in Farming Systems Research and Extension
Published in cooperation with the
Women in Agricultural Development Program,
University of Florida
in Farming Systems
Research and Extension
Susan V. Poats, Marianne Schmink,
and Anita Spring
BOULDER G LONDON
Westview Special Studies in Agriculture Science and Policy
Figures 28.1 and 28.2 are reprinted by permission of Lynne Rienner Publishers, Inc., from Peter E. Hildebrand and Federico Poey, On-Farm Agronomic Trials in Farming Research and Extension. Copyright C 1985 Lynne Rienner Publishers, Inc.
This Westview softcover edition is printed on acid-free paper and bound in softcovers that carry the highest rating of the National Association of State Textbook Administrators, in consultation with the Association of American Publishers and the Book Manufacturers' Institute.
All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher.
Copyright ( 1988 by Westview Press, Inc.
Published in 1988 in the United States of America by Westview Press, Inc; Frederick A. Praeger, Publisher; 5500 Central Avenue, Boulder, Colorado, 80301
Library of Congress Cataloging-in-Publication Data Gender issues in farming systems research and extension.
(Westview special studies in agriculture science and policy)
"Based on the 1986 University of Florida conference"-"Published in cooperation with the Women in Agricultural Development Program, University of Florida"-1. Agricultural systems--Research--Congresses.
2. Women in agricultural--Congresses. 3. Agricultural extension work--Congresses. 4. Farms, Small--Congresses.
5. Agricultural productivity--Congresses. I. Poats, Susan V. II. Schmink, Marianne. III. Spring, Anita. IV. University of Florida. Women in agricultural Development. V. Series.
$494.5.$95G46 1988 630'.88042 87-34315
Printed and bound in the United States of America
OThe paper used in this publication meets the requirements
of the American National Standard for Permanence of Paper
for Printed Library Materials Z39.48-1984
6 5 4 3 2 1
List of Tables and Figures . ix
Marianne Schmink and Susan V. Poats . xv
PART I EOE ICAL AND METHODOLOGICAL IMPLICATIONS
OF THE INCUSION OF GENDER IN FARMI SYSTEMS
RESEARCH AND rMMiSION (FSR/E)
71,l Linking FSR/E and Gender: An Introduction
Susan V. Poats, Marianne Schmink, and
Anita Spring . 1
2 Integration of Intra-Household Dynamics
into Farming Systems Research and Extension:
A Survey of Existing Projects
Rosalie Huisinga Norem . 19
3 Gender Relations and Technological Change:
The Need for an Integrative Framework of Analysis
Alison Evans . 37
4 Problems of Understanding and Communication
at the Interface of Knowledge Systems
Janice Jiggins . 49
5 From Recommendation Domains to Intra-Household
Dynamics and Back: Attempts at Bridging the
Amalia M. Alberti . 61
6 Research, Recommendation and Diffusion Domains:
A Farming Systems Approach to Targeting
Peter Wotowiec, Jr., Susan V. Poats,
and Peter E. Hildebrand . 73
7 Incorporating Women into Monitoring and Evaluation
in Farming Systems Research and Extension
Jonice Louden . 87
8 A Comparison of Rural Women's Time Use and
Nutritional Consequences in Two Villages in Malawi
Lila E. Engberg, Jean H. Sabry,
and Susan A. Beckerson. . 99
9 Correcting the Underestimated Frequency of the
Head-of-Household Experience for Women Farmers
Art Hansen . 111
10 An Evaluation of Methodologies Used in Time
Eva Wollenberg . 127
11 Gender, Resource Management and the Rural
Landscape: Implications for Agroforestry
and Farming Systems Research
Dianne E. Rocheleau M . 149
PART II LIATIN APRICA AND THE CARIBBEAN
V2 Farming Systems Research in the Eastern Caribbean:
An Attempt at Analyzing Intra-Household Dynamics
Vasantha Chase . 171
13 Economic and Normative Restraints on Subsistence
Farming in Honduras
Eunice R. McCulloch and Mary Futrell . 183
14 Phases of Farming Systems Research: The Relevance
of Gender in Ecuadorian Sites
Patricia Garrett and Patricio Espinosa . 199
15 Technological Domains of Women in Mixed Farming
Systems of Andean Peasant Communities
Maria E. Fernandez . 213
PART III ASIA AM THE MILE EAST
16 The Household Enterprise and Farming System
Research: A Case Study from Taiwan
Rita S. Gallin and Anne Ferguson . 223
17 The Contribution of Women to Agriculture in Taiwan
Jane E. Gleason . 237
18 Gender-Differentials in the Impact of Technological
Change in Rice-Based Farming Systems in India
Bahnisikha Ghosh and Sudhin K. Mukhopadhyay . 253
19 Women in a Crop-Livestock Farming Systems Project
in Santa Barbara, Pangasinan, Philippines
Thelma R. Paris . 269
20 Gender Related Aspects of Agricultural Labor
in Northwestern Syria
Andr6e Rassam and Dennis Tully . 287
PART IV AFRICA
21 The Women's Program of the Gambian Mixed
Margaret Norem, Sandra Russo, Marie Sambou,
and Melanie Marlett . 303
<22 Intra-Household Dynamics and State Policies as
Constraints on Food Production: Results of a 1985
Agroeconomic Survey in Cameroon
Jeanne Koopman Henn . 315
-_23 Intra-Household Gender Issues in Farming Systems
in Tanzania, Zambia, and Malawi
Jean M. Due. 331
24 Institutional and Policy Parameters Affecting Gender
Issues in Farming Systems Research in Tanzania
ManasseTimmy Mtoi. 345
25 The Impact of Modern Changes in the Chitemene
Farming System in the Northern Province of Zambia
Mary N. Tembo and Elizabeth Chola Phiri . 361
26 A Diagnostic Survey of Female-Headed Households
in the Central Province of Zambia
Robert E. Hudgens . 373
27 The Gender Factor and Technology Options for
Zambia's Subsistence Farming Systems
Alistair J. Sutherland. 389
28 using male Research and Extension Personnel to
Target Women Farmers V
Anita Spring . 407
-29 The Role of Women Farmers in Choosing Species
for Agroforestry Farming Systems in Rural
Areas of Ghana
Kofi owusu-Bempah. 427
OmmRiBuIns . 444
Tables and Figures
2.1 Projects Responding to Survey . 22
2.2 Types of Intra-Household Data Collected by
Projects Responding to Survey . 25
2.3 Uses of Types of Intra-Household Data by
Projects Responding to Survey . 27
2.4 Most Frequently Used Methods of Data Collection
by Type of Data . 29 2.5 Constraints Influencing Projects . 33 8.1 Production Activities . 103
8.2 Mean Time Spent in Hours by Husbands and Wives in
Two Seasons in Each Farming System . 105
8.3 Mean Amount of Traditional Foods Stored by Type
of Farming Household in Two Seasons . 107
8.4 meals and maize Flour Consumed and Percent of
Children with Normal Weight for Height by Type
of Farming Household and Season . 108
9.1 marital Changes as Expressed in migration
of Women in and out of Rural Areas of
Northwestern Zambia, 1971 to 1972 . 116
9.2 Changes in Gender of Household Head in
Lilongwe Rural Development Project, Malawi
1980/81 to 1982 . 119
9.3 Patterns of Changes of Gender of Household
Heads in Intensively Surveyed Sub-sample . 121
10.1 Distribution of Time by Activity and Gender . 139 10.2 Percent of Time Spent by Gender, Location,
and Field Types . 141 12.1 Communities Sampled by Area Focused Study
in M.aJouya Valley . 176 12.2 AgroecologiCal Characteristics of
the Mabouya Valley . 177 12.3 Socioeconomic Characteristics of
the mabouya Valley . 178 13.1 Annual Average Yields for Upland Intercropping
of Maize, Sorghum and Beans. 187 13.2 Recommended Daily and Yearly Intakes of Nutrients
for Average Family . 190 13.3 Subsistence Activities and Nutritional
Adeuacy . 191 13.4 Foods and Their Contribution to Low Cost Diet .194 16.1 Population of Hsin Hsing by Period and Age,
1958-79 . 226 16.2 occupations of Hsin Hsing Households, 1979 . 229 16.3 Land Use of 61 Farming Households,
Hsin Hsing, 1978-79 . 230 17.1 Income Earned from Off-Farm Employment .241 17.2 Labor Usage (in hours) by Gender and Crop .245 17.3 Crops and Total Hectarage of Farms without
Female Family Labor . 248 17.4 Crops and Total Hectarage of Farms with
Female Family Labor . 249 17.5 Diversification Indices of 30 Farms .250
18.1 Time Allocation of male and Female Population:
Hours per Person per Day. 257
18.2 Male-Female Differences in Labor Input in
Rice Cultivation . 258
18.3 Time Allocation by Male and Female Population
in Economic and Home Production Activities:
Bhandarkona (hours per person per day) . 262
18.4 Labor use Functions: Household Survey Data (Nadia):
Pooled (Log Linear) . 265
18.5 Labor use Functions: FMS Data (Hooghly and Birbhum)
Dependent Variable: Share of Women in Total
Labor Use . 266
19.1 Percentages of men, women, and Children in Sample
Households Providing Different Sources of Labor by
Specific Activity in Rice Production . 275
19.2 Percentages of Men, Women, and Children Providing
Different Sources of Labor by Specific Activities
in Mungbean Production . 276
19.3 Percentages of men, Women, and Children Providing
Different Sources of Labor by Specific Activities in Cowpea Production . 277
19.4 Sample Households Using Different
Sources of Labor by Specific Activities
in Squash Production . 278
19.5 Percentages of Men, Women, and Children in Sample
Households Providing Different Sources of
Labor in Specific Activities in Cattle
and Carabao Production . 279
19.6 Percentages of men, Women, and Children within
Households Using Different Sources of
Labor by Specific Activities in Poultry
and Swine Production . 281
20.1 Contribution of males and Females as Percentages of
the Total Timd Spent in On-Farm Agricultural
Production . 291
20.2 Contribution of Males and Females as Percentages
of Hours Spent in Legume and
Cereal Production . 292
20.3 contributions of Males and Females
as Percentages of Hours Spent in Summer Crop
and Tree Crop Production . 293
20.4 Mean Hours of Labor in Cereal Harvest:
Differences Between Farms Using mechanical and
Manual Techniques . 294
20.5 Regression Results on Labor variables . 298
21.1 Maize/Cowpea Yields of Participating Villages . 310
22.1 Men's and Women's Average Labor Hours . 320
23.1 Percentage of Labor Days Contributed by Females
by operation by Crop, Sazpled Families,
Kilosa, Tanzania, 1980 . 335
23.2 Comparison of Crop Acreage, Income, and
Extension Visits Between Contact, Non-Contact,
and Female-Headed Households, Tanga Region,
Tanzania, 1985 . 342
24.1 Monthly Labor Requirement for Farm Operations
for major Crops in Eastern Uluguru mountains . 347
24.2 Tasks Performed by Gender, by Crop,
and by Activity in the Farming Systems of
Eastern Uluguru Mountains, 1984 . 350
24.3 Estimation of Effective Adult-Equivalent
work-Days Available per month for Farm Work . 352
24.4 Labor Requirements and Yield on the
Alternative System of Production . 353
24.5 Expected Returns and Standard Deviation
of the Two Systems of Production . 354
25.1 Adequate Nutritional Status of Children . 368
25.2 Percentage of Female-Headed Households
by Province . 369
26.1 Main Characteristics of Male and Female-Headed
Households in TRD#3 (Mkushi District) . 378
26.2 Comparison of Female-Headed Households in
TRD#3 (Mkushi District) by Presence of Males
in the Household, oxen ownership, and Size of
Acreage Cultivated. 380
26.3 Comparison of Statistics on Female-Headed
Households in 'Three Recommendation Domains
in the Central Province. 384
28.1 Graduates of Malawi Agricultural Training
Institutions: Bunda College of Agriculture, Colby College of Agriculture, and Thuchila
Farm Institute (Farm Home Assistants),
by Year and sex . 410
28.2 Type of Extension Contacts for Male Household
Heads (MMH), Female Household Heads (FHH), and
Wives from the NSSA Extension Survey, Malawi
(in percentages), 1980-81 . 414
28.3 maize Yields from Farmer-Managed On-Farm Trials
Phalombe, Malawi, 1981-82. 417
29.1 Farmers, Household Type and the Average Number of
Forest Products They Know By Age Group .431
29.2 Goals of Farmers in the Forest Savannah
Transitional Zone of Ghana. 433
29.3 Agricultural Systems (Farm Types) in the Forest
ITransitional zone of Ghana. 434
29.4 Types of Deforestation and Land Degradation
Activities in the Forest Savannah
Transitional zone of Ghana. 436
3.1 A Conceptual Map for Looking at the Farm-Household
System from a Gender Perspective. 46
11.1 Pananao Sierra, Dominican Republic . 154
11.2 Fakot Village, Mhaintain Watershed, India .158
11.3 Sources of Livelihood (Cash & Kind)
in Fakot. 159
11.4 Misamfu, N.E. Zambia . 162
12.1 Agroecological Transect of the Mabouya Valley
of St. Lucia . 175
13.1 Conceptual Model of Subsistence FarmingNutrition System . 188
14.1 Phases of Farming Systems Research: A Synthesis
of INIAP and ICTA Procedures . 201
18.1 Allocation of Time to Different Activities:
Male-Female Differential . 260
18.2 Gender Differential in Allocation of Time
to Different Activities by Socioeconomic
Groups and by Technology . 263
24.1 Conceptual Framework of Institutional,
Cultural and Policy Factors Affecting
Intra-Household Dynamics in FSR . 348
24.2 E,J- Relationships Generated by Traditional
and Alternative Farm Production Systems . 355
26.1 Busiest Months for Male and Female-Headed
Households . 382
28.1 Grain Yield Response for Local Maize (L) and
CCA Composite (C) to Environment, without
Fertilizer, Phalombe Project, Malawi . 419
28.2 Grain Yield Response of Local Maize (L) and
CCA Composite (C) to Environment, with
Fertilizer, Phalombe Project, Malawi . 419
29.1 Diagrammatic Presentation of a Woman Farmer's
Traditional Agroforestry Farm in the Forest
Savannah Transitional Zone of Ghana . 432
29.2 Proposed Field Layout of On-Farm Agroforestry
Research Plots for a Typical Village in the
Forest Savannah Transitional Zone of Ghana . 438
This book is the product of an international conference hosted by the Women in Agricultural Development (WIAD) Program at the University of Florida from February 26 to March 1, 1986. The WIAD Program's general purpose is to promote an understanding of gender and its relevance for agricultural development processes. Women are critical to agricultural production in many places, but access to resources and research technologies may be constrained by gender and this may lead to detrimental effects on the design and implementation of sound agricultural programs. The Program seeks to support and develop expertise related to the roles of women and intra-household dynamics in agricultural production, research, and extension in order to improve the design and implementation of agricultural programs. The Program's primary audience is the faculty and students of the University of Florida, but the Conference and this book permitted us to reach out to a wider national and international audience.
The Women in Agricultural Development Program began as the Women in Agriculture (WIA) group at the University of Florida (UF) in 1983 after five years of informal activities related to women and development. It became an official University of Florida program in 1984 and the name was changed to Women in Agricultural Development in 1986. During early years, several initiatives emerged from the International Programs office and from social scientists, especially those in the Centers for African and Latin American Studies. Momentum built up as a critical mass of faculty came to the UF with experience in the WID field. A group that formed in the fall of 1983, with Anita Spring as Director and Marianne Schmink as Co-Director, served as an ad hoc committee to promote awareness of issues related to WID and, more
specifically, to women's role in agricultural production. The WIAD group received support from the USAID Program Support Grant funds administered by the International Programs office and from other on-campus sources including the Centers for Latin American and African Studies and the Graduate School. Principal program activities include a bi-weekly speaker series and a bibliographic compilation of readings on women in agriculture, as well as work in curriculum development and in technical assistance.
From the beginning, the WIAD Program sought to focus its efforts specifically on the realm of agriculture, in recognition of the UP's strengths in this area. It took advantage of the opportunity to interface directly with a major USAID-funded project, the Farming Systems Support Project (FSSP) that had a worldwide mandate for technical assistance, training, and networking activities. As Associate Director of the FSSP based at the UP, Susan Poats was in a key position to facilitate that linkage. In addition to the institutional presence of the FSSP, the general philosophy and methodology of the "farming systems" approach provided promising avenues for attention to women. Yet gender had not been incorporated effectively and systematically as a variable in most farming systems work. The WIAD Program hoped to stimulate the development of conceptual and methodological approaches that could improve the incorporation of gender issues into farming systems activities. These ideas are discussed in more detail in Chapter 1.
In early 1984, the WIAD Program hosted a visit to the UP by several colleagues involved in an effort by the Population Council to produce useful case studies of gender issues in development projects. This meeting led to the creation of the FSSP/Population Council IntraHousehold Dynamics Case Studies Project, co-managed by Susan Poats and Judith Bruce of the Population Council, with funding from USAID and from the Ford Foundation. The FSSP created a new task force to focus on the integration of household and family concerns into the farming systems perspective. In 1984, Cornelia Butler Flora (a member of both the FSSP's Technical Committee and of the task force on the family) visited UF. She also wrote a position paper on "Intra-Household Dynamics in Farming Systems Research: The Basis of whole Farm Monitoring of Farming Systems Research and Extension." She and other task force members were instrumental in supporting the proposal to develop case study training materials through the PSSP/Population Council project.
The case study project, initially launched from a WIAD activity, generated further momentum. In February of 1985 the project's advisory committee contacted those individuals and projects around the world who might be interested in developing case study materials from their own experiences in farming systems work. A survey questionnaire requested more detailed information from those who responded. The results of this questionnaire provided some indication of the current state of knowledge on intra-household dynamics in on-going farming systems projects (see Chapter 2). The FSSP/Population Council project selected eight proposals to receive support for the development of case study materials. Yet the more than 75 respondents who expressed interest in the project indicated that the incorporation of intrahousehold variables was a greater concern to farming systems practitioners than had been anticipated.
Parallel to the case studies project, the WIAD Program proposed bringing together farming systems practitioners to a conference on the UF campus. The focus of the event would be purposely narrow in order to maximize the potential for improving the integration of gender into the farming systems approach, taking advantage of the interest aroused by the case studies initiative. Sessions would address specific issues of theory, method, and policy related to Farming Systems Research and Extension (FSR/E) in developing areas, systematically comparing African, Latin American, and Asian experiences. Marianne Schmink volunteered to organize the conference and Susan Poats coordinated the format of the program. WIAD's Steering Committee served as an advisory body and a conference task force was appointed. Funding came from the Center for Latin American Studies, the Center for Tropical Agriculture, and the Center for African Studies, as well as from the Ford Foundation and the Rockefeller Foundation. A call for papers circulated in July of 1985 defined specific topical areas of concern and asked potential participants to summarize the data they would present, its relation to FSR/E, the geographic focus, and its relevance to the designated themes.
The response was overwhelming. over a hundred paper proposals arrived from all over the world, expanding the geographic focus of the conference. In an effort to accommodate as many participants as possible, the organizers structured a comprehensive program of concurrent sessions and explored as many sources as possible for travel support, giving priority to visitors from developing countries. over 40 participants received full or
partial funding to attend the Conference. There were 91 speakers in fifteen formal paper sessions, three roundtable discussions, and an after-lunch lecturer. Carmen Diana Deere of the University of Massachusetts at Amherst delivered the keynote address on "Rural Women and State Policy: An Evaluation of the Decade" (later published in revised form in her book with Magdalena Leon, Rural Women and State Policy, Westview, 1987). Films and recent publications on FSR/E and gender were available for review during the Conference and several social events facilitated a high level of information exchange and networking. Attendance at conference sessions was consistently high. A total of 298 persons were registered or on the formal program and many others from the campus and local area attended without registering. The magnitude and complexity of the Conference demanded the assistance of many volunteer helpers. Special thanks are due to Carol Brown, Donna Epting, Jean Gearing, Janet Hickman, Patricia Kuntz, Cindy Lewis, Greg Moreland, Bill Reynolds, Barbara Rogers, Robin Sumner, and Darla Wilkes. Assistance was also provided by Wharton Williams Travel, Classique Cuisine, Renaissance Printing, Farming Systems Support Project, International Programs, International Food and Agricultural System (IFAS), and the College of Liberal Arts and Sciences.
Chris Andrew, Director of the FSSP, formally opened the Conference on behalf of the University of Florida. In his opening remarks he said:
The intellectual capacity assembled in this
room to address gender-related issues in on-farm,
research is second to none. We are fortunate this week to have presentations by people who have traveled from more than twenty countries,
from forty universities (eleven non-U.S.), from
three International Agricultural Research Centers, and many other national, donor, private voluntary organizations, and private entities.
The titles in the conference program mention
twenty-two countries and we know that more will be discussed. Nearly every region of the world
The intra-household concept, as an integral
and dynamic part of the farm system, must consider gender issues. I emphasize integral and dynamic because we must take care not to establish artificial boundaries to accommodate
conventional simplicity. with a broad view of the household, third world food security can be
addressed in a national framework. We need a
conceptual focus that recognizes the farm household as a critical element in successful agricultural policy for research and development.
To enunciate and address the interdependent
needs of women and men in farming systems as
they interact with the bio-physical environment
is an important goal.
The critical importance of agriculture to the
vitality and strength of many of the world's countries is widely recognized. At the same
time, the increasing diversity, complexity, and
intractability of the problems facing agricultural development make it imperative that agricultural research systems change and adapt to
address specific realities. one of these is the
role of women in agriculture. Another is that
gender issues are being considered together with
the farming systems approach, an agenda that
likely could not have found a platform ten years
ago. In the ebb and flow of agricultural
research, this Conference represents something
of a revolution in agriculture and in
The Conference offered an excellent opportunity for exploring and testing new ideas and successful approaches for incorporating gender sensitivity in agricultural research and development. The FSSP/Population Council project used the Conference to test a case study based on FSR/E project activities in Zambia. The two authors of the case, Charles Chabala and Robert Nguiru, were present to view how the case worked in a training context. Hilary Feldstein, Rosalie Norem, Kate Cloud, Susan Poats, Nadine Horenstein, and Mary Rojas conducted the abbreviated training sessions with approximtely sixty conference participants. Their detailed evaluations and suggestions were used to improve the case. Highlighted
repeatedly during the Conference was the need to inventory and assess the usefulness of various methodologies used in dealing with gender issues. An ad hoc special methodology workshop held on Saturday afternoon
following the close of the formal sessions was attended by more than seventy-five people. Janice Jiggins and Hilary Feldstein are using this session as the basis for developing a methodologies handbook to accompany the FSSP/Population Council case studies.
Aside from the high level of interest in the formal and ad hoc sessions, informal networking was intense among those present. These interactions underscored the remarkable level of interest in the practical aspects of integrating attention to gender issues into agricultural research and extension. The Conference provided a meeting place for people with diverse disciplinary and area expertise to learn from one another. The most exciting outcome of the Conference was the discovery of uch a broad base of research already underway on gender issues in agriculture.
A total of sixty-four written papers were available in a three-volume set at the Conference. In order to make conference materials more widely available, the W M Program decided to organize a selection of the international papers for publication as a book. Although the quality of most of the papers presented was high, space considerations made it impossible to publish all of them. This volume includes less than half of the papers presented at the Conference. In making the difficult decision as to which papers to include, the editors used several criteria. Most important was adherence to the original thematic focus on farming systems work, as outlined in the call for papers. Secondly, preference was given to papers that presented new ideas or methodological approaches for the integration of gender into FSR/E projects. Finally, regional balance and the inclusion of non-U.S. authors who might not otherwise find an audience for their work in this country was sought. The overall goal was to produce a coherent reader with a comparative perspective, that would be both stimulating and helpful to people trying to implement gender-aware farming systems projects. we hope the many excellent papers that could not be included in this volume will find publication outlets elsewhere.
This book, and the WIAD Program's overall effort,
have depended on the hard work and support of a dedicated group of faculty, students, and staff from the UF. A special thanks is due to Jean Gearing, who was responsible for much of the editing, correspondence, typing, and organizational tasks associated with the book. The editors are also grateful to Pamela Shaw for her editorial work, to Kathy Gladden for bibliographic and clerical assistance, to Sharon Leslie and Barbara Rogers for their assistance in artwork, and to Lana Bayles, Kenna Hughey, Shirlene Washington, Sabrina Byron, and Dana Whitaker who provided additional help with the word processing. The book and the Conference that produced it are, in turn, part of a larger effort in which the MAD organizers have
been privileged to collaborate with many colleagues, some of whom are mentioned above. It is hoped that the ideas and experiences discussed by the authors of the following chapters will help to stimulate and improve the consideration of gender as a crucial issue in agricultural production and development.
Susan V. Poats
PART 1. THEORETICAL AND METHODOLOGICAL IMPLICATIONS
OF THE INCLUSION OF GENDER IN FARMING SYSTEMS RESEARCH AND EXTENSION (FSR/E)
Linking FSR/E and Gender:
Susan V. Poats, Marianne Schmink, and Anita Spring
The title of this book is like a code. The two terms t'gender issues" and "farming systems research and extension" are shorthand. Each represents an extensive field of research and practice: women and development or WID, and farming systems or FSR/E, respectively. The two fields have much in common. Both emerged relatively recently in response to dissatisfaction with the results of technological change in agriculture in developing countries. Whereas in the 1950s and 1960s development theory andpractice emphasized growth in productivity, by the 1970s there was a renewed concern to implement programs that conceived of development more broadly, to mean the possibility of better lives for most people. This perspective challenged a development field dominated by technical and economic expertise. Efforts to develop more comprehensive approaches that would bring together technical, economic, and social considerations led to the two interdisciplinary fields of WID and FSR/E.
In this brief introduction, justice cannot be done to either field in its own terms. Rather, the historical and practical considerations that favor their interaction and the conceptual problems such a union can help to overcome are reflected. The discussion will indicate how the following chapters in this book contribute theoretical and methodological insights that can help to make agricultural development programs more efficient and equitable.
FARMING SYSTEMS RESEARCH AM EXTENSION
most farms in developing countries are small scale, with few resources other than family labor. Their subsistence activities are multifaceted and their goals
complex, including both market and non-market considerations. Minimizing risk is especially important when family survival is at stake. Given the constraints they face, small farmers actively seek ways to improve their productivity and to maximize the few resources at their disposal. Agricultural technologies (including equipment, inputs such as fertilizers and pesticides, and management practices) are often designed for farmers with greater resources and market orientation based on an essentially economic calculation of costs and benefits. By the 1970s, development practitioners became concerned that the benefits of agricultural innovation accrued most easily to these wealthier farmers. Yet small ' farmers constituted the majority of .producers and, ultimately, those most directly responsible for the welfare of rural families and communities. What were the social and economic costs of neglecting them? The design of technologies appropriate to the majority of lowresource farmers required an understanding of their particular constraints, goals, and practices that went beyond strictly technical and economic criteria. The farming systems concept emerged as a response to this challenge.
FSR/E is not a single approach, but an array of different perspectives and methods. This diversity is a source of debate and dialogue that continually enriches the field. In this book there is likewise no orthodoxy, but rather a collection of different points of view as to how to conceptualize and carry out farming systems work. The common elements that underlay most versions of farming systems include: an explicit commitment to low-resource producers; a systems approach that recognizes the complexity of small farm enterprises; a focus on the farm family or household; and a recognition of the importance of including farmers in the research and extension process. The concept of "domains" is used to denote the specific client group (defined by environmental, ecological, and/or socioeconomic criteria) to whom the project is oriented.
WOMEN IN DEVELOPMENT
The WID field, similar to FSR/E, began with a concern for the distribution of development benefits. Like farming systems, women and development is far from a unified field of knowledge. Not only does it include many strands of research and practice, but the field has evolved rapidly over the approximately 20 years of its existence, since economist Ester Boserup published her ground-breaking work Women's Role in Economic Development in 1970. Boserup's work challenged the preval-REng notion that economic
development, or modernization, would automatically improve women's status by replacing traditional values and economic backwardness with new opportunities and an egalitarian ethos. She argued instead that economic innovations often replaced women's traditional economic activities with more efficient forms of production controlled by men. Examples included the decline of women's cottage industries due to competition from factories hiring predominantly men and, in some parts of the world, the growth of modern service and commerce sectors in which men predominated, in place of women's traditional marketing practices. The recognition that development, as practiced, might actually worsen women's position relative to men's crystallized the new field of women and development around a concern with equity.
By the late 1970s, however, a growing research base on women's economic activities showed that equity was intimately related to more technical problems of efficiency and productivity. If development undermined women's traditional economic contributions, was this loss compensated by the output of new forms of production? Were new economic opportunities opening up for women? What was the impact of these shifts on the welfare and productivity of the poor populations of the world? The new emphasis on the poor focused attention on women's importance as household producers and providers in addition to their domestic roles. No longer were they to be viewed simply as potential welfare beneficiaries whose needs might be neglected by development efforts. Instead, women were a mainstay of family and community welfare, active producers whose potential contributions were often overlooked or undermined. A clearer understanding of changes in women's role in production therefore was essential for the success of agricultural development projects.
A decade of theoretical experimentation and empirical research on women's role in development moved the field from the stage of raising awareness and clarifying issues to a search for practical applications. How could the WID insights be applied to development work? One solution was to create special projects or components devoted to women. while sometimes successful, these all too often emphasized women's domestic responsibilities rather than their productive work. They also distracted from the more general problem of improving the effectiveness of "mainstream" development projects by making them more responsive to gender differences among the client population. By what practical means could such a formidable task be undertaken?
The first attempts to answer this question produced an array of checklists of questions to be asked and data to be gathered in each project setting. A series of case studies were published as examples of how gender affected development projects. Various institutions compiled handbooks that specified how gender issues could be addressed at each step of the project cycle. But there were not enough experts trained in the analysis of such complex and variable matters as household division of labor, decisionmaking, and income management. Some of the basic issues could be specified in advance, but each setting required a unique assessment of their relevance and of the interaction with other important variables. While hiring more women as project staff members appeared to be a good idea, the gender of the researcher or practitioner turned out to be no guarantee of the requisite analytical skills.
In response to this dilemma, WID efforts in the 1980s sought to develop the tools of "gender analysis" and the methods by which development practitioners could learn and adopt them. USAID fostered a major effort to adapt the Harvard Business School's case study teaching method to training on gender issues in development projects. The office of Women in Development sponsored the writing of several new analytical case studies that were compiled in a handbook that also provided a framework and set of basic concepts to be used in the case study analysis (Overholt et al.,1985). The cases and the training method have been widely used in training workshops that provide practice in tackling a set of questions that might otherwise seem hope-lessly complicated. The strength of this approach is its emphasis on the link between project or development goals and gender differences in the client population. This focus helps to clarify the relevant issues and to indicate priorities for research and action.
GENDER ISSUES IN FARMING SYSTEMS RESEARCH AND D[TENSION (FSR/E)
The farming systems perspective is especially appropriate for such a process-oriented approach to gender analysis. The FSR/E methodology consists of a series of stages (diagnosis, planning and design of technology, experimentation and evaluation, and dissemination) that facilitate the specification of steps to be taken to address particular aspects of research and extension and how to make the best use of different kinds of data. But
FSR/E is also conceived to be an iterative, adaptive process in which, once the project is well underway, the various stages of research take place simultaneously. This philosophy is intended to maximize the potential impact of on-going farmer evaluations on the design and dissemination of future technological changes. The research process allows time to learn about the intricacies of farming systems and to incorporate new insights into more refined measures and project adaptations. Other characteristics of the farming systems approach especially important for gender analysis include its-focus,.on.small farm households and on the participation of farmers in the research and extension process.
Disaggregating Development Beneficiaries
The farming systems emphasis on reaching specific lowincome groups helped to illuminate women's roles in agricultural development. Identifying small farmer constituencies required the disaggregation of society into "target" or "client" groups which brought women's activities into greater focus. The interaction of socioeconomic standing and gender was brought home by the growing recognition that women in poor families played essential economic roles that bore little resemblance to the activities of middleclass and elite women in the same societies. These observations were confirmed by mounting evidence from research that documented poor women's multiple economic activities, low earnings and long work hours, and restricted access to productive resources. women pl , ayed a central role in the low-resource farm households that were the focus of farming systems work.
The surprisingly high and growing proportions of
female-headed households dramatized women's economic importance in poor populations and revealed the extra constraints under which they often labored to achieve family welfare (Buvinic and Youssef 1979). Rural out-migration of men was rising in many parts of the world as a result of development, leaving many women either temporarily or permanently in charge of their households (Palmer 1986). Their efforts were often undermined by labor constraints or by lack of access to productive resources, in part because research and extension services were primarily oriented to male farmers. The focus on female-headed households illustrated how disaggregation of beneficiary populations could more precisely delineate appropriate interventions for specific social groups. It also undermined the assumption that development projects focused on male farmers always would have the most effective impact on family welfare.
The Whole Farm Syst
The systems approach endorsed by the FSR/E constituency lent itself well to illuminating women's economic importance. Small farm enterprises encompass multiple activities whose interaction is key to understanding management decisions and practices. The configuration of a given system changes readily over time in response to both internal and external factors. This holistic, dynamic perspective on small farming enterprises provided a framework within which the family division of labor could be a key focus. Social definitions of which tasks would be carried 6xit-by men or women vary from one society, region, class, or ethnic group to another. This variability indicates that the division of labor is determined not by the physical difference between the sexes, but by the social definitions of proper relations between women and men. The concept of "gender" serves to distinguish the social character of these relationships, and the "sexual division of labor" describes the allocation of tasks and responsibilities to men and women in a particular situation. In practice, farming systems practitioners may disaggregated only so far, stopping short at the analysis of the division of labor within the household or family. Agricultural research has historically focused on specific connodities whose production is market-oriented. FSR/E recognized that small farm enterprises combine crops and animals. Yet the perspective still overlooked other essential activities carried out by farm families, including off-farm work, home-based production for use or exchange, and the work Required to maintain the home and its inhabitants. WID research revealed that women were often predominant in these activities, especially those based in the home that Cnded to be overlooked or viewed as merely "domestic" qrk. While men often specialized in income-generating activities, women typically combined household management, child care, and work to generate earnings (both on and off the fam). These competing demands on their time could serve as a significant constraint to the adoption of new forms of production that relied on women's labor.
In small farm households, decisions reflected priorities and constraints related to a variety of activities and goals, not just to those related to cash crop production. The potential trade-offs between resources devoted to agricultural production and investments in improved family nutrition were of particular concern to farming systems practitioners whose objective was to stabilize or enhance rural welfare. The systems approach adopted by FSR/E practitioners provided a starting point for integrating the
diversity of farm and non-farm activities within a more complex model of the whole farm-household system. WID practitioners collaborated by focusing attention on women's importance in agricultural production, but also in focusing on activities not generally defined as "production" that are nevertheless essential to the well-being and economic livelihood of rural households and communities.
The focus on farm families brought development work much closer to the realities of poor families than was possible using the country level statistics. The concept of "household", sometimes used to denote a residential unit, sometimes synonymous with the nuclear family, was useful in the field of development and in the social sciences in general (Schmink 1984). It provided an intermediate level of analysis (between the individual and the aggregate society) and a convenient unit for the collection of empirical data. The existence of such primary domestic units in virtually all societies implied an attractive universality for the concept of household that was familiar to researchers and practitioners from-their own personal existence. These perceived advantages, however, had hidden drawbacks. Development practitioners often generalized from their own experience, presuming that households elsewhere were similar to those in which they lived, when in fact household structure and functioning is highly variable. Whereas in advanced industrial society productive work is largely separated from the home, the same is not true for agrarian communities. Home-based food processing, handicrafts, care of animals, kitchen gardens, and manufacturing of such useful items as soap and clothing for a peasant family are not analogous to the domestic chores of a middle-class urban housewife. They are productive tasks essential to household welfare. Whereas a U.S. household typically depends on one or two monetary wages for its sustenance, rural families in the developing world rely on a diverse set of paid and non-paid activities for survival.
In many societies women and men have quite separate responsibilities, access to distinct resources, and differentiated control over returns from their own activities. In fact, households are themselves systems of resource allocation (Guyer 1981). The unitary neoclassical view of household income inherited from the advanced industrial nations is especially inappropriate for
such complex situations in which household members have access to different resources and work opportunities, and exercise differing degrees of control over separate income streams that flow through the household. Household decision-making is neither necessarily unitary nor harmonious. Different members may decide about production strategies, contribute labor to specific tasks, or bear responsibility for the use of the commodities produced. The complexity of intra-household dynamics implies that the possibility of competing goals or priorities may require negotiation among household members. Households are also fluid; variability stems from responses to exogenous changes (such as male out-migration), from internal differentiation based on class, income, ethnicity, and culture, and from demographic variables within the household unit (that is, the pattern of family formation, or the "life cycle" of the family).
The internal dynamics of small farm households affect the process of client involvement in the research and extension process. If work responsibilities, control over resources, and decision-making are fragmented within the family unit, who are the appropriate partners in the research process and potential beneficiaries of the 'proposed technologies? Since male household heads are typically the public representatives of family groups, it is often assumed that information and resources conveyed to them will "trickle across" to others in their household. But indirect communications strategies are inefficient and may omit the actual "user" from the process of FSR/E. This omission represents a loss of valuable indigenous knowledge and may lead to inadequate or incomplete application of technological innovations. Since women and men may know about different factors relevant to agricultural production, the labor of one may not necessarily substitute for the other. If farming systems projects are to succeed in forging effective collaborative ties with their client population, they must include both women and men farmers as partners in the research process.
CONTENTS OF THIS BOOK
The chapters of this book are a selection of the papers concerned with developing countries in Latin America, Asia, middle East, and Africa presented at the Conference in 1986 at the University of Florida (see Preface). The
Conference's primary objective was to bring together scholars and practitioners with expertise and interest in FSR/E to discuss state-of-the-art issues related to gender in FSR/E. In order to maximize coherence, participants were asked to prepare papers that would address specific issues of theory, method, and policy related to FSR/E across developing regions. The following questions were posed under several topical areas:
The Whole Farming System. How can key components of a farming system, including non-farm activities, livestock, secondary crops, food storage, and food processing, be identified? How does the division of labor by age and gender constrain or facilitate specific economic or productive goals? What is the potential impact of improved agricultural technology on-each of these goals and on household members responsible for specific production activities?
Intra-Household Dynamics. What are the key aspects of internal heterogeneity of household units: differential access and use of resources within households; multiple enterprises and their interactions; substitutability and specialization of labor in agricultural activities; marketing outlets and their relationship to differing or conflicting priorities and needs within farm units; and how might proposed interventions alter the balance of power and advantage?
Institutional and Policy Concerns. How does the surrounding environment beyond the farm gate at household, community, and other social levels differ for men and women? What specific constraints to production are posed by these gender differences? How can FSR/E address constraints such as legal status, restrictions on mobility, domestic obligations, property rights, access to credit, markets, and employment?
Definition of Research Domains. How can the key
components and actors within household and farming systems be identified? What are key constraints to productive activities and how does access to production inputs differ by gender? What is the significance of, and interactions between, multiple enterprises within the farm household and how do they create different labor requirements, goals, incentives, markets, and priorities for different family members?
On-Farm Research and Extension. Who are the specific audiences for direct involvement in on-farm research and extension? How do labor constraints affect proposed solutions and how do labor patterns impact on household members who differ in their access to resources? How can
extension strategies be devised that are responsive to the productive activities of both women and men? Which extension mechanisms are most effective in reaching both male and female farmers?
Monitoring and Evaluation. How can strategies be
designed for monitoring the differential impact of FSR/E interventions on different individuals and enterprises within the farming system? What are the unanticipated effects of technological change? In what ways does misunderstanding of gender issues lead to inadequate planning and design or diminished returns to FSR/E projects? How can these effects be minimized?
The papers contained in this book do not exhaust the answers to these questions, but they do provide a beginning. Authors were asked to include a common "minimum data set" in their case study material to facilitate comparison. The following chapters contain many innovative approaches to conceptualizing and carrying out farming systems projects that effectively take gender into account. They highlight several features of the farming systems approach that could be improved by more attention to gender, and they suggest practical ways that this could be done. The book presents a comparative perspective on.the relevance of gender to farming systems work in the developing regions. Two dimensions run throughout the various chapters: the presentation of site-specific data that will permit in-depth analysis of specific cases and the search for conceptual and methodological innovations.
Part I presents a set of articles that focus on key theoretical and methodological issues relevant to the farming systems approach. In Chapter 2, Rosalie Norem. summarizes the results of a survey ofGrming systems projects that collected data on intra-household dynamics and gender differences. Project staff expressed a need for more intra-household data, especially on the factors determining household variability (such as out-migration and the family "life cycle"), on specific labor constraints stemming from the gender division of labor, and on income management within the household. Her findings also reveal that different kinds of information are useful at each phase of a project. In Chapter 3, Alison Evans discusses some of the problems with FSR/E procedures that impede the effective integration of gender considerations, including the emphasis on market criteria and measure, and the assumed homogeneity of the farm household. She presents a framework of ideas to help in constructing a broader, more dynamic model of farm-household systems. She also discusses the relevance of gender at different points in the FSR/E
process, and institutional constraints that must be overcome to improve attention to gender.
Janice Jiggins, in Chapter 4, continues the discussion by focusing on the problems of communication between researchers, farmers, and extension workers, using examples from Zambia and Lesotho that show the rationality and flexibility within the domestic domain. She explores the difficulties of reconciling scientific knowledge systems with those of indigenous people in the course of conducting on-farm research. Lessons drawn from her examples point to the need for establishing key field-household interactions at an early stage of the diagnostic process and to developing methods for mutual communication of key concepts across researchers and female producers, distinct knowledge systems. She proposes the use of situation-analysis based on critical incident technique and peer group workshops as appropriate methodologies for improving communication and diagnosis.
Amalia Alberti focuses in Chapter 5 on the problem of generating data sufficiently sensitive to gender differences to guide the definition of client groups during the initial phases of a project. Echoing Evans' point that a priority for wealthier, more market-oriented farmers will tend to exclude women, she discusses the pros and cons of different sources of techniques for data collection in FSR/E projects. The following chapter, by Peter Wotowiec, Jr., Susan Poats, and Peter Hildebrand, explores in more detail how definitions of client populations need to be modified in accordance with the problems posed at different stages of the project cycle. They offer a refinement of the conventional FSR/E concept of "domains" to distinguish between "research domains" (that maintain variability), "recommendation domains" (homogeneous for technology testing), and "diffusion domains" (for disseminating new technologies). In Chapter 7, Jonice Louden summarizes the compelling reasons for incorporating gender issues into FSR/E monitoring and evaluation systems, especially in a country such as Jamaica where women play key roles in agricultural production. The FSR/E process presents an opportunity to collect systematically valuable information that can help to inform project implementation and refine gender-sensitive measures of key indicators of development.
Chapters 8, 9, and 10 take up conceptual and methodological issues beyond the farming systems universe. Dissatisfied with the market bias of standard economic models of the household, authors Lila Engberg, Jean Sabry and Susan Beckerson propose an alternative production activity model based on measures of time allocated to
income-generation, subsistence, and home production. The more integrated model suggests trade-off s between labor allocated to cash and to subsistence activities in M~alawi that could have nutritional implications. Also concerned with the concept of household Art Hansen presents data, in Chapter 9, from Africa that suggests caution in conceptualizing and measuring the frequency of female-headed households. His findings show that static surveys may underestimate the probability that a woman will be a head of household at some moment or moments in her life, thereby reinforcing the importance of involving both men and women in development efforts. Eva Wollenberg's Chapter 10 discusses the strengths and weaknesses of various time allocation methodologies and their relevance to farming systems work. She explores how four different approaches to the collection of time use data were used in a Philippine project. Her discussion emphasizes the dynamic nature of the FSR/E process, a theme common to all of the chapters in this section. Gender patterns and intra-household relationships become relevant to different degrees and in different ways at each point in the project cycle. These chapters, and others following, provide concrete suggestions as to how an unfolding farming systems project team can collect and analyze the information that will enable it to develop and adapt production technologies to the needs of different users.
Diane Rocheleau, in Chapter 11, draws upon experience from a broad range of countries; she details a land user perspective as an appropriate method for incorporating women as clients and active participants in agroforestry projects. Her paper and that of Owusu-Bempa (Chapter 29) pliyk an important role in expanding the horizon of FSR/E to consider the rural landscape as the context and focus for projects in order to address the gap between natural resource management and farming systems research. The role and domain of women in the interface of these two areas is clearly laid out as the next critical frontier for expansion of household research and gender-based analysis.
Beginning with Part II, the Chapters explore gender issues in FSR/E on a regional basis. Women's roles in agricultural production are less visible in Latin America than id-6jieri 'developing -regions.- The -same is not true in the Caribbean nations such as St. LUCia,-desaribed in Chapter 12 by Vasantha Chase. While (island womeiin\play a significant role in farm work and decisi-making, they receive less income and fewer extension services than do male farmers. Informal data collection methods reveal that female-headed households face particular labor and input constraints that limit their output, choice of crops, and
amount of land planted. A concern with integrating food consumption into the farming systems approach led the Caribbean Agricultural Research and Development Institute (CARDI) to recommend labor-saving methods of backyard garden production, oriented to improvement of family nutrition. The link between nutrition and agricultural change is also the focus of Eunice McCulloch and Mary Futrell's Chapter 13. Their measures of the nutritional output of cropping activities reveal the "low level steady state" farming system that maintains Honduran families at risk of persistent malnutrition.
In Chapter 14, Patricia Garrett and Patricio Espinosa describe the steps taken by the Bean/Cowpea Collaborative Research Project in Ecuador to adapt project activities to gender and social class differentiation. Their rich discussion of the FSR/E process demonstrates the importance of women's participation in production and decision-making, even in Andean Latin America where farming commonly is assumed to be the domain of men. The same is true in the Peruvian highland community described in Chapter 15 by Maria Fernandez, were women are responsible for most tasks associated with livestock production. Recognizing women as a key source of knowledge on traditional livestock production practices, the project team experimented with a variety of strategies to draw them into active participation. Fernandez' argument reflects the emphasis in Rocheleauls and Owusu-Bempah's papers on agroforestry of the importance of women's knowledge about traditional resource management practices.
Part III presents five case studies from Asia and the Middle East. Chapter 16 by Rita Gallin and Anne Ferguson and Chapter 17 by Jane Gleason present case studies from Taiwan. Gallin and Ferguson use longitudinal data from one village to show that a limited focus on farming activities ignored off-farm work and failed to analyze the interactions between the agricultural and industrial sectors of society. The authors propose the term "household enterprise" as a way of dealing with interrelated farming and off-farm work, and encourage researchers to view off-farm activities as "central rather than tangential to FSR/E analysis." Gallin and Ferguson discuss farm mechanization and note that it did not displace women, but rather concentrated certain tasks among some male specialists and caused some women to assume managerial positions previously restricted to men. Older women also took over tasks of younger women who then sought off-farm employment.
Gleason's detailed labor study in Southern Taiwan
complements the previous work, and together they present a good example of why generalizations about gender and
agriculture should not be made for an entire country. Gleason argues that in Southern Taiwan, as agricultural mechanization increased, more women than men were displaced and forced to other sectors of the economy. In her study, availability of female labor increased the variety of crops grown and the level of diversification, indicating that women will be the users of modern vegetable technology and will be most affected by changes in vegetable production.
Chapter 18 by Bahnisikha Ghosh and Sudhin Mukhopadhyay
studies time allocation by men and women in a rice-based ,- arming system in West Bengal, India. Though female labor
is subject to sociological confitanCs, the contribution of women is often larger than men's. A change-tonew rice
technb-Logy-incrxe 0tfem-ale-labor, however the authors show
that the increased workload falls within the home production sector and was largely unaccounted for in the traditional economic literature.
Chapters 19 and 20 are based on work conducted at two
of the International Agricultural Research Centers. Thelma Paris, in Chapter 19, describes how women were successfully
integrated into a crop-livestock project of the International Rice Research Institute (IRRI), in the Philippines.
Beginning as observers and slowly integrating themselves by collecting and disaggregating data on household and production activities, the section members of the Women in Rice
Farming Systems were able to produce useful information and
become fully participatory members of the project. As a
result of their efforts, the whole team began to recognize that specific production activities are the responsibility
of women and that on-farm research needed to target them.
Women's livestock, particularly swine, rootcrops, and
vegetables that had not been previously addressed by the
project, were proposed as new areas of research as a result
of the incorporation of women's concerns. In addition,
subsequent training courses included a significant number
of women participants, and the course addressed women's
~Andree Rassam and Dennis Tully in Chapter 20, discuss
research at the International Centre for Agricultural
Research in Dry Areas (ICARDA) on gender and agricultural
labor in Syria. They find that though male and female time
contributions to crop production are similar, males are
more often involved in new technologies, especially mechanization, while females are more involved in more traditional ones such as hand labor. The authors expect continuing
mechanization to further reduce female agricultural activities and they propose additional research to determine the impact of these changes.
Part IV contains a number of case studies from Africa that includes the descriptions of cropping systems, labor patterns, and work in on-farm, farmer-managed trials. Some themes that emerge from the papers are: the separate economies of men and women within households; the variability of the sexual division of labor in farm tasks; and increasing numbers of households headed by women and the concomitant increase in work burden due to male migration and divorce. Concerning the sexual division of labor, tasks may be the same or different for both sexes; in
female-headed households the so-called "male tasks" are performed by women out of necessity. Concomitantly, the authors report that researchers and extensionists have failed to recognize women's roles in farming, ignored gender in the design of FSR/E projects, and not included women much as trial cooperators. Women, especially female heads of households, are often low-resource farmers who may have special problems that research and extension need to address. (Indeed, there are male low-resource farmers who have many of the same constraints.) The question as to whether or not gender accounts for separate recommendation domains finds different answers in the papers due to differential ecologies, social organization, and cropping systems.
Chapter 21 by Margaret Norem, Sandra Russo, Marie
Sambou, and Melanie Marlett provides an example from The Gambia of how a women's component was formulated and operationalized as part of a larger, FSR/E project. Existing women's societies were used as a basis for organizing women and a maize-cowpea intercropping package was developed. The women experienced difficulties with the package related to pests, seed varieties, and labor patterns. The project was able to use women's participation in the trials and the problems they encountered to argue for the need to include women in subsequent research and extension efforts.
Jeane Henn in chapter 22, examines how government policies and environmental constraints impact on intrahousehold dynamics in Cameroon. Labor patterns, proximity to roads and urban areas, and farm gate prices affected incomes, differently in two villages. Food sales were very important to women's incomes but marginal to men's incomes that were mostly derived from cash crops. In one village, men withdrew labor from food crops resulting in an increase in women's work. However, the women close to roads and
urban markets were able to increase their labor output and foodstuffs produced, and double their income while women in the other village were not.
Jean Due provides data in Chapter 23 from Tanzania,
Zambia, and Malawi on how gender is important to FSR/E work arguing that unless there are person(s) on the FSR/E team who are sensitive to the issues, important information will be missed. In Tanzania, a diagnostic survey for bean/cowpea research revealed that women select seed and contribute more labor on the crop than men. FSR/E work in Zambia would be hindered by not knowing the extensive labor contribution of women on the one hand, and the extent of offfarm and non-crop income, some of which is generated by women, on the other. Data on extension agents contacts with farmers in Tanzania and farmers, farm income are correlated and show that male contact farmers have seven times the income of female heads of households, who are rarely contacted by extensionists.
,In Chapter 24, Timothy Mtoi discusses labor patterns in one region of Tanzania and uses a model to test the significance of female labor on expected risk and productivity under two cropping systems connected to a FSR project. The analysis shows that farm income would increase if the new alternative farming system had female labor transferred to it. However, policy decisions affect whether or not women can participate in the new technology (i.e., be trial cooperators) and obtain extension advice on the packages.
--Chapters 25, 26, and 27 focus on Zambia and provide a more detailed set on the farming systems there and on the results of FSR/E diagnostic surveys. Mary Tembo and Elizabeth Chola Phiri discuss the traditional chitemene system of shifting agriculture and its sexual division of labor, the results of the colonial period that drained off male labor, and the lack of extension credit services to women. The result of these conditions has affected the diet and nutritional status of the population because women have taken to growing cassava (a crop that is less labor intensive, but also less nutritious than millet or maize), and farmers neglect food crops for household consumption because of growing cash crops.
Chapters by Robert Hudgens and Alistar Sutherland
examine FSR/E diagnostic survey work of Adaptive Research Planning Teams that became sensitized to the need to ta 'rget women farmers in research activities and in the determination of recommendation domains. Hudgens details the charactaristics of male and female-headed households in terms of land holdings, draft power, source of inputs, and cash
sales. The comparisons show that there are both similarities and differences between the two household types and that the female-headed households experience labor shortages. Women tend to be isolated from government services and their production is constrained by lack of exposure to new ideas, inputs, and capital.
As part of the diagnostic phase of FSR work, Sutherland compares women's and men's roles in three regions of Zambia and argues that even within one country, gender roles are influenced by cultural, economic, political, and ecological factors. Labor, cash availability, and draft power tend to divide households into recommendation domains. Gender is a distinguishing factor in one region, but not in the others.
In Chapter 28, Anita Spring reports on two different on-farm, farmer-managed trials in Malawi. In the first, the inclusion of low-resource female farmers along with high-resource male farmers generated two recommendation domains. Improved maize cultivars and use of fertilizer worked well in the better environments and with high resource farmers, but would be disastrous in the lowresource environments and with low-resource farmers. on the other hand, the traditional cultivar was better in -\ these situations. The second set of trials involved all
female cooperators to solve a technical problem of inocul-\ ating soybeans as well as the issue of whether or not male\ extensionists could work with female farmers. It was found that the women could do trials with precision and that male extension and research staff could work with women farmers in terms of training and credit programs.
Kofi Owusu-Bempah, in the final paper, argues for inclusion of farmers in the planning and design of projects, and particularly calls for the involvement of women in the selection of species to be included in agroforestry projects in Ghana. His work represents a largely private sector effort and, like Rocheleau, calls for expanding the framework of analysis to include the landscape perspective and the intersection of crops, livestock and forest enterprizes.
In conclusion, the papers in this volume contribute to an understanding of how gender affects farming systems and the way that FSR/E operates. The papers demonstrate that by linking the two codes - gender and FSR/E - the agricultural research and extension system can become more efficient and effective in dealing with different groups of farmers. The papers provide details of specific cases and the methods used to incorporate gender perspectives and analysis.
There is no single recipe for action. Instead, these studies from an array of ecological, social, and political contexts demonstrate that it is both possible and practical to use gender analysis as a tool in the work of agricultural development.
1970 Women's Role in Economic Development. New York:
St. Martin's Press.
1981 Household and Community in African Studies.
African Studies Review 24: 2/3.
Overholt, C., M. Anderson, K. Cloud, and J. Austin
1985 Gender Roles in Development Projects: A
Casebook. W. Hartford, CN: Kumarian Press. Palmer, I.
1985 The Impact of Male out-migration on Women in
Farming. W. Hartford, CM: Kumarian Press. Schmink, M.
1984 Household Economic Strategies: Review and
Research Agenda. Latin American Research Review
Integration of Intra-Household
Dynamics into Farming Systems
Research and Extension:
A Survey of Existing Projects
Rosalie Huisinga Norem
This paper reports the results of a survey designed to study farming systems projects that include an intrahousehold focus in data collection, design, or implementation. Farming systems models (Shaner et al. 1982) have recognized the importance of the household as a component of the farming system, but until recently little has been done to systematically "open the black box" of the household component in those systems models. Projects responding to the survey being reported here are among those
attempting to gain a more systematic understanding of the inter and intra-household factors influencing farming systems.
The primary purposes of the present survey are to assess the types of information collected and used by projects, to evaluate the methods used for obtaining the information, and to gain some insight into how and why the intra-household information is helpful. In addition, project managers were asked to identity types of information they would like to have, but are not available and the constraints affecting various phases of their projects.
This study evolved from on-going work relating household concerns to farming systems work. When the Farming Systems Support Project (University of Florida/JSAID) was first initiated, a family task force was organized to focus on the integration of household and family concerns into the farming systems perspective. one of the recommendations of this group was to develop case studies and training materials that would promote such integration. In a position paper on "Intra-household Dynamics in Farming Systems Research: The Basis of Whole Farm Monitoring of
Farming Systems Research and Extension," Cornelia Butler-Flora (1984) set the stage for an intra-household dynamics and farming systems case studies project that was subsequently funded and implemented.
Concurrent with the effort to develop training materials that will sensitize people to the importance of intrahousehold factors in farming systems work as well as in other development efforts, there is a need for more knowledge about the kinds of data being collected and the methods used by existing projects that are attempting to focus on intra-household factors. As an attempt is made to understand the complexity of household dynamics, ways to obtain and analyze information within reasonable time and other resource limits also must be found. Questions about how much information on aspects of the household should be obtained from whom have yet to be answered (Norem 1983).
This is not to suggest that one "right" way of focusing on household dynamics and farming systems can be identified. Rather it is to suggest that by examining what is being done and how effective researchers and practitioners involved find current efforts, some guidelines can be identified that will be helpful in future planning.
It may be helpful to consider how information relating to households can be broken into different units depending on the purposes of the individual study. overall, the unit of interest in intra-household dynamics is the household. The unit of data collection might be one or more household members, other informants, or other existing information. The unit of analysis can be an individual, the household or subsystems thereof, a work group, or the farming system, among other possibilities. Designing parsimonious data collection and analysis procedures requires an understanding of how these units relate in various situations.
For example, it may be possible to obtain good data on the unit of interest from only one person if what is required is basic demographic information such as age, gender, and education of household members. The household is also the unit of analysis in this example. Information about the tasks performed in the household is more likely to require data collection from more than one person, or extensive observation or record keeping to permit the collection of enough information to focus on the household as a unit of analysis. As a clearer picture of the state of the art develops, it is hoped a clearer set of guidelines will evolve. The survey summary presented in this paper is a first step.
DESIGN OF THE STUDY
The Farming Systems Support Project and Population
Council Intra-household Dynamics and Farming Systems Case Studies Project was initiated with support from USAID and the Ford Foundation in late 1984. In February of 1985, a request for expressions of interest was sent to projects and individuals on a variety of international mailing lists. over seventy-five expressions of interest were received in response to the request. These expressions of interest were used to develop initial lists of types of data and data collection methods being used in projects. The lists were used in conjunction with the case studies project conceptual framework to draft a survey questionnaire. The questionnaire was reviewed by the case studies project advisory committee and revised using the comittee's suggestion.
The questionnaire was mailed to all projects that had responded to the original request for expressions of interest in the case study project, since those projects had self-selected themselves in terms of interest in intrahousehold concerns. A few other projects were also included in the survey. All questionnaires received (n--19) are included in the summary, regardless of type of project. most are farming systems oriented, with one project being specifically focused on women in farming systems. Seven projects from Asia, seven from Africa, two from the Middle East, and three from Latin America are included. The titles and identifying information about the projects are presented in Table 2.1.
Each project has a different specific target group, but all projects have target groups of farms with multiple crop systems. Fifteen projects report farms in their project also have livestock, most for multiple use, including cash income, food, traction, wealth, and prestige. The average land holdings for farmers in the projects ranged from .89 hectares to 30 hectares, with an overall mean of 6.34 hectares.
Types of Intra-household Data
Each project was asked to indicate whether or not it has data about five general categories: (1) demographic information, (2) household members' participation in activities, (3) household members' access to production resources, (4) household members' participation in decision making, and (5) income and expenditure data, benefits from
PROJECTS RESPONDING TO SURVEY
PR~wv .r TTThFR
India Philippines Philippines
CONTRACTOR AND UNIT IN CHARGE SOURCEE OF FUNDS)
University of Hawaii with University of N. Carolina & Center for Soils Research (USAID)
Ministry of Agriculture and Food and the Virginia State University (USAID)
Winrock International & Ministry of Agriculture (USAID)
Bangladesh Agricultural University (Bangladesh Agricultural Research Council)
Haryana Agricultural University (Haryana)
Silliman University Research Center (Ford Foundation)
Cornell University, Ministry of Agriculture & Food, & the Visayas State College of Agriculture (USAID)
Soil Management CRSP
Farming Systems Development Project Eastern Visayas Non-Farm & Resource Management Institute
Agricultural Research & Production Project
Farming System Development Division
Women in Farming Systems
Role of Farm Women in Decision
Making Related to Farm Business
Farming Systems Development Project Eastern Visayas Non - Farm & Resource Management Institute
PROJECT TITLE CONTRACMR AND UNIT IN CHARGE (SOURCE OF FUNDS)
Burkina Faso Burkina Faso Sierra Leone Extension Ghana Botswana Kenya Sierre Leone
Fulbe Agropastoral Production
in Southern Burkina Faso for USAID Agric. Sector Grant
Income & Agric. Investment in a Bobo Village
Adaptive Crop Research of Sierre Leone
Agricultural Technology Improvement Project
Dryland Farming Research
Phase I: Land Resources Survey Project Phase II: Strengthening Land & Water Division
Frederick Sowers, Univ. of California, Berkeley SAIDI)
University of Illinois, (Wenner Gren Foundation, NSF)
Southern Ill. University, Louisiana State University & Ministry of Agriculture & Natural Resources (USAID, and Government of Sierra Leon) Bureau of Integrated Rural Development (BIRD, REDECASH)
Midwest Int'l Agricultural Consortium (MIAC) (USAID, Kansas State University)
Ministry of Agric. National Dryland Farming Research Station, (FAO/UNDP, Kenya Gov't. ) FAO/Ministry of Agriculture (UNDP)
TABLE 2.1 (continued)
COUNTRY MIDDLE EAST
Mexico Honduras Panama
Irrigation Innovation and Family Farming Strategies in Israel
Syrian Households: Women's Labor & Impact of Technologies
Livestock Production Systems in Central State of Veracruz
Honduras Agricultural Research Project
EMPARAD: Socio-economic (Agricultural Development Bank) Case Study
CONTRACTOR AND UNIT IN CHARGE (SOURCE OF FUNDS)
City Univ. of New York, (Faculty Research Grant)
Andree Rassam, (ICARDA & MEAWARDS)
Centro de Investigacion Ensenanza en Granaderia Tropical, CIEEGT, Facultad de Medicina & Zootechnia (Universidad Nacional Autonomo de Mexico, UNAM) Consortium for International Development, New Mexico State Univ. (USAID)
Miriam Reyes, Martinez Elba, Rosa Hernandes
farm production, food consumption and nutrition information. Each of these categories include several specific kinds of data. Table 2.2 shows the types of data collected and the number of projects that collected each type. Table 2.3 summarizes the ways each type of information was or is used by the projects that responded to the survey.
TYPES OF INTRA-HOUSEHOLD DATA COLLECTED BY
PROJECTS RESPONDING TO SURVEY
Type of Information (N=19) Number of projects
Household Structure, Membership, and Size 19
Ethnic Identity 16
Migration Patterns 9
Variation in Household Structure Over the Life Cycle 7 Household members' Participation in Activities
Cash Crops by Crop 13
Subsistence Crops by Crop 12
Livestock Production 11
Other Primary Income Generating Activities 8
Major Tasks of Household Reproduction 11
Household Member's Access to Production Resources
In General 13
By Tenure Category 9
By Production Potential (Irrigated, Non-Irrigated) 8
Type of Information (N=19) Number of projects
Household members' Access to Production Resources
innovations or Improved Production Inputs
Information (Extension Contacts, Training, Etc.) 14 Technology Inputs Requiring Cash or Credit 10
Household member's Participation in Decision-making
Land use 13
Use of Family Labor 13
Use of Hired Labor 11
Use of Exchange Labor 9
Use of Technology inputs 14
Use of Credit 12
Cropping Choices 13
Cultivation Practices 13
Uses of Harvested Crop and Residue 13
Income and Expenditure Data
Each Household member's Sources of Income 7
Each Household Member's Expenditures 7
Benefits frm Farm Production
Use of End Products From Crop Production 11
Desirable Characteristics of Each Crop
or Crop Product 8
Each Household Member's Access to or
Control of End Products 6
Food Consumption and Nutrition Information
Diet Survey 4Nutritional Adequacy Analysis 4
Food Preparation Practices 6
Food Preferences 6
On-FaM Household Food Production 7
TABLE 2.2 (cont'd.)
TYPES OF INTRA-HOUSEHOLD DATA COLLECTED BY
PROJECTS RESPONDING TO SURVEY
Number of Projects Reporting Use of Information by Type of Information*
Use 1 2 3 4 5 6
Initial Project Design 11 5 5 3 3
Selection of a Target Group 11 4 8 3 4 Identification of
Recommendation Domains 4 5 5 1 4
Choice of Research Topic 7 6 8 3 6
Designing Trials 6 4 6 3 4
Selection of Participating
Farmers for Field Trials 6 5 6 2 4 3 Evaluation of Field Trials 2 3 6 2 4 2 Redesign of Trials 3 6 6 6 4 2
Technology Recommendations 8 5 5 4 6 2 Extension Efforts 7 4 5 6 4
Project Evaluation Design 5 2 4 1 3 1
Assessing Time and Labor
Constraints 13 11 7 8 1 1
Assessing Opportunity Costs
for Innovation 7 5 5 3 3
Note: *Type of Information:
1 Demographic Information
2 Household members, Participation in Activities
3 Household Members, Access to Production Resources
4 Household Members' Participation in Decision-making
5 Income and Expenditure Data, Benefits from Farm
Production, Food Consumption, and Nutrition
6 other kinds of information collected include:
religious affiliation, inheritance data, and
information gathered from husband and wife
USES OF TYPES OF INTRA-HOUSEHOLD DATA
BY PROJECTS RESPONDING TO SURVEY
The most frequently used methods for obtaining demographic information are pre-existing national surveys, formal project surveys, participant observation, and sondeos (rapid reconnaissance surveys). This information is summarized in Table 2.4 for all types of data. Demographic information is also available through pre-existing anthropological studies and local village records for some projects. other projects collected information through farmer records, community informants, time allocation studies, team members' personal knowledge, and in-depth case studies.
Nine projects collected demographic data before the project began, five during the diagnosis stage, and seven parallel with on-farm testing. Ten projects collected demographic data on an on-going basis.
All projects collected demographic data on household structure, membership, and size. most also have information about education and ethnic identity. Nine projects have data on migration patterns and seven have them on variation in household structure over the life cycle. Table 2.3 shows that demographic information was most useful in the early planning stages of projects.
Respondents were asked to identify the most helpful
information for each type and to give an explanation of how or why the information was helpful to their project. They were also asked to indicate any information in each category that they did not have but wished they did. These open-ended questions provide more detail related to intrahousehold concerns than the tabulated results shown in the tables.
The specific demographic information identified as most helpful to a project varies by project, however, some generalizations can be made. Gender and age structure of the household is mentioned by several respondents, sometimes alone and sometimes in conjunction with other information such as labor and income. This information was useful in identifying target groups and in designing trials that ascertain labor bottlenecks and total household activity patterns. Household structure is also reported as an important consideration in designing extension efforts. Ethnic information is mentioned second most frequently as the most helpful demographic information, because farming practices and values about female participation vary according to ethnicity.
TABLE 2.4 MOST F M GENTLY USED METHODS OF DATA COLLECTION
BY TYPE OF DATA
TvDe of Data Data Collection Mefhm
The two kinds of demographic information least often available, migration patterns and variation in household structure over the life cycle, are the ones most frequently named in response to the question, "Are there demographic data you do not have that you wish you had collected?."
Household Members' Participation in Activities
Formal surveys, participant observation, and community informants are the methods most frequently used to obtain information about the participation of household members in various activities (see Table 2.4). As with demographic information, projects obtained this information in a variety of ways. Each of the methods listed above for demographic information is used by at least one project to
National Surveys Formal Survey Participant observation Sondeo
Formal Survey Participant observation Community Informants
National Surveys Formal Surveys Participant observation
Formal Survey Team Member's Personal Knowledge; Participant Observation
Formal Surveys Participant Observation
Household Members, Access to
Household Members, Participation
Income and Expenditure Data, Benefits from Farm Production, Food Consumption, and Nutrition
obtain activity data, group meetings provide an additional source of information about household members' activities.
Three projects collected activity data before the project began, six during initial diagnosis, and six parallel with on-farm trials. Nine projects collect activity data on an on-going basis.
Specific questions were asked about type of activity data collected and method of disaggregation. Twelve projects collected task assignment data disaggregated by gender and age. Nine projects disaggregated by position in the household as well. Five projects have information about time allocation.
Thirteen projects report collecting some information about the participation of household members in various activities. Most frequently collected information (n-13) is about activities related to production of cash crops, with subsistence crops and livestock production information available for twelve out of nineteen projects. Other activities within the household receive less attention as indicated in Table 2.2. Table 2.3 indicated that projects use activity data less often than demographic data. Activity data are used most often to assess time and labor constraints.
Information on household members' activities was most helpful in designing research and targeting interventions. Respondents would like more detailed information about nonproduction activities and several express a desire for activity data covering a period of time up to a year. The complexity of activity data is pointed out and difficulties with processing such data are mentioned.
Household Members, Access to Production Resources
This study breaks production resources into subcategories of land, labor, capital, innovations, and credit. The projects represented use a variety of methods to obtain resource information; the most frequent are pre-existing national surveys, project-conducted formal surveys, participant observation, and team members, personal knowledge.
Six projects collected data on access to resources
before the project began, seven during initial diagnosis, and seven during on-farm trials. Five projects collect these data on an on-going basis.
As indicated in Table 2.2, fourteen of the seventeen projects have some resource information. But examination of Table 2.3 suggests that fewer projects overall use this information than use demographic data. more projects use resource information than activity data for most purposes,
but more projects use activity data to assess time and labor constraints
The answers t questions about the most useful resource data and why and how it is useful indicate that land resource information is perceived as most helpful for more projects than other kinds of resource data. The responses also indicate that the usefulness of resource access data is very project specific. Data on access to resources is likely to be helpful in research design and in selection of field trial locations. There is a pattern among responses about the kind of resource information respondents would like to have had but that was not available. More information about monetary income, including gifts and remittances, is mentioned in several contexts, including credit, opportunity costs for innovations, and access to capital.
Household members, Participation in Decision-Making
Fourteen projects in the survey have some data about decision-making within households. These data are collected most frequently through formal surveys, team members' personal knowledge, and participant observation. other methods are used, but in a project-specific manner. only three projects report having decision-making data to use in initial project design. one project collected data during the initial diagnosis and four collected data parallel with on-farm trials. Six projects are continuing to collect decision-making data.
Table 2.2 indicates that projects which have decisionmaking data also have information about most of the subcategories identified, including land use, labor use, technology use, cropping and cultivation practices, and use of production outputs. Table 2.3 suggests that projects are not using decision-making data extensively. Eight projects use decision-making data to assess time and labor constraints, and this is the most frequent use reported.
Responses to open-ended questions about the usefulness of decision-making data express the theme that a better understanding of household dynamics permits more knowledgeable identification of target groups. Eight respondents indicate their projects could use more detailed decisionmaking data that would reveal more about the effect of decision patterns.
Income and Ejc4nditure Data, Benefits From Production, and food Consumption and NutritioF Information
Seven projects have information about these categories of data. Formal surveys and participant observation are
the most common methods of obtaining the information. There are some differences among he categories. Participant observation is most likely to be the source of information about food consumption and nutrition information, and is not as likely to be a source of production benefits data. Formal surveys are used by several projects for all three categories.
Only one project had data in these categories before the project began. Three projects collected the data during the initial diagnosis, four collected the data parallel with on-fam testing, and eight collect the information on an on-going basis.
Table 2.2 details that eleven projects have information in at least one of these three categories. Income and expenditure data are least frequently available. Table 2.3 shows a fairly equal distribution of the use of specific kinds of data available in these categories over the various phases of the projects, especially in the design and implementation of field trials.
Since this section of the survey combines three
categories, the answers to questions about which information was most helpful, why so, and how, are somewhat complex, but they also point out the need to integrate information about overall production and consumption patterns in the household. For example, respondents noted the importance of looking at off-fam income and cash income from food crops and of understanding the households' reliance on local markets both for food and income. They also mention the need to assess the opportunity costs of innovations based on total inputs and total income generating possibilities.
Respondents answering the question about information they wish they had collected, mention primarily better income data for individual household members monitored over time. Several note the difficulty of obtaining reliable income data and state the importance of finding better ways of obtaining such information.
Most Effective methodologies
Respondents were asked to select the study or activity of their project that was most effective in collecting information about intra-/inter-household variables relevant to farm production and the ones that were most useful in determining project decisions concerning research priorities, cooperating farmers, technology acceptance, etc. Nine respondents name the formal survey as most helpful. This is usually done at the beginning of the project. Eight respondents identify participant observation as the
most useful activity for obtaining household information. This tends to be on-going, Four respondents state that the sondeo as most useful. The sondeo took place anywhere from the beginning to the third year of the project. Ten respondents report the head of household as the primary informant, whether male or female. Seven projects tried to include at least one other adult household member. Three relied on whoever was at home with a preference for the head of household. one case study involved all members of the household.
Constraints to Project
Respondents were asked to identify constraints that
affected the study design, sample selection, conduct of the study, data analysis, or applications of the data to their project or activity. These responses are summarized in Table 2.5. Ten respondents report physical, logistical or resource constraints on sample selection for their projects. Detailed answers to open-ended questions about these constraints show the most common one is transportation, either because of the lack of means or because of the difficulties of the terrain.
CONSTRAINTS INFLUENCING PROJECTS
Number of Projects Reporting Physical/ Cultural/
Phase of Project (N=19) Resource Political
Study Design 9 5
Sample Selection 11 4
Conduct of Study/Activity 7 6
Data Analysis 7 1
Application of Data to
Project/Activity 3 1
In order of descending frequency, other constraints
mentioned are funds, language, personnel, political situations, and ethnic group considerations. In many instances the constraints are named in conjunction with one another, such as ethnic concerns and language difficulties.
Several summary points can be made from the information provided in the survey. First, there is a wide variation in the kind of data being collected about households with a comon focus on the household as a unit of interest. The data are most often collected from heads of household, so for some kinds of data there may be difficulty in using the household as a unit of analysis. For example, data about decision-making describe a dynamic intra-household process, but data involving several household members require complex data collection procedures. It is important to examine alternatives in determining which information is important for which stage of a project and how to obtain it as efficiently as possible. one respondent pointed out that designing more standardized methods of data collection and analysis is difficult because of the unique aspects of each project, but the same respondent also emphasized that efforts to move in this direction will save significant resources and hopefully eliminate the need for each future project to make the same mistakes.
There was general agreement among respondents that more information about intra-household concerns was important. one project is collecting additional information in most of the categories in the final stages of the project with the hope that it can be used in future planning.
This survey gives us one measure of what projects are currently doing and assesses perceptions of the importance of intra-household data. The next step is to suggest a household minimum data set that projects could collect. This would not eliminate the need for specific data for specific projects, but would provide the potential for comparative analysis and the beginning of theoretical generalization.
Flora, C. B.
1984 Intra-HoUsehold Dynamics in Farming Systems
Research: The Basis of Whole Farm Monitoring of
Farming Systems Research and Extension. A Position
Paper. Manhattan, KS: Department of Sociology,
Kansas State University. Norem, R. H.
1983 The Integration of a Family Systems Perspective
into Farming Systems Projects. Conference
Proceedings, Family Systems and Farming Systems
Conference. Blacksburg, VA: Virginia Polytechnic
Institute and State University.
Shaner, W. W., P. F. Philipp and W.R. Schmehl
1982 Farming Systems Research and Development:
Guidelines for Developing Countries. Boulder, CO:
Gender Relations and Technological Change:
The Need for an Integrative Framework
Over the past decade, agricultural research has focused on the rural poor and the need to strengthen the production systems of small farms. In sub-Saharan Africa a substantial proportion of agricultural activity and food production continues to be organized at the household-farm level, and with concern increasing over the region's food producing capabilities, researchers and policy makers have sought to identify ways of raising the productivity and incomes of self-provisioning farming households (Henn 1983; World Food Council 1982).
Farming systems in sub-Saharan Africa are extremely diverse, but a number of shared conditions and characteristics permit some generalizations. First, household members typically comprise the main source of on-farm labor, but households also require large amounts of labor for nonfarm and reproductive tasks. Second, rural women, especially poor rural women, make a substantial contribution to all types of agricultural activity, and in most cases have responsibility for the production, preparation, and distribution of food for the household and for sale, and for the reproduction of family labor (FAO 1984). Thus, if agricultural research and policy is to have a positive ipacft on food prouctio#i-and ecityin sub-Saharah Africa,--reserchrs_.and policy7makers must link new technology and extension advice more directly to the needsandpriorities of smallholder producers-in general and'wome Poduer i particular.'
The record of technological intervention in sub-Saharan Africa has been very variable and the literature contains many examples of the contradictory and unintended effects of technology change for traditional farming households (Dey 1984; Bryceson 1984; Agarwal 1981). It is always the poorest and smallest farming households that are most
vulnerable to the "unplanned" effects of "modernization," but it is the effectiveness and well-being of women producers in poor households that appears to be most at risk (Agarwal 1981; Kisseka 1984; Whitehead 1981).
The fact that technological changes have complex, differential, or even perverse effects on the productivity of small farm systems indicates that agricultural research and extension must be reoriented to identify and deliver solutions to on-farm problems that are reliable and predictable within the whole farm system and above all consistent with the needs and interests of both women and men (CGIAR 1985; Hahn 1985; Fresco 1985; Maxwell 1984).
AGRICULTURAL RESEARCH AND DEVELOPMENT
The driving force behind recent research and development in small farm agriculture has been Farming Systems Research and Extension (FSR/E). Although FSR/E does not follow one model it does tend to pursue a single aim: to develop technology that helps small farmers improve output (yield) and increase marketed food supplies while protecting the natural environment (FSSP 1984). A major advantage distinguishing FSR/E from earlier research and development approaches is its potential for integrating comprehensive, interdisciplinary knowledge about small farmers' technical needs with the wider systemic implications of technological change (Shaner et al. 1982).
The FSR/E approach tries to identify the distinctive components and parameters of the "total" farm system in which a small farm-household functions. In most FSR/E models, the family labor unit or household is at the center of the conceptual framework and the farming system is defined as:
a unique and reasonably stable arrangement of farming
enterprises that the household manages according to well defined practices in response to the physical,
biological and social environments and in accordance
with the household's goals, preferences and resources
(Shaner et al. 1982).
In their functionalist approach to farm systems analysis, FSR/E practitioners try to understand how factors
-under the control of household members (end6ngeous-factors) like family labor and the techniques obf-p-r6du-ctio-njint!eract witphysical,, biological, and so-cioeconomicf actors beyond household-control (exogenous factors) ilkie cli-mate, prices, and the availablity of credit, affect the agricultural welfare of the farm household. By combining
knowledge of the whole farm system with extensive on-farm research and experimentation, systems researchers should be able to recommend technological alternatives that meet the priority needs and preferences of small farmers and evaluate which technologies fit where and why, and monitor who benefits.
In practice, the FSR/E approach is used primarily to
reduce the degree of error in finding the appropriate technical "fix" for market-based problems of agricultural production. Thus, FSR/E practitioners have restricted themselves to finding technological solutions to a limited range of crop and livestock production problems, using mainly market criteria to measure the costs and benefits to farm-households. The methodology used to identify farming households' needs has been implicitly restricted and the choice of recommendations has been limited to suit a narrowly defined, technocratic approach to on-farm development. FSR/E practitioners have not yet built a broad understanding of the diverse needs and preferences of small farms or of the differential effects of technological change for members of farm-households. The FSR/E approach is particularly limiting when considering gender issues in agricultural development. In particular, FSR/E is ill equipped to:
(1) understand the distribution of costs and benefits of
alternative technological choices and extension projects between men and women within farm households and
under what conditions technological change benefits men
but not women, or in fact harms women, individually or
as a group or
(2) identify and meet the technical needs of women as
farmers, particularly as producers and consumers of
FSR/E has not accomplished what it set out to achieve at the-s6-ci-economic level.--Researchers- have -failed to. c(6isd&how_ far rig 'systems are consti-tuted alongeider h-1i-iriad to analyze the influence that gender differenifati'on-haiin shaping technological and economic ouEc6miii-s"in agricultur~al- develomet Th methoUidology contains procedures and assumptions that have tended to reinforce gender blindness.
First, FSR/E uses market or monetary criteria to identify productive components in the farm syte and toNi measure their contribution to household-income-and welfare. Manf-of the value-adding or service activities that are not directly comparable in monetary terms are left out or are considered costless in the economic equation. For example,
FSR/E does not value non-monetized, household-based activities such as food production (especially secondary foods),
processing and preparation for direct consumption, fuel and
water collection, and house maintenance and repair.
From micro-level research, it is shown that preferences
or market criteria have important gender implications
because while men, women, and children all perform some of these non-market tasks, it is women who contribute most of
the labor, energy, and resources (as well as exchanging
labor and resources with other households) to get this work
ttdone. By treating non-market, home-based work as
complementary to the farm unit rather than an essential component of it (Fernandez 1985; Lele 1985), production economists have neglected important links between market
and non-market enterprises. For example, the linkages
ignored include those between water and fuel collection,
that in sub-Saharan Africa is almost always non-monetized,
with crop and livestock production, food production, and
food processing. More importantly, economists have
neglected the significance of gender differentiation, the
division of labor, and resources by gender in shaping
relations between food and non-food production, market and
household-based enterprises, and consequent responses to
economic incentives and new technology.
Second, production economists and technicians have
concentrated on ways of improving-yields_ and-the:evolie-of
marketed output to increase household incomeand welfare.
In this-tprocess-they te-nd to focus on specific sub-stems
like the cropping system, and within that, particular crops with obvious market potential. This selective approach has
often been pursued without carefW~cnsideration of-hepssibilt that-production constraints stem from other
Parts of te farming systemn or that fair e6191Thb-making has 'to -do with- the-6operaioo the7 farminq-systemal
whole anfd niot w4ith isoated-par ts of i-t.
This implies that planned changes within the cropping 1lsystem, for example, that should increase aggregate output rpdand income, may have unanticipated and differential effects
;on individual household members' production. The welfare
effects of planned changes can be predicted only if the
full extent of household members' involvement in the farming system is known, and more importantly, if it is known
how the activities of men, women, and children fit together
within the prevailing technical and social relations of
Systems researchers have often overlooked appropriate
indirect or non-farm solutions to farm-based problems. For
instance, farm output may increase if the productivity of
household-based activities, many of which are women's
responsibility, were to improve. Higher productivity in these tasks may have a positive impact on the amount or quality of labor available for market-oriented production. It has been demonstrated empirically that women's household-based activities will feed directlyinto-the satisfactio-n 6f household basic needs whereas the link between market-based activities and income and household welfare is more spurious (Hart and Popkin 1975). If women can colle-t-ibe-r 6r-fuel more easily, they could expend more time and energy in directly productive activities. Or by raising the productivity of subsistence food crop production and processing women could improve the health and efficiency of labor, especially during the low energy or "hungry" seasons.
Third, the FSR/E approach is based on a stereotyped model of "the household" as a homogeneous unit of production and consumption that is at variance with the reality in many rural areas.
Furthermore, it is assumed that the distributional effect of technological change on farm household members can be read off from shifts in the demand curve for family labor within a specific sub-system. Technical_change, however, does not effect the division of labor and /
resources in any stea ic-way (Whitehed -e98TL The relations between-women ahd-men in farming households are not solidifiedwithinthe technical division of labor nor are their interests and-needs necessarily complementary. Households are nothomogeneous units and the needs and preferences-of women_and men are not easily definecf with respe ct - to, a -singl e household produ-ction -function. Even more importantly, the household decision-making process can be only partially explained by economic factors. The choices of men and women in the household and beyond are the outcome of an intricate bargaining process mediated by normative relations of power and control.
In many West African countries, men and women occupy different positions within the farm economy, not only in terms of task allocation, but also in terms of their different access to factors of production - inputs, credit, others' labor time, off-farm employment opportunities, and markets. In polygynous households, senior wives also have structurally different economic positions from junior wives. In a number of decisions, men and women have a joint economic interest; in many others, often those associated with direct welfare activities, their interests conflict or their economic and technical needs are different. For example in Sierra Leone, women's interests in improving swamp rice farming and intercropping to produce a marketable surplus have been in conflict with
men's interests in upland farming and their use of work gangs and household labor in the production of market crops. Polygyny also compels wives to have independent access to cash income that involves management of non-farm income generating enterprises such as palm oil processing, soap and pottery making, and dyeing cloth. These enterprises are kept separate from men's and require women to maintain separate budgets and resources even when living with men in the same extended household (Richards et al. 1981).
THE SEARCH FOR A NEW FRAMEWORK
Most FSR/E models take a selective and comparatively static view of the farming system. There is a need for more searching concepts and methods to look at gender in farm-household systems, to understand the gender distribution of costs and benefits of technological choices, and to identify women's technical needs. Unfortunately, due to a lack of empirical evidence this paper can only offer some analytical suggestions on how such a framework might be built.
First, a conceptual model that treats the small farming unit as an interlocking system of market production, subsistence, and reproductive activities performed at farm and household levels should be developed. Reproductive activities include primarily the domestic services necessary for family survival. Central to this inter-related model are the multi-activity household and the different sets of economic and social conditions that shape the economic participation and welfare of women, men, and children. These sets of conditions are not immutable. Depending on the total matrix of productive and reproductive activities that comprise any one farm-household system, they are likely to alter with seasonal shifts, the life cycle development of the household, and emerging employment and market opportunities in the local economy.
Such a framework should be able to generate data on the key sets of conditions that shape the differential participation of household members and give explicit attention to the most important interactions between enterprises that vary by gender.
The anthropological literature for sub-Saharan Africa suggests that data needed center on intrahousehold resource allocation and on the relative weights of market, nonmarket, social, and economic forces in shaping the
differential distribution of women's and men's labor, resources, management, and control within the total farmhousehold system. Such information could be used to investigate the possibility for productivity gains in nonmonetarized enterprises that have either a direct or an indirect productive impact on household members. Agricultural researchers and extension personnel also need such information to understand why planned changes in the technological and economic balance of farming systems can have unanticipated and sometimes detrimental effects on other enterprises outside of the production arena.
In Africa, a commonly cited problem for women farmers, particularly in the poorest households (Dey 1984; Guyer 1980; Whitehead 1981), is that technological changes have intensified women's workloads without adequate compensation or have eroded their access to land and non-farm income. without understanding exactly how women's multiple activities interlock with each other and with other enterprises in the-system, neither the aggregate net effect of technological change upon the total workload or returns to labor of individual household members, nor the differential productivity and welfare effects on women and men can be established clearly.
A broader conceptual model raises a number of important methodological considerations for farming systems research.
Recommendation domains. while agronomic, topographic, and general socio-economic data will continue to be crucial in selecting recommendation domains, different household forms, family structures and composition must be increasingly incorporated. In Africa, it is essential to consider polygynous households, consensual family structures, female-headed households, and the complex variation in authority and power structures that determine the location of decision-making and access to resources in household units (Allison 1985). At this stage it is important that researchers be aware of the different economic and social positions that women and men occupy in African households and haw these are manifested in different decision-making processes, separate budget and production functions, and different access to inputs and markets, incentives and off-farm employment opportunities.
Problem identification. Information about the different positions of women and men in the farming household must be linked to precise data about patterns of resource allocation, labor utilization, and the multiple fit of women's and men's responsibilities in order to identify
major constraints and bottlenecks as they vary by gender. The degree of flexibility and substitutability of the organization of farm-household activity must be taken into account as well as the potential trade-offs or conflicts that might emerge between the genders when a constraint within one sub-system creates another elsewhere in the system. Further research should focus on:
(1) the degree of flexibility and substitutability between
the role of women's labor in farm production and
household-based production relative to men's;
(2) the substitutability between women's labor and capital
(productivity enhancing technologies) relative to other
household members; and
(3) the flexibility and substitutability between women's
labor and that of other family or household members,
including husbands, children, neighbors, or hired labor
Much of the information that can be gathered about
these points will be qualitative rather than quantitative, but is nonetheless valuable. Researchers and extension personnel should resist the temptation to devalue the importance of qualitative data in planning on-fam research and in evaluating responses by farm-households.
on-Farm Research, Extension, and Evaluation. Gender differences must be an integral criterion woven- into the analysis of technological constraints and the recomnendation of solutions and must also be considered when choosing target farmers with whom to conduct on-fam trials. It is particularly important that researchers recognize the mixed strategies that women employ to meet basic household needs and consider the potential impact that trade-offs or curtailments within these strategies have for individual and household welfare.
It is also important to trace the distribution of male and female labor time and resources within total production cycles. Thus, crop production from cultivation to processing, consumption, or sale should be disaggregated to understand the possible welfare implications of changes within the cycle. For example, changes in cropping patterns may intensify women's unpaid labor during weeding, harvesting, and processing stages and conflict with their labor needs elsewhere in the farm-household system. Despite creating a greater workload, a new crop variety may offer women potential income gains through the processing and sale of crop by-products over which they have marketing control. The dominant effect cannot be established a priori.
Finally, it is necessary to acknowledge that household services play an economic role in maintaining the productive function of the farming system and to identify the non-pecuniary costs borne by individual household members (usually women) in providing such low productivity services under time-consuming and arduous conditions.
If gender issues are to be given explicit attention within agricultural research and technology development, then an alternativ-e analytical framewQk i s required. Such a framework should identify the, different se ts.-of. cinditions. that characerze- theeconQmi c participation. of-women and men in small farm .systems.and, view .the ,,integratedfarming systems as a rendered system of production and repf_ociion. -The purpose of such a framework is not only o examine existing evidence with greater rigor, but also to generate specific hypotheses, data, and methods that break down gender biases within agricultural research and extension. A number of problems that must be addressed to build an alternative framework include:
(1) Overcoming the methodological division between technocratic and social science research that challenges
the integration of quantitative and qualitative methods
(2) Resolving time conflicts, such as the time scale that
agricultural research institutes or donors demand for
FSR/E but that cannot accommodate the amount of time that sociological or socioeconomic research usually
(3) Channelling research funds and personnel into more extensive data collection and methodological approaches.
(Funding for scientific research into minor food crops and technologies for non-farm activities is extremely
Despite these problems, to have a clear understanding of the relationship between gender relations and the economics of farming systems and technological choice it is important to question the conceptual basis of existing methodological approaches in agricultural research and development (see Figure 3.1) and to build alternative frameworks for gathering empirical data and testing hypotheses in which gender relations are an explicit variable.
Extra H.H. Intra H.H.
H.H Farm Components
Raw materials ' I ~oa1
jManagement I _: ISJ._L e ent ter [ --IVEST cash ii
Male labor Fode I
Child labor (lab
-Domestic chores ino
Consumption goods Payments
(food, durables) (interest, tax)
and services. t
NET INCOME _ GROSS IMXM
EXPEDITURES (wages, cash, kind)
Human Icapital I
security I(formal, informal)
KEY: . flows of family labor in/outside the F-H.H. system.
- flows of non-labor inputs & outputs (& payments).
PLEASE NOTE: For clarity, the elements of the system have not been disaggregated along gender and age lines. These variations are captured when the framework is applied to the situation of each family member in turn and then compared.
Acknowlegement: Many of the ideas for this diagram come from C.D. Deere and A. de Janvry "A Conceptual Framework for the Empirical Analysis of Peasants" American Journal of Agricultural Economics. 1979.
A CONCEPTUAL MAP FOR LOOKING AT THE FARM-HOUSEHOLD SYSTEM FROM A GENDER PERSPECTIVE
1981 Agricultural Modernization and Third World
Women: Pointers From the Literature: An Empirical Analysis. WEP Research Working Paper WEP/1O/WP21.
Geneva: International Labor organization. Allison, C.
1985 Women, Land, Labour and Survival: Getting Some
Basic Facts Right. IDS Bulletin 16:3:24-29. Bryceson, D.
1984 Women and Technology in Developing Countries:
Women's Capabilities and Bargaining Positions.
Prepared for INSTRAW, United Nations.
1985 Women and Agricultural Technology. Bellagio,
Italy: Rockefeller Foundation and International
Service for National Agricultural Research. Dey, J.
1984 Women in Food Production and Food Security in
Africa. Women in Agriculture 3. FAQ: Rome. FAQ
1984 Women in Agricultural Production. Women in
1985 Social Components in Peasant Farming Systems: a
Research Proposal. Unpublished. Flora, C. B.
1984 Intra-Household Dynamics in Farming Systems
Research: The Basis of Whole Farm Monitoring of
Farming Systems Research and Extension. A Position
Paper. Manhattan, KS: Department of Sociology,
Kansas State University. Fresco, L.
1985 Food Security and Women: Implications for Agricultural Research. Presented at the International Workshop on Women's Role in Food Self-Sufficiency and Food Security Strategies. Paris: ORSIOM/CIE,
1984 Newsletter 2:3.
1980 Female Farming and the Evolution of Food
Production Patterns amongst the Beti--of-South-Central Cameroon. Africa 50:4:341-356.
Hart, G. and B. Popkin
1985 A Note on the Interdependence between Ecroomic
and Welfare Factors in Rural Filipino Households.
University of the Philippines, Institute of Economnic
Development and Research. Pp. 75-5. Henr, J.
1983 Feeding the Cities and Feeding the Peasants:
What Role for Africa's Women Farmers? World
Development 11:12:1043-1055. Kisseka, N.
1984 Implications of Structural Changes for African
Women's Economic Participation. Unpublished. Lele, Uma
1985 Women and Structural Transformation. Economic
Development and Cultural Change 34:2:192-221. Maxwell, S.
1984 I. Farming Systems Research: Hitting a Moving
Target. II. The Social Scientist in Farming SystexwS
Research. Institute of Development Studies
Discussion Paper 199.
Richards, P., J. Karimu, et al.
1981 Upland and Swamp Land Rice Farming in Sierra
Leone: The Social Context of Technological Change.
Shaner, W.W., P.F. Philipp and W.R. Schmehl
1982 Farming Systems Research and Development
Consortium for International Development. Boulder,
CO: Westview Press.
1981 A Conceptual Framework for the Analysis of the
Effects of Technological Change on Rural Women.
Technology and Employment Program working Papers WEP
2-22/WP79. Geneva: International Labor
world Food Council
1982 National Strategies to Eradicate Hunger. New
York, NY: World Food Council, United Nations.
Problems of Understanding and
Communication at the Interface
of Knowledge Systems
One of the fundamental justifications for the practice of FSR/E is that its methodologies promote a desirably close relationship between farmers, researchers, and extensionists in the determination of research criteria and the design and choice of interventions. Considerable effort is going into persuading researchers and extensionists to understand that:
The goals and motivations of farmers, which will
affect the degree and type of effort they will be
willing to devote to improving the productivity of, their farming systems, are essential inputs to the'
process of identifying or designing potentially appropriate improved technologies (Norman et al.
much less effort is being devoted to helping farmers to understand the values, rationales, and objectives that lie behind research and extension behaviors. As FSR/E becomes more involved with on-farm trials, particularly farmerdesigned and managed ones, the more important it is for FSR/E practitioners to find ways of articulating their own rationality and making it accessible to farmers.
Some of the FSR/E practitioner rationality is conveyed implicitly in the course of working closely with farmers. A 1985 circular on the Agricultural Technology Improvement Project (ATIP) program, distributed to local agricultural staff in Mahalapaye, Botswana, notes:
In explaining these meetings, it is important the
farmers understand this is a new approach to research in which we want to work together with them to discuss
and evaluate alternatives, rather than just rely on
collecting information from them (ATIP 1985).
But conversations with farmers in areas where FSR/E
teams have been working, reveal that farmers remain greatly puzzled by such things as to why researchers insist that fields or plots be measured in certain ways and what those measurements mean. Or, what it is in the logic of the researchers' world that makes them value, for example, certain livelihood activities such as field cropping above other activities that seem to the farmers themselves ,equally necessary components of their livelihood. The Breakdown of communication and understanding seems greater between women -- as farmers, food processors, traders and consumers - and male researchers. This is due not only to sociocultural distances between them. Male researchers may understand little of the rationality of the domestic domain in their own worlds. The researchers' lack of an explicit frame of reference in this sphere - or use of a partial or biased one -- influences their own set of mental constructs by which they perceive and interpret the world of women within farming systems. Communication difficulties thus ',-'are compounded.
There are a number of threads that might be disentangled here. This-paper will pull out and untwist only one: the logic of flexibility within the domestic domain, illustrated by examples from Lesotho and northern Zambia
(1). The data are not complete from the FSR/E point of
view, having been collected for other purposes, but they do highlight a number of points that FSR/E theory and practice needs to take into account. The paper concludes with suggestions about how this might be done.
The data are drawn from two areas of acute seasonal stress. In the Lesotho site, a longitudinal study of energy flows suggests a bimodal pattern of stress (HussAshmore 1982) while the second area in northern Zambia has a short period of moderately erratic, within-season rainfall and a long, dry, cool period of six to seven or even
seven and a half months, with the time of acute hunger
falling in the January-February period after weeks of heavy labor and declining food stocks (author's unpublished field
Both are areas of high male out-migration and income
insecurity. Risk and loss minimization figure highly in farming system strategies. Women, as household heads and
as farmers within male-headed enterprises, respond to
climatic and income uncertainties by trying to maintain
flexibility, typically in four spheres of the domestic
(1) in production, by maintaining reserve crops and
varieties in household gardens and in wild habitats;
(2) in the timing of operations, volume of product handled,
and technique used, in the spheres of food processing,
storage and food preparation;
(3) by altering the mix, timing, and quality of performance
of their multiple roles; and
(4) by manipulating whatever room there might be for
substitutability of labor and obligation between men and
For reasons of space, the following illustrations are taken from only the first two spheres.
In the peculiarly distorted economic situation of Lesotho, the day to day survival of rural households is largely a matter of how women maintain themselves and their children. The rationality of the farming system is not determined by physical and climatic features - these only set limits to what is possible. It is determined by the rationality of women's lives, that is centered in (though by no means confined to) the domestic domain. Within that domain, there is one resource which is critical: fuel supply. It could be said to be the key both to cropping choices and household food availability. Huss-Ashmore writes:
Because fuel is essential for processing almost all
foods, it can be considered a critical resource for the
maintenance of health and nutritional status. In
Mokhotlong the type of fuel used and the time spent to
procure it, vary according to the seasonal availability
of dung (1982:156).
The preferred fuel is compacted dung, readily available during the winter from the kraal close to the homestead, that, when dried in uniform slabs, burn with the slow, even heat necessary for the long cooking of dried grains and pulses. In households without a kraal, women have to purchase the dung or manipulate k7i-nrelations to obtain it. When the cattle are moved in the summer to the high pastures, women must use horse and cattle dung, picked up from the fields and trails, that is less dense and takes more time to gather. Both sorts are kindled with resinous, woody shrubs that become scarcer as the summer passes but may be the only source of summer fuel if insufficient dung can be gathered. For a short period, kraal dung may be kindled with maize cobs as they are thresFed in the winter.
It is fuel availability rather than food availability that determines which foods are eaten at different seasons:
The supply of slow-cooking protein sources is not used equally throughout the year but is depleted during the
cold season when appropriate fuels are available.
During the summer the population relies heavily on wild
vegetable protein sources, which require more time to
locate and gather but which can be rapidly cooked
one might think that these interactions - and their further entwining with the water/fuel/grain seasonalities of sorghum beer-brewing, a source of income and wage work that is important for being one of the few available to women - are not so terribly difficult for a researcher to discover. However, single visits to households during an exploratory survey may fail to discover seasonalities that are both interdependent across disciplinary boundaries (fuel-forestry; cropping; post-harvest domestic food technology) as well as across gender boundaries (cattle are men's business). The fact that a critical key to the functioning of the farming system operates within the domestic domain may continue to conceal its significance during a verification phase of FSR/E work as well.
A further difficulty arises when researchers try to
measure the quantities involved such as the cooking time of various foodstuffs using different fuels. A~n anthropological study of Sesotho measurement concepts points out:
A woman knows how long to cook vegetables because she
knows when they are ready. One woman, preparing
bread, was asked how she would cook it: Until it is
ready (bo butson). Pressed for precision, she thought carefully -and then said: Five or six hours (li-hora
again the English word). In fact, she cooked the bread
for an hour and a quarter and saw that it was perfect
when she took it from the steam oven. It was ready
both in English terms and her own. At no time did she refer to any kind of time, not even in the sun. It was
not the time that made it ready. It was the cooking
The researcher is concerned with the measurement of time but the woman is concerned with the measurement of "readiness" and there is no reason why the measurement process should not begin with readiness (What does it look like? Is it hard or stiff or does it run? etc.). Instruments such as time allocation studies, useful as they are
as indicators of the range of activities and claims on labor, make invisible whatever it is that women themselves see themselves as allocating or conserving. In HussAshmore's case, women collect wild vegetable proteins not because they are a preference food nor because there is nothing else available, but because women primarily wish to conserve fuel.
The difficulty does not lie in using the measurement units that make sense within the rationality of the user. Researchers and extensionists alike are trained in the concepts of scientific agriculture and these concepts may have no equivalents in the knowledge system of the woman while the woman is trained in the concepts of her indigenous knowledge system and may have no way of knowing the significance of the concepts used by the researcher and extensionist. This has little to do with any differences in ethnic background of the actors and a great deal to do with the difficulty of articulating the rationality of one system in the terms of the rationality of another.
In the northern Zambia location, the fact that local
vegetables and fruits form an important part of the diet is well-established and there are even a few research programs investigating the more important species (MAWD 1983). What is not so readily accepted is that these may have characteristics that yield benefits not provided by modern varieties of the main food crops, however abundant they might be. Interventions that make their production more difficult by switching labor or land use, for example, may also make the seasonal management of diets more difficult, unless the market provides substitutes at affordable prices.
The question of the timing of agricultural innovations with respect to the role of market provision of those goods and services previously supplied within domestic and local economies merits a short digression here. In agriculture in industrial countries and albeit to a lesser degree in irrigated agriculture in developing countries, research organizations work within and for production and knowledge systems that are well-defined, well-organised, and highly interactive. There are at least four main components:
(1) farmers, who are organised and able to contribute to
research progranning through a variety of channels;
(2) powerful industrial organizations engaged in the
business of transforming primary production into a
range of consumer and industrial goods, who are well
able to signal to researchers their own technical requirements or even, by paying for research, to
determine what crop characteristics meet the needs of
their own technical processes;
(3) powerful commercial organizations engaged in the business of wholesaling and retailing produce and processed
foodstuffs, that are able to insist on high quality standards in defence of existing and new sales; and
(4) consumers, who, either through their purchasing power
or through consumer organizations and lobbies, also
signal their preferences to researchers.
The case is quite different in most dryland areas in developing countries. Except perhaps for the richest farmers, producers are not organized nor politically powerful and have few if any links with researchers. The range of transformation processes occurs largely within the domestic domain using local technologies. wholesalers and retailers operate in fragmented and often non-competitive arenas in which the overall level of sales is depressed and quality carries no premium. Finally, consumers have weak purchasing power and few, if any, organized channels for expressing their preferences.
If the inherent yield potential of many dryland areas is judged to be low, with scant chance that the value of the marketed output will ever pay for or induce the kind of infrastructural developments witnessed in irrigated environments, then presumably, it will be necessary to preserve a continuing capacity to derive benefits from the goods and services presently obtained from the biomass through transformation within the local community and the household economy. The challenge becomes that of raising capacity without displacing too many of the benefits presently obtained from within the micro-economy rather than in raising capacity by concentrating on only a few benefits (higher yield) and externalizing the provision of the rest. Varietal characteristics must continue to an unknown extent to meet the demands of domestic transformation processes, technologies, and end uses. The following example describes just such a situation.
The local vegetables (fruits not included here for the sake of simplicity) being produced on one farm during February 1980 at Sambwa in rMpika District of Northern Zambia include: four cassava varieties (masanga uko, matutumushi, muntulunga, ucogo); three finger mle varieties (mwaangwe, mutui a, mwambe); two varieties of groundnuts; and ToalTm-e Eachh~as a very specific
place in seasonal production and food management. For example, mwaangw is a sweet, very early maturing finger
millet that provides one of the first new food crops in the year and a sweet beer for working parties as the main harvesting period approaches. Two of the cassava varieties have palatable leaves (masanga uko and matutumushi - the latter much sought after by the wild pig) but these fall in the cold season (June, July), so some are dried early in the season for later consumption.
In addition to these main food crops, there were five distinct production sites around the compound (and one further away in the dambo or wide valley bottom) tended by the two adult resident women, on which were grown a mix of wild and semi-wild plants and "crops" promoted and officially marketed by the government. A number of the wild plants, such as busoshi (Sesamum alatum) and chimamba (Sphenostylis erecta), also occur n-aturalTy(respectively on disturbed soi and around anthills) but on this farm could be considered as true crops, for they were deliberately planted on chosen sites, protected from chickens and weeds, fertilized with household rubbish, and the product traded in Mpika market. Their utility is partly a reflection of the low and erratic yield of groundnuts but, they have a utility as snack foods at a time of the year when women may cook only once a day or even once in two days. The perennial chimamba ensures that some kind of snack is always going to e available. It would only lose its utility if alternative snack foods were to become available at the critical time of the year when women are busiest or the preparation and cooking time of cassva and millet were to become less or women's cultivation labor were to be reduced.
Another example is provided by the great care the women took to maintain the balance between the availability of the staples (millet, cassava) and the availability of oily or slippery foods for the relish that are added to the nsima or thick, coarse porridge. The nsima is almost ineible in sufficiently large quantitle-swithout such a relish to ease it down. In conditions of scarce and expensive commercial cooking oils, unreliable groundnut harvests in the face of erratic rains, and the time required for shelling and pounding groundnuts, the softness and slipperiness of some local vegetables were highly desired characteristics. P (Fagara chalydea) is an important dry
season resource in this respect. Slippery local vegetables have the additional advantage of needing no blanching or treatment with potash when they are dried for preservation.
Both the men and the women had been experimenting with vegetable production, as the following examples illustrate.
The male head of the compound had been trying white cabbage, tomatoes, and cucumbers from seeds supplied through the Horticultural Marketing Board in Mpika, in a dambo garden at the end of the rainy season. He found Uat the cucumbers grew best but were the least needed for domestic use as they already grew a satisfactory range of cucurbits. The tomatoes were well-liked for their flavor and the softness of the flesh but were tiresome to eat because of their tough skins, the very characteristic that made them suitable for the rough marketing conditions. The women had been experimenting for many years with luManga (Cleome qvnandra) selecting for larger leaf size out7sacri?icing ny of the tenderness. They reported, too, that they could get a higher price in the local market for the larger-leaf variety.
The men in the compound scored consistently lower than the women on the following tests using vegetable sources: identification by sight; recall of the main physical descriptors and husbandry, processing techniques and length of storage; and preparation for eating. Zambian and expatriate members of the nearby agricultural college who were engaged in conducting and supervising trainee extension workers in farm surveys, were asked to share their views of the role of local vegetables in the farming system. They all referred the question to the home economic staff, who were trained to work with "western" vegetables, and with few resources to work in the field, knew only those local vegetables that had been used in the household in their own home areas.
There are further.problems, of communicating knowledge between distinct knowledge systems. The production sites where the local vegetables were grown changed shape and size as women took advantage of rainfall patterns in the advancing season to make additional sowings. At the same time, neither the market value of the product traded nor the opportunity cost of female labor (based on market wage rate) would seem adequate measures of the value of either women's labor time or of the local vegetables to the farming systems. The women themselves used a notion of convenience that appeared to be a combination of characteristics such as: ease of growing near the house; availability (fresh or dried) at moments critical from the point of view of diet management; ease of processing and preparation; timing of labor inputs; and substitutability for other crops. The notion encompassed the principle of flexibility. In this respect, they were reluctant to choose paramount characteristics either for any one crop or between the range of crops. Instead, local vegetables were
viewed as a bundle of biomass that enabled them to manage their resources and responsibilities to the best advantage.
THE IMPLICATIONS FOR FS,/E
There are a number of important "lessons" that can be drawn from this brief review. To summarize, these are reduced to two:
(1) the need to develop methodologies for establishing
the key field-household interactions at an early stag
of the diagnostic process, and
(2) the need to develop methodologies for mutual communication of key concepts across the boundaries of
researchers' and female producers, distinct knowledge_,
Two techniques that might prove to be useful diagnostic instruments for researchers of any background are situation analysis based on the critical incident technique and peer group workshops.
The former is widely used in diagnostic sessions
between researchers and carefully drawn panels of users in industrial and commercial practice. It involves informal but structured interviewing which, as users identify problem areas and describe the boundary conditions, focuses on a "critical incident" that exemplifies one of the problems. The incident is then analyzed in depth, leading into discussion of desirable ways to deal with it. Each of the problems is similar treated in turn.
Peer group workshops are widely used throughout the ESCAP region in the development of local, self-managing groups and income-generating projects and by the Food and Agriculture organization's (FAO) Marketing and Credit division in the promotion of female entrepreneurship. They are based on the understanding that knowledge and expertise exists among local communities, together with a diagnostic capacity attuned to local realities. They draw on the expertise of those who are locally recognized as knowledgeable within the subject problem area by facilitating the preparation of case studies of their successes. These cases are exchanged and analyzed at workshop sessions, leading to identification of interventions that would allow these successes to be replicated. A great deal of experience now exists to guide the preparation and implementation of workshops with those who have little formal education and to facilitate the participation of service officers (agricultural researchers or extension workers).
NOTES AND ACKNOWLEDGEMENTS
(1) The research was carried out between January 1979 and September 1980 in the Central, Northern and Luapula Provinces of Zambia by members of the Rural Development Studies Bureau, University of Zambia, Lusaka.
1982 Seasonality in Rural Highland Lesotho: Method
and Policy. In A Report on the Regional Workshop on Seasonal variations in the Provisioning, Nutrition and Health of Rural Families, pp. 147-161. Nairobi:
ministry of Agriculture and Water Development (MAWD)
1983 Handbooks for Agricultural Field workers,
Zambian Local Vegetables and Fruits. Lusaka:
Department of Agriculture, Ministry of Agriculture
and Water Development.
Both these techniques have the added advantage that
they eliminate some of the stages of "translation" of knowledge concepts and, with careful preparation, it is not too
difficult to identify those items which, though denoted
differently, refer to a standard unit. Returning for example, to the case of the cake that is "ready", the
researcher can measure the hours it takes to cook and the
baker the cooking that is needed to make it "ready". Both
are referring to a standard referent, although the baker might be interested in the number of mouths it feeds and
the researcher in its unit weight and composition. The
difficulty comes when one is using a knowledge concept that
has no referent in the knowledge system of the other. The
-,",,,-difficulty is, in a sense, one-sided. Researchers are
ften keen to learn about and understand the concepts of
producers but have little awareness of the constructs and alues inherent in their own knowledge system. Where the
knowledge system of male agricultural researchers and
extension officers does not encompass either an experiential nor trained understanding of the domestic economy,
the problem seriously undermines FSR/E practitioners,
claims to be conducting systems based technology
Norman, D., E. Simmons, H. Hays
1982 Farming Systems in the Nigerian Savanna,
Research and Strategies for Development. Boulder,
CO: Westview Press. Waliman, S.
1965 The Communication of measurement in Basutoland.
Human organization 3:4:236-243.
From Recommendation Domains to
Intra-Household Dynamics and Back:
Attempts at Bridging the Gender Gap
Arnalia Al. Albert!
One of the concepts to emerge from the Farming Systems approach to agricultural research and extension is that of the recommendation domain. Defined as "a group of roughly homogeneous farmers with similar circumstances for whom we can make more or less the same recommendation" (Byerlee, et al. 1980), the underlying assumption is that farmers of households within the same recommendation domain will have similar responses to proposed technology (Shaner, et al. 1982: 44). Recommendation domains are used to focus the research process and expedite dissemination of the technology thereby facilitating the extension phase.
The debate continues in the farming systems literature and among farming systems practitioners about both the more relevant criteria and the preferred timetable to identify and to elaborate recommendation domains. The position of those who maintain that the early delineation of recommendation domains precludes considerations that are not readily evident or initially salient (Cornick and Alberti 1985), is countered by others who maintain that the early identification of recommendation domains permits their progressive refinement (Franzel 1984). Still others (Norman and Baker 1984) point out that in the last analysis both the target groups identified and the nature of the technology recommended tend to reflect the expertise of the team members in a particular Farming Systems Research/ Extension (FSR/E) project.
This paper argues that, first, recommendation domains sensitive to gender issues are difficult to develop due to scant documentation of women's participation in agricultural and farm-related activities in local areas, and second, if developed, they are difficult to implement due to several features commonto many FSR/E projects. Indeed, it seems that the greater the pressure for prompt
elaboration of recommendation domains, the greater is the likelihood that women's roles, as well as their concerns within the FSR/E context, will be overlooked because of insufficient time to draw them out. The long term solution to satisfactorily addressing gender issues in farming systems, however, lies less in attempts to develop appropriate recommendation domains and more in efforts to revise the FSR/E framework so that gender issues are deliberately and self-consciously entertained. Until these changes occur, several key questions are proposed to assist FSR/E practitioners in assessing what gender related issues are potentially relevant in a particular FSR/E site and whether or not they can be addressed feasibly within the existing project framework.
OBSTACLES TO DEVELOPING
GENDER SENSITIVE RECOMMENDATION DOMAINS
Among the more common techniques suggested for the initial stages of problem diagnosis leading to the formation of recommendation domains are reviews of secondary data, informal interviews with persons such as local officials, residents, and extension workers, and an exploratory survey of farmers sometimes combined with or followed by a formal survey (Harrington and Tripp 1984; Shaner, et al. 1982). The obstacles to uncovering the extent of women's involvement in the total or select phases of a farming system embedded in each of these techniques are discussed briefly.
Secondary Data Reviews
Much of the literature on women in agriculture published within the last ten years underscores the extent to which the involvement and contributions of women in this area have been underrepresented (Deere and Leon 1981; Lewis 1981). Nevertheless, secondary sources such as census data and local agricultural reports that continue to ignore or underestimate female contributions abound. When FSR/E staff consult these materials they are likely to accept the data as factual unless they are aware of the possibility that female participation in agriculture may be masked or otherwise distorted. only when sensitized to this bias may they be persuaded to seek additional corroboration before dismissing gender as a potentially relevant variable.
Informal Interviews'and Exploratory Surveys
Local officials and extension agents can often provide site-specific information that a FSR/E project staff member would be hard-pressed to obtain so efficiently otherwise. Information on female involvement in agriculture, however, is less likely to be obtained for several reasons. First, cultural values may intervene. When female agricultural activity is associated with poverty, not only are male officials unlikely to discuss such activity on the part of female members in their own household, but they may well be reluctant to discuss such activity on the part of female residents in general presuming that it would reflect negatively on the socioeconomic status of the community.
Assuming that these local officials and extension agents are almost always exclusively male, attempts to adjust for these gender-related "blind-spots" by speaking with their wives or other female household members may not yield substantially different results (Alberti 1980). To the extent that these women partake of the elevated social status of their households, they are unlikely to make public their own involvement in farm-related tasks or to imply difficult socioeconomic conditions within the community by referring to such activity on the part of other women unless it is to demean them.
Second, it has been found that male farmers routinely underestimate the degree and undervalue the importance of female involvement in farm-related activities in which they too participate (Bourque and Warren 1981; Deere and Leon 1981; Alberti 1986) and ignore or are unaware of the extent of female involvement in farming activities in which they do not share. Hence asking male farmers about the participation of females in agriculture will not necessarily elicit accurate information.
Finally, the reluctance of the national male FSR/E
staff members, especially if they are from the local area, to ask questions that are deemed inappropriate by local standards must be considered . moreover, cultural norms may restrict male field staff members' access to women for interviews. As yet another possibility, male FSR/E staff members may resist interviewing women because of their own attitudes about female participation in agriculture.
The advantages and limitations of formal surveys have been widely discussed. Within the farming systems literature, Chambers (1981, 1983) Chambers and Ghildyal (1985) is
perhaps the most outspoken and graphic critic as he conjures up visions of "30 pages of questionnaire . which if asked are never coded, or if coded never punched, or if punched never processed . examined . or analyzed. that a number of us have also seen (1980: 4).
Two points about formal surveys on women's involvement in agriculture must be raised. The first is that preparing a questionnaire assumes knowing what is necessary and how to design the queries. While this ought eventually to be true for initial surveys conducted during project development, this is not always the case for women's issues precisely as a consequence of some of the limitations just discussed. Secondly, many formal surveys are designed to be administered to either the male or female head of household, but not to both. Generally, the household member available when the interviewer arrives responds. However, the survey form frequently lacks an item to indicate who was actually interviewed and whether or not that person was male or female. Hence, even if relevant questions about women's involvement in agriculture and farm-related activities are included, it is impossible to disaggregated male and female responses and to analyze them for consistency and comparability.
LOCATING WOMEN IN THE FARMING SYSTEMS CONTEXT
Given these constraints what site-specific information might be readily available that would expedite developing recommendation domains sensitive to gender differences? Readily available refers to information that could be elicited over a few days through informal conversations with local residents, teachers, and other persons working in the area. Rapid Rural Appraisal procedures recommend doing this in conjunction with a field trip around the project area (Chambers 1980; Beebe 1985). The field trip is essential to provide visual information to accompany verbal accounts. Lines of inquiry otherwise not considered may be opened when the information from these two sources does not concur.
The information obtained from responses to the following questions ought to enable FSR/E practitioners to contextualize the situation of women in the FSR/E setting in broad strokes. At the same time, it would facilitate a quick assessment of whether or not the FSR/E project, as it exists or could feasibly be modified, can viably address the gender issues relevant to that site. Where addressing those issues is possible,
collecting the kind of information needed should follow ideally to inform analyses of intra-household dynamics in FRS/E (See for example, Flora 1984; Feldstein 1987).
What Are the Local Cultural Norms Regarding Female Agricultural Activ Is Mo e Than One Culture Represented in the Project Area?
In many parts of Latin America, particularly indigenous regions of the Andes, women work side by side with men in the fields. In other areas, such as Honduras, women are seldom seen working in the fields beneath the direct rays of the sun and may well be embarrassed if they are seen. Asian women such as those from Bangladesh are rarely field workers while many of their Indian counterparts assume the major role in most if not all phases of rice production.
The differences are largely the result of cultural variations whose dominant mode of expression may be religious or ethnic or some combination of the two. What is important is that when a certain portion or subportion of the population of an area shares a particular cultural orientation, it is possible to make certain assumptions about the kinds of roles women are likely to assume within an agricultural setting and to be forthcoming with information about those roles. For example, if visible agriculturally productive activity on the part of women is highly circumscribed, it can be expected that even when women do engage in such endeavors, they will be extremely difficult to document.
When more than one cultural group is represented in an area, additional factors may come to bear on the situation. Is one group dominant and the other subordinate? Is the participation of women in agricultural and farm-related tasks the same for both groups? Are the norms regarding such involvement the same? If the norms vary, which norms do agricultural extensionists and field workers represent?
In culturally complex settings, it is important to specify the cultural group or groups to which a recommendation domain applies. This should help clarify and explain what would otherwise be unanticipated responses to a recommended technology. Factors that might be involved include differential access to extra-household labor by ethnic group or different production objectives despite use of the same traditional technologies.
Does Women's Participatio in Agriculture Vary by Socia CTass? If so, in What Ways?
There is an ever-growing consensus that participation by households, and by women within those households, in the farming system is highly contingent on social class. Women from land-poor households who engage in farming tasks tend to work longer hours at those tasks and generate proportionately lower returns than other women. Often they are the women who have been left behind while their male partners migrate in search of wage employment. Women from landless households are clearly the most vulnerable as they are increasingly dependent on an ever more tenuous agricultural wage labor market that relegates them to more restricted and marginal employment opportunities even as it expands commercially (Hart 1978; Stoler 1977; Sen 1985; Stolcke n.d.; Young 1985; Horn and Nkambule-Kanyima, 1984; Chaney and Lewis 1980).
While these trends may be widespread, they are not universal. Knowing whether or not they are valid for a particular setting should provide some clue as to how candid men or women are likely to be about female involvement in agricultural and farm-related activities.
Do women Specialize in Food Production and Subsistence Agriculture?
Despite broad variations in patterns, the preeminent role of women in the production of food for home consumption appears to cross continental bounds (Chaney and Lewis 1980).
In Latin America the evidence is widespread that the majority of women who directly engage in agricultural production at the household level primarily raise basic crops intended for home consumption though they may also market small portions of those crops. If the household also raises a cash crop, it is likely to be under the control of the male head of household, even when women contribute labor to its cultivation. The more the household's agricultural activities are commercially oriented, the less likely it is that women of the household will be directly involved in agricultural production. However, when hired laborers are present, women of the household are usually expected to provide support services such as food preparation and are occasionally called upon for managerial activities (Deere and Leon 1981; Bourque and Warren 1981; Alberti 1986).
In Asia, the scenario is distinctly different. Despite broad variations in the extent of women's direct involvement in rice-based agricultural economies due to ethnic and religious differences, women are always involved in the processing of rice and frequently bear major responsibility for its transplanting, weeding, and harvesting. When the household's access to rice fields is insufficient to meet its own consumption needs, women as well as men are likely to seek work as agricultural laborers with rice as the preferred medium of payment. Participation in the harvest of kin and neighbors, if not in the planting as well, is another strategy geared to insure a ration of rice (Hart 1978; Sen 1985; Dey 1985). In each instance, the overarching objective is to obtain food that can be immediately used by the household.
In contrast with rice cultivating areas, in the Philippines for example, the cultivation of cash crops such as coconut, tobacco, and commercial varieties of root crops such as cassava and sweet potato, is dominated by men. Root crops grown for home use, however, are often under the immediate control of women (Cornick and Alberti 1985).
Until recently, the situation in Africa presented what had probably been the most consistent association between crops and gender. Even now, food crops are grown almost exclusively by women, though some women, particularly those near urban areas, have begun to cultivate cash crops as well (Ferguson and Horn 1985). In contrast, men continue to concentrate their efforts in cash crop production. Getting a sense of the pattern that predominates for a given FSR/E project should help us identify the crops and animals that women tend to work with as well as to assess the FSR/E project's capabilities in those areas.
UTILIZING THE INFORMATION
WITHIN A FARMING SYSTEMS FRAMEWORK
After a discussion of how this information may be
fruitfully used, the following areas must be considered. First, the knowledge gained should enable researchers to better identify the variables that are particularly relevant in relation to women in farming systems in the project area. Second, it can provide guidelines to estimate the validity of the information and data that does exist. Third, it highlights the kind of information that is available while giving some indication of what is lacking. This should help to assess what additional information is needed and to appraise how sensitive its collection may be.
For example, the knowledge that there are two ethnic groups within the FSR/E project area should immediately prompt the question as to whether or not their attitudes toward agriculture and women's involvement in agriculture are the same. If they differ, ways of systematically distinguishing responses by ethnic group becomes important.
The shortcoming of these illustrations is that real
life situations rarely fall into compartments that vary so neatly along a single dimension. Rather, multiple variables combine and fuse, whether systematically or erratically, resulting in ever more complex relationships. Their salience is heightened as they interact with some of the more common features of farming systems projects. Let us examine some of these characteristics and the way they interact with gender concerns.
FARMING SYSTEMS CONSTRAINTS
AND THEIR IMPLICATIONS FOR GENDER ISSUES
Site selection for farming systems projects often
results from political and economic decisions that occur outside project bounds (Shaner et al. 1982; Harrington and Tripp 1984). marginal areas are less likely to be selected. Not only do they tend to lack political leverage, but projects in such areas are more prone to failure due to their residents' restricted access to resources. women who engage in agricultural and farm-related activities, however, are frequently concentrated among the resource poor who are commonly located in more marginal areas.
Despite occasional efforts to the contrary, farming systems projects are frequently commodity-oriented either as the result of project mandate, team member expertise, or a combination of these factors (Norman and Baker 1984). Furthermore, a commodity orientation is frequently aligned with a commercial orientation. As has been discussed, women are more likely to cultivate food crops with a view to household consumption. Hence, when a FSR/E project has a commodity orientation it may implicitly ignore women by excluding crops of most concern to them.
Another constraint is that FSR/E projects tend to adapt already existing technology, or "shelf technology," to a particular situation, rather than to develop new technology for a specific situation. They justify their approach on the basis of insufficient resources and a time frame inadequate to allow for additional research. However, existing technologies have tended to be capital intensive, and until recently, to give demonstrated results only when adopted as
an entire package, rather than in steps over time. To the extent that the women who engage in agricultural and farmrelated activities are concentrated among the resource poor, they may be unable to adopt the new technology because of insufficient cash resources, or if they have the resources, they may be unwilling to adopt the new technology because it is inappropriate to their goals when they are subsistence rather than commercially oriented.
Finallyto paraphrase Chambers, these factors interlock (1980:3). As Harrington and Tripp note: "domains are formed so that researchers can effectively deal with the majority of farmers in a particular area" (1984: 14). However, the only majority that women tend to constitute as household level agriculturalists is that of the rural poor. Nevertheless, even among these, some women have partners, others are single, some are the only farmer in the household, and still others are the only sources of labor. Though women who directly engage in farming and farmrelated activities are unlikely to be wealthy, it is likely that there is considerable variation in their access to resources, even among those broadly labelled as "poor."
Women in agriculture tend to share a disadvantaged
position in male-oriented agricultural research and development programs: The way they experience that disadvantage, however, is mediated by their culture, resources, and civil status. It is difficult for recommendation domains that depend on homogeneous circumstances in key variables to locate issues that relate to "women" equally despite their diversity. what a true incorporation of gender issues in farming systems implies is a revision of the farming systems unit of analysis from the household to the male and female members within the farming systems household for the stages of problem diagnosis and design. The information thus provided would enable farming systems practitioners to make conscious though difficult choices about where the FSR/E resources will be channelled, knowing full well and in advance how those choices are likely to affect men and women differentially.
Alberti, A. M.
1979 Metodologia Apropriada para el Estudio de la
Mujer Rural en los Andes del Ecuador. Presented at the workshop Metodologia Apropriada para Estudiar a
la Mujer Rural. Quito: CEPLAES/Ford Foundation.
1982 Observations on the Productive Role of Women and
Development Efforts in the Andes. Proceedings of
the First Women in Development Workshop sponsored by HIID/MIT. Cambridge, MA: Harvard University Press.
1986 Gender, Ethnicity, and Resource Control in the
Andean Highlands of Ecuador. Ph.D. Dissertation.
Stanford, CA: Stanford University. Beebe, J.
1985 Rapid Rural Appraisal: The Critical First Step
in a Farming Systems Approach to Research. FSSP
Networking Paper No. 5.
n.d. Factoring New Information into Decisions About
On-Farm Trials. USAID Mission, The Philippines. Bourque, S. C. and K. B. Warren
1981 Women of the Andes: Patriarchy and Social
Change in Two Peruvian Towns. Ann Arbor, MI:
University of Michigan Press. Chambers, R.
1981 Rapid Rural Appraisal: Rational and Repertoire.
Public Administration and Development 1:95-106.
1983 Rural Development: Putting the Last First.
Chambers, R. and B.P. Ghildyal
1985 Agricultural Research for Resource-Poor Farmers:
The "Farmer-First-And-Last" Model. Agricultural
Chaney, E. M. and M. W. Lewis
1980 Women, Migration, and the Decline of Smallholder
Agriculture. Washington, D.C.: USAID Office of
Women in Development.
Cornick, T. R. and A. M. Alberti
1985 Recommendation Domains Reconsidered. Presented
at the Farming Systems Symposium: Farming Systems
Research and Extension: Management and Methodology,
Kansas State University, Kansas.
Deere, C. D. and M. Leon de Leal.
1981 Women in Agriculture: Peasant Production and
Proletarianization in Three Andean Regions. ILO
Rural Employment Policy Research Program:
Unpublished Working Paper. Dey, J. M.
1985 The Concept and Organization of a Research
Network on Women in Rice Farming Systems. In the Report of the Project Design Workshop on Women in
Rice Farming Systems. Los Banos, Philippines:
Feldstein, H., S. Poats, K. Cloud, R. Norem
1987 Intra-Household Dynamics and Farming Systems
Research and Extension Conceptual Framework.
Population Council/FSSP-University of Florida
Ferguson, A. and N. Horn
1985 Situating Agricultural Research in a Class and
Gender Context: The Bean/Cowpea Collaborative
Research Support Program. Culture and Agriculture
Flora, C. B.
1984 Intra-Household Dynamics in Farming systems
Research: The Basis of Whole Farm Monitoring of
Farming Systems Research and Extension: A Position
Frankenberger, T. R.
1985 Adding a Food Consumption Perspective to Farming
Systems Research. Prepared for the Nutrition
Economics Group, Technical Assistance Division,
Office of International Cooperation and Development,
United States Department of Agriculture. Franzel, S. C.
1984 Modeling Farmers' Decisions in a Farming Systems
Research Exercise: The Adoption of an Improved
Maize Variety in Kirinyaga District, Kenya. Human
Organization 43:3:199-207. Harrington, L. W. and R. Tripp
1984 Recommendation Domains: A Framework for On-Farm
Research. CIMMYT Economics Program Working Paper.
1985 Formulating Recommendations for Farmers,
Researchers and Policy-Makers: Issues in the Design
and Use of Socio-Economics Research. Presented at
the ESCAP CGRPT Center Conference "Towards
Recommendations for Research, Policy and Extension: Methodological Issues in Socioeconomics Analysis of
Coarse Grains and Food Legumes," Bandung, 18-23
1978 Labor Allocation Strategies in Rural Javanese
Households. Ph.D. Dissertation, NY: Cornell
Horn, N. and B. Nkambule-Kanyima
1984 Resource Guide: Women in Agriculture in
Botswana. Michigan: Bean/Cowpea Collaborative
Research support Program. Lewis, B. C.
1981 Invisible Farmers: women and The Crisis in
Agriculture. Washington, D.C.: USAID Office of
Women in Development.
Norman, D. W. and D. C. Baker1984 Components of Farming Systems Research FSR
Credibility and Experience in Botswana.
Agricultural Technology Improvement Project,
Department of Agricultural Research. Botswana,
Africa: ministry of Agriculture. Sen, G.
1985 Women Workers and the Green Revolution. In
Lourdes Beneria, ed., Women and Development: The Sexual Division of Labor in Rural Societies. NY:
Shaner, W. W., P. F. Philipp, and W. R. Schmehl
1982 Farming Systems Research and Development.
Boulder, CO: Westview Press. Stolcke, V.
n.d. Social Inequality and Gender Hierarchy in Latin
America. Position paper presented at the SSRC
1977 Class Structure and Female Autonomy in Rural
Java. Signs 3:74-89.,
1985 The Creation of a Relative Surplus Population:
A Case Study from Mexico. In Lourdes Beneria, ed.,
Women and Development: The Sexual Division of Labor
in Rural Societies. NY, NY: Praeger Publishers.
Research, Recommendation and
Diffusion Domains: A Farming Systems
Approach to Targeting
Peter Wotowiec, Jr., Susan V Poats, and Peter E. Hildebrand
TARGETING FARMING SYSTEMS ACTIVITIES: HOMOGENIZING VARIABILITY?
Inherent in the farming systems approach is the recognition of the variability of the complex circumstances farmers face while managing farms that are comprised of inter-related crop, animal, household, and off-farm enterprises. Diversity in farming systems must be recognized in developing appropriate technologies for the farmers that manage those systems. However, it is not practical to conduct research tailored specifically to a few individual farmers. Targeting entails the grouping together of similiar clientele so efforts can be sufficiently focused. Although the concept of targeting might seem contrary to the recognition of heterogeneity among farms, it is an essential component of the farming systems approach. when Farming Systems Research and Extension (FSR/E) practitioners target a group of farming systems as relatively homogeneous based on a few simple factors, the existing variability among farms is often not sufficiently considered. How can FSR/E teams define and target homogeneous groups of farming systems without losing sight of the heterogeneity among them? Farming systems practitioners take different positions on this issue (Cornick and Alberti 1985).
One perspective stresses the early definition of homogeneous groups of farmers using the recommendation domain concept to guide subsequent research activities. Collinson (1979, 1980), Gilbert et al. (1980), and Franzel (1985) advocate ex ante delineation of recomendation domains based on secondary data and preliminary surveys, followed by a formal survey to refine the domain boundaries. Both Collinson and Franzel describe a technique of defining recomendation domains through interviews with extension agents and local authorities before actually initiating activities with farmers. Early definition of
recommendation domains is usually based upon a few relatively easily identifiable factors such as soil type, agroecological zones, crop type, and management (Harrington and Tripp 1985). These authors note the importance of continuing the refinement of domain boundaries.throughout the sequence of on-farm adaptive research, but the subsequent reassessment of recommendation domains is often not vigorously pursued.
A more recent view states that grouping farming systems should not take place until the researchers have an adequate understanding of the variability inherent in local farming systems, usually not accomplished early in the work in an area. Cornick and Alberti argue that recommendation domains established early are frequently poorly conceived and lead to a premature assumption of homogeneity. The failure to consider potential variability'from factors such as long-term climate induced trends in cropping patterns, household decision-making and labor allocation, or relationships between on- and off-farm activities, may bias subsequent technology development. For example, Cornick and Alberti (1985:1) note:
. the roles of women and children that can be critical
factors in the development and subsequent adoption of technologies are often explicitly excluded from consideration in recommendation domains. This occurs because
the usual time frame for development of recommendation
domains is inadequate to the task of understanding intrahousehold dynamics and the importance they hold in the
In particular, socioeconomic factors are often not
fully integrated into domains defined early, either because of the longer period of time necessary to gather this information, or because of the absence of trained social scientists as part of farming systems teams. One area often poorly covered in early definitions of domains is the different agricultural roles of men and women. Proceeding with on-farm research and other activities on the basis of a hastily achieved assumption of homogeneity could result in inefficient subsequent research and the promotion of solutions that are not appropriate to farmers (Cornick and Alberti 1985:25) or technologies that may favor some farmers (male) while causing disadvantages for others (female).
In this paper the issue of variability versus
homogeneity in the targeting of farming systems research and extension activities is explored. After a brief review
of targeting in FSR/E using recommendation domains, problems in the conventional use domains in FSR/E are described in an attempt to bring together the two differing viewpoints and to begin to resolve the question. The refined concept of targeting allows for better inclusion of gender variables in the definition of domains.
OVERVIEW OF TARGETING AND RECOMMENDATION DOMAINS
Targeting for Efficiency and Social Equity
FSR/E must differentiate between various potential
farmer-client groups and determine the particular needs of each, if technologies are to be developed that clearly meet those needs (Byerlee and Hesse de Polanco 1982). Most literature on the subject of targeting in farming systems has stressed the increase in efficiency of FSR/E activities made possible through focusing upon specific, relatively homogeneous farmer groups.
Efficiency in allocation of research resources is
essential if a program is to reach and benefit a maximum number of farmers. By focusing scarce resources upon roughly similar groups of farmers, research programs are able to carry out investigations on a selected number of representative farms and later can transfer the findings to the comparable situations faced by other farmers.
Targeting is also important in justifying the farming systems approach to institutional policy makers who are concerned about social equity in the distribution of resulting benefits. Farming systems practitioners use targeting concepts to assist them in making decisions which increase the likelihood of an optimal distribution of research results among the members of a community.
Conventional Concept of Recommendation Domains
The concept of "recommendation domains" has been widely used in targeting farming systems research since Perrin et al. (1976) first introduced the idea. It is described and defined by Byerlee et al. (1980:899) in the following
. a recommendation domain (RD) is a group of farmers with roughly similar practices and circumstances for whom a given recommendation will be broadly appropriate. It is a stratification of farmers, not area: farmers, not fields, make decisions on technology.
Socioeconomic criteria may be just as important as
agroclimatic variables in delineating domains. Thus
resulting domains are often not amenable to geographical mapping because farmers of different domains may be
interspersed in a given area.
Using this definition, neighboring farm households
might be placed in different recommendation domains because of differences in availability of family labor. In societies where women cultivate different crops than those of the men, female farmers could comprise a recommendation domain separate from male farmers even if they are from the same household.
Expanding Upon the Definition of Recommendation Domain
Perrin et al. (1976) originally conceived of the notion of recommendation domains as an aid to researchers for targeting the development of technologies to specific audiences. The concept has been expanded since then to include a number of additional situations and purposes. Some of the most common applications of recommendation domains include the following gleaned from current literature on the topic:
(1) making policy decisions;
(2) identifying priority issues for research;
(3) specifying clientele for developing recommendations;
(4) selecting representative sites and famercooperators;
(5) focusing analysis of surveys and on-farm trials;
(6) orienting extensionists to groups of similar
(7) transferring adapted technology to appropriate
(8) enhancing equitable distribution of FSR/E benefits.
AS Harrington and Tripp (1985) point out, the domain concept is vital to every stage of the on-fam research process. However, it is apparent from reviewing the literature on the subject that the definition of "recommendation domain" not only changes at each stage, but also varies according to the individual who applies it as well as to the end result. The wide variability among farmers and farms, and the dynamic nature of the farming systems development sequence, contribute to the confusion that
exists among FSR/E practitioners as to the general meaning and use of the term recommendation domain.
On-Farm variability and Conventional Recommendation Domains
The emphasis by Byerlee et al. (1980) upon "farmers,
not fields" as the sole basis for the delineation of recom mendation domains is not always warranted because of the
variability found in some field situations. Cornick and Alberti (1985) cite the case of farmers in the community of
Quimiaq in the mountains of Ecuador who manage different cropping patterns in different agro-ecological zones, a product of altitude, temperature, and rainfall variation on the mountain slopes. Not only does each farm cross agroecological zones, but the cropping patterns found in each field vary greatly from year to year. For example, depending upon a farmer's perception of trends and yearly changes in climatic conditions, bean or fava bean intercrops will be assigned to maize fields located at varying elevations along the slope.
Gender and intra-household variables are often
neglected in the process of defining a recommendation domain because of the relatively more difficult and time consuming task of collecting and analyzing data on these variables. Existing information on gender and household variables often offers few useful insights for defining recommendation domains when compared to the secondary data available on agroecological characteristics. In addition, the gender and household data that may exist may be unobtainable locally. Nevertheless, superficial understanding of these variables or the transfer of erroneous assumptions without continued investigation can hamper design and delivery of appropriate technology.
Refining the Concept of Domains
The argument here is that the issue of targeting in FSR/E has become confusing because the definition of the term "recommendation domain" has been stretched to cover too many situations and too many different purposes. Farming systems practitioners must develop a common understanding of how the use and definition of "domains" change as the farming systems sequence progresses from initial characterization through problem diagnosis, testing, adaptation, evaluation, and finally, to the delivery of the new technology to farmers.
It is essential to account for the heterogeneity in farming systems, even while delineating relatively.
homogeneous groups. Refinement and expansion of the use of domains in targeting will enable researchers to distinguish applications of the domain concept, while still recognizing the diversity among farm households and farming systems.
This can be accomplished by recognizing a problem focus in the definition of the domains, by tying the changing concept of domain more closely to the farming systems sequence, and by stressing a greater inclusion of socioeconomic considerations into the targeting process. The refinements outlined below are a sharpening of focus not a changing of terminology, that will lead to increased utility of this method of targeting in the field.
Any of the three types of domains described below may be defined within specific geographic boundaries for ease in conceptualization, but it is imperative to realize that domains do not necessarily include all the area within the boundaries prescribed. Because domains are based upon a specified problem focus and upon socioeconomic considerations in addition to the more geographically mappable factors of climate, altitude, and soil, they are actually interspersed intermittently in a discontinuous pattern throughout a geographic area.
The examples here will emphasize gender as a key factor in delineating domains; other factors, such as class, education, .language use, or food preferences, could also be used.
Research Domains: Targeting for Variabilit
A "research domain" is a problem-focused environmental (agro-ecological and socioeconomic) range throughout which it is expected that hypothesized solutions to a defined problem could have potential applicability and therefore should be tested. Research domains are determined during the initiation of research activities, largely by consideration of biophysical (agro-ecological) factors, with some attention to socioeconomic and gender issues.
Recommendation Domains: Targeting for Homogeneity
Research domains are comprised of one or more agrosocioeconomic "recommendation domains", that are tentatively defined based upon the response of a specific technology to the actual agro-socioeconomic conditions found on farms. A recommendationn domain" is a group of farmers (or farmers and their fields) with a common problem for whom a tested solution meets their biophysical and socioeconomic requirements for adoption.
In the Ecuadorian case cited by Cornick and Alberti, recommendation domains would be based not only upon farm households, but also upon their separate fields that are not contiguous but widely dispersed in location and altitude. Each household might fall into several recommendation domains depending upon: (1) where their fields are located along the agroecological gradient of the mountainside; (2) the climate-related crop management decisions made for each of those fields; and, (3) the particular problem solutions to be tested.
other examples from West Africa demonstrate how gender can be used to differentiate recommendation domains. in many areas, men and women have separate fields, often inherited from their same sex parent, that are not managed communally by the household. Women traditionally grow rice on their lands while men produce upland crops such as groundnuts or sorghum on their own fields. In this system, fields managed by a household pertain to different recommendation domains depending upon both the cropping system and the gender of the farmer manager. In one area of the Ivory Coast, men plant yams in a cleared field. Women will often care for the yam plants by weeding them while they plant their vegetable crops in the space between the yam plants. In this system, fields are neither men's nor women's, nor would entire fields fall into a single problem-focused recommendation domain. Rather, domains would be determined by crops and their managers, male or female, and contain pieces of many fields.
Recommendation domains are seen as tentative in nature throughout the on-farm adaptive research process. Recommendation domains are initially hypothesized by the FSR/E team on the basis of on-farm exploratory and refinement trials, information collected through directed surveys, and subsequent on-farm verification trials. Over time, as more information is gathered, the recommendation domains are refined and redefined to closer approach reality.
Diffusion Domains: Targeting for Communication
"Diffusion domains" are interpersonal communication networks through which newly acquired knowledge of agricultural technologies naturally flows (Hildebrand 1985). Informal flow of information through a community grapevine is substantial (Rogers 1983). From farmer to farmer, neighbor to neighbor, store operator to patron, information about new ideas moves through a farming community. Awareness of a new technology being verified in on-farm trials