Still waters run deep: Rural Bengali women and small-scale irrigation schemes

ill

Marion Kranenburg University

or

Amsterdam Cultural /\nH1ropology & Non-Western Sociology

MA

thesis, August i 995

Still Waters Run Deep

Rural Bengali women

and small-scale irrigation schemes

The title 'Still Waters Run Deep' refers to the way the installment of irrigation facilities gives rise to complex situations at field level and above, requiring the cooperation and coordination of various hetero-geneous groups of people, including villagers, government representa-tives, politicians, staff of a foreign consultant agency and (foreign) researchers. The metaphor of 'still waters' also applies to women's attitudes and behaviour regarding irrigation: there is more to it than . appears at first sight.

Marion Kranenburg University of Amsterdam Cultural Anthropology & Non-Western Sociology MA thesis

Examiners: Loes Schenk-Sandbergen, University of Amsterdam Barbara van Koppen, Agricultural University Wageningen Nico Kielstra, University of Amsterdam

August 1995 Amsterdam, The Netherlands

Contents

Preface

Acknowledgements

iii

Abbreviations

iv

Glossary

v

List of Annexes

vi

Map of Project Area

vii

l

Introduction

1

1. 1 Purpose and scope of thesis

1.2 Research perspective, organization and methods 5

Research perspective 5

Research organization and methods 8

1.3 Structure of thesis 12

2

Theoretical Framework

14

2. l Changing perspectives in Feminist Anthropology \4

The shift from women to gender 14

The rejection of structuralist dichotomies 16

The deconstruction of the household concept 18

2.2 Conceptualizing South Asian gender relations and ideologies 19

2.3 Gender in development planning 26

2.4 Irrigation planning: a short introduction 33

3

The Terai region of West Bengal

39

3.1 Introduction to the Terai region and the three selected villages 40

The river lift irrigation site 44

• The deep tubewell site 45

The hand tubewell site 47

3.2 Left Front administration:

Recent economic and political developments 48

Political decentralization 51

• Land expropriation and redistribution 54

Registration of sharecroppers 58

.

Irrigation Development 59

The relevance of Left Front policies to an irrigation project 62

3.3 Communities 63

• Rajbansis 64

Namasudras 66

Muslims 68

4

Project organization and

its institutional setting

₇₀

4. l NBTDP history and changes in objectives 71

4.2 The institutional setting of the NBTDP 74

Agricultural extension service

and other supportive policies 76

4.3 Organization and policies of NBTDP 77

• The three schemes: a short introduction 78

Selection of beneficiaries and sites

and the process of installation 78

Operation of RLI and DTW facilities 81

Beneficiary committee's of RLI and DTW facilities 83

Maintenance 83

4.4 The schemes of Jharmagurmari, Paschim Magurmari

and Purba Magurmari 84

The RLI of Jharmagurmari 84

The DTW of Paschim Magurmari 86

4.5 Present and future developments NBTDP: Privatization in the Left Front arena

4.6 Recapitulating the NBTDP approach

5

Gender and irrigation at village level

5.1 Gender relations in rural West Bengal

Gender in spatial organization and mobility Gender in marriage and kinship

Gender ideologies of physiques Gendered control over

and access to economical resources

Gendered control over and access to political resources Gendered control over and access to religious resources

5.2 Gender in agriculture

Responsibilities and interests in household, farm and off-farm production

5.3 Gender in irrigation affairs

Application for RLI and DTW schemes

Selection of HTW beneficiaries and introduction of HTW Construction

Utilization, operation and maintenance Management

5.4 Impact of irrigation on gender:

The variation in types of schemes discussed

Impact of irrigation on gendered division of labour Irrigation and men's control over women's labour Irrigation and incomes

Irrigation and dowries Irrigation and health

The variation in types of schemes discussed

91 94

97

98 98 100 104 108 112 113 113 114 118 118 119 121 121 128 128 128 131 131 132 133 134

6

How can women be supported

in irrigation development?

135

6. l Information 136

Training of rural women 136

Demonstration plots 137

Raising gender awareness among government staff 138

6.2 Organization and representation 139

Credit 139

Representation of women in farmer's organizations 140

7

Conclusion

141

Literature

145

Preface

During the fourth year of my course 'Cultural Anthropology and Non Western Sociology' I had the opportunity to get introduced to Euroconsult in order to study several small-scale irrigation schemes, installed under the North Bengal Terai Development Project (NBTDP) in West Bengal, India. The Terai project is a bilateral development cooperation project of the Dutch and Indian governments. On behalf of the Netherlands the consultant agency Euroconsult acts as a counterpart. I perceived the assignment as an excellent chance to learn about actual situations in develop-ment projects and the people involved in them, the more so as the usefulness and purposes of foreign aid are questioned by right and left wing commentators, in popular media as well as in academical studies. I spent six months, from January to June 1993, in the Terai plains of West Bengal, at the foot of the Himalayan moun -tains. The fieldwork, my first experience, and this thesis enable me to obtain my university degree, comparable to a MA degree.

Euroconsult requested me to perform research on the present functioning of a river lift irrigation (RLI) and a deep tubewell (DTW) scheme, and the socio-economic impact on the local popu-lation in general. I volunteered to study a cluster of hand tube-wells (HTW). As the latter are considered to be successful no students have been asked to perform additional research. Euroconsult was specifically interested in reasons for the apparent underutilization of the larger RLI and DTW 'small scale irrigation' facilities, even though the main factors causing this phenomenon had already been specified in former research. Within the general questions proposed by Euroconsult I integrated my personal interests that finally result in this thesis. Since I take an interest in the ways women are affected by small-scale irrigation schemes and the specific interests they have in irrigation development I considered it relevant to include the handpumps. Their character-istics are in many ways opposed to those of the RLI and DTW. Results of my research in three associated villages have been presented in a detailed report that was completed in 1994. The report offers a description of the overall functioning and utilization of the three different irrigation facilities and an indication of their

socio-economic impact. In the present thesis I shall focus on the role of women in irrigation and the impact of the project on their lives. Finally I shall explore how women can be supported in irrigation development.

As the final accomplishment of my study, this paper embodies more than a mere presentation of the results and interpretation of my fieldwork. I consider it to be an ideal platform to reflect on certain questions related to the meaning and practice of anthro-pology and justify my present attitude towards it. While I am finishing off this thesis more than two years have passed since my stay in India. Both the report and the time-gap have influenced the questions asked in this thesis and the structure of my argument. Physical distance from the project area has gradually and inevitably resulted in a greater sense of mental detachment, which is visible in my efforts to come to terms with the question of why things in the NBTDP are going on in the way they do. Nevertheless, my attachment to the people I studied is still very intense.

Acknowledgments

I would like to thank all the people, friends and family that have been crucial to my well-being and the success of my assignment before, during and after my fieldwork. Several people need to be mentioned in specific.

My research would have been useless without the cooperation of many villagers of Paschim Magurmari, Purba Magurmari and Jharmagurmari. I am really grateful for their time, friendliness and information. They form the essence of my stay in India, the report I have written and this thesis.

With great admiration I would like to thank my interpreter, Papiya Sarkar and her husband Debasis Sarkar. During the first months of Papiya's uneasy pregnancy she showed a lot of strength and character, persevering in completing the assignment. We managed to develop a fine style of cooperation and I am proud to call her my friend. Debasis functioned as a stand-in interpreter and I appreciate his help in realizing and organizing my plans.

As a guest of the West-Bengal government I was looked after in many respects. A comfortable apartment, functioned as a nice home and office during my stay. All government staff has been very friendly and helpful.

When I had just arrived, Professor N.C. Choudhury, Social Anthropologist, introduced me to several project sites, and supplied me with some valuable initial information. I thank him and his wife Mrs. Choudhury, an Anthropologist at the North Bengal University, for encouraging me and ensuring me that I was welcome to visit or consult them when necessary.

Loes Schenk-Sandbergen, my teacher at the University of Amsterdam and research-expert of the NBTDP, introduced me to Euroconsult and guided and stimulated me in doing research. I enjoyed and appreciate the contact we had when she visited the field and like to thank her for her support in general.

I would like to thank Euroconsult for the chance they provide students to learn from their projects and for the financial support they give them. I hope more

students will be able to gain knowledge and experience in this way.

Of great importance were my cook, Bijoya Roy, and her daughter and niece, Minu and Roma Roy. Despite our many differences we managed to establish a special bond.

The numerous meetings I have had with my (ex) student friends (Maria, Dineke, Saskia, Marijn, Anne-Klaas en Marijke) have been of great value and I hope we shall continue our discussions somehow in the future.

Last but not least, I want to thank my friend (in the most profound sense of that word) Richard Bos for his unconditional attention and support.

Abbreviations

ADO Agri-I Agri-M BC

BDO

BJP CA CPI(M) DOA

DTW

DGIS

FHH

HTW

KPS

uw

NBTDP NGO ORU Per REU RLI RW/S

SAE

SC ST

STW

WID

we

WUA

Agricultural Development Officer Department of Agri-Irrigation (DOA) Department of Agri-mechanical (DOA) Beneficiary Committee

Block Development Officer Bharatiya Janata Party, a Hindu nationalist party Command Area

Communist Party ofindia (Marxist) Department of Agriculture

Deep Tubewell

Directoraat-Generaal Internationale Samenwerking

Female Headed Household Hand Tubewell

Agricultural Extension Worker(s) Land and Water

North Bengal Terai Development Project Non-Governmental Organization

Operation Review Unit

Project Coordination Report, Euroconsult Report Evaluation Unit

River Lift Irrigation

Rural Water Supply and Sanitation Sub Assistant Engineer

Scheduled castes Scheduled tribes Shallow Tubewell Women In Development Water charges

Glossary

Adhiar Amon Anchul Aus Bargadar Bari Benami Bhadrnlok Bhut Bi di Bigha Boro Brahmin Chandal Gram Panchayat Holi Izzat Katcha channel Katta Kharif Ksatrya Lodja Mahr Mech Mouza Panchayat Samiti Para Potol Pre-Kharif Prodhan Pucca channel

Sharecropper (North Bengal) Kharif paddy

Area of one Grani (village) Panchayat Pre-Kharif paddy

Sharecropper

Homestead, usually consisting of a group of rooms ('gor'), with a yard in between Fictitious transfer of land

Bengali gentry, middle class, urban and educated

Ghost

Cheap cigarette 1/3 Acre

Rabipaddy(HYV)

Caste of priests and religious teachers, one of the four varnas

Untouchable Village council

Spring celebration, colour festival Honour

Earthen channel 1/20bigha Wet season

Caste of kings, warriors and aristocrats, one of the four varnas

Shame

A sum of money settled upon by the husband at time of Muslim marriage A tribal community

Police unit

Block level council Village quarter

Vegetable, like a mini-gourd Hot season

Chairman of Gram Panchayat Lined channel

Puja Purdah Ra bi Sabhaputi Spout Sudra Vaisya Varnas Zamindar Zilla Parishad

List of Annexes

Annex la

lb

II III N

v

VI VII VIII a VlIIb Religious service

Ideology and practice of veiling and seclusion of women

Dry season

Chairman of the Panchayat Samiti Water Outlet

Caste of peasants Caste of merchants

Colour, theoretical and course division of caste hierarchy in four castes: Brahmin, Ksatrya, Vaisya and Sudra

Superiour landlord/revenue collector District level council

Organigram NBTDP Levels of organization

Livespan and price ofHTW parts Access to RLI land per class and irrigated area per season

Distribution and utilization ofHTW's

Characteristics ofNBTDP irrigation facilities Water charges

Access to DTW land

and irrigated area per season Social structure of households in RLI and D'rw villages Social structure of households in HTW village

<

...

...

...) 111 \ . - . _ 1

\

'~ '-' "'{' . 1

I

~ Q

)

/

i

Lu

j

( J .r I 'l ~ i.. ----J1 , ... ,_'J'J ,,,,/ .,...., , I

'

...

,;,,

,_ 8 NATIONAL HIGHWAY RIVER o 5 10 15 20km -4 I

,,. ... ---,,l ;·"'·'\

I \ . . , ) <.... . ..._/,_ ...

____

,. ~ ... ~

_,,.

8

.,

\

\. H

u

r

. ..._

_____

. ../

'-.,

>'-··/(' ·,. / CHINA I . / .... i-·-. /.·-./ .J!Ep,._-L·v·-· BHUTAN ' ·-·-~--·"'Ass \ l"llO.liCTAl!liA~(·"T·- AM -' CALCUTT A. t'.°\ 4 N \

,,,-·--...

__

'-·

--

... _)

,...

...

c.. .... ,.

_'\

·-·-.

/ e AUPURDUAR

"'f

_I I

/

',

~ I I I I I I ' •

:

"{

.

I .

"·

---· t..

"

_.

_....,

"'

-~

_/

,.

·'""

_\...

.

_\ <...., \S- " ,

L --·

\

.

...

_,,.

L

...

_-·l

...-·,··

Vf

V>

·1

V) ,-• .-'\:'

v-~

\ '-'•

r\:,.

-~~

-·)

,'

('1<1 COOCHBEHAR

/if--'

1v ? ,./- \.

r·-.

'"n

,, "

TUFANGANJ _ ,1 • I A

•·

.

'

-._,

~-

., '\ . .) .

'-

~

.\ r' •'-.)

--

·-·

\

\ ,, J ( ) .A ·-:; "'{" G

Map of Project Area

(Source: Per. Jan. 1993)

4 \ _{\ .... .} ) t,,}

~

t"?

D · - , (

~,

.

l

"

.

~"'

-.r-1

~t

_-....

_s..r"

.

\.. _{_ j .} . I / E:"

s

S T I l L W A T E R S R U N D E E P

1

Introduction

l . l

Purpose and scope of thesis

During my fieldwork I studied the population of three villages in the J alpaiguri district of northern West Bengal. The majority of my informants are beneficiaries of a river lift irrigation (RLI) scheme, a deep tubewell (DTW) scheme or a hand

tubewell (HTW) facility. These small-scale irrigation schemes have been installed within the framework of a bilateral development cooperation project between the Dutch and West Bengal governments, the North Bengal Terai DevelopmentProject (NBTDP). l This project is implemented by the Department of Agriculture (DOA) of the government of West Bengal. The ultimate aim of the project is to improve the lives of poor farmers, by increasing their productivity and income, and to influence the income distribution by targeting the poorest sections of the rural economy, the so-called classes of marginal and small farmer's households, who possess less than two hectares of land. The broader theme of my endeavour is to understand how different technical interventions function at village level and gain insights on the impact of the Terai project on rural Bengali life. I shall specifically focus on rural Bengali women. This focus is rooted in research on gender relations and ideologies in the context of Bengali villages and their manifestation in local irrigation affairs and associated areas of the rural economy. This thesis would not be complete without making an effort to comprehend broader historical processes that reflect important political, economical and cultural developments in the Terai region of northern Bengal. Of crucial importance are the processes promoted by the Terai project and the state of West Bengal. The challenge is to link the broader contextual forces to micro-level, local experiences.

Why a focus on women? In literature on women and development it is shown how women are disadvantaged in comparison to men and how numerous rural development policies and practices have resulted in a further deterioration of women's position. 2 Research points out that rural Indian women, including rural

1 Phase I of the NBTDP functioned as a pilot project and ran during 1985-1986. Phase II ran from June 1988 to June 1993. Phase Ill is running from 1994-1999.

2 Since the effects of agricultural change are diverse and sometimes opposed there is a risk of falling in the trap of easy generalizations. Development not only had many adverse effects on poor rural women, but also on broader economically disadvantaged layers of society. However, though many poor men experienced backlashes of development, women tend to be even worse off, in many cases.

S T I L L W A T E R S R U N

11/;ttt

U1!-<l.,,

Bengali women, are discriminated again;t'it

t~~s 6f''~~~iic;{

over

resources (like land) and have less access to food (indicated by a relatively lower~

calorie-intake), medical care, education and income (Agarwal, 1989, 1994; Banerjee, 1985; Sen and Sengupta. 1985). Little value is attributed to women's labour. Women's wages, for labour that is similar to that of men, are in many cases lower, regardless of laws that prescribe equity. Statistics indicate that women hardly participate in the labour force. However, their unpaid labour contribution, in a direct or indirect way, to the household income and economy in general is not accounted for and is seriously underestimated and socially undervalued. The sex-ratio (=number of women per 1000 men) of most Indian states shows that the subordination of women should not be played down and indicates a deterioration of their position. Nation-wide sex-ratio's are decreasing since 1951, from 946 in 1951 to 917 in 1991. Out of fifteen states West Bengal ranks ninth; six states have worse sex-ratio's. Including West-Bengal all of these states are situated in North and East India. In 1991 the sex-ratio of West Bengal was 917, a slight

improvement in comparison to the ratio of 1981 which was 911.3 (REU 15, 1994b:

11)

The introduction of modern agricultural techniques, marking the development of intensive agriculture, and subsequent changes in the gendered structure of rural life, often had discriminating effects on rural women. Labour-saving technologies, for example those to increase the productivity ofpost-harvesttasks, displaced rural female wage-labour, as occurred in Bangladesh, to the benefit of male employment and incomes. Other agricultural innovations, such as those included in the Green Revolution approach4 have resulted in an increased demand for labour and cash. In the Indian context female labourers appear to have benefited relatively little from increasing employment opportunities. Unpaid female 'family' labour, however, has increased the workload of women from relatively poor farmer's households, though this labour is hardly visualized as work. The greater demand for cash to purchase the modern agricultural inputs that make irrigation effective have caused a rise in indebtedness. Research in African countries has indicated how the promotion of cash-crops, at the expense of subsistence crops, has benefited men relatively more

3 Rural sex-ratio's are better than urban ratio's, 947 and 819 respectively, according to the 1981 Census data of West Bengal (REU 12, 1994a: 14).

4 The concept of 'Green Revolution', from the 1960s and onwards generally enforced as the solution to the global food problem, refers to the HYV-irrigation package and mechanical devices, such as tractors. The HYV-irrigation package consists of bio-chemical inputs, such as 'high yielding variety (HYV)' seeds, fertilizer and pesticides, and a guaranteed supply of water through a variety of irrigation techniques. Especially larger credit-worthy farmers were targeted to pull the agrarian modernization. The approach has been widely criticized for its adverse impact on weaker sections of society and the environment.

S T I L L W A T E R S R U N D E E P

than women. Environmental degradation, that can partly be viewed as a result of adverse effects of development, has affected rural poor in general, but in meeting the daily household needs of water, fuel and fodder women seem to experience the ecological consequences most severely. (Agarwal, 1985; Whitehead, 1985; Shiva, 1988)

Women are more than proportionally represented among the poorest people of the world, and this seems increasingly to be the case. This process, tagged as the feminization of poverty, is aggravated because of eroding family structures that used to offer women some protection in their customarily dependent situations. In South Asia the feminization of poverty is probably less visible than in many African countries, since the 'guardianship' of South Asian males over their women is generally perceived as a matter of honour. However, cases of widowed, deserted or divorced South Asian women who have been left to themselves are more and more referred to (Chen, 1990: 217; Agarwal, 1994: 1). Poverty seems to be a major cause in eroding male support of women. However, while men's support diminishes their control over women is largely maintained. The ideological legitimation of women's subordination to men contributes to continued male control over women, even when no adult males are present in a household to execute this control.

In sum, the persistent discrimination against women, indicated by the nation-wide deteriorating sex-ratio's, and the adverse impact on women of numerous development efforts in the context of eroding supportive structures and persistent gender ideologies that have a suppressive effect on women legitimate, or rather compel research on women, poor women in particular. It is important to support poor women in their daily struggle against poverty, not only from the perspective of their victimization, but also from a more positive perspective: women form a vital link in the survival of poor rural households.

Irrigation development can be an instrument in improving the livelihoods of poor rural people, especially in the context of the Terai region where the population pressure on land is high and where previously only one main crop of paddy could be grown depending on the monsoon rains. Irrigation enables farmers to cultivate multiple crops per year and reduces the risks of crop failures when rains are late. However, as is emphasized by Krishna Bharadwaj, the productivity and

effectivene~s of irrigation do not only depend on the 'supply' side of irrigation: the

technological capabilities of the mode of irrigation, or the organization of

maintenance and delivery. The 'demand; side of irrigation is of critical importance for effective utilization; "In a peasantry that is sharply differentiated, the socio-economic hierarchy among the users produces discriminatory consequences in terms

S T I L L W A T E R S R U N D E E P

of the access to irrigation, its maintenance and utilization" (Bharadwaj, 1990: 4). Potential benefits as well as adverse effects of irrigation development are likely to be spread in an uneven way over a local community. Thus, investigation of varying degrees of irrigation use, its effectiveness and its side effects among different users is imperative.

Whereas class differences are widely acknowledged as being decisive in the distribution of irrigation benefits, gender differences are largely ignored, or concealed in so-called gender-neutral analyses. 5 Studies on gender and (small-scale) irrigation are scarce in general, but especially for the context of South-Asia. Available studies systematically reveal persistent gender biases in project policies and practices, who misunderstand or simply neglect the reality of women's lives.6 Ironically, this occurs despite targeting the poorest people, of whom women are a substantial part.

0stergaard remarks: "Even today the target groups for development projects are often identified as genderless categories, such as 'small farmers' or 'the rural poor'.

In the minds of planners these groups are men. In reality many of them are women." (1992: 2). By neglecting or misunderstanding women's position in a specific society a project may enforce existing discriminating gender biases or have an explicit adverse impact on women's lives, that may in turn affect that of men and children. Schrijvers describes how the introduction of a large-scale river basin scheme in the North Central part of Sri Lanka has contributed to the phenomenon of chronic undernutrition, because women have been cut off from their productive resources. This Mahaweli scheme has been introduced for irrigation and energy purposes and involved resettlement of people. Men who predominantly managed the marketable paddy crop and are generally perceived as the head of the household were favoured in the selection of new landowners. Women's role in the production of other less commercial grains and pulses and the importance of these crops to the local diet in various seasons was neglected. One of these crops, finger-millet, is resistant to fluctuations in rainfall, suitable to store for long periods, of great nutritious value, and therefore most useful when a famine is imminent. (Schrijvers, 1988: 29-51) Sex-stereotyped references to existing local cultural barriers to integrate women in a technical project may falsely legitimate the relative neglect of women's roles and interests and cover up a male-bias in the process of irrigation planning, that is generally managed by male engineers. For example, the idea that women are or should be confined to the domestic sphere may hinder irrigation

SFor example: Smout, 1990; Ahmed and Vuren, van, 1990.

6 On India, Madya Pradesh: Stanbury, 1984; on Sri Lanka: Schrijvers, 1988; on Kenya: Hulschebosch and Van Koppen, 1993; on Indonesia: Casey, 1991.

S T I l l W A T E R S R U N D E E P

planners in understanding women's role in the rural economy. On the other hand do gender ideologies in local communities and institutions form an important factor in responding to a t~chnical intervention and, thus, should be studied seriously. In South Asia specific agricultural activities are taboo for women in general or under certain circumstances, i.e. when they menstruate. This may form a real barrier in their participation in agriculture.

Against this background the following questions have been formulated:

What is the role of women in irrigated agriculture in three villages?

What is the impact of the small-scale irrigation schemes on women?

How can women be supported in irrigation development?

The answers to these questions will evolve from a gender analysis of various project levels, within a broader context of historical, political, economical and cultural developments in the Terai region of northern Bengal. After discussing my

perspective,. organization and methods of research I shall present an overview of the structure of this thesis.

l .2

Research perspective, organization and methods

... in all cases we do intellectual work within particular collegial communicative frameworks, frameworks that are not immune to current political shifts. We need to be aware of the ways

in which they tend to channel and shape our notions of what knowledge is and whom it should serve.

Micaela di Leonardo

Research perspective

My decision to perform my fieldwork within a development project is not only the result of a mere chance offered to me, but also a conscious choice. Since my first year of studying anthropology I am struggling with the purpose of anthropological

S T I L L W A T E R S R U N D E E P

sake of knowing interesting facts about different human beings, increasing an understanding of their lives. On the other hand, realizing this view is not without difficulties, I hold the ideal that anthropological knowledge should be of some use to people in general, but especially the people studied.

The desire to learn and know can be described as a rather selfish interest, which appears a little less selfish once it is shared with an elite group of people with comparable interests. However, this selfishness becomes poignant when it is

realized that the subject of anthropological learning concerns people. These people, studied by anthropologists, are not randomly chosen, but usually have a less advantaged position in this world than most anthropologists studying them. The interest in anthropological knowledge can also be perceived as a powerful interest and luxurious in character. Through history anthropologists had their specific reasons to study marginalized people, varying from recognizing and describing perceived near-to-extinct cultures to describing certain cultural patterns that were of help to secret services in defining their policies. 7 So-called 'advantaged' people have the money to travel and engage in social science.

In consequence of my ide_al concerning the usefulness of anthropological research I decided to concentrate on anthropology and developmental issues, realizing this would not necessarily form the solution to my conflict with

anthropology, nor necessarily be useful to the people studied. Action research has its specific problems and I do not consider it to be soul-saving in any respect. In the first place the question of cultural relativism looms large, since action research focuses on why and how certain developments should be affected. It can be debated what is most ethical. On the one hand, cultural divisions and power differences among people can be studied without questioning the implications of differences. On the other hand these differences may be investigated and questioned explicitly. Both ways of studying people have their own specific impact. I believe that ethics are not reserved to one way or the other. Ethics rather depend on being reflexive on a particular approach. 8

Being reflexive on one's approach includes an understanding of a researcher as a person and his/her interaction with informants, but an examination of the social location of the discipline as well. 9 These play a decisive role in the perception of

7 The most renowned examples concern social research in Latin America (Project Camelot) and Vietnam that has been helpful to the violent politics of the United States (Kloos, 1984). 8 See: Kloos, et al (1978) for a discussion of politics and ethics in relation to social research in the context of Third World countries.

9 Micaela di Leonardo recognizes the importance of examining a researcher's professional and larger intellectual location to contextualize academical work (1991: 31-32).

S T I L L W A T E R S R U N D E E P

cultural 'realities'. Sometimes anthropologists do not realize sufficiently that the discipline itself is shaped in a particular historical context, or that they themselves are part of the people who vary according to time and space, with their own histories and their own cultural biases. Talal Asad, who is renown for his critical review of the rise of anthropology, argues that: " ... anthropology does not merely apprehend the world in which it is located, but that the world also determines how

anthropology will apprehend it (1973: 12)". Further on Asad remarks that: " ... the basic reality which made pre-war social anthropology a feasible and effective enterprise was the power relationship between dominating (European) and dominated (non- European) cultures (1973:17)". I agree with Asad that the anthropologist's claim of politically neutrality supported the status quo, because they did not plead for reforms. It can be argued that despite the claim to be political neutral, this neutrality can not actually exist. Anthropological studies have an impact anyhow, either in supporting or criticizing a political status

Not only the social location of a discipline should be subject to ex:am1rnat11on,.rP<:'5Z·f~·,,"t-:f.

but that of the researcher as a person and his/her interaction with informants as well. Feminist anthropologists have significantly contributed towards the

importance ofreflexivity. l 0 They have revealed biases in mainstream anthropology that had been quite set, they have scrutinized their own biases, and they have contributed towards so-called 'confession literature' on the role of a researchers in fieldwork. A representative of the latter is Carol Warren, who deals with gender issues in performing social field research. Warren concludes her essays with the remark that: "It is not 'any researcher' who produces a particular ethnography, it is you" (1988: 65). I fully agree with her challenge of the myth that a researcher can perform research without being affected by gender, personality, or historical location.11

!'1-s a female fieldworker in India I have regularly experienced real or perceived hindrances that are related to my gender and western background. The perceived hindrances arose from my expectations and desires of how to behave and be accepted in a society where gender-boundaries are more compelling than in my own society. Some perceived hindrances turned out to be less valid for me as a European woman, e.g. riding a bicycle, extensive discussions with men, or smoking in specific contexts. In general I adopted a middle course between adjustment to the local

10 Though they have made valuable contributions to this way of thinking the 'reflexive

movement' is not so much a prerogative of feminist anthropology, but rather a manifestation of the spirit of postmodern time. The emphasis on the importance of gender and its intersection with other principles that shape the general social order may be attributed mainly to feminist anthropology.

S T I L L W A T E R S R U N D E E P

gender norms and adherence to my own values. I observed the local dress-code, but did not dress too classy, as some people expected me to do. I did not conceal my unmarried status, as is sometimes recommended to female fieldworkers of marriageable age. In most cases people wondered why my 'guardians' had not accompanied me, or automatically assumed that the Dutch male student who performed research in the same area was my husband. Generally people respected my different attitudes and ideas as being a logical outcome of my different origins. The contrast in values resulted in numerous discussions, and hilarious incidents, but rarely caused intense disagreements and offense at both sides. In one case an elderly man, a government official, assumed the role of a male guardian - "you are like my daughter" - to prevent me from going home by myself by bicycle at eight o'clock in the evening. I had regularly covered the small distance without any problems or comments from other men. Here, the collision in values was most palpable. As an adult woman I see myself as the only person that can be held responsible for my own actions and I could not accept his fatherly, but in my experience predominantly commanding, claims. Adjustment to his idea of righteousness would undermine my personal integrity. Several additional consequences of my gender shall be discussed in relation to the organization and execution of my fieldwork.

Research organization and methods

My main technique of research has been the execution of in-depth interviews with members of beneficiary and non-beneficiary households, who had access to command area (CA) land, or were able to work on command area land as (landless) labourers. During the different phases of my fieldwork I used a variety of semi-standardized questionnaires. During the initial weeks in the field I gathered general information on the characteristics of households and their agricultural and irrigation practices. At the RLI and DTW sites the initial interviews were usually given by male farmers, though their information has been checked at a later stage with other household members. At the HTW site women were easier to approach since many men were involved in off-farm jobs. I continued to interview both men and women in order to gain insights on their overall division of labour and barriers to out-door female participation, their respective roles in relation to the schemes, their

particular perspectives on the effects of the schemes, their priorities in agricultural and household matters in general and their decision making on these matters. In studying people's activities observations proved a valuable contribution to oral information. Men and women sometimes denied women's participation in fieldwork

S T I L L W A T E R S R U N D E E P

or described it as 'watching the crops'. Whenever I saw these women working in the field, I questioned them on the particular activity right on the spot. Gradually the character of the interviews developed from general to specific. In my diary I call it a switch from a sociological to an anthropological style of research.

In making up the questionnaires I have been inspired by texts and worksheets on gender and irrigation aimed at project planning and evaluation, which tend to emphasize inter- and intra-household differentiation in people's activities, control over and access to resources, benefits and decision making (SAW A, 1992; Feldstein and Poats, 1989; Stanbury, 1984). These frameworks have been used as a tool, their worksheets functioning as checklists. Though these texts off er useful insights, they do not form complete theories, as will be discussed in the theoretical part. They are rather descriptive and not appropriate to explain or understand the broader context of gender in irrigation planning.

The selection of informants has been determined by the location of the small-scale irrigation schemes. The informants of the DTW and RLI schemes live and/or have land within the command area or live very close to it. The HTW facilities have been distributed in a cluster. It was not possible to interview all (potential)

beneficiaries of the three schemes. I intervi.ewed about 37 households per village. About one-third of these households have been interviewed more than once, between two and six times. To acquire more knowledge on the project I also interviewed government staff, mainly those who were interacting with the beneficiaries directly, like the operating staff and the agricultural extension workers (KPS).

The fieldwork was performed during one period of six months (January-June 1993), about two years after the installation of the facilities. An examination of the impact of these facilities appeared somewhat problematic. To study cause-effect processes research at different points in time is required, a so-called 'before-and-after' approach. I solved this problem by asking the villagers to provide their

account of the history of the irrigation schemes and the effects on their lives. Studies of the Research and Evaluation Unit (REU) of the project have been conducted over a longer time-span and provide interesting additional information on similar schemes at other locations. The REU has carried out intensive research since 1989.

Between 1989 and 1993 an in-depth study (273 households) was performed on seven locations with new irrigation facilities. The study focused on the degree target group of marginal and small farmers were reached and the impact of the schemes on cultivation and farmer's incomes. The impact of irrigation on women of beneficiary households was assessed. It has been studied if women have been more

S T I L L W A T E R S R U N D E E P

on landless people has been examined. Attempts have been made to identify the conditions of success and failure of the different schemes. Efforts to engage female research staff have not been successful, but local women and girls were involved to interview female respondents according a special questionnaire for women. Besides the in-depth study a sample survey (1524 households) has been performed in order to back-up the in-depth study. The survey covered the complete project area during three agricultural seasons of 1992-1993. In most households the head of the household acted as the main informant. Other important sources information I consulted are Euroconsult's project coordination reports (Pcr's) and reports written by Dutch students of tropical agriculture and anthropology. An evaluation of the impact of existing irrigation schemes in the Terai region prior to the NBTDP and an evaluation of Dutch-Indian bilateral development cooperation by an independent Operation Review Unit (ORU) provided further information on the project

(Choudhury et al, 1986; ORU, 1994).

During my fieldwork I had to be assisted by a translator. A short language course only provided me with some basic knowledge of Bengali. We communicated in English, a secondary language for both of us. I have selected a female interpreter to decrease the likely hesitation of female informants in answering questions to a strange male person, and to enable myself to move around freely within female domains, both in words and physically. The choice of a female translator proved to be a valuable strategy. Male farmers were more interested to be interviewed or overhear an interview when I worked with a male translator, who worked with me when his wife was unable to come. Female informants felt often uncomfortable and were insecure to express their ideas in presence of their husband or any other adult male member of the household, though some women could not be bothered at all. With my female translator I was able to sit in the kitchen while women were engaged in preparing food. In this ambiance women :were usually very relaxed and frank, while we did not disturb them in their daily routine. Besides, noisy men could be avoided in this way. We also sat down in the field when women were engaged in fieldwork. The cooperation with my interpreter has been fruitful, the more so because she embodied more than just a translator. Her talent to make people feel comfortable was conducive to the atmosphere during interviews. Her married status and pregnancy possibly facilitated discussions on intimate matters, such as family planning, birth practices and menstruation taboos, which enriched our contact with the informants.

S T I L L W A T E R S R U N D E E P

The selection of the particular villages and schemes is mainly based on pragmatic criteria. When the fieldwork was in its design-phase it crossed my mind to select one village where no irrigation facilities had been installed close to a village with a project facility. I realized a straight comparison would not be possible, as these villages would probably differ in many respects, for example in social structure, or physical position in relation to rivers or markets, or soil condition. Ifl had chosen this research set-up the risk that the village without irrigation facilities would have modeled for the situation in the past would be there, whereas this would not be realistic as many changes might have influenced the present-day situation. I decided to study various schemes and localities in their own right. Nevertheless, with proper reserves I consider comparisons at a more abstract level legitimate in terms of how do people utilize and benefit from particular schemes.

The sites of the river lift irrigation (RLI) and the deep tube well (DTW) schemes have been selected in consultation with the project-coordinator of Euroconsult, using the criteria that they should have been functioning for at least one year and that no other in-depth research had been done on the selected schemes. A third and rather practical criterion was the distance from the apartment I lived in to the irrigation sites. Regarding the selection of the hand tubewell (HTW) cluster the following criteria have been observed. Firstly, the cluster should have been functioning for several years. Secondly, the cluster of HTW's should have been allotted to women. However, the selected cluster of 25 HTW's, all owned by females, had not been installed in 1990/1991, as stated officially, but only in November 1992. They had only been in operation since 4-7 months at the time of my fieldwork. Accidentally it turned out that several HTW's, funded under the NBTDP and installed in 1988, were pres_ent in the same area and intermingled with the new ones. I decided to continue my study at this site, hoping to collect interesting data on both old and new HTW's, which had been allotted to both men and women.

The distance between my apartment and the irrigation sites proved to be a major obstacle in performing anthropological research, which is ideally performed by a researcher who lives in a village for some time. I was hosted in an apartment on a state farm, where new approaches to agriculture, horticulture, forestry, and

•

livestock breading are practiced and developed. The advantages of living on the state farm were relative privacy, electricity to run my computer (most of the time) and running and drinkable water. The disadvantages of living at a distance from my research villages are not only connected to waste of time and energy spent on commuting, at least two hours per day when no problems occurred. My main

S T I L L W A T E R S R U N D E E P

concern, especially in the beginning, was my limitation in experiencing the daily cycle of rural life. Besides my female interpreter my female cook proved to be an essential link in filling this gap. Through my cook and her family I got more and more

acquainted with the villagers who lived on the boundary of the state farm.

Gradually I abolished my plans to move closer to my research location, which had

'

probably required a change of interpreter, something I did not fancy. My experiences in this fourth village, that did not have any irrigation schemes, have been important · in understanding the pleasures and worries ofpoor rural people. I have been

participating in their Holi celebrations, a puja for a calf, different phases of marriage negotiation and celebration, group singing, visits to family and fairs, besides of daily chats a:r;i.d the numerous meals we shared. Though the experiences in the fourth village complement my understanding of rural life no comprehensive analysis has been made of this village. However, some of my experiences have been integrated in this thesis

1.3

Structure of thesis

In chapter 2 my theoretical perspective and framework shall be presented. The first part deals with several feminist anthropologist discussions that form the tools to confront purdah, which is generally seen as the key concept in understanding South Asian gender relations, and gender in irrigation planning.

In chapter 3 an introduction to the project area and the three selected villages is provided. Next, the historical context is presented of several political, economical and socio-cultural developments that have changed the project area and its

population through time.

In chapter 4 the history, structure and policies of the NBTDP are explained, including the changes that have taken place since its start, its (revised)

assumptions and its connections to local institutions. The factors in the

iJrganization and institutional setting of the NBTDP that (may) have influenced women's specific position in the project shall be emphasized. The chapter is concluded with a description of the actual state of affairs I encountered in the three villages and a discussion of the present shift in project strategy towards

S T I L L W A T E R S R U N D E E P

In chapter 5 gender relations and ideologies in the rural context of West Bengal shall be analysed, based on the theoretical discussion on purdah, developed in chapter 2. Gender roles in agriculture and irrigation affairs of three different schemes and the impact of irrigation on gender shall be examined.

In chapter 6 I shall elaborate on the question how women can be supported in irrigated development based on their interests.

S T I L L W A T E R S R U N D E E P

2

Theoretical framework

2.1

Changing perspectives in Feminist Anthropology

Several discussions that took place within the context of feminist anthropology are important to demarcate my conceptualization of gender relations in the context of rural South Asia in general and irrigation planning in specific:

the change from a 'women' to a 'gender' as an analytical category the rejection of structuralist categories, in particular the private/public dichotomy

the deconstruction of the household concept

After. considering these discussions their usefulness to my view on South Asian gender relations and gender in irrigation planning shall be explained.

The shift from women to gender

The conceptualization of women by anthropologists has changed through time. Women have not been ignored completely in anthropology, as happened in many other disciplines, because they figured in descriptions of kinship and family affairs. Many ethnographies contain a treasure of information on gender. Since the 1970s feminist anthropologists have criticized the way women have been represented, usually in stereotypical roles. The male bias within anthropology has been disputed extensively and the question of women's subordination has been approached from different angles. Some authors, like Rosaldo and Lamphere (1974) emphasized cultural meanings assigned to men and women, reasoning from a structuralist tradition. Others, like Leacock (1978), used a historic-materialistic perspective and focused on political and economical processes. To correct the male-bias women researchers studied women. Despite the fruitful ethnographic information gathered, it was realized that an anthropology of women by women was not desired. To study women's lives - more or less - separate from men's lives was not considered to be realistic and increased the risk of marginalization. Gradually, the sex-gender concept was developed, separating biological sex from the socio-cultural construction of gender and a shift was made to the study of gender relations, with the concept of power as focal point. Studies of the cultural construction of gender and its

S T I L L W A T E R S R U N D E E P

approaches anymore. In studying gender relations the interdependent character of women's and men's position in society is often neglected. Though I consider power inequality in gender relations to be of major importance it is not my intention to focus exclusively on gender oppositions and conflicts. In practice men and women have to deal with each other, not only in conflicts, but in cooperation as well. Their interests are not always opposed, but shared in many instances. It is not my intention to depict men and women as natural enemies. My aim of studying gender is rather to gain insights on mechanisms that enable women to cope with men on more equal terms, instead of being the eternal underdog.12

At present feminist anthropology is no longer exclusively about women, but rather about the centrality of gender as an analytic concept'in understanding human culture and society. (Moore, 1988; Morgan, 1989) Morgan points out that it is not logical to exclude gender from discussions on social stratification, that are all about power relations. She argues that:· "gender, race, ethnicity, caste, and class are best analysed as they intersect in creating social relations of power, and as different historical, cultural, and political-economic processes shape the configuration of power within.a society" (1989: 11). I would add that other factors that have a role in shaping power relations, such as age or kinship hierarchy, have to be included and analysed ~n an 'intersecting' way as well, depending on the context informants live m. Thus, all forms of patterned inequality deserve recognition (Di Leonardo, 1991).

In sum, in my study the concept of gender is used to delineate cultural meanings and social roles of men and women from the biological characteristics of sex. I understand it as an analytical concept to study relations between men and women, that are characterized by both inequality in power and interdependency. Material and social manifestations of gender have been studied by looking at the distribution of activities, productive sources, benefits and decision making. Gender ideologies have been investigated by finding out why certain gendered arrangements aie adhered to. Since women's lives can not be understood without men a gender perspective in research is required. The concept of gender can not be studied without perceiving other power relations, because they all intersect in determining social stratification. In my fieldwork especially class and communal divisions have been considered.

Class divisions are defined by the size of landholding per household, land being the main productive resource in an agricultural community. In a densely populated and poor area, such as northern Bengal, control over and access to land

12 The interdependent or cooperative aspect of gender is mostly emphasized by gender specialists from the South, e.g. Agarwal (1994).

S T I L L W A T E R S R U N D E E P

forms a crucial source of survival. It must be realized that small differences

between households in terms ofland holding and alternative sources of income have enormous effects in differential access to sources, like credits and sufficient

agricultural inputs, and thus overall developmental potential. I follow the official classification of the Government of West Bengal, which is also used in the

NBTDP13:

Class Hectares Acres Big ha

Marginal 0

-

1 0

-

2,5 0,1

-

7

Small 1 2 2,5

-

5 7 15

Middle 2 4 5

-

10 15 30

Large > 4 > 10 > 30

Communal divisions refer to differences between religious communities, both to differences between Hindus and Muslims and differences between castes. Though caste differences tend to overlap with class differences they are conceptually separated here, as in practice the overlap does not always apply.

The rejection of structuralist dichotomies

Feminist anthropologists have deconstructed many 'taken-for-granted' concepts that ignored women or seriously distorted the diverse realities they live in. For example, the concept of the household has often been (and still is) subjected to assumptions, that were not questioned by researchers, usually with a middle-class and often western background. On the other hand feminist anthropologists themselves have also introduced new concepts that have .not proven to be valuable in analyzing

,,.,,,.,.,,,.~··~~

,Ji~!l.ci~c:i::~l;!tiQQ§:. In a way they fell in similar traps as the ones they were eager to criticize. Thus, the structuralist thinking in dichotomous pairs, that functioned as theoretical frameworks to analyse gender relations (e.g. nature/culture;

public/private) was left behind.

The debate on the publidprivate opposition developed in consequence of inquiries into the reasons for women's subordination and the process of its

development. In this debate the dichotomy of publidprivate originally functioned as the starting point of new analysis of power relations between men and women. It

13 It is debated if these class demarcations reflect the main economical divisions and, thus, if targeting has been adequate. The All India National Sample Survey observes sharper

demarcations. Marginal farmers are those farmers who own less than one acre, small farmers own less than 2,5 acres, middle farmers own less than 5 acres and large farmers own more than 5 acres (REU 15, 1994b: 19-20; Schenk-Sandbergen, 1994: 182).

S T I L L W A T E R S R U N D E E P

was argued that the division of private and public spher~s is universal. The

inferiority of the private sphere to the public sphere was seen as the key explanation of women's subordination to men, because women are generally associated with the private sphere (Rosaldo, 1974). This approach has been criticized for several reasons. Firstly, it is argued that supporters of the dichotomy hold ethnocentric assumptions. Moore suggests that the division between private and public spheres can only be maintained when it is assumed that mothers and the mother-child connection have 'natural' and universal functions, whereas there is no empirical base for such assumptions (1988: .21-30). Secondly, it is theoretically not tenable to analyse men's status and women's status in separate spheres. Empirical evidence shows that female members of households are involved in numerous relations that take place between households, that are 'public' in character. Examples, are women's cooperation in saving and collectively performed (labour), or women's utilization of kinship relations to gain access to resources outside the household or organize marital bonds. Further more, the division of private and public spheres, as far as it is possible to speak of a real division, is not a cause, but rather a result of existing power relations (Mossink, 1981). Thus, a universal framework to analyse women's position the private/public dichotomy should be discarded.

A theoretical framework that can be discarded for similar reasons is presented by Ortner (1974). Ortner explains that the universal subordinate position of women is caused by symbolical association of women with 'nature' and men with 'culture'.

It is presumed that culture is universally perceived as superior to nature. Again, ethnocentrism is playing up. The hierarchical division between nature and culture is not known to all cultures. (Moore, 1988: 13-21)

Especially women are pinned down on their biological capacities and their social role as a mother, as if they form one homogeneous category. Stereotypical, often western middle class, ideas prevail and prescribe the role and tasks of women as 'mothers' and 'housewives', thereby tying them to the household-domain. Men, on the other hand, are often expected to provide for and protect their wives and children and have 'natural' leadership of the household, capable of representing 'the household interest' outside the household. Differences between women, across cultures, communities, classes, etc. are easily neglected in this way. The problems with structuralist dichotomies in relation to studying homogeneous categories of 'men' and 'women' can be summarized as follows: they result in a static picture of a complex and cultural specific reality, they are too universalistic pretentious and, last but not least, too stigmatizing. The critics of dichotomous pairs shows us that analytical categories as 'men', 'women', 'public' and 'private' should be understood

S T I L L W A T E R S R U N D E E P

as cultural specific categories and indicate why a move has been made towards a more relative concept to deal with relations between men and women: gender.

The use of these insights to my study are that the boundaries of public and private spheres need to be crossed in order to comprehend gender relations. The association of women to the private or domestic sphere and their activities and responsibilities with domestic work causes economists to neglect the productivity of women's work14 and prevents irrigation planners from understanding women's roles and interests in irrigation development. In contrast to domestic activities irrigated agriculture is predominantly associated with the public domain and productive activities. Especially in a society that is known for its segregation of men and women assumptions about women's roles and interests in irrigation are easily made.

The deconstruction of the household concept

The deconstruction of the household concept can be seen as a result of the rejection of the private/public framework. The term 'household' refers to the basic unit of society involved in production, reproduction, consumption and socialization. Moore (1988) points at the existence of naturalistic assumptions concerning households and the gender relations within them. She explains that the English word 'household' implies 'co-residence', and is therefore associated with 'intimacy' and 'sharing'. When at the household level everything is assumed to be shared, further questioning on relations at this level might not be considered to be relevant. Furthermore, though different members of this basic unit do cooperate, they do not necessarily have the same interests. The specific organization and composition of households have a direct impact on women's lives, and in particular on their ability to gain access to resources, to labour and to income. The assumption that the productive resources of a household are always controlled by and through men is incorrect. This is illustrated by the growing number offemale headed households (FHH). The control and allocation ofresources within the household is a complex web of rights and obligations. The management oflabour, income and other resources is something which is crucially bound up with household organization and the sexual division of labour. (Moore, 1988: 54-56)

Assumptions concerning the unitary character of a household can influence analyses, in a way that intra-household dimensions become invisible. Within households individuals differ in terms of gender, age, and marital position. Neglect

14 The opposition between unpaid and paid work is largely connected to the domestic versus public opposition.

S T I L L W A T E R S R U N D E E P

of these intra-household dimensions, for example by assuming that benefits are equally shared among the members of one household, may hide certaia negative developments for some members of households, usually the already weakest members (e.g. women, children and/or elderly people).

In my fieldwork a household is defined by those people who usually share their meals that are prepared on one stove.

2.2

Conceptualizing South-Asian gender relations

and ideologies

In this paragraph I will show how the feminist anthropological discussions, explained above, have implications for understanding and describing the

phenomenon of purdah, a key concept in focusing on South Asian gender relations.

Purdah is a broad and flexible phenomenon with multiple meanings and

manifestations across South Asian regions, religious communities, socio-economical classes and age-groups. However, it is in the first place associated with its narrow meaning of female veiling and seclusion, which does not apply to all purdah regions and observers and emphasizes the polarization between men and women. Since, as I will argue below, studying purdah does not have any relevance without perceiving men, I shall suggest to study the phenomenon from a gender perspective, instead of using it as an independent analytical concept. The connection between material manifestations of gender, unequal control over and access to resources (e.g. land, work) and benefits (e.g. incomes, food) and their linkages with social institutions, such as the organization of marriage, kinship, inheritance and residence, and gender ideologies, the perceptions of masculinity and femininity and the relations between them should be clearly established. My analysis of purdah implies crossing the boundaries of private and public spheres. This is less paradoxical than may appear at first sight. Men's control is largely based on women's seclusion.

In India the practice of purdah, generally seen as a symbol and manifestation of the subordination of South-Asian women by men, is most seriously adhered to in the northern regions, both by Hindus and Muslims. In a narrow sense purdah refers to the segregation of women and men, through seclusion and veiling of women.

Purdah literally means 'curtain'. It is a Persian word which traveled to the Indian sub-continent together with the Moguls and the Muslim religion. This does not mean that purdah-like practices did not exist prior to Mogul settlement. Pre-Mogul Hindu purdah-practices may have been reinforced, due to Muslim influences, but can

S T I L l W A T E R S R U N D E E P

not be seen as a copy of Muslim practices (Sharma, 1980: 5; Jacobson, 1970: 7-13 ). In a broader sense purdah refers to the ideologies that support it.

Purdah-ideologies may be defined as wide range of ideas concerning women's role in public life, her position and function within the household and women's physical capacities and characteristics. Ideological adherence to (some form of) purdah does not

necessarily imply purdah-behaviour. Some material wealth is a precondition to withdraw women from labour participation or provide clothing to cover them entirely.

Some social scientists emphasize the function of purdah in maintaining and reinforcing (family) honour and solidarity of the patriarchal family (Mandelbaum, 1988; Vreede-de Stuers, 1968; Jacobson, 1970). In case a woman would not behave according the purdah-ideology inside the house she would endanger the bond and solidarity of male affines. Outside the house a woman's neglect of purdah could endanger herself and thereby her honour, the honour of the family, and most of all the honour of the household males.

A Hindu bride who comes to live with her in-laws, usually in another village than her parental village, has to abide to 'correct' behaviour and know her position at the bottom of the family hierarchy. Her behaviour should not give umbrage to her superiors. Veiling to her in-laws, men and women, may be required. Jacobson shows how 'respect-avoidance' and joking' relations between affines contribute to the harmony of the patriarchal joint-family. In a village near Bhopal, where she performed her fieldwork, Hindu women veiled for elder in-laws, especially men, but not for people of their parental village. Meetings between in-law families were characterized by respectful detachment or obscene jokes. The latter were especially made by natal women about their male affines.

Muslim women veil primarily for men who can or could have been a potential husband. Muslims generally value the social invisibility of women, because of suspected dangers outside the house. In most cases Muslim orthodoxy hardly seems to be the real issue, though it might be perceived in this way. According to Bhatty the purdah-practice is deemed necessary: "because woman is conceived as a sex object, who inevitably arouses, by mere appearance, sensual desires in men over which they are not expected to have any self-control (1976: 104). It is believed that women are physically and mentally weak and should be protected against

themselves, because the family honour may not be endangered by them. The significance of this image of women's physique and sexuality appears from the fact that Muslim girls are supposed to go in purdah when they reach puberty.