. Further Ascending Generations
Consanguineal relatives Affinal relatives

(56) fff, fffb bhuDhadadas [bhuDhadd]*

(57) ffm, ffms, fffs, f ffbw bhuDhi dadi* [bhuDhadl]*

(58) mff, mmf bhuDha nana

(59) mfm, mmm bhuDhi ndni

(60) ffff, etc. jhaRddda

(61) fffm, etc. jhaRdddi

(62) mfff, mmff, mmmf jhaRndna

(63) mffm,mfmm,mmmm jhaRndni

(64) fffff, etc. paRdada, etc.

(65) ffffff, etc. naRdada, etc.

V. First Descending (Children's) Generation •

Cosanguineal relatives Affinal relatives

(66) S, bS nihdl = (70) w; Sws, dhs bwarl

(beta)

(67) d, bd nihdll = (71) h; Swb, dhb jawai

(bed)

(68) sS bhaNju = (72) w bwarl

[bhaNja]

(69) sd bhaNjl = (73) h jawai

(74) stepson or co-wife's son sautelu a

(75) stepdaughter or co-wife's daughter sauteli*
VI. Second Descending (Grandchildren's) Generation

Consanguineal relatives Affinal relatives

(76) SS, bSS, dS ndtl = (78) w bwarl

(77) Sd, bSd, dd ndtiN = (79) h jawai

VII. Other Descending Generations

Consanguineal relatives Affinal relatives

(80) SSS, etc. jhaRnatl = w bwarl

(81) SSd, etc. jhaRnatlN = h jawai

(82) SSSS, etc. paRndtl

(83) SSSd, etc. paRndtlN

e Term of reference; term of address is ddda.
h Term of reference; term of address is dddi.

1 These terms are reportedly the same for man-speaking and woman-speaking.
Karve's data (1953, pp. 93 f., 104) confirm this.

Notes

Introduction

1 As was explained in the "Note on Hindi Terms," "Sirkanda" is a pseu-
donym, as are the names used for other villages in the vicinity, and for all
individuals mentioned hereafter.

Chapter 1 The Setting

1 The western boundary of the Himalaya West is the Tons River (a tribu-
tary of the Jumna), and the eastern is the Kali or Sarda River the boundary
with Nepal. The Himalaya West comprises the present districts of Almora,
Garhwal, Tehri Garhwal, and the hill sections of Naini Tal and Dehra Dun
districts.

2 Following Lowie (1947, p. 111) and Murdock (1949, p. 47), the Pahari sib
can be defined as an exogamous, patrilineal, consanguineal kin group whose
members claim common descent but are often unable to trace the lines of
descent accurately. In much of the anthropological literature, including some
of my own writings, this group has been referred to as the clan (cf. chapter 5).

3 Even the most ethnocentric foreigners find Paharis to have admirable
traits. J. B. Fraser, referring to a Western Pahari group whose polyandry he
found shocking, had this to say:

It is remarkable, that a people so degraded in morals, and many of whose
customs are of so revolting a nature, should in other respects evince a much
higher advancement in civilisation, than we discover among other nations, whose
manners are more engaging and whose moral character ranks infinitely higher.
Their persons are better clad and more decent; their approach more polite and

41 6 NOTES

unembarrassed; and their address is better than that of most of the inhabitants
of the remote Highlands of Scotland; . . . and their houses, in point of con-
struction, comfort and internal cleanliness, are beyond comparison superior to
Scottish Highland dwellings. (Fraser, 1820, p. 209)

Chapter 2 The Economic Context

1 This figure is the one used in official accounting. It includes the 176 acres
around Sirkanda plus 29 acres in odrer areas included with Sirkanda for ac-
counting purposes, and excludes 93 acres which in the accounts are included
with other villages. These 205 acres represent less than one-fourth of all
lands within the official boundaries of Sirkanda. The remainder is unculti-
vated.

2 Rs. is the abbreviation for rupee, the monetary unit worth about 21^ in
United States currency in 1957-1958 (Rs. 4.75 = 11.00). In buying power in
India, a rupee is approximately equal to a dollar in America.

3 Scientific names of plants are obtained from Atkinson (1882), who pro-
vides complete listings by scientific and "vernacular" names of both domestic
and wild plants of the hills.

4 A maund is about 80 pounds. A maund of wheat was worth about Rs. 15
in this area at this time. Wheat is a standard medium of exchange. A common
way to report income is to estimate its value in wheat. If a man gets Rs. 100 in
cash it is often reported as "seven maunds of wheat."

5 Estimates of yearly cash expenditures per family averaged roughly as fol-
lows: gur (brown sugar) Rs. 100, salt Rs. 25, kerosene Rs. 20, cooking oil Rs.
150, molasses Rs. 8, tobacco and cigarettes Rs. 15, cloth and bedding Rs. 110,
utensils Rs. 15, ceremonies Rs. 80, taxes Rs. 12, meat and drink Rs. 70, sweets,
trinkets, and entertainment Rs. 50, proportional year's expenditures on house
construction Rs. 50 (figuring the life of a house at forty years at Rs. 2,000),
livestock Rs. 150. Medical expenses are included in the total for ceremonies,
as medical treatment almost always takes the form of a ceremony. Education
is not a significant extra expense, amounting to only a rupee or so for school
supplies to those families with school-going children.

Low-caste expenditures are significantly less than those reported above —
perhaps half as much. The low castes buy fewer and less expensive clothes
and utensils, eat less sugar, own fewer livestock. Their weddings are inexpen-
sive, their houses are cheap, and they depend for tobacco, sweets, entertain-
ment, meat, drink, and so on largely on handouts from the high castes.

6 In 1958 he withdrew from the village because his servant left and he
could find no suitable substitute. He was gone for several months, and the
chances of his returning were slim because he was ready to retire and his sons
did not want to work in this isolated spot, preferring instead to operate the
profitable family shop in Rajpur.

7 The occurrence in the village of a house belonging to a member of the
Palial sib of Rajputs in a line of houses owned by Jawari sib members is due
to such a case, which also involved a chan at the southwest end of the village

NOTES 417

(see map 3). Two other village houses whose occupant-owners are isolated
from their sib fellows can probably be accounted for in the same way.

Chapter 3 The Supernatural

1 There are Pahari Muslims who are, except for religion, indistinguishable
from other Paharis. One such family lives in Bhatbair, and a boy of that
family attends the Sirkanda school. Other Pahari Muslims live in Tehri
Garhwal. There have also been a few Pahari converts to Christianity among
the Doms of Tehri Garhwal. However, no missionary has ever come to Sir-
kanda or, so far as I could determine, to Bhatbair.

2 The term "pujari" also refers to individuals who are the keepers of par-
ticular temples and who supervise worship and receive offerings to the deities
of such temples. In this account they will be called "temple keepers" to avoid
confusion. Temple keepers are usually of high-caste, and, though they pre-
side over the pujas to a particular god, their functions are entirely distinct
from those of the specialists described in the text.

3 For a description of these gods and their worship see Appendix IA.

4 For a description of the characteristics and manner of worship of these
gods see Appendix IB.

Chapter 4 Calendrical and Life-Cycle Ceremonies

1 For further description of these ceremonies see Appendix IIIA.

2 For a description of each of these types of marriage see Appendix IIIB.

3 Karve notes, "All over India there is the custom of giving bride-price
among the poorer castes and of receiving dowry among the higher castes"
(Karve, 1953, p. 132).

4 M. E. Opler (1959a) has discussed in some detail the religious functions
of the family in Senapur. The situation he described is parallel to that in
Sirkanda.

5 Of course "local Hinduism" can be used to denote local practices regard-
less of their origins and affinities, but the implication of local origin or local
spread must be avoided.

Chapter 5 Kin Groups and Kinship

1 In Sirkanda the word "jat" or "jati" refers to either the endogamous
caste or to the exogamous sib. Since it literally means "birth" or "lineage,"
this is not surprising. Both caste and sib are determined by birth. Hocart
(1950, pp. 32 f.) has discussed the word "jati," pointing out that it "has much
wider and looser a meaning than we have put upon the Portuguese creation
'caste.' It does not refer to any particular kind of division or grouping, but
simply to hereditary status."

As was noted in note 2 of chap. 1, in much of the anthropological litera-

418 NOTES

ture the group here referred to as the sib is called the clan. I have decided to
follow Murdock's usage because his distinction between sib and clan is rel-
evant to an understanding of Pahari social organization, for reasons to be
explained below. Villagers of the plains often refer to the sib as gotrd.

2 The differences between these figures are attributable to records of mar-
riages of now-deceased men or men no longer identified with Sirkanda. The
471 figure includes marriages of Sirkanda women, most of whom have gone
elsewhere at marriage.

3 One of these marriages is the three-wife family, in which two wives occupy
one dwelling and the third another.

4 Perhaps in certain other Pahari areas there is greater laxity, especially
between Rajputs and Brahmins. Saksena states, for Jaunsar-Bawar, "As inter-
marriage between Brahmans and Rajputs is easily possible, but not outside,
they constitute a single endogamous group. . ." (Saksena, 1955, p. 28; cf.
Majumdar, 1944, pp. 115, 138).

5 The 53 Rajput marriages include several with people of these sibs who
come from houses in the immediate vicinity of Sirkanda but who are not
identified as members of the village. Since these cases added to the size of
the sample and brought in no extraneous factors, they have been included in
the discussion.

6 This means that 77 Sirkanda people found mates within the village. The
number represents approximately half as many marital unions, since it takes
two to make a marriage, although polygynous and sequential marriages bring
the total up somewhat. In addition, one barber woman lived as wife of a
weaver man who resided in Sirkanda temporarily.

7 Majumdar found, in an exogamous Pahari village in Jaunsar-Bawar, that
35 per cent of the Rajput marriages were contracted within three miles and
only 8 per cent were contracted over 12 miles away. He found similar pro-
portions among the low castes of the village (Majumdar, 1955, p. 172).

8 The barber caste, being nonindigenous to this area, must go outside to
seek mates or, as is often done, take them from other low castes.

Only two Sirkanda marriages were contracted more than 18 miles distant.
These were of the women sold to outsiders. One of the two women went to
Delhi, the other to the Punjab, each some 150 miles distant.

9 Jain (1948) reports that of 605 women interviewed in Jaunsar-Bawar,
nearly half had been divorced one or more times, and 145 of those undivorced
were still young enough that he felt they still might become divorced. Majum-
dar reports that in a village of Jaunsar-Bawar 62 per cent of the Rajput
women, and 55 per cent of the low-caste (Kolta) women, had been divorced
one or more times. In each case almost as many had been divorced two or
more times as had been divorced once (Majumdar, 1955, p. 172).

10 Some of this material, including fig. 4, has appeared in Berreman, 1962c

11 If a married daughter happens to be visiting her parents at the time of
a funeral or other ceremony in which her natal family participates, she joins
them, even though she would not participate if she were at that time living
in her husband's house. Her sib affiliation obligates her to participate when

NOTES 4 1 9

she is in the local sib group, even if only temporarily; that is, sib affiliation
in such a case overrides clan affiliation. Clan affiliation might be explained as
entirely a matter of place of residence — however temporary — among those
affiliated by consanguinity or affinity with the core sib of the clan. That is,
it may not be inaccurate to say that whether a woman is considered a mem-
ber of her parents' or her husband's clan depends upon whose house she is
staying in at the moment.

12 "Descent, in fine, does not necessarily involve any belief that certain
genealogical ties are closer than others, much less a recognition of kinship
with one parent to the exclusion of the other. ... It merely refers to a
cultural rule which affiliates an individual with a particular selected group of
kinsmen for certain social purposes such as mutual assistance or the regula-
tion of marriage" (Murdock, 1949, p. 16). In Sirkanda maternal ties are recog-
nized and prominent. This is made explicit in the kindred or sapinda rules
of marriage regulation, in kindred participation in marriage and other life-
cycle rites, and in kinship terminology.

Chapter 6 Caste

1 Blunt, 1931, p. 5; Bougie, 1908, pp. 1 ff.; Brown, 1953, p. 30; Davis, 1949,
pp. 378 f.; Duraont and Pocock, 1958, pp. 7 ff.; Ghurye, 1952, pp. 2 ff.;
Hocart, 1950, pp. 3 ff.; Hutton, 1946, pp. 42, 44; Kroeber, 1930, p. 254; Leach,
i960; Nadel, 1954; O'Malley, 1932, pp. 1 f.; Ryan, 1953, p. 19; and Srinivas,
1952, pp. 24 ff. Gilbert (1948) lists over 5,000 references on caste and gives 17
sample definitions as well as one of his own.

2 This definition incorporates the elements most frequently included in
definitions of caste in the literature. See, for example, Blunt, 1931, p. 5;
Brown, 1953, p. 30; Davis, 1949, pp. 378 f.; Ghurye, 1952, pp. 2-30; and
Kroeber, 1930, p. 254. In a survey of definitions of twelve authors, it was
found that in addition to the criteria listed in this minimal definition, only
"distinct occupational and / or ethnic tradition," and "restrictions on eating
(commensality)" were mentioned with comparable frequency.

3 In this area kachchd food is that cooked in water, pakkd food that cooked
in oil or parched. Kachcha food is easily polluted and can be eaten only if
cooked and handled by one's own caste or a higher caste. Pakka food is much
less vulnerable to pollution and can be eaten by a wider range of castes. In
Sirkanda foods cooked in oil are considered less immune to pollution than
parched foods.

4 If a Dom is greeting a non-landowning Brahmin, he may say "Samani
Pundit." The derivation of the word "samani" is not definitely known. One
writer thinks it derives from "Shivaman," made up of the words "Shiva"
(the god) and mandna (to honor). He interprets the greeting as an entreaty
by indigenous Doms for their Indo-Aryan conquerors to honor or worship the
Doms' deity, Shiva (Bahadur, 1916, pp. 73 f.). A. C. Chandola suggests more
plausibly, in a personal communication, that it derives from the word
samnydn, meaning to bow properly. As a greeting, "samani" is used among

4-20 NOTES

all castes and is always given first by the person of inferior status. If two
people are of the same caste, then it is given first by the person of junior
generation or relationship, and if these are equivalent, then by the younger
of the two. Only low castes addressing high castes use the honorific title in
conjunction with the greeting.

5 In such statements the terms "high caste" and "low caste" are literally
translatable as "big caste" and "little caste," that is, big and small in the
sense of important and unimportant, as a "big man" (district magistrate) and
a "small man" (peasant).

6 There is one family of Muslims in Bhatbair. They came from Tehri
Garhwal about three generations ago and bought land from a Brahmin.
They are Pahari agriculturists and are indistinguishable from other Paharis
except in their religion. Their status position is ambiguous. All castes claim
superiority to them and practice ritual avoidance of them. In social relations,
however, they rank about equivalent to weavers who are their neighbors.
They are not really incorporated into the local caste hierarchy, and they
marry exclusively with Muslims in Tehri Garhwal or the valley.

7 William Rowe (in a personal communication) has pointed out that a
similar mechanism for resolution of the low-caste dilemma of mobility aspira-
tions and dharma is to blame one's own unalterable fate for failure to attempt
to move up. There may simply be no use trying to improve one's condition if
it is determined by fate.

Chapter 7 Intercaste Relations

1 At the Bhatbair fair most of the women in attendance were of low castes
while many of the men were of high castes, and the sexual activities among
them were a prominent feature of the fair. There is also the well-known fact
that Pahari girls who become prostitutes are usually of low castes, as are all
of the traditionally available singing and dancing girls.

Chapter 8 The Village Community

1 In the context of informal village social interaction, clans and sibs are
almost synonymous, the sib forming the core and including wives of sib mem-
bers. The named groups are sibs, and the corresponding interacting groups
are thought of as sibs by villagers but are usually clans. This should be
borne in mind throughout the chapter.

Chapter 10 Conclusion

1 The history of Pahari contacts may also be a factor. Seclusion of women
in India is attributable in large part to Moghul rule. Apparently people of
the hills escaped most of this influence, and those plains people who came
to the hills in Moghul times were among the ones who resisted Moghul in-
fluence most strongly.

NOTES 42 1

APPENDIXES

Appendix I

1 The coconut is a sacred object in Hindu ritual and is known in Sirkanda
as Sirlpal (honored fruit). Its sacredness is attributed vaguely by villagers to
its supposed similarity of construction to that of the parts of the universe or
earth. It is used in the Sirkanda area upon four ceremonial occasions: (1) in
worship of Gauril as sacrifice to Gardevi; (2) in worship of the village goddess
Devi (Durga) to win her support in any situation in which it is needed, in-
cluding the weather-influencing rites; (3) in honor of Devi in the village-
protection rite (Mundkile); (4) as a symbolic gift by the bride's father to the
groom in dowry marriage. Only in the wedding ceremony is it not treated in
every way as an animal sacrifice in apparent substitution therefor, even to the
point of being divided among the worshipers to be consumed after the puja.
The coconut is the appropriate offering for Devi at the well-known Kuddu
Kal mandir.

2 Crooke relates such objects to the stones (such as the lingra discussed be-
low) which he says are believed by people in several parts of India to be the
abodes of the spirits of the dead. He describes worship of such stones and
then notes that "instead of stones the Bharvad shepherds use in the rite small
copper plates on which a human figure is engraved by the village gold-
smith . . ." (Crooke, 1926, p. 319).

Appendix II

1 It is reported by Pant (1935, p. 231) for the Kumaon districts as Phooldeyi
(called Goga in the Chakrata Himalayas). He says that it celebrates the
coming of the flowers and the advent of the Hindu New Year.

Appendix III

1 The social considerations taken into account in arranging a marriage —
caste, sib, and so on — are discussed in chapter 5. An excellent account of cus-
tomary law dealing with marriage can be found in Joshi (1929). His descrip-
tion corresponds in almost every detail with my data collected without benefit
of his book.

2 In a marriage witnessed by the author, the bride and groom had been
found to have incompatible horoscopes. Instead of calling off the union as
would normally be done, the Brahmin recommended that the marriage be
performed, with the groom's eldest brother playing the role of groom and
then exchanging places with the groom after the ceremonies. This was done
despite some objection from one or two villagers and considerable incredulity
among outsiders who came to know of it. The intended husband acted as
would any other brother of the groom during the ceremonies and entered the

422 NOTES

picture only after the entire wedding and post-wedding observances had been
concluded. No other such arrangement had occurred in the memory of vil-
lagers. The possibility of a marriage of this kind is noted by Joshi (1929,
p. 109). A similar arrangement but for different reasons is noted by Majumdar
among the Jaunsaris, where in the polyandrous family, he says, "If a brother
wants to marry any particular girl of his choice, the eldest brother goes
through the ceremony of marriage with the girl and he may assign the bride
to the particular brother" (Majumdar, 1944, p. 144).

3 Crooke notes:

It is believed that the stability inherent in a stone is communicated to
those who sit on it or touch it. In the Vedic age the bridegroom caused
the bride to mount a stone, formally grasped her hand, and led her round
the household fire. In modern usage a stone, supposed to represent the
goddess Parvati or Devi, is erected in the marriage shed, the top of it is
smeared with vermilion. . . . The bride approaches the stone and touches
it with her foot, or the bridegroom, stooping down, takes the bride's great
toe in his hand and with it touches the stone. ... In the northern plains
the stone is seldom used, but the bride knocks over a couple of piles of rice
laid on the ground (Crooke, 1926, p. 321).

4 A detailed account of these ceremonies as outlined in the religious books
used by Brahmins in Kumaon can be found in Atkinson, 1884a, pp. 917 ff.
Those in Sirkanda are very similar but were not recorded in detail.

5 This ceremony is not reported for villages in the plains, but Crooke
mentions similar use of stones for the dead in Bombay, the Deccan, Baroda,
and other areas (Crooke, 1926, pp. 234, 319)-

Chodhras in Baroda after the funeral send a relation of the deceased to
fetch a stone from a river-bed. This is kept in the house and the Bhagat
soothsayer is summoned. Under his supervision wine is poured on it, it is
taken in procession with music to the place appointed for such stones, it
is there half buried in the ground or left exposed, rice and wine are poured
on it, a chicken is sacrificed on it, and a lamp fed with ghi is lighted upon
it. (Crooke, 1926, p. 319)

6 This is reminiscent of the milk offerings poured over the lingam in Shiva
temples on the plains.

Index

Accountant, village, 316

Adoption, 163-164, 185-186

Affinal relatives, 167-169, 194. See also
Kindred

Age roles, 238, 253, 260-261, 267

Agriculture, 45-50

Alcohol. See Liquor

Amoral familism, 323-324. See also Senti-
ments

Ancestors: as spirits, 110-111, 386-387;
worship of, 111, 390-391

Animal husbandry, 50-52

Annual ceremonies. See Calendrical cere-
monies

Army: contacts with, 277, 300; villagers
in, 72, 207, 300-301, 310

Artisans. See Low castes

Arya Samaj, 224

Aspirations, 86. See also Sentiments

Atkinson, E. T., 6, 11, 12, 19, 20, 21, 94,
107, 108, 121, 124, 126, 130, 151, 188,
208, 215, 221, 280, 378, 388, 389, 390,
392-393. 395- 397. 404. 416 n. 3, 422
n. 4

Authority. See Caste, dominant; Family;
Government, agents of authority; In-
fluence; Village government

Bacon, E., 348

Baden-Powell, B. H., 26; 283

Bahadur, R. P. R., 7, 12, 21, 29, 30, 107,

181, 202, 346, 419 n. 4
Bailey, F. G., 198, 202, 255, 259
Bajgi, 15, 61; economic role of, 65-66;

rank of, 214, 216. See also Low castes
Baki, 89. See also Shamans
Banfield, E. C, 323, 324
Barber, 61; economic role of, 65; rank

of, 213, 216. See also Low castes
Basham, A. L., 199, 347
Basketmaker. See Bajgi
Beals, A., 315

Beda. See Singers and dancers
Beekeeping, 52-53
Beidelman, T. O., 57
Bendix, R., 223
Berreman, G. D., 3, 23, 31, 57, 78, 82,

89, 102, 115, 133, 143, 150, 159, 171,

186, 198, 199, 229, 243, 244, 257, 264,

337- 338- 343. 346, 35°. 35 >. 353. 355.

386, 418 n. 10
Bhagwan. See Fate
Bhaichara. See Land, tenure on
Bhatbair, 23, 24 (map), 25-27, 294-295,

299
Bhotiya, 7, 14, 18, 23, 299, 349
Bhut. See Ghosts

Birth ceremonies, 126-127, 395-397
Blacksmith, 61; economic role of, 63-65;

rank of, 213, 216. See also Low castes
Blunt, E. H. H., 419 nn. 1 and 2
Bose, N. K., 257
Bougie, C., 419 n. 1

Boundary rite. See Mundkile ceremony
Brahmin, 14, 19-20, 199, 246; priestly

role of, 61-62, 104-105, 114-115, 125,

133-136; rank of, 210-211. See also

Ceremonies

424

INDEX

Bride price, 128-129, 354-355, 417 chap.

4 n. 3
Briffault, R., 347
British Garhwal. See Garhwal
British rule, 12-14, 26-27, 311-312, 317-

318
Brown, W. N., 419 nn. 1 and 2

Calendrical ceremonies, 121-122, 123
(table), 124-126, 388-394

Carpenter-mason, 67. See also Low castes

Cash economy, 48-49, 50-51, 53-54, 55-
57, 61-73 passim, 304-311, 416 n. 5.
See also Wealth

Caste, 197-200; advantages of, to high
castes, 243-254; allied, 208; attitudes
toward, 220-226, 235-236; ceremonial
participation of, 125, 131-133; changes
in, 226-228, 254-256; among children,
165-166; in community development,
325-326; compared to Negro-white re-
lations in U.S., 243-254; councils, 203-
204; deities, 109-110; discrimination
by, 109-110, 209, 237-238, 325-326; di-
visive effect of, 257-258; dominant,
204-209, 326; endogamy, 86, 149, 154-
157; identification with, 339-343; in-
tegrative effect of, 262-263; interaction,
229-242; intermarriage, 154-157, 229;
loyalty, 204-205; maintenance of sys-
tem, 254-256; mobility, 141-142, 210-
211, 220-225, 256-258; ranking, 209-
217; sex relations and, 232-235, 247-
250, 330; socialization, 165-166; stereo-
types, 219-220; vested interests, 242-
254, 343, 344. See also Ceremonies;
Family; Gotra; Low castes; Sib; and
specific castes

Census of India, 22, 208

Central Pahari. See Garhwal; Kumaon

Himalaya; Pahari
Ceremonies: annual, 122-126, 388-394;
life-cycle, 126-131, 395-409; social com-
position of participating groups in,

Chamar. See Shoemakers

Chandola, A. C, 411, 419-420 n. 4

Chans, 34-37

Child rearing, 75-77, 164-166

Childhood ceremonies, 126-127, 395-397

Christians, 417 chap. 3 n. 1

Chula. See Family, joint

Cicisbeism. See Wife sharing

Clan, 131-132, 183-187, 415 n. 2, 418-

419 n. 11, 420 chap. 8 n. 1. See also

Sib

Class differences, 217-219
Climate, 10-11

Cliques, 182, 264, 265-280. See also Dis-
putes
Clothing and ornament, 30-31, 301, 306-

3°7- 310
Coconut in religion, 379, 381, 384, 421

Appendix I n. 1
Cohesion, community, 262-265, 280, 336-

338
Cohn, B. S., 81, 109, 122, 138, 140, 201,

212, 222, 249, 255, 256
Community. See Sirkanda village; Vil-
lage community
Community Development Program, 288-

291, 318-320; analysis of, 321-330. See

also Village level worker
Conflict. See Disputes
Congress Party, 237, 286, 303, 315-318,

325
Contagious magic, 118
Cooperative labor, 77, 324
Corbett, J., 11
Councils. See Panchayat
Courts, 254, 270-275, 277, 282-283, 292-

293. 3l6-3i7
Credit Cooperative Program, 288, 290-

291' 32°-33°
Crime, 73-75, 270-271, 272, 282-283, 322
Crooke, W., 106, 107, 108, 113, 383, 421

Appendix I n. 2, 422 n. 3, 422 n. 5
Crops, 45-50
Cultural drift, 350
Culture area. See Pahari
Curers, 69, 116-117, 369-387 passim

Dancing, 102, 125-126, 392, 402. See also

Kalratra
Dashera, 124, 391
Davis, K., 419 nn. 1 and 2
Death ceremonies, 130-131, 404-409
Dehra Dun: city, 9, 13 (map), 24 (map),

72, 295, 301-302, 304, 307; district, 12,

13 (map), 14, 24 (map); valley, 10
Deities. See Gods
Descent, 176, 193, 194; caste, 222
Devi, 101, 379-381
Devta. See Gods
Dharma, 86, 197, 223-224, 253-254. See

also Fate
Diet. See Food
Disease. See Illness
Disputes, 175, 230, 269-276; tables of,

271, 274
Diviners, 69, 1 15
Division of labor by sex, 75-77

INDEX

Divorce, 160-163, 4J8 n. 9. See also Mar-
riage, second

Diwali, 123, 125, 39!-393

Dollard, J., 235, 243, 250, 255

Dom, 14-15, 20-21, 211-217. See also
Low castes

Domestic animals. See Animal hus-
bandry

Dominant caste. See Caste, dominant

Dowry, 128, 402-403

Drummer. See Bajgi; Pujaris

Dube, S. C, 1, 81, 82, 187, 199, 259, 321,

325
Dumont, L., 326, 419 n. 1
Durga. See Devi
Dwivedi, J. N., 284

Eastern Pahari. See Pahari

Ecology, 348, 351-352

Economic Cooperative Supervisor (ECS),
320-321, 329-330. See also Credit Co-
operative Program

Economy, 38-79, 245-247; minor activ-
ities, 52-53

Education, 165, 290, 300, 302-303, 306,
330-335. See also Teacher

Elections, 285-286

Elopement, 130

Emigration, 35-36, 37, 71-73, 305

Employment outside. See Outside

Endogamy. See Caste

Engagement. See Marriage

Entertainment, 93-94, 125-126, 239, 262,
307-309, 327, 402

Europeans. See Foreigners

Evil eye, 119-120

Exogamy. See Sib; Village community

Exorcists, 90, 112, 115-116

Exploitation of low castes, 59-61, 242-254

Extended family. See Family, joint

Extensions of village, 294-295, 335-338,
340

Factions. See Cliques; Disputes

Fairs, 57, 125-126, 133, 389-390, 420
chap. 7 n. 1

Family: joint, 38, 144-148; authority in,
166-169; ceremonies by, 122, 125; dis-
putes in, 175; income, 53; intrafamily
relations, 166-176; residence in, 145
(figure); sex relations in, 171-174; sta-
bility and division of, 174-175; trends
in, 175-176; worship by, 86

Family: nuclear, 144-148, 174-175

Fate, 83

Felt needs, 327-328, 335

425

Fictive kinship, 195-196

Food, 49-50, 227

Foreigners, 298-300

Forests, nationalization of, 312-315

Fraser, J. B., 415-416 n. 3

Friendship, 266-269. See also Cliques

Funerary rites. See Death ceremonies

Gadgil, D. R., 257

Gaikh. See Jajmani system

Gandhi, M. K., 217, 330

Gatiesh, 400

Garhwal, 11-12, 13 (map), 14, 26, 28, 31,
37, 128, 298. See also Kumaon Him-
alaya

Garhwali, 28. See also Pahari

Ghosts, 111-113, 406-407

Ghurye, G. S., 202, 419 nn. 1 and 2

Gilbert, W. H., 197, 419 n. 1

Gods, 95-110, 369-386; in disputes, 89,
97, 270-274, 282; great traditional, 107-
109; household, 96-98, 168, 369-374;
village, 98-103, 374-386

Goffman, E., 229

Goldsmiths, 67. See also Low castes

Gorer, G., 173

Gotra, 187-191. See also Clan; Sib

Gough (Aberle), K., 199

Gould, H. A., 141, 224, 251, 264

Government: national and state, in Sir-
kanda, 27-28, 301, 311-335; agents of
authority of, 315-318; attitudes to-
ward, 44-45, 311-318; British, 311-312,
317-318; failure of programs of, 321-
330; successful program of, 330-335.
See also Community Development Pro-
gram; Congress Party; Credit Cooper-
ative Program; Education; Village gov-
ernment

Great tradition of Hinduism, 82; and
Pahari Hinduism, 104-109, 122, 137-
142

Grierson, G. A., 6, 13, 15, 16, 17, 19, 21,
22, 347

Gujar, 17

Half-brothers, 146, 176

Halwa, 50, 369-387 passim

Harijan, 208, 217, 225. See also Low
castes

Harper, E. B., 90

Health, 89. See also Curers; Illness;
Medicine

High castes. See Brahmin; Khasa; Raj-
put

Himachal. See Kumaon

426

INDEX

Himachal Times, 313

Himalaya, Himalaya West. See Ruraaon

Hindi terms: transcription of, xii, 411;
used in Sirkanda, 194-195

Hinduism: of Khasas, 19; of Paharis, 80-
82, 80-142 passim, 369-409 passim

History: of Pahari area, 11-21; of Sir-
kanda, 28-30

Hitchcock, J. T., 187

Hocart, A. M., 202, 417 chap. 5 n. 1,
419 n. 1

Holi, 124

Horoscope, 86, 396

Household. See Family, joint

Household gods. See Gods, household

Household protection rite, 98

Houses, 32-33

Hutton, J. H., 419 n. 1

Identification, levels of, in Sirkanda,
295. 296 (figure), 297-299, 341 (figure)

Identification groups. See Reference
groups

Iijima, S., 348, 349

Illegal economic activities, 73-74

Illegitimacy, 153

Illness, 116-117. $ee a^so Curers; Gods;
Health; Medicine

Imitative magic, 118

Immigration, 36

Incest, 86, 157-158

Incorporeal property, 78-79

Independence of India, 27, 301. See also
Government

Influence, 283-284, 325. See also Caste,
dominant

Inheritance, 42, 77-79, 186 (figure)

Integration, community. See Cohesion

Intercaste relations. See Caste, interac-
tion

Intermarriage. See Caste, intermarriage;
Sex relations, intercaste

Irrigation, 39, 47-48

Jain, S. C, 418 n. 9

Jajmani system, 57-61

Jat, jati, 149, 417 chap. 5 n. 1. See also

Caste; Sib
Jaunsar-Bawar, 7-8, 12, 13 (map), 14,

150, 169, 297-298
Joshi, L. D., 7, 30, 78, 129, 149, 158, 188,

421-422 n. 1
Judicial council, 292-293

Kalratra, 91-92, 369-387 passim, 392
Kanagat. See Ancestors; Worship

Karve, I., 158, 179, 188, 189, 193, 414,
417 chap. 4 n. 3

Kawakita, J., 348, 349

Kedarkhand. See Garhwal

Khajidan. See Lineage

Khasa, Khasiya, 14-20, 138-139. See also
Rajput; Brahmin

Kin terms, 194-196, 410-414; for non-
kin, 195-196

Kindred, 128, 158, 179-180

Kinship, 191-196, 296, 339-342

Kitt, A., 140

Krishna, 390

Kroeber, A. L., 55, 419 nn. 1 and 2

Kshatriya, 19, igg. See also Rajput

Kumaon districts, 11, 29

Kumaon hills, 10-11

Kumaon Himalaya, 9-14, 415 n. 1

Lambardar. See Tax collector

Land, 39-45, 245; tenure on, 26-27, 41-

43
Land reform, 312-315
Language. See Pahari
Law. See Crime; Disputes; Government;

Inheritance; Judicial council; Village

government; Panchayat
Leach, E. R., 197, 419 n. 1
Leadership, 281, 283-284, 324-325
Levirate, 79, 152-153
Lewis, O., 81, 110, 122, 127, 187, 247, 249,

258, 264, 265, 295, 325, 390, 397
Life-cycle ceremonies, 126-131, 395-409
Lineage, 176, 177 (figure), 178-179
Lingra, 185, 409, 422 n. 5
Lipset, S., 223
Liquor: consumption of, 94, 227, 239,

240, 242, 402; manufacture of, 64, 66,

73-74

Literature on Paharis, 6-8

Litigation. See Courts

Little tradition of Hinduism, 82

Livestock. See Animal husbandry

Lohar. See Blacksmith

Low castes, 14-15, 20-21, 63-68; accom-
modation of, to status, 236-237, 254-
256; compensatory gains of, 253; dis-
advantages of status of, 242-254; dis-
crimination against, 226, 236, 237-238;
place of, in economy, 41, 59-61, 63-71;
rank of, 211-217; rationale and resent-
ment of status among, 84-86, 220-226;
relations among, 232-233; relations of,
with high castes, 233-242, 255-256; so-
cial organization of, 181-182, 190-191

INDEX

427

Lower Himalayas. See Kumaon
Lowie, R. H., 149, 415 n. 2

Madan (maidari), 92, 102
Mahabharata, 8, 12, 15, 104, 381
Mahadev. See Mahasu; Shiva
Mahar, J. M., 70

Mahar, P. M., 81, 109, 136, 199, 210, 247
Mahasu, 102, 107-109, 385-386
Majumdar, D. N., 8, 18, 19, 20, 21, 80,

81, 143, 160, 169, 200, 201, 249, 346,

347, 348, 418 nn. 4, 7, 9, 422 n. 1
Mandelbaum, D. G., 210
Mandir. See Temple
Mantra, 89
Manu, 19, 138
Marijuana, 74

Markets. See Trade and Markets
Marriage: age at, 153; ceremonies, 127-

130, 397-404, 421-422 Appendix III n.

2; cross-caste, 154-157, 229; kin groups

in, 184-185, 188-190, 397-404; matri-

local, 78-79, 183-187; networks, 24

(map), 159-160, 350; regulations, 148-

150; second, 129-130, 160-163
Marriott, M., 81, 82, 103, 110, 122, 123,

124, 138, 199, 202, 390
Mass media, 303-304
Material culture, 32-33, 310-311, 345-

346. See also Clothing and ornament;

Houses
Matri, 98, 377-379
Matris. See Sprites
Mayer, A. C, 208
Mayne, H., 348
Medicine, 69, 89, 116-117. See also

Curers; Illness
Merchants. See Shops in Sirkanda; Trade

and markets; Vaishya
Merton, R. K., 140
Milk products, 50-51, 317
Missionaries, 417 chap. 3 n. 1
Mistri, 213, 216
Mobility. See Caste, mobility; Plainsward

mobility; Reference groups; Sanskriti-

zation
Moiety, 188, 189
Morgan, K. W., 81, 82, 124
Mortality, infant, 164
Mother goddess. See Devi; Matri
"Mountaineer" [Mr. Wilson of Musso-

orie], 6-7
Mourning. See Death ceremonies
Mukhia, 78

Mundkile ceremony, 102, 133, 383-385
Munshi, K. M., 8, 347

Murder, 282-283

Murdock, G. P., 148, 149, 176, 179, 183,

184, 188, 189, 191, 193, 415 n. 2, 417-

418 n. 1, 419 n. 12
Muslims, 209, 224, 301, 417 chap. 3 n. 1,

420 n. 6
Mussoorie, 9, 13, 24 (map), 31, 295, 301-

302

Nadel, S. F., 419 n. 1

Nai. See Barber

Naming, 126-127, 396

Nepal, 12, 18

Nevill, H. R., 6

Newcomb, T. M., 336

Newell, W. H., 8, 280, 315

Niehoff, A., 257, 305

Non-Pahari contacts, 299-304

Nyaya Panchayat. See Judicial council

Oakley, E. S., 6, 114, 220

Occupational variation, 70-71

Occupations, noncaste, 69

Offerings, religious, 93, 227, 369-387 pas-
sim

Olcott, M., 257

O'Malley, L. S. S., 199, 257, 419 n. 1

Omens, 120

Opler, M. E., 81, 89, 90, 133, 138, 242,
255. 259. 295. 325. 336- 417 chap. 4 n. 4

Outside: contacts with, 31-32, 299-304,
335-338; employment in, 71-73, 305-
307. See also Government; Urban con-
tacts

Outsiders, 295-304, 321-323

Pahari, 13 (map), 14; culture area, 136-
i37> !43- 294-295, 342, 344-351. 35i-
357; language, 14, 16-17, 21-23; peo-
ple, 14-21, 28; self-conceptions of, 85-
86, 251-252, 298-299, 356; sources on,
6-8. See also Dom; Khasa; Kumaon

Panchayat, 28, 280-283, 284-292, 326;
caste and, 203; in disputes, 154, 227,
270-275, 278; in divorce, 161. See also
Judicial council; Village government

Pandavas, 101-102, 104, 178, 381-385

Pant, S. D., 7, 11, 47, 349, 421 Appendix
II n. 1

Pargiter, F. E., 17

Passin, H., 198

Peasant society, 1, 55

Perelman, S. J., 309

Phratry. See Gotra

Physical appearance, of Doms, 21; of
Khasas, 18; of Sirkanda residents, 30

428

INDEX

Pilgrimage, 103

Plains, 10; emulation of, see Plainsward
mobility; Sanskritization

Plainsward mobility, 128-129, 134-137,
141, 227-228

Planalp, J. M., 81, 90, 109, 122, 389, 397

Planning Commission, Government of
India, 208, 292, 318

Pocock, D., 326, 419 n. 1

Police, 245, 315-317. See also Govern-
ment, agents of authority of

Political organization. See Caste, domi-
nant; Government; Influence; Village
government

Pollution, ritual, 198-199

Polyandry, 150, 354-355. 421-422 Ap-
pendix III n. 2

Polygyny, 148, 150-154, 171

Population: by age and sex, 35 (figure);
of Bhatbair, 23, 44; by caste, 208
(table); by family types, 147 (table);
Pahari-speaking, 22; of Sirkanda, 35,

Possession, supernatural: by ghosts, 87,
111-112; by gods, 87, 89, 92, 369-387
passi?n, esp. 374-375

Poverty, 34, 56

Pradhan. See Panchayat; Village govern-
ment

Prestige, 250-253. See also Influence;
Leadership

Priests. See Brahmin; Ceremonies; Puro-
hit

Property. See Incorporeal property; In-
heritance; Land

Prostitution, 68, 75, 214, 307-308. See
also Woman selling

Puja, 89, 369-387 passim, 395-409 passim.
See also Ceremonies

Pujaris, 69, 91, 115, 369-387 passim, 417
chap. 3 n. 2. See also Ceremonies;
Temple keepers

Puranas, 15

Purdah, 31, 167

Purohit, 62, 135, 231. See also Cere-
monies, life-cycle

Raghunath, 100, 107, 376-377

Rajput, 14, 19-20, 205-209, 210-211. See

also Caste, dominant
Ramayana, 107, 124, 391
Ram Lila, 124

Raturi, P. H. K., 7, 21, 30, 180, 202, 346
Redfield, R., 1, 55, 82, 304, 342, 343
Reference groups, 140, 296 (figure), 336-

338, 341 (figure); in Sirkanda, 295-299,

340-342. See also Caste, identification
with

Reincarnation, 84-86, 222-223, 408

Religion, 80-142, 369-409; Plains-Pahari
differences in, 136-137; rewards of, to
high castes, 253-254. See also Worship

Remarriage. See Marriage, second

Research methods and conditions, 3-6

Residence rules, 183-187; joint family
residence, 145 (figure); matrilocal, 163,
184-186; matrilocal and patrilocal,
186 (figure); patrilocal, 144-145, 194

Revenue. See Land, tenure on; Tax col-
lector; Taxation

Rope sliding, 68, 102, 133, 386. See also
Ceremonies

Rosser, C, 8, 351

Rowe, W. L., 159, 160, 218, 255, 420 n. 7

Ryan, B., 257, 419 n. 1

Sacred thread ceremony, 130, 400
Sacrifice. See Offerings
Sakrant, 123, 373, 388-389, 393
Saksena, R. N., 8, 18, 102, 108, 159, 160,

169, 201, 348, 385, 418 n. 4
Sankrityayana, R., 7
Sanskritization: in caste, 227-228, 256-

258; in class, 219; in marriage, 128-

129, 174; in religion, 82, 104-106, 135,

137-142
Sapinda, 158, 179-180. See also Kindred
Sarola Brahmin, 67, 183
Sati, 110-111, 170
Sayana, 26, 283-284
School. See Education; Teacher
Sentiments, 86, 140, 224, 353. See also

Amoral familism
Servants, 69-70

Service castes. See Dom; Low castes
Settlement patterns, 10, 24 (map), 25

(map), 35-36, 39, 44, 190; and cliques,

279

Sex relations: extramarital, 166, 172-173;
in the family, 84, 170-174, 354-355-
404; intercaste, 230, 232-235, 247-250,
261-262; Sanskritization of, 174

Sex roles, 75-77, 260-262

Shamans, 69, 87, 89-90, 115, 133-136,
369-387 passim

Sharma, K. N., 70

Shiva, 100, 102, 104, 107-108, 185, 385-

386- 393
Shoemakers, 68, 155, 214-216, 217-218,

233. See also Low castes
Shops in Sirkanda, 56. See also Trade

and markets

INDEX

429

Shrines, 91, 96-98, 369-374 passim

Shudra, 199, 201-202

Sib, 180-183, 186 (figure), 415 n. 2, 418-
419 n. ii, 420 chap. 8 n. 1; as exog-
amous, 153, 157-158, 280; names of,
30, 180-181. See also Clan

Singer, M., 133

Singers and dancers, 67-68, 214, 216. See
also Low castes

Singh, K. K., 247, 255, 256

Singh, R. D., 242, 255

Sirkanda village, 23, 24 (map), 25 (map),
26-37; caste composition of, 216; co-
hesion in, 336-338; as a community,
259-260, 294-295; external relations of,
335-338; identification with, 35-36;
location of, 9; loyalty to, 204-205;
orientations of residents of, 339-344;
population of, 35, 208; settlement area
of, 32-34, 44, 190. See also Village com-
munity; Village government

Sitala Devi, 101

Sivananda, S., 124

Siwalik mountains, 10

Smith, R. J., 336

Snake worship, 372-374

Socialization. See Child rearing

Songs, 39, 77, 157, 262

Spell casting, 117-118

Spring festival, 388-390

Sprites, 1 13-1 14

Srinivas, M. N., 80, 81, 106, 123, 140,
197- 198, 199' 205, 247, 256, 257, 258,
263, 419 n. 1

Srivastava, R. P., 349

Standard of living, 53-54, 64-65, 217-
2i9' 235, 241

Stevenson, H. N. C, 198, 199, 214, 249

Stowell, V. A., 7, 26, 42, 47

Suicide, 283. See also Sati

Sumptuary laws, 226

Supernatural. See Ancestors; Ceremo-
nies; Ghosts; Gods; Religion; Sha-
mans; Sprites; Worship

Supernatural possession. See Possession,
supernatural

Tailor. See Bajgi
Taulu. See Fairs

Tax collector, 26, 284. See also Govern-
ment, agents of authority of; Sayana
Taxation, 26-27, 41-42, 288-289, 315
Teacher, 332-335. See also Education
Tehri Garhwal. See Garhwal
Teknonymy, 194
Temple, 95, 98, 100, 374-387 passim

Temple keepers, 114-115, 417 chap. 3

n. 2. See also Ceremonies; Pujaris
Tibetan. See Bhotiya
Tibeto-Burmese, 14. See also Bhotiya
Tinker, H., 326
Tirsul, 91. See also Shrines
Tod, J., 30, 181, 187, 188
Topography, 9-10
Trade and markets, 55-57, 6g, 301-302,

309-3n
Traill, G. W., 6, 31, 114, 262, 298
Transportation, 9, 31-32, 51, 301-302
Travel, 9, 304-305
Turner, R., 336
Twice-born, 211. See also Brahmin;

Rajput; Sacred thread ceremony

Untouchable, 198-199, 211. See also

Dom; Low castes
Upreti, P. G. D., 231
Urban contacts, 9, 301-302, 304-311

Vaishya, 69, 199, 201-202

Values. See Aspirations; Sentiments

Varna, 199, 201. See also Brahmin;
Kshatriya; Shudra; Untouchable;
Vaishya

Village accountant, 316

Village community, 1, 259, 352; exogamy
in, 159-160, 180-181, 264-265, 280

Village gods. See Gods

Village government, 245, 284-292. See
also Panchayat

Village level worker (VLW), 288-291,
318-320, 327, 328-330. See also Com-
munity Development Program

Village protection rite. See Mundkile
ceremony

Vishnu, 100, 107-108, 370

Visiting patterns. See Friendship

Voluntary labor, 77, 289-291, 319

Vows, 90-91

Walton, H. G„ 6, 20, 26, 47, 200, 346
Wealth, 54-55, 217-219. See also Cash

economy
Weavers, 67, 213, 216. See also Low

castes
Weber, M., 246, 253
Wedding. See Marriage
Westernization, 140-141
Western Pahari. See Jaunsar-Bawar;

Pahari
Widows, 152-153, 162-163, 170, 408-409.

See also Death ceremonies
Wife sharing, 171-174, 354~355

43o

INDEX

Williams, G. R. C, 6, 23, 27

Wilson, Mr., of Mussoorie ["Mountain-
eer"], 6-7

Wiser, C. V., 59, 144, 219, 310, 315, 321,
322

Wiser, W. H., 57, 59, 63, 144, 219, 310,
315, 321, 322

Witchcraft, 118-119

Woman selling, 74-75, 156, 239. See also
Prostitution

Women: economic role of, 76; family
role of, 144-145, 166-174; social role
of, in general, 31, 167, 260-262, 354

Worship: daily, 121; pattern of, 88-95,
137; placative nature of, 94-95. See
also Ceremonies; Gods; Religion

ODLLFO'

Withdrawn from UF. Surveyed to Internet Archive

Hindus of the Himalayas, main
572.9542B533hC2

3 12b2 0331b 3^51

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Hindus of the Himalayas"

See other formats